We conducted an exhaustive literature searches in the following databases: MEDLINE (via PubMed), The Cochrane Library (CENTRAL), and EMBASE (via embase.com), from their inception until the most recent date. We will design a search strategy that is tailored to the requirements of each database, which will include a combination of controlled vocabulary and search terms related to each gynecological cancer. Appendix displays an exploratory search strategy for MEDLINE. When necessary, we will use validated filters to retrieve the appropriate study designs.

Two independent researchers screened the references yielded by the search to reach an agreement on the inclusion of studies.

The PICOS (Population, Intervention, Comparators, Outcomes, Study Design) criteria are developed a priori to guide the scope of the review, along with the procedures, selection, and synthesis of the literature search. The selection criteria were as follows:

(Population) menopausal women of any age with gynecological cancer receiving HMT; (Intervention) any MHT preparation (oestrogens alone or combined with a progestogen, tibolone or tissue selective oestrogen complex) or any route of administration (oral, transdermal, vaginal or intra-nasal); (Outcome) recurrence and mortality; (Study Design) randomized controlled trials, and related extension studies or follow-up reports. Any complete article that met the inclusion criteria was reviewed in detail.

We described the synthesis of the evidence following the PRISMA guidelines [9]. We assessed the risk of bias of the eligible studies using the Cochrane tool for clinical trials, which takes into account the evaluation of five possible sources of bias (selection, performance, detection, attrition and report bias) [10]. For observational studies, we adapted the ROBINS I tool, focusing on the evaluation of the impact of the confounding variables, selection bias, outcome measures, and attrition [11]. Pooled analyses were conducted using the Mantel-Haenszel method and the random effects model included within the RevMan software statistical package (v 5.3.5) [12].

We made explicit judgements on the certainty of the evidence for each outcome of interest according to GRADE criteria [13]. Quality will be classified as high, moderate, low or very low, based on several factors (including risk of bias, inaccuracy, inconsistency, lack of directionality and publication bias).

Three randomized clinical trials (RCTs) were included on the impact of MHT on ovarian cancer recurrence and mortality [14-16]. Six retrospective observational studies were also considered [17-22]. Most patients received combined oral MHT, primarily with equine conjugated estrogens (ECE) 0.625–1.25 mg/day plus medroxyprogesterone 4 mg/day.

There were no differences in overall survival when all studies were pooled (RR 0.90, 95% CI: 0.73 to 1.10; Fig. 1a), but differences emerged when only RCTs were analyzed [14-16] indicating improvement after MHT (HR 0.71, 95% CI: 0.54 to 0.93; Fig. 1b).

One of the cohort studies [21] presented the results according to the age of the participants. Survival in women under 55 years was higher when using MHT, according to a univariate analysis (HR 0.41, 95% CI: 0.19 to 0.89, p = 0.023), but not in the multivariate model (according to disease stage and chemotherapy administration; HR 0.49, 95% CI: 0.23 to 1.09, p = 0.08). For patients over 55 years old, MHT did not affect survival.

Combined data from two RCTs [14,16] showed an increase in recurrence-free survival with MHT (HR 0.72, 95% CI: 0.58 to 0.90; Fig. 1c), and a reduction in the risk of recurrence (RR 0.81, 95% CI: 0.70 to 0.93; Fig. 1d). These results were maintained when data from retrospective studies were added (RR 0.81, 95% CI: 0.71 to 0.92; Fig. 1d).

Disease-free survival was also higher for MHT users younger than 55 years in the Power 2016 study [21], both in the univariate (HR 0.34, 95% CI: 0.17 to 0.69; p = 0.003) and multivariate models (adjusted HR 0.35, 95% CI: 0.17 to 0.74; p = 0.006). For women over 55 years of age, the use of MHT was not associated with increased disease-free survival.

Fig. 1.

Ovarian cancer. (a) Mortality (data from clinical trials and retrospective studies). (b) Overall survival (data from clinical trials). (c) Recurrence-free survival (data from clinical trials). (d) Recurrence (data from clinical trials and retrospective studies).

One RCT was identified [23] with 1236 participants where neither the dose nor the route of administration of MHT is described, with a mean follow-up of 35.7 months. Eight observational studies were also included, one prospective [24] and seven retrospective studies [25-31] with a total of 1801 endometrial cancer survivors who received MHT and 6015 who did not. Almost all were limited to patients with early stages of the disease.

The RCT [23] found no difference in recurrence according to MHT use (RR 1.17, 95% CI: 0.54 to 2.50), the same being true for the combined analysis of the RCT and the prospective study (RR 1.08, 95% CI: 0.52 to 2.24; Fig. 2a). However, when adding the results of the retrospective studies, a significant reduction in recurrence was observed with the use of MHT (RR 0.64, 95% CI: 0.50 to 0.82; Fig. 2b). In particular, recurrence was reduced with combined MHT (RR 0.31, 95% CI: 0.13 to 0.73; Fig. 2c) but not with estrogen alone (RR 0.57, 95% CI: 0.30 to 1.10). These results were found to be independent of the stage of the disease (RR for Stage I 0.14, 95% CI: 0.03 to 0.70; RR for Stage I–II 0.70, 95% CI: 0.54 to 0.90; Fig. 2d).

Fig. 2.

Endometrial cancer. (a) Recurrence (data from clinical trials and prospective studies). (b) Recurrence (data from clinical trials and prospective and retrospective studies). (c) Recurrence (by type of treatment). (d) Recurrence (by stage of disease). (e) Recurrence (analysis of publication bias).

One RCT was identified [32] with 120 patients under 45 years of age with Stage I and II cancer. Oral MHT was combined with different formulations over an observation period of more than 5 years. This RCT found no significant differences in either mortality (RR 0.62, 95% CI: 0.33 to 1.15; Fig. 3a) or recurrence (RR 0.57, 95% CI: 0.31 to 1.15; Fig. 3b).

Fig. 3.

Cervical cancer. (a) Mortality. (b) Recurrence.

There is considerable confusion regarding the appropriateness of prescribing MHT in women with gynecological cancer, particularly because of the fear of recurrence or increased mortality that may occur with its use.

Women who have suffered from gynecological cancer often present earlier and more intense menopausal symptoms due to ovarian surgery or the effects of certain treatments (RT, QT), which have a severe impact on their quality of life. These long-term risks often overlap with those suffered by women with premature ovarian failure (POF), thus extending the suitability of MHT [8].

This report complements the recommendations contained in a recent position paper published by EMAS and IGCS [33].

This is the first published work where several systematic reviews and meta-analyses are gathered to analyze the recurrence and mortality associated with MHT use in women survivors of gynecological cancer (endometrial, ovarian, or cervical).

This is also the first time that categories of evidence (eligibility criteria) have been distinguished for the use of MHT in these patients, using the strictest methodological tools.

Other systematic reviews have been included [34,40,41] as sources of studies relevant to this report.

The quality of evidence is low overall. Many studies include the generic use of MHT without distinguishing between dose, formulation, or route of administration.

Our report, however, has identified some important areas of improvement for future research. It is expected that the results will contribute to the development of studies that further examine the safety and efficacy of MHT for treating menopausal symptoms in gynecological cancer survivors. Larger RCTs should be conducted, and over a longer follow-up period, to evaluate the various MHT strategies.