Background: Assessing the residual cancer burden (RCB) predictive performance, the potential subgroup effects, and time-dependent impact on breast cancer patients who underwent neoadjuvant therapy in a developer's independent cohort is essential for its usage in clinical routine.
Methods: Between 2011 and 2016, the RCB scores of 184 female breast cancer patients were prospectively collected, and subsequent clinicopathological and follow-up data were obtained retrospectively. Recurrence-free survival (RFS), overall survival (OS), as well as subgroup analysis, and time-dependent variables were calculated with multivariate, complex, or linear statistical models.
Results: A total of 184 patients (HER2 33%, TNBC 27%), with a mean follow-up time of 4 years, treated with neoadjuvant systemic therapy (92% anthracycline-taxane based) were analyzed revealing 43 events (38 recurrences, 28 deaths). High RCB scores were associated with recurrence (median index: 2.34 vs. 1.39 points, rank-sum p < 0.0001), decreased RFS (hazard ratio [HR] = 1.80, 95% confidence interval [CI] 1.44-2.24, p < 0.0001) and reduced OS (HR 1.96, 95% CI 1.49-2.59, p < 0.0001). The RCB score showed proportionality of hazards (interaction HR with linear follow-up time = 1.00, p = 0.896) and good discriminating power (Harrell's c index 0.7).
Conclusions: Our results confirm the RCB score as externally valid prognostic marker and being independent of molecular subtype for RFS and OS in a clinical setting.