Luciferase-Based Detection of Antibodies for the Diagnosis of HPV-Associated Head and Neck Squamous Cell Carcinoma

Burbelo, Peter D.; Chaturvedi, Adrija; Notkins, Abner L.; Gunti, Sreenivasulu

en, E7, also did not detect seropositives in the HV control or CC patients but detected seven (35%) seropositive HNSCC patients which overlapped the E2 seropositive subjects ( Figure 2B) . Testing of the E6 HPV antigen as a Renilla luciferase fusion protein revealed the same pattern with HV and CC showing seronegativity yet detected nine (45%) of the HNSCC as seropositive (data not shown). To further confirm this result, a different reporter, NanoLuc was employed to detect the HPV-16 E6 antibodies. As shown in Figure 2C , the NanoLuc-E6 test also revealed that nine (45%) HNSCC subjects that were seropositive. Testing of twenty SLE patients as another set of disease controls detected no E6 or E7 seropositivity further supporting the observation that the HPV-16 antibodies responses were associated with the HNSCC patients (data not shown). Inspection of the HPV antibody profile in the HNSCC subjects revealed that nine subjects were seropositive for E6, eight overlapping subjects were E2 seropositive and 7 subjects were seropositive for E7 protein. Lastly, the presence of nine E6 seropositive samples in the HNSCC group (45%) was statistically different than the complete absence of E6 seropositivity in either the HV group or subjects with colon cancer (Fischer's Exact test, p = 0.002). Based on the substantial difference in antibody levels seen by LIPS between the HPV-negative and HPV-positive HNSCC samples, we sought to determine whether the rapid LIPSTICKS format, employing the NanoLuc-E6 fusion protein, could be used for detection of these antibodies in the HNSCC group ( Figure 2D ). The one-minute LIPSTICKS assays generated antibody signals that differentiated the HV and HPV-negative HNSCC patients from the HPV-positive HNSCC samples, in which the mean level of anti-E6 antibodies in the HNSCC samples was significantly higher (p = 0.03) than the value seen in the twenty HV controls ( Figure 2D ). Using a cutoff value based on the mean plus three standard deviations of the HV controls revealed a diagnostic performance of 89% (8/9) sensitivity and 100% specificity for detecting the E6 antibody-positive HNSCC patients in the rapid format. Moreover, the E6 antibody levels detected by the LIPSTICKS format were significantly different (p < 0.0001) between the HPV-negative HNSCC and HPV-positive HNSCC cases (data not shown). These findings suggest that the E6 LIPSTICK test has the potential to diagnose HPV-16associated HNSCC.