HPV 18 Subtype in Cervical Cancer Correlates with Pelvic Nodal Involvement at Diagnosis in Patients Undergoing Primary Chemoradiotherapy

Abstract

HPV subtypes correlate with prognostic features and outcomes in cervix cancer patients treated with primary surgery or radiotherapy. We report the incidence of HPV subtypes and outcomes in cervical cancer patients treated with definitive chemoradiotherapy. A single institutional retrospective review was performed on 51 patients from 2/2010-7/2017 treated with definitive chemoradiation, including lymph node (LN) boost when indicated. PCR was performed on patient biopsy samples to determine HPV subtyping. Data on age at diagnosis, race, prior cancer history, smoking status, FIGO stage, tumor grade, distant metastases at diagnosis, pelvic and para-aortic node involvement at diagnosis, tumor size, time to first recurrence, overall survival and follow-up time were collected. Univariate and multivariable analyses were performed based on HPV subtype. Kaplan-Meier method and the log-rank tests were used to compare overall survival and time to first recurrence. HPV was detected in 84% of biopsy samples, with HPV 16 in 51% (26), HPV 18 in 14% (7), HPV 33 in 8% (4), HPV 45 in 10% (5), and HPV 52 in 2% (1) of tumors. HPV grouping was defined as group 1(HPV 18), group 2 (HPV 16), group 3 (HPV 33,45, and 52) and group 4 (no HPV). Based on results from univariate analysis, prior history of cancer, positive pelvic lymph nodes (pLNs) at diagnosis and smoking status were analyzed on multivariate analysis. When comparing group 1 to group 3, there was a statistically significant difference in pLNs found at the time of diagnosis with positive pLNs being found in 86% of group 1 patients and only 40% of group 2 patients (p=0.045). There was also a trend towards a higher rate of pLN incidence in group 1 when compared to group 4, with the latter having positive pLNs 62% of the time. Additionally, there was a trend for group 1 patients to have had a higher rate of other cancers prior to diagnosis with 14% of group 1 patients having a prior non-skin cancer diagnosis compared to 11% of group 2 patients (p=0.156), 0% of group 3 patients (p=0.081), and 0% of group 4 patients (p=0.148). Overall survival and time to first recurrence between the various groups were not statistically significant. Our study demonstrates that HPV subtype correlates with having positive pLNs at the time of diagnosis. HPV 18 subtype had the highest rate of pLN involvement. This may support use of appropriate imaging studies for staging and treatment intensification strategies when treating patients with an HPV 18 subtype.