Colposcopy tends to be subjective given the weak correlations between visual changes and disease severity. It has been suggested that systematic collection of cervical biopsies (SCCB) may increase the detection of high-grade cervical intraepithelial neoplasia and/or cancer compared to the traditional directed cervical biopsies (DCB) approach. We aim to describe through stochastic simulation of cervical dysplasia the methodology and statistical strategy for evaluating the gain in disease detection consequent to the use of SCCB compared to DCB based on data from the FRIDA study.
The FRIDA study, conducted in the Tlaxcala State Ministry of Health, Mexico, was designed to evaluate at the population level new triage alternatives for colposcopy evaluation among high-risk HPV (hrHPV) positive women aged 30 to 64 years. The current analysis uses data from the first 30, 000 women enrolled. All participants were screened by hrHPV testing with a prevalence of 11%. Additional triage tests (hrHPV 16/18, cytology, OncoE6 protein, etc.) were done on hrHPV positive women. After the colposcopy examination in each quadrant, if Reid Index was ≤2 points a biopsy was not required, but necessary if evaluation was >2 points. One biopsy was collected from the most abnormal zone of the squamocolumnar junction; all biopsy collection sites were recorded. This procedure is referred as a SSCB. We performed stochastic models for counterfactual analyses, simulating a comparison between what actually happened and what would have happened in the absence of SCCB. We generated 3 different guideline-based scenarios reconstructing the main results observed in the FRIDA study, starting with the simplest scenario (hrHPV testing and cytology) and adding several levels of complexity (HPV 16/18 genotyping, cytology, DCB and SCCB based-colposcopy) while maintaining preceding scenarios. Stochastic analyses of scenarios 2 (randomization according to HPV genotyping and cytology with DCB) and 3 (randomization according to HPV genotyping and cytology with SCCB) will enable the calculation of detection rates with respect to different collection procedures of cervical biopsies. Histology results were used to reflect the detection rate and to determine the incremental gain in the diagnosis of cervical disease.
This analysis will investigate the benefits of incorporating the SCCB approach meanwhile will provide valuable information to policy makers that will contribute to the introduction of this strategy as a standard practice in colposcopy evaluation in Mexico; this must be guided by evidence obtained from the local context about the cost-effectiveness of the intervention.
1. Torres-Ibarra L y col. Triage strategies in cervical cancer detection in Mexico: methods of the FRIDA Study. Salud Pública de México / vol. 58, no. 2, febrero de 2016.