Amgen Inc. (NASDAQ:AMGN) European Society for Medical Oncology Annual Congress 2022 Conference Call September 12, 2022 1:30 PM ET
Company Participants
David Reese – Executive Vice President of Research and Development
John Strickler – Medical Oncologist, Duke Cancer Institute
Ferdinandos Skoulidis – MD Anderson Cancer Center
Jean-Charles Soria – Senior Vice President of Development
Conference Call Participants
Salveen Richter – Goldman Sachs Group, Inc.
Jay Olson – Oppenheimer & Co.
Umer Raffat – Evercore ISI
Mohit Bansal – Wells Fargo Securities, LLC
Robyn Karnauskas – Truist Securities, Inc.
David Risinger – SVB Securities LLC
Kelsey Goodwin – Guggenheim Securities, LLC
Dane Leone – Raymond James & Associates, Inc.
Colin Bristow – UBS
Brian Skorney – Robert W. Baird & Co.
Operator
My name is Jay Sean, and I will be your conference facilitator today for Amgen’s Investor Conference Call from the European Society of Medical Oncology 2022 Congress. All lines have been placed on mute to prevent any background noise. There will be a question-and-answer session at the conclusion of the last speakers prepared remarks. [Operator Instructions]
I would now like to introduce David Reese, Executive Vice President of Research and Development. Dr. Reese, you may now begin.
David Reese
Thanks, Jay Sean, and welcome, everyone. We had a couple of important LUMAKRAS datasets presented here in Paris earlier today. And I know there is an intense interest in both part of the lung cancer and colorectal dataset. So we will move into the program presentation in just a moment. Can we go to the next slide please?
Here is our Safe Harbor statement in the next slide and then one more slide. So today, I’ll provide a very brief introduction and then Dr. John Strickler from Duke will give an overview of the colorectal cancer data, LUMAKRAS in combination with Vectibix and then Ferdinandos Skoulidis will provide an overview of the CodeBreaK 200 Phase III confirmatory trial in non-small cell lung cancer followed by Jean-Charles Soria from Amgen who will give some concluding remarks.
Why don’t we go to the next slide, please? This is for your reference, again, our broad oncology portfolio, of course, today we are focusing on LUMAKRAS in the two lead indications, G12C-mutated non-small cell lung cancer and colorectal cancer.
So with that, I’d like to turn things over to Dr. Strickler, who will take you through the highlights in the colorectal cancer data from his perspective. John?
John Strickler
Thank you so much. It’s a pleasure to be here and you can advance to the next slide. So I’m going to talk about Sotorasib in combination with Panitumumab in Refractory KRAS G12C-Mutated Colorectal Cancer. This is safety and efficacy for Phase Ib full expansion cohort. You can advance.
So it’s important before we get into the data to review the current standard-of-care in patients with metastatic colorectal cancer. And in patients who have developed intolerance or progression on first and second line therapies, that’s 5-FU, oxaliplatin, irinotecan and anti-VEGF monoclonal antibodies. The existing standard-of-care offers fairly low benefit. So each row here represents a FDA approved therapy. Regorafenib was approved based on Phase III CORRECT trial, which showed a survival improvement of 1.4 months, PFS benefit of 0.2 months, and a response rate of 1%.
The other agent here, which I call TAS-102 Trifluridine/tipiracilv, also FDA approved based on a very small survival benefit of 1.8 months and a response rate of just 1.6%. So this clearly demonstrates that there is an unmet need for these patients. And honestly, as a practicing GI medical oncologist, I can tell you that many of these patients exhaust standard-of-care chemotherapy and still retain good performance status and are looking for options. You can advance to the next slide.
So this is the CodeBreaK 101 Subprotocol H Study Design. This is a Phase Ib multi-center study. The eligibility criteria are shown here: Key eligibility includes KRAS G12C-mutated metastatic colorectal cancer. They are all KRAS G12C inhibitor-naive and they have all progressed after 5-FU, oxaliplatin, irinotecan, and an anti-angiogenic agent. Now this data that I will show you is from part two, which is the Cohort A dose expansion that includes both Sotorasib, 960 milligrams daily and Panitumumab dosed every two weeks. The primary endpoint here is safety and tolerability. You can advance.
Here is shown the baseline characteristics. The characteristics, the demographics you see are very typical for a colorectal cancer patient population at least the trial population. What I’d like to highlight here is that this is a heavily pretreated patient population. The median prior lines of therapy is two. So this is exactly the type of patient that would typically get regorafenib or TAS-102. I’ll also point out that there were 13 patients that received either regorafenib or TAS-102 or both in one case. So that’s 18% both. You can advance.
Next to review the treatment-related adverse events. And once again, this is for all 40 patients enrolled. Treatment-related adverse events of any grade occurred in 93% of patients predominantly attributed to panitumumab. There were nine patients or 23% that had a Grade 3 treatment-related adverse events. There were no Grade 4 or fatal treatment-related AEs. And there were some treatment-related adverse events that led to dose interruptions or reductions. Typically, the panitumumab which occurred in 25% of patients, 15% of patients had a dose reduction attributed to sotorasib.
On the right, you can see that the treatment-related adverse events are typically dermatologic and are also usually Grade 1 or Grade 2. And as a practicing GI medical oncologist, I can tell you that we are quite accustomed to managing anti-EGFR rash with supportive care. You can advance to the next slide.
Next is shown the efficacy and the key thing to keep in mind with this is that this is an intention to treat population. The confirmed objective response rate here is 30% and an additional 63% of patients had stable disease as best response. So the disease control rate was 93%. I personally have enrolled many patients on this trial, and it’s quite frustrating because they will get right up to 28% or 29% and not quite cross that line. And it can be like that sometimes with the RECIST criteria. The other interesting note is that we typically think of right sided cancers that’s having an adverse prognostic signal for colorectal cancer, but there was no difference in response rate. You can advance.
This next slide shows tumor response on the left with the waterfall plot and the key thing to note here is that there was a reduction in RECIST lesions observed in 88% of patients. I can see some of my patients there sitting right at 29%. And I can also tell you from personal experience that I have a patient who is right there at 28%, 29%, somewhere in that range who has been on treatment for well over a year and tolerating treatment well. On the right, you can see the median duration of treatment was 5.9 months with the median duration of response was 4.4 months. You can advance.
This next slide shows the change in target lesions over time and what you can see here is that some of the patients who respond have quite durable responses. So the median duration was 4.4 months, but there are clearly some patients here that are getting out past 10.5 months almost to a year. You can advance.
This next slide shows progression-free survival and with a median follow-up of 11 months, the median PFS was 5.7 months. Also, of note and this was commented in the session today with the median follow-up of 8.8 months, the median overall survival had not been reached yet. So that’s an important finding. I will also point out that there was no difference in PFS, no meaningful difference in PFS based on tumor side despite the fact we think of right-sided primaries as having an adverse prognosis. You can advance.
So in conclusion, sotorasib plus panitumumab was safe and tolerable in these patients with chemotherapy-refractory KRAS G12C-mutated metastatic colorectal cancer and the adverse events that were seen are consistent with the safety profile of sotorasib and panitumumab. Many of these adverse events were quite comfortable and experienced managing in the clinic. The confirmed objective response rate of 30% is clearly higher than what was reported with sotorasib monotherapy. What’s impressive here is a disease control rate of 93%. There was no difference in these outcomes based on tumor location and the medium progression-free survival is impressive at 5.7 months. Clinically meaningful and clearly better than what we have as our FDA approved standard of care. Both of the FDA approved therapies have a median PFS less than two months.
The overall survival data is also early, but appears promising and that overall survival had not been reached. So based on this result CodeBreaK 300 has been launched globally. This is a randomized Phase III study looking at sotorasib plus panitumumab versus investigator choice standard-of-care treatment for KRAS G12C-mutated metastatic colorectal cancer. Great. Thanks.
David Reese
Thank you, Dr. Strickler. Now we’ll move to Dr. Ferdinandos Skoulidis from MD Anderson, who will present key results from the CodeBreaK 200 Phase III trial. Dr. Skoulidis, next slide, please.
Ferdinandos Skoulidis
Good evening. My name is Ferdinandos Skoulidis, and I’m an Associate Professor in Thoracic Oncology of The University of Texas MD Anderson Cancer Center. It is my pleasure to be presenting results from the CodeBreaK 200 Phase III study. So thoracic versus docetaxel for previously treated non-small cell lung cancer with the KRAS G12C-mutation. Next slide, please.
So sotorasib, the first-in-class irreversible oral inhibitor is now approved conditionally in over 40 countries around the world for the treatment of advanced KRAS G12C-mutated non-small cell lung cancer, following a failure of at least one prior line of standard systemic therapy. These approvals are based on results from the Phase I/II CodeBreaK 100 clinical trial in which sotorasib monotherapy yielded a confirmed objective response rate of 41% with a median progression-free survival of 6.3 months, median overall survival of 12.5 months and a favorable safety profile.
Historically, following failure of platinum doublet chemotherapy and immune checkpoint inhibitors, chemotherapy with docetaxel represents a preferred treatment option and is associated with an objective response rate of between 12% and 14%, a median progression-free survival rating between like 2.8 and 4.2 months and the median overall survival between eight and nine months. However, more recently retrospective data in the era of immunotherapy have reported improved clinical outcomes with docetaxel in patients previously treated with immune checkpoint inhibitors, suggesting that perhaps there is a chemo-sensitization effect after immunotherapy.
So here we present results from the primary analysis of the first ever Phase III clinical trial of a KRAS G12C inhibitor, CodeBreaK 200 evaluating sotorasib compared with docetaxel in previously treated patients with KRAS G12C-mutated advanced non-small cell lung cancer. Next slide please.
So this is the clinical trial scheme of CodeBreaK 200. CodeBreaK 200 is a global randomized Phase III clinical trial that was conducted across 22 countries and in five continents. Key eligibility criteria included patients with metastatic or locally advanced and unresectable KRAS G12C-mutated non-small cell lung cancer. That had received and progressed on at least one prior line of prior systemic therapy. That must have included platinum doublet chemotherapy and immune checkpoint inhibitors administered either concurrently or sequentially.
ECOG performance status need to be zero or one. And patients could not have active brain metastases, although patients with a history of CNS involvement and stable and treated brain metastases well allowed to enroll and actually this represented one of the stratification factors for the trial.
Other stratification factors include race and prior lines of therapy. Patients were randomized one-to-one to receive either sotorasib orally at a dose of 960 milligrams once a day or docetaxel intravenously at a dose of 75 milligrams/m2 and once every three weeks and patients were followed and longitudinally with imaging every six for the first year and every nine weeks thereafter.
The primary endpoint of the study was progression-free survival assessed by blind independent central review per RECIST version 1.1 in the intention to treat population. Secondary endpoints included overall survival, objective response rate, duration of response, disease control rate, safety, tolerability and patient reported outcomes. I should note that in February 2021, and in accordance with recommendations from regulatory authorities, the clinical protocol was amended to reduce planned enrollment from 650 patients to approximately 330 patients, and also to enable cross patients randomized to docetaxel to the sotorasib arm. And next slide, please.
Baseline characteristics were generally well matched between the treatment arms and should note that the majority of patients, approximately three quarters of patients were enrolled in Europe and about 12% of patients were enrolled in North America. Approximately one in patients in both arms had a history of CNS involvement and the median number of prior lines of therapy was two. I should, again, emphasize that third eligibility criteria all patients needed to have received and progressed on both platinum doublet chemotherapy and an immune checkpoint inhibitor, unless these were medically contraindicated. Next slide, please.
At a median follow-up of 17.7 months, CodeBreaK 200 met its primary endpoint, demonstrating superior PFS over for sotorasib over docetaxel with a statistically significant and clinically meaningful hazard ratio of 0.66 corresponding to a 34% reduction in the risk of progression or death with sotorasib over docetaxel. The 12-month progression-free survival rate was 24.8% for sotorasib and 10.1% for docetaxel, the median progression-free survival was 5.6 months for sotorasib and 4.5 months for docetaxel. Next slide.
PFS benefit for sotorasib was consistent across subgroups that were pre-specified on the basis of demographics, a decline ECOG performance status number of prior lines of therapy, PD-L1 expression subgroups. Importantly, in patients with a history of CNS involvement, the hazard ratio was 0.53 in favor of sotorasib indicating that perhaps sotorasib has got clinical activity in this patient subgroup. Next slide.
In the intention to treat population, the confirmed objective response rate with sotorasib was 28.1% and was significantly higher than with docetaxel, which had a confirmed objective response rate of 13.2%. The disease control rate was 82.5% with sotorasib and 60.3% with docetaxel. As you can see from the waterfall plots responses to sotorasib were deeper with a median depth of response of 59% compared to a median depth of response of 49% with docetaxel. Next slide.
Responses to sotorasib also occurred more rapidly and were more durable with sotorasib compared with docetaxel with a median time to response of 1.4 months for sotorasib compared with 2.8 months for docetaxel and the median duration of response of 8.6 months with sotorasib compared with 6.8 months for docetaxel. Next slide.
Overall survival did not differ significantly between the treatment arms with hazard ratio of 1.01. I should note here that the study was not powered to detect an overall survival difference and 34% – at least 34% of patients randomized to the docetaxel arm were subsequently treated with a KRAS G12C inhibitor. Next slide.
Moving on to the safety data. It should be noted that despite patients remaining being treated with sotorasib for a longer period of time as can be seen from the median duration of treatment, as sotorasib resulted in a lower incidence of Grade 3 and above treatment-related adverse events and the lower incidence of serious adverse events compared with docetaxel. Next slide.
The toxicity profile of sotorasib was consistent with what had been previously reported in CodeBreaK 100 and the majority of treatment-related adverse events were low grade and easily manageable with standard-of-care and supportive measures. The most common and treatment-related adverse event of all grades with sotorasib was diarrhea. The most common treatment-related adverse event all grades with docetaxel was fatigue. The most common Grade 3 or above treatment-related adverse events with sotorasib were diarrhea and ALT and AST elevation, whereas with docetaxel, it was neutropenia including febrile neutropenia and fatigue. Next slide.
Finally, I think it is important and significant to note that sotorasib significantly slowed deterioration in quality of life metrics, including global health status and physical functioning as well as cancer-related symptoms, including dyspnea and there was a trend towards a slower deterioration, the patient reported outcome of chest pain. Next slide.
So in conclusion, sotorasib, the first-in-class KRAS G12C inhibitor in CodeBreaK 200 showed significant improvement in the primary endpoint of progression-free survival with a hazard ratio of 0.6 months, which was highly statistically significant. The 12 month progression-free survival was also favorable for sotorasib, 24.8% compared to 10.1% for docetaxel. And importantly, the PFS benefit was consistent across all clinically meaningful patient subgroups. This marks the first positive Phase III clinical trial for a KRAS G12C inhibitor in advanced previously treated KRAS G12C-mutated non-small cell cancer.
Other measures of efficacy, objective response rate, disease control rate, duration of response and time to response were also improved for sotorasib compared with docetaxel. No difference was observed in overall survival between the treatment arms, but the study was not powered to detect an overall survival difference and crossover was allowed. Importantly, sotorasib again exhibited a favorable safety profile with fewer Grade 3 and above treatment-related adverse events compared with docetaxel and clinically meaningful patient reported outcomes were superior for sotorasib versus docetaxel.
In my opinion, taking the totality of the data into consideration, sotorasib should be considered the standard-of-care therapy in patients with advanced KRAS G12C-mutated non-small cell lung cancer following failure of platinum doublet chemotherapy and immune checkpoint inhibitor. Thank you very much for your attention.
David Reese
Thank you, Dr. Skoulidis, and thank you, before that Dr. Strickler for those perspectives. Jay Sean, why don’t we go ahead and open things up for questions. Can you remind our attendees how to ask a question?
Question-and-Answer Session
Operator
[Operator Instructions] Our first question comes from Salveen Richter with Goldman Sachs. Your line is now open.
Salveen Richter
Good afternoon. Thanks for taking my question. Just with regard to the data that we’ve seen here, do you expect any change in uptake impact from the confirmatory trial? And could you just talk about the comparison to [Maradi’s] asset?
David Reese
Yes. I’ll start there and then I’ll ask Dr. Skoulidis to comment. Yes. In our view that these data reinforce the use of sotorasib and LUMAKRAS in this patient population previously treated KRAS G12C-mutated non-small cell lung cancer. In terms of comparisons, we’ll let others speak to their data. This is the first global trial. As you heard, it was conducted in multiple countries around the world. It’s clearly the most heterogeneous trial that’s been conducted in this particular disease state. And with that, let me ask Dr. Skoulidis to offer a perspective here as well.
Ferdinandos Skoulidis
Thank you for the question. I would agree with your comments, Dr. Reese. I think CodeBreaK 200 as I mentioned represents a milestone because this is the first randomized Phase III clinical trial for KRAS G12C inhibitor against the standard-of-care that was positive. I think with regards to the comparison between sotorasib and adagrasib obviously in the Phase II clinical trials, both drugs demonstrated efficacy, and by and large, by several efficacy metrics, from my perspective, the efficacy was comparable. However, at this point in time, CodeBreaK 200 represents the only available randomized comparison for a KRAS G12C inhibitor versus the standard-of-care. And the first confirmatory study that met its primary endpoint in this patient population.
David Reese
Thank you. Next question, Jay Sean?
Operator
Our next question comes from Michael Yee with Jefferies. Your line is now open.
Unidentified Analyst
Hi. This is Dina on for Michael Yee. We just had two quick questions for you. The first is what is the interpretation of the clinical meaningfulness of the 1.1 median PFS, and logically the impact on thinking on how this will affect feature results PFS in the first line setting, and how do you think you’ll be able to get LUMAKRAS in the first line setting given this benefit data that you presented today? And then second, in the first line setting right now, you guys are currently running a KRAS chemo study. What is the control arm there and is that relevant and is that control arm that you guys will be using? Is that a relevant comparator and for which patient specifically in the first line? And are you guys going to be looking at OS there since that’s probably required for approval? Thanks so much.
David Reese
Yes. Thank you. A number of questions embedded in there. I’ll ask Dr. Skoulidis first comment on what he thinks of the clinical utility meaningfulness of the data. And then Dr. Jean-Charles Soria will provide a perspective on some of the additional development work that we’re doing. So Dr. Ferdinandos, would you like to comment?
Ferdinandos Skoulidis
Thank you, Dr. Reese. From my perspective, a hazard ratio of 0.66 highly statistically significant corresponding to a 34% reduction in the risk of progression or death in this previously treated patient population with few available treatment options is definitely clinically meaningful. I think we should – I mean, you can also see from the shape of the survival curves, there’s early and sustained divergence of the Kaplan-Meier curves in favor of sotorasib. The one-year rate of – the one-year PFS rate is approximately double, it’s 28% compared to 10%. So significantly improved for sotorasib. And this coupled with improvement in patient reported outcomes, and then overall favorable safety profile sotorasib, I think to my mind establishes it as a standard-of-care in this patient population and I would consider the data highly clinically meaningful.
David Reese
Great. Thank you. Jean-Charles?
Jean-Charles Soria
Yes. I would just like to add on top of the comments of Skoulidis, that when you look at entirety of the efficacy data, you have overall very positive picture, correct? You have these 34% reduction on risk of progression or death, consistent PFS benefiting [also groups] plus 15% to 12 point landmark, response rate is more than double. But very interestingly, when you looked at the kinetics and the death of the response, you have faster responses, deeper responses lasting longer, tolerability is better. And for the side effects that really matter to people fatigue, alopecia, risk of infection neutropenia, is very much in favor of [indiscernible]. We have the choice of appeal, better tolerated, reducing the risk of progression of death by 34% or on IV treatment that meets hospitalization eventually outpatient or inpatient providing significant side effects. So I think overall meaningfully clinically irrelevant.
David Reese
Great. Thank you. And then to the last part of your question, in terms of the planned Phase III trial, and I will remind everyone that we really think of patients as we think about the therapy of first line disease in three broad categories based on tumor PD-L1 expression, those that are PD-L1 negative, those that have low to intermediate expression, and those that have high expression in patients with negative PD-L1 expression, we have some intriguing preliminary data with chemotherapy combinations. What we think is very encouraging data on that basis and after discussions with regulatory authorities, as you indicated, we are launching a Phase III trial that will consist of chemotherapy plus LUMAKRAS versus chemotherapy and a checkpoint inhibitor. So what would be considered standard-of-care, a standard approach in these patients and I’ll have more to say about that trial in terms of the study design and the timing, data availability enrollment as we get up and running. All right. Jay Sean, why don’t we move on to the next question?
Operator
Our next question comes from Jay Olson with Oppenheimer. Your line is now open.
Jay Olson
Hey. Thanks for taking the questions and thank you for this update. Can you just talk about feedback that you’re receiving from practicing clinicians at ESMO especially with regard to the CodeBreaK 200 patient reported outcomes? And then since it seems like one of the major teams at ESMO has been the study of earlier lines of therapy including neoadjuvant across different histologies and in many cases with immunotherapy. Do you have any thoughts on moving LUMAKRAS into a neoadjuvant setting and how you might study that? Thank you.
David Reese
Great. Thanks, Jay. Yes, there is two very good questions. I’ll ask Dr. Skoulidis to comment in just a moment here. Your question about the standards are shifting and moving into earlier lines of therapy. So neoadjuvant therapy is certainly something that we are thinking about quite a bit here. Dr. Skoulidis let me ask you to comment again, on the patient reported outcomes and quality of life measures and what you’re hearing from the community. And then I’ll ask Jean-Charles Soria to comment on that as well, he’s had a chance to talk to quite a number of physicians here in Paris. Dr. Skoulidis?
Ferdinandos Skoulidis
Yes. I think without a doubt the response from the community echoes what has been reported here in CodeBreaK 200. I think both in terms of global quality of life measures, global health status as well as physical conditioning, there is a clinically meaningful and statistically significant improvement or delaying time to deterioration with sotorasib compared with docetaxel, but also some of the key symptoms that patients with lung cancer are troubled by such as dyspnea and cough.
So I think that in terms of patient reported outcomes, these are clearly very, very meaningful metrics. And I think from the community, we – community oncologist are not particularly enthusiastic about prescribing docetaxel to patients leads to hospitalizations, neutropenia, febrile neutropenia, patients develop peripheral neuropathy, they lose their hair, they develop alopecia. And I think that – and obviously it’s an intravenous treatment that requires frequent hospital visits. And in actual fact, I think the rate of hospitalizations is something that has not been reported, but is being looked at. I would anticipate that the rate of hospitalizations would be higher in the docetaxel arm compared to the sotorasib arm, something very, very, and very meaningful as well.
David Reese
Jean-Charles, anything you’d like to add?
Jean-Charles Soria
Yes. I mean, I have had the opportunity to say Dave to discuss with many colleagues here at Paris and many of them consider this a victory, many of them consider this as a very meaningful alternative. And I frankly don’t know any lung cancer doctor that rejoices at the perspective of giving docetaxel, it’s a highly toxic therapy that has activity that many were giving it because they saw shrinkage of tumor at the cost of alopecia and risk of febrile neutropenia. But here we have a pill that shrinks tumors quicker and more meaningfully. So I think it’s an overall positive reception and yet they noticed lack of less, but it wasn’t powered for OS. And OS has not been seen for oncogene addiction therapies, whether it is third generation EGFR inhibitors or third generation ALK inhibitors, they show strong PFS that they don’t show overall survival.
David Reese
Okay. Thank you. Jay Sean, why don’t we go to the next question?
Operator
Our next question comes from Yaron Werber with Cowen. Your line is now open.
Unidentified Analyst
Hi guys. This is Brendan on for Yaron across all the data. And thanks very much for taking the questions. Just a quick one from my side really on the median PFS curves and the NSCLC data, looks like you get pretty nice separation of the curves early on, and then it seems like the two curves more or less kind of converge after about a year or so. Just wanted to see how you think we should look at that. Do you think that’s maybe more due to lower number of patients at risk as you move farther out or because patients transitioning or is it something you would expect based on prior studies? And then just really quickly, we noticed it looks like there might be a little bit fewer PD-L1 high patients in the docetaxel arm, though doesn’t seem too significant, but just wanted to see if that’s something you plan to keep pretty strictly balanced in the pivotal, or is that less of a focus? Thanks very much.
David Reese
Yes. Sure. I’ll handle those questions. In terms of the shape of the PFS curve, it’s generally consistent throughout with some fluctuation. To see curves tail together at the end is not unusual in these studies. In general also, recall that at least 34%, the patients crossed over to receive KRAS G12C inhibitor after progression on docetaxel. And that’s a minimum estimate. Those are the number – those are the patients where we want to absolutely confirm receipt of the KRAS G12C inhibitor also that maybe responsible in part as well for the sort of tail of the curve effect. Great. Thank you. Jay Sean?
Operator
Our next question comes from Umer Raffat with Evercore. Your line is now open.
Umer Raffat
Hi, guys. Thanks for taking my questions. Maybe a couple, if I may, one, if you could just speak to the response rates seen in this trial high-20s versus prior observations of high-30s. Just your comments on that. Secondly, I know you’re making interesting comments on chemo plus KRAS for first line setting and results are good today, but I wonder if a docetaxel plus KRAS might have been a potential combination to try in refractory setting, also curious of your take on that? And then finally, the duration of therapy was 20 weeks in this trial, which I assume is consistent with real world in refractory setting. What would be your expectation for duration of therapy in first line?
David Reese
Sure. I don’t know that we can speculate on first line. But certainly, you would hope to see a longer duration of therapy in previously untreated patients. So you asked about response rates about docetaxel. Let me ask Dr. Skoulidis, who’s got a lot of experience with these patients and the agent to comment on that and response rate of 28% here versus high-30s, low-40s that we’ve observed in other trials? Dr. Skoulidis?
Ferdinandos Skoulidis
Thank you, Dr. Reese. I don’t think – from my perspective, I don’t think it’s at all unusual that as targeted therapy moves from early clinical development Phase I to Phase II into a global randomized Phase III clinical study that involves 22 different healthcare systems across five continents that you observe, kind of a slightly lower and numerically objective response rate with the targeted therapy. We should also be taking into consideration that these are different studies, CodeBreaK 200 and CodeBreaK 100, the patient populations are different.
And just to mention some notable differences in CodeBreaK 200 as part of the eligibility criteria patients needed to have received and progressed on both platinum doublet chemotherapy and immune checkpoint inhibitors. And 98% of patients in both arms had progressed on both platinum doublet and IO. There were just a few patients that were – where it was medically contraindicated. But this compares to approximately 83% patients in CodeBreaK 100 that had received and progressed on both a platinum doublet and then immune checkpoint inhibitor. So this is a different population.
And the second point to note is that the CodeBreaK 200 population also has a higher incidence of CNS involvement, stable and treated brain mets, but in approximately one-third of patients, 34% of patients compared to approximately 23% of patients in CodeBreaK 100. And geographically, enrollment was very different in CodeBreaK 200 with approximately three quarters of patients enrolled in Europe about 12% and fewer in the U.S, whereas CodeBreaK 100 about 70% of patients were enrolled in the U.S. and approximately 18% in Europe. Finally, and we obviously don’t know all the detailed molecular characteristics of these patients, for example, the impact of core mutations, other parameters that maybe impacting ADF, either sotorasib or docetaxel. But overall, I consider this not unexpected at all.
David Reese
Great. Thank you. Jay Sean, next question?
Operator
Our next question comes from Mohit Bansal with Wells Fargo. Your line is now open.
Mohit Bansal
Great. Thank you very much for taking my question. So a couple of questions on each of the assets you presented today to [indiscernible]. So can you just help us contextualize the colorectal cancer trials and in terms of data I mean, [5% of months] sounds really impressive compared to what we have out there. So can you just have – or doctor, please help us understand what – how much unmet need [indiscernible] in that particular market?
The second question is regarding the OS trend that we have seen in lung cancer trials. So yes, this study was not powered for the OS benefit. However, I mean, even medically, it doesn’t appear in favor. Could you please speak to that? And is it the effect of prior IO because we have received that quite a time with checkpoint inhibitors that they don’t show benefit as much in the PFS setting, but they have a survival benefit which depends on latent effect even after you stop the trial, or is it across the whole effect? Thank you.
David Reese
Great. Thanks, Mohit. And I’ll ask Dr. Strickler to comment on the question regarding the colorectal cancer data and then Dr. Soria will comment on the adverse event profile and the potential protective effect of immunotherapy. John?
John Strickler
Yes. I think that there is tremendous enthusiasm on the part of the GI oncology community that KRAS G12C is druggable. We’ve seen now both single agent activity and now activity in combination with anti-EGFR therapies that clearly is superior to our currently available standard-of-care. What is particularly exciting for me is the fact that these therapies, sotorasib with panitumumab is very well tolerated by many patients. And the fact that you can achieve a reasonably high response rate with manageable side effects is very promising. And I think that the data we see for sotorasib is very credible. We’re looking at a pretty large patient population, its intention to treat, its confirmed objective response rates. And I know for my sake that I’m already trying to get access to these therapies off label with variable success because I look at what’s available to me as a GI medical oncologist, and whether it be pancreas cancer or colorectal cancer, this is clearly superior both in terms of efficacy and tolerability.
David Reese
Right. Thanks. And Jean-Charles, do you want to comment on the adverse event question?
Jean-Charles Soria
Yes. Absolutely. I mean, the tolerability of growth really depends on the population that is enrolled, and it’s very clear in the setting of CodeBreaK 200 that we have a 100% of the patients who are treated with IO. But very importantly, this is a heavily pretreated population. It was not a strictly seven, nine-only trial, up to 17% of the patients got third line or more, and we have a sense not only because it’s recruited a lot in Europe, but these are patients with heavy burdens. So all of that contributes to the AU profile that we observed, that is still very clearly and in opinion of vast majority of those who have seen the data very much in favor of sotorasib versus docetaxel.
David Reese
Thank you. Jay Sean, can we move on to the next question?
Operator
Our next question comes from Geoff Meacham with Bank of America. Your line is now open.
Unidentified Analyst
Hi. Thanks for taking the question. This is Charlie on for Geoff. I just have a question, I guess, regarding the safety profile when choosing different KRAS therapy, right? If one – more than one is available, especially how relating to liver toxicity and how would help physician choose in different line of – in the refractory setting and as well as the first line therapy with the safety profile currently in mind? Thank you.
David Reese
Sure. Maybe I’ll ask Dr. Skoulidis to comment on that and how he evaluates and choose the therapies for these patients. Dr. Skoulidis?
Ferdinandos Skoulidis
Yes. Thank you. So the first thing that I would say is that the safety profile of sotorasib in CodeBreaK 200 was very similar to what was previously reported in CodeBreaK 100. I haven’t seen any new safety signals or concerns. I mean, the rate of the hepatotoxicity to all grades was about 10% in CodeBreaK 200 was about 15% in CodeBreaK 100 in with ALT, AST elevation. And the rate of Grade 3 was about 7.8% here. It was between 5% and 6% in CodeBreaK 100. I think the important thing to note is that with standard measures involving those interruption, those reduction and treatment with corticosteroids, the majority of these cases were almost all were resolved and in only five cases was treatment discontinued because of hepatotoxicity.
I think there have been previous reports from a number of different groups regarding the potential time interval from the last dose of immunotherapy and its potential impact on the rate of hepatotoxicity with thoracic. We have not run this analysis. This is something that will be looked at in more detail in the future. So at this point in time, I think all I would say is that the rate of hepatotoxicity is manageable. I think we clearly need guidelines about how to monitor these patients and how to identify hepatotoxicity early and then interrupt treatment those reduce as necessary treat with steroids. If we do that, then in overwhelming majority of cases hepatotoxicity is manageable. And if we get a clear information regarding the impact of previous treatments, then this guidelines will need to be adjusted accordingly. But I think this is still being researched at the moment.
David Reese
Great. Thank you. Jay Sean, next question.
Operator
Our next question comes from Robyn Karnauskas with Truist. Your line is now open.
Robyn Karnauskas
Hi, thanks for taking my questions. So two, so what were the rules for crossing over in the docetaxel arm? And did you see any patients scan drugs beyond progression? I’m trying to get it just thoughts around duration of therapy. I know you talked about 26 weeks, but what might be – maybe a more realistic or real world duration for patients, especially with limited options posts. And I guess I’m long winded, but also thinking about, were there any differences in U.S. versus ex-U.S. duration? And my second question is simple. It’s for the sotorasib plus panitumumab study, those patients with very durable responses, did you tease out any characteristics biologically that would help you identify them? Thanks.
David Reese
Great. Well I’ll ask Jean-Charles to comment on the question about duration and potential U.S. versus ex-U.S. differences. And then Dr. Strickler can comment on some of the duration observations that he made in the colorectal cancer trial. So Jean-Charles?
Jean-Charles Soria
So I’m going to read the crossover first. That crossover was made at the request of the interaction with FDA. And if patients had progression confirmed centrally by [indiscernible], they were allowed to move from the docetaxel to sotorasib. And that is one of the reasons of the crossover, but certainly there are patients who did crossover outside of trial either because the investigator felt that he still wanted to move to KRAS G12C inhibitor if the bigger resource were not provided quickly enough, or if he disagreed with them.
And that’s why we have at the minimum that 34%, those are the crossovers that happened by protocol. There are crossovers that happen outside of the protocol that are documented in the CRS and there might be others that escaped ourselves. And yet there were some patients that were treated beyond progressions small numbers. We are looking into that, and that is the element of the crossover that kind of get share.
David Reese
And I would add that, in terms of duration, I don’t know that we can comment that we saw any dramatic differences between U.S. and ex-U.S. right now. It’s obviously something we will follow-up in the future in other trials. Dr. Strickler, do you want to comment on the duration question in the colorectal cancer cohort?
John Strickler
Yes. I can. I’m speaking as a clinical investigator here, not necessarily speaking on officially Amgen’s position necessarily, but we study targeted therapies quite a bit in the clinic in our GI oncology world. And we think about targeted therapies being most active when there’s oncogene addiction, when there’s pathway addiction. And traditionally in colorectal cancer, in particular, it’s known as a heterogeneous malignancy, which means that you can have rapid acquisition of different variants. And so my expectation, which I’m sure, will be followed up with liquid biopsy data is that you’ll see that those tumors that have less heterogeneity will be more addicted to KRAS G12C and you’ll get longer, more durable responses.
In terms of duration of treatment, the challenge I have as both a clinical investigator and as a clinician is that the next available therapy that I have for patients is not very efficacious and also has considerably more side effects. One of the patterns of failures that I’ve seen in the clinic is just a solitary, what we call oligoprogressive disease. So our standard clinically would be just to ablate that solitary progressing lesion, and then keep the patient on for many more months to come. So I suspect that when you actually bring this into the clinic, the progression-free survivals that you’re seeing out of the trial will not necessarily be what you see in terms of time on treatment because it’s fairly easy to treat that oligoprogressive disease.
David Reese
Great. Thank you for that perspective. Next question, Jay Sean.
Operator
Our next question comes from David Risinger with SVB Securities. Your line is now open.
David Risinger
Thank you. My question has been addressed.
David Reese
Great. Well, those are the ones we like the best. Let’s move on to the next question. Jay Sean?
Operator
Our next question comes from Michael Schmidt with Guggenheim. Your line is now open.
Kelsey Goodwin
Hey, this is Kelsey on from Michael. Thanks for taking our question. I guess regarding the CodeBreaK 200 FDA submission timeline, I guess, does the dosing data that you did after the registrational trial, does that need to be included in this package for the full approval? And then I guess when would you expect to submit this? And then maybe could you just remind us the plan forward for sotorasib monotherapy approaches in frontline non-small cell? Thanks so much.
David Reese
Yes. So of course, we don’t comment on regulatory timelines. Obviously, we just have these data in hand. We look forward to discussions with the FDA, EMA and other regulatory authorities. And obviously, one component of that discussion will be the dose comparison study, which remains on track to report out in the fourth quarter of this year. And so that discussion will determine what the overall package is. And as we have clarity on what a regulatory pathway might look like, of course, will provide guidance to you. So thank you very much. Jay Sean, why don’t we move on to the next question?
Operator
Our next question comes from Dane Leone with Raymond James. Your line is now open.
Dane Leone
Hi. Thank you for taking the questions and congratulations on the data updates at ESMO. For me, it would be helpful if you could provide any detail that you’re comfortable providing at this point and/or additional data on the timeline as you expect to head into frontline studies with sotorasib and non-small cell lung cancer. And maybe add into that, just the learnings from this study as it is the first randomized study, we’re really seeing of this agent and getting a better feel of the outcomes relative to standard-of-care, albeit in a relapsed-refractory setting and how that may inform powering and design? And to kind of in the vein of some of my peer questions earlier, how you would think about handling crossover in a frontline setting where obviously OS is going to be more power – paramount on a statistical outcome basis than it was in this study? Thank you.
David Reese
Yes. Thanks. In terms of the timelines, I don’t think we’re ready to share that yet. As I mentioned, we have the basic study design. We have discussed that with regulatory authorities and as that trial launches, and we begin enroll up, provide guidance in terms of the timelines. In terms of powering, it’s obviously meant to power, it’s powered to determine clinically relevant differences in progression-free survival, overall survival will be an endpoint because these – because the drug is now available, one would anticipate crossover of those who are not randomized to sotorasib in the frontline setting. You know, many of them we would anticipate would go on to receive KRAS G12C inhibitor for subsequent therapy. And so that consideration has also factored into the study design as we’ve designed it and as we’ve had discussions with regulatory authorities. So again, more guidance on that as we get up and running. Great. Jay Sean, why don’t we go on to the next question?
Operator
Our next question is from Colin Bristow with UBS. Your line is now open.
Colin Bristow
Hey, thanks for taking the question. Just to expand on the duration of therapy time that people be exploring. I’m just curious as to what CodeBreaK 200 data there look like versus what you are seeing from your real world experience. And you talked about, just I guess, one person’s experience in treating through progression, but what are you hearing more generally from the frontline about how this is handled?
David Reese
Yes. Maybe I’ll ask Dr. Skoulidis to comment on his real world experience with the drug. He’s treated quite a number of patients with the agent and contrast to that with the clinical trials experience, and then I can provide perspective as well. But Dr. Skoulidis, would you like to start?
Ferdinandos Skoulidis
Thanks. Dr. Reese. I’m not sure MD Anderson quantifies completely as a real world experience, but we currently in the patient treated at MD Anderson, we certainly have seen data that are very similar to what is reported in the clinical side. So duration of treatments along the lines of what is reported here is pretty much consistent with what we are observing here. But I think from colleagues and from discussions, and I think that what we are seeing is broadly in keeping with what has been reported in CodeBreaK 200.
David Reese
Yes. And I would just add that really since the inception of the program, we’ve seen a fairly remarkable consistency of effect from the first 10 patients that we treated and reported at ASCO several years ago to present. So having done oncology drug development for 30 years, it’s one of the most consistent behaviors clinically that we’ve seen. There’s still a lot to learn. Naturally, it took 40 years to get the first KRAS G12C inhibitor. These are clearly not – target is not a receptor tyrosine kinase. The behavior is different. The patterns of resistance or different, there’s a lot to learn. We’ve got a broad-based development program, attempting to address that across indication. So I think still a lot of excitement, but relatively consistent behavior to the specifics of your question. All right, Jay Sean, why don’t we go ahead? We have time for one or two more questions.
Operator
Our next question comes from Evan Seigerman with BMO Capital. Your line is now open.
Unidentified Analyst
Hi there. This is [Connor McKeon] for Evan. Thanks for taking our question. Can you just expand a little bit more on your colorectal strategy given the data you’ve shown at ESMO? What are your expectations for combination therapy going forward? Thank you.
David Reese
Yes. Maybe I’ll ask Dr. Strickler to comment on that. And then in terms of the Phase III trial and what you’re looking for there. And then I can talk about the – more broadly about the development program.
John Strickler
Yes. And as a GI medical oncologist, and as a clinical investigator, I’m involved in CodeBreaK 100 and 101 not involved in the randomized Phase III trial. But in terms of what I’m looking for, number one, we’re looking at response rate as a bonus where we have FDA approved therapies that quite frankly have nearly no responses, progression-free survival four months or greater is better than current standard-of-care and tolerability is of crucial importance, understanding that the median age for metastatic colorectal cancer and pancreatic cancer quite frankly are both right around 70 years of age. And these patients come in after a couple lines of therapy, pretty beat up by treatment. So in the real world setting that question of tolerability becomes absolutely critical to our adoption.
And I think there was an earlier question about, how we choose which treatment, particularly if there’s multiple. I think there’s no doubt in my mind that physicians are creatures of habit. And once we get familiar with the drug and comfortable with a drug, it’s very difficult to get us to use a different one. So I have encouraged colleagues at Amgen to be first because, I think that will be a critical importance as a practicing physician for what I choose. And once I get comfortable with the drug, it will be hard for someone else to come in and convince me to use a different one.
David Reese
Great. Thank you. And in terms of the broader colorectal cancer development strategy, obviously, thinking about earlier lines of therapy would involve combinations with standard backbone chemotherapy, and we’ll have more to say about that as we go along that that’s the logical next step here. All right, Jay Sean, I think we’ve got one more question right in the queue. So why don’t we go ahead and tee that up?
Operator
Our next question comes from Brian Skorney with Baird. Your line is now open.
Brian Skorney
Hey. Good afternoon, everyone. Thanks for taking the question. On the LFTs that we’re seeing in the study, I was just wondering if you’ve got a chance to look at the impact of prior PD-1 usage, particularly in terms of time from the last exposure to initiation of LUMAKRAS. I know there’s been some publications about spacing out exposure there. I just trying to get a feel for how real world this CodeBreaK 200 studies in terms of treating in the post PD-1 progression setting? And then study population here is different as you highlighted, but somewhat similar to CodeBreaK 100, that was the basis of approval. I’m just wondering how you think this really changes utilization or does it just sort of provide validation and indication that where it’s already used with level one evidence? Thanks.
David Reese
Yes. Thanks Brian. I’ll handle those questions. And certainly – so one thing we’re looking at is the duration since prior checkpoints inhibitor in the administration of the drug, I should point out the patients were required to have 30-day washout as an entry criterion onto the trial, so that helps standardize that interval. And that would be largely consistent with clinical practice for many of these patients, if they’re progressing when you get a scan and then they get ready for the next therapy, it’s often in the three or four-week time period.
So – and then our view again, based on the totality of the evidence is that this reinforces what we saw in CodeBreaK 100. And certainly, in our view, this provides irrational that this is an absolute treatment option for patients with second line and beyond KRAS G12C-mutated non-small cell lung cancer. As I mentioned, we will have discussions with regulatory authorities and will provide additional guidance on that as we go forward. And I don’t want to say anything more specific in advance to those discussions.
David Reese
Right. Well, I think we’re out of time. It’s late here for those who are in Paris. And so thank you all for joining our call. As always, the Investor Relations team will be standing by if you have additional questions and we’ll be happy to address those as they come in. Thanks, everyone.
Operator
That concludes the conference call. You may now disconnect.
Be the first to comment