This review found a lack of evidence for the effectiveness of GnRHa in preserving pregnancy rates or ovarian biomarkers. Further trials were recommended. Oral contraceptive pills were not found to be effective in preserving fertility, but conclusions were limited by a lack of good research evidence. The authors' recommendation for further trials appears appropriate.

To evaluate the effectiveness of hormonal interventions for fertility preservation in women with cancer or systemic lupus erythematosus (SLE) receiving chemotherapy

MEDLINE, EMBASE and Web of Science as well as selected relevant conference proceedings were searched from 1982 to March 2010. Published and unpublished studies were sought. There were no language restrictions. Keywords were reported. References of all identified studies were consulted to identify further trials.

Eligible studies needed to be in the form of randomised or nonrandomised prospective comparative studies or have a prospective intervention arm compared with retrospective historical cohort. Studies needed to report on medically confirmed outcome of ovarian preservation (amenorrhea, follicle count, hormonal status, pregnancy or biomarkers). Primary outcomes were ovarian failure rate and pregnancy rates. Secondary outcomes were biological markers for ovarian reserve. Studies varied in their definitions of ovarian dysfunction. Across the studies, participants had SLE or cancers such as lymphoma, Hodgkin's disease, non-Hodgkin's lymphoma, leukaemia or breast cancer. Age of participants varied between studies. Both hormonal therapy regimens and chemotherapy regimens varied across the studies.

Two reviewers independently selected studies for inclusion in the review. Disagreements were resolved through consensus or by referral to a third reviewer.

Randomised controlled trials (RCTs) were assessed using criteria described in The Cochrane Handbook for randomisation, allocation concealment, sample size, exclusions after randomisation and different lengths of follow-up. Observational studies were assessed using the Newcastle-Ottawa scale for selection of cases/cohorts and controls, comparability and exposure (maximum available score of 7).

Trials were assessed for methodological quality by two reviewers. Disagreements resolved through consensus or by referral to a third reviewer.

Authors of studies were contacted when clarification was needed. Individual study results were expressed as relative risk (RR) with 95% confidence intervals (CI).

Two reviewers independently extracted data for the review. In cases of disagreement, a third reviewer extracted the data.

For interventions based on GnRHa (gonadotrophin releasing hormone agonistic analogue), studies were pooled using meta-analysis techniques stratified according to disease type and study design. Heterogeneity was assessed graphically and through use of Χ² and I² tests. Meta-analysis was not conducted for oral contraceptives and a brief narrative summary was presented.

Twenty-one studies (five RCTs) were included in the review (total number of participants not stated). Follow-up in the RCTs ranged from eight months to 5.8 years and across the observational studies varied from six months to 15 years. All RCTs had unclear methods of blinding and three of the five had problems with randomisation and allocation concealment. Observational studies were prospective in design, but used historical controls and scored from 4 to 7 out of 7 on the quality scale.

GnRHa was beneficial in preserving the menstrual cycle (RR 0.26, 95% CI 0.22 to 0.34; 15 studies, 681 patients) and increasing pregnancy rates (RR 1.51, 95% CI 1.01 to 2.28; 10 studies). Results based on RCTs only did not find a statistically significant benefit for either outcome. There was no significant difference in marker levels in patients treated with GnRHa (three studies). Adverse events were not described in the included studies.

One study with 214 participants showed a positive effect for oral contraceptives in reducing amenorrhea rates mainly in women under 30 years. Four smaller studies (sample sizes of 90, 44, 33 and nine) found no significant effect.

There was a lack of evidence for the effectiveness of GnRHa in preserving pregnancy rates or ovarian biomarkers and further trials were recommended. Oral contraceptive pills were not found to be effective in preserving fertility, but conclusions are limited by the lack of good research evidence.

This review was based on broadly defined inclusion criteria. Searching encompassed a range of sources and included unpublished and foreign language material, which minimised potential for publication and language biases. Two reviewers were involved in the selection of studies, quality assessment and data extraction, which helped to avoid bias in these processes. Pooling studies in a meta-analysis may have been inappropriate given variation in participants, treatments and definitions of outcomes. Observational studies are known to be more prone to bias than RCTs, so overall results combining the two types of study may not be reliable. The authors presented the results of RCTs only and noted the differences in outcome.

Overall, the authors' recommendation for further trials appears to be appropriate.

Practice: The authors stated that GnRHa should not be considered as standard of care for fertility preservation in patients treated with chemotherapy, but may represent an alternative method of fertility preservation.

Research: The authors recommended adequately powered RCTs with sufficient follow-up and noted some ongoing trials. They noted that further understanding of chemotherapy-induced ovarian toxicity (in particular the mechanism involved) was essential.

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.