Unique mutational landscape and therapeutic implications in non-small cell lung cancer with comorbid fibrotic interstitial lung disease.

Abstract

8061 Background: Patients with established interstitial lung disease (ILD) have both increased risk of developing non-small cell lung cancer (NSCLC) and higher mortality than patients without a history of ILD. We hypothesized the mutational landscape of NSCLC arising in the setting of ILD would be different than sporadic or smoking-associated cancers, potentially leading to different treatment options given the question of safety of immune checkpoint inhibitors (ICIs) in this population. Methods: We retrospectively identified 330 patients with NSCLC, of which 77 patients had pre-existing diagnoses of ILD with probable or definite usual-interstitial pattern made by radiographic or biopsy findings for comparison to randomly selected patients with NSCLC without clinically apparent ILD. Clinical characteristics, histologic information, mutational data, and treatment regimens were collected using the Stanford Research Repository and compared between patients with NSCLC and established ILD (LC-ILD) and NSCLC without ILD (LC). Statistical comparisons between groups were done with Mann-Whitney testing with significance set at p<0.05. Results: Baseline characteristics including age, sex, race, ethnicity, smoking status, NSCLC type, and NSCLC stage at diagnosis did not differ between the LC-IPF and LC groups. There was significantly lower prevalence of EGFR mutations in the LC-IPF group (33.0% vs 3.0%, p<0.0001) and significantly higher prevalence of KRAS G12D mutations (19.0% vs 41.7%, p=0.05). Baseline tumor proportion score (TPS) between LC and LC-ILD groups was not significantly different (28.2% vs 16.3%, p=0.063) however there were significantly fewer patients with high PD-L1 expression (TPS ≥ 50%) in the LC-ILD group (32.0% vs 16.7%, p=0.036). Occurrence of clinically significant treatment-related pneumonitis occurred in 15 patients in the LC-ILD group with etiologies identified as radiation (n=8, N=20, 40.0%), surgery (n=1, N=40, 2.5%), osimertinib (n=1, N=1, 100%), pembrolizumab (n=2, N=5, 40.0%), docetaxel (n=2, N=4, 50.0%, and pemetrexed (n=1, N=19, 5.7%). Conclusions: These retrospective data highlight the differences in driver mutations in NSCLC in patients with preceding fibrotic lung disease, suggesting potentially divergent biologic underpinnings for tumorigenesis. These results, particularly the under-representation of EGFR -mutations, over-representation of KRAS G12D mutations, and scarcity of therapeutically actionable mutations in the LC-ILD population, lead to limited therapeutic options. Surgically-associated pneumonitis was rare, but radiation, docetaxel, and pembrolizumab appeared high risk for pneumonitis in treated patients, suggesting need careful consideration of risks when treating this unique population.