Hypothesis / aims of study
Gynaecological cancer and breast cancer are prevalent, representing 16% and 26% of all cancers in females, respectively [1]. Females diagnosed with these cancers are at risk of developing pelvic floor disorders and sexual dysfunction due to the cancer itself and consequences of treatment. A previous review examined the effectiveness of pelvic floor muscle (PFM) therapies for gynaecological cancer survivors [2]. Despite the limited number and varying methodological quality of identified studies (n=7), this study revealed low-to-moderate evidence supporting the effect of PFM training or dilator therapy after gynaecological cancer for improving sexual function [2]. New evidence has emerged since this review and an update is important. In contrast, little is known regarding the effectiveness of PFM therapies to treat pelvic floor disorders and sexual dysfunction in the breast cancer population. In addition to PFM therapies, education-based interventions may also contribute to improvements in sexual and pelvic floor function, and they are often delivered in combination with PFM therapies. The aim of this study is to systematically review the effectiveness of conservative therapies (PFM therapies and education-based interventions) on bladder, bowel, vaginal, sexual, psychological function, quality of life, or PFM function in females treated for gynaecological or breast cancer.
Study design, materials and methods
A systematic review was conducted, guided by PRISMA reporting guidelines. Informed by a previous review [2] and with the assistance of a librarian, the search strategy was created based on Participants, Intervention, Comparator, Outcomes eligibility criteria. Six databases (Medline, Embase, CINAHL, Cochrane Library, PsycINFO, and Emcare) were systematically searched from their inception dates to December 2023. Trials included in the previous review [2] were considered for inclusion, and a search update using the same search strategy was performed to identify newly published records since June 2018. Two additional and separate searches were performed, one identifying records on education-based interventions in gynaecological cancer populations and one identifying records on both PFM therapies and education-based interventions in breast cancer populations.
Trials written in English or French and published in peer-reviewed journals were eligible for inclusion. Inclusion criteria were:
Participants: at least 75% of adult females who had undergone any treatment for any type of gynaecological or breast cancer.
Interventions: conservative therapies: (a) PFM therapies, i.e., any therapy that targeted the PFM tissues, e.g., performance of voluntary activation/relaxation of PFM muscles, manual therapy, dilator therapy, desensitisation techniques; or (b) education-based interventions, i.e., any intervention that targeted cognitive aspects related to the pelvic floor by providing information on its structure, function, or exercise. Trials were considered for inclusion if other interventions were also used (e.g., general exercise, application of vaginal lubricant or moisturiser). Trials were excluded if a pharmaceutical ingredient was used.
Comparator: randomised controlled trials (RCTs) with any comparator.
Outcomes: any outcomes related to bladder, bowel, vaginal, sexual, psychological function, quality of life, or PFM function.
One reviewer independently selected the trials by screening titles and abstracts, followed by full texts. One reviewer verified the eligibility of included trials. Two reviewers independently extracted data using a customised data extraction sheet. Extracted data included trial details, population characteristics, treatments arms (intervention group and comparative group) and their characteristics, outcome measures, and results of between-group comparisons. The Template for Intervention Description and Replication (TIDieR) checklist [3] guided data extraction for the intervention group characteristics. The risk of bias of included trials was assessed independently by two reviewers using the Cochrane risk-of-bias tool (RoB2). Any disagreements between reviewers for selection of trials, data extraction, and risk of bias were solved through discussion. The percentage of agreement between reviewers for risk of bias was calculated. A narrative synthesis approach was adopted.
Results
A total of 5069 records were screened. After screening titles and abstracts, 108 full texts were evaluated against eligibility criteria. Sixteen trials were retained. The reasons for exclusion were: out-of-scope language (n=4), trial design (n=39), intervention (n=32), population (n=7), outcomes (n=9). One trial was removed due to inclusion duplication. Among the retained trials, 11 RCTs were conducted in gynaecological cancer populations (n=1288) and 5 RCTs in breast cancer populations (n=1063).
All trials provided interventions as post-cancer treatment rehabilitation, except one trial in which the intervention commenced pre-radiotherapy and continued post-radiotherapy. The number of items from the TIDieR checklist used to describe therapies delivered to the intervention group ranged between 4-11 (out of 12), representing approximately 33%-92% of the total checklist items. The overall risk of bias was assessed as ‘some concerns’ for all RCTs using the RoB2. Agreement between reviewers for all scored items was 86%. Table 1 and Table 2 present the RCTs of PFM therapies and education-based interventions for gynaecological and breast cancer populations, respectively.
Interpretation of results
For females treated for gynaecological cancer, the number of trials investigating the effectiveness of conservative therapies on bladder, bowel, vaginal, sexual, psychological function, quality of life, or PFM function in are increasing, however the certainty of effect remains low due to limited trials in tumour types other than cervical cancer, heterogeneity of interventions tested and outcomes measured, methodological limitations, and poor reporting. As the majority of trials represent participants with cervical cancer, these findings may be generalisable to females following treatment for cervical cancer.
In this review, meta-analysis was not possible given diversity of interventions, comparative groups, and outcomes. Primary outcomes and primary end-points were poorly reported in most trials and few trials provided sample size calculations. Nevertheless, positive signals of clinical improvement following conservative therapies from some trials will help inform power calculations for future trials. Since the previous review [2] more trials have emerged, but no real change in findings is evident for gynaecological cancer, other than an improvement in a bladder function outcome, post-void residual, following PFM therapy plus intermittent self-catheterisation. Combinations of PFM-active therapies and education may provide stronger effects than education interventions alone. Given the paucity of RCTs, results from other study designs (non-RCTs) may build our understanding of feasibility of recruitment and delivery of interventions.
For females treated for breast cancer, five trials were found, and all except one trial were education-based interventions. No interpretation of the strength of evidence is possible at present. However, the direction of findings suggests a promising role for education-based interventions to improve sexual function. More trials investigating PFM therapies are needed. Results from single-cohort pre-post studies suggest that PFM therapies could improve PFM strength, sexual function and reduce urinary symptoms in this population. Results from non-RCTs may provide insight for direction of future trials in this understudied population.
In both gynaecological and breast cancer, more trials with robust methods are urgently needed, with sample size calculations provided, controlled for important patient characteristics (tumour type, cancer treatment type, time since cancer treatment completed, outcomes at baseline) and with well-described interventions.