[The value of p16(INK4a) cytology for early diagnosis of cervical cancer].

To investigate the use of p16(INK4a) immuno-stained cytology as the primary screening for cervical cancer prevention. From March to August 2018, 902 women from Shenzhen and surrounding area were recruited for cervical cancer screening with ThinPrep Cytologic Test (TCT), cobas4800 HPV test, and p16(INK4a) co-test. Colpo/biopsies were performed using the point of interest biopsy protocol of directed and random cervical biopsies plus endocervical curettage for all women, any of whose tests was positive. Two senior cytopathologists interpreted TCT and p16(INK4a) test. The performance of p16(INK4a) for early detection of CIN2+ and inter-observer reproducibility of the interpretation of p16(INK4a) were evaluated. The positive rates of HPV test, p16(INK4a) co-test and TCT diagnosed as LSIL/AGC or higher grade were 8.1% (73/902), 6.8% (61/902) and 4.7% (42/902), respectively. Colposcopy referring rate was 79.6% (109/137), among which 10 cases were diagnosed as CIN2+ (5 cases of CIN2 and 5 cases of CIN3). The sensitivity and specificity for CIN2+ of p16(INK4a) test, TCT (LSIL/AGC or higher grade) and HPV test were 90.0%, 80.0%, 100.0% and 90.9%, 91.9%, 82.5%, respectively. Compared to TCT and HPV test, there was no significant difference in sensitivity and specificity between p16(INK4a) and TCT/HPV test (0.05). The Kappa value of the 2 cytopathologists in interpreting p16(INK4a) and TCT was 0.944 and 0.425, respectively (0.05). p16(INK4a) for cervical cancer screening is equally sensitive to HPV test and specific to TCT while subjective difference of cytopathologists' interpretation of p16(INK4a) is small. Therefore, p16(INK4a) can be used as a new cervical cancer screen method for its better diagnostic performance.