Effect of Capivasertib in Patients With an AKT1 E17K-Mutated Tumor: NCI-MATCH Subprotocol EAY131-Y Nonrandomized Trial | Breast Cancer | JAMA Oncology | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  CONSORT Diagram for NCI MATCH Subprotocol
CONSORT Diagram for NCI MATCH Subprotocol

aReasons participants did not receive an assignment: screened outside of the subprotocol group window (n = 8), also have sequence variation excluded (n = 12), prior treatment (n = 1), and an annotation issue (n = 5).

bReasons assigned participants were not enrolled: prior treatment (n = 2), deteriorating performance status (n = 3), inadequate organ function (n = 1), and reason unknown (n = 2).

Figure 2.  Response to Capivasertib in Patients With an AKT1 E17K-Mutated Tumor
Response to Capivasertib in Patients With an AKT1 E17K-Mutated Tumor

The subprotocol EAY131-Y trial included 35 eligible, evaluable, and analyzable patients. Eight patients were screened outside of the subprotocol group window, 12 had excluded mutated tumors, 1 had previous treatment, and 5 had an annotation issue. Two patients had previous treatment, 3 had deteriorating Eastern Cooperative Oncology Group Performance Status score, 1 had inadequate organ function, and 2 had unknown reasons. A, The waterfall plot demonstrates the best percentage change from baseline in the sum of the diameters of the target lesions for patients with follow-up target lesion measurements (n = 28). The best responses labeled below the bars demonstrate the confirmed overall response rate by Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1: complete response (CR; n = 1), partial response (PR; n = 9), stable disease (SD; n = 16), and progressive disease (PD; n = 2). Four patients are labeled as SD because of missing follow-up scans confirming PR. One patient with hormone receptor (HR)-positive/ERBB2 (formerly HER2)-negative breast cancer (BC) missed cycle 2 measurements, and at cycle 4, the patient experienced a 74% reduction with a new lesion after cycle 4. Of the 3 evaluable patients with HR-positive/ERBB2-negative BC who continued hormone therapy along with capivasertib, 2 had a PR and 1 was a patient missing cycle 2 measurements. B, The swimmer plot of the objective responses from the start of treatment to disease progression is shown. At the time of the analysis, 2 patients continued on therapy. TNBC indicates triple-negative breast cancer. C and D, In the subprotocol EAY131-Y trial, the median progression-free survival (PFS) was 5.5 months (C) and the median overall survival (OS) was 14.5 months (D).

aNew lesion.

bMissing follow-up scan confirming PR.

cHormone therapy along with capivasertib.

dReduction with a new lesion after cycle 4.

Table 1.  Clinical Features of 35 Eligible Patients
Clinical Features of 35 Eligible Patients
Table 2.  Tumor Histologies of Eligible Patients
Tumor Histologies of Eligible Patients
Table 3.  Any Adverse Event Occurring in at Least 10% of Patients
Any Adverse Event Occurring in at Least 10% of Patients
1.
Vivanco  I, Sawyers  CL.  The phosphatidylinositol 3-kinase AKT pathway in human cancer.   Nat Rev Cancer. 2002;2(7):489-501. doi:10.1038/nrc839 PubMedGoogle ScholarCrossref
2.
Mundi  PS, Sachdev  J, McCourt  C, Kalinsky  K.  AKT in cancer: new molecular insights and advances in drug development.   Br J Clin Pharmacol. 2016;82(4):943-956. doi:10.1111/bcp.13021 PubMedGoogle ScholarCrossref
3.
Carpten  JD, Faber  AL, Horn  C,  et al.  A transforming mutation in the pleckstrin homology domain of AKT1 in cancer.   Nature. 2007;448(7152):439-444. doi:10.1038/nature05933 PubMedGoogle ScholarCrossref
4.
Stemke-Hale  K, Gonzalez-Angulo  AM, Lluch  A,  et al.  An integrative genomic and proteomic analysis of PIK3CA, PTEN, and AKT mutations in breast cancer.   Cancer Res. 2008;68(15):6084-6091. doi:10.1158/0008-5472.CAN-07-6854 PubMedGoogle ScholarCrossref
5.
Khoury  JD, Wang  WL, Prieto  VG,  et al.  Validation of immunohistochemical assays for integral biomarkers in the NCI-MATCH EAY131 clinical trial.   Clin Cancer Res. 2018;24(3):521-531. doi:10.1158/1078-0432.CCR-17-1597 PubMedGoogle ScholarCrossref
6.
She  QB, Halilovic  E, Ye  Q,  et al.  4E-BP1 is a key effector of the oncogenic activation of the AKT and ERK signaling pathways that integrates their function in tumors.   Cancer Cell. 2010;18(1):39-51. doi:10.1016/j.ccr.2010.05.023 PubMedGoogle ScholarCrossref
7.
Kalinsky  K, Jacks  LM, Heguy  A,  et al.  PIK3CA mutation associates with improved outcome in breast cancer.   Clin Cancer Res. 2009;15(16):5049-5059. doi:10.1158/1078-0432.CCR-09-0632 PubMedGoogle ScholarCrossref
8.
Razavi  P, Chang  MT, Xu  G,  et al.  The genomic landscape of endocrine-resistant advanced breast cancers.   Cancer Cell. 2018;34(3):427-438.e6. doi:10.1016/j.ccell.2018.08.008PubMedGoogle ScholarCrossref
9.
Davies  BR, Greenwood  H, Dudley  P,  et al.  Preclinical pharmacology of AZD5363, an inhibitor of AKT: pharmacodynamics, antitumor activity, and correlation of monotherapy activity with genetic background.   Mol Cancer Ther. 2012;11(4):873-887. doi:10.1158/1535-7163.MCT-11-0824-T PubMedGoogle ScholarCrossref
10.
Davies  BR, Guan  N, Logie  A,  et al.  Tumors with AKT1E17K mutations are rational targets for single agent or combination therapy with AKT inhibitors.   Mol Cancer Ther. 2015;14(11):2441-2451. doi:10.1158/1535-7163.MCT-15-0230 PubMedGoogle ScholarCrossref
11.
Ribas  R, Pancholi  S, Guest  SK,  et al.  AKT antagonist AZD5363 influences estrogen receptor function in endocrine-resistant breast cancer and synergizes with fulvestrant (ICI182780) in vivo.   Mol Cancer Ther. 2015;14(9):2035-2048. doi:10.1158/1535-7163.MCT-15-0143 PubMedGoogle ScholarCrossref
12.
Mullard  A.  NCI-MATCH trial pushes cancer umbrella trial paradigm.   Nat Rev Drug Discov. 2015;14(8):513-515. doi:10.1038/nrd4694 PubMedGoogle ScholarCrossref
13.
Eisenhauer  EA, Therasse  P, Bogaerts  J,  et al.  New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1).   Eur J Cancer. 2009;45(2):228-247. doi:10.1016/j.ejca.2008.10.026 PubMedGoogle ScholarCrossref
14.
Lih  CJ, Harrington  RD, Sims  DJ,  et al.  Analytical validation of the next-generation sequencing assay for a nationwide signal-finding clinical trial: Molecular Analysis for Therapy Choice clinical trial.   J Mol Diagn. 2017;19(2):313-327. doi:10.1016/j.jmoldx.2016.10.007 PubMedGoogle ScholarCrossref
15.
Hyman  DM, Smyth  LM, Donoghue  MTA,  et al.  AKT inhibition in solid tumors with AKT1 mutations.   J Clin Oncol. 2017;35(20):2251-2259. doi:10.1200/JCO.2017.73.0143 PubMedGoogle ScholarCrossref
16.
Smyth  L, Oliveira  M, Ciruelos  E,  et al.  Abstract P5-21-32: AZD5363 in combination with fulvestrant in AKT1-mutant ER-positive metastatic breast cancer.   Cancer Res. 2018;78(4)(suppl):5-21. doi:10.1158/1538-7445.SABCS17-P5-21-32Google Scholar
17.
Smyth  LM, Zhou  Q, Nguyen  B,  et al; AACR Project GENIE Consortium.  Characteristics and outcome of AKT1E17K-mutant breast cancer defined through AACR Project GENIE, a clinicogenomic registry.   Cancer Discov. 2020;10(4):526-535. doi:10.1158/2159-8290.CD-19-1209 PubMedGoogle ScholarCrossref
18.
Turner  NC, Alarcón  E, Armstrong  AC,  et al.  BEECH: a dose-finding run-in followed by a randomised phase II study assessing the efficacy of AKT inhibitor capivasertib (AZD5363) combined with paclitaxel in patients with estrogen receptor-positive advanced or metastatic breast cancer, and in a PIK3CA mutant sub-population.   Ann Oncol. 2019;30(5):774-780. doi:10.1093/annonc/mdz086 PubMedGoogle ScholarCrossref
19.
Jones  RH, Carucci  M, Casbard  AC,  et al.  Capivasertib (AZD5363) plus fulvestrant versus placebo plus fulvestrant after relapse or progression on an aromatase inhibitor in metastatic ER-positive breast cancer (FAKTION): a randomized, double-blind, placebo-controlled, phase II trial.   J Clin Oncol. 2019;37(15)(suppl):1005. doi:10.1200/JCO.2019.37.15_suppl.1005Google ScholarCrossref
20.
Altomare  DA, Testa  JR.  Perturbations of the AKT signaling pathway in human cancer.   Oncogene. 2005;24(50):7455-7464. doi:10.1038/sj.onc.1209085 PubMedGoogle ScholarCrossref
21.
Banerji  S, Cibulskis  K, Rangel-Escareno  C,  et al.  Sequence analysis of mutations and translocations across breast cancer subtypes.   Nature. 2012;486(7403):405-409. doi:10.1038/nature11154 PubMedGoogle ScholarCrossref
22.
Mosquera  JM, Varma  S, Pauli  C,  et al.  MAGI3-AKT3 fusion in breast cancer amended.   Nature. 2015;520(7547):E11-E12. doi:10.1038/nature14265 PubMedGoogle ScholarCrossref
23.
Parikh  C, Janakiraman  V, Wu  WI,  et al.  Disruption of PH-kinase domain interactions leads to oncogenic activation of AKT in human cancers.   Proc Natl Acad Sci U S A. 2012;109(47):19368-19373. doi:10.1073/pnas.1204384109 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    December 30, 2020

    Effect of Capivasertib in Patients With an AKT1 E17K-Mutated Tumor: NCI-MATCH Subprotocol EAY131-Y Nonrandomized Trial

    Author Affiliations
    • 1Department of Medicine, Columbia University Irving Medical Center, New York, New York
    • 2Now with Winship Cancer Institute at Emory University, Atlanta, Georgia
    • 3Department of Biostatistics, Dana-Farber Cancer Institute, Harvard T.H. Chan School of Public Health, Boston, Massachusetts
    • 4Department of Obstetrics and Gynecology, Washington University School of Medicine in St Louis, St Louis, Missouri
    • 5Department of Medicine, TGen/HonorHealth Research Institute, Scottsdale, Arizona
    • 6Department of Medicine, Thomas Jefferson University Health, Philadelphia, Pennsylvania
    • 7Division of Cancer Treatment and Diagnosis, National Cancer Institute, Bethesda, Maryland
    • 8Department of Biostatistics, Dana-Farber Cancer Institute–ECOG (Eastern Cooperative Oncology Group)–ACRIN (American College of Radiology Imaging Network) Biostatistics Center, Boston, Massachusetts
    • 9Division of Cancer Therapeutics and Diagnosis, Molecular Characterization and Assay Development Laboratory, Leidos, Frederick, Maryland
    • 10Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston
    • 11Department of Pathology, City of Hope National Medical Center, Duarte, California
    • 12Department of Medicine, University of Pennsylvania, Philadelphia
    • 13Department of Medicine, University of Texas Southwestern Simmons Cancer Center, Dallas
    • 14Department of Medicine, Massachusetts General Hospital, Boston
    JAMA Oncol. 2021;7(2):271-278. doi:10.1001/jamaoncol.2020.6741
    Key Points

    Question  What is the clinical activity of capivasertib in patients with an AKT1 E17K-mutated metastatic tumor?

    Findings  In this nonrandomized trial of 35 patients with an AKT1 E17K-mutated tumor treated with capivasertib, the confirmed overall response rate was 28.6%, and 1 complete response and 9 partial responses were achieved. Grade 3 treatment-related adverse events included hyperglycemia and rash, and 1 patient had a grade 4 hyperglycemic adverse event.

    Meaning  The results of this trial indicate that capivasertib had a clinically significant objective response rate in patients with an AKT1 E17K-mutated metastatic tumor of various histologies, including rare cancers.

    Abstract

    Importance  In the National Cancer Institute Molecular Analysis for Therapy Choice (NCI-MATCH) trial, agents targeting genetic tumor abnormalities are administered to patients. In the NCI-MATCH subprotocol EAY131-Y trial, patients with an AKT1 E17K-mutated metastatic tumor received the pan-AKT inhibitor capivasertib.

    Objective  To assess the objective response rate (ORR) of capivasertib in patients with an AKT1 E17K-mutated tumor.

    Design, Setting, and Participants  Between July 13, 2016, and August 10, 2017, patients in the NCI-MATCH trial were enrolled and assigned to the subprotocol EAY131-Y nonrandomized trial. Patients included adults with an AKT1 E17K-mutated metastatic tumor that had progressed with standard treatment, and these patients were assigned to receive capivasertib. Tumor assessments were repeated every 2 cycles. Data analysis of this evaluable population was performed from November 8, 2019, to March 12, 2020.

    Interventions  The study treatment was capivasertib, 480 mg, orally twice daily for 4 days on and 3 days off weekly in 28-day cycles until disease progression or unacceptable toxic effect. If patients continued hormone therapy for metastatic breast cancer, the capivasertib dose was 400 mg.

    Main Outcomes and Measures  The primary end point was the ORR (ie, complete response [CR] and partial response) according to the Response Evaluation Criteria in Solid Tumors criteria, version 1.1. Secondary end points included progression-free survival (PFS), 6-month PFS, overall survival, and safety.

    Results  In total, 35 evaluable and analyzable patients were included, of whom 30 were women (86%), and the median (range) age was 61 (32-73) years. The most prevalent cancers were breast (18 [51%]), including 15 patients with hormone receptor (HR)–positive/ERBB2-negative and 3 with triple-negative disease, and gynecologic (11 [31%]) cancers. The ORR rate was 28.6% (95% CI, 15%-46%). One patient with endometrioid endometrial adenocarcinoma achieved a CR and remained on therapy at 35.6 months. Patients with confirmed partial response had the following tumor types: 7 had HR-positive/ERBB2-negative breast cancer, 1 had uterine leiomyosarcoma, and 1 had oncocytic parotid gland carcinoma and continued receiving treatment at 28.8 months. Sixteen patients (46%) had stable disease as the best response, 2 (6%) had progressive disease, and 7 (20%) were not evaluable. With a median follow-up of 28.4 months, the overall 6-month PFS rate was 50% (95% CI, 35%-71%). Capivasertib was discontinued because of adverse events in 11 of 35 patients (31%). Grade 3 treatment-related adverse events included hyperglycemia (8 [23%]) and rash (4 [11%]). One grade 4 hyperglycemic adverse event was reported.

    Conclusions and Relevance  This nonrandomized trial found that, in patients with an AKT1 E17K-mutated tumor treated with capivasertib, a clinically significant ORR was achieved, including 1 CR. Clinically meaningful activity with single-agent capivasertib was demonstrated in refractory malignant neoplasms, including rare cancers.

    Trial Registration  ClinicalTrials.gov Identifier: NCT00700882

    Introduction

    The phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway is one of the most commonly altered pathways in cancer, including sequence variations, somatic copy number abnormalities, overexpression, and aberrant signaling.1 AKT represents an important signaling hub, modulating cell survival, cell cycle progression, and cellular growth through activation of downstream targets.2 An AKT1 (OMIM 164730) point alteration in the pleckstrin homology domain that replaces a glutamic acid with lysine (E17K) at residue 17 is the most commonly reported AKT sequence variation across tumors.3 This alteration confers increased activity by promoting localization of AKT to the plasma membrane through a PI3K-independent mechanism.4 AKT-mutated tumors are rarely found in human cancers.5,6 For example, AKT1 E17K-mutated tumors have been reported in 2% to 7% of breast tumors.4,7,8 Measurement of p-AKT levels in cell lines and tumors confirms a quantitative increase in activated AKT as a consequence of this point alteration.4

    The oral competitive kinase domain inhibitor capivasertib (AZD5363) has been shown to inhibit all 3 isoforms of AKT in vitro.9 Capivasertib has also been associated with inhibiting AKT signaling in MCF10A-AKT1 E17K cells and substantially reducing tumor growth in 2 AKT1 E17K-mutated breast cancer explant models.10 Furthermore, the addition of capivasertib has been shown to resensitize a tamoxifen citrate–resistant cell line to the antiproliferative effects of tamoxifen.11 These preclinical data suggest that AKT inhibition with capivasertib may provide a novel therapeutic approach and serve as the rationale for investigating this agent in AKT-mutated tumors.

    The National Cancer Institute Molecular Analysis for Therapy Choice (NCI-MATCH) trial opened in 2015. It is the largest national signal-finding trial in the United States for patients with tumors that are refractory to standard treatment that incorporates genomic testing to direct patients to investigational, molecularly targeted treatments.12 The trial was designed as a master protocol with nearly 40 single-group phase 2 trials as well as both completed and ongoing groups. In the present NCI-MATCH subprotocol EAY131-Y trial, we analyzed the clinical outcomes and safety of capivasertib in patients with an AKT1 E17K-mutated tumor. We aimed to assess the objective response rate (ORR) among these patients.

    Methods

    This nonrandomized trial was approved by the institutional review boards at all participating sites. All patients were required to sign a written informed consent form. We followed the Transparent Reporting of Evaluations With Nonrandomized Designs (TREND) reporting guideline.

    Eligibility Criteria

    The entry criteria for the NCI-MATCH trial master protocol included patients 18 years or older who had histologically confirmed solid tumors, lymphoma, or multiple myeloma with progression after at least 1 line of standard systemic therapy or a malignant neoplasm for which no known effective standard therapy existed. Patients were required to have measurable disease according to Response Evaluation Criteria in Solid Tumors (RECIST), version 1.113; an Eastern Cooperative Oncology Group (ECOG) Performance Status score of 0 to 1 (scale: 0-5, with higher scores indicating worse functional status); life expectancy of at least 3 months; and adequate hematologic, kidney, and liver function. Patient-reported race/ethnicity were captured. Enrolled patients must have had a tumor that was amenable to biopsy or tumor specimen that was obtained within the past 6 months without intervening response to therapy. Tumor sequencing (Oncomine Comprehensive Assay [formerly Oncomine Cancer Panel]; ThermoFisher Scientific) and immunohistochemical assays for phosphatase and tensin homologue (PTEN) expression were performed with validated assays in central laboratories accredited for the Clinical Laboratory Improvement Amendments.5,14

    In 2017, the NCI-MATCH Trial was amended to allow for recruitment using designated outside laboratories through next-generation sequencing (NGS) results determined by an NCI-MATCH–designated outside laboratory. Patients were not allowed to receive nonprotocol therapy after the biopsy or tumor specimen submission results were received. Any systemic anticancer therapy, radiotherapy (except palliative radiation therapy of 30 Gy or less), or major surgical treatment must have been completed more than 4 weeks before study treatment initiation. Patients with underlying cardiopulmonary dysfunction or resting corrected QT interval greater than 480 milliseconds were excluded. Patients with stable brain metastases were included.

    For this NCI-MATCH subprotocol EAY131-Y nonrandomized trial (Supplement 1), only patients with an AKT1 E17K-mutated tumor were enrolled and assigned to the subprotocol between July 13, 2016, and August 10, 2017. Patients with concomitant KRAS (OMIM 190070), NRAS (OMIM 164790), HRAS (OMIM 190020), and BRAF (OMIM 164757)–mutated tumors were excluded because these variations may limit response or result in resistance to therapy.9 In cell lines, for instance, the presence of RAS (OMIM 606614) sequence variation was statistically significantly associated with resistance to capivasertib,9 potentially because of ERK-mediated AKT independence.6 Patients were not allowed to have previously received PI3K, AKT, or mammalian target of rapamycin (mTOR) inhibitors with the exception of US Food and Drug Administration–approved rapalogs (1) in the metastatic setting, such as everolimus for patients with hormone receptor (HR)–positive/ERBB2 (formerly HER2)-negative metastatic breast cancer, or (2) if administered during a short-term preoperative trial (<4-week duration). Patients with diabetes were allowed, provided the fasting glucose value was less than 8.9 mmol/L or 160 mg/dL and as long as the patient did not require insulin or used 2 or fewer oral hypoglycemic medications for diabetic control. Patients with HR-positive/ERBB2-negative breast cancer whose tumor most recently progressed while using fulvestrant or aromatase inhibitors were allowed to enroll and receive the study medication concurrently with that specific endocrine therapy. Strong inhibitors or potent inducers or substrates of CYP3A4 or substrates of CYP2D6 must not have been received within 2 weeks before the first dose of study treatment (3 weeks for St John’s wort).

    Study Treatment and Sequence Variation

    Patients were assigned to receive capivasertib, 480 mg, orally twice daily for 4 days on and 3 days off in 28-day cycles. Capivasertib was administered on an open-label, single-group basis, with adjustment for toxic effect and with the first dose reduction at 400 mg and the second reduction at 320 mg on the same dosing schedule. For patients who continued fulvestrant or an aromatase inhibitor concomitantly, capivasertib was administered at 400 mg twice daily for 4 days on and 3 days off in 28-day cycles, with 1 dose reduction at 320 mg on the same dosing schedule. Patients were required to be fasting for 2 hours before and 1 hour after each dose. There were no breaks between cycles.

    The NCI-MATCH trial NGS assay used the Oncomine Comprehensive Assay as an investigational assay under the Investigational New Drug application held by the NCI. The assay can reliably identify the presence or absence of greater than 3000 known sequence variations in 161 unique genes (87 genes with hot-spot coverage, 48 genes with full coverage, and copy number variants and 51 selected fusions). Assay validation and performance analysis has been published previously.14

    End Points, Assessment, and Toxic Effects

    The primary end point was the ORR (complete response [CR] and partial response [PR]) according to RECIST, version 1.1.13 Key secondary end points were progression-free survival (PFS), PFS at 6 months, overall survival (OS), and safety. Progression-free survival was defined as the time from subprotocol registration to the determination of disease progression or death from any cause or censored at the date of last disease assessment for patients who had not progressed. Overall survival was defined as the time from registration to death or censored at the date of last contact. Imaging was performed every 8 weeks (2 cycles) for the first 2 years and then every 12 weeks (3 cycles) thereafter. Toxic effects and adverse events (AEs) were classified according to the Common Terminology Criteria for Adverse Events, version 4.0 (NCI). Study participation was discontinued if patients developed AEs that led to a delay of more than 4 weeks or if patients could not tolerate the capivasertib study dose of 320 mg twice daily for 4 days on and 3 days off.

    Statistical Analysis

    An agent was defined as promising if at least 5 responses were observed in 31 evaluable patients (ie, 16% ORR). This criterion allowed for 92% power to distinguish a 25% ORR from a null rate of 5%. The 1-sided type I error rate was 1.8%.

    Descriptive statistics were used to summarize patient characteristics, treatment, and study outcomes. A 1-sided P value for the ORR was calculated using a 1-sample binomial test against the null rate of 5%, and P < .05 was deemed as statistically significant. All statistical analyses were performed with R, version 3.5.0 (R Foundation for Statistical Computing). Data analysis of this evaluable population was performed from November 8, 2019, to March 12, 2020.

    Results
    Patient Population

    Of the 5548 patients in the NCI-MATCH trial who had central testing of their tumor, 68 patients (1.2%) had an AKT1 E17K-mutated tumor (Figure 1 and eFigure 1 [Study Schema] in Supplement 2). Forty-three patients with an AKT1 E17K-mutated tumor were assigned to the subprotocol group, of whom a total of 8 patients were not enrolled because of previous treatment (n = 2), deteriorating ECOG Performance Status score (n = 3), inadequate organ function (n = 1), or unknown reason (n = 2) (Figure 1). An AKT1 E17K alteration was identified by an Oncomine NGS assay panel except for 1 patient with breast cancer screened by a designated laboratory NGS panel that identified an AKT sequence variation. All 35 enrolled patients, of whom 30 were women (86%) and 5 were men (14%) with a median (range) age of 61 (32-73) years, were eligible and initiated capivasertib and therefore were included in the statistical analysis.

    Baseline clinical features of the 35 patients are described in Table 1. Most patients (24 [69%]) had an ECOG Performance Status score of 1 compared with 11 patients (31%) with an ECOG Performance Status score of 0. The median (range) number of previous lines of therapy was 4 (0-10), and 19 patients (54%) had more than 3 previous lines of therapy.

    Table 2 describes the tumor histologies of eligible patients. Patients with breast cancer comprised approximately half of the cohort (18 of 35 [51%]), including 15 with HR-positive/ERBB2-negative disease and 3 with triple-negative disease. Four patients (11%) with HR-positive/ERBB2-negative disease continued hormonal therapy (letrozole or fulvestrant) concurrently with capivasertib. Three patients (9%), also with HR-positive/ERBB2-negative disease, had previously received the mTOR inhibitor everolimus. Eleven patients (31%) had gynecologic tumors, including 7 with endometrioid adenocarcinoma of the endometrium (n = 4) or ovary (n = 3). The remaining tumor histologies, including rare tumors, were represented with 1 patient (3%) each, such as leiomyosarcoma arising from previous low-grade adenosarcoma of the endometrium, myxofibrosarcoma, and an oncocytic carcinoma of the parotid gland.

    Clinical Outcomes

    The ORR was 28.6% (10 of 35 patients; 95% CI, 15%-46%; P < .001; against the null rate of 5%). One patient with endometrioid endometrial adenocarcinoma achieved a CR and remained on therapy more than 35.6 months as of November 20, 2019 (Figure 2A). Of the 9 patients with a confirmed PR, 7 patients had HR-positive/ERBB2-negative breast cancer, 1 had uterine leiomyosarcoma, and 1 had oncocytic carcinoma of the parotid gland who remained on treatment for 28.8 or more months (Figure 2B). The median (range) duration of response was 4.4 (3.1 to ≥31.7) months. Four patients were labeled as having stable disease (SD) because of a missing follow-up scan confirming PR (Figure 2A). Of these patients, 3 had tumors that progressed within 1 to 2 cycles with a newly identified lesion. The fourth patient had HR-positive/ERBB2-negative breast cancer and achieved 31% reduction after cycle 2 but was measured with 27% reduction at cycle 4 and clinical progression after cycle 5.

    Of the remaining patients, 16 (46%) had SD and 2 (6%) had progressive disease (PD). Seven patients (44%) had stability for at least 5.5 months (168 days). Of the 3 evaluable patients with HR-positive/ERBB2-negative breast cancer who continued fulvestrant or an aromatase inhibitor after progression, 2 had a PR with fulvestrant, and 1 with SD combined with letrozole achieved a 31% reduction after cycle 2 but 27% reduction at cycle 4. Of the 2 evaluable patients with previous everolimus treatment for HR-positive/ERBB2-negative breast cancer, 1 experienced PR and 1 had PD. The third patient who previously received everolimus also continued use of the aromatase inhibitor exemestane after progression with capivasertib, although the patient was unevaluable because she discontinued treatment before any evaluation. Seven patients (20%) were coded with a best response as unevaluable because of death in cycle 1 (n = 2), baseline scan outside of the 4-week window (n = 1), no disease evaluation and discontinuation of treatment after 1 to 2 cycles because of AEs (n = 2), and no disease evaluation and discontinuation of treatment after cycle 1 because of patient withdrawal (n = 2).

    With a median (range) follow-up of 28.4 months, the median PFS was 5.5 (0-35) months (95% CI, 4.6-11.3) (Figure 2C). The median PFS was 1.9 months (95% CI, 1.58-not applicable [NA]) for 6 patients with breast cancer, 11.3 months (95% CI, 4.57-NA) for 11 patients with a gynecologic tumor, and 5.5 months (95% CI, 6.18-NA) for 18 patients with non–breast cancer or non–gynecologic cancer. The overall 6-month PFS was 50% (95% CI, 35%-71%). No patient started new, nonprotocol anticancer treatment before progression. The median OS was 14.5 months (95% CI, 10.2-NA) (Figure 2D).

    All 35 treated patients were included in the toxic effect analysis. Grade 1 or 2 AEs that occurred in 10% or more of patients and/or any grade 3 or 4 AEs are listed in Table 3. Grade 1 or 2 AEs that occurred in 20% of patients or more included diarrhea (17 [49%]), fatigue (15 [43%]), nausea (13 [37%]), proteinuria (10 [29%]), hyperglycemia (8 [23%]), and anorexia (7 [20%]). Grade 3 AEs that occurred in at least 10% of patients included hyperglycemia (8 [23%]) and maculopapular rash (4 [11%]). One patient experienced a grade 4 hyperglycemic event. Fifteen patients (43%) had at least 1 AE-related dose modification during treatment, including delay, reduction, discontinuation, or omission of a dose. Median (range) treatment duration was 5 (1 to ≥38) cycles. All dose modifications occurred in patients who received single-agent capivasertib without hormone therapy. Five patients (14%) had at least 1 AE-related dose reduction. Capivasertib was discontinued because of AEs in 11 of 35 patients (31%), 5 of which occurred during cycle 1. These AEs were mostly grade 3 events, with the exception of 1 non–grade 3 anorexia and 1 grade-4 hyperglycemia. Of the 3 patients who discontinued treatment because of hyperglycemia, only 1 had preexisting diabetes. One patient death occurred because of PD and was not attributable to capivasertib.

    Sequence Variation Landscape

    Co-occurrence variant data were available for 34 patients (eFigure 2 in Supplement 2). TP53 (OMIM 191170) alteration in tumors was present in 35% of patients (n = 12 of 34), with 4 of those patients achieving a PR. No association was found between the response to AKT inhibition and the presence of a TP53-mutated tumor, although the number was low. In addition, a PIK3CA E726K (OMIM 171834) sequence variation was identified in a patient with HR-positive/ERBB2-negative breast cancer (best response: unevaluable), and 1 patient with endometrioid adenosarcoma had a PIK3R1 N564D-mutated tumor (OMIM 171833) (SD). PTEN loss by immunohistochemistry5,14 was not identified in any tumor. Amplification in AKT1 in addition to an AKT sequence variation was observed in 1 patient with HR-positive/ERBB2-negative breast cancer (PR). ESR1 (OMIM 133430) alteration was also identified in 7 patients with an AKT1 E17K-mutated tumor: 5 with an HR-positive/ERBB2-negative breast tumor (2 PRs and 3 SDs) and 2 with endometrioid endometrial adenocarcinoma (1 SD and 1 unevaluable). All 5 patients with ESR1-mutated breast cancer received previous endocrine therapy, whereas only 1 of the 2 patients with ESR1-mutated endometrial adenocarcinoma had previous endocrine treatment (ie, megestrol acetate). Although ESR1 sequence variation often reflects an acquired resistance to antiestrogen therapy, no association was found between response to AKT inhibition and the presence of an ESR1 alteration, albeit with a small sample size.

    Discussion

    The pan-AKT inhibitor capivasertib met its prespecified primary end point, with an ORR of 28.6% in patients with an AKT1 E17K-mutated tumor. We observed a CR and a prolonged disease stabilization in a heavily pretreated, heterogeneous patient population. Despite the low rate of AKT1 E17K sequence variation (1.2% [68 of 5548]) in patients whose tumor was tested centrally, this group was brought together within 1 year. The clinical activity of capivasertib in the setting of this infrequent sequence variation appeared promising.

    The median (interquartile range) duration of response for those with a CR or PR was 4.4 (3.1 to ≥31.7) months, and of those with SD, 7 of 16 (44%) had stability for at least 5.5 months. Only 2 patients had PD as their best response. Results of this nonrandomized trial are consistent with findings of a previously reported multicohort basket trial of capivasertib with the same dose and schedule15 in which 17.3% of patients (9 of 52) with an AKT1 E17K-mutated tumor had a confirmed PR. Although the ORR in the NCI-MATCH trial is numerically higher, this rate should be interpreted with caution because these trials enrolled a comparatively small population with heterogenous previous treatments and tumor histologies. In both studies, most tumors were metastatic AKT1 E17K-mutated breast or gynecologic tumors, potentially limiting the generalizability of the clinical activity of AKT inhibition in a tumor-agnostic approach.

    Given the preclinical data that capivasertib can resensitize endocrine-resistant cells to endocrine therapy,11 we allowed the continuation of antiestrogen therapy at the time of progression in HR-positive/ERBB2-negative metastatic breast cancer in 3 patients with evaluable disease; 2 had a PR and 1 had SD. These data are in accordance with initial results from a small cohort that reported a confirmed response rate of 33% (4 of 12) with fulvestrant plus capivasertib in fulvestrant-resistant disease.16 Of the 2 evaluable patients with previous everolimus treatment for HR-positive/ERBB2-negative breast cancer in the present study, 1 patient experienced PR and 1 had PD. Although the presence of an AKT1 E17K sequence variation has been associated with a clinical advantage from mTOR inhibition in a genomics registry,17 prospective randomized studies with AKT inhibitors in HR-positive/ERBB2-negative metastatic breast cancer excluded previous mTOR inhibition.18,19 Additional data are needed to determine whether previous mTOR inhibition is associated with clinical activity with AKT inhibition.

    In this subprotocol trial, 15 of 35 patients (43%) required at least 1 AE-related dose modification, and 11 (31%) discontinued capivasertib because of AEs. Although the AE profile, including grade 3 or higher hyperglycemia, rash, and gastrointestinal events, was consistent with AEs previously reported using a similar capivasertib dose, the discontinuation rate was higher than the previously reported rate of approximately 12%.15 All dose modifications were observed in patients who started at 480 mg twice daily for 4 days on and 3 days off weekly. This discontinuation rate may be associated with NCI-MATCH trial site participation that included both academic and community centers, reflecting the potential real-world tolerability of AKT inhibition. The present trial is also notable for its inclusion of patients with well-controlled diabetes; however, only 1 patient with diabetes required discontinuation because of glycemic-related toxic effects. Awareness of these potential toxic effects is critical in the management of patients receiving these agents, and AEs (particularly hyperglycemia) should be proactively managed in future trials.

    Limitations

    Interpretation of these results should be restricted to patients with an AKT1 E17K-mutated tumor. Patients with a tumor with concomitant KRAS, NRAS, HRAS, and BRAF sequence variations were excluded because of potential resistance concerns. In addition to AKT alterations, other means of augmenting AKT activity include AKT gene amplification and AKT rearrangements. However, patients with only AKT amplifications or fusions in their tumors were excluded given the current lack of preclinical data to support their inclusion.20-22 Although other activating sequence variations in AKT1, AKT2 (OMIM 164731), and AKT3 (OMIM 611223) have been described,15,23 further data are needed to determine the potential benefit of AKT inhibition in tumors with AKT alterations other than AKT1 E17K-mutated tumors. Currently, AKT inhibitors remain investigational, and various strategies are being evaluated in clinical trials, including registration trials, in combination with hormonal therapy or chemotherapy.

    Conclusions

    In this nonrandomized trial, capivasertib demonstrated clinically meaningful activity in various histologies, including rare cancers, in patients with an AKT1 E17K-mutated tumor.

    Back to top
    Article Information

    Accepted for Publication: September 30, 2020.

    Published Online: December 30, 2020. doi:10.1001/jamaoncol.2020.6741

    Corresponding Author: Kevin Kalinsky, MD, MS, Winship Cancer Institute at Emory University, 1365 Clifton Rd, Ste B4112, Atlanta, GA 30322 (kkalins@emory.edu).

    Author Contributions: Drs Kalinsky and Hong had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Kalinsky, McCourt, Sachdev, Mitchell, Zwiebel, Doyle, McShane, Li, Gray, Rubinstein, Patton, Williams, Hamilton, Conley, O'Dwyer, Harris, Arteaga, Chen, Flaherty.

    Acquisition, analysis, or interpretation of data: Kalinsky, Hong, McCourt, Sachdev, Mitchell, Zwiebel, McShane, Gray, Patton, Williams, Hamilton, O’Dwyer, Harris, Chen, Flaherty.

    Drafting of the manuscript: Kalinsky, Hong, Sachdev, Mitchell, Patton, Williams, Hamilton, Harris, Chen.

    Critical revision of the manuscript for important intellectual content: Kalinsky, Hong, McCourt, Sachdev, Mitchell, Zwiebel, Doyle, McShane, Li, Gray, Rubinstein, Patton, Hamilton, Conley, O’Dwyer, Harris, Arteaga, Chen, Flaherty.

    Statistical analysis: Kalinsky, Hong, Li, Hamilton.

    Administrative, technical, or material support: McCourt, Mitchell, Doyle, Gray, Patton, Williams, Hamilton, O’Dwyer, Harris, Arteaga.

    Supervision: Kalinsky, McCourt, Doyle, Gray, Williams, Hamilton, Conley, Harris, Chen, Flaherty.

    Other - Reviewed statistical analyses: McShane.

    Conflict of Interest Disclosures: Dr Kalinsky reported receiving personal fees from Seattle Genetics, Merck, Immunomedics, Novartis, Eisai, Eli Lilly, Pfizer, and AstraZeneca; receiving nonfinancial support from Immunomedics, Novartis, Incyte, Genentech, Calithera Biosciences, Acetylon, Seattle Genetics, Amgen, Zentalis Pharmaceuticals, and CyomX Therapeutics outside the submitted work; and having a spouse who was previously employed by Array Biopharma and Pfizer and is currently employed by Grail. Dr Sachdev reported receiving personal fees from AstraZeneca, Celgene, Ipsen, PUMA, Novartis, Pfizer, and Tempus outside the submitted work. Dr Mitchell reported receiving grants from Genentech, Taiho Oncology, and Exelexis that were paid to her institution during the conduct of the study; serving on the Corvus Board of Directors; and receiving personal fees from Genentech Advisory Board, Bristol Myers Squib Advisory Board, and Novartis Advisory Board outside the submitted work. Dr Zwiebel reported receiving personal fees from Boston Pharmaceuticals for part-time consulting on unrelated products outside the submitted work. Dr Gray reported receiving grants from the National Cancer Institute (NCI) during the conduct of the study. Dr O'Dwyer reported receiving grants from the National Institutes of Health (NIH) during the conduct of the study. Dr Arteaga reported receiving grants from Lilly, Pfizer, and Takeda; serving in an expert advisory role to Novartis, Lilly, Immunomedics, Merck, Daiichi Sankyo, Taiho Oncology, AstraZeneca, and OrigiMed outside the submitted work; holding minor stock options in Y-TRAP and Provista; and serving in the Scientific Advisory Board of the Susan G. Komen Foundation. Dr Flaherty reported receiving grants from the ECOG-ACRIN during the conduct of the study; personal fees from Loxo Oncology, Clovis Oncology, Strata Oncology, Vivid Biosciences, Checkmate Pharmaceuticals, Kinnate, X4 Pharmaceuticals, PIC Therapeutics, Amgen, Asana Biosciences, Adaptimmune, Aeglea, Shattuck Labs, Tolero Pharmaceuticals, Apricity, Oncoceutics, Fog Pharma, Neon Therapeutics, Tvardi, xCures, Monopteros, Vibliome, Lilly, Genentech, Bristol Myers Squibb, Merck, Takeda, Verastem, Boston Biomedical, Pierre Fabre, and Debiopharm outside the submitted work; and grants and personal fees from Sanofi and Novartis. No other disclosures were reported.

    Funding/Support: The NCI-MATCH trial was coordinated by the ECOG-ACRIN Cancer Research Group and was supported by grants U10CA180820, U10CA180794, UG1CA233339, UG1CA233341, UG1CA233329, UG1CA233302, and UG1CA233180 from the NCI of the NIH.

    Role of the Funder/Sponsor: The NCI and ECOG-ACRIN sponsored and funded the trial. AstraZeneca supplied capivasertib for the NCI-MATCH (Molecular Analysis for Therapy Choice) trial subprotocol EAY131-Y trial but had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation of the manuscript; and decision to submit the manuscript for publication. Before manuscript submission, AstraZeneca offered reviews and comments of the paper and approved the manuscript for submission.

    Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Mention of trade names, commercial products, or organizations does not imply endorsement by the US government.

    Meeting Presentation: The initial results of the NCI-MATCH trial were presented as an oral abstract during the plenary 1 session of the 30th EORTC (European Organisation for Research and Treatment of Cancer)-NCI-AACR (American Association for Cancer Research) Symposium; November 13-16, 2018; Dublin, Ireland.

    Data Sharing Statement: See Supplement 3.

    References
    1.
    Vivanco  I, Sawyers  CL.  The phosphatidylinositol 3-kinase AKT pathway in human cancer.   Nat Rev Cancer. 2002;2(7):489-501. doi:10.1038/nrc839 PubMedGoogle ScholarCrossref
    2.
    Mundi  PS, Sachdev  J, McCourt  C, Kalinsky  K.  AKT in cancer: new molecular insights and advances in drug development.   Br J Clin Pharmacol. 2016;82(4):943-956. doi:10.1111/bcp.13021 PubMedGoogle ScholarCrossref
    3.
    Carpten  JD, Faber  AL, Horn  C,  et al.  A transforming mutation in the pleckstrin homology domain of AKT1 in cancer.   Nature. 2007;448(7152):439-444. doi:10.1038/nature05933 PubMedGoogle ScholarCrossref
    4.
    Stemke-Hale  K, Gonzalez-Angulo  AM, Lluch  A,  et al.  An integrative genomic and proteomic analysis of PIK3CA, PTEN, and AKT mutations in breast cancer.   Cancer Res. 2008;68(15):6084-6091. doi:10.1158/0008-5472.CAN-07-6854 PubMedGoogle ScholarCrossref
    5.
    Khoury  JD, Wang  WL, Prieto  VG,  et al.  Validation of immunohistochemical assays for integral biomarkers in the NCI-MATCH EAY131 clinical trial.   Clin Cancer Res. 2018;24(3):521-531. doi:10.1158/1078-0432.CCR-17-1597 PubMedGoogle ScholarCrossref
    6.
    She  QB, Halilovic  E, Ye  Q,  et al.  4E-BP1 is a key effector of the oncogenic activation of the AKT and ERK signaling pathways that integrates their function in tumors.   Cancer Cell. 2010;18(1):39-51. doi:10.1016/j.ccr.2010.05.023 PubMedGoogle ScholarCrossref
    7.
    Kalinsky  K, Jacks  LM, Heguy  A,  et al.  PIK3CA mutation associates with improved outcome in breast cancer.   Clin Cancer Res. 2009;15(16):5049-5059. doi:10.1158/1078-0432.CCR-09-0632 PubMedGoogle ScholarCrossref
    8.
    Razavi  P, Chang  MT, Xu  G,  et al.  The genomic landscape of endocrine-resistant advanced breast cancers.   Cancer Cell. 2018;34(3):427-438.e6. doi:10.1016/j.ccell.2018.08.008PubMedGoogle ScholarCrossref
    9.
    Davies  BR, Greenwood  H, Dudley  P,  et al.  Preclinical pharmacology of AZD5363, an inhibitor of AKT: pharmacodynamics, antitumor activity, and correlation of monotherapy activity with genetic background.   Mol Cancer Ther. 2012;11(4):873-887. doi:10.1158/1535-7163.MCT-11-0824-T PubMedGoogle ScholarCrossref
    10.
    Davies  BR, Guan  N, Logie  A,  et al.  Tumors with AKT1E17K mutations are rational targets for single agent or combination therapy with AKT inhibitors.   Mol Cancer Ther. 2015;14(11):2441-2451. doi:10.1158/1535-7163.MCT-15-0230 PubMedGoogle ScholarCrossref
    11.
    Ribas  R, Pancholi  S, Guest  SK,  et al.  AKT antagonist AZD5363 influences estrogen receptor function in endocrine-resistant breast cancer and synergizes with fulvestrant (ICI182780) in vivo.   Mol Cancer Ther. 2015;14(9):2035-2048. doi:10.1158/1535-7163.MCT-15-0143 PubMedGoogle ScholarCrossref
    12.
    Mullard  A.  NCI-MATCH trial pushes cancer umbrella trial paradigm.   Nat Rev Drug Discov. 2015;14(8):513-515. doi:10.1038/nrd4694 PubMedGoogle ScholarCrossref
    13.
    Eisenhauer  EA, Therasse  P, Bogaerts  J,  et al.  New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1).   Eur J Cancer. 2009;45(2):228-247. doi:10.1016/j.ejca.2008.10.026 PubMedGoogle ScholarCrossref
    14.
    Lih  CJ, Harrington  RD, Sims  DJ,  et al.  Analytical validation of the next-generation sequencing assay for a nationwide signal-finding clinical trial: Molecular Analysis for Therapy Choice clinical trial.   J Mol Diagn. 2017;19(2):313-327. doi:10.1016/j.jmoldx.2016.10.007 PubMedGoogle ScholarCrossref
    15.
    Hyman  DM, Smyth  LM, Donoghue  MTA,  et al.  AKT inhibition in solid tumors with AKT1 mutations.   J Clin Oncol. 2017;35(20):2251-2259. doi:10.1200/JCO.2017.73.0143 PubMedGoogle ScholarCrossref
    16.
    Smyth  L, Oliveira  M, Ciruelos  E,  et al.  Abstract P5-21-32: AZD5363 in combination with fulvestrant in AKT1-mutant ER-positive metastatic breast cancer.   Cancer Res. 2018;78(4)(suppl):5-21. doi:10.1158/1538-7445.SABCS17-P5-21-32Google Scholar
    17.
    Smyth  LM, Zhou  Q, Nguyen  B,  et al; AACR Project GENIE Consortium.  Characteristics and outcome of AKT1E17K-mutant breast cancer defined through AACR Project GENIE, a clinicogenomic registry.   Cancer Discov. 2020;10(4):526-535. doi:10.1158/2159-8290.CD-19-1209 PubMedGoogle ScholarCrossref
    18.
    Turner  NC, Alarcón  E, Armstrong  AC,  et al.  BEECH: a dose-finding run-in followed by a randomised phase II study assessing the efficacy of AKT inhibitor capivasertib (AZD5363) combined with paclitaxel in patients with estrogen receptor-positive advanced or metastatic breast cancer, and in a PIK3CA mutant sub-population.   Ann Oncol. 2019;30(5):774-780. doi:10.1093/annonc/mdz086 PubMedGoogle ScholarCrossref
    19.
    Jones  RH, Carucci  M, Casbard  AC,  et al.  Capivasertib (AZD5363) plus fulvestrant versus placebo plus fulvestrant after relapse or progression on an aromatase inhibitor in metastatic ER-positive breast cancer (FAKTION): a randomized, double-blind, placebo-controlled, phase II trial.   J Clin Oncol. 2019;37(15)(suppl):1005. doi:10.1200/JCO.2019.37.15_suppl.1005Google ScholarCrossref
    20.
    Altomare  DA, Testa  JR.  Perturbations of the AKT signaling pathway in human cancer.   Oncogene. 2005;24(50):7455-7464. doi:10.1038/sj.onc.1209085 PubMedGoogle ScholarCrossref
    21.
    Banerji  S, Cibulskis  K, Rangel-Escareno  C,  et al.  Sequence analysis of mutations and translocations across breast cancer subtypes.   Nature. 2012;486(7403):405-409. doi:10.1038/nature11154 PubMedGoogle ScholarCrossref
    22.
    Mosquera  JM, Varma  S, Pauli  C,  et al.  MAGI3-AKT3 fusion in breast cancer amended.   Nature. 2015;520(7547):E11-E12. doi:10.1038/nature14265 PubMedGoogle ScholarCrossref
    23.
    Parikh  C, Janakiraman  V, Wu  WI,  et al.  Disruption of PH-kinase domain interactions leads to oncogenic activation of AKT in human cancers.   Proc Natl Acad Sci U S A. 2012;109(47):19368-19373. doi:10.1073/pnas.1204384109 PubMedGoogle ScholarCrossref
    ×