Clinical and immunologic biomarkers for histologic regression of high-grade cervical dysplasia and clearance of HPV16 and HPV18 after immunotherapy

Purpose: As previously reported, treatment of high-grade cervical dysplasia with VGX-3100 resulted in complete histopathologic regression (CR) concomitant with elimination of HPV16/18 infection in 40.0% of VGX-3100-treated patients compared with only 14.3% in placebo recipients in a randomized phase IIb study. Here, we identify clinical and immunologic characteristics that either predicted or correlated with therapeutic benefit from VGX-3100 to identify parameters that might guide clinical decisionmaking for this disease. Experimental Design: We analyzed samples taken from cervical swabs, whole blood, and tissue biopsies/resections to determine correlates and predictors of treatment success. Results: At study entry, the presence of preexisting immunosuppressive factors such as FoxP3 and PD-L1 in cervical lesions showed no association with treatment outcome. The combination of HPV typing and cervical cytology following dosing was predictive for both histologic regression and elimination of detectable virus at the efficacy assessment 22 weeks later (negative predictive value 94%). Patients treated with VGX-3100 who had lesion regression had a statistically significant >2-fold increase in CD137 ⁺ perforin ⁺ CD8 ⁺ T cells specific for the HPV genotype causing disease. Increases in cervical mucosal CD137 ⁺ and CD103 ⁺ infiltrates were observed only in treated patients. Perforin ⁺ cell infiltrates were significantly increased >2-fold in cervical tissue only in treated patients who had histologic CR. Conclusions: Quantitative measures associated with an effector immune response to VGX-3100 antigens were associated with lesion regression. Consequently, these analyses indicate that certain immunologic responses associate with successful resolution of HPV-induced premalignancy, with particular emphasis on the upregulation of perforin in the immunotherapy-induced immune response.