Welcome to the ISLC webinar, 2023 ASCO Highlights. On behalf of ISLC Educational Committee, we have planned this webinar focusing on the three important abstracts from ASCO 2023. I'm Hidehito Horinouchi, a thoracic medical oncologist at National Cancer Center Hospital, Tokyo, Japan. I will be your presentation moderator today. We will start this activity with some brief housekeeping items. This activity has been planned and implemented in accordance with the accreditation requirements and policies of the Accreditation Council for the Continuing Medical Education, ACCME. The International Association for the Study of Lung Cancer, ISLC, is accredited by the ACCME to provide continuing medical education for physicians. The International Association for the Study of Lung Cancer designates the live format for this educational activity for a maximum of 1.00 AMA PLA, Category 1 credit. Physicians should only claim credit commercial with the extent of their participation in this activity. So all the faculty, planners, and the reviewers for the webinar today have disclosed their conflicts of interest, COI. Jody Patel is a speaker for Takeda and advisor for AstraZeneca, Unheard Therapeutics, and Lilly. Stephanie So is a speaker for AstraZeneca and MSD and received research funding from AstraZeneca and Garden Health. Stephanie also is a consultant for Pfizer. I, Hidehito, am a speaker for AstraZeneca, ABI, MSD, Roche, Chubai, Bristol-Myers, Squibb, Ono, and also receive research funding from them. I'm also a speaker for Hunger. Our first presenter is Dr. Jody Patel. Dr. Patel is a director of thoracic oncology at Northwestern University, which is located in Chicago, US. Dr. Patel, I can't wait for your lecture. Please start. Thank you. Thank you so very much. I'm so pleased to present Keynote 671. It's perioperative pembrolizumab in early-stage non-small cell lung cancer. This was presented by Dr. Heather Wakeley at ASCO just several months ago and I think presents an exciting opportunity for improving cure for many of our patients. This is the presentation. This is a randomized phase three trial and as background, we understand that immunotherapies have really become a standard of care for patients with advanced and metastatic non-small cell lung cancer. A number of phase three trials have been reported in the recent past that demonstrate checkpoint inhibitors either given before or after resection of early-stage non-small cell lung cancer improve outcomes. But despite this, unfortunately, many patients still experience recurrence. Investigators felt that perhaps a perioperative approach with both neoadjuvant and adjuvant inhibition could provide benefit over just three cycles of nivolumab neoadjuvantly or one year of immune checkpoint inhibition post-resection. Moreover, there are two perioperative trials that have been recently reported, the phase three AGEAN trial and NeoTorch trials, which have shown, again, improvements in EFS for neoadjuvant or perioperative immune checkpoint inhibition. So this work builds upon these studies. This is the trial design. As stated, this is a randomized double-blind phase three trial. There were dual primary endpoints of event-free survival per the investigator and overall survival. So this is the first analysis which focuses on EFS. In terms of eligibility criteria, these patients had at least stage two to 3B disease, and 3B was identified as only N2 non-small cell lung cancer. They'd never undergone prior therapy for lung cancer, were able to undergo surgery, and had enough tissue for PD-L1 evaluation. So there were 786 patients randomized in a one-to-one fashion to receive either a cisplatin-based regimen with gemcitabine or pemetrexid and pembrolizumab versus the same with placebo patients for four cycles, up to four cycles. Patients then underwent surgery and then continued adjuvant therapy for up to 13 cycles every three weeks. The pre-specified stratification factors were stage two versus three. We're looking at high PD-L expression, so greater than 50% versus less, histology, and then geographic region. As in many neoadjuvant trials, many more patients were screened than actually went on study. And so 1,300 patients were screened, about half of them ended up being randomized with the intent to treat population. So recognize this is an early analysis. The median follow-up is just over two years. But nevertheless, what we can see is the treatment disposition. And so I think it's important to see that the majority of patients were able to complete all four cycles of neoadjuvant therapy, so 74% in each arm, again, well-matched. Only 13 in the pembro arm and 16% in the placebo arm had to discontinue all study therapy. And then the number of patients who underwent surgery was close to 80% in both arms. In terms of adjuvant therapy, again, an early analysis. So we know that 10% of patients in each arm were still undergoing treatment. And so this will certainly impact later analyses. But I would call out that 20% of patients discontinued adjuvant therapy before completion. So that's 30% of all patients who received at least adjuvant therapy. And so there is truly a discontinuation rate that we need to keep track of as the data mature. These are the baseline characteristics. So the majority of the patients were male, but again, very well-balanced between arms. 30% of patients were Asian. A small number of patients were Black across most of these immune therapy trials. 30% of patients were from East Asia. The majority of patients had non-squamous histology. And about 13% of patients in the pembro arm and 12% of patients in the placebo arm were never smokers. The majority of patients had stage 3 disease with over 40% of patients having N2 disease. As in other previous trials, about a third of patients had PD-L1 scores over 50%. A small number of patients had EGFR and ALK translocations. This was not mandated in the study. In terms of surgical details, most patients underwent R0 resections, a higher number in the pembrolizumab arm, which has been demonstrated on other studies. So greater rates of response. Most patients underwent lobectomy. But over 10% of patients were able to undergo pneumonectomy and then after immune therapy, which is certainly important. And all-cause mortality in both arms was quite low. Here's the event-free survival, and this is certainly striking. At 24 months, we see that 62% of patients remain free from recurrence or other event on the pembrolizumab arm versus 40.6% of patients on the placebo arm. That leads to a highly statistically significant hazard ratio of 0.58. Certainly there are a fair number of patients who have not had an event still, but that median EFS has not been reached in the pembrolizumab arm. And certainly 17% in the placebo arm. And we see a leveling in terms of early indications. And so one, I think, given that robust hazard ratio, is quite enthusiastic about what this will look like at future analyses. These are the event-free survivals in different subgroups. And I'd call your attention to a couple of things. So a small number of women certainly tended to fare a little bit better. Patients who are not East Asian fared a little bit better, and this might be because of molecular heterogeneity. Patients who had high PD-L1 scores had the most significant impact with PD-L1 greater than 50, a hazard ratio of 0.42. And it really didn't matter by N2 status, but certainly by pathologic stage, those with higher risk of recurrence fared a little bit better with pembrolizumab. One more thing to point out are the EGFR and ALK cohorts. These numbers are very small. And certainly patients with EGFR mutations have a very wide confidence interval. And so I would just be careful in terms of interpretation. The patients that, again, had known EGFR mutations or had no known EGFR mutation or were wild-type tended to fare quite well with pembrolizumab. This is early overall survival. And so, again, there haven't been many events. This is quite early after that median follow-up of about 25 months. But, again, it seems that pembrolizumab and patients who have been treated with pembrolizumab are doing a little bit better. We look at the median for placebo is 45.5 months, and it has not been reached in pembrolizumab. That hazard ratio is 0.73 but has not reached statistical significance. You can see a large number of patients have not been evaluated yet. Another exploratory outcome was pathologic response. So major PRs were significantly higher in the pembrolizumab arm, 11 versus 30%. And then path-complete responses were very close to what we've seen in some of the other trials with chemotherapy at 4% and then significantly higher, a change, a difference of a delta of 14% in patients who received neoadjuvant pembrolizumab. In terms of toxicity, most patients tolerated the pembrolizumab. Clearly, there were treatment-related AEs with chemotherapy as well, but really the immune-mediated AEs, 6% that were grade 3 to 5 in the pembrolizumab arm, 5% were educated as serious. And then one patient, unfortunately, led to a fatality. So in summary, certainly neoadjuvant pembrolizumab plus chemotherapy followed by surgery and adjuvant pembrolizumab is feasible and provided superior EFS compared to chemotherapy and surgery alone. And that comes with, I think, a very significant hazard ratio. And the EFS benefit was across all subgroups that were analyzed. Path CRs were higher with pembrolizumab. And then I think this, we'll talk a little bit about the EFS benefit for perioperative pembrolizumab, whether or not patients had a path CR. We're still understanding what OS benefit will look like, and we'll continue to watch this when events have been reached. There were no new AEs, and certainly I think this is a very promising regimen for perioperative pembrolizumab. So really, how do we put this into context? Because we have a number of options. So we know that over the past several months, across the world, a number of adjuvants or immunotherapy strategies have been incorporated. So we have adjuvant atezolizumab for patients that have stage 2 to 3 and at least PDL1 tumor proportion scores of 1%. Checkmate 816 led to rapid adoption, I think, of neoadjuvant nivolumab for patients with 1B tumors greater than 4 centimeters. And then more recently, we have adjuvant atezolizumab, excuse me, adjuvant pembrolizumab for 1B to 3 tumors, regardless of PDL1 status. And then in the past two months, we've seen all these perioperative strategies with Neotorch, Aegean, and Keynote 671. So where does this fit? We can certainly say without hesitation that there's impressive EFS on Keynote 671, but is this really truly better than Checkmate 816? Clearly, we hate to do cross-trial comparisons, but we have more mature data with Checkmate 816, and we know that in Checkmate 816, we have three-year overall survival data now in which the hazard ratio for nivolumab is 0.62. Certainly in Keynote 671, we're seeing this impressive EFS, but is that really being driven by the neoadjuvant component? It's really difficult with this trial design to tease out what's being driven with the adjuvant therapy of one year. One other consideration is, can we use pathologic response to help really inform these treatment decisions? So we know that patients with PATH-CRs tended to do quite well in terms of event-free survival. This is reminiscent again of Checkmate 816. Those even without PATH-CR, however, tended to do a little bit better. So is this delta between the two lines really significant, and how can we further improve those patients' outcomes who have not reached PATH-CR? Is that really driven by the adjuvant therapy? So I think our questions are really, what additional therapy should we do for patients without PATH-CR? Then I think the most pressing question for our field right now is that we have a number of opportunities to improve outcomes, but really, what are the cumulative toxicities? Certainly one year of therapy after surgery is very different than three cycles of neoadjuvant therapy. What are the chronic toxicities that we see in terms of immune events over time? Then clearly the financial toxicity, certainly one year of adjuvant therapy is not inexpensive and puts strains on all health systems. So we can say, I think without, again, hesitation, 671 and other perioperative studies really are a major advance in a new standard of care for the treatment of patients with resectable non-small cell lung cancer. What patients benefit from neoadjuvant therapy versus adjuvant therapy really needs to be done in a multidisciplinary setting, and then the decisions about further adjuvant therapy, I think, will be based likely, again, cooperation with our pathologists and advances in predicting biochemical or biomarkers that predict good outcome. We need to further study this additive impact of adjuvant immunotherapy after neoadjuvant chemoimmunotherapy. I think that is sort of the pressing decision, and ultimately we would imagine that optimization of treatment selection will really incorporate escalation and de-escalation strategies based upon biomarker analysis and clinical response. Thank you, Dr. Patel. I'd like to discuss with our faculties about which patient subset will be suitable for the perioperative ICIs rather than preoperative alone. Based on the information we have right now, we have plenty of trial data. So I would like to start this question to Navneet first. Yeah, thank you. So I think this is an exciting era in which we have seen the advent of, you might say, integration of immunotherapy along with chemotherapy in both the neoadjuvant and the adjuvant setting. It's very hard right now to make a call as to which strategy to adopt as a uniform strategy, and it's ultimately, as Dr. Jyoti said, it has to be a multidisciplinary decision. There are going to be a subset of patients in whom you might want a neoadjuvant approach of chemoimmunotherapy together to downsize tumors, tumors in which the surgeon might feel that doing upfront surgery may be a difficult decision, and who obviously have to be fit enough to tolerate the combination of chemo immuno upfront. And I think, so this is basically a shared decision-making, both inputs from multidisciplinary teams, as well as patient and caregiver preferences as to what they would like. The ultimate aim is going to be to have the most effective treatment given, which is best tolerated and having the best results. Thank you. So I'd like to pass to Stephanie, the same question. Yeah, so I think if we're thinking about which patients that we want to choose for a neoadjuvant versus perioperative neoadjuvant chemo I.O. strategies, it's a little bit difficult now. I think we're in a bit of a conundrum about, you know, which combination of drugs we should use upfront. What I would say is that I think it is quite clear that I think different patients actually benefit from different approaches. And I think it's nice that we can use pathologic response to categorize them into different risk baskets. So those that achieve pathological complete response, perhaps maybe a lot of them don't even need for the therapy after surgery, whereas those that don't achieve FCR, then I think, you know, they probably fall into a bag of patients that, again, need to be risk stratified further, and those that maybe benefit from I.O. alone, or maybe they need something more than the adjuvant I.O. after surgery. And I think, you know, future trials will be really important to try and dissect what will be the optimum strategies for these different patients. Thank you. So I have one great question from the audience. So Dr. Akashi Desai presented one great question. So I'd like to request Jyoti to answer this comment so that I will share the question in the live answering mode. So what does the panel think is the role for the adjuvant therapy in patients who did not have PCR and so on? So Jyoti, could you respond to your question? It's a great question. So that's a wonderful question. So unfortunately, there are a significant subset of patients who don't attain path CR, and even those that have minimal treatment response. So my sort of piece would be that patients who have major pathologic response or major response, those patients likely would benefit from immunotherapy alone. However, in patients who've had minimal response to induction off of a trial, I may want to give additional chemotherapy based upon what I'm seeing. Certainly after four cycles of platinum-based or cisplatin-based induction on a trial such as this, additional therapy after surgery might be difficult. My hope is that we'll have then complete biomarker analysis. And then perhaps this is when we can start thinking about bringing other agents into the mix. And that may be sort of our next strata of trials. Another important piece of patients who get perioperative strategies versus just adjuvant strategies is what biomarker analysis we have. So patients with stage three disease by and large have had EVAS and had mediastinal staging. And so we can do NGS. And the patients with stage two disease might represent another group of patients in which we haven't complete information. And so it may be that those patients end up getting more or more likely to get resection. And then we can sort of tease out based on molecular analysis what their best therapies would be. Yeah, thank you for your summary. So the time is limited so that I would like to move to the next session. Thank you for sending the questions. So before proceeding to the next session, our next section will be the Adola. Then we will go into the keynote 789. And we will have the similar question and answer session after the Adola presentation by Dr. Singh. So if there is any questions from the audience, please send it even during the presentation. So thank you, Dr. Patel. And I would like to proceed to the next presentation. Our next presenter is Dr. Navneet Singh. Dr. Singh is a professor of pulmonary medicine and a faculty in charge of the lung cancer clinic in Chandigarh, India. Thank you. Thank you, Dr. Singh. So good morning, good afternoon, good evening to all the attendees, depending where you are located in the world. And thank you again for joining us for the ASCO 2023 highlights. I'll be focusing on the Adola trial, which is basically looking at targeted therapy in resectable disease. This was presented by Dr. Roy Herbst in the plenary session of the ASCO annual meeting last month. And we are all aware of the phase three Adola trial design. It's enrolled patients with completely resected stages 1B to 3A NSCLC, patients who had to be 18 years or older, a performance status of zero to one, a confirmed non-squamous histology and harboring one of the two common EGFR mutation types, that is exon 19 deletions or the Elliot-Fibrate R-point mutation on exon 21. They should have undergone complete resection with negative margins. And the maximum internal between surgery and randomization was 10 weeks if they did not receive adjuvant chemotherapy or 26 weeks if adjuvant chemotherapy had been given. Patients were stratified by stage, the type of EGFR mutation and race. And they were randomized into receiving either OC-multinib 80 milligram once daily or placebo. The planned treatment was for three years and treatment was continued until disease recurrence, treatment completion or discontinuation criteria was met. The primary end point for this trial was disease-free survival for stages two to 3A. And the key secondary end points was DFS in the overall population, which was stage 1B to 3A, overall survival and the safety and health-related quality of life. Now, if you look at the baseline characteristics of the overall population, you can see that the majority were females, never smokers, Asian, and had a performance status of zero. Now, importantly, the stages 1B, 2 and 3A were all very equally batched, approximately one third each and a slightly higher proportion of exon 19 deletions. And 60% of patients in each group had received adjuvant chemotherapy. Now, we are all aware of this, that adjuvant OC-multib significantly improved disease-free survival. The primary analysis was published initially in NEJM in October, 2020 at 29% maturity, in which case there was a very impressive hazard ratio of 0.2. An updated analysis of this was published in the Journal of Clinical Oncology earlier this year with a 45% maturity and OC-multimedian survival was 66 months versus 28 months in the placebo arm, again, with a hazard ratio, which was very impressive of 0.27. As expected, adjuvant OC-multimed also significantly improved CNS disease-free survival because this drug has excellent brain penetration. And although the maturity was only 13%, the hazard ratio was 0.24 in favor of OC-multimed. And this is what everybody was waiting for, the overall survival in stage 2 to 3A disease. As you can see that this was read at 21% maturity with a median follow-up of 62 months for OC-multimed and 60 months for placebo. And the 5-year OS rate was 85% with OC-multimed and 73% in the placebo arm, again, with an overall survival hazard ratio of 0.49, which was very highly statistically significant. Even for the overall population of stage 1B to 3A, the hazard ratio was almost identical at 0.49 and the 5-year overall survival rate was 88% in OC-multimed arm and 78% in the placebo arm. If you try to look at the subgroup analysis, it did seem that the benefit was seen in almost all subgroups, but we will talk about this in a little more detail shortly. So if you look at overall survival by disease stage, what you may be able to appreciate is that for stage 1B and stage 2, the incremental benefit was approximately 6% to 7% for OC-multimed as compared to placebo, but this benefit was much more marked for stage 3A disease, approximately 18%. And that is why the hazard ratios were the best for stage 3 of 0.37. In fact, for stage 1B and 2, this did not reach statistical significance on subgroup analysis. If you try to compare the benefit in patients with and without adjuvant chemotherapy, what you can see is both subgroups benefited almost equally with a 10% difference between OC-multimed and placebo for survival and the hazard ratios being almost identical at 0.49 and 0.47 respectively. Now coming to subsequent treatments, 22% in the OC-multimed arm and 54% in the placebo arm received subsequent therapy. EGFR-TKI is one of the most common treatment given and OC-multimed actually was the most commonly used drug even post-progression. Coming to the safety summary, what you can see is that grade 3 or higher AEs were significantly more common with OC-multimed, 23% versus 14% for placebo. AEs leading to discontinuation, 13% with OC-multimed versus 3% for placebo. Those reductions, 12% with OC-multimed, 1% with placebo, and those interruptions, 27% with OC-multimed versus 13% for placebo. And therefore the authors had concluded that the DFS benefit at Adora has translated into a statistically significant overall survival benefit with use of adjuvant OC-multimed, both for the primary population, which was stage 2 to 3a, and for the overall population, which was stage 1b to 3a, with almost identical hazard ratios of 0.49. And therefore Adora became the first phase three study to demonstrate clinically meaningful overall survival benefit with use of targeted therapy in resected early stage NSCLC harboring common EGFR mutations. But like all important things to be done, one has to put the findings into perspective and by looking deep into the study. Now this is a busy slide, but it summarizes three other important studies in the adjuvant study. The E1 trial, which was a phase 2 trial, and the IMPACT and the ADJUVANT trial, which were phase 3 trials. So the patient numbers in all of these was much less than the Adora trial, but the important thing to realize is that they used a first generation drug, either Arlotinib or Jefitinib, as the EGFR TKI, and compared it to chemotherapy, which was minorelbin cisplatin. And this is in contrast to OC-multinib, which was compared to placebo in the Adora trial with or without adjuvant chemotherapy. Again, the treatment duration was two years for the other three trials, and it was three years for Adora. Now the long-term DFS benefit of five years was not significant in IMPACT or ADJUVANT, but it was significant in the E1 trial. And similarly, long-term overall survival benefit was significant in the E1 trial, but not in the other two trials. As we have seen for Adora, the hazard ratios were very impressive for both four-year DFS as well as for five-year overall survival in both the overall population, which is 1b to 3a, and the target population, which was stage 2 to 3a. Now, as I said earlier about the subgroup analysis, it did seem that stages 1b and 2, as well as patients with the exon 21 L858R mutation did not benefit from the adjuvant OC-multinib approach. But one has to realize that subgroup analysis, unless specified a priority, are not power to detect statistical differences. The other thing one has to be aware of is the health-related quality of life and safety from this trial. Now, this has been published in Journal of Thoracic Oncology and it shows that the baseline QL scores were similar in both arms. The differences in the physical domain of the QL questionnaire between OC-multinib and placebo were minimal at all time points, including a treatment discontinuation, and there were no clinically meaningful changes from baseline in either group. And the conclusion was that the quality of life was maintained with three years of adjuvant OC-multinib treatment, and this is reflected in the graphs on the right side. But you could also argue that this may raise the question that why was quality of life not better in the adjuvant OC-multinib despite phenomenal benefit in DFS and overall survival? There are some other unanswered questions as well. Is the three-year adjuvant OC-multinib duration appropriate? Can you have a lesser duration of treatment for a subgroup of patients? If so, it would lead to a reduction in both treatment costs and potential treatment-related adverse effects. And on the other hand, are there patients in whom stopping at three years may not be warranted? Should patients with exon 19 deletion be treated differently from those with exon 20 L858R mutation? And finally, can a similar adjuvant-targeted therapy approach be followed for ALK rearrange and SCLC? Now, we are aware that the FLORA trial was a landmark trial which looked at OC-multinib versus the first-generation EGFR drugs in the advanced metastatic disease. And so if you look at the subgroup analysis from the FLORA trial versus the ADORA trial and focus on two things, and that is the race and the type of EGFR mutation, what you may be able to appreciate that for some reason, Asian patients in both the trials and the L858R mutation in both trials did not seem to benefit as much as did exon 19 deletions and non-Asian patients. And therefore, this may raise the question that perhaps one size doesn't fit all, and there is a need for a precision strategy while adopting precision oncology. And therefore, the way forward may be actually to look at ctDNA and do ctDNA-based treatment decisions, both for unreceptible and receptible disease. And in case of receptible disease, look at MRD or minimal residual disease to guide the choice and the duration of adjuvant therapy. And this was an excellent discussion by Dr. Benjamin Solomon at ASCO on this trial. He had said that there are going to be a subgroup of patients who will have poor outcomes despite OC multilevel, and whom intensification of therapy and perhaps combining with other treatments may be warranted and there is going to be a big group of patients who will have good outcomes in whom deintensification of therapy can be considered. We also need to think that EGFR mutations are not just the exon 19 deletions and the L-rate 5-8Rs. There are so many other uncommon mutations and there are so many other oncogenic driver alterations for which targeted drugs are available. And in fact, if you talk of India, ALK prevalence is approximately 10% and represents the second largest group of oncogene-driven NSCLC. And therefore, the question is whether you can use the same approach in ALK-positive NSCLC or not. So the Alchemist trial had a form in which ALK-rearranged NSCLC following surgical resection were randomized into crizotinib or placebo. And similarly, there's the ALENA trial, which is ongoing, which is looking at adjuvant electinib in ALK-positive NSCLC following surgical resection. And we hope to get answers from these to guide us in the future. Now, this was the summary from the presenters at ASCO and they said for the future, OC Mertinib trials are going to look at the new adjuvant setting, look at stage 1A patients, and look at the five-year outcomes. But we also need to keep into perspective that OC Mertinib may not be accessible in all areas of the world and may have a big economic burden. And therefore, our future scenarios are trying to determine the optimal duration of adjuvant EGFR TKI therapy. And if needed, looking at the more detailed genomic profile and use MRD to guide this tailored therapy. So the key takeaways from the ADORA trial are that it is the first phase three trial of a targeted agent in the adjuvant setting to demonstrate an overall survival benefit in resected non-small cell lung cancer. Adjuvant OC Mertinib is the standard of care for resected NSCLC, harboring common EGFR mutations. But one has to keep in mind three A's related to treatment, which is access, affordability, and adverse effects. And all of this is possible only if you do biomarker testing. And therefore, we at ISLC as it came out with an expert consensus recommendation on biomarker testing for both metastatic and non-metastatic diseases has been published in the Journal of Thoracic Oncology. And it highlights the importance of testing for at least EGFR and PD-L1, even for early stage resected NSCLC. Thank you very much. Thank you, Dr. Shin. So now it's open for the discussion and the question. So I already have the two questions from the audience so that I will answer this in the live mode. So the first one is from Dr. Devala-Bruno. So Dr. Shin, in your opinion, what is the role of adjuvant chemotherapy in this patient population? So do you have any comments? Yeah, so I think this is a very important point and this is something which really everybody needs to be aware of. Biologically targeted therapies, they are not as, they are cytostatic as compared to chemotherapy which is cytotoxic. And therefore, when we are trying to achieve a cure, that is the aim. Patients who are eligible and warrant chemotherapy should get chemotherapy first and should be followed up with adjuvant osematinib. In fact, that is what we came up with our stage three NSCSC guideline update at ASCO. The temptation to avoid giving adjuvant chemotherapy is strong, but one must not do it at the cost of reducing the success of a cure. And therefore, patients who are fit for adjuvant chemotherapy and need chemotherapy, and that would include all stage three NSCSC patients and the vast majority of stage two patients should get adjuvant chemotherapy and then followed up with osematinib as adjuvant treatment. Obviously, stage one B patients typically do not need adjuvant chemotherapy and therefore they don't need to be offered adjuvant chemotherapy as such and you could go directly to adjuvant osematinib. And I think that is my take on it, but I would be happy to welcome suggestions from our other esteemed faculty. So how about Stephanie and Jyoti about this same question? Yeah, I would agree with Dr. Singh that I don't think adjuvant osematinib negates the benefit of adjuvant chemotherapy. I think we've used it for many years and the benefit and overall survival has been proven with longitudinal mature follow-up. So I would still recommend adjuvant chemotherapy for the majority of stage two and three lung cancer patients after surgery if they are fit enough, yeah. Thank you. How about you, Jyoti? I would absolutely agree with that recommendation and I think another twist on it might be in a patient with known stage three disease and an EGFR mutation, I would give chemotherapy up front, right? To, in the same idea of giving something that is cytotoxic to that patient. Yeah, I also agree. This is a very great time for that or even for the cytotoxic because we have many antiemetics effective ones in comparison with the 20 years or before in the cisplatin adjuvant trial of the ages. So I'd like to move to the last one question. So for the time limitation, so how do you, so this question is from Dr. Junko Tanizaki. How do we determine the appropriate postoperative period in the future? So I'd like to ask this question to Stephanie. So, yeah, I think that is a great question about the duration of adjuvant osmotinib. I think another important concept here is to recognize that not all patients are the same. Again, they probably do have different risk baskets and I really liked that slide by Dr. Solomon, which showed that some patients actually probably need something more or something other than adjuvant osmotinib. Probably a good number of patients do benefit from three years, but some could even get away with potentially less than that. So I do think that the future generation of trials should focus on this concept of risk stratification so that we can identify the optimum duration of therapy rather than use a one size fits all approach depending on the patient's risk profile. And I think this would be a combination of things like MRD, potentially pathological response and also other multi-omic approaches. Thank you. So time flies, so we should go ahead. So I would like to move to the next speaker. So thank you, Dr. Singh. So now I am going to turn it over to our final presenter, Dr. Stephanie So, who is a consultant in the Division of Medical Oncology and Clinical Assistant Professor at the National Cancer Center Singapore. So could you please start your speech talk? Okay, thank you very much. I'd like to thank the organizing committee for inviting me to discuss Keynote 789. This was a phase three trial of pemtrexid and platinum with or without pembrolizumab for TKI resistant atrial front mutated metastatic non-squamous lung cancer that was recently discussed at ESCO. So this was presented by Professor James Yang and this slide details the study design. So again, this was a phase three randomized study looking at stage four non-squamous lung cancer patients with an EGFR exon 19 or L858R mutation. They needed to have a good performance status with progressive disease after first or second generation EGFR TKI without a T790M mutation or after first and second EGFR TKI with a T790M mutation and osmotic failure. So the second group of patients they included were those who had osmotic failure with a T790M mutation after progressing on first or second generation EGFR TKI. And last group was those with osmotic failure as first line therapy, regardless of T790M. So patients were stratified by PD-L1 TPS score using a cutoff of 50% whether they had been treated with prior osmotic as well as their geographic region East Asian versus non. Patients were then randomized in a one-to-one ratio to receive pembrolizumab or placebo in combination with pemotrexate and platinum therapy. They were given maintenance after four cycles and those in the placebo arm were given the option to cross over at the time of progression. So the study adopted a dual primary endpoint study design of progression-free survival, as well as overall survival. Key secondary endpoints included overall response rates and duration of response, safety, as well as patient reported outcomes. So this is the consort diagram. About 492 patients were randomized. A vast majority of them received the allocated treatment. At the time of data reporting, most patients had discontinued on their treatment largely due to disease progression. And in the placebo group, approximately a third of them crossed over to receive subsequent pembrolizumab. So this slide details their baseline demographics as well as patient characteristics. I think between the study as well as the control group, they were pretty well balanced in terms of their characteristics, including their age, their gender, prior treatment for chemotherapy and TKI. About more than half of them had an exonitin deletion and about 40% of them had a T sub-90M mutation. There was a slight preponderance of patients with the ECOP performance status of one in the pembrolizumab versus the placebo arm. So this was the first primary endpoint of PFS at the interim analysis too. So this was a negative study. The addition of pembrolizumab to chemotherapy did not significantly improve progression-free survival over placebo. This was a median PFS of 5.6 versus 5.5 months, translating to a hazard ratio of 0.8 that did not meet the pre-specified boundary of statistical significance. We then have the second endpoint, overall survival. So again, this was negative. So the addition of pembrolizumab only resulted in a numerical LPA modest improvement in overall survival of 15.9 versus 14.7 months, translating to a hazard ratio of 0.84 that again was not statistically significant. So this forest plot illustrates the overall survival across the pre-specified subgroups. And I think unfortunately there was no real subgroup that seemed to benefit from the addition of pembrolizumab to chemotherapy. So the authors did an exploratory analysis of overall survival by their PD-L1 TPS score using a cutoff of 1% and above. So what they observed was that in the PD-L1 positive population, these patients seem to derive more benefit from the addition of pembrolizumab with a median PFS of 18.6 versus 14.1 months, translating to a hazard ratio of 0.77. This is in contrast to the PD-L1 negative population where these patients did not seem to benefit as much from the addition of pembrolizumab. As this was an exploratory analysis, this difference was not formally tested statistically. So looking at antitumor activity, the overall response rates were pretty similar between the study as well as the control arm at 29% versus 27.1% respectively. In terms of duration of response, again, pretty similar at 6.3 months versus 5.6 months in the control arm. Okay, so this slide summarizes the adverse events. Pretty similar incidents of adverse events in both arms. More than half of the patients had a grade three to five adverse event, and most of them were attributed to be treatment-related. However, there was a generally low incidence of discontinuation of any treatment as a result of adverse events. I highlighted here, there was a low incidence of grade three and above immune-mediated adverse events in the pembrolizumab arm. So in conclusion, the authors concluded that the addition of pembrolizumab to chemotherapy only resulted in a numerical but not statistically significant prolongation in PFS as well as OS versus placebo in combination with chemotherapy among patients with TKI-resistant EGFR-mutated metastatic non-squamous lung cancer. The AEs were generally manageable with no new safety signals, and they observed that these results were generally consistent with prior findings that TKI-resistant EGFR-mutated metastatic lung cancer, patients tend to derive less benefit from PD-1 or PD-L1-based therapy as compared to the EGFR wild-type population. So I'll now go through some studies going through data for the use of checkpoint inhibitors in EGFR-mutated lung cancer. So this was a meta-analysis published in 2016 highlighting the limited efficacy of immune checkpoint inhibitors in the EGFR mutant population over docetaxel. This was in contrast to the EGFR wild-type population where most patients seem to derive more benefit from the use of checkpoint inhibitors over chemotherapy. We then have the studies looking at the quadruplet regimens of a checkpoint inhibitor in combination with the anti-VEGF as well as chemotherapy. So there are two studies looking at this. The first is Empower 150. So just to highlight that, this was a small population of approximately 50 patients with EGFR mutations who had received prior TKI, and this was the ABCP regimen that seemed to achieve a significant PFS benefit over bevacizumab in combination with chemotherapy, but it has a ratio of 0.42, and this difference was statistically significant. However, this did not translate into a significant OS benefit, but this could potentially be confounded by the small numbers. So next we have the ORION31 study. So this was a Chinese study with three arms. The quadruplet regimen was centilumab with a bio similar to bevacizumab in combination with chemotherapy. And again, the quadruplet regimen also significantly improved progression-free survival versus chemotherapy alone, translating to a hazard ratio of 0.51, and the result was statistically significant. Unfortunately, this quadruplet regimen, again, did not translate to an improvement in overall survival significantly with a hazard ratio of 0.98. So the next question is whether we should add a checkpoint inhibitor to chemotherapy, and we do have three studies which try to address this question. The first one is ORION31, which we alluded to in the previous slide. So this was a three-arm study. One of the study arms was centilumab in combination with placebo and chemotherapy, which they compared against chemotherapy. Next, we have the CHECKMATE722 that was presented at ESMO Asia last year, as well as KEYNOTE789 that we recently discussed. These two studies had pretty similar study designs. Both of them were looking at the addition of nivolumab and pembrolizumab, respectively. Next, we have CHECKMATE722, pemtrexid platinum chemotherapy. So ORION31 was actually a positive study. The authors observed a significant PFS benefit with the addition of centilumab to chemotherapy, resulting in an improvement in the median progression-free survival from 4.3 months to 5.5 months, although this margin of benefit was more modest as compared to the quadruplet therapy with a hazard ratio of 0.72, which is very significant. And again, unfortunately, this improvement in PFS did not translate to a significant OS benefit. The forest plot on the right highlights that patients with the eGFR axonitin deletion or T790M mutation, as well as those who had received two pyrolines of eGFR TKI, unfortunately did not seem to benefit from the addition of centilumab to chemotherapy. We then have the results of CHECKMATE722 and KEYNOTE789, which both observed no significant benefit in terms of PFS or OS with the addition of nivolumab or pembrolizumab to chemotherapy among patients who had progressed on eGFR TKI in the eGFR mutant population. So I'd like to highlight some differences in the study design in these three studies. I will not go through in too much detail in the interest of time, but just to highlight some key differences, including the number of study arms, as well as the inclusion of the types of eGFR mutations. So KEYNOTE789 only included axonitin or L858R, whereas the other two studies also included uncommon eGFR mutations. There were some differences in the choice of stratification factors, primary endpoint, the use of placebo or platinum in the control arm. And in terms of statistical power, just to note that for CHECKMATE722, they had difficulties with slow accrual, so the trial had to undergo a protocol amendment midway to amend it to a smaller number of patients that inadvertently resulted in some degree of underpowering of the study. So I think some take home points here, the addition of pembrolizumab to chemotherapy did not significantly improve PFS nor OS among patients with TKI resistant eGFR mutated lung cancer. And the results of the study was consistent with the prior results in CHECKMATE722 that was presented last year. And I think we would all agree that there are better biomarkers needed to identify patients who may potentially benefit from checkpoint inhibitors post TKI resistance in the eGFR region population. Thank you very much. Thank you. Thank you. Great presentation. So I would like to take some questions from the audience. And I already have the one question from Dr. Takayuki Takahama from Japan, and I will live this question. So this question is especially about the Fitch patient, about the role of the ICIs in the eGFR mutation positive patient population. And also if there is some lowly of ICI of each patient population subset, should we use this? So Stephanie, could you please answer this question? I think with both the results from KEYNOTE789 as well as CHECKMATE722, I think they're quite convincing that we should not add a checkpoint inhibitor to chemotherapy, if we choose to use that as second line therapy post TKI resistance. In terms of whether we should use the quadruplet regimen, again, I think that is, it looks promising, but I'm not certain yet of which patients I would be more inclined to use a quadruplet regimen over platinum doublet chemotherapy. Are there good enough biomarkers to help inform which patients may potentially benefit from checkpoint inhibitors? At this moment, I do not think so. I think in both CHECKMATE722 as well as KEYNOTE789, authors in both studies did seem to observe maybe some slightly more benefit in the PD-L1 positive population. But again, I don't think the signal is strong enough to convince us of the robustness of PD-L1 as a biomarker here. So I do think there is more work to do in this space. Thank you. Any other comments from the other faculties? Yeah, I'd just like to reemphasize that immunotherapy alone or perhaps even a combination is not something that really works for EGFR mutated NSC-LC. We know that from the metastatic setting and the pointers from all of these new adjuvant and adjuvant studies also seems to indicate more or less the same thing, because the PD-L1 expression really does not matter in the presence of an EGFR, sensitizing EGFR mutation. It is not the PD-L1 which is driving the tumor, it is the EGFR mutation. And what we have seen in the metastatic setting is also to some extent being seen with the use of immune checkpoint inhibitors if used in EGFR positive early stage. Thank you. Thank you for joining us today.

webinar

ASCO 2023

pembrolizumab

non-small cell lung cancer

chemotherapy

adjuvant osmirtenib

EGFR-mutated

disease-free survival

immune checkpoint inhibitors