Poster Presentation Clinical Oncology Society of Australia Annual Scientific Meeting 2019

Prognostic models for predicting ovarian function in young breast cancer patients after chemotherapy: a systematic review and meta-analysis. (#386)

Zobaida Edib 1 2 , Yasmin Jayasinghe 1 2 3 , Martha Hickey 1 2 , Alexandra Gorelik 4 5 , Michelle Peate 1 2
  1. University of Melbourne, Parkville, VICTORIA, Australia
  2. Department of Obstetrics and Gynaecology, The Royal Women's Hospital, Parkville, Victoria, Australia
  3. The Royal Children's Hospital, Parkville, Victoria, Australia
  4. Department of Medicine, The Royal Melbourne Hospital, University of Melbourne, Parkville, Victoria, Australia
  5. School of Behavioural & Health Sciences, Australian Catholic University, Melbourne, Victoria, Australia

Aim: The potential reduction of ovarian function due to cancer treatments is a high priority for young breast cancer patients, with some patients willing to opt for less effective treatments to reduce the impact on fertility. Thus, reliable and accurate prediction of post-treatment ovarian function is important to optimise adjuvant endocrine treatment and individualise fertility counselling to facilitate informed oncofertility decision-making. The aim of this systematic review and meta-analysis was to identify, conduct a critical appraisal and summarise existing prognostic models developed for predicting ovarian function in young breast cancer patients.

Methods: An extensive literature search was conducted using combination algorithm of controlled vocabulary and search terms through MEDLINE, EMBASE and Cochrane Library without any date or language restriction up to May 2019. Included were any articles developing or validating prognostic models of post-treatment ovarian function in premenopausal breast cancer patients. Two authors independently screened the studies, extracted data and assessed risk of bias. Results were summarised qualitatively, and meta-analysis was conducted using random-effects model.

Results: The systematic review found nine prognostic models predicting ovarian function of premenopausal breast cancer patients, with age and ovarian markers reported as the most common predicting factors. Many baseline and clinical factors influencing post-treatment ovarian function of young breast cancer patients have not been explored. Although the average discrimination performance of the models is good (C statistic 0.81, 95% CI 0.78-0.86), the quality assessment reported a high risk of bias across all the included studies. These limitations have an impact on development and validation of these existing models, with less applicability in clinical practice.

Conclusions: Further research is needed with a large sample to identify and assess the added value of baseline and missing clinical predictors to ovarian function recovery risk prediction models for young breast cancer patients, alongside with clearer reporting and validation.