Pregnancy and survival after primary breast carcinoma in young women: A meta-analysis of healthy mother effect studies.

9122 Background: Due to a trend towards postponing childbearing in the western world and the increased rate of infertility among young couples, diagnosis of breast cancer in nulliparous women is expected to rise over the next decades. Thereafter an increased number of these women is expected willing to conceive after breast cancer. Based on literature, most physicians usually recommend pregnancy two years after diagnosis of breast cancer. Nonetheless, the prognostic significance of pregnancy following treatment of breast cancer is limited to trials with small patient cohorts. Furthermore, a healthy mother effects (HME) selection bias appears to be consistent in most of these studies, since women undergoing childbearing after treatment were healthier when compared to the control group. Aim: To perform a systematic review and meta-analysis of published trials corrected for HME bias in order to assess the effect of pregnancy (at least 10 months after diagnosis) vs. no pregnancy on overall survival of primary breast cancer patients aged <45. Methods: We searched MEDLINE and ISI (last search January 2010) for eligible studies. From each study we extracted relative Hazard Ratio (HR) or, if not provided, all the necessary data to impute it. In case period from diagnosis was not reported we extract relative data to estimate it. Results: Electronic searches strategy yielded 1606 hits pertaining to 20 potentially eligible studies involving 49370 premenopausal breast cancer patients. Ten studies (total screened 41415, pregnants 1157, matched-controls 13187) were eligible since considering HME potential bias in matching controls. Among these 9 studies (pregnants 1089, matched-controls 13051) presented data for analyses. Overall survival was statistically higher among patients getting pregnant compared to no pregnancy: fixed effect model estimated pooled-HR 0.51 (95% CI: 0.42-0.62). No study heterogeneity was observed: Q=10.4 p.=0.17; I∧2 = 48%. Conclusions: Pooled available evidence indicates thatin primary breast cancer patients younger than 45 years old, pregnancy (at least 10 months after the diagnosis) does not jeopardize prognosis and appears to provide significant survival benefits No significant financial relationships to disclose.