Positive homologous recombination signature (HRDsig+) and the genomic landscape of intrahepatic cholangiocarcinoma (iCCA).

627 Background: Defective DNA repair has not been a major focus in iCCA clinical research. In this comprehensive genomic profiling (CGP) study, we queried whether the presence of a scar-based HRDsig biomarker could identify a subset of patients with unique genomic alterations (GA) and potential for responsiveness to PARP inhibitor-based treatment regimens. Methods: 6,271 cases of clinically advanced iCCA underwent hybrid capture based CGP to evaluate all classes of GA. Microsatellite instability (MSI) status, tumor mutation burden (TMB) levels, genomic ancestry and genomic trinucleotide signatures were determined from the sequencing data. HRDsig status was calculated using a broad set of genome-wide copy number features (PMID 37769224). PD-L1 was determined by IHC using the Dako 22C3 tumor proportional score (TPS). Results were compared using the Fisher exact system with the Benjamini-Hochberg procedure. Results: 288 (4.6%) of the sequenced iCCA featured an HRDsig+ status. Gender distribution (46-49% male) and median age (67 years) were similar in both HRDsig+ and HRDsig- iCCA. More pathogenic GA per tumor were found in HRDsig+ vs HRDsig- iCCA (5 vs 4; P<0.0001). Genomic ancestry distribution was similar; European ancestry ranged from 73% to 76% in the two cohorts. The APOBEC signature was uncommon but slightly more frequent in the HRDSig+ cases (1.7% vs 0.4%; P=.034). MSI High status was slightly more frequent in the HRDsig- group (1.8% vs 0.0%; P=0.034). Median TMB was higher in the HRDsig+ cases (3.6 vs 1.2 mut/Mb; P<0.0001) as was the frequency of TMB > 10 mut/Mb (9.1% vs 3.5%; P<0.0001). PD-L1 expression was similar with low level (1-49% TPS) ranging from 19% to 23% in the 2 groups. As anticipated, GA in genes associated with HRD including BRCA1 (6.6% vs 0.8%; P<0.0001), BRCA2 (25.0% vs 1.3%; P<0.0001), PALB2 (8.7% vs 0.3%; P<0.0001), and ATM (6.3% vs 3.4%; P=0.033) were all more frequent in the HRDsig+ iCCA. In contrast, GA in genes associated with iCCA targeted therapies were less frequent in the HRDsig+ cases including FGFR2 (8.0% vs 12.5%; P=0.033), IDH1 (3.5% vs 14.0%; P<0.0001) and ERBB2 (3.5% vs 5.7%; NS). In the HRDsig- group, 83.3%/74.1% of BRCA1 / BRCA2 mutated iCCA were mono-allelic likely non-driver GA respectively. MTOR pathway activating GA were more frequent in the HRDsig+ cases including PTEN (7.3% vs 3.3%; P=0.003) and NF1 (8.0% vs 2.8%; P<0.0001). MTAP deletion, an emerging iCCA target, was more frequent in the HRDsig+ iCCA (23.3% vs 16.9%; P=0.024). Conclusions: Nearly 5% of advanced iCCA feature HRDsig positive status which is associated with both significant differences in the frequencies of therapy associated genomic targets and the potential for introducing PARP inhibitors for these patients. HRDsig- iCCAN = 5983 HRDsig+ iCCAN = 288 P-value ATM 3.4% 6.3% 0.033 BRCA1 0.8% (83.3% mono-allelic) 6.6% <.0001 BRCA2 1.3% (74.1% mono-allelic) 25.0% <.0001 PALB2 0.3% 8.7% <.0001