ES09.03 The Role of Biomarkers for Immunotherapy in the Adjuvant and Neoadjuvant Setting

Abstract

For patients with NSCLC and absence of oncogenic addiction immune checkpoint inhibitors targeting PD1/PD-L1 change the therapeutic field. These treatments became standard of care in metastatic setting and are rapidly emerging in adjuvant and neoadjuvant setting. While more than 50% of patients with operable NSCLC will experience recurrence after surgery, neoadjuvant or adjuvant chemotherapy provides only a modest improvement in overall survival and could be highly toxic. Recent clinical trials underline the high rate of major pathological response in neoadjuvant setting and increase progression free survival in adjuvant setting. While such data will rapidly be practice changing immune checkpoint inhibitors are not effective in all patients and like in metastatic setting biomarkers remains unmet need. Various biomarkers are currently testing going from simple clinical biomarkers, to complex genomic and transcriptomic tests. Most biomarkers seem to behave similarly in adjuvant/neoadjuvant setting and in metastatic setting. Clinical patient’s characteristics are important to predict efficacy of adjuvant or neoadjuvant treatment. In IMPOWER 10 trial, testing adjuvant treatment with atezolizumab, oncogenic addiction with EGFR mutation or ALK translocation are associated with absence of efficacy of atezolizumab (1). Tumor with EGFR and ALK genomic alteration tumor are frequently poorly infiltrated in CD8 T cells and present poor level of mutations which render them resistant to immune checkpoint inhibitors. PD-L1 is also an important biomarker of efficacy in early NSCLC. In Checkmate 816 trial (2) which tests neoadjuvant chemotherapy versus nivolumab plus chemotherapy underlines that PD-L1 >1% is associated with higher rate of pathological complete response but nivolumab enhance efficacy of chemotherapy in either PD-L1 positive or negative tumors. In contrast in IMPOWER 10 trial (1), atezolizumab improves progression free survival only in patients with PD-L1 positive tumors. Efficacy proportionally increase with PD-L1 expression. Interestingly, similar observation was previously performed in PACIFIC trial (3). Importantly, a post-hoc analysis underlines that durvalumab give survival benefit only in PD-L1 positive patients raising question on this utility in this subset of patients. Smaller studies testing anti PD-1 alone or in association with anti CTLA-4 in neoadjuvant setting observed major responses irrespective of PD-L1 status (4,5). So, PD-L1 expression is an imperfect surrogate marker of hot tumors richly invaded by anti tumoral CD8 T cells. Such data underline the difficulty to use PD-L1 to avoid immune checkpoint inhibitors usage. In addition, PD-L1 assessment in neoadjuvant setting only relies on a tumor biopsy which is an imperfect surrogate marker of PD-L1 expression in the whole tumor. Tumor mutational burden (TMB) represents the total number of mutation found in coding region of cancer cell genome. High number of mutation is correlated with high capacity of cancer cells to present neoantigens, and cancers cells which present high level of neoantigen are more easily recognized by T cells. In CHECKMATE 816 (2) high TMB is associated with better response to neoadjuvant immunotherapy (pathological complete response of 22.44% for TMB low and 30.8% for TMB high). Similarly, in a small study using nivolumab and ipilimumab in neoadjvuvant setting, there was a significant correlation between the pathological response and the pretreatment TMB (5). Additional works are awaiting to determine if HLA status or intratumoral heterogeneity are also important to predict response to adjuvant or neoadjuvant immunotherapy like in metastatic setting. In addition to such classical biomarkers some emerging biomarkers are described in small cohort of neoadjuvant immunotherapy. Gut microbiome could influence response to checkpoint inhibitors by different mechanisms like modulation of tumor microenvironement or bacterial mimicry. In NEOSTAR trial (4), higher abundance of Akkermansia in baseline samples of gut microbiote is associated with better response to nivolumab + ipilimumab and is positively correlated with intratumoral immune response. Similarly, Akkermansia was also associated with favorable clinical outcomes in patients with melanoma, NSCLC and renal cell carcinoma receiving immunotherapy in metastatic setting thus confirming the important role of gut microbiota and akkermansia in the response to immunotherapy.