Alpha-smooth muscle actin expression in the stroma predicts resistance to trastuzumab in patients with early-stage HER2-positive breast cancer.

Purpose The companion diagnostic test for trastuzumab has not changed much in the last 25 years. We used high-plex digital spatial profiling to identify biomarkers besides HER2 that can help predict response to trastuzumab in HER2-positive breast cancer. Experimental design Fifty-eight protein targets were measured in three different molecularly defined compartments by the NanoString® GeoMx® Digital Spatial Profiler (DSP) in a tissue microarray containing 151 breast cancer patients that received adjuvant trastuzumab as part of the HeCOG 10/05 clinical trial. Promising candidate biomarkers were orthogonally validated with quantitative immunofluorescence (QIF). RNA sequencing data from the NeoALTTO study were accessed to provide independent cohort validation. Disease-free survival (DFS) was the main outcome assessed. Statistical analyses were performed using a two-sided test (α=0.05) and multiple testing correction (Benjamini-Hochberg method, FDR Results By DSP, high expression of alpha-smooth muscle actin (α-SMA), both in the leukocyte and stromal compartments, was associated with shorter DFS in univariate analysis (P = .002 and P = .023, respectively). High α-SMA expression in the stroma was validated by QIF after controlling for ER and PR status (HR, 3.12; 95% CI, 1.12-8.68; P = .029) showing recurrence on trastuzumab in the same cohort. In the NeoALTTO cohort, elevated levels of ACTA2 were predictive for shorter DFS in the multivariate analysis (HR, 3.21; 95% CI, 1.14-9.05; P = .027). Conclusions This work identifies α-SMA as a novel, easy-to-implement biomarker of resistance to trastuzumab that may be valuable in settings where trastuzumab is combined with other therapies.