Sapanisertib plus Fulvestrant in Postmenopausal Women with Estrogen Receptor–Positive/HER2-Negative Advanced Breast Cancer after Progression on Aromatase Inhibitor

Endocrine therapy has been the cornerstone of therapeutic intervention for estrogen receptor–positive (ER⁺) breast cancers for several decades, but many patients with metastatic disease have primary resistance to endocrine therapy, and many more who initially respond will ultimately develop secondary resistance (1, 2).

Historically, the basis of treatment for ER⁺/HER2-negative (HER2⁻) advanced breast cancer in postmenopausal women has been sequential endocrine therapy with various agents targeting the estrogen/ER pathway via different mechanisms, including selective ER modulators, aromatase inhibitors, and selective ER downregulators (1). In recent years, treatments targeting other molecular pathways have been explored in a bid to prevent resistance, restore sensitivity to endocrine therapy, or otherwise arrest growth. Targeting the cell cycle survival pathway has proven successful: phase III trials of the cyclin-dependent kinase-4/6 (CDK4/6) inhibitors palbociclib, ribociclib, and abemaciclib demonstrated significant improvements in progression-free survival (PFS), and even overall survival (OS) in some circumstances, when these agents were added to either nonsteroidal aromatase inhibitors, tamoxifen, or fulvestrant in the first- and second-line settings (3–9). The approval of the CDK4/6 inhibitors in combination with aromatase inhibitors or fulvestrant has changed the treatment paradigm (10), but development of resistance to CDK4/6 inhibitors remains an issue and therapies beyond these agents to overcome endocrine resistance are still needed (11).

Cross-talk between ER signaling and the PI3K/AKT/mTOR pathway is strongly implicated in endocrine resistance (12), as well as in resistance to CDK4/6 inhibitors in preclinical models (11). This pathway provides an escape mechanism that allows tumor cells to proliferate under conditions of abrogated ER signaling and/or CDK4/6 inhibition (12, 13). The pathway is also often hyperactivated in breast cancer tumor cells through genetic alterations in various components of the pathway that are present in approximately 30% to 50% of cases (14, 15).

The principle of targeting the PI3K/AKT/mTOR pathway to restore endocrine sensitivity is supported by promising results in clinical trials in the second-line setting of agents targeting different components of the pathway in patients with ER⁺/HER2⁻ breast cancer with resistance to prior endocrine therapy. In the phase III SOLAR-1 trial, the selective PI3Kα inhibitor alpelisib significantly improved PFS when added to fulvestrant compared with fulvestrant plus placebo [median, 11.0 vs. 5.7 months, respectively; HR, 0.65; 95% confidence interval (CI), 0.50–0.85; P < 0.001], following prior endocrine therapy in women with ER⁺/HER2⁻ breast cancer harboring PIK3CA mutations (16). In the phase III BOLERO-2 trial, addition of the mTOR complex 1 (mTORC1) inhibitor everolimus to exemestane achieved clinically meaningful improved outcomes compared with exemestane plus placebo in patients with ER⁺/HER2⁻ breast cancer that was refractory to nonsteroidal aromatase inhibitors (median PFS, 11.0 vs. 4.1 months, respectively; HR, 0.38; 95% CI, 0.31–0.48; P < 0.0001; refs. 17, 18). Benefits of adding everolimus to endocrine therapy in patients with aromatase inhibitor–resistant metastatic breast cancer have also been reported in phase II studies in which everolimus was combined with tamoxifen (19) or fulvestrant (20). However, a possible limitation of targeting mTORC1 is that this inhibits negative feedback that would usually suppress activation of AKT, which can paradoxically increase activity of the PI3K/AKT/mTOR pathway (12, 21, 22). Preclinical studies have shown that inhibition of mTOR via both mTORC1 and mTORC2 protein complexes leads to greater suppression of cancer cell proliferation and migration, compared with selective mTORC1 inhibition, in various malignancies (23–25).

Sapanisertib is an investigational, oral, potent, and highly selective adenosine triphosphate–competitive inhibitor of mTOR kinase with dual specificity against mTORC1 and mTORC2 (26, 27). Efficacy of sapanisertib in combination with exemestane or fulvestrant was explored in a phase Ib/II study (28), in which the combinations were well tolerated and exhibited clinical benefit in postmenopausal women with ER⁺/HER2⁻ advanced or metastatic breast cancer, including groups who had shown both resistance and sensitivity to previous everolimus treatment. Based on these data, we hypothesized that dual mTORC1/2 inhibition with sapanisertib may restore sensitivity to endocrine therapies in patients with ER⁺/HER2⁻ metastatic breast cancer. We now report the findings of a phase II, open-label, randomized, three-arm study in which daily or weekly sapanisertib was added to fulvestrant to evaluate whether the combinations improved outcomes, compared with fulvestrant alone, in postmenopausal women with ER⁺/HER2⁻ advanced or metastatic breast cancer that had progressed during or after aromatase inhibitor therapy.

Eligible patients were postmenopausal women with histologic confirmation of ER⁺ (≥1% positive stained cells)/HER2⁻ metastatic or advanced breast cancer that was not amenable to resection or radiotherapy with curative intent, and who had measurable disease or bone lesions. All enrolled patients had experienced disease progression on prior aromatase inhibitor therapy, defined as recurrence during or within 12 months after discontinuation of adjuvant therapy or progression during or within 1 month after the end of therapy in the metastatic setting. Patients were excluded if they had received prior treatment with mTOR inhibitors, PI3K inhibitors, dual PI3K-mTOR inhibitors, AKT inhibitors, or fulvestrant. Prior treatment with CDK4/6 inhibitors was permitted; enrollment was managed to reach a target of approximately 40% of patients with previous exposure to CDK4/6 inhibitors, with enrollment of CDK4/6 inhibitor-naïve patients halted if it exceeded approximately 60% of the study population. Other exclusion criteria included progressive disease (PD) on more than two endocrine therapies or more than one chemotherapy for advanced or metastatic disease, an Eastern Cooperative Oncology Group performance status ≥2, or life-threatening metastatic visceral disease (extensive hepatic involvement or symptomatic pulmonary lymphangitic spread).

All patients provided written informed consent. The study was conducted according to the protocol, the ethical principles that have their origin in the Declaration of Helsinki, the International Conference for Harmonisation E6 Good Clinical Practice Guideline, and all applicable local regulations. The protocol and informed consent form were approved by the relevant Institutional Review Board or Independent Ethics Committee prior to study commencement.

This was a phase II, open-label, three-arm, multicenter study (NCT02756364) conducted across 50 study sites in the United States and Spain between July 2016 and November 2019. Patients were randomized 1:1:1 via an interactive response technology (IRT) system to receive: (i) fulvestrant 500 mg intramuscularly every 28 days (with an additional loading dose on day 15 of the first 28-day treatment cycle); (ii) fulvestrant as described plus sapanisertib 4 mg orally daily; or (iii) fulvestrant as described plus sapanisertib 30 mg orally weekly, administered on days 1, 8, 15, and 22 of each treatment cycle. Patients were stratified according to the presence/absence of visceral metastases, prior sensitivity to endocrine therapy, and previous exposure to CDK4/6 inhibitors. Prior sensitivity to endocrine therapy was defined as ≥24 months of endocrine therapy before recurrence in the adjuvant setting (i.e., in patients who have not received previous endocrine therapy in the metastatic setting) or a response or stabilization for ≥24 weeks of the most recent endocrine therapy for advanced/metastatic disease. Patients were treated until disease progression, discontinuation due to unacceptable toxicity, or withdrawal of consent. Patients who were benefiting from treatment at the end of the study were permitted to continue treatment in a posttrial access program. Patients in the single-agent fulvestrant arm who experienced PD could be rerandomized to sapanisertib daily or weekly plus fulvestrant. This was offered on a compassionate basis and there were no preplanned analyses of efficacy following crossover to combination treatment in these patients.

The primary endpoint was PFS (defined as the time from the date of randomization to the date of first documentation of progression or death due to any cause, whichever occurred first). Secondary efficacy endpoints included: TTP (defined as the time from the date of randomization to the date of first documentation of progression); OS (defined as the time from the date of randomization to the date of death); overall response rate [ORR; the proportion of patients with a complete response (CR) or partial response (PR)]; clinical benefit rate [CBR; the proportion of patients with a CR, PR, or stable disease (SD)]; and CBR ≥6 months (CBR with SD lasting ≥6 months). The proportion of patients with treatment-emergent adverse events (TEAE) was assessed as a safety endpoint. Other endpoints included changes from baseline to end of study in scores on the European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire Core 30 (QLQ-C30) and the 23-item, breast cancer–specific EORTC QLQ-BR23. Exploratory endpoints included genetic markers or biomarkers in circulating tumor material associated with response or treatment resistance.

Response was assessed by investigators according to Response Evaluation Criteria In Solid Tumors (RECIST) version 1.1 based on radiologic tumor evaluations using CT or MRI scans of the chest, abdomen, and pelvis obtained at screening, every other treatment cycle from cycle 2, and at an end-of-treatment (EOT) visit (within 30–40 days of last dose of randomized study drug). Patients entering PFS follow-up after discontinuing treatment for reasons other than PD continued to have scans every 2 to 3 months until PD or starting another anticancer treatment. They were subsequently followed-up every 6 months to monitor OS, as was the case for patients who progressed on study treatment. Safety was evaluated using the NCI Common Terminology Criteria for Adverse Events (CTCAE) version 4.03. Quality of life (QoL) questionnaires (EORTC QLQ-C30 and EORTC QLQ-BR23) were completed before administration of treatment at the start of each 28-day cycle, and at the EOT visit.

Assuming an increase in median PFS from 4 months with fulvestrant alone to 8 months with the fulvestrant/sapanisertib combinations (with a target HR of 0.5), a total of 72 PFS events were required to achieve approximately 90% power for each pairwise comparison based on a two-sided log-rank test at a significance level of 10%. With an assumed dropout rate of 10%, it was estimated that 51 patients per treatment arm were required (for a total study population of approximately 153 patients).

The full analysis set (FAS) comprised all patients randomized and the safety analysis set comprised all patients who received at least one dose of study drug. PFS, OS, and TTP were evaluated in the FAS using Kaplan–Meier analysis. For pairwise comparisons of fulvestrant plus sapanisertib daily or weekly versus fulvestrant alone, HRs and 95% CIs were estimated from a stratified Cox regression model with treatment arm and stratification factors as covariates; two-sided P values were generated from a stratified log-rank test and assessed at a 10% significance level. Tumor response (ORR, CBR, and CBR ≥6 months) was analyzed in the safety analysis set and the response-evaluable set (defined as all patients who received at least one dose of study drug, had measurable disease at baseline, and had one postbaseline disease assessment), and compared between treatment arms using a stratified Cochran—Mantel–Haenszel test. For patient-reported outcomes, change from baseline was analyzed using a mixed effects model, including treatment group, visit, treatment-by-visit interaction, and stratification factors as fixed effects, baseline score as a covariate, and intercept as a random effect with autoregressive covariance structure.

Genomic analysis of circulating tumor DNA (ctDNA) was performed for patients who were randomized to fulvestrant or fulvestrant plus sapanisertib 4 mg daily; data for the fulvestrant plus sapanisertib 4 mg daily group were pooled with data from patients who received fulvestrant plus sapanisertib 4 mg daily in a previous phase Ib/II study (28). Plasma samples collected at baseline were analyzed with a custom PlasmaSelect-R next-generation sequencing breast cancer gene panel (Personal Genome Diagnostics, Inc.) designed to evaluate 46 genes and 7 amplifications, as described previously (28). Associations between genetic biomarkers and PFS/OS were analyzed using Cox regression.

The datasets, including the redacted study protocol, redacted statistical analysis plan, and individual participants' data supporting the results reported in this article, will be made available within 3 months from initial request to researchers who provide a methodologically sound proposal. The data will be provided after its deidentification, in compliance with applicable privacy laws, data protection, and requirements for consent and anonymization.

A total of 141 patients were enrolled and randomized to receive single-agent fulvestrant (n = 46), fulvestrant plus sapanisertib daily (n = 47), or fulvestrant plus sapanisertib weekly (n = 48; Fig. 1). One patient in the fulvestrant plus sapanisertib weekly arm was randomized but withdrew before receiving treatment due to symptomatic deterioration; therefore, the safety population comprised 140 patients. The most common reason for discontinuation of treatment in all arms was disease progression, which occurred in 76.1% of patients in the single-agent fulvestrant arm, and 59.6% and 53.2% of patients in the daily and weekly combination treatment arms, respectively. Eighteen patients whose disease progressed on fulvestrant were rerandomized to fulvestrant plus sapanisertib daily or fulvestrant plus sapanisertib weekly (9 to each treatment).

Patient baseline demographics and disease characteristics were generally well balanced among study arms (Table 1). Across all treatment arms, median age was 58.0 years (range, 33–84); 84.4% of patients had demonstrated sensitivity to prior endocrine therapy and 34.0% had received prior treatment with CDK4/6 inhibitors. Two thirds (64.5%) of patients had visceral metastases.

Median (95% CI) PFS was 3.5 (1.9–5.6) months in the single-agent fulvestrant arm, compared with 7.2 (3.9–10.6) and 5.6 (4.1–9.0) months in fulvestrant plus sapanisertib daily and fulvestrant plus sapanisertib weekly arms, respectively (Fig. 2). While there was a trend for prolonged PFS with combination treatment versus fulvestrant, it was not statistically significant for either fulvestrant plus sapanisertib daily (HR, 0.77; 95% CI, 0.47–1.26; P = 0.537) or fulvestrant plus sapanisertib weekly (HR, 0.88; 95% CI, 0.53–1.45; P = 0.849). Figure 3 shows forest plots of PFS in patient subgroups stratified according to the presence/absence of visceral metastases, sensitivity to prior endocrine therapy, prior treatment with CDK4/6 inhibitors, and demographic variables. The subgroup of patients with a history of CDK4/6 inhibitor treatment (n = 16 in each treatment arm) had the greatest improvements in PFS with fulvestrant plus sapanisertib daily versus single-agent fulvestrant (HR, 0.34; 95% CI, 0.14–0.82; P = 0.042) and fulvestrant plus sapanisertib weekly versus fulvestrant (HR, 0.48; 95% CI, 0.21–1.12; P = 0.116).

The majority of PFS events were progression events (1, 1, and 2 patients had events of death in the time-to-event analysis of PFS in the single-agent fulvestrant, fulvestrant plus sapanisertib daily, and fulvestrant plus sapanisertib weekly arms, respectively). TTP results were, therefore, similar to PFS, with median TTP (95% CI) values of 3.5 (1.9–5.6), 7.2 (5.5–10.6), and 5.6 (4.1–9.0) months, respectively.

OS data were immature at the time of database lock, with 43.5%, 31.9%, and 37.5% of patients having died in the single-agent fulvestrant, fulvestrant plus sapanisertib daily, and fulvestrant plus sapanisertib weekly arms, respectively. Median OS was 30.5 months in the fulvestrant arm, not evaluable in the fulvestrant plus sapanisertib daily arm (HR vs. fulvestrant, 0.71; 95% CI, 0.36–1.40; P = 0.276), and 34.2 months in the fulvestrant plus sapanisertib weekly arm (HR vs. fulvestrant, 0.89; 95% CI, 0.47–1.68; P = 0.470).

Tumor responses are shown in Table 2 and Supplementary Fig. S1. The ORR in the safety population was 10.9%, 21.3%, and 12.8% in single-agent fulvestrant, fulvestrant plus sapanisertib daily, and fulvestrant plus sapanisertib weekly arms, respectively. The CBR was 60.9%, 74.5%, and 66.0%, and CBR ≥ 6 months was 32.6%, 48.9%, and 25.5%, respectively. Numerically higher ORR and CBRs in the fulvestrant plus sapanisertib daily arm were not statistically significant compared with fulvestrant alone.

Patients received a median of four cycles of single-agent fulvestrant (range, 1–40; mean, 9), five cycles of fulvestrant plus sapanisertib daily (range, 1–33; mean, 8), and four cycles of fulvestrant plus sapanisertib weekly (range, 1–39; mean, 6), resulting in median (mean) treatment exposure lasting 16 (36), 20 (33), and 17 (24) weeks, respectively. Dose modifications, including dose delays, omissions, and reductions, were recorded for 79% and 81% of patients receiving sapanisertib 4 mg daily and 30 mg weekly, respectively. Mean relative dose intensities were 73.9% and 75.0%, respectively, of planned cumulative sapanisertib dosages, while fulvestrant was administered with mean relative dose intensities of 96.9% to 99.6% across single-agent and combination treatment arms. Overall rates of TEAEs are shown in Table 3, along with a summary of the most common treatment-related TEAEs (reported by ≥20% of patients in any treatment arm). The most common treatment-related TEAEs occurring more frequently in the combination treatment arms than in the single-agent fulvestrant arm were gastrointestinal TEAEs, including nausea (4.3%, 48.9%, and 83.0% in single-agent fulvestrant, fulvestrant plus sapanisertib daily, and fulvestrant plus sapanisertib weekly arms, respectively), vomiting (4.3%, 31.9%, and 66.0%, respectively), diarrhea (0, 38.3%, and 21.3%, respectively), and stomatitis (2.2%, 34.0%, and 31.9%, respectively). Grade ≥ 3 treatment-related TEAEs occurred in 68.1% and 53.2% of patients in the fulvestrant plus sapanisertib daily and fulvestrant plus sapanisertib weekly arms, respectively, while none of the grade ≥ 3 TEAEs occurring in the fulvestrant-only arm were judged to be treatment-related. The most common grade ≥ 3 treatment-related TEAEs occurring in the sapanisertib arms were vomiting, diarrhea, asthenia, and rash (Table 3).

The rate of treatment discontinuation due to TEAEs was higher in the combination treatment arms (31.9% and 36.2%) than in the single-agent fulvestrant arm (4.3%). Three patients died on study (after the first dose and within 30 days of the last dose of study drug): 2 patients in the single-agent fulvestrant arm and 1 in the fulvestrant plus sapanisertib daily arm. All on-study deaths were attributed to underlying disease. No patients who crossed over from single-agent fulvestrant to combination treatment died during the combination treatment phase.

There was a high level of compliance with QoL assessments (95%–98%), although the numbers of patients with QoL data declined with successive treatment cycles as patients discontinued treatment. Mean EORTC QLQ-C30 summary scores showed no significant change from baseline in any study arm (Supplementary Fig. S2A). Mean change from baseline in scores on the global health status/QoL subscale showed a trend for deterioration with successive fulvestrant plus sapanisertib daily treatment cycles, while scores remained close to baseline in the single-agent fulvestrant and fulvestrant plus sapanisertib weekly arms (Supplementary Fig. S2B).

There were no apparent differences between treatment arms in scores on most functional subscales, although social functioning scores declined in the combination treatment arms relative to single-agent fulvestrant (Supplementary Fig. S2C). The impact of side effects was revealed by increased (worsened) scores on the nausea and vomiting, appetite loss, and diarrhea EORTC QLQ-C30 symptom subscales in the combination treatment arms relative to single-agent fulvestrant (Supplementary Fig. S2D–S2F). Patients in the combination treatment arms also had higher (worse) mean scores on the EORTC QLQ-BR23 systemic side effects subscale (Supplementary Fig. S2G).

Genomic analysis of ctDNA incorporated data from 22 patients in the single-agent fulvestrant arm who did not cross over to combination treatment, and 31 patients in the fulvestrant plus sapanisertib daily arm, who were pooled with 15 patients who received fulvestrant plus sapanisertib 4 mg daily in a previous study (28). Cox regression analyses showed that baseline TP53 mutations were associated with worse PFS and OS outcomes compared with wild-type (WT) tumors (Supplementary Fig. S3). Additionally, the impact of TP53 mutations on PFS was greater in patients treated with single-agent fulvestrant compared with sapanisertib 4 mg daily (Supplementary Fig. S3A). Similar associations were seen for AKT1 mutations and PFS (Supplementary Fig. S3A), and MTOR mutations and OS (Supplementary Fig. S3B).

This phase II study investigated the efficacy and safety of the combination of fulvestrant plus sapanisertib, a strategy to simultaneously inhibit the estrogen/ER and PI3K/AKT/mTOR pathways, in patients with ER⁺/HER2⁻ advanced breast cancer who had received prior endocrine treatment. The patient population comprised postmenopausal women with a history of resistance to aromatase inhibitor treatment; the majority had acquired resistance to endocrine therapy and approximately one third of patients had previously been treated with CDK4/6 inhibitors.

PFS was numerically longer with both sapanisertib plus fulvestrant combination regimens than with fulvestrant alone; however, the treatment differences did not reach statistical significance, so the primary study endpoint was not met. The clear trend for prolonged PFS, coupled with numerical increases in ORR and CBR in the combination arms, suggest a modest clinical benefit of adding sapanisertib to fulvestrant. Limited interpretation can be made based on the OS results, as the data were immature at the time of data cutoff.

Overall, the results of this study were consistent with the findings of a phase II study with another dual mTORC1/2 inhibitor vistusertib, also administered in combination with fulvestrant in a similar patient population (the MANTA trial; ref. 29). That study failed to demonstrate a significant benefit of adding daily or intermittent vistusertib to fulvestrant compared with fulvestrant alone, with only a modest numerical improvement in PFS (median, 7.6–8.0 months vs. 5.4 months, respectively) that was significantly lower than that seen with fulvestrant plus everolimus (median, 12.3 months).

The safety profile of sapanisertib was consistent with the mTORC1/2 inhibitor mechanism of action and no new safety signals were detected (28, 30, 31). However, the sapanisertib plus fulvestrant combinations had increased toxicity, including a higher incidence of gastrointestinal and dermatologic side effects, leading to more treatment discontinuations due to TEAEs compared with single-agent fulvestrant. A similar side-effect profile was seen with vistusertib in combination with fulvestrant in the MANTA study (29). That study also investigated daily and intermittent dosing regimens, and found that intermittent dosing reduced the incidence of rash and stomatitis, but that this came at the cost of higher rates of nausea and vomiting (29). That finding is consistent with observations in the present study, in which the safety data did not clearly indicate whether sapanisertib daily or weekly dosing was optimal. With each regimen, efficacy may have been tempered by suboptimal dosing (mean relative dose intensities of approximately 75% with each sapanisertib regimen).

Patient-reported outcomes suggested a negative impact of sapanisertib daily on QoL, based on the deterioration in mean global health status/QoL score in this arm that was not seen with fulvestrant alone or sapanisertib weekly. This observation was surprising given the high rates of nausea and vomiting in the sapanisertib weekly group, and we speculate that the difference between the two combination treatment regimens, in terms of QoL impact, may relate to the dropout of patients who experienced the worst side effects in the weekly dosing arm (who had worse mean scores on several symptom subscales including diarrhea/vomiting, and a faster attrition rate over successive treatment cycles; Supplementary Fig. S2) leaving a subgroup of patients who tolerated treatment relatively well. Patients in the sapanisertib daily arm remained on treatment for a median of 20 weeks (mean, 33 weeks) compared with 17 (24) weeks in the weekly arm and 16 (36) weeks in the single-agent fulvestrant arm.

CDK4/6 inhibitors have advanced the treatment of ER⁺ breast cancer in recent years (10), but long-term success continues to be limited by acquired resistance (11). Optimal sequencing of endocrine therapy, CDK4/6 inhibitors, and other targeted therapies is yet to be determined (10). In this study, patients with a history of prior CDK4/6 inhibitor treatment appeared to derive greater PFS benefit from the addition of sapanisertib to fulvestrant than CDK4/6 inhibitor-naïve patients. Although this observation was based on a small subgroup of patients (n = 16 in each treatment arm), the findings suggest that further investigation of the interactions between pathways is warranted. Preclinical studies suggest that targeting the PI3K/AKT/mTOR pathway can overcome resistance to CDK4/6 inhibitors (32) and that triple therapy targeting ER, CDK4/6, and PI3K/AKT/mTOR could prevent the onset of resistance to endocrine plus CDK4/6 inhibitor therapy; although the additional toxicity seen with dual combination therapy compared with single-agent treatment indicates that optimizing tolerability of triple therapy would be a challenge. A phase I/II study of ribociclib, everolimus, and exemestane in postmenopausal women with ER⁺/HER2– advanced breast cancer who had progressed on prior CDK4/6 inhibitor therapy demonstrated clinical benefit and tolerability (33), and a phase II/III study of inavolisib, palbociclib, and fulvestrant versus placebo, palbociclib, and fulvestrant in patients with PIK3CA-mutated, ER⁺/HER2⁻ locally advanced or metastatic breast cancer is currently ongoing (NCT04191499).

Sapanisertib targets the PI3K/AKT/mTOR pathway, and genomic analysis in this study showed that patients with mutations in components of this pathway, including AKT1 and MTOR mutations, exhibited increased sensitivity to sapanisertib treatment. This was evidenced by a more pronounced treatment effect of fulvestrant plus sapanisertib versus single-agent fulvestrant on PFS in patients with AKT1-mutated tumors than in those with WT tumors, and a similar effect on OS in patients with MTOR mutations. This result is consistent with the findings from a previous study of sapanisertib in combination with fulvestrant or exemestane, in which a positive association between AKT1 mutation status and best treatment response was observed (28). TP53 mutations were also associated with relative improvements in PFS and OS with fulvestrant plus sapanisertib versus fulvestrant alone, although TP53 mutations are common in many tumor types and may be a marker for underlying tumor biology rather than being directly implicated in PI3K/AKT/mTOR pathway dysfunction.

The efficacy and safety findings from this study, in which a modest clinical benefit of adding sapanisertib to fulvestrant was accompanied by significant additional toxicity, do not suggest a favorable risk–benefit profile for this combination in the overall study population. Future research should focus on the role of mTOR inhibition in relation to other strategies, such as CDK4/6 inhibition, with a view to optimizing treatment benefits, since patients with a history of CDK4/6 inhibitor treatment appeared to derive the greatest benefit from mTOR inhibition with sapanisertib. The results of the genomic analyses also support a strategy of targeted treatment based on genetic markers, to identify patients who are likely to benefit most from such treatment, especially those with mutations in the genes encoding proteins in the PI3K/AKT/mTOR pathway.

J.Á. García-Sáenz reports grants and personal fees from Seagen and Gilead; grants from Novartis, Celgene, Eli Lilly and Company, and Eisai; grants and other support from AstraZeneca; and personal fees from Daiichi Sankyo and MSD outside the submitted work. N. Martínez-Jáñez reports grants from Roche, Pfizer, Novartis, Eisai, Daiichi Sankyo, AstraZeneca, and LYLLI during the conduct of the study. Y. Jerez reports personal fees from Novartis, Roche, Pfizer, and AstraZeneca during the conduct of the study, as well as personal fees from Novartis, Roche, and Pfizer outside the submitted work. S. González-Santiago reports personal fees from Clovis, Eli Lilly and Company, Seagen, GSK, Roche, Pfizer, MSD, AstraZeneca, Novartis, and Pharmamar outside the submitted work. M. Ramos reports speaker's bureau with Novartis, Pfizer, and Roche. A. Antón reports grants and other support from Takeda Pharmaceutical Company, Ltd. during the conduct of the study, as well as personal fees from Bayer Hispania, Eli Lilly and Company, Gilead, and Pfizer outside the submitted work. E. Carrasco reports grants and nonfinancial support from Takeda Pharmaceutical Company, Ltd. during the conduct of the study, as well as grants and nonfinancial support from Roche, PUMA, Pfizer, Novartis, Seagen, AstraZeneca, and Eli Lilly and Company outside the submitted work; in addition, E. Carrasco has stocks of Eli Lilly and Company. J. Chen reports other support from Takeda Pharmaceutical Company, Ltd. during the conduct of the study, as well as other support from Takeda Pharmaceutical Company, Ltd. outside the submitted work. R. Neuwirth reports other support from Takeda Pharmaceutical Company, Ltd. during the conduct of the study. K. Galinsky reports personal fees from Takeda Pharmaceutical Company, Ltd. during the conduct of the study. S. Vincent reports other support from Takeda Pharmaceutical Company, Ltd. during the conduct of the study. E.J. Leonard is an employee of the Sponsor, Takeda Pharmaceutical Company, Ltd. (TDC America). D. Slamon reports other support from BioMarin, Pfizer, Novartis, Eli Lilly and Company, Amgen, and Seattle Genetics during the conduct of the study. No disclosures were reported by the other authors.

J.Á. García-Sáenz: Formal analysis, supervision, validation, investigation, visualization, writing–original draft, writing–review and editing. N. Martínez-Jáñez: Writing–review and editing. R. Cubedo: Resources, investigation, writing–review and editing. Y. Jerez: Conceptualization, writing–review and editing. A. Lahuerta: Investigation, writing–review and editing. S. González-Santiago: Investigation, writing–review and editing. N. Ferrer: Writing–review and editing. M. Ramos: Validation, investigation. J.L. Alonso-Romero: Validation, investigation, writing–review and editing. A. Antón: Supervision, validation, investigation, writing–original draft, writing–review and editing. E. Carrasco: Supervision, investigation, project administration, writing–review and editing. J. Chen: Formal analysis, writing–review and editing. R. Neuwirth: Formal analysis. K. Galinsky: Formal analysis, visualization. S. Vincent: Data curation, writing–review and editing. E.J. Leonard: Data curation, supervision, methodology, writing–review and editing. D. Slamon: Conceptualization, supervision, investigation, writing–review and editing, conceptualization based on the initial preclinical concept and data generated regarding triple therapy with hormonal blockade + cdk4/6 inhibition + pi3k inhibition (26).

The authors acknowledge Spanish Group for Breast Cancer Research (GEICAM) and Translational Research In Oncology (TRIO). The authors also thank the patients and their families, as well as the clinical study teams, for making this study possible. This study was funded by Millennium Pharmaceuticals, Inc., Cambridge, MA, a wholly owned subsidiary of Takeda Pharmaceutical Company, Limited. The authors acknowledge Marcel Kuttab, PharmD, of Takeda Pharmaceuticals U.S.A., Inc., for editorial assistance. Medical writing support for the development of this manuscript, under the direction of the authors, was provided by Samantha Stanbury, PhD, and Helen Wilkinson, PhD, of Ashfield MedComms, an Ashfield Health company, funded by Takeda Pharmaceuticals U.S.A., Inc., and complied with the Good Publication Practice 3 ethical guidelines (Battisti WP and colleagues; Ann Intern Med 2015;163:461–4).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.