Replimune Group, Inc. (REPL) Earnings Call Transcript & Summary

Good morning, and thank you for joining our R&D investor event. We are excited to have this opportunity to share with everyone a number of important data and strategic program updates. Before we begin, I need to highlight our forward-looking statement. On Slide 3, I'm Philip Astley-Sparke, Chief Executive Officer of Replimune. And I have with me today from management, Robert Coffin, our President and Chief R&D Officer; and Sushil Patel, our Chief Commercial Officer. We're also very pleased to have with us today several leading physicians to present data and provide their insights on our development plan. This include Dr. Kevin Harrington, Professor in Biological Cancer Therapies at The Institute of Cancer Research, London, and Consultant Clinical Oncologist at The Royal Marsden; Dr. Nikhil Khushalani, Chief of Medical Oncology at the Moffitt Cancer Center; Dr. Muneeb Ahmed, Interventional Radiology Division Chief at the Beth Israel Deaconess, Associate Professor of Radiology at Harvard Medical School and President Elect at Society of Interventional Radiology; and Dr. Tony Saab, Leader of Gastrointestinal Cancer Program at the Mayo Clinic. Slide 4 shows the agenda, which will start with a brief further introduction from me, followed by 3 content sections. The first section will cover our plans to establish a major RP1 skin cancer franchise, starting with cutaneous squamous cell carcinoma where we will provide a detailed overview of the indication and our thoughts in relation to commercialization. We will also be presenting updated data from the completed cohort from the Phase II IGNYTE trial that continue to support our registrational studies in anti-PD-1 naive cutaneous squamous cell carcinoma and anti-PD-1-failed melanoma. Additionally, we are presenting today new RP1 data from the cohort being enrolled in anti-PD-1-failed non-melanoma skin cancer and from the ARTACUS clinical trial with RP1 as monotherapy in solid organ transplant recipients with skin cancer. In the second section, we will present on the feasibility of deep injections with a focus on injections into the liver, as we look to establish franchises beyond predominantly superficial tumor types before presenting on the indications we intend to pursue with RP2/3 in the final segment. Moving to RP2/3, we have seen strong initial data with RP2, which supports the advancement of RP2/3 into mid-stage development in new and difficult-to-treat tumor types. As a result, we're announcing today a broad RP2/3 Phase II development plan, including new clinical trials in squamous cell carcinoma of the head and neck, hepatocellular carcinoma and colorectal cancer. Following the presentation, we'll be holding a live Q&A. To ask a question, you'll note there's a question submission bar on your screen that will be open throughout the presentation this morning and during the Q&A session. We will collect those questions throughout the morning and take as many as time allows. All questions will be held until the Q&A portion of the event. Moving to Slide 5. We are continuing to make great progress towards our ambition of making our products a cornerstone of immuno-oncology treatment regimens as the most practical and effective way to initiate a systemic antitumor immune response. Our mission is to deliver transformational results for patients across cancers using tumor directed oncolytic immunotherapy to induce a powerful and durable systemic antitumor immune response resulting in quality survival and a chance for cure. Moving to Slide 6. We are the leader in tumor directed oncolytic immunotherapy with 3 wholly owned programs. As aforementioned, we plan to establish a broad-based RP1 skin cancer franchise and look forward to reading data from 2 skin cancer registration studies in late '22, early '23. With a broad-based, mid-stage development plan now in place for RP2/3 beyond skin cancers, the company has the potential to generate substantial revenues in the '25 to 2030 time frame. We've also made solid progress putting in place the building blocks to build an entity capable to transform the immuno-oncology landscape. Our own manufacturing facility is fully operational, filling GMP batches. Our comparability work to the contract material used in our clinical studies to date is largely complete. In addition, our Chief Commercial Officer and his team are continuing the planning process to ensure, if approved, our products are widely adopted in the marketplace. Finally, we have a strong balance sheet to execute on our vision. As a reminder of our technology and MOA on Slide 7, tumor directed oncolytic immunotherapy is the use of viruses that, when injected into tumors, partially or completely destroy them through virus replication, [ bursting ] tumor cells open and exposing all of the released cancer antigens to the immune system in an environment of necrotic cell death. This leads to the activation of a patient-specific systemic immune response and the disruption of uninjected deposits. On Slide 8, our platform is based on the herpes simplex virus, which we believe is an optimal viral species for oncolytic use, it being high lytic inflammatory and having the ability to carry multiple immune-stimulating proteins into the tumor microenvironment to further amplify the immune response generated. Our strain of HSV has been deliberately selected for its lytic properties in human tumor cells. And from all our products as our base platform, we express a fusogenic protein that greatly increases direct tumor killing, images to cell death and systemic immune activation. These design features ensure maximum presentation of antigen on the MHC of antigen-presenting cells. We then expressed various proteins from the virus, intended to maximize post-imagery signals at the APC T cell interface to ensure optimal T cell priming. Our lead product RP1 additionally expresses GM-CSF; RP2 additionally expresses an anti-CTLA-4 antibody to start a negative feedback loop for the CD20 HC [ CTLA-4 ] access; and RP3, 2 further immune co-symmetry pathway activating proteins, targeting CD40 and 4-1BB, which also leads to downstream inflammatory cytokine release. Our confidence in our ability to establish a broad skin cancer franchise with RP1 is underpinned by broad activity across a variety of skin cancers as monotherapy and in combination with checkpoint drugs, including in the checkpoint refractory setting, where clear systemic long-term responses have been generated. RP2 has generated high rates of response in difficult-to-treat tumors as monotherapy and in combination with anti-PD-1 in checkpoint failed patients. Our intended program for RP2/3 is targeted patient groups that are relatively immune insensitive and will be detailed later in the presentation. As such, moving to Slide 9, we believe we have the ability to drive growth across all segments of an IO portfolio. To improve IO effectiveness in the anti-PD-1 naive setting, we have generated a very high rate of complete response. We've always said from the get-go that our modality and products can result in high rates of complete response. It is these responses that transform lives, give patients the potential to cure and relieve them of disabilities and/or dispigments caused by disease. While this further supports our expectation for a positive outcome in our registration directed studies, including our lead CERPASS study in cutaneous squamous cell carcinoma, we also believe that if our well-tolerated products are pushed earlier into disease courses, the rate of complete response could go higher still and result in many patients never developing any type of end-stage disease we are currently treating. In terms of overcoming and IO resistance, as mentioned previously, we've seen activity in very hard-to-treat patients who have failed anti-PD-1 in a number of different tumor types, not least melanoma, where we've seen a high rate of durable response and where enrollment is ongoing in the study with registration intent. Finally, in terms of turning cold tumors hot, we've seen activity in immune insensitive tumors with RP2, including as monotherapy in uveal melanoma and salivary gland cancer. This data, combined with encouraging reproducible activity in patients with liver metastases with very poor prognosis, gives us confidence in moving beyond skin cancer to establish a franchise with RP2/3 in GI cancers and head and neck cancer, where underlying response rates to IO approaches alone are generally low. We are planning additional studies in each of these categories, which will be further detailed in the presentation. I will now hand on to Dr. Harrington to summarize today's data sets with RP1, including improved data in our completed IGNYTE cohorts in anti-PD-1 naive non-melanoma skin cancer and anti-PD-1-failed melanoma. In addition, Dr. Harrington will present new data in hard-to-treat patients, supporting our broader RP1 skin cancer ambitions, with first looks at anti-PD-1-failed non-melanoma skin cancer data and RP1 monotherapy data in transplant patients with skin cancers. Kevin?

Philip, many thanks, indeed. And if I could just maybe skip past the biography slide and go straight to Slide #12, please. Well, it gives me great pleasure to join you. So I'm going to provide you with an update of data initially from the IGNYTE clinical trial, on which I'm privileged to be one of the investigators, and that study looks at patients with melanoma, non-melanoma skin cancers together with initial data with RP1 monotherapy in patients with solid organ transplants with non-melanoma skin cancer. So first, I provide you with the updated aggregate safety data from RP1 combined with nivolumab. Now please excuse the fact this is rather a busy slide. You can see from the summary in looking at the Grade 3 and the Grade 4/5 toxicities that this treatment is extremely well tolerated and the side effects are generally a self-limiting, constitutional flu-like illness, typical of a virus exposure that lasts between 1 and 3 days, with Grade 3 events being seen relatively rarely. And it highlights the potential of the tolerability of this agent to allow combination regimens. Next slide, please. So first of all, I'm going to take you through data relating to RP1 in melanoma and give you some updates on the data that you've previously seen perhaps in earlier presentations. So here, I provide you with an updated slide on Slide 14, looking at the melanoma response data. You can see here across the top of the table the labels indicating the various scenarios in which patients are being treated. And you can see the program covers cutaneous, mucosal and uveal melanoma and also covers patients with PD-1 naive and also PD-1-failed tumors. What you can see here that the data at the last cut in June 2021 were already quite mature. So any changes have been relatively modest in terms of the scale. But what I can reassure you of, and I'm pleased to be able to announce, that we've seen that there have been a further uplift in the documented overall response rate, that has increased now in PD-1-failed cutaneous melanoma to 37.5% from the initial 31.3% that we've seen. And in mucosal melanoma, we've seen a PR, a partial remission, convert into a complete remission of disease. On the next slide, which is Slide 15, I show you data from the spider or the spaghetti plot. And you can see here that the various categories of patients are picked out in color coding. What I'd like to draw your attention to, first of all, is the depth and the durability of response, which continues to be excellent. And certainly for patients with anti-PD-1-failed cutaneous melanoma, indicated by the red lines, you can see that there are very durable responses and durable treatment benefit. In fact, the outcomes appear to be rather similar, depending on whether or not the patient has had or has had not received a prior PD-1 therapy. I draw your attention also to the highlighted patient, that blue arrow in the middle of the slide. This is a patient who had achieved a durable period of minor response or stable disease who gradually began to progress on single-agent PD-1 inhibition and was reinitiated on RP1 and demonstrated a further response. So this is really a very impressive and intriguing result. Next slide, please. So here you see an example of a patient treated, having had prior exposure to anti-PD-1 and anti-CTLA-4. This is melanoma disease in a patient who has had progressive disease to those treatments. You can see here a series of scans for baseline, indicating the presence of lesions in many sites, both in the liver, the spleen and also in the lung. The injected lesion is highlighted in red in the second column. And you can see over time the evolution of a response in the injected lesion. But more compelling, in the yellow circles, you can see in liver, in spleen and in lung, in uninjected lesions, you can see a patient here who is developing an excellent response in all of these sites. The patient remains as an ongoing partial remission with no metabolic evidence of disease by PET scan, now out to over 1.5 years, at 19 months, in fact. Next slide, please. So in this relatively brief update, I have been able to show you that in patients with both naive and PD-1-failed melanoma, we've seen incremental improvements in the data sets since our last data cut with additional responses observed, further deepening of responses and impressive increasing durability of response. The response rate in the anti-PD-1-failed cutaneous melanoma has now gone above 1 in 3, further highlighting the potential for this therapeutic opportunity for patients. So on Slide 18, I highlight to you the plans for registration directed study. So the treatment of anti-PD-1-failed melanoma remains a considerable problem in the clinic, as you know. And as many as half of patients initially treated with either PD-1 monotherapy or CTLA-4 PD-1 combination therapy will fail to respond to initial treatment or will ultimately progress on this treatment. And the response rate to further immune challenge with anti-PD-1 therapies in this group of patients is only around 6% to 7%, and there are as yet no approved therapies. Therefore, we believe that the promising data with RP1 combined with nivolumab in these patients represents an important opportunity for patients, and Replimune has an ongoing registration directed clinical trial in 125 patients with anti-PD-1-failed cutaneous melanoma. And we're going to see an interim output from those data towards the end of 2022 and are optimistic that these will be positive data based upon the data I've already shared with you. So next slide, Slide 18, please, takes us on to the next topic for conversation, which is another important opportunity for RP1 non-melanoma skin cancers. So next slide, Slide 20, please. Again, in this rather busy slide, I show you data looking at anti-PD-1 naive non-melanoma skin cancer, looking at the responses. And you can see in the red columns, that is the recent data, the updated data we're presenting to you. And in the other columns, you can see the data from June of 2021. So you can see that now the cohort is fully enrolled as compared to when we last showed you these data. And as with the melanoma data I've already shown you, we've seen incremental improvement in the response rates. Cutaneous squamous cell carcinoma is particularly of interest, bearing in mind the ongoing registrational CERPASS study. And combining RP1 with cemiplimab versus cemiplimab alone, the overall response rate in this group of patients has increased from 60% to 64.7%. And I want you to just pay attention to the very high rate of complete remissions, which we have seen is maintained at 47%. The activity continues to be seen also in other non-melanoma skin cancers, basal cell carcinoma, Merkel cell and angiosarcoma, with activity being particularly apparent in the latter 2, Merkel cell and angiosarcoma, both tumors have unmet need. And this again highlights the potential broad utility of RP1 across a range of different tumor types. So in the next slide, I will show you an example of a response to an extremely unpleasant cutaneous skin cancer. This is, in fact, the last patient enrolled in the RP1 plus nivolumab cohort. And you could see here, the patient presenting with a deeply unpleasant, deeply ulcerating lesion in the region of the left temple. And over the course of 5 months, you can see that this is evolving towards an excellent partial remission with every expectation indeed that this patient may achieve a complete response. And Slide 22 gives you the summary of the waterfall plot data for these patients, and you can see the depth of response. Notice this isn't just in cutaneous squamous carcinomas, but you're seeing excellent responses to angiosarcomas and also Merkel cell carcinomas. And the black circles indicate patients ongoing, and you can see many of the patients with complete remission remain ongoing, deriving further benefit. Slide 23 demonstrates the spider or the spaghetti plot, again, really giving very impressive indications of the durability and the depth of the response here. I also want to draw your attention to the patient in the middle of the slide. You can see on the spider plot the growth of the tumor that's occurring out to 600 days, and then the patient being reinitiated after a progressive disease episode, demonstrating a further response of treatment. This is a cutaneous carcinoma -- squamous carcinoma. So again, excellent responses. Now on Slide 24, just to conclude this part of the talk, I hope I've been able to share with you that as for melanoma, there have been incremental improvements in our data since the last data cut, with additional responses observed, deepening of those responses over time and the durability continues to be impressive and demonstrable. In patients with cutaneous squamous cell carcinomas, the overall response rate has increased to now nearly 2 in 3 patients; and the complete remission rate, almost 1 in 2. These are remarkable results. We're seeing improved activity in other tumor types, and I've shown you the data from Merkel cell and angiosarcoma, which may be particularly compelling. And again, this highlights potential for this agent across a range of different cutaneous cancer opportunities. So on Slide 25, I remind you of the design of the randomized Phase II study of the CERPASS study. This is a randomized controlled clinical trial of RP1 combined with cemiplimab versus cemiplimab alone. This has dual independent endpoints of overall response rate and complete remission rate. It is actively recruiting in the United States and Europe and in Australia, and it is expected that it will complete enrollment around the middle of the year with top line data being available, we hope, at the beginning of 2023. From the results that we have seen thus far with nivolumab, we would be optimistic that there would be a positive outcome for this study, both in terms of the overall response rate endpoint and the complete remission rate endpoint and also in terms of key secondary endpoints, including durability response. So many chances for this study to win. Next slide is #26. So we now turn to new data with RP1 combined with nivolumab in patients with non-melanoma skin cancer who have progressed while on prior anti-PD-1 therapy. These numbers of patients are rather smaller. As you can see for the anti-PD-1-failed melanoma, this is a difficult disease to treat. There are no approved therapies that are known to be effective for these patients. And while this is a relatively early data set and the numbers of patients are small, the initial data cut that I share with you looks very promising. You can see here that there have been responses in anti-PD-1-failed cutaneous squamous carcinomas, Merkel cell cancers and angiosarcomas, with all of the responding patients in a maintained response to date, which is impressive. The overall response rate for this group of patients as a whole is currently 1 in 3, 33%. And a number of other patients who have not been on study for long are still in stable disease, and we hope, indeed, anticipate that these may evolve into further responders. So as for the melanoma scenario, it seems that RP1 combined with nivolumab may be able to generate clinically meaningful responses in patients with prior anti-PD-1 exposure and failure of their disease, and it may represent an important new treatment opportunity for these patients. I'm going to show you some examples of these different tumor types in the next slide. So on Slide 27, I show you a patient who has had prior anti-PD-1 therapy for a cutaneous squamous cell carcinoma. You can see this unfortunate patient has a deeply unpleasant recurrent lesion in the preauricular area with involvement of the parotid bed. You can see that within 6 months of treatment, there has been a remarkable response of disease. You can also see in the biopsies taken at screening and at week 6 on treatment, there is a substantial increase in the number of CD8+ T cells infiltrating into the tumor. Indeed, the paucity of T cells at the outset of treatment may well have been one of the reasons underlying the failure of PD-1 therapy in this patient initially, which has been turned around by co-treatment, in this case, with RP1. So on Slide 28, I show you a patient who has had anti-PD-1 therapy for a Merkel cell carcinoma who has developed recurrent or progressive disease. And again, you can see here, again, with the same color coding, red circles indicating injected lesions, yellow circles indicating uninjected lesions. And you can see that this patient with multiple deposits on the forearm has had a dramatic response of treatment within 3 months in the setting of prior anti-PD-1 failure. And finally, we go for another example in Slide 29, a patient with an anti-PD-1 previously treated angiosarcoma of the scalp, a typical patient with this disease, a rather elderly patient with disease on the scalp. And you can see 2 baseline photos taken from slightly different angles, indicating the nodularity and the widely disseminated recurrence in the head and neck region. And then you can see within 4 months, this patient has had a superb response. So as for cutaneous melanoma, the treatment effects seen with PD-1 combined with -- RP1 combined with PD-1 therapy appears to be impressive, and it doesn't appear to be affected by prior treatment status of the patient, meaning we can see responses in those who had prior anti-PD-1 therapy. So on the next slide, I'm going to briefly introduce the program ARTACUS, which is looking at the difficult problem of treating patients with solid organ transplants who develop skin cancers. So this is a difficult scenario that has been addressed in a Phase I study with RP1 where it's been used as a monotherapy. Due to the immune-suppressed nature of patients who have received organ transplantation, they are particularly at risk of developing cutaneous malignancies, especially cutaneous squamous cell carcinomas. And these occur in up to 70% of patients by 20 years after their transplantation. And while these tumors may well respond well to anti-PD-1 therapy, it raises the risk of rejection of the transplanted organ. And this is clearly not something that's broadly acceptable to the patient or indeed their treating physicians. So under these circumstances, there's an opportunity to test monotherapy to see whether or not that can deliver responses. Here, I show you data from the first 6 patients. And again, as for the anti-PD-1-failed non-melanoma skin cancer data I just presented, these are rather early data. In the first 6 patients, we have achieved 2 responses, 1 partial remission and 1 complete remission so far. Regrettably, 1 other patient who was already responding to treatment contracted COVID and succumbed to that disease dying of COVID and before we could assess the true benefit of the Replimune agent in that patient. This high risk of COVID, of course, in these patients is another example of why we have to be very cautious in how we treat these patients. We also know that COVID has significantly impacted the recruitment into the trial. And this has led to some slowing down of the initial recruitment, but I'm pleased to say that this is now picking up as the pandemic is waning. 1/3 of the patients, 2 out of 6, have had a response, and these data should be regarded as promising but early in nature. All of the patients presented here had cutaneous squamous cell carcinomas, and all of them were kidney transplant recipients. So on Slide 32, we look at the safety of this treatment. Clearly, when treating patients who are immune suppressed having had an organ transplantation, we might be worried about the administration of a replication-competent virus. What you can see here is that there is no clear evidence of a different safety profile of RP1 compared with the population of patients who are not immune suppressed. And there has been no evidence reassuringly of organ rejection. Indeed, we wouldn't expect that. Slide 33 just allows me to summarize this. So initial data in both anti-PD-1-failed non-melanoma skin cancer treated with RP1 combined with nivolumab and RP1 monotherapy in solid organ transplant recipients with cutaneous squamous cell carcinoma, again, shows clinical activity with responses observed in 1/3 of patients. This again further highlights the potential for RP1 as a treatment for patients with anti-PD-1-failed and other difficult-to-treat tumors and also highlights the potential for monotherapy activity. So with that, I have the pleasure of handing back to Rob Coffin to continue with the presentation. Rob?

Thank you very much, Kevin. So I'll now just briefly summarize our overall program in skin cancers with RP1 on the next slide, which is intended overall to establish a major skin cancer franchise, as Philip outlined it and as Sushil will further discuss later. So from the data Kevin just presented, we believe that the data in anti-PD-1 naive cutaneous squamous cell carcinoma and anti-PD-1 naive melanoma continue to support our registration directed activities, represented on the top 2 boxes on this slide. For CERPASS, we expect this study to complete enrollment around midyear with the trigger for the primary analysis 6 months later. Final data cleaning and central review of response data will mean the top line results should become available during Q1 next year. The anti-PD-1-failed melanoma interim data is expected at the end of the year, with top line primary analysis data towards the end of next year. In addition to anti-PD-1 naive cutaneous squamous cell carcinoma and anti-PD-1-failed melanoma, we've also seen activity in basal cell carcinoma, Merkel cell carcinoma and angiosarcoma, including with anti-PD-1-failed disease, as Kevin just presented, within each area, the likely approach commercialization being through compendia listing. Similarly, we've also now demonstrated RP1 monotherapy activity in solid organ transplant recipients with CSCC where we also plan for potential future registration or compendia listing. Finally, we're now planning for a study in the neoadjuvant setting in cutaneous squamous cell carcinoma, which is a very substantial potential market, which likewise will be discussed a bit more later. Thus, overall, we have quite a broad range of activities ongoing in skin cancers, which captures nearly all of the potential segments in these tumor types and which we believe can deliver substantial value to Replimune in the medium to longer term. So I'll now hand over to Dr. Nikhil Khushalani, who is the Vice Chair in the Department of Cutaneous Oncology at the Moffitt Cancer Center in Florida and also an investigator on the CERPASS clinical trial as well as being an investigator on many prior studies in cutaneous squamous cell carcinoma and related areas, including the study which led to the approval of cemiplimab in the first place. So Dr. Khushalani will now provide an overview of CSCC, including disease background, current treatment options and unmet need. So over to Dr. Khushalani.

Thank you very much, Rob. We can skip through the bio here. I appreciate the invitation and the broad strokes that were provided previously by Dr. Harrington, set the stage very well for my portion of this discussion. So what I've titled this is Going Beyond the Classic Management of Cutaneous Squamous Cell Carcinoma. As many of you would recognize, this Janus, the 2-faced Roman god. And looking back, looking forward, essentially, this is a disease that traditionally has not seen much traction until recently in the last, I would say, 3 to 5 years, where immune checkpoint inhibition has really transformed the management of these patients, particularly those with locally advanced unresectable disease or those with metastatic disease. And there is a lot of biologic rationale for the efficacy of these agents in this disease, primarily related to a high tumor mutational burden. So it is very obvious that this would be one pathway that one would follow. But the key issue then becomes this immune checkpoint does not work for everyone and how can we actually better improve on the existing standards of care. Next slide, please. These are my disclosures. Next slide, please. So essentially, from an epidemiology standpoint, skin cancer, keratinocyte carcinomas, which is the bucket term that we use typically for a combination of basal cell carcinoma and squamous cell carcinoma, are the most common cancers that are seen. And in the United States, the 2 of them together equate more than the sum of all other cancer incidences in the U.S. If you specifically parse this down to CSCC or cutaneous squamous cell carcinoma, it is the second most common skin cancer that we see with roughly approximately 0.75 million patients on an annual basis. It is hard to accurately estimate the true incidence of this disease. Because it is so common, it often, in many countries, including the United States, is not reported to tumor registries. There have been now concerted efforts in several European countries to really get a better sense and a better handle as to how many of these new patients or new tumors we see because many patients have a multitude of tumors over their lifetime. The primary etiologic factors, exposure to ultraviolet radiation, both ultraviolet light A and B, and particularly accumulative exposure over time. What we have known is there has been a dramatic increase in the incidence of CSCC over the last 3 decades, an almost 250% or greater increase. And many of these patients have either locally recurrent disease that is evident on the skin or have evidence of nodal involvement, so regional nodal metastases in the absence of distant systemic metastases, which is where the neoadjuvant therapy, and I'll discuss this a little bit later in my presentation, really offers us an opportunity not only to intervene with systemic and intralesional therapy ahead of time prior to definitive surgical resection or prior to radiotherapy, but actually, from a scientific perspective, also allows us to study the biology of these diseases by obtaining serial biopsies as we go along. Similar to the incidence, it's a little hard to truly estimate the mortality related to CSCC, but most of us believe that it is somewhere in the 7,000 to 15,000 deaths on an annual basis, and this again varies from publication to publication. I think it's important to recognize that many of these patients unfortunately succumb to local regional progression, so locally recurrent disease or uncontrolled locally advanced disease rather than distant metastases. Another important etiologic factor is underlying immune deficiency in immunosuppression, where about 15% to 30% of patients fall within this category, particularly with solid organ transplant or other immunosuppressive illnesses, including autoimmune disorders as well as hematologic malignancies, such as chronic lymphocytic leukemia or HIV infection, and this predisposes patients to developing CSCC. Next slide, please. So the typical patient presentation, these tend to be older individuals. The median age is somewhere between 70 to 75 years. And that's important to recognize because these individuals then have associated comorbid factors, particularly from a cardiac as well as a renal standpoint. Previously, we used to utilize chemotherapy, typically platinum-based for these patients. But with renal insufficiency, many of them are not candidates for cisplatin or similar type of drugs. So newer treatment modalities are critically important. The vast majority of these develop from what we call the classic sun-damaged skin or actinic keratosis. These are individuals who go to their dermatologist, often will have cryotherapy to a multitude of lesions that are pretty cancerous, and many of them eventually will end up getting biopsy. What you see on the right-hand side is a micrograph of an individual with locally advanced recurrent disease on the scalp and clearly disfiguring, clearly a problem. Many of these patients have foul-smelling discharge from their tumors, painful due to evidence of perineural involvement. And then over a period of time, and this has not been very well studied, unfortunately, from a psychosocial aspect, there tends to be a delay in seeking care. And part of it may directly be associated with the social stigma when these diseases are so visible on the head and neck area where it's most common. And given the local regional progression that occurs in many patients with high-risk disease with superficial involvement or visible involvement, intralesional intratumoral approaches offer an excellent avenue to try and help with disease control. Next slide, please. So really, this slide summarizes what the unmet needs are. We know that anti-PD-1 therapy works very well for these patients. It is an effective option, and there are 2 approved drugs in the United States, cemiplimab and pembrolizumab. The key question is, can we actually improve on the overall and complete response rates with anti-PD-1-based combination therapy? So using anti-PD-1 therapy as the backbone. Many patients, unfortunately, will still not respond to anti-PD-1-based therapy. Is there an effective second-line treatment post anti-PD-1, so in the PD-1 refractory patients? Dr. Harrington presented some early data on that population, both in melanoma as well as in CSCC. The specific subgroup of individuals who are high risk but who unfortunately cannot get immune checkpoint therapy upfront, is that an area that we can clearly develop additional treatment options? Again, a high area of unmet need, and I'll show you some of that transplant data as well. And certainly, try to move the needle earlier in the case of disease course. So to try to mimic advanced disease success in earlier-stage disease, it offers us the opportunity to consider this for neoadjuvant therapy where there is still visible disease, but potentially resectable and try to obtain long-term durable disease control in that setting. Next slide, please. So this is a schematic of the original EMPOWER study that led to the approval of cemiplimab in this otherwise poorly studied disease or I would have called it an orphan disease 5 years ago, not so much anymore because there's certainly a lot more attention being paid now to this particular group of diseases in keratinocyte carcinoma. But just to recap, this was a Phase I, Phase II trial. And what this slide primarily highlights is the adult patients with locally advanced or metastatic disease who are not candidates for surgery. And in the Phase II portion that was presented and published in The New England Journal of Medicine back in 2018, and I was privileged to be part of the team that participated in this trial as an investigator and saw some of the dramatic responses. So at that time, these individuals got weight-based dosing, 3 milligram per kilogram of intravenous cemiplimab every 3 -- every 2 weeks and then up to 96 weeks in the Phase II setting. And again, the important highlight here is at the bottom in terms of exclusion, patients with autoimmune disorders that were treated with systemic immunosuppressive therapy, patients with solid organ transplant recipients were essentially excluded from this trial. Next slide, please. And similarly, we had a locally advanced cohort of 78 patients, and these were patients who had locally recurrent, locally advanced disease and were not candidates for additional curative-intent surgery and/or radiotherapy. And between the 2 arms or 3 arms of the study in the metastatic setting, the response rates were between 45% to 50%, 47% to 50% specifically between arms 1 and 3. And in the locally advanced setting, which is arm 2 of that trial, the overall response rate has measured through clinical observation as well as RECIST traditional criteria was 44%. The typical toxicity of cemiplimab was that with any other anti-PD-1 drugs, so really not different from that standpoint. And what you can see here on the top left is the waterfall plot of the responses and then the bottom was the swimmers plot demonstrating durability of response as well. Next slide, please. This slide highlights the progressive increase with subsequent longer-term follow-up on these patients where we have seen across all 3 arms of this study, Group 1 and Group II are the metastatic patients, in the middle group 2 are the locally advanced patients, where with the metastatic patients we have seen a progressive improvement in the complete response rate, albeit small numbers, but now approximating anywhere between 18% to 20% complete response rate. I think that's important because that is one of the endpoints on the CERPASS trial as well in terms of trying to assess what is the complete response rate to combination therapy. But more importantly, from a practical standpoint, does it even matter? If one looks at data in melanoma, it is very clear that patients who have deeper responses tend to have more durable responses. I think we're still scratching the surface in terms of understanding this phenomenon in cutaneous squamous cell carcinoma. But as a treating oncologist, to me it is far more gratifying to see a complete response that occurs in these patients. I then tend to be far more altruistic in aiming to stop therapy in those patients who achieve a complete response and have already completed the designated time frame of therapy. Next slide, please. This was a similar trial that was done with pembrolizumab that led to the approval of this agent in the similar type of population. What's highlighted here is the objective response rate to pembrolizumab in 105 patients, with locally recurrent or advanced disease was 35%. Now one may ask, why is there a difference between pembrolizumab and cemiplimab? Quite frankly, I don't believe that there is any difference. It is important to note that patients on the KEYNOTE-629 trial were far more heavily pretreated compared to those on the EMPOWER study. Almost 87% of patients had some form of systemic therapy and/or radiotherapy, which may have blunted the second-line response rate to pembrolizumab. If one specifically looks at the subpopulation of patients who did not have any prior systemic therapy, the response rate is approximately 50%. So very similar to cemiplimab. So I think utilizing one or the other is very reasonable. Similarly, another smaller trial called the CAR Skin study showed very similar efficacy for pemromizumab in this disease. Next slide, please. So the question that really then comes up is PD-1 anti-PD-1 therapy works for a subset of patients and it can induce durable responses. What happens in those patients that either have primary resistance to anti-PD-1 therapy or who develop secondary or acquired resistance, i.e., in the course of their treatment, eventually stop responding? What do we have for them post anti-PD-1 treatment? And I think this is a very broad arena and certainly ripe for further investigation. Currently, there is no regulatory approval for any specific agent in this particular setting that have failed anti-PD-1 therapy. We, in the clinic tend to use anti-EGFR agents such as cetuximab. We use chemotherapy, which is typically platinum-based or taxane-based. In some cases, we also use 5-fluorouracil based regimens but all of these tend to have finite responses and certainly are replete with toxicity insofar as chemotherapy goes. So that can limit the use, particularly in an aging population. So new options are clearly needed for these patients, preferably those with better tolerance, lesser toxicity and preferably, obviously, those with a higher overall response rate, complete response rates and potentially durable benefit. So again, a lot of room for investigation in this area of high unmet need. Next slide, please. What about immunosuppression in skin cancer? I alluded to this earlier in the talk. If you look at patients who have undergone solid organ transplant, the incidence of cutaneous squamous cell carcinoma is almost a hundred-fold increase compared to those who have not undergone a solid organ transplant. And this is very well documented data in a cohort study across 26 transplant centers around the country, coordinated through the University of California in San Francisco. And when you look at the skin being the most common site of malignancy post solid organ transplant, with an incidence rate of approximately 1,400 per 100,000 patient years, CSCC accounts for more than half of that. Putting it in a different perspective, when you look at an immunocompetent population, the incidence of basal cell carcinoma is almost 4 is to 1 compared to cutaneous squamous cell carcinoma. In the transplant population, that's completely reversed. So CSCC is a high area of unmet need within the transplant population as are other skin cancers as well. Next slide, please. What is also known is that in these patients who are immunosuppressed to primarily preserve their allograft, the overall prognosis is noticeably worse compared to those who are immunocompetent. And this is admittedly a small study that was published a few years ago, where they looked at 23 immunosuppressed patients versus 46 immunocompromised patients, organ transplant or hematologic malignancy, and what they clearly found was that those individuals who are immunosuppressed did much worse compared to those who are immunocompetent. So this highlights the fact that newer treatments are clearly the need of the hour for this population. I think in terms of numbers, it is also important to recognize that we are doing many more transplants in the United States compared to 2 or 3 decades ago, almost to the extent of threefold. For example, in 2015, there were approximately 31,000 solid organ transplants performed, and that would include a plethora of kidney transplant, liver transplant, lung, heart, pancreas transplant and so on and others. And this was about threefold higher compared to the transplant registry data from 1988. Next slide, please. So clearly, in these patients, the use of immunotherapy, specifically anti-PD-1 or anti-CTLA-4, is very limited. These were patients who were not included in the original clinical trials primarily because of the concern for organ rejection or transplant rejection. And case reports and smaller case series have shown an approximately 50% risk of allograft rejection, which is extraordinarily high. So we tend to back into consideration of these drugs for this particular group of patients only after they have exhausted other treatment options. What about other immunocompromised statuses, for example, patients with autoimmune disorders, patients with concurrent hematologic malignancies? Again, these were individuals who were excluded from the original registration trials. We in the clinic certainly use anti-PD-1 therapy on a case-by-case basis. But again, this is an area of unmet need where we can consider newer treatment options, including intralesional approaches for them. Next slide, please, neoadjuvant therapy. This is a fourth arena where one could certainly consider intralesional along with systemic therapy. I just wanted to clarify what is the true terminology that I utilize, and obviously many others? The neoadjuvant therapy, specifically the administration of systemic therapy prior to surgery in patients who have accessible, regionally advanced or metastatic disease, most importantly, with a plan to carry out resection after a finite period of time on systemic or intralesional therapy. You can assess the response rate radiographically or traditional means. And then postoperatively, i.e., adjuvant therapy may or may not be used. But again, the intent is these are surgically operable patients in whom the typical paradigm originally may have been surgery followed by consideration of adjuvant therapy. But in the arena of immunotherapy with evident tumor and potentially tumor-associated antigens in sight, neoadjuvant therapy certainly offers a lot of biologic rationale to pursue. Next slide, please. Again, this is palpable notes, which we see in our patients with CSCC. Again, it's hard to truly estimate the total number of patients who present in this manner. But looking at a multitude of studies, pioneered right here in Boston through Dana-Farber and Mass General, approximately 3.7% of patients with cutaneous squamous cell carcinoma will end up developing nodal metastases. If you look at the entire spectrum of over 700,000 patients presenting with this disease, that would equate somewhere between 20,000 to 25,000 patients with nodal metastases on an annual basis. Similarly, the disease-specific death related to cutaneous squamous cell carcinoma from prior cohort studies is approximately 1.7% to 2%. And that's how we tend to estimate some of the mortality rates associated with this disease as well. Again, neoadjuvant therapy would be very appropriate for patients with palpable nodes or radiographically evident nodes. Those with in-transit metastases, so again, cutaneous or subcutaneous metastases that are visible and palpable on the skin. One could biopsy these tumors, image these tumors, proceed with systemic therapy, either systemic or intralesional. We could rebiopsy them, if necessary, to understand the biology of the disease. That I think is critical from a scientific aspect to understand the immune microenvironment and [ Milio ] and then eventually take these patients to the operating room, assess the pathologic response. What does the pathologist see under the microscope? Is there any evidence of disease or not? And then based on that, even adapt adjuvant strategies to this. Next slide, please. Our colleagues at M.D. Anderson published a small cohort of patients of 20 in Clinical Cancer Research last year. And essentially, what they did was treated patients with neoadjuvant, cemiplimab, 2 doses of therapy prior to planned surgery. In the vast majority of these patients, in 14 of these patients, 70% of them who underwent surgery, the pathologic response was either a complete pathologic response or a major pathologic response that was identified, 11 out of those 14 had essentially a pathologic complete response. Why is that important? Again, it's a little too early to tell whether in CSCC a pathologic complete response equates long-term disease control. In melanoma, we have already identified that pathologic complete response to immunotherapy, or MAP kinase targeted therapy, portends long-term disease control in a large number of patients. I believe, personally, a similar scenario will play out over time in CSCC as well. I think this is also important because we can then potentially limit the amount of treatment that is administered to patients prior to surgery and then risk-stratify or risk-adapt postoperative therapy based on the response that we actually see. So we may be able to change paradigms and clearly benefit patients with very specific tumor-directed therapy in this particular setting. A larger trial of neoadjuvant cemiplimab in CSCC has already been completed, results will be anticipated sometime in this calendar year. Next slide, please. This slide essentially highlights what other avenues exist and where the field is going. And it's a busy slide. I won't belabor the point, but essentially, it shows you that this is almost like melanoma deja vu, is given the propensity for CSCC to experience immune manipulation we can utilize multiple other agents that are targeting the immune microenvironment to try and see is how can we build on our existing blocks that we utilized for the management of this disease. So a lot of room for improvement and certainly a lot of room for additional clinical and scientific investigation. Next slide, please. So in summary, what I hope to have been able to show you in the last 20 minutes or so is cutaneous squamous cell carcinoma clearly has a rising incidence. It is an increasing public health problem and certainly greater urgency, both in the treatment of patients that I tend to see which would have advanced disease, but I think a lot of avenues for consideration of prevention strategies as well. While we do know that anti-PD-1 therapy as standard of care for the appropriate patient with advanced CSCC, almost 2/3 of these patients eventually still progress or may not benefit at all from this avenue of therapy. Oncolytic immunotherapy clearly provides a promising option to help a broader spectrum of these patients, and you saw a lot of objective data on that previously. And certainly, our understanding at a molecular standpoint will eventually hopefully lead to therapeutic advances, setting the stage into combination therapy, adjuvant therapy and neoadjuvant therapy. I thank you for your time and attention.

Thank you, Nikhil, for that very nice overview of cutaneous squamous cell carcinoma. So now I'm going to just spend a couple of slides talking about our initial launch of RP1 in skin cancers. Next slide, please. So we believe that if we can replicate some of the transformative data that we've seen from some of our earlier studies in our pivotal data sets, this will allow us to nicely progress towards many aspects of our vision that you see outlined on this slide here. Firstly, it's important, it's a really critical 12 months for us on the commercial side as we prepare for our initial launches. And importantly, our first goal is to rapidly establish RP1 in cutaneous squamous cell carcinoma. And I'll explain in the following slides why we believe this is an ideal first launch for our modality. We plan to have a second launch in checkpoint-failed melanoma shortly after our cutaneous squamous cell launch, which will allow us to continue to build the momentum, customer experience and enable us to develop a meaningful skin franchise. We also, as you have seen earlier, a compelling data in other skin cancers, basal cell, Merkel cell angiosarcoma, et cetera, which I think provides an opportunity to further expand our skin franchise. Next slide, please. So we are really excited about advanced cutaneous squamous cell carcinoma as our first launch, as we believe our product and the mode of administration really makes a lot of sense in this tumor type. Given the unmet need in CSCC that you heard from Dr. Khushalani, we believe we can provide a potential option, either alone or in combination, for all key segments of the market, including providing an important option for those who have failed checkpoint inhibitors. Ultimately, if we can produce strong CRs in CSCC, RP1 could be a powerful option for many more patients in the neoadjuvant setting and really help redefine the treatment paradigm and achieve our goal of cure. In totality, these various segments represent an estimated U.S. treated population of around 40,000 patients. Next slide, please. So one of the questions we often get is how often -- how are you different to T-VEC and why do you believe you're going to be successful in your launch? And I just want to say a couple of comments here. Firstly, I think T-VEC was great proof of concept, but it didn't launch with strong data or a label. And we aim to come to market with a strong data set and a compelling value proposition. The whole reason for being for Replimune was really to build a more potent oncolytic immunotherapy platform, and that really starts with RP1, including developing [indiscernible] strain as well as additional modifications, really aimed to drive a stronger systemic immune response. Let's also remember -- it's also important to remember that RP1 can be used in combination with standard of care. And so we're not really asking customers to replace their I-O, which we know is a bigger burden. So when we think about our critical success factors, there's really 2 areas we need to focus on. The first being establishing really strong HCP confidence. Beyond the obvious awareness and education of our key data, we need to make sure that there's trained injection champions at every key account, and that includes nurses, nurse practitioners or MDs. Secondly, what's going to be really important is to have a positive experience, making it as easy as possible for the whole care team by simplifying dosing, logistics and cleanup. One of the things that we've seen that's different to T-VEC is that we have improved stability at 2°C to 8°C, which is a normal refrigeration; versus T-VEC, which was stored at -70°C. Also, I would say, given the superficial nature of the tumors that Dr. Khushalani mentioned, our initial launch is really going to allow health care professionals and patients to actually see the therapy is doing to them. And so that's a powerful modality and a powerful option for us to bring this to market. So next slide, please. So as we think about the next stage of this presentation, we're going to focus on -- we're going to move away from RP1 and now focus on RP2 and 3. And this transitions into some of the clinical -- before we get into some of the clinical development discussions, we're going to discuss some of the considerations of deeper injections, which we believe will be key to enabling our use of our assets in more solid tumors. So next slide, please. Now one of the important aspects of our modality will be making intratumoral injections, and as we like to say, tumor-directed treatments easy and routine. We think about this in 3 steps. As mentioned, we start with CSCC, where it's largely an outward growing cancer, where patients die a local regional progression with a high percentage of patients who have lesions that can be injected. However, we do believe that to drive even stronger immune and systemic immune activity with RPx, we will need to inject deeper locations such as lymph nodes or the visceral mets, and we are also doing this in a stepwise fashion. And today, you're going to -- we're going to focus on why step 2, that you see on this slide here, is an obvious gateway to getting to step 3. I think importantly, in step 2, this includes indications where delivery is part of routine medical practice and can be easily adopted, e.g., primary liver cancer or liver metastases or use of ultrasound and repeat injections of the liver already occurs. Hence, in these indications, we're really swimming with the tide versus against it. Next slide, please. So firstly, I just want to reiterate that we believe skin cancer alone represents a sizable opportunity, and that's about 50,000 patients if you now include, in addition to the CSCC number I gave you before, checkpoint-failed melanoma patients. In the next wave of our development strategy, one of the important things we've considered was how to best move along this continuum and what has become clear in our market research and talking to key opinion leaders, such as the ones you're hearing from today, is that not only our HCC, CRC and locally advanced head and neck areas of unmet need, but are indications where we can more easily do deep injections. Now if we specifically consider liver cancer and liver mets, there's a strong rationale for doing deep injections in addition to significant commercial opportunity. In HCC, CRC and uveal melanoma that you see on this slide, patients die with liver failure or liver mets, and the liver tumor microenvironment really is an issue for existing I-O treatments. Deep injections of RP2 and 3 has the ability to really alter the tumor microenvironment. Now if we turn to second-line HCC and third-line CRC, these account for about 30,000 treated patients. And again, given the unmet need, if we can show proof-of-concept activity in these later-line settings, it will allow us to rapidly move into much larger opportunities in earlier disease. Ultimately, we believe deeper injections in other solid tumors, especially in the neoadjuvant locally advanced settings, could provide significant patient opportunity and unlock a large additional commercial value as this is really where the whole market is moving to. Next slide, please. On this slide, I want to just orient you to the top of the slide initially. And if we think about intratumor injections, they really are not one size fits all. And so if we're thinking about injecting accessible lesions, this really should not be any harder than doing an IV. And as I mentioned, a nurse or another trained health care professional can very much do this. Now if we think about deeper injections, we've done detailed buying process and market research and certainly there are some barriers to overcome. But this is not complex like cell therapies such as TILs or CAR-Ts. And importantly, these barriers can be able overcome. You're going to hear more about this shortly, but firstly, it's important to note that technically, this is not particularly challenging. And in fact, they are far more complicated intra procedures that interventional radiologists do such as Yttrium-90, et cetera. And a lot of deep injections can actually be done by ultrasound. Now what we've seen from our research is the hurdles really fall into 3 categories as outlined here. And while some of these are going to require some new behaviors and new processes to be adopted, again, nothing came through in the research is insurmountable without appropriate education and training. Ultimately, if the data is compelling, this modality really addresses unmet needs for patients, we believe HCPs will find a way to incorporate RPx into the treatment paradigm. Oncolytic immunotherapy importantly provides an opportunity for radiologists to become part of the treatment of systemic I-O treatments, which is something we're really excited about. Many of the procedures they do such as TACE and TARE are locally -- are largely local treatments, and so this again provides a new opportunity to bring them into the treatment paradigm. Next slide, please. So to wrap up, if we bring together our initial skin launches and worked on feasibility of deep injections, there's a logical and stepwise adoption ladder. It starts with CSCC, and we recognize that the first launch really does set the trajectory for the portfolio and the company, and we believe CSCC is a great starting point as it will showcase our product attributes and provide ACP's hands-on experience with RP1. We then build with more experience in skin, this time in checkpoint-failed melanoma and start to get adoption of some deeper injections, including lymph nodes. Finally, as discussed, given the unmet need and feasibility, we feel liver cancer and liver mets for a strong gateway to broader adoption in other solid tumors for RP2 and 3. And now next slide, please. It's now to my great pleasure and this -- to introduce Dr. Muneeb Ahmed, who's the Division Chair or Chief of Interventional Radiology at Beth Israel and is the President-elect of the Society of Interventional Oncology. Over to you, Muneeb.

Thank you, Sushil, and I'm happy to be participating in this. Next slide, please. So as we have heard, when we think about deep/visceral tumor injections, these are really tumors that cannot be injected via direct visualization or palpation, and often will require imaging guidance in order to be injected and that includes CT or ultrasound guidance, which I'll talk about. When we think about this, it's often solid organs, such as the liver, the lung, kidneys, the spleen, but also includes lymph nodes and deep locations in the retroperitonium, in the pelvis, in the chest, lung, thorax. If we think about targeting some of the deeper injections, liver tumors present a compelling first place to start. There's a rising incidence of liver metastases overall. And these often determine outcome of patients, regardless of the primary cancer of origin with an associated 50% higher risk of mortality. Liver metastases themselves are often very resistant to current immunotherapy options, in part due to the unique immune resistant landscape of the liver. And the direct injection concept for liver tumors has actually been around for decades, so it's not a new concept to implement in clinical practice. And therefore, targeting liver metastases really represent an opportunity to treat multiple cancer types. Next slide. When we think about what this involves from a technical perspective, again, these are often done with ultrasound or CT guidance. And approximately 80% of liver tumors can be injected with ultrasound guidance, meaning that this is an easy-to-do and easy-to-implement technique. The needle is placed in the target lesion. As it -- when it comes to RPx injections, up to 10 mls can be injected and the agent is distributed across the tumor through needle repositioning. And as an example of this, these 2 images demonstrate that. The image on the left is a grayscale ultrasound image. It demonstrates a dark hyperechoic focus of metastases within the center of the liver. And we see a thin echogenic white line representing the needle with the tip positioned in that. On the image on the right, this has been injected with an agent, and we can see that white effect from local injection that covers the entire tumor. Now it's important to remember that this type of technique can be done by any radiologist, either a diagnostic or an interventional radiologist. These are outpatient procedures usually with conscious sedation. Logistics are similar to commonly used FNA or biopsy procedures. So the infrastructure and process to do this is in place at almost all centers and hospitals. And this is a 20-minute procedure, monitoring is required for 2 to 4 hours afterwards. And all of this speaks to the potential wide transferability and scalability of this type of technique at later stages of commercialization for this technology. And the formal injection protocol for RPx has actually been developed with input from a number of KOLs and Interventional [indiscernible]. Next slide. And so this is just an overview of safety profile and adverse events. This was actually presented at the recent Society of Interventional Oncology Annual Meeting. Really looking at early data from RP1 and RP2 on liver metastases injection. And I'll call out that essentially, overall, the side effect profile is very similar to other types of injectable therapies. There is an increased incidence of fever, nausea and chills afterwards, which can be seen in liver injections regardless of the therapy due to the vascularity of the organ, but these are extremely manageable as well. There's also slightly increased injection site pain, again, well within what might be expected for liver injections. Next slide. And this is an example of a patient with liver metastases who was treated with RP2 monotherapy. It's a patient who had uveal melanoma, had extensive liver metastases that are shown here and had been previously treated with ipi and nivo and progressed at 15 months. And just to walk you through these images. The image on the left-hand side demonstrates a large liver metastases circled in red and several smaller liver metastases circled in yellow. The large one was injected with RP2 monotherapy. And over time, we can see over 3, 6 and 9 months, that the injected lesion has decreased in size from direct injection. But as well, we can see a progression of multiple other sites of uninjected lesions really speaking to both the local and off-target systemic effects of this agent. Next slide. Now one of the keys to success is really engaging key stakeholders and understanding where they live, and we've heard Sushil outline that very nicely in earlier slides. But really tumor-directed therapy requires a success of key stakeholders. And there are 2 groups in particular when it relates to deep visceral injections. The first is the medical oncologists, obviously, general or specialized to their specific cancer type are often the main point of contact. But interventional radiologists as well are a key group to engage. They are often already treating liver cancers primarily through liver-directed therapies, such as tumor ablation and embolization, and they're going to be a gateway for all image-guided procedures, in particular, deep injections, whether it's in the liver or elsewhere. And one of the ways to really engage them is through societal involvement. In particular, Replimune is engaged with the Society of Interventional Oncology, of which I'm the president-elect, to really identify these key stakeholder groups early as a way to ensure both identification of clinical trial sites that are going to be successful, but also for a more longer-term engagement with this for commercialization as well. And it's important to remember that many of these -- sorry, if you go back, many of these individuals do engage commonly in the form of virtual and multidisciplinary tumor boards or clinics. We have a well-established one at [indiscernible] as well. So there are several ways into these key stakeholder groups for early clinical trial site engagement. Next slide. And so when we think about intratumoral therapies, they're actually really increasing overall use at multiple centers. This is an example of that with clinical trial data that was published by M.D. Anderson from their Interventional Radiology Group, really reporting the use of direct intratumoral injection across a number of immunotherapy trials and a total of 1,300 injections. And it speaks to the wide range of potential injection sites. We can see here on the left they injected multiple organs and multiple sites. They also were able to perform 4 biopsies, both at the injection site and at non-injection sites. So the ability to collect non-target data is part of clinical trials and routine practice. And overall, the serious adverse rates were very, very low, regardless of which trial they were in, just speaking to the fact of the overall safety profile of a direct intra-tumoral injection strategy. Next slide. And so to summarize, really intra-tumoral immunotherapy injection is a very compelling proposition for liver metastases as this is a frequent site across multiple tumor types, and usually associated with both poor prognosis regardless of the primary tumor type and with associated reduced activity of systemic immunotherapies as well. RP1 and RP2 can be safely administered repeatedly to hepatic metastases and have demonstrated activity both locally and systemically. And ultimately, intra-tumoral immunotherapy injection of these organs can likely be safely escalated to other sites as well. And I will pass it back to Rob. Thank you.

So thank you very much, indeed, Muneeb, for that overview. So I'm now going to provide an overview of the RP2/3 program, including the data to date and the development plans for Phase II, which is really the first time we've presented these. So as a reminder, RP2 and RP3, which are really one program, leverage our ability to express multiple transgenes from our high oncolytic potency backbone enhanced through the expression of the fusogenic protein, as Philip already described. RP2 and 3 focus on delivering potent immune activators to tumors and draining lymph nodes delivering molecules which function at both the time and the place of immune response activation and which also have a level of clinical validation, which is part of the thought process behind which proteins to include. So RP2 expresses an anti-CTLA-4 antibody, as has been clinically validated by both ipilimumab and tremelimumab, with RP3 further expressing immune co-stimulatory molecules CD40 ligand and 4-1BB ligand. Agonistic antibodies against both CD40 and 4-1BB have shown evidence of activity in the clinic, but also have been shown to be somewhat toxic. So as for anti-CTLA-4, we believe the direct delivery of each of these into tumors should enhance not only their activity, but also minimize systemic toxicity. So altogether, this approach seeks to maximize immunogic signal 1, that being antigen presentation through the lytic effects of the virus killing the tumor and releasing antigens; and also signal 2, that being immune co-stimulation; and resulting signal 3, the release of inflammatory cytokines. And each of these should be enhanced through the expression of both the anti-CTLA-4 and the co-stimulators used. So preclinical data demonstrates the promise of the approach. In this experiment, we've grown tumors in both the left and right flanks of mice. The right tumor has then been injected with either vehicle or a low dose of RP1, which is itself not highly effective, particularly in uninjected tumors at that low dose. So that's the second panel down on the left. However, when we further encode an anti-CTLA-4 antibody, CD40 ligand or 4-1BB ligand or indeed in another co-stimulator, OX40 ligand in the virus and then inject at that same low dose, we see that not only are the antitumor effects improved in the injected tumors but effects are improved even more in the uninjected tumors as indicated by the 3 other red circled groups on the left and the right. Therefore, we can conclude from this experiment and other similar experiments that the local delivery of each of these molecules into tumors increases not only the local benefit but also the systemic immune-mediated benefit, which we achieve. So if we come on to clinical data, as we previously presented both last year and the year before at SITC, in a Phase I population of 9 patients using RP2 monotherapy at 2 different dose levels to determine the recommended Phase II dose, while 5 patients of those 9 rapidly progressed as would be expected, in an advanced Phase I population, 3 of the patients achieved an overall objective response. This was durable out to 15 months in uveal melanoma -- in uveal melanoma patients, and ongoing at the last data cut in the other 2 patients who had esophageal cancer and mucoepidermoid carcinoma, a type of salivary gland cancer. And this durability further speaks to the immunologic potency of the approach because without durability you wouldn't expect the immune system to be kicking in. So all patients have failed standard of care, including checkpoint blockade for both esophageal cancer and uveal myeloma patient. So this is one of -- the first example in mucoepidermoid carcinoma, which is actually one of Kevin's patients who will speak again in a minute on head and neck cancer, who's achieved an ongoing complete response following injections into only the upper lesion in the neck. The lower lymph node wasn't injected, but the patient overachieved an overall durable complete response, which is still ongoing out now, I think, at around 2 years. So this is the esophageal cancer patient who failed multiple therapies, including an anti-PD-L1 antibody on a clinical trial. The patient had relatively limited liver disease and also other lesions in the abdomen, and it was just the liver lesion which was injected. All achieved a response, and this patient actually has an ongoing metabolic complete response, although a RECIST partial response out now to over 18 months. So very clear single-agent activity with RP2. When we combined RP2 with nivolumab, we also saw clinical activity and with no new safety signals as compared to when we combined RP1 with nivolumab. So here, we had a 30-patient Phase I cohort of patients with advanced disease, all of whom having failed standard of care, including immune checkpoint blockade where they were eligible for that. And out of that group of 30 patients, we saw 7 overall responses as of the last data cut, which was presented in November. So this included responses in patients with uveal melanoma, cutaneous melanoma and head and neck cancer, with all of those responding patients having failed prior anti-PD-1 therapy. So biomarker data with RP2 as for RP1 has also been promising, indicating broad and potent immune activation. A snapshot of some of that data is shown here. In biopsies on the top left panel, we see very strongly increased expression of CD8 and PD-L1 in patients whose tumors had initially low levels of these and gene expression analysis on the bottom left panel shows that all of the classes of genes associated with immune activation are indeed increased as we would hope. We also see that there's no association of either baseline PD-L1 levels or CD8 levels with response, as shown on the right panels, indicating the ability of RP2 combined with nivolumab to induce responses, whether or not tumors are immunologically hot. So as you can see, the data with RP2 clearly demonstrates clinical activity also with promising supportive biomarker data warranting movement moving into Phase II. So I'll now hand back over to Professor Harrington to run through the Phase I clinical trial with RP3, on which Kevin was one of the investigators along with many other of our trials, in fact. So over to Kevin.

Many thanks, once again. And if I could go straight to Slide #83, please, which gives us a summary of the initial arrangement that we had for this RP3 Phase I trial. So trial design, as you might anticipate, the inclusion criteria were typical for a Phase I study, offering this treatment to all-comers patients with any tumor type having failed all standard of care or other known treatments that were available to them. Now in the first instance, we dosed 3 patients at the low dose, and you can see on the slide that's indicated by an initial dose of 10 to the 5 pfu/mL, followed by 4 doses at 10 to the 6 pfu/mL up to 10 milliliters of injectate. And then subsequently, we followed those patients and were treated with 3 patients at the higher doses, as you can see, that was a log higher across each of those dose levels. Both groups could have been expanded to 6 patients in the event that we'd seen a DLT. That actually wasn't necessary. Safety was assessed and scans were performed at 1 and 3 months following the last dose of treatment, with 5 doses being the total dose of injectate that was given. There was no prolonged dosing in this group of patients. Dosing could be given both by direct intratumoral injection, which indeed we did for many of the patients, and also by image-guided injections, as you've heard so elegantly earlier on in this set of presentations, and that was for deep tumors. And once we had achieved the recommended Phase II dose, we have the opportunity to enlarge the population of patients with Herpes Simplex Virus seronegativity in order to expand that group to at least 3 patients so that we would be able to understand the safety in that group of patients. And then subsequently, the opportunity to recruit a 30-patient cohort using the same inclusion criteria, and now combining with nivolumab, opening that study, which is now recruiting. I'm pleased to say. So on Slide 84, I give you some data regarding what's happened with these patients. So the current status is that the first 3 patients enrolled at the low dose did not demonstrate dose-limiting toxicity. And we then went on to the higher dose. And again, we did not see dose-limiting toxicity. The recommended Phase II dose, therefore, was declared as the higher dose, so 10 to the 6 followed by 10 to the 7 x 4. And the combination with nivolumab enrollment has been initiated at that dose level. And as I've told you, we were able to recruit an initial seronegative patient to complete the data set that we have around safety with that. The current available data is for these first 6 patients dosed. So it's a relatively early view of the data dosed as monotherapy, I'm not going to share with you the data around the combination therapies, and this is currently being expanded both in the U.S. and in Europe. So on the next slide, please, Slide 85, I show you the summary of the safety data. And again, you can see readily for yourselves on this relatively busy slide. If you focus on the grade 3 and above toxicities you see a porosity of toxicity there. So again, what we've seen is flu-like constitutional symptoms that tend to resolve within the first 72 hours, the vast majority of those being grade 1 and grade 2. However, I think one thing that doesn't really come through very clearly on this slide. And I think those of us who've been investigators on the RP1, RP2 and now on the RP3 program would attest to, is the fact that this virus is significantly potent and we do see patients developing very significant reactions. Indeed, shortly before coming to give this presentation, I injected a patient in the clinic this morning in London. And this patient previously, with their last cycle of treatment had really quite profound fever and rigors, which settled with conservative management and the patient was entirely fine. But this is a potent virus I think it tokens the fact that it is very well endowed with payload agents, which are really designed to activate immune responses. So on the next slide, I'll show you an example of a pleomorphic sarcoma, and you can see at baseline and then at day 3. The impression that you get here really for this lesion, which is superficial enough for us to be able to capture this phenomenon, is really localized inflammation at this injection site, again, suggestive of immune activation. And if I had been able to show you pictures from the patient I treated today, again, you would see really quite impressive peritumoral inflammation in this patient following his injections. So these agents are potent. So I now go on the next slide, in 87, to show you the nature of the patients recruited. And this was a cohort of patients thus far, as you might expect in a Phase I study, where the disease burden was extremely high. You can see here across the various patients really very widespread disease, both in terms of the dispersion of the disease across multiple organ sites. But also within organs, you can see heavy disease burden. So for instance, I draw your attention in the bottom right-hand corner to the volume of liver disease in this patient. So this is typical of a Phase I population. Next slide, please. And here you see, in really quite detailed analysis, the various features of those first 6 patients and you can see a range of different tumor types and not just the ones that you would expect or have been tested or were a part of the initial RP1 program. So you can see here, sarcoma represented, esophageal carcinoma, head and neck cancer, about which I will say more in a few moments. You can see that the patient population is heavily pretreated, with extensive multiorgan, multi-site disease. And you can see that injections were given both superficially and into deep sites. So although we have established a recommended Phase II dosing demonstrated safety, it's probably a little early to be able to comment on responses. And you can see that a number of patients didn't complete the all 5 injections due to early progressive disease given the advanced nature of their disease. Indeed, unfortunately, 3 of the 6 patients, 2 of the -- both of the melanoma patients and a patient with colorectal cancer died of progressive disease within 4 months of starting treatment. However, one patient has achieved stable disease for a year then regrettably progressed in liver lesions. And one my head and neck cancer patients is still in survival follow-up. So although the overall population that has been enrolled, has given us the information around safety and allowed us to determine this recommended Phase II dose, I think for the subsequent efficacy endpoints within the trial, we need to look to the later population of patients being recruited. I'm now going to hand back to Rob, who's going to discuss the intended Phase II development program with RP2 and RP3. So Rob, back to you.

Thank you very much. Indeed, Kevin, for that great overview of where we are with RP3 to date. So I'll -- I'm now going to resummarize I suppose, the data with RP2 and RP3. So in summary, both of these agents have been well tolerated, with, as you've heard, maybe slightly more markside effects with RP3 as might be expected with the additional payloads encoded. These generally resolved within 72 hours, and we think indicate that RP2 and RP3 would be expected to be able to be combined across a spectrum of different anticancer modalities, which we do indeed aim to do in our Phase II program with RP2/3. RP2 has shown durable clinical activity, both as monotherapy and combined with nivolumab, including a clear signal in uveal melanoma, which we're further exploring now. We've also enrolled -- we are also enrolling a cohort of patients with GI cancers, lung cancer, breast cancer and head and neck cancer, rather than just the all-comers patients enrolled so far with RP2, although these are still Phase I patients. RP3 has shown very good tolerability although somewhat more robust side effects than RP1 and RP2, and is expected to provide enhanced efficacy as compared to RP2. So while safety has been able to be provisionally established and the RP2D declared for RP3, it's too early to draw any conclusions as to efficacy. As for RP2, a focused cohort of patients with lung cancer, breast cancer, GI cancers and head and neck cancers is to be enrolled, rather than just the all-comers patients, as we've enrolled so far in the 2 monotherapy cohorts just reported upon. Overall, we believe it appropriate to keep our options open as to which of RP2 and RP3 to progress in particular situations as the data with RP3 catches up, which is really a year or so behind RP2. So we're, therefore, writing the Phase II protocols, I'm about to come on to, to use both RP2 and RP3, the intention being to open one or other of the cohorts with RP2 or RP3 based on the totality of the data available at that time. This also provides optionality to open both if we so choose. So we'll follow the data and decide which to use in each case as we get towards actually opening the protocols I'm about to come on to. So with RP3 clearly ready for Phase -- for broad Phase II development and the data with RP3 running somewhat behind, these are the overarching principles, which helped us determine the range of Phase II indications, which have already been alluded to earlier in the various talks, and the combination drugs to pursue. First, we wish to choose settings where in general, there's the opportunity to show a signal in a single arm setting compared to the combination drugs used, usually the standard of care. There needs to be a clear unmet need and an attractive underlying landscape for development, and feasibility of injection clearly needs to be taken into account too. We aim to develop across a spectrum of areas where an oncolytic immunotherapy, which can directly kill tumors and ignite a systemic antitumor immune response to cold tumor spots might be used. This includes truly immunologically cold tumors, where I-O doesn't work; tumor types were only patients with high PD-L1 response; and patients with responsive tumor types where I-O has failed. This includes developing in settings with locally advanced disease and the neoadjuvant use, too. Finally, we aim to combine with current standards of care to move earlier in treatment, ideally to the first-line setting in at least some cases, with the potential for synergy with these other anticancer approaches, too. Following from that, we've also identified some high-conviction, high-reward indications for initial prioritization as you'll see in the coming slides. So the first area where we wish to establish, develop RP2/3 is in the liver cancer and liver metastases setting as both Muneeb and Sushil have previously discussed for reasons we've set out indeed before. The liver is a very common site of metastases. And patients with liver mets have a very poor prognosis and also responds very poorly to IO. Liver mets appear to eliminate tumor-specific T cells through the action of tumor-associated macrophages. We expect that the destruction of the tumor by RP2/3 and the induction of new tumor reactive T cells will help to overcome that. Primary liver tumors and liver mets are also routinely injected as standard of care, as Muneeb discussed, indicating the practicality of targeting those tumor types. The areas we've identified for development here, a first-line hepatocellular carcinoma combined with standard-of-care immunotherapy, which is atezolizumab combined with bevacizumab, and second-line hepatocellular carcinoma combined with anti-PD-1. The second-line setting is an area where there's no effective standard of care and where any level of activity would provide a clear signal and a potential path to registrational development. Additionally, we're also conduct a study in third-line colorectal cancer where there's also no standard of care and where any activity would provide a clear signal and a potential path to registration development. This, likewise, would be in combination with anti-PD-1. So I'll come back to some more details of all of these towards the end of the session. So the next area of interest is to move into treating patients with earlier disease. Here, we believe there's considerable opportunity and where often the objective can be to achieve actual cures. Tumors are also often easily accessible earlier in disease. And we believe the safety profile of RP2 and 3 is appropriate for us both in patients with earlier disease and in combination with the other therapies, which might be used in earlier disease. So as already mentioned, we're planning for a study in the neoadjuvant setting in CSCC, which will be with RP1. But as part of the Phase II program with RP2 and 3, we're planning a study in local regional advanced head and neck cancer, combined with standard-of-care chemoradiation with curative intent to be followed, in fact, by adjuvant anti-PD-1 therapy. So following an initial safety run-in phase in that clinical trial, we intend to rapidly move to a randomized controlled trial with registration intent. Additionally, in patients with early disease, we'll also be supporting a number of investigator-initiated trials, about which you'll hear more over the year, including early breast cancer and additional trials in skin cancer, including in particular settings, some of which Nikhil discussed earlier in CSCC. So the next area -- sorry, the final area is that of overcoming IO resistance. So as shown, RP viruses can turn cold tumors hot and deliver responses irrespective of baseline PD-L1 and CD8 T cell status, including in patients who failed prior IO. Here, we'll further work in head and neck cancer, targeting first-line recurrent patients who, while eligible for anti-PD-1 therapy, don't respond well to it, those being patients with low levels of PD-L1. Here, we'll combine with standard-of-care chemotherapy and anti-PD-1. I already mentioned the Phase II study in second-line HCC, which while being in the liver category, also falls into the IO-resistant category as well. We'll also be planning for some additional signal finding studies in some other cancer types as well. So I'll now hand back to Professor Harrington, who will talk about the background, the current treatment landscape in head and neck cancer and the areas of unmet need. Following which, Dr. Tony Saab of the Mayo Clinic, with whom we've been working on the Phase II development plans in the eye cancers, will do the same for HCC and CRC. So we with that, handing back over to Kevin.

Thanks again, Rob. And the audience will be pleased to hear, I think, this is my last segment just for now. So next slide, please. I'm going to try to explain to you some of the issues relating to the treatment of head and neck cancer, and where there may be opportunities. So for those of you not particularly familiar with head and neck cancer as an entity, on the next slide please, I'll show you just the background. So these are tumors that arise between the skull base and the clavicle, affecting areas of the pharynx, the oral cavity and the larynx. And this is a common disease, actually, worldwide, 1 million cases per year projected by 2030. The standard historical etiological events, being alcohol and tobacco consumption, but more recently, we've been seeing an epidemic of patients relating to infection with typically type 16 human papillomavirus for oropharyngeal cancers. And the majority of these tumors are of squamous cell histology. We know that these agents are active against squamous cell histology, of course. Next slide, please. So on this slide, I show you a summary, and this is a rather complicated scheme, and I'm not going to walk you through all aspects of this. I ask you to focus on patients with locally advanced squamous carcinoma of the head and neck, the non-resected population, where the gold standard of treatment in the vast majority of centers around the world is concurrent chemoradiation. And you can see that the NCCN guidelines give Category 1 optionality for 3 of those approaches. But by far, the majority of patients will be treated with a combination of platinum and radiotherapy, usually platinum on a 3-weekly schedule. Next slide, please. And here, you can see data which have been increasingly maturing over the last 20 years from the MACH-NC Meta-analysis Group, in which we look at the addition a systemic therapy to locoregional therapy in the main radiation, so radiation plus chemotherapy. And you can see that there is a highly statistically significant improvement in overall survival to the tune of about 6.5% at 5 years. But I think the thing that I take away from this as a head and neck oncologist is that there is a very significant unmet need here. Most of the patients regrettably still succumb to their disease within a relatively short period of time. Next slide, please. So we have tried to integrate immunotherapy into standard-of-care chemoradiotherapy regimens. And I'm going to rehearse for you 1 or 2 studies that have been conducted and published and presented within the last year or 2. The first of these is the JAVELIN Head and Neck 100 study. And in this rather complicated schema, you can see that patients with high-risk, locally advanced disease, almost 700 patients randomized 1:1 between gold standard treatment, which would be platinum plus 7 weeks of radiation, with either placebo given a week beforehand, during and then after for 1 year as adjuvant therapy, or the experimental arm in which we interceded with the use of an anti-PDL1 antibody avelumab. And the endpoint of this study, looking for an early positive readout, was progression-free survival measured by investigator per RECIST 1.1, looking at this as an early 1- and 2-year landmark. Next slide, please. So the groups are well balanced. But what I show you here is the data -- are the data for the primary endpoint. And you can see, much to our disappointment, the placebo curve lies superior to the experimental arm of avelumab here for the progression-free survival endpoint. So you can see that the addition of [ concomitant ] induction, concomitant and then adjuvant immunotherapy and that schema was not successful. The other message to take away from this is, this is a disease truly of unmet need. At the 24-month landmark analysis, you can see that up to 50% of patients in each arm has relapse of their disease. Next slide, please. And when we look at the overall survival analysis from these patients, you can see again, the curves from the point of view and the investigators are in the wrong place. The avelumab arm is doing inferiorly to the placebo arm. No benefit, indeed. A hazard ratio of 1.31, not a statistically significant detriment, but nonetheless, a clear indication that there would not be a survival advantage. And again, the take-home message here is that at around about 36 months, 1/3 of our patients have died of their disease. So this represents a significant medical problem. I'll show you very briefly on the next slide another study that has been conducted this time with the drug pembrolizumab, looking with a slightly different approach in a group of patients less well able, less fit to receive concomitant platinum-based chemoradiation. You can see the randomization is between cetuximab radiation versus pembrolizumab radiation. The projected local regional control of 15 months with cetuximab radiation by the investigators was 60%. They pretty much nailed that. It was 59% in the study. But pembrolizumab did absolutely no better. And again, you can see that the odds ratio and the p-value indicate no benefit here from the addition of pembrolizumab to radiation when compared with another gold standard cetuximab radiation. So where should we go? And what are the opportunities? On the next slide, I just want to rehearse with you some data published most recently in full paper form, but this is the slide design from the original ESMO Congress. You can see a small randomized Phase II study of a novel immunotherapy agent. This is Xevinapant, which is a SMAC memetic drug that modulates the way cells die. It's an inhibitor of inhibitors of apoptosis. Here, you can see Xevinapant combined with chemoradiation versus placebo plus chemoradiation. And again, the primary endpoint of this study was a local regional control endpoint done at 18 months. And if we go to the next slide, I show you really the stunning data for progression-free survival, albeit with relatively small numbers. You can see here a marked difference between the experimental arm and between the control arm of chemoradiation Now this can be chemoradiation arm is admittedly performing less well than we may have expected from historical controls, but this was a highly advanced population of patients. I think the take-home message from me from this study is that we have an opportunity here truly to combine novel immunotherapeutic approaches above and beyond the simple PD-1/PD-L1 approach and maybe to use other agents. So this is an approach to Rob Coffin and I have explored before. And on the next slide, I show you the design of a clinical trial that we published now fully 12 years ago, in which we designed a study with standard-of-care chemoradiation in which we gave a herpes simplex virus, an oncolytic herpes simplex, Talimogene laherparepvec, T-VEC as you all know. During that treatment given at each dose of platinum and then a further dose 3 weeks after the end of chemoradiation delivered intratumorally into patients with neck disease and heavy nodal burden. Next slide, please. And here, I show you an example of one such lesion that was injected a patient with a hypopharyngeal tumor, the white arrow indicating the large nodal mass. And you can see on the right-hand side of the slide, Panel C indicating a complete response. Next slide, please. And here, I summarize you the data from that Phase I/II clinical trial. And treatment was extremely well tolerated. No grade 3 to 5 toxicities. Significant number of patients showing response by RECIST. Pathological complete remission confirmed in the neck dissection specimens of 93% of the patients. We were also able to detect the fact that the virus was replicating in and spreading within the nodes that we had injected. We showed that there were very impressive disease-specific and overall survival data, locoregional control achieved in every patient and a relapse-free survival rate in the initial publication of 76.5%. And at the latest update of this, more than half of the patients still in remission on long-term follow-up. Next slide, please. So the conclusions from this approach, which is a concomitant approach with a novel immunotherapeutic agent, is that although we have the opportunity of having effective primary therapy with chemoradiation, a large number of patients will relapse, especially those with high-risk disease. It doesn't look as if anti-PD-1, anti-PD-L1 therapeutic approaches concomitantly with chemoradiation are going to be the solution to this. And I think that means that the field is wide open and in urgent need actually of developments of better therapeutic opportunities. And I believe that oncolytic immunotherapy is just such a treatment opportunity, the ability to kill and sterilize tumor cells through the oncolysis, but also the activation of the immune response to help clear up the disease and protect against subsequent relapse. We would anticipate synergy, both with the radiation and the chemotherapy. Indeed, those data have been published. And in addition, the treatment beyond the completion of chemoradiation with adjuvant nivolumab as per the PACIFIC study in Stage III non-small cell lung cancer may be another way to amplify this benefit. And in the final session, I'm just going to discuss about unmet need in relapsed metastatic head and neck cancer on Slide 110. And on Slide 111, I rehearse for you the way that we have been developing immune checkpoint inhibition therapies in head and neck cancer. Typically, of course, as you know, working from the right-hand side of the slide from the bottom to the top, the direction of development is usually in relapsed end-stage disease progressing back towards curative opportunities. And that's exactly what we've been doing with approvals in the second line for pembrolizumab and nivolumab. And in the first-line setting for pembrolizumab monotherapy or pembrolizumab plus chemotherapy. But as yet, there is a wide open opportunity both in adjuvant therapy and in concomitant therapy with chemoradiation for the development of a novel therapeutic approach. Next slide, please. So here, I show you some summary data the KEYNOTE-048 study. Now this is the study that demonstrated that the addition of pembrolizumab to platinum and 5-fluorouracil was superior to gold standard platinum 5-FU cetuximab. And here, you see the data for the patients with the PD-L1 CPS greater than or equal to 1. This is the most broadly applicable license for this treatment globally. You can see that there's a benefit, clear cut. But what you also see is that the vast majority of patients, unfortunately, still succumb to their disease. So there's a need for better treatments. And on the next slide, #113, I show you data from a long-term follow-up analysis where we've looked at patients with very high CPS, greater than or equal to 20, excuse me. CPS greater than or equal to 1 on the right-hand side. You can see, in fact, most of the benefit here is deriving in the high CPS values for patients treated with chemotherapy plus pembrolizumab. But even for that small group of patients who benefit, we have to remember the vast majority of patients who sadly dying of their disease and the need for further better treatments. Next slide, please. So here, I conclude where I think the opportunities might lie both in the first-line setting -- setting for first-line and recurrent metastatic disease. So we may have 2 new standards of care, but they're not yet good enough. And although we see benefit for chemotherapy plus pembrolizumab combinations, there is an opportunity to do more. We believe that the use of chemotherapy can reduce the rate of early progression. There's a significant unmet need for the patients who have a relatively low CPS value, less than 20, so between 1 and 19, and indeed less than 1. And we have a real opportunity here for intratumoral RPx agents directly killing tumors, turning cold tumors into hotter tumors, increasing expression of PD-L1 and inducing systemic immune responses. And we have every expectation and the data thus far from the program suggest that this will be potentiated by anti-PD-1 therapy, and there may indeed be a synergy with chemotherapy drugs, including taxane-based chemotherapy. So we believe that there are new opportunities here to intratumoral RPx platform viruses RP2 and RP3 with chemotherapy and immune checkpoint blockade, especially targeting the CPS less than 20 population. So with that, I hand you back to -- I think back to Philip, who will introduce our next speaker, I believe.

Yes. That's Tony Saab, leader of gastrointestinal at the Mayo Clinic. Over to you, Tony.

Yes. Thank you. All right. So we'll go through data regarding hepatocellular carcinoma and colorectal cancers, specific considerations for unmet needs. The next slide, please. These cases are certainly both colorectal and liver cancer are some of the most common cancers in the world. They're in the top 7, and you can see colorectal cancer is in the third spot and liver cancer in the sixth spot. When we look actually at cancer deaths, now these 2 cancers are actually in the top 3 cancer killers in the world and does continue to represent a significant unmet need despite the improvements that have been occurring over the last few years. Next slide, please. So first, I'm going to start with liver cancer and go a little bit through the landscape and where the need lies. Next slide. So when looking at the high-level overview for the treatment landscape in HCC, and we typically use this Barcelona Liver Clinic Classification to look at the various stages from the 0s to A and C D. 0, of course, is the very early stage where mostly it's a surgical or transplantation issue, liver transplantation. The Stage D is, unfortunately, the last stage of the cancer. The focus of the discussion will be primarily around the advanced stage, the C. Although a lot of what is applying now to the C is moving a little bit into the [ B ] as well as we continue to enhance our likelihood to see responses in a disease where we haven't traditionally seen responses with therapeutics. Therapy overall has advanced relatively well with immunotherapy. For the longest time, we didn't have any standard. After 2007, '08, we had multikinase inhibitors, sorafenib. And for many, many years until just 2020, we haven't had any meaningful regimens in the first line, immunotherapy specifically, to move the needle further. And we have this with atezolizumab and bevacizumab and now with durvalumab and tremelimumab in advanced HCC. There's still a significant need. As we go through the data, you'll see that although we are improving, we certainly are hitting a plateau. There are novel approaches that are being investigated, primarily local interference or direct interferes with the local immune media or macro environment. Now we believe we'll continue to enhance further the likelihood of seeing meaningful responses. And then beyond the first line, it becomes quite tricky. We have agents mostly multikinase inhibitors, mostly tested post sorafenib. However, these have moved the needle very little for the most. And this represents, again, another significant unmet need and important to continue to explore the role of immunotherapies. Next slide, please. So the first experience with first-line PD-1 inhibitors immunotherapy have been with single-agent nivolumab, CheckMate 459, which compared that to sorafenib, suggesting that essentially a small subgroup of patients may end up being benefiting. But it wasn't meaningful enough to actually lift the curves beyond sorafenib. And unfortunately, it was negative. It was a setback. But then the thought is that in HCC, a single agent PD-1 or PD-L1 will probably not move the needle enough for most patients. And therefore, combinations that were rationally designed based on early clinical data as well as significant preclinical data suggesting that adding essentially agents like anti-VEGF agents with bevacizumab or multikinase inhibitors would essentially enhance the immune competency of the milieu and help cells migrate to the tumor and therefore, enhance the activity. Next slide, please. Additional data also suggested that where PD-1 blockade on its own may not be sufficient, adding a CTLA-4 inhibitor will actually enhance further the activity and therefore, increasing significantly the likelihood of seeing more meaningful responses in more patients. Next slide, please. And in fact, this certainly has translated into improvements. And we'll see the results next. But in this situation now, we have 2 standards -- standard of care. The preferred one is atezolizumab and bevacizumab for first-line HCC. And then for select patients, those that are at risk for severe bleeding, and I can tell you for the selected group, it's very few patients perhaps tried tremelimumab and durvalumab. And there's still a significant proportion of patients. You've heard quite a bit from previous discussions about patients who essentially either are transplant -- solid organ transplant patients or those patients essentially with, in this case, with severe autoimmune diseases, there's a contraindication to PD-1 or CTLA-4 inhibitors. Those patients, at least in HCC, and remember, HCC is a disease where we transplant the liver for the early, early stages, yet many of those patients can have local recurrences, and they're not eligible for IO therapy anymore. Those patients would be eligible for lenvatinib or sorafenib. And it also create opportunities for local manipulation of the immune microenvironment without going systemic with PD-1 and CTLA-4 inhibitor, an area that could be explored certainly with direct injections into the tumor site. Next slide, please. This just shows you where we are today. We have 2 studies -- 3 studies, Phase III studies, that we're going to focus on, that essentially it changed the landscape of how we treat patients in the first-line setting of hepatocellular carcinoma. So COSMIC-312, the one on the right, actually look at cabozantinib and atezolizumab versus sorafenib. This study was positive on the progression-free survival, but was negative on the overall survival. These were 2 co-primary endpoints. So unfortunately, the lack of survival benefit will make this a very difficult proposition to move forward with, especially the toxicities appear to be amplified. And the response; rate was 11%. And that's, frankly, no better than what you would expect a single agent PD-1, PD-L1 inhibitors. Certainly a disappointment. IMbrave150, which looks at atezolizumab and bevacizumab, bevacizumab monoclonal antibody to VEGF and atezolizumab PD-L1 inhibitor has shown essentially improvements across all efficacy parameters from survival, to progression-free survival, to response rate. Now notice the response rate is 30% with 8% complete responders. And that's fantastic. For liver cancer given the fact that we previously haven't expected responses above 11%, but certainly, we're not there yet. We're not even close to the 50% mark of response. It's an improvement, but certainly -- and significant, but certainly not where we want to be yet. With HIMALAYA, with durvalumab and tremelimumab, a combination called STRIDE, there was a little bit more modest improvements historically, at least, with median overall survival. But disappointingly, the progression-free survival was not as good. In fact, at the median at least, was a little bit lower than sorafenib, but the hazard ratio about the same. The responses were observed than 20% of the patients with only 3% of complete response. So this is why atezolizumab and bevacizumab would remain the primary standard with backup for a few patients for the STRIDE regimen. The toxicities are relatively tolerable and mostly acceptable in this group of patients. Next slide, please. Now when we hit the second line, it becomes more obvious that -- well, 2 things become obvious is: one, the outcomes remain very modest when improved; two, the responses are far few between; and then three, the toxicity, specifically with the multikinase inhibitors can be very limiting for this group of patients. And as you can see in the CELESTIAL, RESOURCE; CELESTIAL was with carbozantinib; RESOURCE with regorafenib; and REACH-2 was with ramucirumab, although the alpha-fetoprotein marker was -- in this study was above or equal to 400, which is about 40% of patients with HCC. There were no responses, almost no responses in those -- with those agents. Improvements with both survival and progression-free survival were incredibly modest. And ultimately, these were in the setting of prior sorafenib. So today, in a world where we've moved to atezolizumab and bevacizumab, so IO primarily in the first line, it's very difficult to put in perspective how exactly those studies continue to fare in a landscape that gets more complicated. KEYNOTE-224 look at pembrolizumab. And then one as a follow-up study KEYNOTE-240 with pembrolizumab single agent versus placebo, that although numerically did show some improvements, but these improvements was not statistically significant versus placebo. There's a China study looking at the same that appears to be positive. And so the combination of the 2 studies may actually keep pembrolizumab as an option. Again, those studies were done in the setting of post sorafenib, and not post atezolizumab and bevacizumab. And then CheckMate-040 did have nivolumab and ipilimumab arm. The response rate was very promising. This was a single-arm study. The first-line study comparing it to standard-of-care TKI is ongoing. However, one consistent element here is, all these studies were performed following the failure of sorafenib. We continue to see that IO is, if I may use this, is the way to go. Meaning IO in this disease, it seems to move the needle a little bit further historically than tyrosine kinase inhibitors. And especially as you amplify the utilization of CTLA-4 inhibitors or VEGF inhibitors, as you move even further, that there was an improvement of outcomes versus PD-1 or PD-L1 inhibitors. Next slide, please. It's also important to note that a lot of these tyrosine kinase inhibitors that are used to have significant toxicity, especially in this patient population. So how can we continue to expand the benefits of immunotherapy? I mean we've had a plateau. We have improvements, but we've needed that 20%, 30% response rate plateau. And so one of the thoughts is, essentially, is to continue to move forward with enhancing essentially the tumor-associated antigen exposure and primarily targeting intratumoral treatment. And there are different ways to do that. The less specific, meaning applying radiation, whether externally and internally, that certainly appears to improve outcomes somewhat, but these are very nonspecific ways to do this. Next slide, please. And others, and again at the spirit of all the discussions today, is applying oncolytic viral therapy, virus infection directly to the tumor. Now these are tumors that primarily in HCC stay in the liver. A few metastasize outside the liver. So liver remains the primary target. We know that liver tumors intrinsically, across multiple malignancies, are tend to be immunosuppressive. And so local milieu manipulation makes a lot of sense. In this instance, injecting oncolytic virus directly into the tumor that may end up releasing antigens and recruiting in effector cells, which essentially brings those carpeted T cells and other immune cells to the milieu, the prime expanded, they travel to the tumor from the lymph nodes. And ultimately, with these activated T cells, as they infiltrate the tumor cysts, ends up exposing the area to higher level of antigens and cells that are essentially ready to kill those cancer cells. And this antitumor response can be boosted by immune checkpoint inhibitors. So that presents essentially a great asset with the hope to enhance further the benefits we've seen in HCC with immune therapies. Next slide, please. So thinking about this in perspective, so immunotherapy, obviously, with atezolizumab and bevacizumab and somewhat with STRIDE, has moved the needle for us. But we've had a plateau in first-line HCC. Beyond first line, there is absence of data post IO. So now that our world in the first line moved to atezolizumab, bevacizumab or STRIDE, we do not have that data in the second line. We don't have much data in the second line. Lo and behold, knowing that IO seems to be driving the post here, whether reversing resistance, such as with -- priming with injected oncolytic viruses into tumors, whether that essentially can prolong the benefit that we've seen with IO and enhance it further. So the tumor directed oncolytic immunotherapy appears to have promised to improve IO effectiveness in first line. And we hope that it may actually overcome resistance in the second line, and therefore, extending the life of IOs and enhancing the response. And as you've heard, liver procedures by interventional radiology is a pretty routine in HCC. In fact, one can foresee how we can continue to move this into that Stage B, also earlier stage -- disease stages and even into the neoadjuvant settings to continue to enhance outcomes. Next slide, please. Now I'm going to move on to colorectal cancer, specific focus on refractory colon cancer. Next slide, please. So we've recently published this summary, but essentially improvement in outcomes with the advent of a lot of these agents that are either specifically and biologically targeting certain subgroups of colorectal cancer patients, or in the case of a very small percent of patients, about 2% to 3%, with MSI high with immune therapies, as you can see, essentially, the response rates improve further up as we are able to identify the targets better. And that also is translated into an improvement in progression-free survival. Now many things here that are important to keep in mind. One, is although we're starting to break down colon cancer into theses subgroups, so the totality of these subgroups, if you put all these together, remains less than 20% of patients where we have an identified target, that we can go after, or an immune target that we can go after. About 80% of patients with colon cancer in the refractory setting will have liver metastasis. In fact, that you can make that argument for most metastatic patients with metastatic colorectal cancer. And these liver -- the liver metastatic disease for those patients with heavy involvement in the liver end up being one of the main reasons why patients die from colon cancer. Next slide, please. When we look at the landscape for the other 80% of the patients who, frankly, do not have a targeted therapy, you can see that the outcomes are relatively poor, modestly improved at best, but there also remains significant toxicities for patients who are not as performing as they reach this line of therapy. And 2 agents are available to us. One is a multikinase inhibitor; the other one is a cytotoxic agent, TAS-102. Regorafenib, the multikinase inhibitor. And as you can look -- as you can see in the box down there, there are no responses typically seen with these agents. And the improvements in both survival and progression-free survival are incredibly modest, and the toxicities can be quite pronounced with these agents. Next slide, please. There has been tremendous effort to try to move IO from those patients what we've seen with MSI high colorectal cancer, significant improvements. But this remains limited to 2% to 3% of patients. The majority of the patients are microsatellite stable or proficient MRR. And for those patients trying to move IO into the settings where we can actually see similar responses has been incredibly challenging. We've recently published a study in JAMA Open Networks and essentially look at the contribution of atezolizumab, a PD-L1 inhibitor, to standard chemotherapy. And the one arm had the atezo, the other arm had a placebo. What we've seen interestingly is that there is a small subgroup of patients that seems to benefit in the microsatellite stable, that seems to benefit from the addition of IO from atezolizumab. The majority don't. Next slide, please. And these trends appear to be similar, whether it's regorafenib plus IO, whether it's lenvatinib plus IO. You do see these trends. This, by the way, is the first randomized study prospectively assessing that. But what appears to be a common theme across the board is that patients who do have liver metastases, do not appear to benefit as well from the addition of atezolizumab versus placebo, unlike the patients with no liver metastases that appear to drive some relative benefits. So when we look at response rate, for example, it's less than 10%. It's 5.8% to be exact if the patients have liver metastases. And it's 23% if the patients did not have liver metastases. And the main reason for that, we think, and that's been actually shown in multiple studies that the immune milieu and with liver metastatic disease, not just in [indiscernible] but here, since we're discussing [indiscernible] across the board, tends to be immunosuppressive, and therefore, challenging in terms of going directly with PD-1/PD-L1 inhibitors. So you really need to act on the immune milieu to enhance the likelihood of activation of these agents. Next slide, please. So in summary, how can we turn these microsatellite stable colorectal cancer tumors from cold to hot, understanding that this is 96-plus percent of patients with colorectal cancer will have microsatellite stable tumors. And more than 80% of those patients will not have an identifiable target to go after. So it's a significant unmet need. And we've seen that the agents available to us in the setting right now are not -- certainly, are not meeting the need. They're toxic. They end up giving very little improvements in outcomes. In immunotherapy, specifically the PD-1, PD-L1 inhibitors and VEGF-targeting agents do show some limited promise, except the promise that appears to be mostly limited to the non-liver metastatic disease. Because the presence of liver metastases is consistently associated with resistance to PD-1 and PD-L1 inhibition, as I said, in microsatellite stable colorectal case, but with also we've seen these trends across the board in other cases as well. And there's this immunosuppressive microenvironment that's present in this liver metastatic disease. And it appears from preclinical data and early clinical data, the tumor-directed oncolytic immunotherapy may reverse this immunosuppressive microenvironment and liver metastases. And specifically, when we look at colorectal cancer and microsatellite stable tumors, it gives us hope that this may enhance the role of IO in this disease. With this next slide, please, I finish this discussion and pass it back on to Rob.

Thank you very much, indeed, Tony. So I'm going to provide a little bit of a summary in relation to the intended Phase II development program RP2 and 3. So in summary, we believe that the Phase II development plan outlined by myself and Tony and Kevin, hits all of our areas of interest and provides a risk-balanced approach to further developing RP2 and 3. This includes potentially very large markets with unmet high needs, such as third-line colorectal cancer, as Tony has just described. On improving IO effectiveness, on the top left, we're adding first-line hepatocelullar carcinoma to our portfolio, which currently includes first-line CSCC. On overcoming IO resistance, on the top right, we're adding second-line HCC, and also first-line recurrent head and neck cancer with low PD-L1. In bringing IO to new tumor types, on the bottom left, we'll be working, as already said, in third-line colorectal cancer and also continuing to develop the signal we've already seen in uveal melanoma. Finally, in the early disease setting, on the bottom right, as well as planning for a study in neoadjuvant cutaneous squamous cell carcinoma, we'll also, as high priority, we work in locally advanced head and neck cancer in combination with standard-of-care chemoradiation in each of those cases with the objective of achieving higher rates of cure. So these clinical development plans are further summarized here in my last slide. In locally advanced head and neck cancer, at the top, we'll initially have a safety running in combination with chemoradiation and then rapidly aim to move to a randomized controlled phase. This is intended then following an interim analysis expand to registrational size or sort of a 2-step or even 3-set approach. In the first-line recurrent head and neck cancer setting, we'll conduct an approximately 30-patient initial additional signal-finding trial to look at response rates in combination with chemotherapy and anti-PD-1, i.e., the standard-of-care. Then depending on that data, the intent would be to move to a registrational setting. In hepatocellular carcinoma, one box down, we'll run an initial clinical trial with 2 signal finding arms. In the first-line setting, that will be in combination with standard-of-care atezolizumab and bevacizumab; and then in the second-line setting, in combination with anti-PD-1, whereas Tony indicated, there's no effective standard of care with which to combine. Both of those would then be intended to move to a registrational development phase depending on the data seen. Finally, we'll run a signal-finding trial in third-line colorectal cancer in combination with, again, anti-PD-1, where again, there isn't a known to the effective standard of care in which to combine, which again is intended to move to registrational development dependent on the data seen. So we think that really does provide a good risk-balanced approach with multiple shots on goal, including targeting areas of very significant unmet needs and also commercial potential. So with that, I'll hand back over to Philip to sum up what we've heard before we move to a Q&A session.

Yes. Some brief concluding remarks. With RP1, we're on course to establish a major skin cancer franchise. We are maintaining guidance that we expect to complete accrual from our registration-directed CERPASS study in cutaneous squamous cell carcinoma midyear and to release top line data in early '23. Further, we expect to release directional data from our registration directed study in anti-PD-1 failed melanoma in late '22. Launch scale manufacturing has been established, and commercial planning for launch in the U.S. is advancing. With RP2/3, we have an exciting mid-stage program to pursue in GI cancers and head and neck cancer, where expedited approvals in some of these settings could be feasible. Finally, we have a strong cash position to drive value through multiple meaningful data catalysts. We will now turn over to Q&A.

Thanks, Philip, and thanks again for everyone for joining this morning. We'll start with our Q&A. [Operator Instructions] Let's start with a question that's addressed to the KOLs that are on the line here. There's a debate on the logistics and value of the intra-tumoral approach that persists on The Street. What are your views on this modality, particularly in indications like CSCC and anti-PD-1 failed melanoma?

As the question is directed to skin cancers, let's ask Nikhil and then Kevin to comment.

Sure. So I think the -- if I understood this correctly, it's a question of what is the value of an intralesional approach. I think there's a two-pronged benefit to that. Number one is you have accessible tumor. Tumor that can be easily injected and with little or no morbidity at all. It can be done in the office-based setting. Admittedly, in the United States, I think a key issue becomes is if I take time out of my day to actually inject these patients in the clinic does it take away from doing -- seeing other patients? I think that is a real issue. But within a multidisciplinary setting, where we have surgeons who help us with this, but more importantly, taking advantage of the expanding, what I call advanced practice professionals, so our APPs, so our nurse practitioners as well as our physicians assistance, who can be easily trained and we've already done that at market. We routinely have that in terms of clinical trials as well as standard of care. So it is very doable and very targeted.

Kevin, anything to add?

I would agree with all of those comments and specifically the question asked about cutaneous squamous carcinoma and melanoma. And for me, those are, by and large, the most easily accessible tumors. We tend to do those routinely in our outpatient facilities, I guess, what in America is called the office, and these are done really very routinely. I think again, I would fully endorse the comments that you can train people to be experts in this relatively straightforwardly with appropriate educational programs. And actually, for your workforce, I think it's a really strong development because, of course, it gives that level of autonomy. It gives that level of job satisfaction, that skilled allied health care professionals are basically upskilling and they're able to take part in the treatment of the patient in a way that either to they haven't been able to. So I think it's actually -- it is not the problem that people imagine it might be. It's actually very doable.

And actually, we do have a request for perhaps Dr. Ahmed to respond to that same question.

Yes, absolutely. So I mean, I think as we obviously have talked a little bit about more palpable and visible lesions as it relates to cutaneous cancers. Obviously, in the metastatic phase, there are going to be deeper lesions that might need to be injected. Those are easily accessible and visible through kind of standard techniques as we've talked about that are already implemented widely in clinical practice, including biopsies, direct access via ultrasound and CT guidance. And even in some of these cancers that have metastatic nodes that might be getting [indiscernible] biopsies. Again, that's part of routine clinical practice. It's a very short step towards using those methodologies to inject into those tumors as well. And I think with the widespread development of portable. What's happening in parallel is development of portable ultrasound technologies as well that make this very cost-effective and accessible even in potentially office situations.

Great. Our next question is for Dr. Harrington and Dr. Khushalani. Do you think the approval of relatlimab for first-line melanoma could change how patients respond to second-line PD-1 combination therapies?

Sure. That's a very, I think, a loaded question, if I may call it that, given that nivolumab plus relatlimab as combination in the same vial was actually approved about 10 days ago on March 18 in the United States, for the treatment of patients with advanced unresectable melanoma. Do I think that this will impact subsequent line therapy? I think it's too early to tell. The way I look at this is they will likely become 3 buckets of patients that we will be utilizing various anti-PD-1-based regimens in the frontline setting for advanced melanoma. And I certainly would be curious to see what Dr. Harrington has to say in his opinion, but 1 bucket will be clearly the NIVO plus RELA patients based on the approval. I anticipate that, that will take primarily from those patients that we were previously utilizing anti-PD-1 monotherapies via the nivolumab or pembrolizumab. There will still be a place for anti-PD-1 monotherapy in a small subset of patients in the frontline, where we, as clinical practitioners feel that they may not be able to tolerate a combination therapy for a variety of reasons. Now one could make the argument because the manuscript said that the incidence of adverse events, treatment-related adverse events was not noticeably higher. I would actually caution that interpretation because almost 15% of patients in the NIVO/RELA arm discontinued treatment for toxicity as opposed to 8% of patients in monotherapy arm. So there is an increase in toxicity, particularly hepatitis, particularly adrenal insufficiency, both of which could have longer-term consequences for patients. The third bucket will still be the combination of ipilimumab plus nivolumab. I think many of us in the field would still like to see more data on NIVO/RELA or what we traditionally consider the poor prognostic or poor [ actors ] that is high LDH, bulky disease, CNS metastatic disease where ipi/nivo has clearly been shown to be beneficial, particularly in asymptomatic patients or patients with metastatic mucosal or metastatic acral melanoma, where anti-PD-1 therapy doesn't work as well. So I think in those patients, we would still utilize ipi/nivo combinations based on a longer-term data on that. In terms of subsequent therapy, I don't think we clearly understand the immune microenvironment following PD-1 plus LAG-3 inhibition and therefore, using an ipilimumab-based regimen or an alternate regimen potentially on the backbone of anti-PD-1. That could be a second drug to restore or overcome primary or secondary resistance to anti-PD-1, and that could include intralesional therapy as well.

So Dr. Harrington, could you particularly focus on whether or not you think the approval will affect the second line -- the need for second-line options?

Yes, indeed. I mean that was a fabulously comprehensive answer, which makes my job very easy here. I mean I guess I'd point out a couple of things. There are jurisdictions around the world sadly that do not necessarily adopt very quickly these relatively new treatment alternatives. And so it may well be that there will be plenty of patients around the globe who do not have the opportunity of the combination of PD-1/LAG-3 first line. But the question is really about the mechanisms of resistance that will emerge to patients who have been exposed to dual blockade with PD-1 and LAG-3-targeted therapies. And the honest truth is we don't know the true answer to that at the moment. But I would imagine that the sorts of things that we're likely to see in biopsy studies will need to be done in order to document this, is that we will see T cell exclusion. We will see loss of immune heat, and we will see lack of inflammatory micro environmental changes all of which, of course, are changes that we can reverse by the use of intra-tumoral therapies of the potency that we see with RP1 potent enough, but RP2 and RP3 really enhanced potency agents. So I would still be confident that we have a suite of agents available to us that can reverse that profile of resistance when we understand what it is. So I don't see this as an overwhelming threat, but it's certainly something that needs to be studied.

I think if I could just add one more. I don't know how many of you saw the ASCO plenary session from the March series, where ASCO has started monthly plenary sessions. And Georgina Long presented the first look at response rate in NIVO/RELA and the first look at the interim analysis for overall survivorship. The combination of NIVO/RELA the response rate is approximately 43% compared to 35% with NIVO alone. So certainly numerically higher. But overall survivorship at that look did not achieve statistical significance. The hazard ratio was 0.8. So while the combination clearly improves the time to progression, progression-free survival, we have not yet seen any tangible benefits yet in overall survivorship. Now admittedly, some of that could be due to second-line therapy. And I think that's where we will clearly need to parse out how did patients do with second-line therapy, and that clearly remains that opportunity to improve on our second-line options for refractory disease.

Our next question is for Sush. Sush, how do you see oncolytic virus adoption in the academic setting versus the community setting?

Yes. I think firstly, which is a great question. I think firstly, we're going to have to see the compelling data. I mean any time you want anyone to adopt anything, whether it's in the academic setting or the community setting, we're going to need to generate that compelling data set. And as we said and talked about, we believe cutaneous squamous cell is a nice opportunity to do that in addition to the data we're looking to generate in the [ immune ] checkpoint refractory melanoma space. In terms of the adoption, I think it's going to be difficult to say in terms of if you need from a technology perspective, I don't think that when we talk about community nowadays at least in the U.S., most community practices are either associated with a hospital or some of the hospital network and integrated delivery systems. So in many ways, they have the capabilities to do injections if that's the question and so especially deep injections. So I don't think that will be a major hurdle right now. If you've got a community practice, it doesn't have that accessibility to more of an integrated delivering network. Then I think what Muneeb mentioned around even private practice, radiology practitioners, should going to do this with portable ultrasounds and other equipment. And maybe Muneeb, I'll just ask you to comment if there anything else around the community practice that you would see as a hindrance to adoption versus academic?

Yes, absolutely. Well, I'll make a couple of points. I think the first thing is that even in early trials success, engagement with kind of the key stakeholder groups in particular, both medical oncology but interventional radiology will also be critical to that success. I think that's something that this team, the leadership team here understands very well. I think the second piece is that translating that practice into a broader more immunity-based setting also includes review of protocols and establishment of sort of standard operating procedures that account for potential barriers in that and engaging early in sort of key stakeholder groups, getting input from interventional radiologists, both in the academic center and private practices will really facilitate that transition and understanding what needs to happen and also sort of dealing more proactively with barriers to and concerns around kind of oncolytic virus injections. But overall, I think there is a large community of interventional and procedural radiologists, including diagnostic radiologists who are -- have the skill set, have technology and capability as well to perform these. And then the last thing to a point that was mentioned I think, earlier in this discussion is really some attention to reimbursement strategies and regulatory strategies to ensure that intra-tumoral injections and the reimbursement it goes with them properly reflects the effort level of work, I think, will be critical as well to product adoption.

To clarifying our first potential launch in cutaneous squamous cell carcinoma, we'll be launching in the community and academic centers in time.

In that situation just as Nikhil said, these injections are not particularly complicated even in a community practice that can be done by nurse practitioners are, in fact, that was already happening with T-VEC when they launched as well. So again, we need to make sure there's appropriate training for cutaneous squamous cell to do this safely and effectively, but from a technical perspective. I don't know, Nikhil, if you have any other comments on that?

No, I think you're absolutely right. I think encouraging the allied health professionals to practice a top of license is absolutely critical. So I think investing in their education and investing in techniques of the delivery would be important. But I would preface this by saying that it all comes down to data. And if the data is very compelling and it clearly demonstrates that intralesional plus systemic therapy offers benefit to patients, I think we follow the data. And we say that is how we would modulate or modify our practices if we needed to carve out based on obviously reimbursement as well, carve out a selected time. This could be our time for intralesional office-based injections. That's exactly how we do it right now as opposed to interrupting my day to go and give an injection. That is not very efficient use, but actually carving out time within the day just for these patients, we can easily make that work.

Our next question is on RP1. Initial data for PD-1 failed NMSC. You mentioned that some nonresponders were showing signs of tumor reduction. How many does this apply to? And how much are you expecting the ORR could uptick?

I think it's probably a little too early to comment. I think there's 5 patients with stable disease indicated in that data and a number, more than 1 and less than 5 are in the group of patients who have clearly seen a reduction. But obviously, we just need to wait for the data to mature a little bit to see where they really get to, or I would say is the early impressions of the investigators is that some of their additional patients are clearly doing well and their tumors are beginning to reduce.

Next couple of questions are on future development. What could a registrational study look like in neoadjuvant CSCC? And when could you be in a position to start that trial?

Well, so again, it's a little too early to answer that question in detail. We are in the planning stages of a clinical trial design in the neoadjuvant setting. The intention is to get that going around the end of the year and a number of different trial designs including with different potential control arms and combinations are currently in discussions. So unfortunately, I can't answer in detail at the moment until our planning process is complete.

Thanks, Rob. And another one, do you plan to run registration-enabling studies for MCC, BCC and/or angiosarcoma since RP1 has shown robust efficacy in these tumor types?

So we're still working out what the appropriate route forward to potential commercialization would be in those tumor types. Obviously, one could run actually registrational directed studies in them, although they are relatively rare, at least in the latter 2 tumor types -- types of tumor. So we're also thinking about compendia listing as a route forward and likewise, all will become clear in the fullness of time. But there's 2 potential avenues both could be employed and it's really sort of commercial consideration as to which is the best approach to commercialize in those tumor types. So maybe Sush, can comment further, too.

Yes. Like you said it, Rob, I think it really depends on the size of the opportunity. We haven't done a lot of work on the angio opportunity size, for example, angiosarcoma based on the incidence numbers, [ obviously published ] in a very, very small market that being said, we think a lot of patients come on to our studies, and it makes you wonder if there's more patients out there. So I think we need to do more work on that, and it could be the compendia listing of 20 or so patients could be sufficient.

Could I add something from a clinical perspective for the angiosarcoma team. In my previous life, in addition to cutaneous, I did sarcoma as well. So I saw a lot of angiosarcoma at a referral academic center. And certainly with angiosarcoma that is considered unresectable. Our treatment options are primarily over systemic taxanes which works very well, but it works for a finite period of time. And then almost all of those patients eventually develop either local regional progression or distant metastatic disease. Similarly, anthracycline, so adriamycin or equivalent drugs are utilized in angiosarcoma, but cannot be given to everyone because of the obvious cardiac toxicity and comorbidity. So given that many of these tumors are again on the scale and easily accessible and with at least some of the early data in a rare subtype of tumor that this would qualify for, I think the bar for potential regulatory approval is certainly lower.

I think our observation and Sush suggest that the sort of official numbers may underrepresent the reality of the disease because as Sush said, we do tend to get quite a number coming in, which suggest there must be [indiscernible] more out there than people think in a similar way as it turned out to be many more out there as it were in CSCC once an effective drug became available to treat it.

I don't know if you be meant as well.

Great. Another question with RP1. What's the potential regulatory path in solid organ transplants? And could the ARTACUS study be that registrational study?

So the ARTACUS study was actually designed as to be at a size, which could in a self-support registration in a tumor type where there's not vast number of patients that still a clear unmet need. And we could therefore, with the full 65 patients recruited, choose to actually file a supplemental BLA for that indication. However, the alternative approach is still being considered as well, which again could be compendia listing. So again, it's a sort of commercial consideration as to which approach is appropriate, but we have the flexibility for either based on the current development plan.

Maybe just add we're definitely looking at this market from a commercial perspective and doing some research. It's quite hard. Nikhil said, we'll get an actual handle on what are the opportunities, there's clearly a lot of patients out there, but are they going to be actually treated. This is something we're really thinking through, and it also depends on what organ transplant you're talking about, and these patients can develop cutaneous squamous cell carcinoma at any different time point and maybe like years after their transplant. So it's hard for us to get a sort of exact number today, but we're doing that work. And again, it could be that we go forward with registrational study or again the compendia listing could be sufficient to treat those patient given the unmet need. I don't know Nikhil, if you have anything?

No, I would agree. I think having seen these patients in our center, even though Moffitt Cancer Center does not have its own transplant program. We are part of the University of South Florida and the large teaching hospital, which is Stanford General Hospital does the solid tumor organ transplants there. If you look at the transplant registry data, the number of solid organ transplants on an annual basis continues to increase by roughly about 5% compared to the preceding year. So for example, in 2021, about 41,000 solid organ transplants were performed. The vast majority admittedly kidney care -- kidney transplants. And if you then extrapolate further, about 8% to 9% of those patients over time will develop CSCC and certainly about 15% depending on various registries, both from the United States and the United Kingdom, will develop keratinocyte carcinoma, so including basal cell cancers as well. So that certainly -- and those patients certainly cannot receive or at least upfront should not receive immune checkpoint inhibitor therapy for compromise of the allograft in. So I think that certainly serves as an opportunity.

We will also separately looking at in particularly in one of our investigator initiated studies, other immune suppressed populations as well as solid organ transplant recipients, i.e., patients with autoimmune diseases, HIV, et cetera, to gather data beyond just solid organ transplants.

I think in ARTACUS, where we're talking about sort of expanding beyond some of the different organs, right, that's a discussion we're having on.

Yes.

Great. We've got a question for Dr. Harrington, a bit of a long one, so bear with me. You provided data of responses in skin cancer, slightly more than 1 in 3 for anti-PD-1 failed melanoma and so much similar outcomes demonstrated in anti-PD-1 failed to CSCC. The question is around the roughly 25% of patients that have progressive disease. And are there any characteristics you can speak to of patients that progress?

That's a good question, and I wish I had a really clear good answer to it. The fact is that the patients who respond, and this I think gets a little bit to the question that was asked to Rob Coffin a moment ago about what's happening to the patients currently with stable disease. We often see quite an early response but the actual full resolution of the lesion, and I think it was shown in one of my slides as the tumor regress towards complete remission, it can take many months to achieve that state. For the patients with progressive disease, we tend to see that, that progressive disease is relatively slow but it tends to be relatively early and relentless is our experience with these patients, I'd be interested to hear if others have a similar experience. We don't see a true hyperprogressive response that is sometimes spoken about, certainly in head and neck cancer with anti-PD-1 therapies. We don't see that occurring. But we do tend to see that patients who are responding, we often get an early readout of that often with the patients saying they've got symptomatic benefit and then we see the reduction in the tumor. But for patients where the disease is not responding, we see a relentless progression albeit rather slowly.

Could I slightly add which isn't directly answering the question, but I think an observation is that when patients get through to their certain scan, whether or not they're responding by that point, tend to do really very well in the longer-term with often at the first scan or first assessment, some evidence of increase or [indiscernible] then addressed by the second scan. So while it doesn't tell you which patients were those which were going to get to that second scan, it certainly is the case that if you do get that second scan, you've really got a good chance of the long-term benefit.

Great. Our next question for anti-PD-1 refractory CSCC patients. Is there a spider plot for patient responses over time? Or is it too early in their treatment?

It's too early. It's not big enough and yet for a spider block. And as the table says, there's still 4 patients with stable disease on study that could go either way. So we'll do that when we next update the data, probably late in the year.

Yes. It really is the first thing to cast an early data, which is definitely directionally promising that more patients are among the follow-up there or needed or is it appropriate to call out in those ways, which we set we'll be doing sort of in set. From what we've seen so far, I think it will look very similar to this part of the block, we've already shown in the other skin cancer situations.

Great. Thanks, Rob. The next question is thank you all for the updates in skin cancer. When do you -- when do we get updates on non-skin cancer programs?

Well, we're running a Phase I expansions for RP2, RP3 focusing down on the tumor types but more interested in for Phase II development that we've unveiled today, including GI cancers and head and neck cancer. We don't have an update on those Phase I expansion patients around the end of the year. Obviously, the Phase II program doesn't start until the year-end itself, and therefore, data we pushed out to in early days in 2023 for the Phase II program.

We've got a manufacturing question. And how much more work do you need to do to be ready to file the BLA on time?

Whilst the work is progressing on track, we have completed the comparability work from our Framingham in-house materials been used in our studies and put the report back into the FDA, where we're having further discussions with the FDA over the months to come. And we've also had good clarifying discussions with the FDA on release assays and it's all on track, the bulk of the work, the back has been broken .

Great. Philip, this is a question directed towards you. It's a broader strategic question for the company. The slides noted that you are capitalized to build U.S. commercial infrastructure. And I guess the question is, how are you thinking about current development status for RP1 through 3 and balancing that with potential timing of an OUS partnership?

Well, we have a very broad ambition on cornerstone of IO regimens, which involves going into a number of different cancer types, some of them being quite large cancer types. We've also seen activity from the -- all the way from the anti-PD-1 failed setting. I also have reason to believe there's compelling reason to go to the neoadjuvant setting. So there's also a broad spectrum of potential development. So I think it's kind of intuitive that to extract the most value across all those stages of disease and different disease types at some point, a partner would help us extract more value than continues the default of progressing alone. However, we have as you noted, we have the funds and the plans to build U.S. infrastructure for our initial launches. What that does is allows us to have destiny in our own hands. So quite when the timing of such a development path may take place, the development commercialization path, we're beholden to any particular time line. So in terms of the value creation, we will see at any given point in time depends on the value created, whether or not in terms of a more holistic commercialization and development plan makes sense to the company. As I reiterate, we're in a very strong position, $420 million on the balance sheet that's our destiny in our own hands and the default pathway at the moment is to build our U.S. commercial infrastructure for the launch of CSCC and PD-1 failed melanoma.

We have a next question. Have you looked for the biomarkers in responders versus nonresponders over the treatment with RP1?

Well, so I think really that we don't have yet enough data from enough patients to draw any hard and fast conclusions as to what that were predicted biomarkers that could be for response. We did see some characteristics in responding patients, which are not present in non-responding patients that they're probably after the effects -- post changes rather things you can necessarily see at a baseline. However, I do think to surpass a data set, which will be really rather large when we finish the trial will be an excellent place, which we will be doing to look at a biomarker effects in both responding and non-responding patients in a truly homogenous population where we may then really begin to see something which is predicted. So the answer at the moment is it's too early. We definitely see very positive changes indicative of broad immune activation, but we haven't seen any clear baseline prognostic characteristics, which could suggest which patients respond and which don't. And as you've seen, patients whether or not they have PD-L1 or CD8 or other immune markers at baseline can respond in a sort of a non-correlated way.

Next question is for Dr. Harrington. Can you provide any insight into how toxic chemotherapy is in first-line metastatic HNSCC, and if you think using RP2, 3 plus PD-1 makes sense given that HNSCC virus could provide rapid tumor killing.

So for the first-line treatment of relapsed metastatic head and neck cancer, we know that the standard of care regimens such as the extreme regimen, which was previously the gold standard are associated with very appreciable Grade 3 toxicity rates in excess of 60%, 70% in many studies. So we know that this is a toxic regimen and therefore, patients need to be reasonably robust with a performance status of at least 1, preferably 0 in order to withstand those. We also know that single agent immunotherapy, pembrolizumab for instance, is well tolerated and that the Grade 3 toxicity rates are something like 1/2 to 1/3 of the rates that we see with standard of care chemotherapy. The tricky thing, of course, is to combine both immunotherapy and the immune checkpoint inhibition. And we see that the toxicity looks very similar but with the absence of the cetuximab-related toxicity. So again, in order to deliver cytotoxic chemotherapy, plus an immune checkpoint, patients need to have a strong and a good performance status and be able to withstand the toxicities. Our experience previously, not yet with the RP platform of viruses but with our previous experience around the work within the T-VEC agent is that the addition of a viral therapy to chemotherapy combinations adds little or no additional toxicity. So I would feel that it would be highly likely that for patients who are fit enough to withstand chemotherapy plus IO combinations, the addition of intra-tumoral oncolytic viral therapy with an RP2 and RP3 platform would be entirely sustainable.

The next question is for Dr. Saab. How excited would you and your colleagues be about a novel approach like oncolytic immunotherapy to move the needle in HCC and CRC? And how quickly would something like that enroll based on need and anticipated interest?

I think the approach itself is very exciting. As I mentioned, if you want to talk about HCC. So in HCC, we obviously have hit a plateau although we have moved the needle, we have moved far enough to be -- to essentially have done the job for that right. So the -- an approach that essentially target specifically the liver tumor, which the liver tumors would certainly be desirable, especially in a disease. So let me start with HCC first, a disease that has already shown some level of immune competence, meaning that out of all the therapy options that we've had over the last 20 years, IO seems the one to hold the most promise. So in that sense, oncolytic viral application holds quite a bit of a promise. And there is some precedence not just in preclinical settings, but then certainly early clinical settings that does show that we can enhance that activation. So certainly very exciting, very feasible. As I mentioned, those [indiscernible] are mostly in the liver. In our setting and other settings as well, easily reachable by ultrasound injection. We've done quite a few of these studies in the past and we know they're very feasible. And so certainly, looking at the promising data, both early clinical as well as preclinical. I mean that's a very exciting area in HCC as to colorectal cancer. Now this one has been a bit more challenging. But again, looking at those microsatellite stable colorectal cancer, certainly seems that outside the liver, we are seeing those responses. So immunotherapy, in my viewpoint, remains a promising option for patients as long as we are able to move those liver metastatic lesions from immunosuppressed to essentially immunocompetence. And this is where, again, looking at the totality of the data we have today, and where we may be moving in the near future, it appears that tumor-directed oncolytic immunotherapy may actually reversed that local immune suppressive microenvironment and therefore, bringing those tumors into where we're seeing, the non-liver metastases in colorectal cancer responding. And so in that setting, again, I'm pretty excited about the opportunity. As I said, 80% of those tumors will be present in the liver or 80% of metastatic or rectal cancer will have some level of liver involvement from -- some to extensive liver involvement. And again, reaching those tumors with ultrasound-guided strategies to inject the virus are done pretty routinely on clinical trials in our practice and others as well.

Great. We've got a number of questions regarding RP3. First, how do you compare the patient baseline?

First. We can't hear you unfortunately.

Can you hear me now?

That's perfect.

Great. So we got a number of questions on RP3. First, how do you compare the patient baseline prior line disease lesion for RP3 versus RP2 monotherapy?

Rob, can you make some opening comments.

Yes. I mean these were both standard Phase I population. And to some extent, it's a luck of the draw, sorry, who is enrolled into those populations. However, I don't think it's a very obvious conclusion that the patients enrolled into RP3 did through luck of the draw, happen to have more widespread, both with regard to a number of disease sites and the burden of disease than the Phase I populations enrolled with RP2 monotherapy. There were certainly similar patients in the Phase I monotherapy for RP2, which was 5 out of the 9 who did progress rather rapidly. So yes, by luck of the draw, they had more disease or widespread and a greater burden. And as a result, they're also very heavily pretreated.

Great. And the second one is compared to RP2 monotherapy, how do you view RP3's IO PD marker and abscopal effect?

Really with all of our RP1, RP2 and RP3, we're expecting sort of similar but more of the same with RP2 compared to RP1 and more of the same, again with RP3 compared to RP2. So those should be up regulating the same sort of things in both cases, reversing the immune suppressive effects on tumors, turning cold tumors hot. But in each case, both -- each doing more than the predecessor and therefore with the potential for greater local benefit and also greater systemic i.e., scope of benefit is obviously too early to conclude. Conclusively that we are seeing that but the early data with RP2 certainly suggests we are seeing a higher frequency of responses in patients with PD-1 failed disease that we see with RP1. As the RP3 data catches up, we would expect that we will begin to see promising results there as well. While it is only anecdotal, I do think the slightly more enhanced side effects we see with RP3 as compared to RP2, which likewise has slightly more side effects as compared to our RP1 is relevant in drawing a conclusion that it does seem to be doing what it's intended to do, which is to provide more potent immune activation, which as Philip and - otherwise said, we're exploring in more patients over the course of this year in a more focused way.

And the last one on this RP3 is what is the reason for not going for a higher dose of RP3 monotherapy given that you've not seen any dose-limiting talks?

Kevin can further comment, too. However, we are seeing substantial information. We do see, if we go to higher levels than 10 to the 7 with RP1. So with RP1, we did go up to 10 to the 8 and concluded that there was no benefit based on biomarkers and other effects being seen and also based on the fact there was a considerably greater degree of shedding from the tumor with 10 to the 8 as compared to the 10 to the 7. So with RP1, we've gone up 10 to the 8, we decided the RP2D was appropriately 10 to the 7, and we would expect to see similar, RP2 or RP3, we went up to 10 to the 8. I now benefit based on biomarker but probably increased shedding, which on balance, we believe, means the 10 to the 7 is the most pragmatic and appropriate level to go forward with.

I think, Rob, if we did go to a further log to the 10 to the 8 with the RP3 agent, I think we might begin to see cytokine release syndrome type DLTs. We might actually get into that territory at those sorts of levels from what we're seeing at the 10 to the 7 dosings. I think the important thing to stress with oncolytic viral therapy or oncolytic immunotherapy is that the paradigm of they shall escalate the dose until you can't give any more is not proven and it's not necessarily clear cut. And I think that getting a dose that achieves a biologically meaningful effect, for instance, at the 10 to the 7 level, but is appropriately tolerable and allows you to work around by adding other agents to this without causing overlapping toxicities is really what for me is very attractive about these agents. And I think we can achieve a huge amount with the RP platform without necessarily having us to expose patients to excessive toxicities. So that would be my take on this is that I think we're in the sweet spot without necessarily having to overdo the talks in order to prove that we can do that.

Thanks, Dr. Harrington. And perhaps one more question. Can you expand on the go-forward plan for RP3?

As it relates to kind of RP2 on whether we're going to [indiscernible] RP2, 3 can you take that Rob?

Yes. Probably it wasn't in the slides -- on the slide, I did discuss that the intention is to keep our offering open for the moment as the RP3 data catches up. So Phase III protocols, which are being written are written to have groups or cohorts for both RP2 and RP3. And then the intention is to open one or the other based on availability of data at the time we begin to open those trials, which will be towards the end of the year. So we are anticipating over the course of this year, the data with RP2 -- RP3, sorry, will begin to catch up. Therefore, we'll have a greater level of information upon which to make the decision. But it sounds that we've not slowed down, as I said, the protocol has been written with both to maintain flexibility up until the last moment and also potentially to provide the opportunity to test both and at least maybe one of the protocols if we think that would be advisable based on what we know when we need to make the decision. So we are able to say today whether it would be RP2 or RP3 in any particular protocol, but we will be saying when we start those protocols towards the end -- towards the end of the year as the data catches up.

Thanks, Rob. We do have other questions but we are at a lot of times. We will do our best to follow up and answer the remaining questions. But Philip, I'll turn it back over to you.

Yes, obviously happy to answer questions individually after the event. We are past top of the hour, so just some quick thank you. Thank you to the key contributors. Thank you to all those that supported the event on the outside. And a very special thank you to the leading positions that have joined us for the session, including with me here today are Dr. Khushalani.

Thank you.

Dr. Ahmed, and virtually, Dr. Harrington and Dr. Saab, very much appreciated. And with that, we look forward to giving you further data updates towards the end of the year as the year progresses. And I wish everyone a good rest of the day. Thank you for joining.