In the age of personalized medicine, the multitude of scientific publications describing new biomarkers has raised a lot of criticism, 1 as many of these results have been rather difficult to validate. As a consequence to this criticism, we are observing a considerable improvement in the design of clinical biomarker studies, which are increasingly based on large sample cohorts from clinical trials. The article by Loi et al2 that accompanies this editorial exemplifies this. Loi et al2 report a prognostic role for tumor-associated lymphocytes in triple-negative breast cancer in a large prospective clinical trial, evaluating more than 2,000 tumor samples from patients with nodepositive breast cancer treated within the Breast International Group (BIG) 02-98 trial. The main finding of the study was that in estrogen receptor–negative/human epidermal growth factor receptor 2 (HER2)–negative breast cancer (n 256) an increased lymphocytic infiltrate was linked to a reduced relapse rate and improved survival. This association was independent of the type of chemotherapy and was not observed in hormone receptor–positive tumors (n 1,078). In the subgroup of HER2-positive tumors (n 297) there was a significant interaction with benefit from anthracycline-only chemotherapy. It should be noted, however, that trastuzumab was not part of the adjuvant treatment in BIG 2 to 98, which limits the interpretation of theresultsintheHER2-positivesubsetinthecontextofcurrenttherapeutic approaches.

The study by Loi et al2 shows that tumor-infiltrating lymphocytes can be evaluated as a distinctive histological parameter using the predefined standardized definitions of intratumoral lymphocytes, stromal lymphocytes and lymphocyte-predominant breast cancer (LPBC). We have evaluated the nearly identical parameters in tumor samples from the neoadjuvant GeparTrio study, with a slightly different definition of LPBC. 3 In this study, we have shown that a pretherapeutic immunological infiltrate is a predictive factor for response to neoadjuvant therapy. In the neoadjuvant situation, an increased rate of pathological complete response was observed in different tumor subtypes including hormone receptor–positive tumors. In contrast, the survival effect in the study by Loi et al was present only in the triple-negative subgroup. This difference could be explained by the fact that pathological complete remission is a surrogate end point for survival for triple-negative, but not for hormone receptor–positive breast cancer. 4