Adoptive T-cell therapy (ACT) is a highly intensive immunotherapy regime that has yielded remarkable response rates and many durable responses in clinical trials in melanoma; however, 50–60% of the patients have no clinical benefit. Here, we searched for predictive biomarkers to ACT in melanoma. Whole exome- and transcriptome sequencing and neoantigen prediction were applied to pre-treatment samples from 27 patients recruited to a clinical phase I/II trial of ACT in stage IV melanoma. All patients had previously progressed on other immunotherapies. We report that clinical benefit is associated with significantly higher predicted neoantigen load. High mutation and predicted neoantigen load are significantly associated with improved progression-free and overall survival. Further, clinical benefit is associated with the expression of immune activation signatures including a high MHC-I antigen processing and presentation score. These results improve our understanding of mechanisms behind clinical benefit of ACT in melanoma.

Melanoma is currently divided on a genetic level according to mutational status. However, this classification does not optimally predict prognosis. In prior studies, we have defined gene expression phenotypes (high-immune, pigmentation, proliferative and normal-like), which are predictive of survival outcome as well as informative of biology. Herein, we employed a population-based metastatic melanoma cohort and external cohorts to determine the prognostic and predictive significance of the gene expression phenotypes. We performed expression profiling on 214 cutaneous melanoma tumors and found an increased risk of developing distant metastases in the pigmentation (HR, 1.9; 95% CI, 1.05-3.28; P=0.03) and proliferative (HR, 2.8; 95% CI, 1.43-5.57; P=0.003) groups as compared to the high-immune response group. Further genetic characterization of melanomas using targeted deep-sequencing revealed similar mutational patterns across these phenotypes. We also used publicly available expression profiling data from melanoma patients treated with targeted or vaccine therapy in order to determine if our signatures predicted therapeutic response. In patients receiving targeted therapy, melanomas resistant to targeted therapy were enriched in the MITF-low proliferative subtype as compared to pre-treatment biopsies (P=0.02). In summary, the melanoma gene expression phenotypes are highly predictive of survival outcome and can further help to discriminate patients responding to targeted therapy.

Cancer genome sequencing has shed light on the underlying genetic aberrations that drive tumorigenesis. However, current sequencing-based strategies, which focus on a single tumor biopsy, fail to take into account intratumoral heterogeneity. To address this challenge and elucidate the evolutionary history of melanoma, we performed whole-exome and transcriptome sequencing of 41 multiple melanoma biopsies from eight individual tumors. This approach revealed heterogeneous somatic mutations in the range of 3%-38% in individual tumors. Known mutations in melanoma drivers BRAF and NRAS were always ubiquitous events. Using RNA sequencing, we found that the majority of mutations were not expressed or were expressed at very low levels, and preferential expression of a particular mutated allele did not occur frequently. In addition, we found that the proportion of ultraviolet B (UVB) radiation-induced C>T transitions differed significantly (P < 0.001) between early and late mutation acquisition, suggesting that different mutational processes operate during the evolution of metastatic melanoma. Finally, clinical history reports revealed that patients harboring a high degree of mutational heterogeneity were associated with more aggressive disease progression. In conclusion, our multiregion tumor-sequencing approach highlights the genetic evolution and non-UVB mutational signatures associated with melanoma development and progression, and may provide a more comprehensive perspective of patient outcome.

Large cell carcinoma with or without neuroendocrine features (LCNEC and LC, respectively) constitutes 3–9% of non-small cell lung cancer but is poorly characterized at the molecular level. Herein we analyzed 41 LC and 32 LCNEC (including 15 previously reported cases) tumors using massive parallel sequencing for mutations in 26 cancer-related genes and gene fusions in ALK, RET, and ROS1. LC patients were additionally subdivided into three immunohistochemistry groups based on positive expression of TTF-1/Napsin A (adenocarcinoma-like, n = 24; 59%), CK5/P40 (squamous-like, n = 5; 12%), or no marker expression (marker-negative, n = 12; 29%). Most common alterations were TP53 (83%), KRAS (22%), MET (12%) mutations in LCs, and TP53 (88%), STK11 (16%), and PTEN (13%) mutations in LCNECs. In general, LCs showed more oncogene mutations compared to LCNECs. Immunomarker stratification of LC revealed oncogene mutations in 63% of adenocarcinoma-like cases, but only in 17% of marker-negative cases. Moreover, marker-negative LCs were associated with inferior overall survival compared with adenocarcinoma-like tumors (p = 0.007). No ALK, RET or ROS1 fusions were detected in LCs or LCNECs. Together, our molecular analyses support that LC and LCNEC tumors follow different tumorigenic paths and that LC may be stratified into molecular subgroups with potential implications for diagnosis, prognostics, and therapy decisions.