posted on 2024-09-16, 11:18authored byBiagio Ricciuti, Arielle Elkrief, Joao Alessi, Xinan Wang, Yvonne Li, Hersh Gupta, Daniel M. Muldoon, Arrien A. Bertram, Federica Pecci, Giuseppe Lamberti, Alessandro Di Federico, Adriana Barrichello, Victor R. Vaz, Malini Gandhi, Elinton Lee, Geoffrey I. Shapiro, Hyesun Park, Mizuki Nishino, James Lindsay, Kristen D. Felt, Bijaya Sharma, Andrew D. Cherniack, Scott Rodig, Daniel R. Gomez, Narek Shaverdian, Mehrdad Rakaee, Chaitanya Bandlamudi, Marc Ladanyi, Pasi A. Janne, Adam J. Schoenfeld, Lynette M. Sholl, Mark M. Awad, Michael L. Cheng
Supplementary Figure 2. Proportion of samples with PD-L1 TPS ≥1%, ≥50%, ≥90% according to ATM mutation status (ATMMUT versus ATMWT).
History
ARTICLE ABSTRACT
Introduction: ATM is the most commonly mutated DNA damage and repair gene in non-small cell lung cancer (NSCLC), however limited characterization has been pursued. Methods: Clinicopathologic, genomic, and treatment data were collected for 5172 patients with NSCLC tumors which underwent genomic profiling. ATM immunohistochemistry (IHC) was performed on 182 NSCLCs with ATM mutations. Multiplexed immunofluorescence was performed on a subset of 535 samples to examine tumor-infiltrating immune cell subsets. Results: A total of 562 deleterious ATM mutations were identified in 9.7% of NSCLC samples. ATMMUT NSCLC was significantly associated with female sex (P=0.02), ever smoking status (P<0.001), non-squamous histology (P=0.004) and higher tumor mutational burden (DFCI: P<0.0001; MSK: P<0.0001) compared to ATMWT cases. Among 3687 NSCLCs with comprehensive genomic profiling, co-occurring KRAS, STK11, and ARID2 oncogenic mutations were significantly enriched among ATMMUT NSCLCs (Q<0.05), while TP53 and EGFR mutations were enriched in ATMWT NSCLCs. Among 182 ATMMUT samples with ATM IHC, tumors with nonsense, insertions/deletions, or splice site mutations were significantly more likely to display ATM loss by IHC (71.4% vs 28.6%, P<0.0001) compared to tumors with only predicted pathogenic missense mutations. Clinical outcomes to PD-(L)1 monotherapy (N=1522) and chemo-immunotherapy (N=951) were similar between ATMMUT and ATMWT NSCLCs. Patients with concurrent ATM/TP53 mutations had significantly improved response rate and progression-free survival with PD-(L)1 monotherapy. Conclusion: Deleterious ATM mutations defined a subset of NSCLC with unique clinicopathologic, genomic, and immunophenotypic features. Our data may serve as resource to guide interpretation of specific ATM mutations in NSCLC.