Cancer Medicine (Jan 2023)

Genomic landscape of non‐small‐cell lung cancer with methylthioadenosine phosphorylase (MTAP) deficiency

  • Prashanth Ashok Kumar,
  • Stephen L. Graziano,
  • Natalie Danziger,
  • Dean Pavlick,
  • Eric A. Severson,
  • Shakti H. Ramkissoon,
  • Richard S. P. Huang,
  • Brennan Decker,
  • Jeffrey S. Ross

DOI
https://doi.org/10.1002/cam4.4971
Journal volume & issue
Vol. 12, no. 2
pp. 1157 – 1166

Abstract

Read online

Abstract Introduction New treatment strategies for advanced non‐small‐cell lung carcinoma (NSCLC) include synthetic lethality targets focused on protein arginine methyl transferases such as PRMT5 that exploit the impact of genomic loss of methylthioadenosine phosphorylase (MTAP). Methods Twenty nine thousand three hundred seventy nine advanced NSCLC cases underwent hybrid‐capture based comprehensive genomic profiling between June 1, 2018 and May 31, 2020. PD‐L1 expression was determined by immunohistochemistry (Dako 22C3 PharmDx assay). Results 13.4% (3928/29,379) NSCLC cases exhibited MTAP loss distributed in adenocarcinoma (59%), squamous cell carcinoma (22%), NSCLC not otherwise specified (16%), and 1% each for large‐cell neuroendocrine, sarcomatoid, and adenosquamous carcinoma. Statistically significant differences in mitogenic driver alterations included more KRAS G12C mutations in MTAP‐intact versus MTAP‐lost (12% vs. 10%, p = 0.0003) and fewer EGFR short variant mutations in MTAP‐intact versus MTAP‐lost NSCLC (10% vs. 13%, p < 0.0001). Statistically significant differences in currently untargetable genomic alterations included higher frequencies of TP53 (70% vs. 63%, p < 0.0001) and RB1 inactivation (10% vs. 2%, p < 0.0001) in MTAP‐intact compared to MTAP‐lost NSCLC. SMARCA4 inactivation (7% vs. 10%, p < 0.0001) was less frequent in MTAP‐intact versus MTAP‐lost NSCLC. Alterations in ERBB2, MET, ALK, ROS1, and NTRK1 did not significantly differ between the two groups. Predictors of immunotherapy efficacy were higher in MTAP‐intact versus MTAP‐lost NSCLC including tumor mutational burden (9.4 vs. 8.6 mut/Mb, p = 0.001) and low (30% vs. 28%, p = 0.01) and high PD‐L1 (32% vs. 30%, p = 0.01) expression. Alterations in biomarkers potentially predictive of immune checkpoint inhibitor resistance (STK11, KEAP1, and MDM2) were similar in the two groups. Conclusions MTAP loss occurs in 13% of NSCLC, supporting the development of targeted therapies to exploit PRMT5 hyper‐dependence. MTAP loss is accompanied by small differences in targeted and immunotherapy options which may impact future combination strategies.

Keywords