Abstract 5010: Prediction of therapy resistance by targeted massive-parallel sequencing in primary HER2-positive breast cancer

Abstract

Abstract Background: Massive parallel sequencing is a promising tool to investigate key molecular events in cancer. Genomic alterations, such as PIK3CA mutations, are important for response to therapy in HER2+ breast cancer (BC). We have investigated genomic alterations in 364 pretherapeutic core biopsies from two prospective clinical trials with or without anti-HER2 therapy. Methods: A total of 417 formalin-fixed paraffin embedded samples from HER2+ tumors of the neoadjuvant GeparTrio (G3 without anti-HER2 treatment) and GeparSepto (G7 with dual blockade and randomization for paclitaxel vs nab-paclitaxel) study were analyzed by deep targeted massive parallel sequencing, which was successful in 364 tumors (87%). We interrogated hot spot regions of 22 genes (including TP53, PIK3CA, CDH1, FBXW7, PTEN, AKT1, ATM, BRAF, EGFR, ESR1, FGFR2, HRAS, KRAS, NRAS, SF3B1) with a minimum coverage of 500. Only non-synonymous mutations with allele frequencies ≥10% were taken into consideration. Results: A total of 291 non-synonymously mutated genes were detected in the 364 tumor samples, the most commonly mutated genes were TP53 (47%), and PIK3CA (24%), respectively. EGFR, KRAS, NRAS, HRAS were combined to the XRAS group with 9 (2.5%) mutations. 151 tumors had no, 148 one, 57 two, and 8 tumors three or more mutated genes. There were no statistical significant differences between HR+ and HR- HER2+ tumors; overall 141/252 HR+ and 72/112 HR- tumors had a mutation. HR+ tumors had TP53 mutations in 112/252 (44.4%); PIK3CA mutations in 53/252 (21.0%) and XRAS mutations in 5/252 (2.0%). HR- tumors had 60/112 (53.6%) TP53 mutations, 34/112 (30.4%) PIK3CA mutations and 4/112 (3.6%) xRAS mutations. Response rates (pCR, ypT0 ypN0) were evaluated separately for the two trials. In G7 the pCR rate was 64.7% in the group with a TP53 mutation vs 60.6% in the group without (p = 0.545). The pCR rate was significantly lower in the PIK3CA mutant vs wild type (wt) group (47.7% vs 66.7%; p = 0.009). This effect was seen in the HR+ (43.9% vs 61.3%; p = 0.052) and the HR- cohort (54.2% vs 80.0%; p = 0.029). In the nab-paclitaxel cohort, pCR rates were significantly lower in patients with PIK3CA mutations compared to those without (38.7% vs. 72.0%; p = 0.001), whereas in the paclitaxel group, there was no significant difference between patients with and without PIK3CA mutations (55.9% vs. 60.9%; p = 0.690; interaction p = 0.039). Neither TP53 nor xRAS mutations showed a significant effect on response and treatment effect. In G3 the pCR rate was 16.3% in the PIK3CA wt cohort compared to 23.7% in the mutant cohort; p = 0.339. Conclusion: Targeted NGS on FFPE core biopsies reliably identified the most common genomic alterations in HER+ BC. PIK3CA mutation in HER2+ BC predicts resistance to anti-HER2 therapy. PIK3CA mutations were found to be predictive for response to nab-paclitaxel in G7. The results show that mutational alterations are relevant for response in HER2+ BC. Citation Format: Sibylle Loibl, Jan Budczies, Wilko Weichert, Jenny Furlanetto, Albrecht Stenzinger, Nicole Pfarr, Gunter von Minckwitz, Christian Jackisch, Andreas Schneeweiss, Peter A. Fasching, Sabine Schmatloch, Bahriye Aktas, Valentina Nekljudova, Karsten Weber, Michael Untch, Carsten Denkert carsten.denkert@charite.de. Prediction of therapy resistance by targeted massive-parallel sequencing in primary HER2-positive breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5010.