FP12.05 Molecular Landscape of Primary and Acquired Resistance to Immune Checkpoint Inhibitors in Chinese Advanced Non-Small Cell Lung Cancer

Xiaomin Niu,Zhen Zhou,Yu Yang,Lan Shen,K. Liu,Jian Bai,Aixia Yang,Lin Wu,Shun Lu
DOI: https://doi.org/10.1016/j.jtho.2021.01.131
IF: 20.121
2021-01-01
Journal of Thoracic Oncology
Abstract:Immune checkpoint inhibitors (ICIs) have improved the prognosis of non-small cell lung cancer (NSCLC) patients. However, about 80% of patients do not respond at all ("primary resistance"), whereas others initially respond to immunotherapy, later relapse and develop therapy resistance ("acquired resistance"). Absence of PD-L1 expression is well regarded as a biomarker of primary resistance to ICIs, and tumor mutation burden (TMB), microsatellite instability (MSI), HLA genotype and tumor neoantigen burden (TNB) have also been reported in recent studies to be related to the resistance of immunotherapy. This research is to explore the molecular landscape of primary and acquired resistance in Chinese advanced NSCLC treated with anti-PD-L1/PD-1 antibodies based on results of whole exome sequencing (WES) and targeted sequencing (TS) containing comprehensive immune molecular markers. WES (∼39Mb CDS of over 18,000 genes) or TS (∼1.6Mb CDS of 654 genes) were conducted on 50 samples from 45 patients received anti-PD-L1/PD-1 antibodies, including 18 baseline pretreatment and 32 resistant tumor samples (Figure 1A). The primary endpoints include objective response rate (ORR) and progression-free survival (PFS), and statistics was analyzed using Fisher’s exact test and Kaplan-Meier analysis. About 31% of patients (N=14) responded to anti-PD-L1/PD-1 antibodies in Chinese advanced NSCLC (Figure 1A, 1B), and 40% (N=18) had stable disease. TNB was a stronger prognosis predictor of ICIs than TMB in pretreatment samples (N=18, Figure 1C, 1D). Patients with high TNB had a significantly better objective response (P=0.007) and longer PFS (P=0.003) (Figure 1C). HLA supertype B27 (HR=0.350, P=0.021) and HLA B*15:02 (HR=0.330, P=0.018) were individually associated with a poor prognosis in total 45 patients received ICIs (Figure 1E, 1F). These results indicate that low TMB or presence of HLA B27 and HLA B*15:02 were associated with primary resistance to ICIs. For acquired resistance to ICIs, known mutations were found when disease progressed on ICIs (3/25), including B2M p.W80*, B2M p.L15Ffs*41 and JAK1 p.L235I (Figure 1G). One patient had PFS of 11.2 months with acquired B2M mutation p.L15Ffs*41 detected only in post-progression sample (Figure 1H). In Chinese advanced NSCLC, low TNB and presence of HLA B*15:02 and supertype B27 are biomarkers of primary resistance to ICIs, implying TNB and HLA genotypes can help screen patients with greater clinical benefit from immunotherapy in addition to PD-L1. Known mutations of acquired resistance to ICIs are found in few samples, other molecular mechanisms need to be explored to understand acquired resistance to ICIs.
What problem does this paper attempt to address?