Abstract PS5-01-04: Clinical predictors for first-line treatment duration in HER2-positive metastatic breast cancer Results from the AGMT_MBC-Registry

View at publisher
Authors: Gabriel Rinnerthaler, S. P. Gampenrieder, Angelika Pichler, Walter Herz, Renate Pusch, Clemens Dormann, Christoph Suppan, Margit Sandholzer, Thomas L. Winder, Sonja Heibl, Lukas Scagnetti, C. A. Schmitt
+13 more August Felix Zabernigg, Daniel Egle, C. Hager, P. Pichler, Florian Roitner, J. Andel, Kathrin Strasser‐Weippl, R. Bartsch, Vanessa Castagnaviz, M. Hubalek, Michael Knauer, Christian Fridolin Singer, R. Greil
Cited by: 0

Abstract

Abstract Background: Based upon the CLEOPATRA trial, a combination of docetaxel plus trastuzumab and pertuzumab (THP) has been the standard treatment for patients with metastatic HER2-positive breast cancer. However, this standard has recently been challenged by the results of the DESTINY-Breast09 study, which compared trastuzumab-deruxtecan (T-DXd) plus pertuzumab with THP. In this analysis, we aimed to develop a risk model based on known parameters that identifies patients with a long first-line (1L) therapy duration, as these patients may not need 1L T-DXd therapy. Patients and Methods: The AGMT_MBC-Registry is a nationwide, multicenter, ongoing retrospective and prospective registry for metastatic breast cancer (MBC) patients in Austria. Patients with HER2-positive breast cancer treated with a 1L chemo, trastuzumab plus pertuzumab combination were included in this analysis. Time to next treatment (TTNT) was defined as the interval from initiation of 1L to the initiation of the next subsequent treatment line. To identify prognostic factors associated 1L TTNT a multivariable Cox proportional hazards model was performed. Clinical, demographic, and tumor-related covariates - including metastatic sites, age, hormone-receptor (HR) status, and tumor grade - were included in the analysis. Variable selection was performed using a bidirectional stepwise approach based on the Akaike Information Criterion. Patients with missing values were excluded. A risk score was derived from the linear predictor of the Cox model. Survival-optimized cut-off points were identified using maximally selected rank statistics to stratify patients into low-, intermediate-, and high-risk groups. Hazard ratios (HRs) with 95% confidence intervals (CIs) are reported. Time-to-event analyses were performed using Kaplan-Meier estimates, with group-specific survival probabilities reported at 12, 18, and 36 months. Results: As of March 11, 2025, the registry included 3,094 patients, and 218 patients with HER2-positive MBC were available for this analysis. Median age at diagnosis of MBC was 58 years (range 26-85) and 62% had hormone receptor (HR) positive disease. Bone (57%), liver (43%), and lung metastases (35%) were the most common metastatic sites; brain, pleural, and peritoneal metastases were rare (

Fields of Study & Subjects

HER2/EGFR in Cancer Research Breast Cancer Treatment Studies Advanced Breast Cancer Therapies

Publication details

Journal / source
Clinical Cancer Research
Volume
32
Issue
4_Supplement
Pages
PS5–01
Publication date
17 February 2026
Language
EN