Krystle Kuhs

Ph.D., M.P.H.

University of Kentucky

Associate Professor of Epidemiology

PLCO (Learn more about this study)

2021-0002

Sep 9, 2021

Circulating tumor HPV DNA and risk of HPV-driven oropharyngeal cancer

HPV16 E6 antibody positivity has been identified as a promising early biomarker of HPV-driven oropharyngeal cancer (HPV-OPC). HPV16 E6 antibodies are present in up to 90% of HPV-OPC patients and appear more than 10 years prior to diagnosis. However, given the long lag time between seroconversion and cancer diagnosis, additional markers more proximal to cancer diagnosis are needed. Circulating tumor (ct)HPVDNA detection is a promising marker for detection of HPV-OPC. Sensitivity and specificity estimates at the time of HPV-OPC diagnosis range from 92-96% and 97-100%, respectively. However, prior studies have focused solely on using ctHPVDNA for post-treatment surveillance of recurrence. To date, the potential of ctHPVDNA for early detection of HPV-driven has not been explored. A barrier to evaluating ctHPVDNA in a pre-diagnostic setting is that the assays developed to date require a high volume of input (>1ml). Leveraging our experience in ctDNA, we have developed a highly sensitive (100%) and specific (100%) ddPCR assay for detecting ctHPVDNA in small volumes of archived specimen. The objective of this study is to evaluate whether ctHPVDNA is detectable in the blood prior to diagnosis and to compare this biomarker to HPV16 E6 seropositivity. We propose to test 81 pre-diagnostic blood samples from OPC cases and matched controls (N=162; 2:1 ratio) from prior studies of HPV16 E6 seropositivity nested within PLCO. For OPC cases, the blood sample closest to the time of diagnosis will be selected. The aims of this study are to: (1) To determine the association between pre-diagnostic detection of ctHPVDNA and development of OPC and (2) To evaluate the ability of ctHPVDNA to discriminate OPC cases and controls compared to HPV16 E6 seropositivity. We hypothesize that ctHPVDNA will be detectable prior to cancer diagnosis, and that ctHPVDNA will be more sensitive than HPV16 E6 seropositivity at the timepoint closest to diagnosis. This will be the first study to evaluate ctHPVDNA in a prediagnostic setting.

(1) To determine the association between pre-diagnostic detection of ctHPVDNA and development of OPC. The conditional logistic regression model will be used to evaluate the association between pre-diagnostic detection of ctHPVDNA and OPC development. We hypothesize that ctHPVDNA is detectable in the blood prior to OPC diagnosis, i.e. pre-diagnostic detection of ctHPVDNA is associated with development of OPC.

(2) To evaluate the ability of ctHPVDNA to discriminate OPC cases and controls compared to HPV16 E6 seropositivity. We will calculate the sensitivities of ctHPVDNA and HPV16 E6 for discriminating OPC cases and controls. The McNemar test will be used to test the difference between the two sensitivities. We hypothesize that ctHPVDNA will have a higher sensitivity than HPV16 E6 seropositivity. We will also calculate the specificity of ctHPVDNA and hypothesize that ctHPVDNA will have a very high specificity, i.e. specificity>96%.