Impact of somatic PI3K pathway and ERBB family mutations on pathological complete response (pCR) in HER2-positive breast cancer patients who received neoadjuvant HER2-targeted therapies

Background: The Cancer Genome Atlas analysis revealed that somatic EGFR, receptor tyrosine-protein kinase erbB-2 (ERBB2), Erb-B2 receptor tyrosine kinase 3 (ERBB3) and Erb-B2 receptor tyrosine kinase 4 (ERBB4) gene mutations (ERBB family mutations) occur alone or co-occur with somatic mutations in the gene encoding the phosphatidylinositol 3-kinase (PI3K) catalytic subunit (PIK3CA) in 19% of human epidermal growth factor receptor 2 (HER2)-positive breast cancers. Because ERBB family mutations can activate the PI3K/AKT pathway and likely have similar canonical signalling effects to PI3K pathway mutations, we investigated their combined impact on response to neoadjuvant HER2-targeted therapies.

Methods: Baseline tumour biopsies were available from 74 patients with HER2-positive breast cancer who were enrolled in the phase II TCHL neoadjuvant study (ICORG 10-05) assessing TCH (docetaxel, carboplatin, trastuzumab) (n = 38) versus TCL (docetaxel, carboplatin, lapatinib) (n = 10) versus TCHL (docetaxel, carboplatin, trastuzumab, lapatinib) (n = 40), each for six cycles. Activating mutations in PIK3CA and ERBB family genes were identified using mass spectrometry-based genotyping. Phosphatase and tensin homolog (PTEN) expression was assessed by immunohistochemistry.

Results: PIK3CA and/or ERBB family mutations were detected in 23 (31.1%) tumour samples tested, whereas PTEN expression was low in 31.1% of cases tested. Mutation frequency was similar in each treatment arm (31.3% in TCH arm, 30% in TCL arm and 31.3% in TCHL arm) and was not influenced by oestrogen receptor (ER) status (27.6% in ER-negative patients, 33.3% in ER-positive patients) or progesterone receptor (PR) status (32.6% in PR-negative patients, 29% in PR-positive patients). There was no significant difference in pathological complete response (pCR) rates between 47 patients with wild-type (WT) tumours and 22 patients whose tumours carried mutations (in either PIK3CA or ERBB family genes) (42.5% vs. 54.5%; p = 0.439). Similarly, there was no significant difference in pCR rates between patients with PIK3CA/ERBB family mutated/PTEN-low (i.e., PI3K-activated) tumours and patients without PI3K activation (50% vs. 44%; p = 0.769). However, in the TCHL (but not the TCH) group, the pCR rate was higher for 9 patients with PIK3CA/ERBB family mutated tumours than for 20 patients with PIK3CA/ERBB family WT tumours (77.8% vs. 35%; p = 0.05).

Conclusions: Our results indicate that patients who receive neoadjuvant TCHL and have PIK3CA/ERBB family mutated tumours may be more likely to have a pCR than patients with WT tumours.

Trial registration: ClinicalTrials.gov, NCT01485926 . Registered on 2 December 2011.