Introduction/Background Standard treatment of high-grade cervical intraepithelial neoplasia (CIN) is conization. Merely one conization out of four is an overtreatment due to regression of lesion from biopsy to cone. Beside increasing unnecessarily the costs, CIN overtreatment might hamper the feasibility of follow-up and negatively affect reproductive outcomes. The aim of this study is to develop a new approach for high-grade CIN treatment.

Methodology Consecutive women with a diagnosis of high-grade CIN undergoing laser CO2 conization were recruited at the Outpatient Service of Central Tuscany (Florence, Italy) from September 2015 to October 2018. Before conization, cervical samples were collected for each patient and viral load of HR-HPV was assessed with Hybrid Capture 2 (HC2), which considered as positive only samples with viral load above a defined threshold. Histology reports of both biopsy and cone, as well as clinical data, were collected for each patient. Statistical analysis was performed with IBM SPSS statistics 23.0 software, using contingency tables, Pearson’s chi-square test and nonparametric tests.

Results 295 patients were enrolled. Cone histology showed a lesion regression (negative for high-grade CIN) in 40,5% of CIN II at biopsy (62/153) and in 26,9% of CIN III (25/93). Viral load in cervical samples at conization was statistically associated with CIN grade at cone histology (p<0,001): 75,7% of negative samples resulted in CIN I at cone histology, whereas 72,8% of positive ones resulted in high-grade CIN or worse at cone histology. Furthermore, all the lesions that progressed from biopsy to cone were positive at HC2 and presented higher viral load compared to those that regressed (p<0.001).

Conclusion HR-HPV testing with viral load assessment at the time of scheduled conization might be used to stratify patients referred to the procedure, identifying those who are eligible to repeat biopsy versus those who have indication to proceed with conization.