p16INK4A flow cytometry of exfoliated cervical cells: Its role in quantitative pathology and clinical diagnosis of squamous intraepithelial lesions

Abstract Background P16INK4A is a surrogate signature compensating for the specificity and/or sensitivity deficiencies of the human papillomavirus (HPV) DNA and Papanicolaou smear (Pap) co‐test for detecting high‐grade cervical squamous intraepithelial lesions or worse (HSIL+). However, traditional p16INK4A immunostaining is labour intensive and skill demanding, and subjective biases cannot be avoided. Herein, we created a high‐throughput, quantitative diagnostic device, p16INK4A flow cytometry (FCM) and assessed its performances in cervical cancer screening and prevention. Methods P16INK4A FCM was built upon a novel antibody clone and a series of positive and negative (p16INK4A‐knockout) standards. Since 2018, 24 100‐women (HPV‐positive/‐negative, Pap‐normal/‐abnormal) have been enrolled nationwide for two‐tier validation work. In cross‐sectional studies, age‐ and viral genotype‐dependent expression of p16INK4A was investigated, and optimal diagnostic parameter cut‐offs (using colposcopy and biopsy as a gold standard) were obtained. In cohort studies, the 2‐year prognostic values of p16INK4A were investigated with other risk factors by multivariate regression analyses in three cervicopathological conditions: HPV‐positive Pap‐normal, Pap‐abnormal biopsy‐negative and biopsy‐confirmed LSIL. Results P16INK4A FCM detected a minimal ratio of 0.01% positive cells. The p16INK4A‐positive ratio was 13.9 ± 1.8% among HPV‐negative NILM women and peaked at the ages of 40–49 years; after HPV infection, the ratio increased to 15.1 ± 1.6%, varying with the carcinogenesis of the viral genotype. Further increments were found in women with neoplastic lesions (HPV‐negative: 17.7 ± 5.0–21.4 ± 7.2%; HPV‐positive: 18.0 ± 5.2–20.0 ± 9.9%). Extremely low expression of p16INK4A was observed in women with HSILs. As the HPV‐combined double‐cut‐off‐ratio criterion was adopted, a Youden's index of 0.78 was obtained, which was significantly higher than that (0.72) of the HPV and Pap co‐test. The p16INK4A‐abnormal situation was an independent HSIL+ risk factor for 2‐year outcomes in all three cervicopathological conditions investigated (hazard ratios: 4.3–7.2). Conclusions FCM‐based p16INK4A quantification offers a better choice for conveniently and precisely monitoring the occurrence of HSIL+ and directing risk‐stratification‐based interventions.