Conservative Management of HSIL in Patients With Future Pregnancy Aspiration

Overview

Conservative management of high-grade squamous intraepithelial lesions (HSILs) seems safe and justified in young women (\<30 years), but evidence is insufficient on whether it is also advisable for older women. This study will be conducted to analyze spontaneous HSIL regression rates in women of reproductive age and establish whether conservative HSIL management could be safely recommended to women of childbearing potential, irrespective of age. This is a single-center prospective observational study that will include consecutive women of reproductive age, referred to a tertiary hospital due to HSIL between March 2021 and December 2025, who prefer conservative management rather than immediate cervical conization. All patients will be followed-up regularly with colposcopy, cytology, human papillomavirus (HPV) testing and biopsies. In case their lesions progress or HSIL persists after 24 months of follow-up, conization will be indicated. Rates of spontaneous regression or resolution, as well as progression rates, will be assessed. Furthermore, the association between potential predictive factors and HSIL resolution will be analyzed.

In most developed countries, two main strategies are currently used for cervical cancer prevention: primary prevention with human papillomavirus (HPV) vaccination, and systematic screening with periodic cytology or HPV testing. Nevertheless, a significant number of women are still diagnosed with high-grade squamous intraepithelial lesions (HSIL), with higher prevalence rates among younger women (0.6% between ages 20 and 29; 0.2% between 40 and 49). Since the average age at first pregnancy in Catalonia is 32.2 years, many patients have not yet fulfilled their reproductive intentions when they are diagnosed with HSIL. Most HSIL cases are treated surgically with cervical conization, although previous studies have indicated that this may increase the incidence of preterm labor in future pregnancies. Therefore, many women diagnosed with HSIL will undergo treatment that may affect their future reproductive life. In patients diagnosed with HSIL, conservative management with periodic evaluations would be an efficient alternative to surgery, with the expectation of spontaneous resolution. Recent studies have shown that 40-88% of HSIL lesions can regress spontaneously within two years, especially in women under 25. Furthermore, a meta-analysis of 36 studies involving 3160 patients concluded that active surveillance of patients diagnosed with cervical intraepitelial neoplasia (CIN) grade 2, rather than immediate surgical treatment, is justified, especially in young women (\<30 years). Recent clinical guidelines include active surveillance as a valid option for management of HSIL in fertile women, with no age restrictions. Although the biological behavior of CIN2 without surgical treatment has been described in numerous studies, few included lesions classified as CIN3. Since the latest anatomical and pathological classification of the World Health Organization does not make any distinction between CIN2 and CIN3, classifying both as HSIL, treatment guidelines should be uniform for all HSILs, without discriminating between CIN2 or CIN3. Furthermore, although several studies have demonstrated the safety of conservative management for HSIL in women younger than 25-30 years, there is insufficient information in women aged 30-40 years. Therefore, further evidence is still needed on the rate of spontaneous regression of HSIL (both CIN2 and CIN3) and on the safety of active surveillance, so that conservative management may be considered in all women who wish to become pregnant in the future. The main goal of this study will be to analyze whether spontaneous HSIL (CIN2 and CIN3) regression rates in women of reproductive age are high enough to support conservative management without age restrictions. We aim to describe the rates of spontaneous HSIL regression, HSIL persistence and HSIL progression in women under conservative management. Furthermore, we aim to describe predictive factors associated with HSIL regression rates. If this study confirms the safety of conservative management of HSIL, it could be possible to reduce the number of cervical conizations in these patients. Aim This will be a prospective observational cohort study performed at a Lower Genital Tract (LGT) unit of a tertiary hospital (Hospital de la Santa Creu i Sant Pau, Barcelona, Spain) affiliated with a cytology-based cervical cancer screening program. In this center, all colposcopic examinations are performed by two senior colposcopists, accredited by the Spanish Association for Cervical Pathology and Colposcopy (AEPCC). Pathology samples are analyzed by specialist technicians from the Pathology department of the same hospital and reviewed by expert pathologists. HPV tests are performed for all patients in the same laboratory using Genómica's Clart Human Papillomavirus 4. Patients All women referred to our LGT unit with biopsy-confirmed histological diagnosis of HSIL (CIN2 or CIN3) are assessed with colposcopy and counseled by expert gynecologists. Patients considered suitable for conservative management (instead of immediate surgical treatment with cervical conization) are routinely informed of the potential risks and benefits of both cervical conization and active surveillance of HSIL. Then, a shared decision is made for either surveillance or immediate conization. To be considered suitable for conservative management, women must: be of reproductive age; have transformation zone type 1 or 2 (fully visible squamous-columnar union) at colposcopy, with a fully visible lesion and no atypical glandular cells (AGC) at cytology; and commit to attend follow-up visits at regular intervals. This study will start in March 2021 and will be concluded in December 2027. Consecutive women of reproductive age referred to our center from March 2021 to December 2025 with a diagnosis of HSIL that are considered suitable for conservative management and freely chose active surveillance will be invited to participate in this study. Patients will be followed-up for a minimum of 2 years. Eligibility criteria include positive high-risk (hr) HPV cytology, reproductive age and a wish to pursue pregnancy in the future. Women who are pregnant at diagnosis, immunosuppressed or diagnosed with carcinoma in situ of the cervix or cervical cancer will be excluded. This study was approved by the institution's ethics committee. Written informed consent will be obtained from all patients. All follow-up visits and clinical tests will be performed according to standard patient care and no tests or visits will be conditioned to participation in the study, nor will any additional treatment be given to patients enrolled. Verbal informed consent to continue conservative treatment will be reviewed at each follow-up visit. At any time during follow-up, patients can choose to have conization. Follow-up visits will be performed every 4 months, including colposcopy and cytology at each visit. Biopsies will not performed routinely, but on a case-by-case basis according to changes observed in colposcopy. The follow-up interval will be extended to 6 months if cytology and biopsies indicated low-grade squamous intraepithelial lesion (LSIL) or atypical squamous cells of undetermined significance (ASCUS) and this result is consistent with colposcopy. HPV will be determined every 8 12 months. This follow-up protocol will be continued until resolution of HSIL. Patients will be referred back to primary care when no HSIL is detected by cytology, colposcopy or biopsies and if HPV is negative at two consecutive visits. Conservative management will be discontinued and conization will be indicated if HSIL persists after 24 months of follow-up, if the patient no longer meet criteria for surveillance, or under the patient's request. All information will be retrieved from electronic patient files. Demographic data, smoking status, parity, use of contraceptives and colposcopy will be recorded. Information from pathology reports of cytology, biopsies and conization specimens will be collected, and time to HSIL resolution, regression or progression will be recorded. Patients will be classified into two groups according to the outcome of active surveillance: spontaneous HSIL regression or conization (women who had cervical conization for any reason during follow-up). Patients with spontaneous HSIL regression will be further classified into 3 subgroups: total resolution (no colposcopic lesion, normal pathology by biopsy and cytology, and negative HPV for the HPV type initially detected); partial resolution (regression of colposcopic lesion, negative cytology and biopsies, but persistence of the initial hrHPV detected); and lesion regression (HSIL no longer detected, but persistent LSIL in either cytology, histology or colposcopy). Patients who eventually have conization will be further classified according to indication criteria: failure to meet criteria for conservative management or persistence of HSIL after 24 months of follow-up. Data analysis Patient characteristics will be described according to whether their lesions regress spontaneously or conization is indicated throughout the follow-up period. Numbers and percentages will be provided for categorical variables, whereas means and standard deviations will be used for quantitative variables. The association between potential predictive variables and HSIL resolution will be analyzed with chi-square tests (or Fisher's exact test when appropriate) for categorical variables and t-tests for quantitative variables. Variables assessed will include: age and ethnicity, smoking status, number of sexual partners within the previous 6 months, HPV type, alternative therapies, HPV vaccination. The characteristics of patients who have conization will be summarized, comparing patients referred to surgery because they failed to meet surveillance criteria to those referred to surgery due to persistence of HSIL at end of follow-up. The statistical analysis will be performed with SPSS version 26.0 (SPSS Inc., Chicago, IL, USA). Differences between groups were considered statistically significant at p \< 0.05.