Mersana Therapeutics, Inc. (0M40.F) Earnings Call Transcript & Summary

Welcome everyone to the Annual Crédit Suisse Healthcare Conference. We're joined today by Anna Protopapas, CEO of Mersana Therapeutics. And she's going to give us a presentation on all the recent developments for Mersana. You can take it away, Anna.

Thank you, Tiago, and thank you for having us today and for the opportunity to tell you more about Mersana and why we're so excited about the potential of our current clinical programs and robust pipeline of ADC candidates. Before I start, I do want to remind you that I will be making forward-looking statements. Mersana was founded on the vision that innovative new ADC platforms that address the limitations of the first generation approaches could lead to therapies that could significantly improve outcomes for cancer patients in need. These platforms are proprietary DolaLock payload with controlled bystander effect; our Dolaflexin platform carrying the DolaLock payload with a high drug-to-antibody ratio; our Dolasynthen platform also carrying the DolaLock payload but with a precise homogeneous and customizable drug-to-antibody ratio; and our Immunosynthen platform for targeted immune stimulation. All of these platforms represent significant advancement in the field. I will not go through the details of these platforms in this presentation. But you will find much of the data associated with the performance of these platforms on the publications' section of our website. Importantly, we have leveraged these platforms to build a very innovative pipeline of first-in-class therapeutics and are rapidly bringing this pipeline to patients. I will focus today's presentation primarily on 1536, our lead asset, a first-in-class Dolaflexin ADC targeting NaPi2b. We have demonstrated proof of concept of XMT-1536 in heavily pretreated ovarian cancer patients and proof of activity in lung adenocarcinoma as well as a favorable safety profile without the severe neutropenia, neuropathy or ocular toxicity observed with other ADC platforms. The data not only demonstrate the potential of 1536 as a therapeutic, but also demonstrate the potential of a highly differentiated platform to deliver subsequent programs with improved therapeutic benefit to patients. 1592 is our second ADC, targeting NaPi2b and was created using our Dolasynthen ADC platform. Preclinically, XMT-1592 has shown a differentiated profile, particularly in lung cancer. We now look to validate the potential of XMT-1592 in the clinic and have initiated patient dosing in a Phase I dose escalation study. B7-H4, our next DolaLock ADC is in IND-enabling studies, another first-in-class molecule addressing areas of high unmet medical need. Lastly, our Immunosynthen platform has been validated across multiple targets, and we're moving our first Immunosynthen ADC to IND-enabling studies. With a promising lead asset, a first-in-class pipeline, innovative and differentiated platforms, all of these position Mersana well to achieve our vision of bringing life-changing ADCs to cancer patients in need. This year has proven transforming for Mersana. Next year could be just as transforming with multiple important milestones. For XMT-1536 in ovarian cancer with compelling proof-of-concept data in hand, we are moving full speed ahead and are planning for an FDA interaction and the initiation of a registration-enabling study as well as life cycle management studies. In lung cancer, we have observed activity, as I mentioned, and are actively enrolling patients with a focus towards establishing proof of concept. In parallel, we're also dose escalating 1592 to gain a better understanding of its profile and the potential to translate the preclinical differentiation into the clinic. The decision as to which molecule we take forward in lung adenocarcinoma will be informed by both data sets and is expected in 2021. We're also keenly focused on advancing two new and exciting programs towards IND: Our DolaLock ADC candidate targeting B7-H4 and our first Immunosynthen ADC. We have the financial resources and the team to execute on this plan to advance our pipeline and demonstrate the potential to benefit patients in need. We'll begin with XMT-1536, our first-in-class Dolaflexin ADC, targeting NaPi2b. Let me start by describing the target. NaPi2b is an excellent ADC target broadly expressed in ovarian cancer and lung adenocarcinoma, but with limited expression in healthy tissues. This makes it a great target for an ADC. NaPi2b is a lineage marker that is responsible for transporting phosphate across the cell. Interestingly, in lung adenocarcinoma, there's a correlation between expression of NaPi2b in EGFR and KRAS mutations. NaPi2b was first described as an ADC target by Genentech, who pursued in using the vcMMAE platform from Seagen, a program that provided a target validation but was terminated for platform-related toxicities. In preclinical studies, highlights of which are shown in this graph, we have demonstrated a strong relationship between response and H-score, H-score being a measure of biomarker expression. We have developed a proprietary and robust assay to measure H-score in clinical trials and are in the process of validating with a commercial vendor, a companion diagnostic in anticipation of potential patient selection strategies in the upcoming registration-enabling studies. Before getting into the exciting data on XMT-1536, I would like to outline the current treatment landscape for ovarian cancer patients. The unmet medical need is significant in ovarian cancer, with platinum therapy remaining the backbone of treatment in earlier stages of the disease. Recently, bevacizumab and PARP inhibitors have been introduced and are being increasingly used in earlier lines of therapy. Unfortunately, patients inevitably become resistant to platinum therapy and having previously received bevacizumab or PARP at that point, their options are limited to single agent chemotherapy, such as pegylated doxorubicin and topotecan. These agents have limited efficacy. In platinum-resistant patients with 1 to 3 prior lines, the response rate is between 4% and 12%, and the PFS is about 3 to 4 months with a median survival of less than a year. We are actually studying an even more heavily pretreated population in the expansion portion of the study that includes patients with 4 lines of therapy and patients who have more bev and PARP pretreatment than historical studies. This is a patient group with high unmet medical need as demonstrated by the fact that the FDA has granted us Fast Track designation. I'll start with a brief review of the results of the dose escalation portion of the Phase I study which we presented at the Society of Gynecologic Oncology in March of this year. This data showed that 1536 is well tolerated without the severe toxicities of other ADC platforms, such as neutropenia, peripheral neuropathy or ocular toxicity. The most common treatment-related events were Grade 1 and 2 nausea, fatigue, headache; and the most frequent Grade 3 treatment-related AE was transient AST elevation without associated changes in bilirubin or cases of Hy's law. We now understand this lab value to be related to inhibition of AST clearance by specialized macrophages. This is an effect noted for other therapies and is not associated with hepatic injury. In this portion of the study, we determined the MTD to be 43 milligrams per meter square. These patients had a median of 5 prior lines of therapy, and in most cases had run out of other treatment options. Longitudinal studies have shown that as ovarian cancer patients progress, the potential for a response decreases rapidly with line of therapy, and that the expected response rate is close to 0 for these ovarian cancer patients. The fact that we saw multiple confirmed partial responses and durable stable disease in this patient population is very exciting. We also saw our first confirmed partial response in a lung adenocarcinoma patients. In addition, a favorable biomarker response trend was observed. For a subset of the valuable patients treated at 30 milligrams per meter square or above who had higher NaPi2b expression, 5 out of 15 or 33% achieved partial responses, and 6 out of 15 or 40% of patients achieved stable disease for an overall disease control rate of 73%. In contrast, there were no responders in the no-NaPi2b expressing group. This is important because it indicates the potential for a patient selection strategy which can identify patients more likely to benefit from XMT-1536. Turning now to the expansion portion of the Phase I study presented first at ASCO in May and then updated at ESMO into -- in September. The safety profile is consistent with previously presented data at ASCO, and in the dose escalation. XMT-1536 was generally well tolerated without the severe toxicities observed with other antibody-drug conjugates such as neutropenia, peripheral neuropathy or ocular toxicity. The most frequently reported treatment-related AEs included generally Grade 1 and 2 fatigue, nausea, decreased appetite, vomiting and transient AST elevation, which recovers by the next dose. No changes in bilirubin or Hy's law cases were observed. And these transient AST elevations are easily monitorable with standard lab tests. Discontinuations for treatment-related AEs were rare. 29 were evaluable for resist response with at least one scan. Encouraging antitumor activity was observed in patients with ovarian cancer with an overall response rate of 34% and a DCR of 79%. Responses, including complete responses were observed in those failing platinum and previously treated with bevacizumab and PARP inhibitors. It is very uncommon to see complete responses in this ovarian cancer patient population. Median duration of response has not yet been reached in patients with higher NaPi2b expression. As a reminder, NaPi2b expression is being evaluated by immunohistochemistry in the archival as well as fresh tissue where available, and a correlation between biomarker expression in clinical responses are being explored as part of a companion diagnostic development. This data continue to support the potential for NaPi2b biomarker-based patient selection strategy. And we will define the threshold and the patient selection strategy once we have the complete data set. Slide 10 outlines XMT-1536 favorable safety profile -- as I mentioned in the summary slide, the profile without severe neutropenia, neuropathy or ocular toxicity. This safety -- this differentiated safety profile is important because it provides the opportunity to combine with platinum, still the backbone of ovarian cancer treatment in the earlier lines, and with checkpoint inhibitors. These are combinations we will seek to explore in 2021. This is also -- our tolerability profile is also a key differentiator from the now terminated Genentech program, which was shown to not be combinable with platinum because of overlapping neutropenia. Turning now to the safety and efficacy data. This data continued to show that XMT-1536 is an active drug in ovarian cancer. Of the 47 patients with ovarian cancer 29 were evaluable for resist response with at least one scan. Given the accelerated speed of recruitment, we have 15 patients that are non-evaluable because they did not yet harvest this assessment at the time of the data cut. Of the 29 evaluable patients, we observed 2 complete responses and a partial response for an objective response rate of 34% and a disease control rate of 79%, very consistent with the data reported at ASCO. As I mentioned, complete responses are very uncommon in disease -- in this disease setting, and both the patients with complete responses had prior treatment with -- were platinum-resistant and had prior treatment with both bevacizumab and PARP Inhibitors. The responses were observed in platinum-resistant and platinum-refractory ovarian cancer. The majority of patients, actually over 70% of patients have had bevacizumab and a large number have had PARP inhibitors as well. Responses were observed within 2 cycles of treatment in 70% of patients and within 4 cycles of treatment in 100% of patients. This data continued to demonstrate promising activity and the potential of XMT-1536 to provide a meaningful clinical benefit for patients. For consistency purposes, we bifurcated the data into higher and lower NaPi2b expression based on the lowest NaPi2b expression score at which we observed a response in the dose escalation. We will define the patient selection strategy once we have completed the data set. Moving on to the waterfall plot. The waterfall plot shows the changes in tumor dimensions of target lesions for the patients, as characterized by Napi2b expression, and those were expression is not yet determined. As you can see, we have seen deep responses and even though patients who had stable disease represented in many cases tumor shrinkage. The emerging profile of XMT-1536 continues to show important activity with these deep responses. On the spider plot, we observed multiple patients who have shown deepening response over time, including several patients with stable disease who experienced to are shrinking in their target lesions. Moving on to the swimmers plot. The swimmers plot is broken up -- out into higher, lower and not yet determined NaPi2b expression. The data is non mature as 50% of patients remain on study and several have not yet received resist assessment. Importantly, one of the 2 complete responses observed in 2 patients with platinum-resistant ovarian cancer disclosed at ASCO, one patient remains ongoing on study at week 42. Interestingly, the other complete response patient who came off study around week 19 from tolerability and logistics issues associated with moving out-of-state due to the COVID-19 lockdown, were still in complete response when we scanned at day 28 without any additional intervening therapy. Turning to the next steps for 1536. In ovarian cancer, we have exceeded our goal of enrolling approximately 40 to 45 patients in ovarian cancer expansion cohort. And we'll continue to enroll patients in the ongoing expansion study until the launch of the anticipated registration-enabling study. We plan to provide an additional update on XMT-1536 in ovarian cancer at an Investor and Analyst Day around the end of the year. Given the significantly larger number of patients and longer follow-up, this update will include more details on our plans for a registration-enabling study involved -- informed by planned FDA interactions and our life cycle management studies for XMT-1536. As mentioned earlier, the current standard of care in the platinum-resistant ovarian cancer setting is single agent chemotherapy with an established response rate of 4% to 12%, median PFS of 3 to 4 months and a survival rate of less than 12 months. This is in patients with up to 3 lines of therapy. We've observed robust and consistent activity in patients that are even more heavily pretreated than the data I just -- in the studies I just described. In lung cancer expansion cohort, our objective is to enroll 40 to 45 patients. We plan to discuss our data disclosure plans for this cohort when we lay out our goals and milestones for 2021 in early January. In lung cancer, the standard of care following progression on checkpoint inhibitors and platinum-based chemotherapy or failure of targeted therapy for patients with tumors harboring oncogenic driver mutations is docetaxel alone or in combination with platinum therapy. The standard of care has an overall response rate of 14% to 23% and a median response rate of 3 to 4 months. So again, a lot of opportunity to improve upon the standard of care with XMT-1536. I will now briefly turn to XMT-1592, our ADC product candidate, which uses our new platform Dolasynthen. Preclinically, XMT-1592 has shown a differentiated profile, particularly in lung cancer, that if validated clinically, can extend our leadership in the NaPi2b area. In lung tumor models, XMT-1592 has fourfold more efficacious than XMT-1536, consistent with increased exposure to the DolaLock payload in the tumor. Our primary motivation for bringing XMT-1592 forward is increasing our short-term goal in lung. Although we have seen some activity in lung, we have more work to do to better define the profile of the drug and the path to the market. This gives us an opportunity to evaluate XMT-1592 in parallel. If clinically, XMT-1592 proves to have such a significant improvement in efficacy over the signals we have already observed with XMT-1536, that would be very exciting. So this is another shot at goal for this important indication with a high unmet medical need. We started dose escalation with 1592 in May, and we are -- we believe we can move through dose escalation fairly rapidly, given our deep understanding of the patients, the -- and the expression of the antigen, and we will be disclosing data on the dose escalation in 2021. I would now briefly turn to some of our earlier pipeline. Our next DolaLock ADC will be an ADC targeting B7-H4, another first-in-class ADC, which is now in IND-enabling studies. B7-H4 is an excellent ADC target. It has a unique expression profile. It is expressed on tumor cells, but it is also expressed on immunosuppressive tumor-associated macrophages. This represents an ideal expression profile for a DolaLock ADC. We often focus on the anti-tubulin cytotoxic mechanism of our DolaLock payload but we have also demonstrated that our payload causes immunogenic cell death and can activate dendritic cells and contribute to a secondary immune response in the tumor. Targeting B7-H4 on tumor cells provides a direct cytotoxic effect, while targeting B7-H4 on tumor-associated macrophages can lead to catabolism of the ADC and release of the DolaLock payload in the tumor environment, where it can contribute to bystander killing as well as additionally further contribute to the secondary new activation. We have compelling preclinical data. We are in IND-enabling studies, and we will be disclosing our data around year-end at the Analyst and Investor Meeting. And finally, a very brief overview of our Immunosynthen platform. Being able to activate the innate immune system in a controlled targeted manner without the systemic liabilities has the potential to be a huge leap forward in immuno-oncology. We have leveraged our significant expertise in ADC to design our Immunosynthen platform from the ground up, taking into account the optimal payload linker and drug-to-antibody ratio specifically for this application. We have generated a compelling data set, validating our approach across multiple targets and are on track to select our first Immunosynthen ADC to move into our IND-enabling studies. Our Immunosynthen platform has devastated clear efficacy using a variety of targets across different models. As shown here, a single injection at a low dose is sufficient to induce deep and durable regressions across several targets, a much greater effect than the systemically-administered STING agonist currently in clinical development, even at a 50-fold higher dose. Consistent with an immune response, rechallenged or cured mice through reimplantation of tumor cells shown here for Target 3, demonstrate that the tumor does not regrow, consistent with the immunological memory of the tumor. We are excited about the significant progress to date for our Immunosynthen ADC platform, employing a variety of targets. In addition, we are presenting some very exciting data today at SITC that further characterize what we believe are significant advantages of our STING ADCs relative to other innate immune activating approaches, and we are hosting a webinar on Monday, November 16 at 8:00 a.m. to provide a comprehensive overview of the Immunosynthen STING agonist ADC platform and communicate the data behind our first Immunosynthen ADC we have selected to move into IND-enabling studies. Lastly, I will summarize our goals and milestones for this year. We have reached 3 important data disclosures on 1536. As I mentioned, we are planning to provide an additional data disclosure in ovarian cancer at an Analyst and Investor Day around the end of the year. On 1592, we achieved our goal of initiating patient dosing in the dose escalation portion of the study. We continue to work to achieve rapid dose escalation throughout the remainder of the year, and we'll discuss our plans to disclose dose escalation data when we lay out our goals and milestones for 2021 in early January. We also continue to make progress in advancing B7-H4 and our first STING agonist ADC candidate. And as I mentioned, we'll be disclosing data and our first candidate on STING next Monday and our [ DolaLock ] B7-H4 at the Analyst and Investor Day around year-end. We will continue to maintain a strong foundation with top talent. Our people are our biggest asset and leverage strategic partnerships to continue to build value. It's been a very productive year for us, and we look forward to updating you on our continued progress. Thank you very much. With that, I will conclude my presentation.