Abstract Background Aberrations of phosphoinositide-3-kinase (PI3K) pathway are extensively found in many human cancers through several mechanisms, including mutation or amplification of PIK3CA and loss of phosphatase and tensin homolog (PTEN) and inositol polyphosphate 4-phosphatase-II(INPP4B). In breast cancer, a number of studies have suggested the putative mechanism of resistance to trastuzumab therapy in terms of PI3K pathway activation. We aimed to evaluate the predictive relevance of these biomarkers to trastuzumab efficacy in HER2-positive disease. Patients and Methods A total of 43 breast cancer patients with HER2-positive who received both neoadjuvant treatment and surgery at Kumamoto University Hospital between 2004 and 2012 were selected. The regimens of chemotherapy included anthracycline or taxane-containing drugs in combination with trastuzumab. Using pretreatment tumor tissues, PIK3CA mutations (E542K, E545K, and H1047R) were analyzed by direct dideoxynucleotide sequencing and digital PCR methods. Additionally, the expressions of PTEN, pAkt, and INPP4B were assessed by immunohistochemistry (IHC). Results The overall pathological complete response (pCR) rate was 60%. Direct sequencing detected PIK3CA mutations in 21% of all patients, whereas digital PCR detected them in 26 % when the cutoff point of the mutation was set at 1%. In some cases, it was difficult to differentiate mutant DNA from artifact by direct sequencing, but we could identify the mutation clearly using digital PCR. We found the correlation between the proportion of the PIK3CA mutation and the pCR rate; the pCR rates in the patients with PIK3CA mutations with cut-off of 1%, 10% and 20% were 55%, 29%, and 0%, respectively. There were no significant correlations of clinicopathological features with PIK3CA mutations, copy number status, PTEN, and pAkt expression. Low INPP4B expression was associated with larger tumor size (P = 0.035), and higher nuclear grade (P = 0.031) compared to high expression. We evaluated the contribution of biomarkers related to the PI3K pathway to the prediction of pCR by logistic regression models. In multivariate analysis, activation of the PI3K pathway due to either PIK3CA mutation or low PTEN expression were related to poorer response to trastuzumab (OR of predictive pCR was 0.11, P = 0.041). Similarly, high activation defined as PIK3CA mutation or low expression of PTEN or INPP4B tend to have lower pCR (OR was 0.14, P = 0.064). Conclusions 1. Digital PCR has potential to complement the direct sequencing data, leading to more accurate measurement of the mutation frequency. 2. Our findings provide additional support for the recently published studies regarding activating mutation in PIK3CA in HER2-positive breast cancer, and further suggest that integrated biomarkers of PIK3CA mutation, PTEN, INPP4B are stronger predictors of trastuzumab response than either one alone. Citation Format: Sueta A, Yamamoto Y, Takeshita T, Yamamoto-Ibusuki M, Iwase H. High activation of PI3K pathway defined by PIK3CA mutation, PTEN, and INPP4B expression are associated with trastuzumab efficacy in HER2-positive breast cancer. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P3-07-11.