Abstract

Introduction: Altered expression of ERRFI1 (Epidermal Growth Factor Receptor Feedback Inhibitor 1), NKX2-1 (also known as TTF1), and CLU, play a role in the response to targeted therapies and immune checkpoint inhibitors (ICI) in lung cancer. Here, we evaluate the prevalence of ERRFI1 and NKX2-1 mutations, changes in their RNA levels, and CLU RNA levels prior to therapy (PT) and upon disease progression (UP), focusing on EGFR and ALK tyrosine kinase inhibitors (TKIs).

Methods: De-identified records of 34,362 patients (pts) with a primary diagnosis of lung cancer and subjected to Tempus xT/xR NGS were identified. Gene expression levels of ERRFI1, NKX2-1, and CLU were compared between NSCLC vs. SCLC, squamous vs non-squamous histology, and sample collection time (PT vs UP). The somatic landscape was assessed for ERRFI1 and/or NKX2-1 pathogenic/likely pathogenic (P/LP) or variants of unknown significance (VUS) somatic mutations. Somatic differences were compared using Pearson’s Chi-squared test or Fisher’s exact test, as appropriate. The median log2 (TPM+1) gene expression was compared between groups using the Wilcoxon rank sum exact test.

Results: Among the total cohort, NSCLC had significantly higher expression levels of ERRFI1 (7.8 vs 6.5, p<0.001) and CLU (7.3 vs 6.3, p<0.001) compared to SCLC, while NKX2-1 levels were higher in SCLC (6.7 vs 5.5, p<0.001). ERRFI1, NKX2-1, and CLU expression were higher in non-squamous NSCLC compared to squamous NSCLC (p<0.001). CLU expression was significantly higher in UP samples with EGFR TKI (p=0.024), ALK TKI (p=0.033), and ICI treatments (p=0.048) compared to PT samples. NKX2-1 expression was significantly lower in UP samples for EGFR TKI (p<0.001), ALK TKI (p=0.002), and ICI treatments (p<0.001) compared to PT samples. ERRFI1 expression was also lower in UP samples with ALK TKI (p=0.013) but not EGFR TKI or ICI treatments. Of those samples, 1% (403/34,362) had somatic mutations in ERRFI1 only, 6% (1,980/34,362) in NKX2-1 only and 0.01% (39/34,362) in both. The top co-mutated genes in ERRFI1 and/or NKX2-1 mutated tumors included TP53 (64%), KRAS (36%), FOXA1 (25%), CDKN2A (24%), NFKBIA (24%) and EGFR (21%). EGFR co-mutated tumors from pts with a history of TKI had a significantly higher number of NKX2-1 amplifications compared to those without (p<0.001), while NKX2-1 deletions were more prevalent in the TKI naive cohort (p=0.002).

Conclusions: Lung cancers with ERRFI1 and/or NKX2-1 mutations exhibit distinct genomic profiles, with significant variations in ERRFI1, NKX2-1, and CLU expression across different subtypes. CLU expression was higher UP after EGFR TKI, ALK TKI, and ICI, while NKX2-1 and ERRFI1 expression were lower UP. Understanding the dynamic expression across disease states may inform knowledge of resistance mechanisms and putative treatment strategies. This could be particularly informative as clusterin inhibitors are undergoing evaluation in clinical trials.

VIEW THE PUBLICATION