Revolution Medicines, Inc. (RVMD) Earnings Call Transcript & Summary

Good morning, and welcome to Revolution Medicines Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I'd now like to hand the conference over to your first speaker to Erin Graves, Senior Director of Corporate Communications and Investor Relations. Please go ahead.

Thank you, and welcome, everyone to today's Investor Webcast. Joining me on today's call are Dr. Mark Goldsmith, Revolution Medicine's Chairman and Chief Executive Officer; Dr. Steve Kelsey, our President of R&D; and Dr. Wei Lin, our Chief Medical Officer; Peg Horn, our Chief Operating Officer; and Jack Anders, our Chief Financial Officer will join us for the Q&A portion of today's call. As we begin, I would like to note that our presentation will include statements regarding the current beliefs of Revolution Medicines with respect to our business and the proposed acquisition of EQRx, including statements regarding our development plans and time lines for our portfolio and pipeline and the expected timing and benefits of the proposed acquisition, all of which are intended to be covered by the safe harbor provision of the Private Securities Litigation Reform Act of 1995 for forward-looking statements. These statements are subject to a number of assumptions, risks and uncertainties. Actual results may differ materially from these statements, and except as required by law, the company undertakes no obligation to revise or update any forward-looking statements. I encourage you to review the legal disclaimer slide of our corporate presentation and all our filings with the SEC concerning these and other matters. During this call, we will be referring to slides from our corporate presentation, which was posted to our website prior to this call. With that, I will turn the call over to Dr. Mark Goldsmith, Revolution Medicine's Chairman and Chief Executive Officer. Mark? On behalf of the employees of Revolution Medicines and our many collaborators and partners, it's my pleasure to welcome you to this R&D review following multiple exciting scientific presentations by our collaborating investigators at the Triple and ESMO meetings. Our company's mission is to revolutionize treatment for patients with RAS-addicted cancers through the discovery, development and delivery of innovative targeted medicines. Today, I'll begin by briefly reviewing our pioneering class of drug candidates that target the oncogenic RAS(ON) drivers of common life-threatening cancers. Next, Dr. Steve Kelsey, our President of R&D will recap the recent presentations on our unprecedented RAS multi inhibitor RMC-6236 and RASG12C selective inhibitor, RMC-6291, our first 2 RAS(ON) inhibitors that have shown differentiated and promising initial clinical profiles. And then Dr. Wei Lin, our Chief Medical Officer will summarize our initial single agent and combination development strategies aiming to deliver durable clinical benefit broadly to patients with RAS-addicted cancers. The fundamental biologic principle behind our work is that excessive RAS(ON) signaling caused by tumor RAS mutations drives malignant growth in many human cancers. These oncogenic mutations can occur in any of the natural isoforms of RAS, KRAS, NRAS or HRAS, and most frequently occur at mutational hotspots at amino acids G12, G13 or Q61. Such mutations are responsible for over 200,000 new cancer diagnoses in the U.S. alone each year, and in particular, our frequent causes of lung, pancreatic and colorectal cancer. Innovative work by Kevan Shokat and John Ostrem paves the way for first-generation inhibitors that bind to the KRASG12C mutant in the inactive or off form and prevent conversion into the active or on form. We have taken a very different approach, developing oral compounds that bind to and inhibit the active or on form of RAS. We built an industrial strength product engine, incorporating a vision and early work by Greg Verdine colleagues at Warp Drive Bio, a company we acquired 5 years ago. We call our compounds tri-complex inhibitors, since they are designed to bind selectively and with high affinity to a relatively flat surface on the disease target by forming tri-complex structures that contain the disease target, a highly abundant cellular chaperone protein such as cyclophilin A and the tri-complex inhibitor bound in a pocket between them. Drug candidates in our collection of more than 20,000 design-based RAS inhibitors directly and rapidly inhibit the intended oncogenic RAS(ON) cancer drivers and tumor cells and induce deep and durable suppression of RAS cancer signaling by a mechanism of action rectifies common ways that tumors can resist RAS(OFF) inhibitors. The studies we will review today have provided compelling clinical validation of the first 2 RAS(ON) inhibitors as single agents, and we believe for this new class of inhibitors broadly. Today, we will touch on several of our 6 named development stage RAS(ON) inhibitors, that as a group were designed to treat nearly all patients with RAS cancers. The first of these investigational drugs RMC-6236 is clearly unique in that it potently binds to and inhibits every form of K, N or HRAS we've ever tested, including nearly all known oncogenic variants as well as the wild-type of normal RAS proteins. We call this bold compound, a RAS multi-ON inhibitor. The other RAS inhibitors in our pipeline are designed as mute-selective compounds targeting individual oncogenic RAS variants. These 2 distinctive complementary approaches an inhibitor of the RAS family of proteins and specific inhibitors of individual RAS mutants are intended to enable multiple monotherapy and combination strategies against RAS-addicted cancers. You will hear more about how we plan to implement these strategies later in this presentation. I'll take a moment now to introduce the remarkable drug candidate RMC-6236 that is summarized here, and then Steve will take it from there into the clinical data. As shown in this example of a preclinical RAS-dependent non-small cell lung cancer model, the oral potent tri-complex RAS family inhibitor, RMC-6236 induces dose-dependent regressions in RAS-addicted cancers even at low doses. Regressions representing partial or complete responses have been observed most frequently in the dose range of 10 milligrams to 25 milligrams per kilogram. Preclinically, RMC-6236 is well tolerated in this active range. And as you'll see, plasma exposures in humans corresponding to this preclinical dose range are achieved at doses of 80 milligrams to 120 milligrams per day with objective tumor responses. As recapitulated here, we've shown examples of compelling response rates to RMC-6236 in xenograft tumor models carrying 5 different KRASG12 mutations and representing the 3 major epithelial RAS-addicted human cancers. Such responses to RMC-6236 were also characterized by impressive durability in most models as indicated. Earlier this year, we disclosed a small preliminary set of encouraging clinical findings from the RMC-6236-001 single-agent dose escalation study that in many ways appear to mirror the preclinical profile. I'm pleased to invite Dr. Steve Kelsey to review our significantly larger preliminary clinical data set that is consistent with and substantially extends the encouraging trends seen in those initial findings. Steve?

Thanks, Mark. I'm going to summarize the key data from the 3 presentations on RMC-6236 and RMC-6291 that were recently given at the Molecular Targets and Cancer Therapeutics or Triple Meeting, and earlier today at the ESMO meeting. Slide 11 is to remind you of the study design for the RMC-6236 Phase I dose escalation study, which currently focuses on patients with advanced solid cancers that harbor KRAS mutations that positions G12, currently excluding G12C mutations. We have escalated the dose of RMC-6236 up to 400 milligrams daily. The color coding of the boxes is synchronized with the data slides that we will show shortly. Dose optimization are still ongoing, particularly for patients with non-small cell lung cancer and pancreatic ductal adenocarcinoma. We have not yet decided whether to test the 500 milligram dose level as we are still evaluating 400 milligrams daily. Recall that doses below 80 milligrams daily were not expected to be associated with tumor regressions in patients based on preclinical exposures in human PK modeling, which turn out to be reasonably concordant with clinical reality. As of the 11th of September, 131 patients have been treated with RMC-6236, the majority with either pancreatic ductal adenocarcinoma or non-small cell lung cancer. The spectrum and frequency of the G12 mutations in the study were consistent with the epidemiologic distribution. Patients have received a median of 2 prior regimens for their RAS-mutant cancer. The pharmacokinetics of RMC-6236 as a single agent showed good oral bioavailability, kinetics consistent with once daily dosing and dose dependent increase in exposure. Beginning at 160 milligrams daily and more consistently at 200 milligrams day or above, steady state exposures were within the range of those observed in mice at 25 milligrams per kilogram daily, a dose that you've just heard delivered significant and broad activity in the tumor xenograft study shown on Slide 10. I'd also like to note that dosing at 80 milligrams in patients provided exposures below, but approaching that of the 10 milligram per kilogram mice exposure at which tumor regressions were observed in sensitive to clinical models. Dosing at 120 milligrams or above in patients achieved exposures more consistent with and frequently exceeding that of 10 milligrams per kilogram daily in mice. As we have previously reported, RMC-6236 despite its ability to potently inhibit both mutant and wild-type RAS was well tolerated across dose levels and with drug exposures that were well within the range required for potent inhibition of mutant and wild-type RAS. We have seen a few Grade 3 or greater adverse events overall and have not yet defined a maximum tolerated dose. Skin rash remains the most frequent treatment-related adverse events and the most common dose limiting toxicity as defined by the protocol. At higher dose levels, some gastrointestinal toxicity has been observed, although Grade 3 events were likewise are common. Only 1 treatment-related Grade 4 event has been observed, a gastrointestinal perforation of 80 milligrams daily that has been previously described in some detail and is believed to be due to shrinkage of the full fitness solid tumor. The safety and tolerability profile of RMC-6236 compares favorably with that of previously reported for some KRASG12C inhibitors and with historical profiles for standard of care cytotoxic chemotherapy. The overall toxicity profile, including only infrequent and relatively modest increases in transaminases increases our confidence that RMC-6236 will be tolerated in combination with standard of care treatments. Potential drug combinations include immune checkpoint inhibitors, such as pembrolizumab and our own in portfolio mutant selective RAS(ON) inhibitors, such as RMC-6291 and RMC-9805. Recall that the RMC-6236-001 study is a first in-human Phase I dose escalation study, necessarily, the first cohorts of patients were treated at doses below those that were expected to result in significant anti-tumor activity. In fact, as discussed on Slide 13, exposures in patients reach those at which tumor regressions could be anticipated at around 80 milligrams to 120 milligrams daily and above. This prediction was elegantly supported and exemplified by Dr. Spira in his recent presentation and poster on RMC-6236 at the Triple meeting on the 13th of October. The case history of a patient with ovarian cancer was reported. This patient who had progressed on prior therapies had stable disease when it started on RMC-6236 at 20 milligrams daily. But on escalation to 80 milligrams daily, the tumor reduced in size and the patient achieved a partial and sustained response. Therefore, we have reported anti-tumor activity for efficacy evaluable patients with non-small cell lung cancer and pancreatic cancer, who started treatment with RMC-6236 at a dose of 80 milligrams daily or greater, at least 8 weeks prior to the data cut-off for ESMO, which was October the 12. The timing of this analysis ensures that all reported patients had received at least 1 follow-up CT scan reported to us at the time of the analysis. You can see from the waterfall plot of 40 available patients with KRASG12 mutant non-small cell lung cancer, the overall response rate, including unconfirmed partial responses was 38%. There was one confirmed complete response. The overall response rate observed so far compares favorably with inhibitors of KRASG12C that have been tested in KRASG12C non-small cell lung cancer, and is above historical response rates to standard of care chemotherapy exemplified by docetaxel. The majority of unconfirmed partial responses in our non-small cell lung cancer cohort have been subsequently confirmed and some RES's responses were observed for the first time after initial 6 week scan, due to the relatively short follow-up for patients enrolled at the higher dose levels in this study, it is possible that the observed overall response rate for RMC-6236 in lung cancer is an underestimate of the eventual outcome, and we will continue to assess this. As predicted by the preclinical work, multiple tumors with either KRASG12D or KRASG12V-mutant patients were responsive to RMC-6236. To our knowledge, RMC-6236 is the first RAS inhibitor described with clinical activity against lung cancer with either of the 2 most common RAS mutations in human cancer. Furthermore, in lung cancer, the G12D and G12V genotypes together are about as common as the G12C genotype or roughly 12.5%. Although, we haven't tested RMC-6236 in patients with the G12C mutation yet, preclinical data shows that RMC-6236 is highly active in tumors harboring the KRASG12C mutation, and we'd expect it to be so in patients. We also expect tumors harboring other mutations to be sensitive as well, but none have been represented sufficiently in the lung cancer patient cohort so far to be able to assess this. Furthermore, while the data set is too small and early to determine whether any differences exist and outcomes across the different G12 genotypes, overall, RMC-6236 could well prove to be applicable in 25% to 30% of human non-small cell lung cancer patients. Slide 16 shows more temporal information on each patient with non-small cell lung cancer. The median follow-up is relatively short of 3.1 months, and the majority of patients remain on study therapy. To reinforce the anti-tumor activity of RMC-6236 in KRAS mutant lung cancer, I would like to highlight this case presented by Dr. Arbor at ESMO earlier today of an 83 year old female with a heavy smoking history that was diagnosed with KRASG12V mutant non-small cell lung cancer. Her prior treatment for non-small cell lung cancer included a checkpoint inhibitor doublet and subsequently chemotherapy, but she progressed and presented with dyspnea at rest and cough before starting RMC-6236 of 300 milligrams daily. Within 1 week, she noticed a marked decrease in both symptoms and she started playing golf again. Her first CT scan showed a complete response, which was confirmed on her next scan, and she currently continues on treatment. Slide 18 shows the waterfall plot of 46 patients with KRASG12 mutant pancreatic cancer, again, only including patients at doses above 80 milligrams daily and who started treatment with RMC-6236 at least 8 weeks prior to the data cut off on October 12, so as to ensure that information from at least one follow-up CT scan was captured on each patient. The overall response rate, including unconfirmed partial responses was 20%, although one of these patients did discontinue RMC-6236 prior to response confirmation. This overall response rate compares favorably with historical data for salvage chemotherapy in pancreatic cancer where response rates are typically around 10%. The disease control rate was 87%, meaning that only 13% of pancreatic cancer patients treated with RMC-6236 had progressed with their first response evaluation. Typically, around 30% to 40% of patients with advanced pancreatic cancer would have been progressed at their first response assessment when receiving salvage chemotherapy after their initial chemotherapy has failed. As with non-small cell lung cancer, pancreatic adenocarcinoma tumors with KRASG12D and G12D mutations were responsive to 6236. In addition, 2 patients with pancreatic tumors harboring KRASG12R mutations, a mutation almost exclusive to pancreatic cancer and the third most common RAS mutation in pancreatic cancer also responded. With the 3 RAS mutations that are most commonly found in this tumor-type showing sensitivity, we believe the breadth of activity is consistent with the preclinical experience and highly encouraging for our efforts to help patients. The swimmer plot on Slide 19 confirms the short follow-up for patients in this study. As with the lung cancer cohort, it is too early to calculate a median progression free survival. But it is encouraging that only 2 patients with pancreatic cancer who had an initial response have progressed after 18 and 25 weeks on treatment, respectively. Median progression-free survival for patients treated with chemotherapy has been reported to be in the range of 12 to 14 weeks. Durability of anti-tumor effect irrespective of RECIST response will be an important factor in determining the clinical impact of RMC-6236. To confirm the anti-tumor activity of RMC-6236, we evaluated mutant KRAS allelic burden in circulating tumor DNA before and on treatment. Slide 20 shows that treatment with RMC-6236 was associated with a significant reduction in circulating mutant KRAS DNA. Consistent with previous reports, reduction in ctDNA tracked with clinical response by RECIST. So to date, and while dose optimization is still ongoing, RMC-6236 has been well tolerated at active dose levels with significant anti-tumor activity in both non-small cell lung cancer and pancreatic cancer and against at least 3 common RAS genotypes. The early overall response rate in both lung cancer and pancreatic cancer compares favorably with current standard of care for patients with advanced tumors who have failed prior systemic therapy. Now let's turn to RMC-6291, our mutant selected KRASG12C inhibitor that binds covalently to the on-state of KRASG12C and rapidly and irreversibly inhibits it. This compound is highly patent, inducing tumor stasis in preclinical models at 3 milligrams per kilogram daily and regressions, including frequent complete responses at 10 milligrams per kilogram daily or higher. RMC-6291 has a number of distinguishing properties from the RAS(OFF) inhibitors based on its unique binding and mechanism of action. For example, it is highly active in some preclinical tumor models that are resistant to first-generation RAS(OFF) inhibitors. Shown here are 2 examples of significant anti-tumor activity by RMC-6291 in tumors that are resistant to sotorasib on the left or adagrasib on the right, carrying associated resistant mechanisms that are believed to be clinically relevant. Further, in some preclinical models in which RMC-6236 or RMC-6291 as single agents were only partially active. The combination of these 2 compounds delivers significant and high sustained anti-tumor benefit. This observation supports an important part of our overall strategy for the RAS(ON) inhibitors that will be discussed later. Slide 25 shows the study design for the RMC-6291 single-agent Phase I dose escalation study. As a reminder, the study focuses on patients with advanced solid cancers that harbor KRASG12C mutations, the majority of which have either non-small cell lung cancer or colorectal cancer. To date, we have escalated the dose of RMC-6291 up to 400 milligrams twice daily. All of the doses tested were predicted to have some degree of anti-tumor activity based on preclinical modeling. As with the RMC-6236 results shown earlier, the color coding of the boxes is synchronized with the dose levels on the data slides that we will show shortly. Dose optimization for RMC-6291 is still ongoing, and the data are preliminary, but we believe providing a meaningful early snapshot of the potential of our first mutant selective tri-complex RAS(ON) inhibitor. To date, 63 patients have been treated with RMC-6291, of which 23 have non-small cell lung cancer and 33 have colorectal cancer, the remaining 7 have various tumor types, including cholangiocarcinoma and gynecologic cancers. 25 or 40% of these patients had previously received with KRASG12C(OFF) inhibitor. Most of these patients receive their KRASG12C(OFF) inhibitor as their most recent prior therapy before being treated with RMC-6291. All of the patients previously treated with the KRASG12C(OFF) inhibitor have progressed on their prior KRASG12C(OFF) inhibitor and the gap in the time between discontinuing RAS(OFF) inhibitor and starting RMC-6291 was a median of 9 weeks, although with a wide range of 2 weeks to around 2 years. As with many Phase I dose escalation studies in this space, patients have received a median of 3 prior therapies for that cancer. With particular relevance to interpreting the safety and tolerability of RMC-6291 in the non-small cell lung cancer population, please note that 1/3 of the lung cancer patient is treated with RMC-6291 has received an immune checkpoint inhibitor in 12 weeks of starting RMC-6291. Prior immune checkpoint inhibitor therapy within the previous 12 weeks was reported by investigators at Memorial Sloan Kettering Cancer Center as a predictor of significant toxicity with KRASG12C(OFF) inhibitor therapy, particularly hepatotoxicity associated with sotorasib. The pharmacokinetic profile of RMC-6291 shows that it is orally bioavailable and exposures increased in a dose-dependent manner. The clearance of RMC-6291 from plasma is consistent with the twice-daily dosing regimen, which was initiated at the 200 milligram dose level. The exposure to target engagement relationship in preclinical studies predicts greater than 90% average cross-linking of total KRASG12C in human subjects receiving 100 milligrams twice daily or higher. Please note that distinct from KRASG12C(OFF) inhibitors, most or all of cross-linked intratumoral G12C should represent direct inhibition of the active or on state of mutant KRASG12C. RMC-6291 was generally well tolerated up to 400 milligrams twice daily. Grade 1 or Grade 2 gastrointestinal toxicities, predominantly nausea and/or diarrhea were the most frequent treatment-related adverse events. QTC prolongation has been observed, although only one case representing 2% of patients overall showed an increase in QTC to greater than 500 milliseconds. And there have been no cardiac toxicity associated with this electrocardiographic atonality. Only 1 treatment-related AE, a Grade 3 QTC prolongation in a patient who is tolerating RMC-6291 well otherwise required discontinuation of RMC-6291 per protocol. Increases in liver enzymes were low grade and low frequency, supporting the potential for combination with immune checkpoint inhibitors. Slide 29 shows the clinical activity of RMC-6291 in oral patients with non-small cell lung cancer. 3 out of 7 or 43% of lung cancer patients who have not previously received a KRASG12C(OFF) inhibitor responded. 5 out of 10 or 50% of the lung cancer patients who had previously received the KRASKRASG12C(OFF) inhibitor responded. The disease control rate for both sets of patients was 100% with no early progression. And no patient who has responded viruses has yet progressed or discontinued treatment. To confirm the anti-tumor activity of RMC-6291, we're also showing the reduction in circulating KRASG12C mutant allelic frequency by dose and by response. With the caveat of small numbers and considerable heterogeneity, molecular responses are as expected associated with tumor shrinkage. Molecular responses do not seem to increase in depth with dose levels greater than 200 milligrams twice daily. Overall, the initial data show encouraging clinical activity for RMC-6291 in patients with KRASG12C mutant non-small cell lung cancer and colorectal cancer, in particular, the response rates in patients with non-small cell lung cancer that previously received a KRASG12C(OFF) inhibitor compared very favorably with recently reported data on the experience with other KRASG12C inhibitors. And the overall response rate for single-agent RMC-6291 in colorectal cancer is similar to that reported for other KRAS inhibitors in combination with anti-EGFR antibodies. While preliminary, these data are consistent with the differentiated mechanism of action for RMC-6291 that has previously been presented and support the potential of RMC-6291 for further development. I will now turn back to Mark.

Thank you, Steve, for that clear review of important data from the first 2 RAS(ON) inhibitor investigational drugs in our pipeline with encouraging clinical activity across multiple common KRASG12 tumor genotypes and tumor types, data that we believe provide clinical validation of this exciting and highly differentiated class of drug candidates. I'd also like to remind everyone that a third investigational drug from this class is also undergoing initial clinical evaluation, although, we don't have sufficient experience to report on it today. This compound, RMC-9805 is an unprecedented oral inhibitor that selectively and covalently engages KRASG12D, the most common mutated driver of RAS-addicted cancers. It is highly active preclinically, and we look forward to reporting on its clinical profile at the appropriate time. I'm also pleased to introduce formally RMC-5127, the newest exciting mutant selective compound from the RAS(ON) inhibitor class to enter development at RevMed. This oral inhibitor is highly selective and potent against the second most common driver of RAS-addicted cancers, KRASG12V. This compound is in IND-enabling development, and we look forward to bringing it into the clinic. And a reminder that previously, we've described 2 other mutant selective RAS(ON) inhibitors in development, RMC-0708, targeting the KRASQ61H variant and RMC-8839 targeting the KRASG13C variant. We are proud of our extensive portfolio of R&D assets focused on RAS-addicted cancers now dominated by this deep collection of RAS(ON) inhibitors. We've reviewed the current status of our 3 clinical stage compounds, including the first 2 for which we have initial safety, tolerability and clinical activity data as well as 3 additional development candidates and a further pipeline of preclinical and discovery assets. In a moment, we'll bring on Dr. Lin to give us a preview of what's next for the exciting compound RMC-6236. Let me comment here that we also believe RMC-6291 is quite promising, and we await additional monotherapy data to help guide prioritization of the multiple potential paths by which we can drive clinical impact, and ultimately pursue commercial value from this compound. In the meantime, you'll see that we are already moving forward with the study of the combination of RMC-6291 and RMC-6236. Let me now update you on 2 RAS companion inhibitors we have in clinical development. First, RMC-4630 is a selective inhibitor of the RAS regulator SHP2 that we envisioned as a potential combination drug with RAS inhibitors. Despite encouraging early data from the CodeBreaK 101 study in the global Phase II [ RMC-4630 ] study, the combination of RMC-4630 dosed at 200 milligrams day 1, day 2 with sotorasib dosed at 960 milligrams QD, showed additive side effects compared to either agent alone. Clinical benefit and response rate or durability was not observed, likely limited by dose interruptions and discontinuations due to intolerability. Given the exciting clinical profile of RMC-6236, we had given this compound priority over RMC-4630 for evaluation in combination with our mutant selective RAS(ON) inhibitors, but RMC-4630 continues as an option for possible evaluation in other combinations in the future. Second, RMC-5552, a first-in-class bi-steric, selective mTORC1 inhibitor was designed to treat cancer patients with RAS mutant tumors bearing co-mutations that upregulate in TORC1 signaling. The ongoing RMC-5552 01 study has provided objective evidence of single-agent clinical activity in patients at tolerated doses as recently updated at the Triple meeting, including partial responses. Consistent with our initial vision for this compound, future development will focus on combination treatment with RAS(ON) inhibitors, particularly for patients with tumors carrying both RAS mutation and markers of mTORC1 hyperactivation. Let's now return to our headline asset, RMC-6236. As Steve showed, this inhibitor has proven to be broadly active in patients with multiple common tumor types and driven by multiple common RAS gene types. We plan to execute further development of RMC-6236 with both speed and breadth by making parallel investments in both late-stage monotherapy trials and clinical studies of combination approaches. As you'll see, we are committed to expanding the scope of development for RMC-6236 by evaluating the broadest possible range of RAS genotypes of lines of treatment and of tumor types. Dr. Lin will provide this overview of our current thinking, which, of course, remains subject to data-driven decisions and consultation with investigators and regulatory authorities. Wei?

Thank you, Mark. Our vision for the evolvement of RMC-6236 is to bring the benefit of direct RAS(ON) inhibition, the greatest number of patients with RAS-addicted cancers. To achieve this vision, we will pursue in parallel the initial registration of RMC-6236 as monotherapy in multiple later line indications with high unmet needs, exploring combinations and a significantly improved the standard of care in the first-line and adjuvant settings. To broaden the impact of RMC-6236 on the lives of patients with RAS-addicted cancers. We aim to build on the proof-of-concept emerging from our first in-human study and expand the development of RMC-6236 in multiple dimensions. These plans include going beyond KRASG12 mutations, going to earlier lines of therapy and going into all solid tumor types. Let me describe in more detail these 3 dimensions. First, graft mutations. In our first in-human trial, the preliminary safety and anti-tumor activity of RMC-6236 have been established in patients with lung and pancreatic cancers, specifically harboring KRASG12 mutations. However, preclinical data indicates that RMC-6236 is broadly active across all 3 canonical RAS isoforms, KRAS, NRAS and HRAS, as well as against all 3 common mutation hotpots, G12, G13 and Q61. Therefore, in our potential future registration trials, we aim to broaden the biomarker selection to include all 3 RAS isoforms and all 3 mutation hotspots. As we expand from population is where, we have established clinical proof-of-concept, 2 populations where we are relying on pre-clinical proof-of concept. We use nested designs and hierarchical testing to mitigate potential risks from including expanded populations. I'll illustrate this approach in our proposed monotherapy randomized trial design later in the presentation. Second, lines of therapy. While the highest unmet needs are in previously treated patients, the untreated first-line and adjuvant patients represent the largest number over 200,000 patients with RAS mutated cancers newly diagnosed in the United States each year. Furthermore, agent therapy presents the only setting where we may be able to offer cancer patients the potential for cure. Our strategy expanding into earlier lines of therapy involves a multipronged approach that includes development of our RAS(ON) inhibitor doublets, as well as combinations of a RAS(ON) inhibitor with standard of care and other anti-cancer therapies. Third, tumor types. Lung, pancreatic and colorectal cancers are the most common RAS-addicted cancers, many other solid tumors such as melanoma, ovarian, endometrial and cholangiocarcinoma contained RAS mutations at a high frequency. The most significant approach -- efficient approach to expand the evaluation of RAS(ON) inhibition to these patients is a tissue-agnostic registration strategy via a potential single-arm Phase II trial. And our current strategy is to explore initiating this type of trial. Now I want to delve more deeply into our monotherapy development plan in non-small cell lung cancer. Each year, approximately 200,000 people in the United States and 2 million people worldwide are diagnosed with non-small cell lung cancer. The majority will die from their disease at some point, making it the most lethal of all cancers. RAS plays a role in the majority of non-small cell lung cancer. And RAS is the most common oncogenic driver in non-small cell lung cancer with approximately 30% of newly diagnosed non-small cell lung cancer patients, having tumors to harbor RAS mutation. The current global standard of care for second-line patients will be treated with immunotherapy and platinum-based chemotherapy is docetaxel plus ramucirumab, while approved, it's not widely used. Sotorasib and Adagrasib are indicated for use only in patients with KRASG12C mutated non-call cell lung cancer, because they have not gained full approval, they're not formally considered standard of care on major health authorities. At the table on the right illustrates, our Sotorasib has shown a higher response rate, improvement in median progression free survival was only 1.1 month from 4.5 to 5.6 months, and there was no overall survival benefit. Therefore, the entire population of patients with RAS mutated non-call cell lung cancer share a common global standard of care in docetaxel and expect median survival of less than 1 year represents a major unmet need. To address this high unmet need, we're planning a global randomized Phase III trial comparing RMC-6236 against docetaxel in patients with RAS-mutated non-small cell lung cancer who have been treated with immunotherapy and platinum-containing chemotherapy. Biomarker selection will include KRAS, NRAS and HRAS. Will include G12C and non-G12C mutations and will include G13 and Q61 mutations. The study endpoints are progression-free survival, overall survival and patient report outcomes. In our currently contemplated trial design, we've taken into account the potential complexity associated with including patients with RAS mutations, spanning positions, G12-G13 and Q61. We expect that a package to support a broad approval, we need to include sufficient data to demonstrate efficacy in each mutation subgroup. We have a number of approaches to serve this goal, including a nested design with higher statistical testing that began with a focus on the patient group with the highest probability of success, and sequentially expands into additional groups. We will be discussing our proposed analysis plan with the FDA before finalizing our Phase II design. The design presented here is preliminary and has not been reviewed by health authorities. The final design we change or will not change or I believe that the high M&A exists and our commitment to evaluate RMC-6236 broadly in patients with RAS-mutated non-small cell lung cancer, as we try to improve the standard of care for these patients. Next, let's review a potential monotherapy bond plan for RMC-6236 in pancreatic cancer. This disease is notorious for having the worst prognoses among all major cancers, most of the approximately 50,000 patients in the United States and 500,000 patients were blocked for needy diagnose with pancreatic cancer each year who die from the disease. More than 90% of pancreatic cancer harbors a RAS mutation, making it a dominant oncogenic driver in this disease. Among the chemotherapy regimens, commonly used in the second-line setting, the median progression-free survival is between 3 and 3.5 months. And median overall survival is between 6 and 7 months, making this indication one of the highest unmet need in oncology. While the first-generation KRASG12C(OFF) inhibitors have shown anti-tumor activity, KRASG12C mutation occurs in only 1% of pancreatic cancer. So these inhibitors are not relevant for most pancreatic cancer patients. However, the initial evidence of efficacy provides the first proof-of-concept that human pancreatic cancer is a RAS-addicted cancer. And together with the clinical data we presented at ESMO and the Triple meeting strongly supports the continued devaluation of RMC-6236 in the broader population of RAS-mutated pancreatic cancer. To address this high unmet need, we're considering a potential global randomized Phase III trial comparing RMC-6236 against a physician's choice of chemotherapy regimens in patients with graft-mutated pancreatic cancer. These are second-line patients whose tumors harbor the RAS mutation. The broad biomarker selection will include all RAS isoforms and G12,-G13 and Q61 mutations. The study endpoints would include progression-free survival, overall survival and patient report outcomes. Similar to our non-small cell lung cancer Phase III trial. This trial is -- has to account for the complexity associated with including multiple mutations. At the appropriate time, we'll be discussing our proposed analysis plan, including a potential nested approach with the FDA before finalizing our Phase III design. Therefore, the final design may change based on regulatory feedback. Our goal is to provide a robust data set to support a broad approval in RAS-mutated pancreatic cancer. The high prevalence of RAS mutations in pancreatic cancer is a strong indicator that this is a RAS-addicted cancer. Therefore, we are committed to evaluating the potential of RMC-6236 to improve the standard of care for these patients broadly. So far you've heard how we are thinking about the development of RMC-6236 as a monotherapy using non-small cell lung cancer and pancreatic ductal adenocarcinoma as priority examples. Of course, we believe that the potential clinical utility of RMC-6236 and our entire RAS(ON) incubator pipeline is much broader. Now I want to share 4 strategies: expanding RMC-6236 development from monotherapy to combinations. Combinations underlie our plan in the first line setting, where effective treatment options may exist and will consider combinations with standard of care of the development option. Combinations also underline our plan in the active setting where we may have the opportunity to evaluate a potent of cure for patients. and combinations underlying our plan to expand into multiple tumor types. Our first strategy is RAS(ON) indicator doublet. We hypothesized that in RAS-addicted tumors holding the ambition of RAS signaling drives efficacy. And an in-resistance mechanism is recovery of RAS in tumor cells. This hypothesis is supported by data on resistance to the first-generation KRAS inhibitors. Therefore, our first combination strategy is to develop RAS(ON) inhibitor for doublets, which based on pre-clinical results, we presented, we prevent or delay the emergence of resistance and they translate to longer durability of treatment benefit and longer survival for patients. Given our differentiated mechanism action based on tri-complex inhibitor technology and our deep pipeline of RAF(ON) inhibitors, RevMed is uniquely positioned to execute on this strategy in EORTC space, we have already initiated the development of RMC-6236 plus RMC-6291 in a Phase I trial and are actively recruiting patients. Our second strategy is to replace chemotherapy with RAS(ON) inhibitors, first-line non-small cell lung cancer, where intensive care is an chemotherapy plus immunotherapy is an ideal indication to explore this strategy. RAS(ON) inhibitors for RAS(ON) inhibitors doublets, aimed to deliver better outcomes for patients by targeting the oncogenic diction with greater efficacy while improving on the safety profile of chemotherapy. The combination development of RMC-6236 plus anti-PD-1 and RMC-6291 plus anti-PD-1 is currently in the back timing stage. Our third strategy is to combine with other target therapies. In colorectal cancer, receptor tyrosine kinase reactivation, especially eGFR is a primary mechanism that appears to limit the efficacy of the first-generation KRASG12C(OFF) inhibitors. Combination with anti-eGFR therapy has shown benefit in addressing this limitation. While KRASG12C mutation represents only approximately 3% to 4% of colorectal cancer. RAS mutation overall occur in approximately 50% of colorectal cancer. We, therefore, expect to evaluate the combination of RMC-6236 plus anti-eGFR antibody in this large patient population. Our fourth strategy is to combine RMC-6236 plus standard of care. Pancreatic cancer is a true example for this strategy. For patients with RAS mutated pancreatic cancer, neither immunotherapy nor target therapy is approved, and the senator remains chemotherapy for both first-line and the acid settings. Combination with chemotherapy offer RMC-6236 an alternative path, especially for patients with tumors harboring an additional mutation not addressed by RAF(ON) inhibitor doublets. The combinations of RAF of RMC-6236 plus chemotherapy are currently in advanced planning stage. Of course, these 4 strategies may evolve as we learn more about each of our RAF(ON) inhibitors; their combinability, their efficacy in different tumor types and lines of therapy. We believe that this multi-pronged approach will improve our chance to change the standard of care in RAS-addicted cancers. Back to you, Mark.

Thank you, Wei. As you can see, we've made a great deal of progress with our rich portfolio of RAS(ON) inhibitors and we have exciting ongoing activities and development plans. To put the clinical findings to date in perspective, together, the first 2 of our investigational drugs from this class have already shown significant antitumor activity in patients against the 4 most common oncogenic RAS G12X mutations, which we believe validates our ability to target the large G12X subset. Together, these compounds have also shown significant activity across the 3 major types of RAS-addicted epithelial cancers. Projecting the validated RAS G12X subset on to the epidemiology of these common solid tumors, it can infer that these first 2 compounds in aggregate, credibly may be able to serve a remarkable 3/4 of the RAS-mutated forms of these 3 major cancers. This compelling scope of opportunity may become even more inclusive as we expand clinical evaluation of RMC-6236 to the G13 and Q61-mutated RAS genotypes and to other tumor types and continue advancing other assets in our pipeline. We are most grateful to the patients and their families who have participated in our trials so far or planned to, to our investigators and collaborators, to our employees, partners and advisers and to our investors for helping us sustain our ambitious program in order to outsmart cancer.

Thank you. [Operator Instructions] Our first question comes from the line of Matt Frahm (sic) [ Marc Frahm ] from TD Cowen.

Congrats on the data over the past week. Maybe to start off with -- one with the data today, it does look like there's maybe a bit more delayed responses sometimes in pancreatic cancer than maybe what you're seeing in lung cancer, the biological rationale that based on your preclinical data you're seeing that might explain that?

Thanks, Marc, for your question. Really appreciate it. I'm going to ask Steve Kelsey first to comment on that and we'll see where we go from there.

It's not at all clear that the data to support the delayed response in pancreatic cancer comes from preclinical models necessarily. I think it comes from a wealth of clinical experience with pancreatic cancer generally, that it is a very different disease from non-small cell lung cancer, both in terms of the way it's constituted with a lot of the bulk of the tumor often being not actually tumor tissue. So that's the first thing. So even if you eradicate all the tumor, you may still end up with a significant lump on a CT scan, which takes a while to remodel because it's mainly fibrous tissue. The second obvious difference impacting between pancreatic cancer and non-small cell lung cancer is the involvement of the immune system that a lot of non-small cell lung cancers have a lot of immune effector cells in there and those can significantly debulk the tumor whereas pancreatic cancer is notoriously difficult for the immune effector cells to access and so it may just take a lot longer. And I think that's what we're seeing. I think that's what we're seeing, Marc with the difference between how long it takes for a patient to achieve a partial response to RMC-6236 in the pancreatic cancer versus the lung cancer patients.

And then maybe just on the idea of expanding out beyond G12X, can you remind us preclinically if you see any meaningful differences in the level of activity against some of these other hotspots? But then also while you're kind of finalizing designs with regulators, are there any plans to start enrolling some of those patients on a single-arm basis just to kind of reconfirm clinical activity?

Why don't I comment on the first question and then Steve can hit on the second one. The first question, differences between G12 mutants and G13Q61, in the preclinical work, you may recall, we studied a very large panel of cell lines, which is sort of the best way we have to get at this and across many different genotypes in many different tumor types. Really the only difference we could see in terms of response rates was between the G12 mutants and all the other mutants. They were all sensitive. As we've mentioned, we've not really found a mutant, oncogenic mutants for variants of RAS isn't sensitive to some degree to RMC-6236, but there was a quantitative difference between the G12s and all the others and that's why we prioritize G12 in the study. Of course, as I mentioned, we're probably covering 3 quarters or more of all RAS cancers, even just with the G12 mutations, but there just seems to be some difference in sensitivity. And I think Wei's comment about the nested design with hierarchical testing, statistical testing is a really convenient and practical way for us to both focus on the highest probability subgroup, representing 3/4 of cancers and then also testing the others without taking a penalty for including them in the trial. With regard to your second question, could you repeat the second part of the question?

It was just if there was going to be any plan to, at least in the interim while you're finalizing that design enrolled from these patients with G13 or Q1 (sic) [ Q61 ] mutations.

Sure. Maybe, Steve, do you want to comment on that?

Yes is the answer. We have been focused very much on pancreatic ductal adenocarcinoma non-small-cell lung cancer up to now. I think as we are moving -- well, firstly, we're more confident that we have an active drug and secondly, moving closer to dose optimization, we are planning to do a number of things and Wei described some of those, particularly the combinations that will allow us to move into earlier lines of therapy, but also is a high priority for us to get some data in the other mutations beyond the G12 mutations and also in some of the other tumor types that are the G12 mutations as well, particularly colorectal cancer and some of the gynecologic malignancies. So that's very much part of our plan for the next sort of 6 to 12 months.

Yes. If I could add to that. Investigator and patient demand for RMC-6236 remains very high, as we've talked about before and that continues to be the case based on feedback as recently as today from ESMO investigators are very interested. They have many ideas for other patients and the needs are quite deep. And so we will certainly be making the compound available in a variety of contexts, expansion contexts both to serve patients while we're still evaluating things, but also to collect the kinds of data you're talking about. So I think we're about to hit sort of more of an inflection point and opening things up to a broader group of patients. Wei, maybe he could comment also though on the tumor-agnostic trial concept, which he talked about briefly, but is an important expansion tool and something to move on relatively quickly, although you'd like to do that at a recommended Phase II dose, but to move relatively quickly on and to run in parallel with the more dedicated randomized Phase III trials. Wei, do you want to comment a little bit further on how that works and what the value of it is.

Sure. Happy to Mark. Yes, as Mark mentioned and Steve as well that we are going to be enrolling in the expansion cohort of our -- in our first-in-human study, additional solid tumor that cover other histologies and other mutations to expand on the pre-observation that RMC-6236 is active, much more broadly than in the patient population. We've shown data so far. I think if the data continue to support that, that could potentially lead to a discussion with health authorities for a potential [ tissue-agnostic ] registration in both solid tumors. Thanks.

And our next question comes from the line of Jonathan Chang from Leerink Partners.

First question, given the early stage -- still early stage nature of the RMC-6236 efficacy data, can you talk about whether you think the data could improve from here, whether it's a function of higher dose levels, longer follow-up or otherwise? And then second question, if the data evolve over time in a way that's largely consistent with the data presented today, can you talk about your thoughts on whether the data are sufficiently compelling versus standard of care benchmarks?

Sure. Thanks. Maybe Steve can comment on both of those questions.

Yes. So the first question about what do we expect to happen to the data over time, I think our expectation is that the data will at least stay as is, if not improve and there are a number of reasons for that. The first is that the -- we've shown you data for all patients treated at 80 milligrams daily and above. And there's a big difference between 80 milligrams daily and 400 milligrams daily. So the overall response rates right now, don't forget are driven in large part by the patients that are at lower dose levels and not so much by the patients or rather the high dose levels for whom we have less data. The second reason is, as you can see from those waterfall plots, there is a tendency for patients for their disease to continue to shrink and for some patients who have not achieved the criteria for partial response to move into the criteria for partial response and actually, when we have done the analysis, we've actually looked at the response rates for patients that have been followed for longer than those that were included in this ESMO analysis, the response rates are actually higher than the response rates that were reported at ESMO. So it is quite possible that the response rates will improve. Now the thing that will really be telling, I think, and we'll have to wait some more months for the data is the durability, not just the durability of responses, but the durability of freedom from progression for patients who have had some degree of tumor shrinkage and have not progressed. And I think that will be probably more important and will be a bigger driver of the long-term impact of RMC-6236 and whether or not the response rate goes up or down by a few percentage points.

Yes. Maybe I can add to that. I think a couple of things that tend to get lost in that particular topic. One is the tolerability and safety profile is viewed by investigators' expectations as very strong and that has real impact on so many things going forward. And so from a -- do we need to improve in that regard, I don't think we need to improve in that regard. Secondly, the breadth of genotype activity is unprecedented and I don't think we need to improve in that regard. And third, so far, the durability that we've measured is very strong. We just don't have enough observation period. So as Steve said, well, we'll just need to see and we'll watch it, but there's really no indication today that we have anything but a compound that will provide durability. And ultimately, that is really going to drive most of the approvals that Wei spoke about earlier. The only one, I think, in which the response rate matters is in that tumor-agnostic study, where that will be an important endpoint. So we're pretty pleased with where we are today and with the collection of data and patients patient outcomes so far. We can't predict the future, but we're quite encouraged by it. And I'd also add that, as I mentioned earlier, investigators are very excited about this. They're clamoring to be part of the studies of RMC-6236 and they want to put their patients on the compound. So I don't think that there is much for us to be concerned about from the point of view of the caregivers who are responsible for these patients recognize the deep unmet needs and frankly, the very strong and highly differentiated outcomes that we described today, for which there simply is no other comparator.

Yes. So you also asked specifically about benchmark comparisons. I think the easier one here is non-small cell lung cancer. I mean the comparators docetaxel, the overall response rate in lung cancer is more tightly linked to the temporal endpoints of progression-free survival and possibly even overall survival. And so I think we have a fairly compelling case there and that's why we're intended to move forward. The difficulty of pancreatic cancer, I mean, look, firstly, the response rate is double -- anything that's ever been seen in G12X-mutant pancreatic cancer, right? I mean there have been reports of G12C specific inhibitors being used in G12C mutant pancreatic cancer, which is extremely small patient population. The outcomes are similar to what we think -- what we've shown for RMC-6236 and our investigator base, as Mark said, is really excited about this. The challenge there is not so much whether the overall response rate is better than standard of care. It clearly is. The challenge is whether or not the durability because overall response rate is not so tightly linked to progression-free survival, we have to wait until we're confident that the progression-free survival is of a nature that is going to be significantly impact patients by being significantly better than standard of care. And so we're confident at the moment with the preliminary data that we have that the trajectory is in the right direction, but we just don't have enough information right now and particularly not at the -- what will ultimately be the recommended Phase II dose to be able to report out what we are like to have.

Our next question comes from the line of Eric Joseph from JPMorgan.

Let me add my congrats on the data. I guess with the dose selection in the planned Phase III is still pending, I'd be curious to know sort of what additional follow-up you're seeking to arrive at dose selection and sort of perhaps if there's a time frame that you might attach to that? And Mark, you've previously talked about the possibility of advancing tumor-specific dose regimens, one dose selection into lung cancer, another for pancreatic. I guess any additional thoughts on that potential strategy in light of the data that you're presenting here today?

Great. Thanks very much, Eric. Let me preface the first part of the answer to your first question. Let me provide a first part of the answer with regard to time line, which is that we have indicated and I can continue to believe that we have a very good shot at starting the first randomized Phase III global trial in at least one indication in 2024. We can't be more specific than that for the reasons that are kind of embedded in your question, but we're well on our way towards designing such a trial. And of course, we need a little bit more information. Maybe Wei can then fill in a little bit, not so much in terms of time, but just what are we doing to get to that dose selection for lung cancer and for pancreatic cancer as well, which, by the way, could end up being the same dose. I think I've earlier just said it is always possible that one would end up proceeding with a dose that's working very well in lung cancer, while you're still evaluating higher doses in pancreatic cancer and that's sort of a TBD right now. We're just being transparent about it. But maybe Wei can comment on that further. And then we'll come back to the second part of your question.

Sure. Happy to Mark. Yes. As you know, FDA has published guidance called Project Optimus on how to optimize dose. And we're following that guidance very closely in expanding patients who have non-small cell lung cancer and pancreatic cancer at more than 1 dose level to really optimize and identify the optimal dose to move forward and to discuss with FDA as the recommended Phase III dose. And I think like Mark mentioned, it's possible that these 2 tumor histology would arrive at the same dose or different, I think will be very data-driven in that regard. And so that would be one of the key gating identified [indiscernible] for opening up our first [ dose ] hopefully many registrational trials in the coming year.

And then your second question, really, I guess I already commented on and Wei did as well, which is do we end up with different doses. And I think that was sort of embedded a little bit in Jonathan's question about response rates and I think Steve gave you a very clear view that we're in a strong place today with response rates, but it's also not the main thing that we're looking for, which is something I've talked about for the last number of months, as you know. The disease control rate is very high for both lung and pancreatic cancer even at the doses that we tested. And those, we believe, are more likely to be a predictor of durability, which is going to be the key to approvals rather than response rate per se. And I think Steve also mentioned the response rates. They're calculated at a moment in time. They're calculated to include patients who had only one scan and have even had the opportunity to convert and we've seen very high rates of conversion. So I don't think we're too focused on those specific numbers. But there is a difference in the unmet needs between lung and pancreatic cancer and patients and their caregivers maybe willing to pick their time and tolerate higher doses. So there may actually be a justification for pushing those sort of more aggressively in one population versus another. Again, I'm not advising that that's what we will do, but that's a subject of ongoing discussion and we haven't reached an MTD yet in either population. So there certainly is room to go. And in general, with RAS inhibitors once observed that more is better, a phrase I've used many times in the past. And so that's something for us to keep on the table and we'll see whether things converge for diverge.

And maybe just a quick follow-up, if I could. I guess is there anything in this RMC-6236 data set that speaks to the potentiality for mutant KRAS heterogeneity or a mix of KRAS mutations within a given patient's tumor, whether any cases in this data set so far? And if so could you speak to the activity of RMC-6236 in that type of setting?

I don't think we have -- since for RMC-6236, there are no patients in the study who have previously received a RAS inhibitor. So there hasn't been any real selection pressure for the emergence of those rare or second mutations that tend to become enriched over time in patients who had a RAS(OFF) inhibitor. But we haven't evaluated any patients like that simply because the protocol is designed. So I don't think, to my knowledge, I don't think we've seen any second RAS mutations within the data set. Steve or Wei, if they want to add to that. I think we just don't know.

Not in the RMC-6236, no. The RMC-6291 data set, we've seen a number of patients with colorectal cancer who progressed on a KRAS G12C inhibitor that have had multiple additional mutations and amplifications. And that's, I think, one of the -- I know we talked about RMC-6236 but we did mention that we haven't reported outcomes for patients with colorectal cancer who are treated with RMC-6291 and with the increase you treat it with another G12C inhibitor, and that's because the G12C inhibition with the G12C inhibitors frequently induces other mutations in colorectal cancer. But in the RMC-6236 data set, we have not seen that. And I think that for the reasons that Mark mentioned is that because there are currently no other RAS inhibitors being used in these patients that selection pressure is very different, but these patients are undergoing selection pressure with cytotoxic chemotherapy. And so the primary driver RAS mutation tends to still be the predominant driver in these patients.

Which I think does set up, just to sort of close out that point, it does set up the concept here that RMC-6236, which we continue to believe is going to be active against many different mutant forms and the preclinical evidence strongly supports that, that it may, in fact, provide a much more robust and durable suppression of the RAS pathway because those clones that might be lurking as ultra-rare clones and heterogeneity in the background simply don't have the opportunity to emerge. And we'll take every day of benefit that, that might be able to provide. I think that could really serve to our patients. But that's a concept today based on the preclinical data. Of course, we just don't have any clinical data yet to validate it.

Well, congrats again, and thanks for taking the questions.

And our next questions comes from the line of Chris Shibutani from Goldman Sachs.

This is Charlie on for Chris. And I'll add my congratulations as well. Really exciting. So first, we were wondering just in terms of the ongoing work to pursue lung and registrational studies, whether you are considering the inclusion of active brain metastases considering the preclinical evidence for activity within the CNS? And then as a follow-up, I was just wondering, with the lung complete responder, can you provide some clarity on when that patient had dosed down to 200 mgs?

Sure. Active brain mets, Wei, do you want to comment on what we're contemplating with regard to active brains mets in a registration study.

Yes, sure. Happy to. So the current plan is to actually follow -- we plan to follow a very direct-driven approach. And so the current Phase I study has not included a patient with active brain mets. They all have included patients with treated brain mets. And so since that's the data type that's been generated and we'll follow that. So the patient population planned for the Phase III would also include only patients with treated brain mets and not active brain mets. However, in the future, we do at the right moment, trying to really possible explore patients who have active brain mets and then evaluate the activity of RMC-6236 as well as RMC-6291 in those populations.

Yes. And just to build on Wei's comments and as you sort of built in your question, we have shown preclinical data for all 3 of our clinical stage RAS(ON) inhibitors that showed activity in a model of brain metastases RMC-6236 showed activity in an identical model to that used with, I think, Adagrasib and KRASG12C tumors and RMC-6236 was as good or superior to that. So I think it is active. On the other hand, the 2 mutant selective inhibitors showed even greater activity in the preclinical model. We don't know exactly how well either of those predicts clinical behavior, but we're encouraged by them. And as Wei said, we will, in parallel, I think a key point to all of this is that we're going to study all of these different questions in parallel, while we're also driving towards the earliest possible and strongest possible approval. So more to come, but I think parallel, parallel, parallel should be the theme of the day. Your second question had to do with the patient who had a complete response, a pretty amazing response, probably had a response faster than we measured it simply because based on her clinical behavior. But I think the first can was done per protocol at 6 weeks. Wei, do you know -- I think the question being asked here was whether -- yes, when did the dose reduce and specifically, I infer from that, that you're trying to relate that to the antitumor activity of 200 versus 300. But maybe Wei can clarify that.

Yes. So patient who experienced a complete response was dose reduced in the middle of cycle 3.