Clinical utility of circulating tumor tissue-modified viral HPV DNA testing in HPV-driven oropharyngeal cancer arising in women.

Abstract

e18032 Background: HPV+OPSCC is rare in womenand may indicate a different biology. The purpose of this study was to evaluate the clinical utility of circulating tumor tissue modified viral (TTMV)-HPV DNA testing in women with HPV+ OPSCC, and to compare the results to our male cohort. Methods: A total of 88 TTMV-HPV DNA tests were collected on 23 women with OPSCC, treated with curative intent. 87% pts (20/23) had pre-treatment TTMV testing. Initial HPV status was assessed by p16 marker IHC and then by TTMV-HPV DNA assay. There were 22/23 pts with p16+ by IHC. One pt with p16- IHC had a positive pre-Tx TTMV score of 185. Post treatment TTMV testing was performed at 3-month intervals. The pre-Tx TTMV median score was 1047 normalized frags/ml. (range: 7-16909). Median number of surveillances TTMV tests were 2.5 (range 1-7). Test results were correlated with imaging to assess clinical utility. Results: 23 women were included. Median age 65.9 (Range: 49-79). Median follow-up was 25.6 months (Range 1.4-52.7). There were 20 pre-treatment tests, 19/20 pts were p16+ IHC. 14 TTMV tests were true positive, with a per-test and per-patient sensitivity of 76.5% (95% CI: 56.3 – 96.6%). For p16+ males, the per-test sensitivity was 54/59, 91.5%, (95% CI: 84.4 – 98.6%) and the per-pt sensitivity is 50/55, 90.9% (95% CI: 83.8 – 98.5%). Fisher’s exact test was used to determine if there is a relationship between biological sex and pre-treatment sensitivity. The per-test p-value was 0.0575, and the per-pt p-value was 0.1119. 5 pts with p16+ IHC had negative baseline TTMV tests, one of which had TTMV-HPV16 DNA subsequently confirmed by FFPE. Clinical staging included T1-T2 (N=19; 82.6%), T3-T4 (N=4, 17.4%), N0-N1 (65.2%: N0=3, N1=12), N2-N3 (34.8%: N2=7, N3=1). There was no correlation between pre-treatment TTMV score and T stage. There was a significant difference in pre-treatment score for pts with N0 vs N+ (Mann-Whitney U test, p = 0.0392). However, the N0 group only had 3 tests, compared to 15 tests in the N1/N2/N3 group. The median score of the N0 group was 7 and the median score of the N1/N2/N3 group was 371. All TTMV tests resolved to zero after treatment completion. 9/14 (64.3%) TTMV scores resolved within 3 months after treatment completion. These pts had post treatment imaging and clinical examination without evidence of disease. We have not observed to date any rise in TTMV-HPV ctDNA in this cohort. Conclusions: This is the first single institutional report evaluating the clinical utility of TTMV-HPV DNA in a female cohort treated for HPV+ OPSCC. Utilization of this testing in a female population appears feasible in a community-based hospital setting. All pts had TTMV resolution after treatment and remained undetectable without recurrence. Despite the apparent trend toward lower sensitivity in females, the difference was not statistically significant and warrant further investigation in a larger multi-institutional study.