Low Interferon Expression Coupled With 9p21 Loss in Non-Small Cell Lung Cancer (NSCLC) Patients Associated With Distinct Somatic Landscape, Altered Immune Population, and Poorer Response to Immune Checkpoint Inhibitor (ICI) Therapies

AACR 2025

Mar 25, 2025
Oncology
Presentation

Natalie Vokes, Jessica L. Symons, Brooke Rhead, Pooja A. Shah, Santiago Trevino III, Konstantinos Levantakos, Stamatina Fragkogianni, Calvin Chao, Jordi A. Rodon

Background – Deletions in 9p21 are common across multiple cancer types including NSCLC. Notably, 9p21 deletions have been implicated in an immune-excluded tumor immune phenotype and worse outcomes following ICIs. Whether this effect is due to CDKN2A loss, collateral loss of MTAP, or loss of type I interferon genes (IFNKIFNA) has not been fully adjudicated.

 

Methods – The Tempus Database (Tempus AI, Inc., Chicago, IL) was queried for de-identified records of patients with NSCLC and DNA (Tempus xT) and RNA (Tempus xR) molecular profiling. CDKN2A/MTAP deleted samples were defined as those containing biallelic loss of CDKN2A/B and MTAP. As IFNK is not included in the DNA sequencing bait set, IFNK status was defined as low in any sample with expression ≤ the first quartile of IFNK expression from samples with intact CDKN2A/B and MTAP. Samples with tumor purity < 40% or mixed loss of CDKN2A/B and MTAP were excluded. Association analysis of CDKN2A/MTAP/IFNK status with other genomic events and clinical outcomes was performed. N=16,947 patients were included in the overall analysis, of whom n=1,991 had CDKN2A/B/MTAP deletion (11.7%, _CM_del).

 

Results – Significantly more _CM_del patients had low IFNK expression compared to CDKN2A/B/MTAP intact (_CM_wt) patients (n=14,956); of _CM_del (n=1300), 65.3% had high IFNK expression (_IFNK_hi) and 34.7% had low IFNK expression (_IFNK_lo). Among _CM_del patients, IFNKhi samples were enriched in never smokers compared to IFNKlo patients (24% vs 18%, p=0.002). _CM_del was enriched in samples with EGFR and ALK driver alterations, and depleted in samples with KRAS or TP53 alterations. In both _CM_del and _CM_wt contexts, KRASEGFR, and ALK alterations associated with _IFNK_hi expression, whereas TP53NFE2L2STK11KEAP1SMARCA4, and KMT2D associated with _IFNK_lo expression. _IFNK_hi expression within _CM_del patients was associated with an increase in the estimated proportion of immune cell subpopulations, including CD8 T cells and macrophages (M1 and M2), and a decrease in neutrophils and NK cells. In metastatic/unresectable stage III patients with recorded ICI-containing 1L therapies (n=1,795), patients with _CM_del and _IFNK_lo trended toward higher rates of progressive disease (PD) (37%), whereas _CM_del/_IFNK_hi had comparable rates of PD to _CM_wtl/_IFNK_hi and _CM_wtl/_IFNK_lo (28%, 27%, 30%, respectively). _CM_del/_IFNK_lo patients treated with non-ICI/TKI therapies (mostly consisting of chemotherapy) had a non-significant trend toward increased rate of PD (63%).

 

Conclusions – _CM_del associates with driver alterations in EGFR and ALK, along with worse ICI and chemotherapy outcomes in the context of _IFNK_lo but not _IFNK_hi expression. Further work to disentangle the impact of immune infiltrate and the co-mutation landscape will help clarify the underlying associated biology.