145P Prognostic value of tumor-infiltrating lymphocytes in HER2-positive breast cancer treated with neoadjuvant chemotherapy and dual HER2-blockade: A TRAIN-2 sub study

Abstract

BackgroundPreviously, it has been demonstrated that stromal tumor-infiltrating lymphocytes (sTILs) provide independent prognostic information in early-stage HER2-positive breast cancer. The aim of our study was to validate these findings by studying sTILs and outcome measures in early-stage HER2-positive breast cancer patients who were randomized to receive neoadjuvant chemotherapy with or without anthracyclines plus trastuzumab and pertuzumab in the TRAIN-2 study.MethodsFollowing sTIL scoring on pre-treatment biopsies by two pathologists, the geometric mean was calculated. When the two scores differed by more than 10%, a final score was assigned after revision by one pathologist. The association between sTILs and pathological complete response (pCR) was studied using logistic regression analyses adjusted for age, hormone receptor (HR) status, clinical T-stage, clinical N-stage, tumor grade and treatment arm. Similarly, the relation between sTILs and event-free survival (EFS) was evaluated by Cox regression analyses.ResultsTable: 145PStromal TILs, pathological complete response and 5-year event-free survival ratesGroupNumber of events5-year EFS (%)95% CILow sTILs, no pCR11/7383.875.3 - 93.2High sTILs, no pCR8/4581.771.0 - 94.1Low sTILs, pCR13/12981.569.7 - 95.2High sTILs, pCR3/12397.594.7 - 100 Open table in a new tab ConclusionsWe did not observe a significant association between sTILs and pCR or EFS after adjustment for clinical variables. However, patients with pCR and high sTILs show excellent 5-year EFS rates compared to patients with no pCR or low sTILs. Early-stage HER2-positive breast cancer patients with high sTILs and pCR may be candidates for de-escalated adjuvant treatment, whereas patients with low sTILs may benefit of additional treatment.Clinical trial identificationNCT01996267.Legal entity responsible for the studyBOOG Study Center.FundingRoche.DisclosureA.E. van Leeuwen-Stok: Financial Interests, Institutional, Funding, Roche funding for the TRAIN-2 study: Roche . H. Horlings: Financial Interests, Institutional, Funding: Roche; Non-Financial Interests, Institutional, Advisory Role: SlideScore.com. R.F. Salgado: Non-Financial Interests, Institutional, Other: Merck , Bristol Myers Squibb; Financial Interests, Institutional, Funding: Puma Biotechnology, Merck , Roche; Financial Interests, Personal, Advisory Board: Roche , Exact Sciences, Bristol Myers Squibb . G.S. Sonke: Financial Interests, Institutional, Funding: Roche , Merck , Novartis , Agendia, AstraZeneca , Seagen; Financial Interests, Institutional, Advisory Role: Seagen, Biovica. All other authors have declared no conflicts of interest. BackgroundPreviously, it has been demonstrated that stromal tumor-infiltrating lymphocytes (sTILs) provide independent prognostic information in early-stage HER2-positive breast cancer. The aim of our study was to validate these findings by studying sTILs and outcome measures in early-stage HER2-positive breast cancer patients who were randomized to receive neoadjuvant chemotherapy with or without anthracyclines plus trastuzumab and pertuzumab in the TRAIN-2 study. Previously, it has been demonstrated that stromal tumor-infiltrating lymphocytes (sTILs) provide independent prognostic information in early-stage HER2-positive breast cancer. The aim of our study was to validate these findings by studying sTILs and outcome measures in early-stage HER2-positive breast cancer patients who were randomized to receive neoadjuvant chemotherapy with or without anthracyclines plus trastuzumab and pertuzumab in the TRAIN-2 study. MethodsFollowing sTIL scoring on pre-treatment biopsies by two pathologists, the geometric mean was calculated. When the two scores differed by more than 10%, a final score was assigned after revision by one pathologist. The association between sTILs and pathological complete response (pCR) was studied using logistic regression analyses adjusted for age, hormone receptor (HR) status, clinical T-stage, clinical N-stage, tumor grade and treatment arm. Similarly, the relation between sTILs and event-free survival (EFS) was evaluated by Cox regression analyses. Following sTIL scoring on pre-treatment biopsies by two pathologists, the geometric mean was calculated. When the two scores differed by more than 10%, a final score was assigned after revision by one pathologist. The association between sTILs and pathological complete response (pCR) was studied using logistic regression analyses adjusted for age, hormone receptor (HR) status, clinical T-stage, clinical N-stage, tumor grade and treatment arm. Similarly, the relation between sTILs and event-free survival (EFS) was evaluated by Cox regression analyses. ResultsTable: 145PStromal TILs, pathological complete response and 5-year event-free survival ratesGroupNumber of events5-year EFS (%)95% CILow sTILs, no pCR11/7383.875.3 - 93.2High sTILs, no pCR8/4581.771.0 - 94.1Low sTILs, pCR13/12981.569.7 - 95.2High sTILs, pCR3/12397.594.7 - 100 Open table in a new tab ConclusionsWe did not observe a significant association between sTILs and pCR or EFS after adjustment for clinical variables. However, patients with pCR and high sTILs show excellent 5-year EFS rates compared to patients with no pCR or low sTILs. Early-stage HER2-positive breast cancer patients with high sTILs and pCR may be candidates for de-escalated adjuvant treatment, whereas patients with low sTILs may benefit of additional treatment. We did not observe a significant association between sTILs and pCR or EFS after adjustment for clinical variables. However, patients with pCR and high sTILs show excellent 5-year EFS rates compared to patients with no pCR or low sTILs. Early-stage HER2-positive breast cancer patients with high sTILs and pCR may be candidates for de-escalated adjuvant treatment, whereas patients with low sTILs may benefit of additional treatment.