HPV integration as a prognostic biomarker for metastatic anal cancer: A next-generation sequencing ctDNA-based approach.

Abstract

3525 Background: Anal cancer is an HPV-associated malignancy with increasing prevalence in the United States. Immune checkpoint blockade demonstrates limited efficacy in patients with metastatic squamous cell cancer of the anal canal (mSCCA), and no prognostic or predictive biomarkers have demonstrated clinical utility in this population. Available circulating tumor DNA (ctDNA) assays have utilized ddPCR methodologies to report only the most common HPV types (HPV-16 and HPV-18) and HPV copy numbers. To extend the utility of an HPV ctDNA assay, we created a novel next-generation sequencing HPV ctDNA assay to analyze HPV integration as a prognostic biomarker for patients with mSCCA. Methods: Using an IRB-approved protocol, ctDNA isolated from the plasma of patients with mSCCA was sequenced using a 1.4Mb hybrid-capture target-enrichment panel covering the whole genome sequences of all 193 HPV types. HPV type, HPV copy number, and HPV integration status/locus were determined using a bioinformatic pipeline developed in-house. A Fisher's exact test was used to compare radiographic and ctDNA changes in response to systemic treatment. Unpaired t-tests were used to compare demographics between patients with HPV integration (I) or no integration (NI) status in the ctDNA. Median survival was estimated (Kaplan Meier) and compared according to integration status using a log-rank test. Results: 68 plasma samples from 27 patients with mSCCA were analyzed. While HPV-16 was detected in ctDNA of all 27 patients, five patients had additional oncogenic HPV types [HPV-18 (2), HPV-45 (2), HPV-73 (1)]. Radiographic changes in metastatic disease volume were concordant with HPV copy number changes in the ctDNA (odds ratio 12.2, 95% confidence interval (CI) 2.0-75; p = .007). HPV integration events were detected in the ctDNA for 16 (59%) patients (median 2, range 1-7 unique events per patient). For the 23 patients with mSCCA who received anti-PD1 immunotherapy (15 I, 8 NI), there were no differences in age (60 vs 59 years, p = .88) or prior lines of treatment (3.0 vs 3.5, p = .72). There was a trend towards improved median progression-free survival for NI ctDNA status (5.8 vs 3.2 months, hazard ratio (HR) 1.8, 95% CI .71-4.5). Relative to NI status, I status was associated with worse overall survival rates (OS) (22.5 vs 44.9 months, HR 5.0, 95% CI 1.5-17, p = .005). Conclusions: Detection of HPV integration status for all HPV types is feasible using a ctDNA-based approach for patients with SCCA. HPV integration was prognostic for worse OS in patients with mSCCA. Identification of HPV integration as a novel prognostic biomarker for survival outcomes in mSCCA warrants further evaluation in larger clinical trials.