Introduction: Inhibition of phosphatidylinositol-3-kinase (PI3K) induces apoptosis when combined with estrogen deprivation in estrogen receptor (ER)-positive breast cancer. The aims of the present study were to identify effective PI3K pathway inhibitor and endocrine therapy combinations, to evaluate the effect of PI3K pathway mutations and estrogen dependency on tumor response, and to determine the relevance of PIK3CA mutation in recurrent disease.Methods: The PI3K catalytic subunit inhibitor BKM120, the mammalian target of rapamycin (mTOR) inhibitor RAD001 and the dual PI3K/mTOR inhibitor BGT226 were tested against ER-positive breast cancer cell lines before and after long-term estrogen deprivation (LTED). The impact of estradiol deprivation and the ER downregulator fulvestrant on PI3K pathway inhibitor-induced apoptosis was assessed. PIK3CA hotspot mutation analysis was performed in 51 recurrent or metastatic breast cancers and correlated with ER status and survival.Results: Drug-induced apoptosis was most marked in short-term estrogen-deprived cells with PIK3CA mutation and phosphatase and tensin homolog loss. Apoptosis was most highly induced by BGT226, followed by BKM120, and then RAD001. Estradiol antagonized PI3K inhibitor-induced apoptosis following short-term estrogen deprivation, emphasizing a role for estrogen-deprivation therapy in promoting PI3K inhibitor activity in the first-line setting. ER-positive MCF7 LTED cells exhibited relative resistance to PI3K pathway inhibition that was reversed by fulvestrant. In contrast, T47D LTED cells exhibited ER loss and ER-independent PI3K agent sensitivity. PIK3CA mutation was prevalent in relapsed ER-positive disease (48%) and was associated with persistent ER positivity and a late relapse pattern.Conclusions: Estrogen deprivation increased the apoptotic effects of PI3K and dual PI3K/mTOR inhibitors in ER-positive disease, providing a rationale for PI3K/aromatase inhibitor combinations as first-line therapy. In LTED cells, differential effects on ER expression may be a relevant consideration. When ER was persistently expressed, fulvestrant strongly promoted PI3K drug activity. When ER was lost, PI3K inhibitor monotherapy was sufficient to induce high-level apoptosis. Although tumors with PIK3CA mutation had a late recurrence pattern, these mutations were common in metastatic disease and were most often associated with persistent ER expression. Targeting PIK3CA mutant tumors with a PI3K pathway inhibitor and fulvestrant is therefore a feasible strategy for aromatase-inhibitor-resistant ER-positive relapsed breast cancer.