HER2 Reactivation through Acquisition of the HER2 L755S Mutation as a Mechanism of Acquired Resistance to HER2-targeted Therapy in HER2+ Breast Cancer

Abstract

Purpose: Resistance to anti-HER2 therapies in HER2⁺ breast cancer can occur through activation of alternative survival pathways or reactivation of the HER signaling network. Here we employed BT474 parental and treatment-resistant cell line models to investigate a mechanism by which HER2⁺ breast cancer can reactivate the HER network under potent HER2-targeted therapies.Experimental Design: Resistant derivatives to lapatinib (L), trastuzumab (T), or the combination (LR/TR/LTR) were developed independently from two independent estrogen receptor ER⁺/HER2⁺ BT474 cell lines (AZ/ATCC). Two derivatives resistant to the lapatinib-containing regimens (BT474/AZ-LR and BT474/ATCC-LTR lines) that showed HER2 reactivation at the time of resistance were subjected to massive parallel sequencing and compared with parental lines. Ectopic expression and mutant-specific siRNA interference were applied to analyze the mutation functionally. In vitro and in vivo experiments were performed to test alternative therapies for mutant HER2 inhibition.Results: Genomic analyses revealed that the HER2L755S mutation was the only common somatic mutation gained in the BT474/AZ-LR and BT474/ATCC-LTR lines. Ectopic expression of HER2L755S induced acquired lapatinib resistance in the BT474/AZ, SK-BR-3, and AU565 parental cell lines. HER2L755S-specific siRNA knockdown reversed the resistance in BT474/AZ-LR and BT474/ATCC-LTR lines. The HER1/2-irreversible inhibitors afatinib and neratinib substantially inhibited both resistant cell growth and the HER2 and downstream AKT/MAPK signaling driven by HER2L755S in vitro and in vivoConclusions: HER2 reactivation through acquisition of the HER2L755S mutation was identified as a mechanism of acquired resistance to lapatinib-containing HER2-targeted therapy in preclinical HER2-amplified breast cancer models, which can be overcome by irreversible HER1/2 inhibitors. Clin Cancer Res; 23(17); 5123-34. ©2017 AACR.

Fig. 1. The HER2L755S mutation reactivates HER2 signaling in BT474 models and is associated with acquired resistance to L-containing regimens

A, Western blot analyses of L/LT-treated BT474 parental and resistant cells. BT474/AZ-P were treated 6h with DMSO or 1μM L. BT474/ATCC-P cells were treated with DMSO or 1μM L + 10μg/ml T. BT474/AZ-LR cells were cultured in the presence of 1μM L and BT474/ATCC-LTR cells were cultured in the presence of 1μM L+ 10μg/ml T. B, Number of mutations detected using whole exome sequencing identified in single samples or multiple samples in the BT474/AZ-P, BT474/AZ-LR, and BT474/ATCC-LTR lines using the BT474/ATCC-P line as reference. C, Pathogenic mutations identified by whole-exome sequencing (supplementary methods) in the BT474/AZ-P, BT474/AZ-LR, and BT474/ATCC-LTR lines using the BT474/ATCC-P line as reference are shown. The shade of blue in each block represents the cancer cell fraction (CCF) of each mutation in each line. Orange squares represent mutations that were found to be expressed using RNA sequencing. Mutations associated with the loss of the wild-type allele are indicated by a diagonal bar. The HER2 (encoded by the gene ERBB2) L755S mutation is highlighted in red.

Fig. 2. The HER2L755S mutation but not the G572V mutation is the driver of acquired resistance in the two BT474 LR/LTR derivatives with HER2 reactivation

A, Dox-inducible ectopic expression of C-terminal HA-tagged WT and mutant HER2 constructs were validated by western blot. B, WT-, G572V-, L755S- and G572V/L755S-HER2-expressing BT474/AZ-P cells were selected with Dox+L for 5 weeks. Pictures were taken at 4X magnification using an Olympus IX70 microscope with a RETIGA 1300R Fast 1394 camera and analyzed with Image-pro plus software (version 5.0). Scale Bar: 50μ. C, WT-, G572V-, L755S-, and G572V/L755S-HER2-expressing BT474/AZ-P cells were treated with or without 1μM L for 6h followed by western blot. L755S- and G572V/L755S-HER2-expressing BT474/AZ-P cells which survived the Dox+L selection for >5 weeks (L755S-LR and G572V/L755S-LR) were analyzed by western blot for HER2 and downstream signaling. D, BT474 parental and HER2-reactivated LR/LTR cells were transfected with siRNA (seq1) targeting the HER2L755S mutant. Culture medium was replaced the next day with regular medium or drug-containing medium, and replaced again at 4 days. Cell growth was assessed at six days by methylene blue assay. Relative percent (%) growth was normalized to mock transfection. Statistical analyses were performed for AZ and ATCC separately. Model–estimated group means and 95% confidence limits were plotted combining two independent experiments.

Fig. 3. The irreversible HER1/2 inhibitors overcome acquired resistance to HER2-targeted therapy conferred by the HER2L755S mutation in BT474 models in vitro

A, Responses of BT474/AZ-LR and BT474/ATCC-LTR lines and their relative parental lines (BT474/AZ-P and BT474/ATCC-P) to lapatinib (L), afatinib (Afa), and neratinib (Nrb) were measured by cell growth assay. Data was analyzed by GraphPad Prism (version 6.05) to generate drug response curves and relative IC50 values using the Log (inhibitor) vs. response-variable slope model (Bars=SEM) with normalization of data defining the biggest number in each dataset as 100% and the smallest number in the same dataset as 0%. B, BT474/AZ-P, BT474/AZ-LR, BT474/ATCC-P, and BT474/ATCC-LTR lines were treated with or without 6h of 50nM Afa or Nrb followed by Western blot.

Fig. 4. Afatinib effectively overcomes acquired resistance in the BT474/AZ-LR xenografts

A, Mice prepped with estrogen (E2) pellets were injected with 5×10⁶ BT474/AZ-LR cells and treated with L until randomization to 5 groups: E2+L, E2+Afa, ED+Veh, ED+L, and ED+Afa (for details see Materials and Methods). B, Kaplan-Meier analyses of progression-free survival within 16 days of treatment of E2+Afa or E2+L. Tumor progression was defined as tumor size tripling since randomization. TTP: time to tumor progression. C, Kaplan-Meier analyses of progression-free survival and tumor regression (graph showing change of proportion of non-regressing tumors) within 85 days of ED+Veh, ED+L, or ED+Afa treatment. Tumor regression was defined as tumor size halving since the day of randomization. TTR: time to tumor regression. D, p-HER2 (Y1221), p-AKT (S473), and p-p44/42 MAPK (T202/Y204) levels of xenografts in each treatment arm was assessed by IHC and scored as H-score by a pathologist. Two tumors (Suppl. Fig. S8B and S8C) that were resistant to the ED+Afa regimen were not included in the analysis. *: p<0.05, **: p<0.01, ***: p<0.001.