The genomic landscape and prognostic significance of 11.q.13 amplification in breast, head & neck, and lung cancer.

Abstract

e18037 Background: The 11.q.13 (11q) chromosomal band is amplified in many cancers and is associated with poor outcomes in Head & Neck (HN) cancer. The 11q amplicon contains several putative oncogenes, but it’s unclear which amplified genes drive clinical outcomes. We compared the genomic landscape and prognostic significance of 11q amplification in Breast (BC), HN, and non-small cell lung cancer (NSCLC) tumors and identify genes whose expression may be driving clinical outcomes. Methods: DNA (592 genes or WES)/RNA (WTS) sequencing was performed for BC (N = 6,967), HN (N = 2,337), and NSCLC (N = 13,433) tumors submitted to Caris Life Sciences (Phoenix, AZ). 11q amplification (CNA) was defined as tumors with concurrent copy number amplification with copies of FGF3, FGF4, FGF19, and CCND1. Non-amplified tumors (copy number normal, CNN) were defined as a < 4 copies for all four genes. Her2/Neu (+: ≥3+ and > 10%) and HR (ER or PR+: ≥1+ and ≥1%) expression was tested by IHC. tests were applied as appropriate, with P-values adjusted for multiple comparisons. Real-world overall survival (OS) data was obtained from insurance claims, and Kaplan-Meier estimates were calculated for a molecularly defined subset of tumors. 11q genes that were two-fold over expressed in CNA vs CNN were further investigated. The hazard ratio (HR) was calculated for subgroups stratified by median expression of 11q genes. The genes with the five largest HRs (worse OS for CNA, p < .05) were identified as potential driver genes. Results: 11q CNA was observed in 10% of BC, 12% of HN, and 3% of NSCLC. 11q CNA rates were enriched in HR+ BC (HR+/HER2-: 17%, HR+/HER2+: 19%) as compared to HR- tumors (HR-/HER2+: 6%, TNBC: 3%, p < .05). TP53 mutations were less frequent in CNA BC (32 vs 53% CNN) but more frequent in CNA HN (91 vs 52%) and NSCLC (89 vs 66%, p < .05 all). In NSCLC, CNA tumors had a lower rate of EGFR (4.3 vs 12%), KRAS (6.7 vs 28%) and STK11 (5.7 vs 13%) mutations ( p < .05). Over 34 gene were amplified as compared to CNN ( > 3% increase, p < .05), including FGFR3 (11 vs 0.2%) and PDCD1 (10 vs 0). MEN1 (5 vs 0%) was the only 11q gene amplified. CNA was associated with worse OS in HR-/HER2+ (HR 6.3 [2.6-15.5], p = .001), while no difference in OS was observed in BC (HR 1.0 [.9-1.1], p = .62). Compared to CNN, CNA had worse OS in HN (HR 1.5 [1.3-1.7], p < .001) and NSCLC (HR 1.3 [1.2-1.4], p < .001). Stratification by 11q gene expression identified DHCR7, GAL, ANO1, MRPL21, and FADD as the strongest predictors of OS in HN tumors, while GAL, MYEOV, MRPL21, OVOL1, and DHCR7 were strongest in NSCLC (HR: 1.3-1.2). Conclusions: 11q CNA shows differential association with key driver mutations across cancer types that provides new insight to patient subpopulations harboring these alterations. Co-amplification of non-11q genes in NSCLC suggests a unique mechanism of genome instability. The role of DHCR7 and GAL as predictors of clinical outcomes in HN and NSCLC CNA tumors warrants further investigation.