Background: With increasing interest in value based care multiple value calculators have been introduced that rely on clinical trial efficacy and toxicity data as inputs. However, clinical trial patient mix and outcomes may vary from those patients treated in the community oncology setting. When sufficient time has elapsed since approval, real-world studies of patient populations may be needed to augment trial data for realistic value assessment. Methods: A chemotherapeutic drug approved in 2007 for metastatic/locally advanced breast cancer was examined through a community oncology specific EMR database. Patients with a confirmed diagnosis of advanced breast cancer (N = 199), treated with the drug of interest that conformed to the labeled indication, during the period (December 2007 –November 2015) and with at least 6 months of continuous follow up were included in the study. Patients with treatment for a secondary cancer or participation in any clinical trial were excluded. Progression free survival (PFS) was defined by time from drug initiation to either initiation of a subsequent line of therapy (LOT), discontinuation of therapy (no subsequent LOT), or death from any cause. Results were compared against package insert referenced clinical trial data. Results: The pivotal clinical trial data consisted of patients (N = 375) that were 67% white, median age 53, baseline Karnofsky 70-100%. Median PFS was 5.7 months. Patients identified through EMR data were 45% white (although 30% unknown race), median age 58, baseline ECOG performance status of 1.2 (available for 47% of patients). Median PFS of 4.3 months (95% confidence interval, 3.9–5.5). Conclusions: This real-world EMR data analysis affirms prior reports with patients older, sicker, and possibly less diverse than their trial cohort. A 25% shorter PFS is a significant finding that may impact on value calculation. Further study is needed to compare real world patient populations to clinical trial populations and account for differences in demographics and expected treatment outcomes, especially given that clinical trial data is informing newly developed value calculators.