Methods: We performed a systematic review and meta-analysis to compare the effects of PMRT to no PMRT for elderly patients with intermediate-risk breast cancer. We searched PubMed for eligible studies from Jan 2008 to Dec 2018. We assessed the methodological quality of the included studies using the ROBINS-I tool and performed the meta-analysis with random effects model. The primary outcome of interest was overall survival (OS); secondary outcomes were breast cancer specific survival (BCSS), loco-regional (LRR) and distant disease recurrence (DDR).

Results: We found 2 retrospective cohort studies with 743 patients directly comparing PMRT to no PMRT. Both studies were judged to have serious risk of bias in their methodological quality. The pooled results suggest that PMRT was associated with a 20% relative reduction in the hazard in death, ranging from 41% relative reduction, a substantial negative association to 10% relative increase, a small positive association (HR 0.80, 95% CI: 0.59-1.1, P=0.62, I2=0%). PMRT was also associated with a 17% relative reduction in the hazard for breast cancer related death, ranging from 52% relative reduction, a substantial negative association to 41% relative increase, a substantial positive association (HR 0.83, 95% CI: 0.48-1.41, P=0.48, I2=0%). One study did not observe any significant differences in LRR and DDR between the two groups.