Kalpit Shah, Dieu An H Nguyen, Max Gold, Eva Gracia Villacampa, Leire Alonso Galicia, Hejin Huang, Xian Marissa Chen, Jiaxin Sun, Yu Guo, Shreya Gaddam, Stephanie Hilz, Zhiqian Zhai, Lihua Zou, Anupriya Tripathi, Hartmut Koeppen, Jason Cheng, Matthew Wongchenko, Steven Gendreau, Marcos Abreu Machado, Emily Chan, Yang Xiao, Hsiao-Yen Ma, Ning Ding, Cloris Xue, Lieke Kusters, Mark Merchant, Paolo A Ascierto, Caroline Robert, Grant A McArthur, Tommasso Biancalani, Joakim Lundeberg, David Richmond, Yibing Yan
Using the largest melanoma clinical-molecular data set, we have defined a transcriptional classification framework that defines biologically distinct patient subgroups associated with response or resistance to ICB and its combination with MAPK-targeted therapies. These findings support biomarker-driven approaches to identify patients who may benefit from therapeutic escalation by adding MAPK inhibition to ICB rather than ICB alone, as well as a subgroup with persistent unmet need.
PURPOSE: In BRAF-mutant melanoma, clinical trials evaluating combination and sequencing strategies of MAPK-targeted therapy with immune checkpoint blockade (ICB) have yielded mixed results. We sought to define tumor biology associated with resistance to ICB and response to ICB and its combination with MAPK-targeted therapy, and determine biomarkers to guide treatment strategies.
PATIENTS AND METHODS: We integrated clinical with multi-omics data from IMspire150, BRIM, AVAST-M, and publicly available immunotherapy-treated melanoma cohorts to identify transcriptional states associated with response or resistance to ICB and BRAF/MEK inhibition combined with ICB. Functional assays and spatial transcriptomic analyses were used to evaluate and validate the underlying mechanistic hypothesis associated with response or resistance to BRAF/MEK inhibition combined with ICB.
RESULTS: Two major ICB-resistant states were identified: differentiated and undifferentiated tumors with distinct clinical and microenvironmental features. Differentiated tumors were associated with improved outcomes with BRAF/MEK inhibition combined with ICB, despite limited benefit from ICB alone. In contrast, undifferentiated tumors were less responsive to both ICB and MAPK-targeted combinations and were enriched for angiogenic and stromal programs. These tumors were associated with improved outcomes following anti-vascular endothelial growth factor therapy in retrospective analyses and showed sensitivity to CDK7 inhibition in preclinical models.
CONCLUSION: Using the largest melanoma clinical-molecular data set, we have defined a transcriptional classification framework that defines biologically distinct patient subgroups associated with response or resistance to ICB and its combination with MAPK-targeted therapies. These findings support biomarker-driven approaches to identify patients who may benefit from therapeutic escalation by adding MAPK inhibition to ICB rather than ICB alone, as well as a subgroup with persistent unmet need.