Women treated with cervical loop electrosurgical excision procedure require follow-up to detect residual or recurrent HSIL+, defined as high-grade squamous intraepithelial lesions, adenocarcinoma in situ or cervical cancer. Currently, co-testing with cytology and human papillomavirus (HPV) analysis is usually recommended. This study investigates whether HPV testing alone is comparable to co-testing in detecting HSIL + up to three years after treatment. Recurrence rates of HSIL + are also presented, with follow-up extending up to 18 years. This retrospective cohort study included all 3,540 women treated with a cervical excision in Uppsala County between 1 January 2005 and 31 December 2019. Women with cancer identified in the cone biopsy were excluded. The main outcome was HSIL + detected within three years of follow-up. Sensitivity, specificity and negative predictive value were calculated for the 1,938 women who had a co-testing result as part of their test of cure. Thus, the analysis for the main outcome could finally be performed on 1,938 out of the total number of 3,540 women. Additionally, long-term data on recurrence and time to HSIL+, along with a separate analysis of results prior to cervical cancer diagnosis, were collected for the whole cohort of 3,399 women. The sensitivity and negative predictive value for detecting HSIL + were 69% and 97% for HPV alone, and 74% and 98% for co-testing, respectively. These differences were not statistically significant. Specificity was higher for HPV alone than for co-testing. The negative predictive value of HPV testing for excluding cervical cancer (n = 5) within three years was 100%. Recurrence rate of HSIL + in the three-year follow up was 8%, and the total recurrence rate of HSIL + with a mean follow-up of nine years was 10%. Mean time to recurrence was 28 months. None of 19 cervical cancer cases identified in the long-term follow-up had a co-testing result showing negative HPV but positive cytology. HPV testing alone, as a single test, is comparable to co-testing in detecting HSIL + up to three years after treatment independently of margin status, and demonstrates a higher specificity. Cytology plays a very limited role in the test of cure analysis and could therefore be omitted.
Håstad et al. (Wed,) studied this question.