Cardiff Oncology, Inc. (CRDF) Earnings Call Transcript & Summary

Good day, and welcome to the Cardiff Oncology KOL call on KRAS in clinical practice for colorectal cancer. [Operator Instructions] As a reminder, this conference is being recorded, and a replay will be made available on the Cardiff Oncology website following the conclusion of the event. I would now like to turn the call over to your host, Dr. Mark Erlander, Chief Executive Officer at Cardiff Oncology. Please go ahead, sir.

Thank you, operator, and good morning to all of you that have joined us today for today's key opinion leader event. Before I begin, I'd like to thank LifeSci Advisors for putting this event together as well as our key opinion leaders, Dr. Afsaneh Barzi and Dr. Heinz-Josef Lenz, for their time. Today's event will focus on KRAS in colorectal cancer, beginning with a discussion by Dr. Barzi on the history of KRAS in clinical practice, the challenges of drug development and effectively targeting KRAS mutations and the current unmet need for second-line metastatic colorectal cancer treatment. Following Dr. Barzi's presentation, Dr. Lenz will review the recently presented ESMO data from our ongoing Phase Ib/II trial evaluating Onvansertib in combination with FOLFIRI and bevacizumab for the second-line treatment of KRAS-mutated metastatic colorectal cancer. Next slide, please. Let me remind everyone that during the presentations and discussions to follow, forward-looking statements may be made as disclosed on this slide. Next slide, please. I'd like to go first, just introduce you to Cardiff at a glance. We are a clinical stage biotech company that's developing Onvansertib, which is an oral highly selective PLK1 inhibitor. We are on the NASDAQ Exchange as CRDF. As of June 30 of this year, we reported $30.5 million in cash and clinical trial funding. Our quarterly cash burn is about $4 million per quarter, and we are located in sunny San Diego, California. Next slide, please. So before I turn the podium over to our distinguished key opinion leaders and also to really just set the stage, I would like to give you a brief overview of Cardiff Oncology and our investigational drug, Onvansertib. We licensed Onvansertib from Nerviano Medical Sciences, which is a lead developer of kinase inhibitors, in March of 2017, and we quickly moved into the clinic with filing 3 INDs and initiating 3 trials. What makes -- we believe what makes Cardiff Oncology an attractive investment opportunity are 4 key pillars. First, Onvansertib is a third-generation first-in-class oral PLK1 inhibitor, which is a proven therapeutic target and overexpressed in most cancer types. Second, we have a strong lead program in KRAS-mutated colorectal cancer, which represents a significant unmet medical need for new effective treatment options. And to this end, at the end of May, the FDA granted fast track designation for our clinical development program. And shortly thereafter, in mid-June, we initiated an expanded access or what used to be called compassionate use program, so that patients who are not eligible for our clinical trial but who perhaps could benefit clinically from treatment of Onvansertib could have access. This program has been met with considerable enthusiasm from oncologists and patients across the country whose treatment options are very limited. Third, we have integrated a biomarker strategy into all of our clinical programs, and we believe that we may be able to identify patients who are most likely to respond to treatment with Onvansertib across our different indications. And finally, we have a diversified pipeline with this one molecule, Onvansertib, and the opportunity to address a number of different solid tumor and hematological cancers. Next slide, please. So let me just dive a little bit about Onvansertib, so that would be the next slide. So PLK1, what is it? It's an enzyme. It's a serine/threonine kinase. It's a master regulator of the cell cycle progression. As I mentioned earlier, it's also overexpressed in tumor types -- tumors versus normal cells. Tumor cells are, in essence, hijacked PLK1 pathway. And because of that, tumor cells are particularly vulnerable when you inhibit PLK1 and induces cell death, the tumor cell death. It's also interesting that there's emerging data demonstrating that PLK1 also is a key regulator of other cellular functions beyond mitosis that are essential for tumor growth, and those include the biosynthesis of DNA as well as the DNA damage response. Over to the right is a cartoon of the cell cycle and really showing you where Onvansertib when it inhibits PLK1 that causes that G2/M arrest where they no longer can go on and divide. The next slide, please. On the left, you'll see Onvansertib and really some of the key properties of Onvansertib. We've already talked a little bit about the fact that it's highly selective for PLK1, which is, we think, is extremely important. We also in all 3 -- across all 3 of our clinical trials and with over 100 patients enrolled in these trials, we have demonstrated that this drug is well tolerated and is safe. We also -- another key thing about Onvansertib is that it is oral. And also, it has a 24-hour half-life, which is very different from previous generation PLK inhibitors, which had, for example, half-lives of 5.5 days, which made them very unwieldly in the clinic. And because of that, we have flexible dosing and scheduling and also ideal pharmacokinetics. Finally, we also have synergy with many different FDA-approved targeted drugs as well as chemotherapies. Today, we'll be focused on 2 on the right-hand side there, 5FU and irinotecan, which are both used in treatment of second-line metastatic colorectal cancer. Next slide, please. And with that, I would like to now introduce our key opinion leaders. First, Dr. Barzi, who comes from City of Hope Comprehensive Cancer Center, she is the Associate Clinical Professor of Medical Oncology, GI Oncology and Clinical Director of Access Hope. And also our PI, Dr. Lenz, who comes from the USC Norris Comprehensive Cancer Center, he is Professor of Medicine and Preventative Medicine, also the Associate Director of Clinical Research, the J. Terrence Lanni Chair in Cancer Research and also the Co-Director of the USC Center for Molecular Pathways & Drug Discovery. With that, Dr. Barzi, take it away. You're on mute. You're on mute, Dr. Barzi.

Sorry. I want to thank everybody for joining the meeting. And Mark, thank you so much for the introduction. Over the next few minutes, I will be discussing KRAS-mutant metastatic colorectal cancer, a unique entity. Next slide, please. Now looking at the epidemiology of colorectal cancer. Colorectal cancer is the third most commonly diagnosed cancer in females and males in the United States. There are about 150,000 new cases per year and it remains the third cause of cancer-related death in the U.S., with 50,000 deaths per year. At diagnosis, 20% of the patients have metastatic disease. And of those without metastatic disease, 30% develop metastatic disease during the course of their cancer. Therefore, annually, an estimated 50,000 to 60,000 patients are new to treatment for CRC. Patients with metastatic disease have a poor outcome with a 5-year survival rate of 14% compared to a survival rate of 90% for localized disease. Next slide, please. RAS mutation is common in cancer. RAS is a protein that's upfront in signal transduction in cells and basically involved in regulation of differentiation, growth, migration and apoptosis. In 1982, mutationally activated RAS were detected in human cancer, making it the first discovery of mutated genes in cancer. Beyond the long history, RAS gene represents one of the most frequently mutated oncogene in human cancers with high prevalence among pancreatic cancer, colorectal cancer and lung cancer. RAS protein cycles between GTP-bound activated state and GDP-bound inactive state, and this is regulated by RAS GAPs and RAS GEFs. And when RAS is mutated, it basically stays persistently in GTP-bound activated state, which results in activation of downstream effectors and oncogenic pathways. Next slide, please. KRAS is critical in the development of colorectal cancer. In the model of chromosomal instability as the biology of colorectal cancer, it is believed that the development of cancer is a gradual transition from normal epithelium to adenoma and eventually to carcinoma. During this process, multiple genomic alteration can happen that allow the formation of the cancer and KRAS mutation is an early and very frequent event in this development. Next slide, please. Therefore, KRAS mutation is common and frequent in colorectal cancer, with roughly 45% of the colorectal cancers harboring KRAS mutation, and there are multiple different variants of KRAS mutation, 3 most common in colorectal cancer are G12D, G12V and G13D. Next slide, please. Now if RAS is so frequent and we've known about it so long, how come we have not been able to target it effectively? To date, no effective RAS inhibitor has been approved. And it's considered an undruggable target. Part of this reason is the affinity for the RAS to GTP is very high, and there is an abundance of GTP in the cell and therefore, disrupting the activation of RAS has been an impossible task. The exception is G12C that currently has inhibitors in clinical development. However, G12C represents only 8% of the KRAS mutation in CRC. So if we cannot directly target RAS, we should look at alternative strategies to inhibit the activated pathway. There are 2 logical ways to think about pathway inhibition. One is to look at the inhibitors of the downstream effect from RAS such as PI3K, AKT, MEK. This has been tried in clinical trials. And combination of these agents have been used with very limited clinical activity and significant toxicity that had resulted in abandoning this practice. Alternatively, we can look at the RAS upstream pathway and development of farnesyl transferase inhibitors were actually intended to disrupt the KRAS plasma membrane localization, but unfortunately, that also had to be abandoned because of lack of clinical activity. There are other ways to target RAS pathway, including metabolic pathways. The KRAS mutated are dependent on glutamine for metabolism, and there are currently efforts to develop drugs to target this pathway. Last but not least, it's looking at synthetic lethality and synthetic lethal partners for mutated KRAS. Next slide, please. Now synthetic lethality and development of drugs on this concept is not new to the cancer drug development. PARP inhibitors for BRCA mutation are based on this synthetic lethality. What it is, is basically, it is the concept that when a gene is mutated, it may require partnership with another gene to perform its function. And if the partner gene were to be removed from this equation, then the mutated gene becomes basically nonfunctional. In the graph that's shown, if you look at the RAS wild-type and look at the partner gene X, because the RAS wild-type is active, even if you remove the gene X, the cell remains viable. However, when RAS is mutated, removing the gene X results in cell death and this lack of viability of the cell. Next slide, please. Searching for synthetic lethality genes in KRAS-mutated CRC identified PLK as a potential target. How was it done? A genome-wide RNA screen was performed in KRAS-mutant and wild-type CRC cell lines to identify synthetic lethal partners for KRAS mutant. PLK1 were identified to have synthetic lethality with KRAS mutant, meaning that KRAS-mutant cells were hypersensitive to PLK1 inhibition. Now what do we know about Onvansertib? We know that it induced a stronger cell growth inhibition and mitotic arrest in KRAS-mutant cells compared to the wild-type cells, as shown in the graphs to the right of this slide. Next slide, please. Now let's look at the interaction between KRAS and PLK1. So when KRAS is targeting the mitosis, it basically collaborates with CRAF to localize to the myotonic spindle pole and that is where PLK1 plays a critical role to basically permit the mitotic progression, proliferation and survival. So if one were to basically inhibit the PLK1, in this case, through use of Onvansertib, this process of progression in mitotic -- in mitosis and proliferation of cells would come to a halt. Next slide. Thus far, I discussed the epidemiology of colorectal cancer, the relevance of RAS and the concept of synthetic lethality. But what is the meaning of KRAS in clinical practice and how KRAS had shaped clinical practice for colorectal cancer? KRAS is a pivotal diagnostic biomarker in colorectal cancer. And this goes back to the development of EGFR antibodies, which initially were developed for all colorectal cancer. However, shortly after the development and FDA approval, it was discovered that KRAS mutation is actually a predictor of benefit of these agents. In KRAS-mutated patients, no improvement in overall survival, progression-free survival and response rate was observed when EGFR antibodies be used. And in fact, mutation of KRAS was identified as the most frequent mechanism of resistance of this class of drugs. This resulted in shaping the clinical practice but also forced the FDA to go back and change the approval of these antibodies and limit their use to only KRAS wild-type subset of patients. Now how has it shaped the clinical practice of metastatic colorectal cancers? At least in the U.S., the most common agents to be used in the first-line setting in metastatic colorectal cancer is FOLFOX and bevacizumab. However, in the second-line setting, patients with KRAS wild-type have the option to use chemotherapy plus EGFR antibodies. While for KRAS-mutated patients, the option is limited to chemotherapy plus anti-angiogenic therapy. Now the question is what agents are commonly used in second-line setting and how effective they are. Next slide, please. In a large real-world evaluation of the patterns of care in metastatic colorectal cancer using Flatiron Health data, 14,000 with metastatic CRC patients were identified. Among those, 7,000 received second-line therapy. The most common second-line therapy is FOLFIRI plus bevacizumab followed by FOLFIRI and looking at other anti-angiogenic therapy, including Ziv-aflibercept and ramucirumab, there is a significantly lower rate of utilization of these agents. Although we don't have any head-to-head comparison of effectiveness of bevacizumab with other anti-angiogenic agents, including Ziv-aflibercept and ramucirumab, a poster that's presented in ESMO this year suggest that bevacizumab may actually be the best partner with FOLFIRI in treatment of these patients. The poster is a presentation on a study that's called [ Deslico ] and shows that combination of FOLFIRI plus Ziv-aflibercept has a significantly lower rate of overall survival and progression-free survival in the second-line therapy in patients with metastatic colorectal cancer. Next slide, please. Now what do we know about the efficacy of second-line therapy in metastatic colorectal cancer? Generally speaking, outcomes of patients in the second-line setting is poor. Efficacy of FOLFIRI in second-line setting was established many years ago, but it was shown to have a 4% response rate and 2.5 months progression-free survival. Addition of bevacizumab to FOLFIRI improves the outcomes. However, when looked at in the subset of patients with KRAS wild-type versus KRAS-mutated patients, while both groups had improvement in the progression-free survival, the improvement in overall survival by addition of bevacizumab was limited to the subset of patients with KRAS wild type. Here is when we see the divide in prognosis of patients with KRAS-mutated subset. Next slide, please. We generally look at the response rate as the first signal for effectiveness of an agent. And as shown in the previous slide, the response rate to FOLFIRI plus bevacizumab is around 5%. The response rate with other anti-angiogenic therapies, including aflibercept and ramucirumab are reportedly higher. However, these trials have differences in terms of patient population that they enroll, which may explain this higher response rate. Next slide, please. In conclusion, development of treatment for KRAS-mutated CRC patient is an area of unmet need. I presented data that shows that KRAS-mutated CRC represents a large and diverse population with nearly half of CRC patients harboring KRAS mutation. Efforts to develop therapies targeting RAS have been unsuccessful and newer strategies, such as synthetic lethality, provide an opportunity to advance the care in this patient population. Second-line treatment for CRC patients have generally very poor outcome. And current data suggests that KRAS mutation is associated with poor overall survival and perhaps progression-free survival. There is paucity of focused study for KRAS-mutant population, and given that this is a very unique entity, it is time to focus trials, specifically for this patient population. With this introduction, I'll hand over the microphone to my colleague, Dr. Lenz, to discuss the trial of Onvansertib in KRAS-mutated patients in second-line setting. Thank you.

Thank you so much, Dr. Barzi, and thank you so much, Dr. Erlander. I think you set an incredible stage in identifying that KRAS-mutant patients need other treatment options. It's a significant clinical unmet need. And that PLK1 is a very attractive new target for patients with mutant RAS. So I'm going to present the clinical data on the Phase I, Phase II Onvansertib in combination with FOLFIRI and bevacizumab in second line for metastatic colon cancer with the KRAS mutation. These data were just presented at ESMO last weekend. Next slide. So the rationale was actually very nicely laid out by Mark and Afsaneh showing that the KRAS-mutant cells are very sensitive to inhibition of Polo Kinase 1, through the extensive preclinical data with synthetic lethality. Now very importantly, in preclinical models using xenograft mouse models, Onvansertib shows significant synergy with drugs we use in second line, such as Irinotecan and 5-FU as well as bevacizumab. So the clinical design would be very rationally developed to combine Onvansertib with FOLFIRI and bevacizumab. You can see in the bottom of the slide very nicely the synergisms of the active cytotoxic agents we use in standard of care for this patient population. Next slide. I think it's very important, and Afsaneh made that very clear. The response rate in second-line are very disappointing. They hover around 10%, most of them under 10%. And KRAS mutant do overall works with MTBHS strategies compared to wild type. The treatment schedule was based on a 28-day cycle, 2 treatments with FOLFIRI and 5 days in the first 14 days, patients received the oral Onvansertib. The primary end point in the Phase Ib was the characterization of dose-limiting toxicity, adverse events and tolerability. In the Phase II, we would like to see the objective response rate in patients who failed FOLFOX/bevacizumab treatment. Secondary end points included progression-free survival, and as Mark mentioned, the very extensive biomarker evaluation to look at the mutant allele frequency over time under treatment pressure. Next slide. You can see we have so far involved 13 patients in the Phase Ib cohort, 6 patients at 12 milligrams per square meter, 3 patients at 15 and 4 patients with 18, and we are just working up 5 -- patients 5 and 6. The median age is 62. We have predominant males with 8 versus 5. Patients have an ECOG performance of 0 and 1. The colon primary site is usually colon, but we have also some rectums. Many patients have liver metastasis, 46%; liver only 15%. The number of metastatic sites involved is more than 2 or equal in the majority of the patients. Patient had 70% prior bevacizumab treatment. Next one. The clinical trial showed also here the safety and tolerability of Onvansertib. It is actually extremely well tolerated. The most toxicity we see when you look in the Grade 3, Grade 4 column is neutropenia. Now I have to make very clear because of FDA, we needed to do and include the standard FOLFIRI regimen, which includes 5-FU bolus. Everyone knows 5-FU bolus creates significant Grade 3 and Grade 4 neutropenia. So it's very difficult to really identify whether this neutropenia we see is actually related to Onvansertib. And Dr. Barzi and I agree that in most cases, this is actually the 5-FU bolus because when we take it away, the neutropenia did not recur. So we know indirectly that it was caused by 5-FU bolus regimen. But you can see that otherwise, it's very, very well tolerated with Grade 3 with fatigue and as well as one increase of a liver enzyme, which could be also a progression of disease. And the abdominal pain more related to the disease than to the drug, very consistent with the other ongoing clinical trials for this particular drug. Next slide. Here are the first glimpse of the efficacy. Now you have heard from Afsaneh, response rate may be around 10%, probably lower. PFS may be between 2 and 5 months in the best case scenario. When you look here, when you go under the 24 weeks, this is the one 6-month interval. You can see that most of the patients reach 6 months, so much longer than we would see from similar treatment. However, you understand this is a small sample size, but giving us some hope that this combination has some promising efficacy. Some of these patients are still ongoing, when you look at the first 2 and the last 2, which are already beyond 6 months. We had only 2 progression of disease in this patient population. So we are very excited about the potential efficacy of these combinations because 73% showed a durable response beyond 6 months. And 4 of them are still on treatment. Next slide. You can see here the waterfall plot very clearly the extent of tumor shrinkage in this patient population. With the 5 responders, 4 of them were confirmed. And even the patients with stable disease had some tumor shrinkage, and some of them are on for a long time. So we think there is some evidence that this novel combination may be effective in patients with mutant RAS for -- in second-line colorectal cancer. You can see that the distribution of the liver metastasis, the metastatic organ for the patients who achieved partial response, are well distributed among the liver and other sites, including the peritoneum. Next one. Actually, I asked them to do a better slide because I still don't fully understand this slide thing. But I think what I think is very important that we are looking at very sophisticated monitoring of biomarker with liquid biopsies. The liquid biopsy allow real-time molecular monitoring of the efficacy of the ongoing treatment. And you can see that we saw in the patients who responded, significant decrease of the KRAS mutant allele frequency. And this was seen across the different variations of KRAS mutations. So we may have a biomarker with liquid biopsy monitoring the allele frequency of KRAS in these specific mutations. Next one. You can see that the patients who have a partial response have significant decrease. And it's actually extremely exciting to see that some patients even with stable disease, they continue to drop even -- we didn't reach partial response, but it still continue to drop the allele frequency. And you can see 9 out of 11 had a KRAS where in which we could detect at baseline with digital PCR. All patients showed decrease of the mutant allele frequency after the first cycle. And the biggest change was in the patients, as expected, who has partial response. Obviously, we need more data to better understand what the degree of decrease means and is there any time of progression, a reversal, an increase of this mutation or other mutations at the molecular scape. But so far, we have not seen that. Next one. So at USC, we did actually a little bit more work because it seems like Polo Kinase 1 is a very important target and a downstream event of RAS signaling. So we wanted to know what the -- really the value or the clinical significance of Polo Kinase 1 expression is. We use the [ CARIS ] database with over 7,500 metastatic colon cancer patients with complete NGS analysis and 6,000 of these patients had whole transcriptome analysis. And I think these are very interesting data to better understand the distribution of high expression of Polo Kinase 1. The expression was divided in 4 quartiles and quartile 4 is the highest expression. And you can see that the expressions seem to be different based on the metastatic site. The highest is the liver, then brain and ovary and bladder have particular high expression in these metastatic sites. And there seems to be not so much difference as such the lung and the peritoneum seems even the opposite. So these are early data, but giving us a little bit a landscape on the metastatic site and the upregulation of this particular gene, Polo Kinase 1. Next one. There are some differences in association with other pathways because it may be very important for the future of developing this promising target to combine others. And Mark and Afsaneh mentioned that potential other combinations may be very valuable depending on the cancer type. We could clearly see that in the high expressers of Polo 1 Kinase, you have upregulation of WNT signaling and TP53 and chromatin remodeling as well as of DNA mismatch -- of DNA repair, including mismatch repair, as shown here with the Lynch Syndrome. There are also some association with notch signaling and as you can see, response to immunotherapy. So I think that gives us a glimpse maybe where in the future, potential combination maybe or should be developed. The next slide shows very exciting data and linking Polo Kinase 1 with immune signaling. And in a very dramatic way, you can see that the upregulation of Polo Kinase 1 is significant associated with basically all immune cells or cells involved in immune signaling, including the T cells, the CD8, the cytotoxic, the NK, the B lineage, the monocytic, the dendritic cells so anticancer-associated fibroblasts. So we are very, very interested in looking for the role of this particular target and using Onvansertib in combination with immunotherapy, and I'm sure Mark and Vicki are working on potential concepts in preclinical models. When we look particularly at the immune checkpoint genes with PD-1 and PD-L1 and PD-L2 and CTLA-4 and the Tim-3 and the Lag-3, you can see significant association. So I think there is a unique potential for combination in the immune checkpoint arena or other immune checkpoint modulators. Next one. So in conclusion, I think I hope I could convince you that the combination of Onvansertib and FOLFIRI-Bev is very well tolerated. Very few toxicities. And the one we see are more related to the 5-FU bolus of the FOLFIRI regimen. Only 9% reported AEs, Grade 3, Grade 4. Most of them, the neutropenia I mentioned. We saw 5 out of 11 with a partial response, including 4 confirmed and 1 patient proceeded to a curative surgery. So it -- we couldn't confirm it, but we cured him, so I don't think that's a dilemma. Eight patients had a durable response over 6 months. And as mentioned from Dr. Barzi, the PFS in this patient population is between 2 and 5 months. Four of these are still on treatment and the median of progression-free survival has not been reached. Dr. Erlander gave you a little bit insight that maybe biomarker may be very helpful in identifying the patients who do very well with the partial response to these significant decreases and the most significant decreases of mutant allele frequency of the KRAS variance. So I think -- and I showed you some glimpse that this particular target of Onvansertib may be having other very promising combination partners in the immune therapeutic arena and other potential chemotherapy agents targeting the DNA repair pathways as well as modifier of chromatin remodeling, which I think is a very exciting new field to develop. So I'm going to stop here and I'll give it back to the operator to take over and coordinate the question-and-answer session.

[Operator Instructions] The first question is from Terry Smith, asking if we could talk about the competitive landscape around KRAS-mutated CRC. Let me start with Dr. Barzi, and then if Dr. Lenz would also like to add.

So speaking about -- I'm assuming the reference is to drug development. I would go back to say there are many efforts to develop drug for KRAS mutated. Majority of the effort is focused on the third, fourth-line therapy because a lot of them are single-agent therapies or targeted therapies. The uniqueness of this combination is that it's actually placed in second-line therapy. And there aren't any KRAS mutation trials that are looking in the second-line therapy. So the uniqueness of this approach is actually the line of therapy, and the fact that the -- it is going in combination with chemotherapy and biological therapy, which is identified as an active therapeutic option in this patient population.

So I may jump into that because I think there has been incredible efforts made because Frank McCormick leads the RAS program at NCI, which is working for many years to identify treatment options for this patient population because of the high frequency of RAS mutation in many cancers. And this is not an easy task. They have not been really successful moving forward. I think we have learned a lot about RAS signaling. We now know there are 2 major subsets, the epithelial and mesenchymal form. But treatment strategies are very limited. We all know about the [ ancient drug ], the [ sheets ] you have seen Dr. Barzi mentioned. I think there are very interesting data in lung cancer. Unfortunately, the colon data are not as promising. Very low response rate and the frequency of this mutation is very low, less than 10%. So I don't think this will impact most of the cancers, including colon. RAS protein is extremely difficult. When you look at the structure, you don't have a target to really inhibit or interfere. That's the problem with this particular mutant protein. So I think the major focus is really to inhibit downstream events. And one of the downstream events that next linking to the signaling is actually the protein SOS, which has been now being developed as a target. These trials are ongoing. The data are coming in. Maybe there is some effect, but we still need to see more data. The PLK1 is a very unique downstream event. And that's one of the reasons I also showed that it has significant interaction with other pathways opening it up for very unique combinations. And what's very important for our proof to see the signatures with Irinotecan and bevacizumab to make it a valuable really treatment options on top of the standard of care, where we don't have to start in third or fourth line refractory, where it's much more difficult to find a place for an effective drug to be evaluated. So I think this is a very unique opportunity to go into second line with these now promising results. We are very excited to see where this data will lead us.

Dr. Lenz, another question for you. How common is it for a second-line patient to proceed to curative surgery?

Very, very rare. Dr. Barzi can confirm. We don't have it very often in first line. So in second line, usually much more difficult. And this patient had peritoneal disease so this is even more difficult and challenging. So I think we are very lucky that this patient underwent this particular surgery. And I think the treatment was very instrumental to have control of disease and tumor shrinkage to make him eligible for this curative resection. So I think when we expand this clinical cohort, hopefully, we see more of these. But we are very excited. And this is the guy also, and Mark can comment on, who had significant disease of the mutant allelic frequency. It's all matched. Clinically, no side effects. Clinically doing extremely well. Decrease of the tumors. Decrease of the liquid biopsy, KRAS, MAF and curative resection. So we couldn't have been more excited for this gentleman.

Thank you. This is for both Dr. Lenz and Dr. Barzi. Maybe we'll start with Dr. Barzi. Perhaps you can share your view on rechallenge with Avastin. Do patients respond better after initial response or they develop resistance in the second-line setting?

So the definition of resistant to anti-angiogenic therapy is a very hard definition. We really don't have any clinical reliable mechanism to identify resistant to anti-angiogenic therapy. As far as we can say, if you can continue with anti-angiogenic therapy through lines of therapy in colon cancer, we have first line, then second line, we have -- first line, bevacizumab; second line, bevacizumab, aflibercept, ramucirumab; and potentially, third line, regorafenib. We show that basically, sensitivity to anti-angiogenic therapy remains throughout the course of treatment and the course of disease. So if I want to respond to that, I would say, number one, we don't have any clear mechanism for resistance or clear way to identify resistant to anti-angiogenic therapy clinically. There does not appear to be a resistance to this agent across lines of therapy. In terms of the response, it's believed that bevacizumab is not a major player in response and tumor shrinkage in these -- at least in colorectal tumor type. And it's primarily the improvement in the time to event, progression-free survival, and overall survival that drives the use of anti-angiogenic therapy, including bevacizumab in this population.

So I can only agree with Afsaneh. I think the interesting data are that the approval of bevacizumab beyond progression in first line was based on a randomized clinical trial. The median survival was only a benefit of 1.4 months. There was heavy criticism how patients were selected in this second-line treatment driven by the German cooperative group. Later subsequent meta-analysis actually showed that mutant RAS did not benefit from bevacizumab in this line of treatment. And this was already mentioned in the overview by Dr. Barzi. So I think there is limited benefit from bevacizumab in second line, probably driven more from the wild-type than from the RAS-mutant patient population. Because the data are very clear that when you have bevacizumab resistance, you have upregulation of many genes, including PDGF receptor and IGF receptor and others, which overcome potential VEGF blockage. And the hope was that with ramucirumab and aflibercept to address this, but they have not shown higher efficacy and have more toxicity. So overall in the world, the bevacizumab is considered standard of care for all patients. But I think many of the key opinion leaders agree that there is very limited benefit, particularly in RAS mutant. And I think a very good important point from Afsaneh is that this bevacizumab does not do anything with the response within second line, we see.

Thank you. Thank you, both Dr. Barzi and Dr. Lenz. The next question again, either of you can answer, maybe we'll start with Dr. Barzi. Well, it's a small cohort right now. How does the baseline characteristics of patients enrolled into the Onvansertib trial compared to a general second-line KRAS CRC population when we look at historical data sets?

So I would look at a few things to say the population is very similar to the second-line CRC trials in terms of age and gender distribution. These are patients who have significant burden of metastatic disease. I think Dr. Lenz showed the data with regards to the presence of more than one metastatic sign, the presence of liver metastatic disease, which is the major challenge in management of this disease as well as the performance status, which majority of the patients have good performance status. So the limited data mirrors the clinical trials in the second-line setting, and that makes it reliable. That said, it's a very small data set. And we have to wait for additional data to see if we see the same signal with more patients included in this trial.

I think I have nothing to add. I think you answered that very perfectly. I think I agree.

Dr. Barzi, could you talk a little bit about the AE that led the patient with the unconfirmed PR to go off study? Any additional color you can provide on this patient?

So -- I mean unrelated AEs that resulted in patients going off trial. There -- some of them concurrent illnesses or concurrent condition that resulted in a change in the lab value or patient condition that basically was recognized as a safety limit for continuation of this treatment. I'm not sure I can go into more detail because it's...

Yes, this was the hep B patient.

I know. That's what I was trying to say. I'm not sure I can disclose more, given that it's one patient, and there is potential for identification. But it was a concurrent illness that resulted in a change in lab values that would not have been safe for continuation of Onvansertib.

Thank you, Dr. Barzi. Can you also speak to the rationale for the 2 patients' decision to stop therapy? Was there something in particular [ talk ] was that stopped these patients? What changed between this ESMO data and the data released prior, so that would have been the ASCO data?

So I mean a patient's decision for withdrawing consent is recognized as one of the reasons for leaving any trial. This one is not an exception. These patients left the trial for personal reasons. Some basically were not in the city that this trial was available to them. Others had family reason or other reason that forced them to quit enrollment on the trial. The reason they selected to a stop or quit the trial was not the side effects, was not the treatment, was actually environmental variables that was outside of the control of the investigator and the patient.

Thank you, Dr. Barzi.

Just confirming the patient that went on [indiscernible].

We cannot hear you Vicki and Mark, okay?

The next question, just confirming that the patients that went on to curative surgery had peritoneal disease. And is that 12% of stage IV disease with and without RAS, Dr. Lenz?

Yes. So there is no correlation with RAS mutant and involvement of the peritoneal disease. Other genetic characteristics are associated with more terminal disease such as MSI-high and BRAF that this patient did not have microsatellite instability or BRAF mutation. He really had a KRAS mutation. I think we were very fortunate that he had only intra-abdominal disease and mesenteric involvement and no systemic disease in liver and lung, which would have prevented a curative intervention.

Dr. Lenz and Dr. Barzi, based on your experience and practice, would you mind commenting on Onvansertib's durability? And share with us your opinion on the clinical significance of stable disease in this setting?

Afsaneh, do you want to start?

Sure. So I think a stable disease is actually very good. I know that this trial is designed to detect objective response rate. But in clinical practice, a stable disease is very good, and it implies that the growth of the tumor is halted. I want to bring to the attention of the audience that we look at the response by looking at the imaging. We see a mass, let's say, in the liver, and one can imagine that not all of everything what we see on the imaging is actually tumor cells. This mass is supported by connective tissue, by blood vessels or other elements that are part of the host. Even if the treatment is very successful in terms of shrinkage of the tumor cells, the supportive elements, the connective tissue will still remain and that may be the reason that we are seeing maybe a stable disease in some and more response in others. So speaking of the experience and scientifically, a stable disease is considered a very good event in cancer. In terms of durability of Onvansertib, as Dr. Lenz pointed out, the duration of -- that these patients have stayed on therapy and the duration that they have maintained their response is much better than the reported literature and much better than the individual clinical experience that we have. Dr. Lenz?

Yes, I think that stable disease is underrated because it is a very heterogeneous group or you can have a decrease of 20%, have increase of 20%, and it's all the same. I think that's the reason when I showed the waterfall plots that many of the stable disease actually had some shrinkage, 20% and 30%. And one of them is now on, I think, for 7 or 8 months with further decreasing of the liquid biopsy, KRAS allele frequency. So I think altogether, this is very important and stating this can be very effective in order to have prolongation of PFS and overall survival. If the tumor does not grow, you will not die. And I think if you have a long control of disease, it's very important. You see that, by the way, between one or that, you don't have to have a response to have prolonged incredible PFS and overall survival. So it is not a new thing. I think we saw that more with classical chemo that we don't have long stabilization of disease. I think the new target agents may kind of put a dent on this, that this disease will regain a different value in the future. Only more larger trials can show how the Onvansertib will fit into that equation.

Thank you, Dr. Barzi and Dr. Lenz. Dr. Lenz, this question is for you. Could you please elaborate a bit more on neutropenia and the influence of a 5-FU bolus versus maybe using a 48-hour infusion?

Yes. So I think that's a very important point. And I think a drug developer will have a hard time because there's always work from FDA. You have to use the approved combination treatment. So the reason we know 5-FU very well, because 5-FU was developed at USC more than 30, 40 years ago. So we have a lot of experience how 5-FU should be developed. When you give 5-FU bolus, okay, you actually inhibit DNA synthesis through inhibition of RNA if and it's incorporated and so on. You have completely different side effect profile. You have nausea, vomiting, diarrhea, neutropenia, significant impact on bone marrow suppression. When you give 5-FU over continuous infusion, it's all about inhibition of TS and DNA synthesis. You do not see any neutropenia, none. You see basically skin toxicity, a skin rash and some mucositis, but you don't see neutropenia. Now the French put us in this difficult position all because they put the pump with the bolus together, and that was one and showed a survival benefit, and that's the reason we're stuck with this regimen. The Germans use the 24-hour infusion. They don't see the Grade 3 neutropenia the same. So I think there are many different regimens for 5-FU. When you give it continuously without bolus, you have no neutropenia. So we were fighting with the FDA to get this off the protocol and we couldn't. However, what we have done is that when we see the neutropenia that we can stop the 5-FU bolus. And these patients did not have any further dose reduction. So I think we know exactly why they got it. We cannot 100% exclude that neutropenia is an effect of the investigation drug, but it seems very consistent with the 5-FU bolus administration. So it's a clear definition. And just to make clear, if you take the bolus out, the efficacy does not change, very clearly demonstrated. So I hope that answers the question. And the oncologists around the world struggle with that all the time when we do clinical trials.

That was really helpful. Next question is, can you please describe how patients are selected for inclusion in the trial? Is there any cherry-picking of these patients? What is the inclusion criteria?

Afsaneh, do you want to take that? It's a standard.

So if we think in the broader concept, the inclusion criteria for every trial is, to some extent, cherry-picking. And this is not an exception. However, when we look at the inclusion criteria of this trial, this is not different from the colorectal trials in the second-line setting. We basically define the patient, the molecular requirement, we exclude patients with MSI-high and BRAF mutated. They must clearly have a KRAS mutation on the tissue to be eligible. They must have received first-line therapy, either for metastatic setting or either in metastatic settings or receive therapies in -- for adjuvant therapy, but progress very quickly on adjuvant therapy. They must have acceptable organ function, meaning they're neutrophil counts, their liver function, their kidney function must be at an acceptable level to proceed with the therapy. So I mean, this is very standard inclusion criteria for every trial. I don't see that we are doing anything in terms of cherry-picking the patients. So that's my perspective. Dr. Lenz, what do you think?

Yes, I agree. I think everyone who does clinical trial, we cannot select patients when poor performance in [ hepatocellular liver ] function because we cannot attribute any side effects after that. So it has to be very clean. That goes along with standard. There was no additional criteria where we try to pick the patients who do particularly well. These are standard eligibilities for any clinical trials in order to make sure that the data we see are related to the investigation drug we use and not confounded by comorbidities or organ dysfunction or performance status because that would make it very difficult to evaluate the safety as well as the efficacy of a new drug. So -- but nothing special done outside the standard we would do.

Dr. Lenz, how likely is it that PLK1 inhibition in combination with chemotherapy would also improve outcomes in KRAS wild-type CRC patients?

Yes, that's a very good question. And whoever asked that is a very smart person because I think I have the same question. I think there is a particular sensitivity in mutant. But it does not mean it will not work in wild type. And I think that's a different question and development. And I think I already told the Cardiff to look into potential preclinical models for wild-type situations because you can have upregulation of PLK1 in wild type. So the question is, how do we select the patients who benefit the most? And who should do that maybe even for wild type, okay? So there is a difference in sensitivity, but does not mean it does not work in wild type.

Yes. And I would add, remember, we are identifying KRAS-mutated CRC as a unique entity. If we combine these patients, we may lose the signal. And so the advantage of this trial is it's actually very focused on KRAS-mutant patient, which provides the best opportunity for advancing the care for this population.

Next question is for Dr. Barzi. The treatment multiline study notes ORR of 4% of bev and FOLFOX and but other studies of FOLFOX plus bev in second-line in patients progressing on bev and chemo note response rate of 16% to 25%. What is the range of potential ORRs that has been seen in smaller historical trials? And what gives you confidence incremental effect of PLK1 inhibition is real?

So my answer to that question is you have to look at several different things. One is what is the compound that's used in the trial that reports the response rate and how relevant it is to the current trial we have. I showed you that the response rate in bevacizumab second-line therapy is in the range of 5%. And I think the response rate to other anti-angiogenic therapy, including ramucirumab and Ziv-aflibercept are actually reportedly better. And part of the reason that those response rates are better may have to do with the patient population. For example, inclusion of Asians is much higher in the ramucirumab trial, which may explain the higher response rate that's observed in that trial. The VELOUR trial for aflibercept is criticized for how it selected the patients with a high number of patients actually being rapidly progressing, going from FOLFOX adjuvant therapy to second-line therapy. So the caveat of interpretation of the response rate is partly related to the population selected, partly related to the agent selected. And in this case, given that the chemotherapy backbone is going to be FOLFIRI and bevacizumab, we want our responses to be comparable to that. I will go one step further and say, we should look at the boundary and [ 95% ] confidence interval for the reported response rate. In the ramucirumab trial, the upper bound of the response is something around 16%. And in the lower initial trial, the upper bound of the response is 23%. However, when they reanalyze their population with a population that's similar to the bevacizumab trial, the upper bound of the response comes down to something around 16%. So even if we think that the upper bound of the response is 16%, the current ongoing trial is aiming for a response rate of 20%, which is significantly higher than the historical upper bound response that's reported in the literature. I would also add that there are some retrospective studies that look at the response rate. And if one is worried about the concept of cherry-picking, the cherry-picking is more likely to happen when we have retrospective studies that don't follow the same paradigm of patient selection as prospective studies.

Yes. So I agree with Afsaneh from the evaluation. I think the eligibility and patient entry criteria may be a big difference because not all patients had bevacizumab before in these trials or they allowed a long interval before reintroduction of chemotherapy in second line. We don't know about the quality of progression before. So there are a lot of factors going in. Overall, there is no doubt that the second-line treatment is limited effective, and we need better treatment options, particular for the RAS mutant because none of these have adjusted for RAS mutation in a very significant way. When you look at RAS mutations in these groups, they seem to do worse, and that is a very consistent picture. So we have to be also not only careful looking at the differences of patient entry, but also what patient population do we compare, all wild-type or KRAS. And I think here, there may be differences. Because we see with wild-type better outcomes, more response rate, longer PFS and overall survival.

Dr. Lenz, let's start with you on this next question and then go to Dr. Barzi. There's a clear predominance of FOLFOX in the first-line setting. But it looks like irinotecan-based regimens are used in the first-line setting. What drives the decision to use FOLFIRI over FOLFOX in first line?

So it's a historic reason, okay? Because in the U.S., FOLFOX was compared to a combination with irinotecan with 5-FU bolus, the IFL regimen. And this FOLFOX was superior in efficacy. So the U.S. grew up with understanding that modified FOLFOX-6 is a more effective regimen. They got used to it and know how to deal and administer and monitor them. So FOLFOX become historically the backbone in the U.S. for first line. Does not mean that a modified FOLFIRI is inferior. So in Europe and in Asia, about at least half of them start with a modified FOLFIRI regimen. So it is not that one is more effective than others. There are some differences in the toxicity profile. Irinotecan, maybe a little bit more GI toxicity. In Oxaliplatin, what we are afraid of is the neurotoxicity, which usually starts to develop after 3, 4 months of treatment there. It is a dose-limiting toxicity and Oxaliplatin needs to be dose modified or even stopped if the neurotoxicity does not improve. So I think it's not one is better than the other. I think for patients who have neuropathy with other comorbidities, we choose FOLFIRI. But in the U.S. all the clinical trial designs are laid out and assumed that first-line is FOLFOX best. Now there is an increasing use of modified FOLFOXIRI in first-line because this has shown significant response and PFS and potential overall survival benefits. So I can be very honest when I have a KRAS mutant or BRAF mutant and I think the patient may be resectable or become curative resectable, I often use FOLFOXIRI bev in first line. We had a long time learning curve to how to use it because everyone was afraid of the toxicity. But we have managed it. It's that now standard of care in pancreas cancer. It's a standard of care of colon cancer. So I think we learn it depends a little bit on the treatment goal, how we choose and the overall condition of the patient and their potential side effect profile. If somebody works with their profession using their hands, you need to be more careful with your Oxaliplatin treatment. So it could be -- we still use in a majority of FOLFOX in order to actually allow these patients for subsequent clinical trials, that FOLFIRI is absolutely a treatment options and not -- it's not used because of lack of efficacy or lower efficacy.

I mean I think I completely agree that it's a matter of patterns of practice. In the U.S., FOLFOX is adopted as the first-line therapy. And in patterns of care study, 70% of the U.S. providers use actually FOLFOX in the first-line setting. And the why of that may have to do with trial design, contracts. I think there are some other factors related to that. I think we always discuss with the patient what side effects they can expect. FOLFIRI actually is associated with hair loss and diarrhea, something that most patients do not like. So when you present them with the option, many patients gravitate towards using FOLFOX because of the fact that it doesn't result in any hair loss, and it doesn't result in as bad of a diarrhea as FOLFIRI does.

Next question is again for either or both of you. What is more important in these patients typically? Inducing a deep response initially or maintaining a durable response? And how does Onvansertib factor into this?

Can I start? Let me start, okay? Because I think it's a very important point. It's because we have learned to look at partial response and think this is the ultimate goal, 50% or more volume shrinkage. But I'm not sure if this is correct anymore and maybe the most sensitive measure. In the recent large clinical trials in the 5-FU and [ some 80405 ], it became very apparent that with target agents, we may see a completely difference in the deepness of response as well as early tumor shrinkage. And the Germans led the way and established for the first time that if you have a shrinkage of 20% or more within 6 weeks, this is a very important prognostic marker. And it seems to be much more with the cetuximab in wild-type RAS because you see deepness of response, more than with bevacizumab. And this translated into a very important predictor of PFS and overall survival. But it was not linked to cetuximab. When you see the deepness of [ onset ] of that, you just see it less. Or with FOLFOXIRI, where you see this is a little bit more than with bev, it is still a biomarker, which predicts outcome. Now that does not mean and distinct with the ones who have early tumor shrinkage do overall better, but I cannot say who will have the longest progression duration of response or duration of PFS. But overall, it is a very sensitive marker, which is the reason the waterfall plot I showed you in the tumor shrinkage is very important. You can see that there is significant deepness of response. The Germans also showed that the deepness of response in second line is a significant prognostic marker. So the dynamic and the extent of the response is critical. Now how important PLK1 is for that, it's very early, but it looks promising when you look at the waterfall data. I think anything to say more would be just hypothesis or hopeful wishing. And I hope we are right. But this is kind of where we are.

I completely agree. I would add that the persistence or maintaining the response speaks to the biology of success in terms of controlling the pathway. If you see a high response, that then in a few weeks you lose that, it suggests development of resistant mechanisms. And the development of that mechanisms are rapid enough to overcome the response that you initially achieve. And therefore, I think if you can sustain the response, it only shows that you're successful and there aren't redundant pathways that are going to uprise and create the progression. So that's one perspective. The other one is colorectal cancer is being considered a chronic disease. And for a chronic disease, you're going to have long-term treatment. And the longer you can maintain the patient on the same treatment is actually better psychologically for the patients and for the physicians to not have to transition from one treatment to another. So from both biological and epidemiological reason, it is better to have longer duration of response than to have a very deep response then gets reversed shortly after that.

Next question is, Dr. Lenz, could you take this to start, please? What percent of patients who are KRAS wild-type in frontline develop KRAS mutation as a resistance mechanism to first-line therapy?

So I think -- this is an important question, which is where liquid biopsy play an increasing role. The majority of patients develop a molecular mechanism of resistance. We can explain resistance to cetuximab in about 80% of panitumumab with molecular mechanism, but KRAS mutations, KRAS amplifications are the most frequent mechanism of resistance. But it's funny because it's not only KRAS. In about more than 1/3 of these patients, they have more than one mechanism of resistance. It's c-MET amplification, HER2 amplification, [ PSV ] kinase mutation. I mean it's about 10 established molecular mechanism resistance. So KRAS is the most common and the dominating, but not the only one. So we need to really focus when we see these KRAS mutations that this is a new now targetable drug with this new agent. So it is the most common mechanism of escape under cetuximab pressure.

And if I learned from Dr. Lenz that KRAS is an early development and if you have KRAS, you have KRAS. And if you don't have KRAS, unlikely that you will develop KRAS. It is possible that it's subclonal. It's small in terms of quantity or population of the sub assay in KRAS who don't have a KRAS mutation and who may not be able to see it upfront. And then we see it later, then there's a change or there's a shift in the clonal distribution of the tumor. But for those who have KRAS mutation, they have it. For those who don't have KRAS mutation, it is unlikely to see development of KRAS mutation, not impossible.

Yes. We did extensive analysis in thousands of patients. We know exactly what the mutant allele frequency in the tumor needs to be that cetuximab will fail. So if it's higher than 0.1, cetuximab will fail very quickly. When it's lower than 0.01, actually, cetuximab has a benefit. So when you do all the efficacy analysis, you see that at 0.1, it shifts, that the benefit is getting lost. But I think you need to also consider the other way around, which is the reason I love that Cardiff is looking for the liquid biopsy. You can lose KRAS mutation under treatment pressure in some of the patients. The numbers are not clear, maybe 20% to 30%, where you have a KRAS mutant, you start with FOLFOX bev, and you may lose it, okay? There are very early clinical trials who are showing when you lose it and you treat with an EGF receptor, you can actually have response. So the dynamic of the KRAS in the liquid biopsy goes both ways, which is the reason we are very careful and glad that we can monitor these mutation allele frequency in this clinical trial to make sure it is present.

And just changing gears just a little bit here. Are there other combinations that you would think would be useful to test with an anti-PLK1 mechanism. So other combinations with Onvansertib that you would recommend considering for testing?

If I may take this one. It's -- the development of drugs is one step at a time. And if we are looking at the combination of drug development, FOLFIRI provides an excellent backbone because there is synergy with the 2 chemotherapy agents, both 5-FU and irinotecan. Can we actually look at the development of Onvansertib with Capecitabine or 5-FU alone? Absolutely. That remains an option and an open door, and I know that Cardiff is going to look into that. Can we look at development of this drug with other targeted therapies, including -- I showed you that PLK is to a non-NIK-related pathway. Could we look at combination with NIK pathway? Absolutely and a possibility. But like I said, the development is one step at a time. This seems to be the most logical step given the synergism between the Onvansertib and the 2 chemotherapy agents. And hopefully, if this is proven successful, it would be efforts to look at other chemotherapy combination and also other targeted therapy and immunotherapy combinations.

I mean it's a very good question, and that's the reason we actually did the extensive analysis within [ CARIS ] to see the interaction with activated pathways, the interaction with mutations, the interaction with different metastatic sites. And it becomes clear that certain activated pathways are associated with it. And one of them is really the immune signaling pathway. The other one is DNA repair. So this would actually suggest potential combinations with PARP inhibitor, immunotherapeutic agents. They need to be checked preclinically in mouse models as the Cardiff group has done to see what is promising to move forward. It may be also dependent on cancer-specific drugs available and so on. So we are GI oncologists. So I cannot speak for the prostate cancers and the leukemias where this pathway is also highly upregulated. But I think the targeting DNA repair, targeting immune pathways, but also chromatin modeling, which was -- I was very excited about it, maybe -- or should be tested in preclinical models.

Thank you, Dr. Barzi and Dr. Lenz, for taking the time to be with us today and for sharing your insights. This will conclude the question-and-answer session. However, if there are questions that we were not able to address during this call, please feel free to reach out to us directly or via Life Science advisers, and we'd be happy to address all questions outside of this KOL event. Thank you, everyone. We appreciate your time today.

Thank you so much. Bye-bye.