Oncology Frontier: At the Poster Discussion Session of metastatic non-small cell lung cancer， you are the discussant of abstract 9011， 9012 and 9013. Could you please make some comments on the three studies?

Dr. Velcheti: I am Dr Vamsidhar Velcheti， Associate Director of Immunotherapy Research at the Cleveland Clinic. I will be discussing three very important posters presented at the Poster Discussion Session at the ASCO Annual Meeting 2017.

The first poster reviewed was the updated survival information from the KEYNOTE-001 study， the three-year follow-up of overall survival presented by the investigators of the KEYNOTE-001 study. The study was previously published in the New England Journal of Medicine in 2015. This is one of the biggest phase I studies that has been done and perhaps the most important phase I study ever done， that led to the approval of pembrolizumab in two different diseases， melanoma and non-small cell lung cancer. The phase Ib component of the study was presented at the Annual Meeting by Dr Leighl and colleagues， where they updated the overall survival at the three-year time point. This study included chemo-na?ve and previously treated patients and there was a striking overall survival of 26% in patients who were previously chemo-na?ve and 19% in patients who were previously treated at the three-year time point. What is interesting in this study is that there were no long-term， grade 3-5 toxicities noted in any of those patients. However， it doesn’t surprise me or any of my colleagues because of the uniqueness of this category of immunotherapy treatment across clinical trials with these agents in lung cancer as well as melanoma， that we are consistently seeing patients respond to these drugs for a longer duration of time. As a matter of fact， Dr Julie Brahmer has presented a five-year survival update from the CHECKMATE-003 study at the Annual AACR Meeting this year， where we are seeing a five-year survival in the 16% range. What is interesting in that study is that most of the patients in that study went off treatment with nivolumab at the two-year time point， and despite going off the treatment， they still had a three-year survival in that study of 18% which is very comparable to the KEYNOTE-001 study where patients actually continued on with pembrolizumab treatment until disease progression. It raises the question whether we really need to treat until progression with these agents. At this point， I think it is an open question， but the emerging trend in melanoma survival outcomes as well as in lung cancer is that patients may not be requiring treatment beyond a certain time， maybe 1-2 years. When you have patients on these treatments for longer durations of time， they are at risk for autoimmune side effects.

The next abstract to be discussed was from Dr Santini and colleagues and presented at the Poster Discussion Session. This was a retrospective study in 495 patients treated at Memorial from 2011 to 2016. They evaluated the cohort of patients treated with anti-PD-1/PD-L1 and also some patients who received a CTLA-4 inhibitor. In this cohort， they found that about 15% of patients went off the study because of immune-related adverse events， and of those 15%， some were rechallenged with immunotherapy and some were not rechallenged with immunotherapy. It turns out that patients who were rechallenged did not all show severe autoimmune side effects. There were grade 1-2 autoimmune toxicities and， understandably， the investigators were comfortable rechallenging those patients. This begs the question whether those patients were having a pure autoimmune condition. As clinicians treating patients with immunotherapy， often the autoimmune toxicities and side effects are based on a clinical diagnosis， for example， a mild diarrhea. That may or may not be causally associated and may be a mere temporal association that may still lead to discontinuation. That may explain the low prevalence of autoimmune toxicities in the rechallenge cohort， which was 24%. This may be due to a bias in patient selection， so the data from this study should be interpreted with great caution. The most important takeaway from this retrospective study is that for patients who discontinued therapy because of immune-related adverse events and patients who were rechallenged， the overall survival for each cohort is essentially identical. Another question we need to ask is do we really need to rechallenge somebody who has immune-related adverse events， especially if they had a response prior to developing that immune-related adverse event leading to discontinuation in the study? Overall， the study should be taken in the context of the patient’s clinical situation in my opinion. There are some patients where you have very severe autoimmune conditions and we should be very cautious about using immunotherapy in those patients. If there are mild autoimmune conditions， you can safely rechallenge them but with great caution and to importantly have a multidisciplinary approach， with a GI specialist on board where there is autoimmune colitis， for example.

The last abstract is very important. It is a pharmacoeconomic analysis by Dr Goldstein et al from Emory University. The question they were asking was whether the new label for pembrolizumab of 200mg flat dose is more cost-effective than using an individualized weight-based dose. To do this analysis， investigators looked at the US Medicare database and looked at a population-based analysis determining how many of those patients would be technically eligible for frontline pembrolizumab considering the inclusion criteria for the KEYNOTE-024 study. Based on their assessment evaluating the cost of the drug and looking at base care analysis models， they found there was an overwhelming 0.8 billion dollar saving when using a flat dose rather than individualized dosing of pembrolizumab. This is a remarkable finding， but we have to take into consideration that the clinical relevance of this study is contingent on the presumption that the efficacy between the flat dose and the individualized dose is comparable. We do know from the KEYNOTE-010 study that the 2mg/kg and 10mg/kg dosing essentially performed identically in terms of efficacy and toxicity. There were a lot of other analyses performed in terms of biological endpoints like PD-1 receptor saturation and so on， and there were no significant differences based on weight. So I think it is very important that this is considered for future drug development as we use these targeted therapies and immunotherapies， to not focus on the traditional drug development paradigm of having a maximum tolerated dose， but instead using the biologically effective dose and trying to minimize the financial toxicity in these patients.

It has been a great session and I am happy to have been able to provide this update from the meeting.

Vamsidhar Velcheti，MD是克利夫兰医学中心免疫研究部门的副主任。他在6月3日的肺癌壁报讨论专场点评了三项研究（摘要编号9011、9012和9013）。

Vamsidhar Velcheti: ASCO报道了KEYNOTE-001中Pembrolizumab一线/二线治疗晚期非小细胞肺癌的3年总生存结果。根据2017ASCO的报道，550例晚期NSCLC患者纳入该试验，101例患者一线进行免疫治疗，其3年总生存率是26.4%；449例为经治患者的3年总生存率是19%。这项研究的有趣之处在于，任何患者都没有出现3~5级迟发毒性。

摘要9012是关于肺癌患者发生免疫治疗的不良反应（irAE）后再次使用免疫药物的安全性分析。这是一项回顾性研究，在482例接受抗PD-1/PD-L1±抗CTLA-4治疗的患者中，15%出现了延迟治疗进程的不良反应，在这15%的患者中，54%患者再次使用免疫药物治疗，其中24%的患者出现了原来的irAE复发，26%出现了新的irAE，50%的患者未再发生irAE。患者出现影响治疗的AE，经处理改善后，医生可继续使用免疫疗法。问题的关键是，所谓的AE是不是纯粹的免疫治疗导致的不良反应？其他原因造成的AE也可能导致免疫治疗延迟。最重要的一点，对于第一次irAE前获得CR/PR的患者，停止治疗组和再次免疫治疗组的PFS和OS是相似的。那么，在导致治疗延迟的irAE发生前若患者已经得到缓解，再次免疫治疗还有必要吗？总体来说，解读这项研究时应当考虑患者的临床情况，如果irAE非常严重，再次予以免疫治疗应谨慎；如果irAE轻微予以免疫治疗是安全的。

第三项研究非常重要：固定剂量和个体化剂量Pembrolizumab用于PD-L1阳性NSCLC患者一线治疗的药物经济学分析。研究对比了2 mg/kg和200 mg固定剂量。相比FDA推荐的200 mg Q3W，个体化剂量每年为美国节省8亿2500万美元。当然得出这一结论的前提是推荐剂量和个体化剂量的疗效是一样的。在KEYNOTE-010研究中，2mg/kg和 10mg/kg剂量的疗效和毒性相似。研发靶向药物和免疫治疗药物时，不应像传统药物研发那样专注最大耐受剂量，而是应探索生物有效剂量，减少经济毒害性(financial toxicity)。