Low CD86 expression is a predictive biomarker for clinical response to the therapeutic human papillomavirus vaccine IGMKK16E7: results of a post hoc analysis.

Abstract

Background: Although therapeutic human papillomavirus vaccines could offer a noninvasive treatment for patients with cervical intraepithelial neoplasia, none has been clinically implemented. Oral administration of the therapeutic human papillomavirus vaccine IGMKK16E7 results in the histological regression of human papillomavirus 16-positive cervical intraepithelial neoplasia 2/3 to normal (complete response). We investigated biomarkers that could predict complete response after oral administration of IGMKK16E7.

Methods: Forty-two patients administered high-dose oral IGMKK16E7 in a phase I/II trial were included. Cervix-exfoliated cells were collected before vaccine administration. Gene expression of CD4, CD8, FOXP3, programmed cell death 1 protein, CTLA4, CD103, CD28, CD80, CD86, and programmed cell death 1 ligand 1 in the cells was measured by quantitative reverse transcriptase-polymerase chain reaction. Receiver operating characteristic curve analysis and Mann-Whitney tests were used to explore potential biomarkers. Pearson correlation coefficient analysis was used to correlate gene expression profiles with clinical outcome.

Results: The only predictive biomarker of vaccine response for which receiver operating characteristic curve analysis showed significant diagnostic performance with histological complete response was CD86 (area under the curve = 0.71, 95% confidence interval = 0.53 to 0.88, P = .020). Patients with complete response had significantly lower CD86 expression (CD86-low) than patients with no complete response (P = .035). The complete response rates for CD86-low and CD86-high patients were 50% and 19%, respectively, and CD86-low patients had a significantly higher complete response rate (P = .047). Compared with all patients, the CD86-low group had a 1.5-fold increase in the complete response rate. Gene expression of CD86 and CTLA4 showed the strongest positive correlation with clinical outcomes in the incomplete response group (P < .001).

Conclusion: Low expression of CD86 in exfoliated cervical cells can be used as a pretreatment biomarker to predict histological complete response after IGMKK16E7 administration.