Resistance to Checkpoint Blockade Therapy Through Inactivation of Antigen Presentation

Although immune checkpoint blockade (CPB) leads to prolonged responses in 15-40% of patients with metastatic melanoma, treatment refractory disease and progression after initial response remain major causes of mortality. While predictors of response have been reported, the common mechanisms of both primary and acquired resistance are poorly understood. To identify mechanisms of resistance and examine the evolving landscape in response to CPB, we performed whole exome sequencing (WES), immunohistochemistry (IHC), and RNA-sequencing (RNAseq) of longitudinal tumor biopsies from 17 metastatic melanoma patients treated with various CPB therapies. We found no significant changes in both mutational and neoantigen loads over time between responders and nonresponders. However, we identified abnormalities in one gene, beta-2-microglobulin (B2M), an essential component of MHC Class I antigen presentation, that were present in samples during disease progression but not regression. In total, we identified B2M aberrations in 29.4% of patients, including multiple early frameshift mutations, loss of heterozygosity (LOH) overlapping B2M, and absence of tumor-specific B2M protein expression. Additional defects in the antigen presentation and IFNγ pathways were identified but were not restricted to progressing lesions in our cohort. In two independent cohorts of 105 and 38 melanoma patients treated with ipilimumab (anti-CTLA4) and pembrolizumab (anti-PD1) respectively, we found that B2M LOH was enriched 3-fold in nonresponders (~30%) vs. responders (~10%) and associated with poorer overall survival (log-rank p=0.01, p=0.006). Loss of both copies of B2M was found only in nonresponders. We also found evidence for association of LOH overlapping IFNGR1 with poorer overall survival exclusively in the anti-PD1 cohort. Thus, B2M loss is likely a common mechanism of primary and acquired resistance to therapies targeting CTLA4 or PD-1.]]>
Metastatic melanoma patients (stage III-IV). Treatment with one of the following checkpoint blockade inhibitors: anti-CTLA4, anti-PD1, anti-PDL1. All patients have at least one time point in which a biopsy was collected. The exclusion criterion for this study is: Samples with low tumor purity were excluded. ]]>
2012-2016: Patients with metastatic melanoma provided written informed consent for the collection of tissue and blood samples for research and genomic profiling, as approved by the Dana-Farber/Harvard Cancer Center Institutional Review Board (DF/HCC Protocol 11-181). Matched tumor and normal blood samples were obtained from 17 patients at baseline and after CPB treatment between 2012-2016. 2015-2016: DNA from matched samples was extracted followed by library construction, QC and WES sequencing. 2016: Data was analyzed and results were associated to the clinical outcome of patients involved in this study.2017-2019: Additional longitudinal data including WES and RNAseq from additional matched samples from a patient in the study were generated, QC’d, and analyzed in relation to clinical outcome. ]]>