Addition of Ipatasertib to Dual Anti-HER2 Maintenance Therapy in HER2-Positive Metastatic Breast Tumors with PIK3CA Mutations: The Phase Ib SOLTI-1507 IPATHER Trial

Abstract

Purpose:

The aim of this study was to evaluate the safety and feasibility of ipatasertib combined with trastuzumab and pertuzumab (HP) as maintenance therapy after first-line treatment in patients with HER2-positive metastatic breast cancer harboring PIK3CA mutations (PIK3CAmut).

Patients and Methods:

This prospective, multicenter, single-arm, phase Ib study evaluated the safety and preliminary efficacy of ipatasertib, an AKT inhibitor, combined with HP, with or without endocrine therapy as maintainance therapy, in patients with unresectable locally advanced or metastatic PIK3CAmut, HER2-positive breast cancer following first-line induction chemotherapy and HP.

Results:

Seventeen patients were enrolled, with a median follow-up of 27.7 months. During the dose-selection phase, ipatasertib at 400 mg daily (21 days on and 7 days off) with standard HP was established as the recommended phase II dosage. This decision was based on the absence of dose-limiting toxicities in the first six patients treated at this dosage during the initial 28-day cycle, which constituted the primary endpoint. Grade 3 treatment-related adverse events (TRAE) occurred in seven patients (41.2%), most commonly diarrhea and nausea. Two (11.8%) reported four serious TRAE (diarrhea, vomiting, ischemic stroke, and pneumonitis, one case each) related to ipatasertib, from which they recovered. The confirmed overall response rate was 31.1% [95% confidence interval (CI), 12.1%–58.5%], clinical benefit rate 84.6% (95% CI, 53.7%–97.3%), and median progression-free survival 16.4 months (95% CI, 9.4–NR); 47.3% of patients were progression free at 18 months.

Conclusions:

These results support ipatasertib plus HP as a safe and promising maintenance strategy for HER2-positive breast tumors harboring PIK3CAmut.

Translational Relevance.

The IPATHER study addresses a critical therapeutic gap in the management of HER2-positive metastatic breast cancer harboring PIK3CA mutations (PIK3CAmut), a subgroup associated with resistance to standard anti-HER2 therapies. By integrating ipatasertib, an oral AKT inhibitor, with maintenance trastuzumab and pertuzumab following first-line chemotherapy, this trial explores a strategy grounded in the biological rationale of targeting the PI3K/AKT pathway. The combination was well tolerated, with manageable toxicities and no dose-limiting toxicities at the recommended dose. Importantly, promising preliminary efficacy was observed, with a median progression-free survival of 16.4 months, favorable as compared with historical data in PIK3CAmut tumors. These findings suggest that AKT inhibition may help overcome intrinsic resistance mechanisms and enhance the durability of anti-HER2 treatment. This study sets the foundation for future trials evaluating PI3K/AKT pathway targeting, either alone or in combination with other therapies, and emphasizes the need to personalize treatment strategies based on tumor genomic profiles.

Introduction

HER2-positive breast cancer accounts for 20% to 30% of all breast tumors and remains a serious disease (1–3). The combination of trastuzumab and pertuzumab (HP) plus chemotherapy (typically, a taxane) is currently the preferred standard of care for first-line treatment of HER2-positive metastatic breast cancer (4–8).

One of the major determinants of resistance to anti-HER2 treatments is the hyperactivation of signaling pathways downstream of HER2, especially the PI3K/AKT/mTOR pathway (9–12), which is dysregulated in up to 50% of cases, primarily because of activating mutations in PIK3CA (PIK3CAmut) or the loss of the tumor suppressor PTEN (13–16). Data from multiple studies with anti-HER2 therapy in the neoadjuvant, adjuvant, and metastatic settings have consistently shown that patients with PIK3CAmut tumors benefit less from anti-HER2 treatments like trastuzumab and pertuzumab, compared with those with wild-type PIK3CA (PIK3CAwt) tumors (17–24). In the phase III CLEOPATRA trial, patients with PIK3CAmut, locally recurrent, unresectable, or metastatic breast cancer treated with induction taxane plus HP had a shorter progression-free survival (PFS) than those with PIK3CAwt tumors (12.5 vs. 21.8 months, respectively; ref. 20).

Ipatasertib is an ATP-competitive, small-molecule pan-AKT inhibitor with high selectivity for activated AKT (25). The clinical use of ipatasertib has been assessed in multiple cancer types with a high prevalence of PI3K/AKT/mTOR pathway activation signature, including HER2-negative/hormone receptor (HR)–positive and triple-negative breast cancer trials exploring a variety of therapeutic combinations (13). We hypothesized that combining ipatasertib with HP would be safe and could be explored as a maintenance strategy after first-line treatment with chemotherapy induction plus dual HER2 blockade in patients with PIK3CAmut HER2-positive metastatic breast cancer.

Patients and Methods

Study design

The IPATHER study was a multicenter, nonrandomized, single-arm, prospective, open-label, phase Ib dose-finding study conducted in Spain between September 2020 and June 2023. The study aimed to assess the safety and preliminary efficacy of the combination of ipatasertib with HP as maintenance therapy in patients with PIK3CA-mutant HER2-positive unresectable locally advanced or metastatic breast tumors after first-line induction therapy for metastatic disease. The representativeness of our study population compared with the general HER2-positive breast cancer population is detailed in Supplementary Table S1.

The study included five sequential phases: molecular prescreening, screening, dose-limiting toxicity (DLT) assessment (i.e., cycle 1), and dose-expansion phase (i.e., cycle 2 onward) until the end of treatment (EOT) visit (Supplementary Fig. S1). At prescreening, patients signed the molecular prescreening informed consent form to perform central determination of HER2 and presence of PIK3CAmut, which occurred before or during induction therapy with a taxane or vinorelbine plus HP. Once chemotherapy was stopped because of other reasons than progression (as per clinical judgment), screening evaluations were performed over 28 days to confirm whether patients met all eligibility criteria before allocation.

A rolling-six design was implemented for dose escalation during the DLT period, allowing inclusion of two to six patients per cohort. DLT were defined as adverse events (AE) occurring in cycle 1 (28 days) and were adjudicated by an independent safety monitoring committee. Based on observed toxicities, the ipatasertib dosage could be de-escalated from 400 to 300 or 200 mg once daily (Supplementary Fig. S2). Enrollment into the expansion cohort was contingent on determining the recommended dose for expansion (Supplementary Figs. S1 and S2).

Ethics statement

The study protocol was approved on November 22, 2019, by the Ethics Committee (EC) at the leading hospital (EudraCT number: 2019-001526-94; Vall d’Hebrón EC, Barcelona, Spain). In total, 10 Spanish centers participated in this study, which was conducted in accordance with the ethical guidelines of the 1975 Declaration of Helsinki, the International Conference on Harmonization for Good Clinical Practice, and all applicable local laws and European Directives for personal data protection. All patients provided written informed consent to participate in the study. This article has been written following the CONSORT-DEFINE 2023 guidelines.

The present study was registered at clinicaltrials.gov with the code NCT04253561.

Participants and study settings

Eligible patients had PIK3CAmut, HER2-positive, unresectable, locally advanced or metastatic adenocarcinoma of the breast and were allocated for maintenance HP after at least four cycles of first-line treatment for metastatic disease with a taxane or vinorelbine plus anti-HER2 regimen (HP). Other inclusion criteria were (1) female (pre- or postmenopausal) or male patients aged 18 years or older, (2) Eastern Cooperative Oncology Group performance status of 0 or 1, (3) prior induction chemotherapy discontinued for a reason other than progressive disease (PD), (4) in patients with unresectable locally advanced disease, recurrent disease or PD with no available standard curative options at the moment of induction therapy with chemotherapy plus HP, (5) after induction therapy, the absence of PD documented in the screening CT scan, (6) in patients with bilateral breast cancer, HER2 positivity in both breasts or in a metastatic biopsy, (7) independently of prior neo/adjuvant therapy, ≥6-month disease-free interval between completion of anti-HER2 therapy and metastatic diagnosis, (8) last dose of taxane or vinorelbine plus HP administered <9 weeks before cycle 1 day 1 (C1D1), (9) no baseline diarrhea or diarrhea grade ≤1 within the last 28 days, (10) life expectancy of ≥6 months, (11) negative pregnancy test result for premenopausal female patients of reproductive capacity and for female patients <12 months after the menopause, and (12) having provided signed consent to participate in the study. As per the study protocol, patients were required to have a fasting total serum glucose ≤150 mg/dL and glycosylated hemoglobin ≤7.5% at study entry. The study recruited both males and females; however, all enrolled patients were female. Gender-related data were not collected beyond the self-reported male–female dichotomy. Key exclusion criteria are detailed in the Supplementary Data S1.

Estrogen receptor (ER) status was assessed locally by IHC, defined as ER and/or progesterone receptor expression in >1% of tumor cells. Central determination of HER2 (IHC/ISH) defined by the American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) guidelines, mutational analysis of PIK3CA [using next-generation sequencing (NGS) on the Illumina platform], and central determination of the PAM50 tumor-intrinsic subtype (nCounter assay, NanoString Technologies) were performed on the most recent formalin-fixed, paraffin-embedded tumor sample (primary breast, local recurrence, or metastatic site) or, if unavailable or of poor quality, fresh tumor biopsy was obtained upon patient consent (Supplementary Fig. S1).

The same samples could be used to confirm HR status, defined by the ASCO/CAP guidelines.

Radiologic imaging during the screening period consisted of a CT scan and/or MRI of the chest/abdomen/pelvis, a bone or PET scan, and any other imaging studies (CT neck and plain films) as clinically indicated by the treating physician and performed ≤28 days before C1D1.

The recruitment period was February 2020 to November 2023. Results are reported until the data cutoff of July 30, 2024. The trial is expected to end in December 2025. No attrition was observed after treatment allocation; all enrolled patients either completed the study treatment or discontinued as per protocol-defined criteria.

Interventions

Ipatasertib was administered orally once a day, starting with the first dose on C1D1 through day 21, followed by 7 days off treatment in each 28-day cycle. The initial planned dose of ipatasertib was 400 mg; dose reductions to 300 or 200 mg were permitted according to tolerability. No dose de-escalation below dose level (DL) 2 was explored in this study (Supplementary Fig. S2). DL1 and the range of planned ipatasertib DL were selected based on safety and pharmacokinetic data from previous clinical studies (26–28). Subcutaneous trastuzumab was administered at a fixed dosage of 600 mg every 3 weeks. Patients received 420 mg pertuzumab as an intravenous infusion every 3 weeks. The doses of trastuzumab and pertuzumab were selected based on the currently available safety data as single agents with no dose modifications permitted. No evidence of drug–drug interactions between trastuzumab and pertuzumab was observed in previous trials (29, 30). Concomitant prophylactic antidiarrheal medication (loperamide) was recommended. Endocrine therapy (ET) was started from cycle 2 day 1 (C2D1) to avoid interference with the DLT assessment in cycle 1. Dose modifications and treatment discontinuations followed standard clinical practice and protocol-specific criteria.

In patients with HR-positive tumors, standard ET (aromatase inhibitors, tamoxifen, or fulvestrant, alone or in combination with luteinizing hormone releasing hormone analogues) could be added at the investigator’s discretion from C2D1 (Supplementary Fig. S1). Patients with isolated central nervous system (CNS) progression while on study treatment could continue therapy after completing local treatment (radiotherapy or surgery) for brain or dural metastases, provided they were asymptomatic, had no prior isolated CNS progression, and had unequivocally stable systemic disease. Patients remained in the study until unacceptable toxicity (i.e., a grade 4 AE or life-threatening serious AE with a possible, probable, or certain drug causality as determined by the investigator), a delay in scheduled treatment of >28 days due to drug-related toxicity, PD or second isolated CNS progression, withdrawal of consent, any protocol violation (e.g., intake of prohibited medication or poor compliance with study procedures or treatment), study termination by the sponsor, the need for other treatments for breast cancer, or any medical condition or personal circumstance (including pregnancy) that could expose the patient to risk if continuing on the study medication. If a patient withdrew or terminated early from the study, an EOT visit was conducted, and collected data were analyzed for the purpose of the study, unless the patient had withdrawn consent.

Investigational medicinal products (IMP; ipatasertib, trastuzumab, and pertuzumab) were manufactured and provided by F. Hoffmann-La Roche Ltd. (Roche). Non-IMP treatment was supplied by the study site pharmacy. Further medication and administration details are provided in the study protocol (Supplementary Data S2).

Objectives and outcome measures

The primary endpoint was to determine the maximum tolerated dose (MTD), defined as the highest dose at which no more than one patient of the first six enrolled patients experienced a DLT during the first treatment cycle (DLT period) with ipatasertib and HP. Secondary endpoints included the overall safety and tolerability profile of the combination, objective response rate (ORR), duration of response (DoR), clinical benefit rate (CBR), and PFS.

The main objectives of the study were to assess the tolerability of combining ipatasertib with HP (with or without ET), particularly the incidence and severity of diarrhea, and to determine the recommended phase II dose (RP2D) of the combined therapy. The primary endpoint was the MTD, defined as the highest DL of ipatasertib when administered in combination with HP, at which no more than one patient experiences a DLT during the DLT assessment window (i.e., cycle 1) among the first six enrolled patients. A DLT was defined as grade ≥3 diarrhea lasting more than 72 hours, grade ≥2 diarrhea lasting more than 5 days, or other grade ≥3 nonhematologic or hematologic toxicities that occurred within the first 28 days of treatment, assessed by the investigator as probably or unequivocally related to the study medication and unrelated to disease, disease progression, intercurrent illness, or concomitant medications. A full list of the criteria for defining DLT is shown in Supplementary Table S2. After the DLT assessment period, any toxicities that occurred during the dose-expansion phase did not influence the MTD.

Secondary objectives included evaluating the safety and tolerability of the combined treatment (with or without ET). Secondary safety endpoints included the incidence and severity of DLT during cycle 1 and the overall incidence and intensity of AE, treatment-emergent AE (TEAE; i.e., new AE or those that worsened in severity after C1D1, except for PD), and treatment-related AE (TRAE; AE judged by the investigator as possibly, probably, or definitely related to treatment). Special emphasis was placed on diarrhea from C1D1 until 28 days after EOT or the start of new anticancer treatment, whichever occurred first. The NCI Common Terminology Criteria for Adverse Events version 4.03 was followed. Secondary tolerability endpoints were dose reductions, treatment interruptions, and discontinuations. Stringent dose reduction criteria were applied to de-escalate ipatasertib to the next lower DL if needed (Supplementary Fig. S2).

Additionally, the trial assessed the preliminary antitumor activity of ipatasertib plus HP (with or without ET) as maintenance therapy after first-line induction chemotherapy with a taxane or vinorelbine plus HP. Secondary efficacy endpoints included objective response rate [proportion of patients achieving complete response (CR) or partial response (PR) at 18 weeks after C1D1], DoR [time from the first documented objective response to PD or death from any cause (until 30 days after EOT), whichever occurred first], CBR [percentage of patients achieving CR, PR, stable disease (SD), or SD/non-PD (for patients with nonmeasurable disease or CR after induction therapy) for ≥24 weeks after C1D1], and PFS [time from C1D1 to PD or death from any cause (until 30 days after EOT), whichever occurred first]. ORR and DoR were assessed only in patients who had not achieved CR after induction therapy. Patients with no progression or ORR were censored at the date of their last tumor assessment within 35 days after EOT.

All measurable and evaluable tumor lesions were assessed and documented at screening and reassessed at each subsequent tumor evaluation—every 8 weeks (±1 week) for the first 12 months after enrollment and every 12 weeks (±1 week) thereafter—until disease progression, withdrawal of consent, or study termination by the sponsor, with additional scans performed as clinically indicated. Tumor response was evaluated according to version 1.1 of the RECIST guidelines. Plasma samples (30 mL each) for future analyses, including circulating tumor DNA (ctDNA), were collected at prescreening, C1D1, C2D1, PD, and EOT visits. ctDNA analyses have not yet been performed, and the mutational data presented in the current article derive from tumor tissue. A complete list of study procedures and assessments at screening, DLT, and remaining treatment periods, as well as the EOT visit, can be found in the study protocol (Supplementary Data S1).

Sample size and statistical methods

Approximately 15 evaluable patients per DL were considered sufficient to have preliminary results to assess the safety of the combination of ipatasertib with HP. Given the reported incidence of PIK3CA mutations (38%; ref. 31), a screening failure rate of 86% was anticipated, and approximately 108 patients were expected to undergo prescreening. No formal power calculation was performed; the sample size was primarily determined based on (i) feasibility within the recruitment period and available resources and (ii) the need to detect early safety signals and guide dose selection for subsequent studies.

A descriptive analysis of the variables included in the study was performed. Continuous variables were expressed as mean and standard deviation, and categorical variables were expressed as absolute values and percentages. Time-to-event variables were analyzed according to the Kaplan–Meier method. Kaplan–Meier survival curves, and the median value were reported, along with associated 95% confidence intervals (95% CI) calculated using the SE derived from the Greenwood formula. In addition, survival rates at 6, 12, and 18 months were estimated. Categorical endpoints (ORR and CBR) were reported with numbers, percentages, and the corresponding 95% CI calculated using the Clopper–Pearson method. No data imputation was performed, and as no formal comparisons were predefined for this study, no P values were reported. The reverse Kaplan–Meier method was used to estimate the median follow-up time. All statistical analyses were performed using R statistical software version 4.3.1.

Research resource identifiers

HER2 status was centrally determined using IHC and/or ISH with anti-HER2 antibody (RRID: AB_2335975, cat. #790-2991, Ventana).

Mutational analysis of PIK3CA was performed by NGS using the Illumina platform (RRID: SCR_016381).

PAM50 intrinsic subtype was centrally determined using the nCounter assay (NanoString Technologies, RRID: SCR_023912).

All statistical analyses were conducted using R version 4.3.1 (R Foundation for Statistical Computing, RRID: SCR_001905).

Results

Study population

Between February 2020 and November 2023, 116 female patients with locally advanced or metastatic breast cancer were assessed for eligibility (Fig. 1; Supplementary Fig. S3). Of these, 24 harbored HER2-positive PIK3CAmut tumors, and 17 were included in the study (Fig. 1).

Figure 1.

Patient disposition. CONSORT diagram of the IPATHER study. Reasons for not meeting eligibility or inclusion criteria are indicated. The efficacy population included patients with ≥2 complete treatment cycles and ≥1 tumor assessment. CT, chemotherapy; IC, inclusion criterion; PI, principal investigator.

The type of sample, disease characteristics, and genomic landscape of all the tumors assessed for eligibility are shown in Fig. 2. Most samples (13 of 17) were obtained from breast tumors, whereas the others were from metastatic sites, including lymph node (n = 1), brain (n = 1), and bone (n = 2). The most frequent genomic alteration identified beyond PIK3CAmut (100%) was TP53 mutation (64.7%), followed by PTEN, FSR2, FOXA1, FGFR2, and ATR mutations, each detected in 5.9% of cases (Fig. 2). Among the 17 patients included, 16 samples had a single PIK3CAmut and one had two (Fig. 2). Most alterations were in the helical domain (n = 9; seven H1047R, one H1047Y, and one H1047L) or in the kinase domain (n = 5; four E545K and one E542A). Two additional mutations were observed (M1043I and G12D). One patient had concurrent PIK3CAmut and PTEN mutations. With regard to intrinsic subtypes (Fig. 2), PAM50 classified tumors as luminal B (n = 2), luminal A (n = 1), HER2 enriched (n = 5), and normal-like (n = 1). In eight patients, the PAM50 subtype could not be determined because of insufficient tumor material for molecular testing.

Figure 2.

Translational genomic analysis of included patients (N = 17). Oncoplot depicting the genomic landscape of included patients. MBC, metastatic breast cancer; NA, not available; PR, progesterone receptor.

The safety population consisted of all 17 patients included in the study who received at least one dose of ipatasertib and HP. The efficacy population consisted of 16 enrolled patients who completed two or more treatment cycles and underwent at least one tumor assessment. At data cutoff, five patients remained on the study treatment (Fig. 1).

Baseline patient characteristics are summarized in Table 1. All participants were female and White with a median age of 54 years (47.1% premenopausal). Ten (58.8%) patients each had ER-negative tumors, visceral metastasis, and stage IV disease at diagnosis. All patients received a taxane in combination with HP as induction therapy. The median duration of taxane plus HP before study enrollment was 3.8 months (range, 2.1–5.6), and the best response to the induction therapy was PR in nine patients (52.9%) and SD in eight patients (47.1%). ET was added to ipatasertib and HP at the investigator’s discretion in two of the seven patients with HR+ tumors (Supplementary Fig. S1).

Table 1.

Baseline characteristics of study participants with PIK3CAmut, HER2-positive locally advanced or metastatic breast cancer.

Characteristic	Overall (N = 17)
Median age, years (min, max)	54.0 (40.0, 77.0)
Female, n (%)	17 (100)
White, n (%)	17 (100)
ECOG performance status, n (%)	​
0	11 (64.7)
1	6 (35.3)
Menopausal status, n (%)	​
Premenopausal	8 (47.1)
Postmenopausal	9 (52.9)
HR status, n (%)	​
HR−	10 (58.8)
HR+a	7 (41.2)
ER+	7 (41.2)
PgR+	5 (29.4)
Stage at diagnosis, n (%)	​
I–III	7 (41.2)
IV	10 (58.8)
Visceral metastasis, n (%)	​
Yes	10 (58.8)
No	7 (41.2)
Median duration of induction therapy, months (standard deviation)	3.9 (1.2)
Best response to induction therapy, n (%)	​
CR	0 (0.0)
PR	9 (52.9)
SD ≥ 24 weeks/<24 weeks	4 (23.5)/4 (23.5)
PD	0 (0.0)

Abbreviation: PgR, progesterone receptor.

^aOnly two patients received subsequent ET.

Determination of the recommended dose of ipatasertib with HP

No DLT or dose reductions were reported in the first six patients at the initial DL1 of ipatasertib (Supplementary Fig. S2). For this reason, the combination of 400-mg ipatasertib (C1D1 through day 21) with the standard dose of HP was established as the MTD and the RP2D.

Dose expansion and secondary safety objectives

Eleven additional patients were enrolled in the expansion phase at the same dose, for a total of 17 patients evaluable for safety. A total of 16 patients (94.1%) reported TRAE, which were mainly grade 1 or 2 (Table 2). TRAE occurring in >20% of patients included diarrhea (76.5%), nausea (47.1%), decreased appetite (35.3%), fatigue (29.4%), rash (29.4%), and vomiting (29.4%). There were no hyperglycemia events. Seven (41.2%) patients reported grade 3 TRAE (Table 2); diarrhea and nausea were the most frequent (11.8% each). Two (11.8%) patients reported four serious TRAE (diarrhea, vomiting, ischemic stroke, and pneumonitis, one case each) related to ipatasertib (Table 2). One patient experienced grade 3 diarrhea and vomiting, requiring intravenous hydration at the hospital, which resolved within 5 days; 2 years later, the same patient developed grade 3 pneumonitis that required hospitalization and resolved with corticosteroids and antibiotics in less than a month. The other patient had a grade 2 ischemic stroke that was considered serious because it required hospital admission; the patient recovered after 6 days. There was one grade 3 AE of special interest (rash), which was related to ipatasertib. All TEAE, independently of causality, are summarized in Supplementary Table S3.

Table 2.

TRAE and serious TRAE in the study safety population (N = 17).

By PT, n (%) of patients	All grade	Grade 1	Grade 2	Grade 3
With any TRAE	16 (94.1)	16 (94.1)	11 (64.7)	7 (41.2)
Diarrhea	13 (76.5)	12 (70.6)	6 (35.3)	2 (11.8)
Nausea	8 (47.1)	8 (47.1)	2 (11.8)	2 (11.8)
Decreased appetite	6 (35.3)	6 (35.3)	2 (11.8)	0 (0.0)
Fatigue	5 (29.4)	3 (17.6)	1 (5.9)	1 (5.9)
Rash	5 (29.4)	4 (23.5)	1 (5.9)	1 (5.9)
Vomiting	5 (29.4)	5 (29.4)	1 (5.9)	1 (5.9)
Anemia	3 (17.6)	3 (17.6)	0 (0.0)	0 (0.0)
Stomatitis	3 (17.6)	3 (17.6)	1 (5.9)	1 (5.9)
Dysgeusia	2 (11.8)	2 (11.8)	0 (0.0)	0 (0.0)
Increased transaminases	2 (11.8)	2 (11.8)	0 (0.0)	0 (0.0)
Dizziness	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Ischemic stroke	1 (5.9)	0 (0.0)	1 (5.9)	0 (0.0)
Lymphopenia	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Thrombocytopenia	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Abdominal rigidity	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Anal incontinence	1 (5.9)	1 (5.9)	1 (5.9)	0 (0.0)
Constipation	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Dyspepsia	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Gastroesophageal reflux disease	1 (5.9)	1 (5.9)	1 (5.9)	0 (0.0)
Hemorrhoids	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Mucosal dryness	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Pyrexia	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Nail dystrophy	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Dry eye	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Visual impairment	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Xerophthalmia	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Arthralgia	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Myalgia	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
ADHD	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Disorientation	1 (5.9)	0 (0.0)	1 (5.9)	0 (0.0)
Breast pain	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Pneumonitis	1 (5.9)	0 (0.0)	0 (0.0)	1 (5.9)
Increased INR	1 (5.9)	0 (0.0)	1 (5.9)	0 (0.0)
Decreased blood potassium	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Decreased blood sodium	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
Decreased ionized calcium	1 (5.9)	1 (5.9)	0 (0.0)	0 (0.0)
With any serious TRAE	2 (11.8)	0 (0.0)	2 (11.8)	1 (5.9)
Diarrhea	1 (5.9)	0 (0.0)	1 (5.9)	0 (0.0)
Vomiting	1 (5.9)	0 (0.0)	0 (0.0)	1 (5.9)
Ischemic stroke	1 (5.9)	0 (0.0)	1 (5.9)	0 (0.0)
Pneumonitis	1 (5.9)	0 (0.0)	0 (0.0)	1 (5.9)

No grade 4/5 TRAE were observed in the study population. More than one event may have been reported per patient.

Abbreviations: ADHD, attention deficit hyperactivity disorder; INR, international normalized ratio increased; PT, preferred term.

Ipatasertib-related toxicities led to six (35.3%) dose interruptions, six (35.3%) dose reductions, and two (11.8%) treatment discontinuations. Toxicities related to ipatasertib leading to discontinuation of any study treatment were grade 1 diarrhea, grade 2 ischemic stroke, and grade 3 pneumonitis, one patient each (Supplementary Table S4).

Efficacy outcomes

With a median follow-up of 27.7 months, the confirmed ORR was 31.3% (95% CI, 11.0%–58.7%), including two (12.5%) patients with CR and three (18.8%) patients with PR. Of note, these responses all occurred during the study treatment (i.e., after the induction therapy with taxane plus HP). The CBR was 87.5% (95% CI, 61.7%–98.4%; Table 3). Of the patients included in the efficacy analysis, five (29.4%) were censored: four because of ongoing treatment at data cutoff and one because of the investigator’s decision to discontinue without progression. The remaining 12 (70.6%) experienced an event: nine (52.9%) had documented disease progression, one discontinued because of pneumonitis, one withdrew consent, and another discontinued per the investigator’s decision following clinical deterioration.

Table 3.

Clinical efficacy of ipatasertib in combination with HP (N = 16).

​	Overall (N = 16)
Median follow-up period, months (95% CI)	27.7 (23.9–NR)
Confirmed ORR, % (95% CI)a	31.3 (11.0–58.7)
Best overall response, n (%)	​
CR	2 (12.5)
PR	3 (18.8)
SD, ≥24 weeks/<24 weeks	9 (56.3)/2 (12.5)
PD	0 (0.0)
CBR (CR + PR + SD ≥ 24 weeks), % (95% CI)a	87.5 (61.7–98.4)
Median DoR, months (min; max)	14.8 (11.0; 24.1)
PFS (from C1D1)	​
Median PFS, months (95% CI)	16.4 (7.6–NE)
6-month PFS, % (95% CI)	81.3 (52.5–93.5)
12-month PFS, % (95% CI)	66.2 (36.3–84.6)
18-month PFS, % (95% CI)	47.3 (18.9–71.4)

Abbreviations: NE, not estimable.

^a95% exact binomial CI; Clopper–Pearson method used.

Time to progression is shown in Fig. 3A. The median DoR and PFS were 14.8 (min, 11; max, 24.1) months and 16.4 (95% CI, 7.6–not estimable) months, respectively (Fig. 3B). The PFS rates at 12 and 18 months after starting ipatasertib and HP were 66.2% (95% CI, 36.3–84.6) and 47.3% (95% CI, 18.9–71.4), respectively (Table 3; Fig. 3B). PFS from the first line of treatment for advanced disease, including the induction treatment with chemotherapy + HP, is shown in Supplementary Fig. S4.

Figure 3.

Overall response and PFS from C1D1 (N = 16). A, Swimmer plot of baseline patient characteristics (left-hand side of vertical axis) and PFS times (horizontal bars), colored according to RECIST guidelines. The length of the bars is proportional to the time from C1D1 to PD or the end of the study (data cutoff). B, Kaplan–Meier curve from C1D1. MBC, metastatic breast cancer; THP, taxane, trastuzumab, and pertuzumab.

Discussion

IPATHER is the first trial to assess the combination of an AKT inhibitor with anti-HER2 therapy in HER2-positive advanced breast cancer. Ipatasertib 400 mg from day 1 to 21 every 28 days plus standard-dose HP was identified as the RP2D for maintenance therapy following first-line induction chemotherapy plus HP in PIK3CAmut HER2-positive advanced breast cancer. Our findings suggest that this combination is both feasible and tolerable, with promising efficacy.

The rationale to combine a PI3K pathway inhibitor with dual anti-HER2 therapy comes from several preclinical and clinical data indicating that PI3KCAmut and/or dysregulated PI3K pathway associate with decreased efficacy of anti-HER2 agents, particularly HER2-directed antibodies. In the pivotal CLEOPATRA trial testing docetaxel, pertuzumab, and trastuzumab in the first-line treatment of HER2-positive metastatic breast cancer, patients with tumors harboring PIK3CAmut had shorter PFS and ORR than those without PIK3CAmut (20). In the early setting, PIK3CAmut was also associated with lower rates of pathological complete response (pCR) in patients treated with neoadjuvant docetaxel and HP in NeoSphere (24), in those treated with trastuzumab-based regimens with lapatinib in CHER-LOB (32), in those treated with lapatinib and trastuzumab among HER2-enriched tumors in PAMELA (33, 34), and in patients treated with neoadjuvant T-DM1 plus pertuzumab in the KRISTINE trial (35). Similarly, PIK3CAmut correlated with poorer outcomes in the NeoALTTO trial (17), although no clear effect was observed in the ExteNET trial of extended neratinib (36), suggesting alternative pathways influencing resistance mechanism to this drug. The association of PIK3CAmut status and response to HER2-directed antibody–drug conjugates in the metastatic setting is less clear. In MARIANNE, PFS among patients with PIK3CAmut tumors was 8.3 months on T-DM1 alone and 11.0 months with T-DM1 plus pertuzumab versus 16.6 and 18.8 months, respectively, in wild-type cases (37). In TH3RESA, in turn, PFS for patients with PIK3CAmut tumors was 6.2 months on T-DM1 versus 3.1 months on physician’s choice, whereas wild-type cases had 6.8 versus 3.4 months, respectively (38). Similar results were observed for patients treated with T-DM1 and with capecitabine plus lapatinib in the EMILIA trial, in which PFS was 10.9 versus 4.3 months for PIK3CAmut tumors and 9.8 versus 6.4 months for PIK3CAwt tumors, respectively (39). Interestingly, the DESTINY-Breast03 trial demonstrated that trastuzumab deruxtecan (T-DXd) maintained superior efficacy over T-DM1 regardless of PIK3CAmut status, suggesting T-DXd’s potential to overcome PIK3CAmut-associated resistance (40).

Although chemotherapy combined with HP is currently the standard first-line treatment for HER2-positive metastatic breast cancer, there is a clear need to optimize treatment for patients with tumors harboring a PIK3CAmut. IPATHER is one of the first trials to investigate targeting of the PI3K/AKT/mTOR pathway in this setting and the first one specifically evaluating the addition of an AKT inhibitor to dual anti-HER2 maintenance therapy. Other agents tested in combination with HP include the PI3K inhibitors alpelisib (41, 42) and inavolisib (43). In part 1 (safety run-in) of the phase III EPIK-B2 trial, for instance, alpelisib plus HP showed promising preliminary efficacy results, including a 50% ORR and 100% CBR in patients with PIK3CAmut tumors (42). Nevertheless, part 2 of this trial, which exclusively enrolled patients harboring HER2-positive PIK3CAmut tumors, was terminated early. Currently, another phase III study (INAVO122) is enrolling patients with HER2-positive PIK3CAmut advanced breast cancer to determine whether maintenance inavolisib plus HP improves outcomes relative to HP alone (43).

Recently presented results from the PATINA trial suggest that the addition of palbociclib to anti-HER2 therapy and ET may improve outcomes in HER2-positive/HR-positive metastatic breast cancer. In PATINA, the addition of palbociclib to anti-HER2 therapy and ET significantly improved PFS in HER2-positive/HR-positive metastatic breast cancer, with a hazard ratio of 0.74 (P = 0.0074) and a median PFS of 44.3 versus 29.1 months in the control arm, with anti-HER2 therapy and et alone (44). Although these data are promising, the full publication is awaited, and it will be important to understand whether patients enrolled in PATINA with PIK3CAmut tumors derive the same benefit from this regimen as those without PIK3CA mutations in order to ascertain the potential benefit of targeting PI3K and/or the development of triplets with CDK4/6 inhibitors and PI3K inhibitors in this context.

In IPATHER, the full dosage of 400 mg ipatasertib daily (administered 21 days on and 7 days off), together with HP, was deemed tolerable, with no DLT observed in the first six patients treated at this DL. Importantly, there were no reported grade 4 or 5 TRAE. However, some patients required dose modifications after the first cycle of treatment, and two patients discontinued ipatasertib because of toxicity. The observed safety profile of ipatasertib and HP combination, with the most frequently reported grade 3 AE being diarrhea, nausea, rash, and vomiting, is consistent with prior studies of AKT inhibitors (27). The observed incidence of gastrointestinal AE must be taken into consideration if this combination moves forward in order to initiate appropriate mitigation prophylactic and therapeutic strategies. Importantly, the RP2D of 400 mg was established according to protocol-defined criteria and was consistent with previous clinical studies, ensuring adequate target inhibition. Nonetheless, alternative dosing strategies, such as lower starting doses or earlier prophylactic interventions, could be considered in future studies to further improve tolerability.

Although the primary objectives of IPATHER were safety and feasibility, the study yielded encouraging preliminary efficacy data, with a median PFS from the time of study inclusion of 16.4 months and with 47.3% of patients remaining progression free at 18 months after initiating maintenance treatment. In the landmark CLEOPATRA trial, the median PFS in patients with PIK3CAmut tumors was 12.5 months since the start of induction chemotherapy plus HP (7, 20). Although promising, the interpretation of PFS in IPATHER is limited by the study design, which excluded patients whose disease progressed during or shortly after induction therapy, introducing an inherent selection bias and precluding cross-trial comparisons. Of note, in patients enrolled in IPATHER, the observed ORR to prior induction chemotherapy and HP was relatively modest, which can likely be explained, at least in part, by the presence of PIK3CAmut tumors. In contrast, the observed ORR with ipatasertib and HP was encouraging, although we cannot exclude that some contribution to this ORR could still come from induction chemotherapy.

This study has several limitations that should be acknowledged. First, it is a phase Ib/II single-arm study with a relatively small size, which may limit the generalizability of the findings to broader clinical populations. Second, the absence of control groups and the relatively short duration of follow-up further constrain the ability to assess long-term outcomes or comparative effectiveness. However, the favorable PFS highlights the potential of this strategy to address an unmet need in a genomically defined subgroup known to respond poorly to conventional anti-HER2 therapies. Third, ET was not mandated for patients with HR+ tumors, and just two patients received ET in combination with ipatasertib and HP. As this was a phase Ib safety-focused study, the priority was to evaluate the tolerability and preliminary efficacy of the combination of ipatasertib and HP, but it should be acknowledged that the cross-talk between the ER and the HER2 pathway may be more adequately targeted with the combination of ET and PI3K-targeting drugs. Finally, the biomarker analyses were exploratory, and it is conceivable that additional prognostic/predictive biomarkers may have affected the results. Despite these limitations, this study lays the groundwork for future trials aiming to validate predictive biomarkers and define long-term outcomes in this patient population.

In summary, the IPATHER study provides preliminary evidence that incorporating ipatasertib into maintenance HP is a feasible and promising strategy for patients with HER2-positive PIK3CAmut metastatic breast cancer. Although further studies are required to validate these findings, AKT inhibition stands out as a compelling therapeutic approach for this patient population, which is known to develop resistance to standard anti-HER2 therapies.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S1 | IPATHER trial design. aPatients could be pre-screened before or during induction treatment with chemotherapy and dual anti-HER2 blockade regimen. Patients who were HER2-positive and harbored a PIK3CA mutation, identified in archival or fresh tumor tissue samples (from primary breast, local recurrence, or metastatic lesions) or in ctDNA, qualified for the study. ctDNA, circulating tumor DNA; EOT, end of treatment; ER, estrogen receptor; HR, hormone receptor; MBC, metastatic breast cancer; PO, oral administration; PR, progesterone receptor; SC, subcutaneous.

Supplementary Figure S2

Supplementary Figure S2 | Dose de-escalation design and criteria to dose expansion. aIf ≥2 DLT occur in the first six patients, a decision had to be made on whether to expand DL1 to 10 additional patients or to de-escalate ipatasertib dose to DL-1. DL, dose level; DLT, dose-limiting toxicity.

Supplementary Figure S3

Supplementary Figure S3 | Translational genomic analysis of screened patients (N = 116). Oncoplot depicting the genomic landscape of all screened patients. ER, estrogen receptor; IHC, immunohistochemistry; ISH, in situ hybridization; NA, not available; PR, progesterone receptor.

Supplementary Figure S4

Supplementary Figure S4 | Progression-free survival in the efficacy cohort since the initiation of first-line treatment for advanced disease (N = 16). Kaplan-Meier curve from the first-line treatment for metastatic breast cancer with trastuzumab plus pertuzumab plus chemotherapy. CI, confidence interval; NR; not reported; PFS, progression-free survival.

Supplementary Table S1

Supplementary Table S1 | Representativeness of Study Participants

Supplementary Table S2

Supplementary Table S2 | Criteria for defining DLT. AE, adverse events; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CTCAE, Common Terminology Criteria for Adverse Events; DLT, dose-limiting toxicities; GI, gastrointestinal; LLN, lower limit of normal; SOC, standard of care; ULN, upper limit of normal.

Supplementary Table S3

Supplementary Table S3 | Treatment-Emergent Adverse Events (TEAE) and serious TEAE in the study safety population (N = 17)

Supplementary Table S4

Supplementary Table S4 | Drug-related dose interruptions, dose reductions, and treatment discontinuations (N = 17)

Supplementary Data S1

Supplementary Data S1. Detailed description of the experimental and analytical methods used in this study

Supplementary Data S2

Supplementary Data S2. Study Protocol

Supplementary Data S3

Supplementary data S3. Analysis outputs

Data Availability

Most of the data generated in this study have been included in the article and its supplementary materials, including the official statistical analyses for the final analysis provided in Supplementary Data S3. However, certain data are not publicly available because of confidentiality restrictions. Deidentified participant genomic and clinical data supporting the findings of the present study will be made available for academic use and within the limitations of the related informed consent upon request by submitting their request to the corresponding author, Mafalda Oliveira (moliveira@vhio.net).

Authors’ Disclosures

M. Oliveira reports grants, personal fees, and nonfinancial support from Roche during the conduct of the study; grants, personal fees, and nonfinancial support from AstraZeneca and Gilead Sciences; personal fees from Eli Lilly and Company, Curio Science, Daiichi Sankyo, Eisai, i-One, Medscape, MSD, ProteinQure, and Relay Therapeutics; grants and personal fees from Pfizer; and grants from Immutep outside the submitted work. E. Ciruelos Gil reports personal fees from Roche, Novartis, Eli Lilly and Company, AstraZeneca, Daiichi Sankyo, and Menarini outside the submitted work. G. Villacampa reports personal fees from GSK, AstraZeneca, Pierre Fabre, PharmaMar, Pfizer, and Reveal Genomics outside the submitted work. J. Gavilá reports grants and personal fees from Roche, AstraZeneca, and Pfizer and personal fees from Daiichi Sankyo outside the submitted work. P. Tolosa reports personal fees from AstraZeneca, Daiichi Sankyo, Adamed, Novartis, Pfizer, Eli Lilly and Company, Esteve, Gilead Sciences, and Roche outside the submitted work. M.A. Bergamino Sirvén reports grants from Fundación BBVA Beca Josep Baselga/Joan Rodés and personal fees from Eisai, Novartis, AstraZeneca, and Pfizer outside the submitted work. F. Salvador reports personal fees from AstraZeneca outside the submitted work and employment with AstraZeneca. X. Gonzalez Farre reports personal fees from AstraZeneca, Gilead Sciences, and Novartis and other support from Eli Lilly and Company during the conduct of the study. T. Pascual reports personal fees from AstraZeneca, Novartis, Daiichi Sankyo, Pfizer, Veracyte, Argenetics, Gilead Sciences, and Eli Lilly and Company; other support from Gilead Sciences, Daiichi Sankyo, and Roche; and nonfinancial support from SOLTI Cancer Research Group outside the submitted work. C. Saura reports personal fees from AstraZeneca, Aventik Medical S.L.U., Boehringer Ingelheim, Bristol Myers Squibb, Daiichi Sankyo, Exeter Pharma, F. Hoffmann-La Roche Ltd., Gilead Sciences, Eli Lilly and Company, Merck Sharp & Dohme, Novartis, PeerVoice PSL, Pfizer, Puma, Roche, and Sanofi during the conduct of the study; and nonfinancial support from the Cooperative Breast Cancer Group SOLTI; the Advisory Board on Cancer, Infertility, and Pregnancy; the American Society for Clinical Oncology; the European Society for Medical Oncology; GEICAM (Spanish Breast Cancer Research Group); the Oncology Society of the Official College of Physicians of Barcelona; the Spanish Society of Medical Oncology; and the European Association for Cancer Research outside the submitted work. No disclosures were reported by the other authors.

Authors’ Contributions

M. Oliveira: Conceptualization, resources, data curation, supervision, funding acquisition, validation, investigation, visualization, writing–original draft, writing–review and editing. E. Ciruelos: Conceptualization, resources, data curation, validation, visualization, writing–original draft, writing–review and editing. G. Villacampa: Conceptualization, data curation, software, formal analysis, validation, writing–review and editing. S. Morales: Validation, investigation, visualization, writing–review and editing. J. Salvador Bofill: Validation, investigation, visualization, writing–review and editing. V. Quiroga: Validation, investigation, visualization, writing–review and editing. J. Gavilá: Validation, investigation, visualization, writing–review and editing. A. Cortegoso: Validation, investigation, visualization, writing–review and editing. E. Vega: Validation, investigation, visualization, writing–review and editing. F. Henao: Validation, investigation, visualization, writing–review and editing. S. Servitja: Conceptualization, validation, investigation, visualization, writing–original draft. P. Tolosa: Validation, investigation, visualization, writing–review and editing. S. Chillara: Software, formal analysis, supervision, methodology, writing–review and editing. M. Bergamino: Validation, investigation, visualization, writing–review and editing. F. Salvador: Formal analysis, investigation, visualization, writing–review and editing. M. Paes Dias: Resources, data curation, formal analysis, visualization, methodology, writing–review and editing. J.M. Ferrero-Cafiero: Conceptualization, validation, investigation, visualization, methodology, writing–review and editing. X. Gonzalez Farre: Validation, investigation, visualization, writing–review and editing. T. Pascual: Conceptualization, validation, investigation, visualization, writing–review and editing. C. Saura: Conceptualization, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing.