Telaglenastat plus Everolimus in Advanced Renal Cell Carcinoma: A Randomized, Double-Blinded, Placebo-Controlled, Phase II ENTRATA Trial

Clear cell renal cell carcinoma (RCC) is the most aggressive form of RCC and represents ∼75% of all cases (1). In the first- and second-line setting, systemic therapy often includes use of a tyrosine kinase inhibitor (TKI), targeting VEGFR, and/or checkpoint inhibitor (CPI), either as monotherapy or in combination (2). In the third-line setting, evaluation of treatment regimens of 1,012 patients with metastatic RCC (mRCC) across 25 international cancer centers from 2005 to 2015 showed that everolimus monotherapy was the most prevalent therapy, with an associated median progression-free survival (PFS) of 3.7 months (3). After progression in the third-line, treatment options are limited and often include agents targeting similar pathways. Few studies have explored new therapeutic agents in heavily pretreated patients (3).

Many cancers are characterized by altered glucose and glutamine metabolism that manifest as increased glycolysis (i.e., Warburg effect) and glutamine utilization (4, 5). As glucose is diverted away from the tricarboxylic acid (TCA) cycle, the increased demand for an anapleurotic source of carbon is met by glutamine, which is converted into glutamate and α-ketoglutarate (a TCA cycle intermediate) through the enzymatic activity of glutaminase and glutamate dehydrogenase, respectively (6, 7). Telaglenastat (CB-839) is an investigational, first-in-clinic, small molecule, reversible oral inhibitor of glutaminase. In preclinical studies, telaglenastat showed antiproliferative activity across a wide range of tumor cells and demonstrated antiproliferative, pro-apoptotic activity in several solid tumor models [e.g., non–small cell lunch cancer (NSCLC)], RCC, breast cancer, myeloma], both alone and in combination with other anticancer therapies (8, 9). Clear cell RCC has been shown to have elevated expression of glutaminase and is sensitive to glutamine withdrawal, knockdown of glutaminase gene expression, and inhibition of glutaminase activity (8, 9).

Everolimus is an inhibitor of the mTOR, a master regulator of metabolic pathways, playing central roles in lipid and nucleotide synthesis, mitochondrial biogenesis, metabolic homeostasis, and glycolysis, while suppressing catabolic pathways such as autophagy, lysosome biogenesis, and proteasome assembly (10, 11). Everolimus has well-documented inhibitory effects on glucose metabolism (10, 11). In normal cells, glucose undergoes glycolysis and is further metabolized in the TCA cycle to generate energy. However, in tumors, cells undergo a metabolic shift where growth factors drive abnormal glucose metabolism. This abnormal glucose metabolism, which occurs even under aerobic conditions, is also known as the Warburg effect and generates lactate while depriving the TCA cycle of a key source of carbon (in the form of acetylCoA). To compensate for this metabolic shift, glutamine metabolism is increased to sustain the TCA cycle for growth and proliferation. Although the Warburg effect has been a well-described phenomenon, to date there have not been agents that specifically target this metabolic shift. mTOR inhibitors, such as everolimus, inhibit glucose utilization and lactate production (12).

Combination of telaglenastat with everolimus represents a dual-targeted strategy that inhibits two key tumor metabolism pathways needed to support increased growth and proliferation of tumors (Supplementary Fig. S1). This inhibition leads to decreased extracellular acidification rates and oxygen consumption rates, which has demonstrated synergistic anticancer activity in preclinical studies, including synergistic anticancer effects in cell lines and in xenograft models of RCC (8, 9). These effects have translated to clinical observations in an early-stage trial; in a phase I cohort of patients with RCC receiving telaglenastat in combination with everolimus in the second- or later-line setting, the combination showed a tolerable safety profile with encouraging efficacy outcomes (13, 14).

To further explore whether addition of telaglenastat could enhance the anticancer effects of everolimus alone, the randomized, double-blind, placebo-controlled ENTRATA study compared telaglenastat plus everolimus (TelaE) versus placebo plus everolimus (PboE) as 3L+ treatment for patients with mRCC.

Eligible patients were ≥18 years of age and had advanced or mRCC with clear cell component, measurable disease per RECIST v1.1, at least two prior therapies including at least one prior VEGFR-targeting TKI, KPS ≥70%, and adequate organ function. Patients with prior mTOR inhibitor use were excluded. Treated/stable brain metastases were allowed. All patients provided written informed consent.

Patients were randomly assigned in a 2:1 ratio to receive either TelaE or PboE. Randomization was stratified as per the number of prior TKIs (1 or >1) and prognostic risk category (favorable or intermediate/poor) according to the Memorial Sloan Kettering Cancer Center (MSKCC) criteria. Telaglenastat was administered orally at a dose of 800 mg twice daily. Everolimus was administered orally at a dose of 10 mg once daily. Dose reductions for telaglenastat (600 mg, then 400 mg, then discontinue) and everolimus (according to package insert) were specified for management of adverse events (AE). Treatment was continued until disease progression or the development of unacceptable toxicity.

The primary endpoint was investigator-assessed PFS (per RECIST v1.1), defined as the time from randomization to disease progression or death due to any cause, whichever occurred first. The secondary endpoint was overall survival (OS; time from randomization to death due to any cause). Additional endpoints included overall response rate [ORR; percentage of patients with complete response (CR) or partial response (PR) per RECIST v1.1], duration of response [DoR; time between first documentation of a PR/CR to first documentation of progressive disease (PD) or death], and percentage of patients with CR, PR, and stable disease (SD) according to RECIST v1.1, documented at least 8 weeks posttreatment initiation.

Safety assessments conducted on the Safety Analysis Patient Set included assessment of number and proportion of patients reporting AEs by treatment group and by worst severity reported. Adverse events collected included those occurring during or after the first dose of the study drug and up to 28 days after the last dose.

The study was approved by the Institutional Review Boards or Ethics Committees at all participating sites and conducted in accordance with the ethics principles of the Declaration of Helsinki and the International Council of Harmonization Guidelines on Good Clinical Practice. The trial was registered at www.clinicaltrials.gov (NCT03163667).

For the primary endpoint of PFS, with 80% power, one-sided α of 0.2, and 2:1 randomization, approximately 48 events (progression or death) were required to detect a HR of 0.6 (minimum detectable HR = 0.77). An estimated 3.7-month median PFS for the everolimus control group was based on published data for patients with clear cell mRCC after two prior lines of therapy (15, 16). On the basis of an estimated sample size of 63 patients, the primary analysis was to be conducted when approximately 48 PFS events (PD or death) had been reached (data cutoff, April 26, 2019). The final OS analysis, which was not statistically powered, was performed after 42 OS events (data cutoff, September 30, 2020).

Data will be made available from the corresponding author upon reasonable request.

Between September 6, 2017 and January 28, 2019, 69 patients were randomly assigned to either the TelaE arm (n = 46) or the PboE arm (n = 23; Fig. 1). Median age was 65 years, 83% of patients were male, and 68% were categorized as having intermediate or poor MSKCC risk. Baseline demographics and disease characteristics were generally balanced between arms, with the exception that more patients in the TelaE arm had adrenal and lymph node metastasis (35% vs. 13% and 74% vs. 61%, respectively), compared with the PboE arm (Table 1). Patients in both arms were heavily pretreated, having received a median of three prior therapies in the advanced/metastatic setting; 23% of patients were on their fifth- or later-line of therapy. Seventy percent had received at least two prior TKIs, and 88% had prior anti-PD-(L)1 therapies.

By the data cutoff of the primary analysis (April 26, 2019), 39 patients had discontinued treatment in the TelaE arm, and 18 in the PboE arm. At final analysis (September 30, 2020), all patients had discontinued treatment. The most common reason for treatment discontinuation was disease progression, which occurred in 26 patients in the TelaE arm and 14 in the PboE arm.

The primary PFS analysis was conducted after 48 PFS events had occurred. Median PFS was 3.8 months (95% confidence interval (CI), 2.4–7.4 months) in the TelaE arm and 1.9 months (95% CI, 1.9–8.3 months) in the PboE arm (HR, 0.64; 95% CI, 0.34–1.20; P = 0.079; Table 2; Fig. 2A). Kaplan–Meier estimates of PFS probabilities at 3 months were 63.5% for TelaE and 37.0% for PboE. At 6 months, they were 39.1% and 31.7%, and at 9 months, they were 31.3% and 15.8%, respectively. The median duration of follow-up was 7.5 months. The improvement in PFS met the prespecified significance threshold of P ≤ 0.2 for PFS at the primary analysis.

Best overall responses were 1 (2.2%) PR and 26 (56.5%) SD in the TelaE arm, and 11 (47.8%) SD in the PboE arm; 16 patients (34.8%) had PD in the TelaE arm and 10 (43.5%) had PD in the PboE arm (Table 2; Supplementary Fig. S2). Correlative analyses between response and telaglenastat exposure were inconclusive due to the small sample size and high interpatient variability in telaglenastat PK (Supplementary Table S1).

The OS data had not reached maturity at the primary analysis data cutoff date. At final analysis (September 30, 2020), median OS was 14.4 months [95% CI, 7.9 months–not reached (NR)] in the TelaE arm and 9.7 months (95% CI, 8.2–25.4 months) in the PboE arm (HR, 0.80; 95% CI, 0.42–1.50; P = 0.2401; Table 2; Fig. 2B). Median duration of follow-up was 21.5 months.

The median duration of exposure to TelaE was 99.5 days (range, 2–752), and to PboE 81.0 days (range, 20–590). Overall incidence of AEs was 100% for both arms (Table 3). Grade 3 or 4 treatment-related AEs occurred in 34 (74%) patients in the TelaE arm and 14 (61%) patients in the PboE arm. AEs leading to treatment discontinuation (any study treatment) occurred in 12 (26%) patients in the TelaE arm and 8 (35%) patients in the PboE arm; these included two events of pneumonitis that led to discontinuation of everolimus (one grade 2 in the TelaE arm and one grade 4 in the PboE arm). AEs leading to everolimus dose interruptions or reductions occurred in 78% of patients in the TelaE arm and 65% in the PboE arm. AEs leading to telaglenastat or placebo dose interruptions or reductions occurred in 74% of patients in the TelaE arm and 65% in the PboE arm.

Rates of AEs were similar between arms, the most common including fatigue (48% TelaE vs. 52% PboE), anemia (41% vs. 35%), cough (35% vs. 44%), dyspnea (35% vs. 39%), serum creatinine increase (35% vs. 35%), and diarrhea (22% vs. 48%). The most common grade 3 or 4 AEs were anemia (20% vs. 17%), fatigue (4% vs. 9%), abdominal pain (9% vs. 0%), thrombocytopenia (7% vs. 0%), and hyperglycemia (7% vs. 4%; Table 4).

Fatal events occurred in 2 patients in the TelaE arm (cardio-respiratory arrest, respiratory distress) and in 1 patient in the PboE arm (cardio-respiratory arrest). None were deemed related to any study treatment.

As a proof-of-concept phase II study, ENTRATA was designed statistically to pick up signals of efficacy using a smaller than traditional sample size. Unlike two-sided α of 0.05 used for hypothesis testing, our study used a one-sided α of 0.20 (type I error rate). A one-sided log-rank P < 0.2 in favor of TelaE treatment is regarded a positive result worthy of further exploration. ENTRATA met its primary endpoint of improved PFS at primary analysis: the median PFS increased from 1.9 to 3.8 months and met the one-sided significance threshold of (P = 0.079) in patients with heavily pretreated metastatic clear cell RCC. Although it was an unpowered secondary endpoint, OS did trend in favor of the telaglenastat arm with a numerically higher median OS. Preclinically, telaglenastat is cytotoxic to several RCC cell lines and has synergistic anticancer effects when combined with everolimus (8). However, tumoricidal effects seen in cell culture or in vivo may not translate to what occurs in patients with RCC. The low response rate and high SD rate observed in this study is consistent with those from the phase I study in clear cell RCC patients [1/21 (5%) PR; 19/21 (90%) SD; ref. 14]. The lower rate of SDs in ENTRATA may be attributed to having sicker and more heavily pretreated patients than the population that was enrolled in the phase I study.

The overall safety profile showed that telaglenastat was generally well tolerated and did not appear to contribute substantially to the toxicity profile of everolimus alone. As the first randomized, controlled trial for telaglenastat, data from this study demonstrate proof-of-concept for telaglenastat in RCC, while building on findings from the phase I study that included cohorts of patients receiving telaglenastat monotherapy and combinations with signal transduction inhibitors, including everolimus.

Everolimus was initially approved for second- and third-line treatment of patients with advanced RCC, following the pivotal phase III RECORD-1 study that demonstrated an improvement in median PFS of ∼2 months compared with placebo in patients who previously had disease progression on sunitinib and/or sorafenib (17). For several years, single-agent everolimus was the dominant agent used for second-line treatment, until the regulatory approvals of nivolumab, cabozantinib, and lenvatinib plus everolimus, which have all demonstrated superiority to everolimus monotherapy (18–20). At the time the study was initiated, everolimus was the most common treatment for third- or later-line metastatic RCC, functioning via a distinct mechanism of action from the TKIs and CPIs that are used in earlier lines of therapy (3). New therapies have since been introduced to the RCC treatment landscape, offering alternative approaches with next-generation VEGFR TKIs (e.g., lenvantinib and tivozanib) and novel combinatorial therapies. For example, combination regimens of lenvantinib plus everolimus or pembrolizumab have recently demonstrated high PFS and OS rates in a phase III study of patients with advanced RCC treated in the first-line setting (21), although tivozanib secured FDA approval as monotherapy for relapsed or refractory RCC after two or more therapies (22). With the evolving RCC treatment landscape, everolimus has lost its initial popularity in the third-line setting, often being relegated to patients who are much sicker than those evaluated in pivotal trials. Nevertheless, although the RCC treatment landscape is entering an era where therapeutic options are growing rapidly, there remains a need to continue developing new agents with different mechanisms of action beyond TKIs, mTOR inhibitors, and immune-CPI therapies.

In this double blind, randomized controlled study, the control arm of PboE had a shorter PFS (1.9 months) compared with historical data for everolimus. This discrepancy may in part be due to differences in everolimus prognostic risk scores and treatment in a particularly very heavily pretreated population in the current trial, reflecting the changes in the RCC treatment landscape following the approvals of nivolumab and cabozantinib in the second- and third-line settings. Fifty-nine percent of patients received the study treatment on this trial in the fourth- or later-line settings (with 39% being fifth-line or beyond), more than two-thirds had received at least two prior TKIs, and nearly 90% had received prior CPIs. In other studies, the median PFS for everolimus has previously been reported at ∼3.7 months in the third-line setting for clear cell mRCC (3). Likewise, in the METEOR study in patients who had disease progression on a VEGFR-TKI, everolimus was associated with a median PFS of 3.8 months for the overall population, which decreased to 2.7 months in the subpopulation of patients with bone metastases at baseline (15, 23). Compared with the METEOR study, our study population had a higher proportion of patients with intermediate or poor MSKCC prognostic risk (68% vs. 54%; ref. 15), and more patients with very advanced disease—more than half having three or more sites of metastasis. In this study of heavily pretreated patients with predominately intermediate- and poor-risk disease, the addition of telaglenastat met its primary endpoint of prolonging PFS compared with everolimus alone based on prespecified statistical design. In addition, although not sufficiently powered to evaluate OS, the median OS trended higher in the TelaE arm than the PboE arm, supporting the clinical activity of dual metabolic inhibition with this combination regimen.

TelaE's safety profile and AE data were similar to the PboE arm (15), suggesting that telaglenastat did not contribute substantially to additional toxicity or safety concerns over that of everolimus alone. Few additional grade 3 or 4 toxicities were seen in the TelaE compared with the PboE arms. Adverse events leading to dose discontinuation were similar between the two arms (26% and 35% for TelaE and PboE, respectively). Fatigue was the most commonly reported AE in both the TelaE and PboE arms (48% and 52%, respectively). In the phase I study of TelaE, which evaluated this combination in earlier lines of therapy for patients with mRCC, fatigue was reported at a lower frequency (22%), perhaps reflecting a population of patients who were less ill than those enrolled in ENTRATA (13, 14).

Limitations of this study include its small sample size, in addition to the enrollment of heavily pretreated patients, the vast majority of whom had intermediate- or poor-risk disease. Although the phase I telaglenastat monotherapy cohort demonstrated high telaglenastat exposure (AUC: 19,900 ng ⋅h/mL, C_max 2,960 ng/mL; C_min 1,490 ng/mL) in a patient with clear cell RCC who achieved a PR (24), correlative analyses in ENTRATA were inconclusive due to high interpatient variability in PK parameters. Further exploration is warranted to understand the scientific basis for why ENTRATA met its primary endpoint, whereas CANTATA, another randomized phase II trial in patients randomized to cabozantinib plus telaglenastat versus cabozantinib plus placebo, did not show significant differences between treatment arms in outcomes of PFS or responses (25). Indeed, the partner agents for each study (everolimus and cabozantinib, respectively) have distinct mechanisms of action. It will be important to establish whether the dual metabolic targeting mechanism proposed for telaglenastat plus everolimus can be considered a viable treatment combination strategy for developing future regimens.

In conclusion, this randomized phase II trial demonstrated improvement in PFS for TelaE versus PboE in patients who were heavily pretreated (median fourth-line of therapy) and included those who developed refractory disease to multiple TKIs and CPI therapy. TelaE was well tolerated and telaglenastat did not contribute substantially to toxicities seen in the PboE arm. ENTRATA supports proof-of-concept to further assess glutaminase inhibition with telaglenastat as a novel strategy for the treatment of mRCC.

C.-H. Lee reports grants and nonfinancial support from Calithera during the conduct of the study. C.-H. Lee also reports grants and personal fees from BMS, Exelixis, Merck, and Pfizer; grants, personal fees, and nonfinancial support from Eisai; personal fees from EMD Serono, AiCME, Intellisphere, and Research to Practice; and grants from Calithera and Eli Lilly outside the submitted work. R. Motzer reports personal fees from Calithera during the conduct of the study; R. Motzer also reports personal fees from Genentech/Roche, Pfizer, Eisai, Exelixis, EMD Serono, Merck, and Aveo, as well as grants from Genentech/Roche, Pfizer, Eisai, Exelixis, Merck, Aveo, and Bristol Myers Squibb outside the submitted work. H. Emamekhoo reports personal fees from BMS, Exelixis, and Seattle Genetics outside the submitted work. M. Matrana reports personal fees from AstraZeneca, Seagen, Astellas, Bristol Meyers Squibb, Janssen, Merck, Amgen, Eisai, EMD Serono, and Pfizer outside the submitted work. J.J. Hsieh reports other support from Calithera during the conduct of the study. A. Hussain reports grants from Clovis during the conduct of the study, as well as personal fees from Bayer and Merck outside the submitted work. U. Vaishampayan reports personal fees from AAA, Aveo, Bayer, Sanofi, Gilead, and Pfizer, as well as grants and personal fees from BMS, Exelixis, and Merck outside the submitted work. S. Liu reports personal fees from Exelixis, EMD Serono, Seagen, and Eisai outside the submitted work. S. McCune reports other support from Calithera during the conduct of the study, as well as other support from Bristol Myers Squibb outside the submitted work. J. Bendell reports grants from Gilead, Genetech/Roche, BMS, Five Prime, Lilly, Merck, MedImmune, Celgene, EMD Serono, Taiho, Macrogenics, GSK, Novartis, OncoMed, LEAP, TG Therapeutics, AstraZeneca, Boehringer Ingelheim, Daiichi Sankyo, Bayer, Incyte, Apexigen, Koltan, SynDevRex, Forty Seven, AbbVie, Array, Onyx, Sanofi, Takeda, Eisai, Celldex, Agios, Cytomx Nektar, ARMO, Boston Biomedical, Ipsen, Merrimack, Tarveda, Tyrogenex, Oncogenex, Marshall Edwards, Mersana, Calithera, Blueprint, Evolo, FORMA, Merus, Jacobio, Effector, Novocare, Arrys, Tracon, Sierra, Innate, Arch Oncology, Prelude Oncology, Unum Therapeutics, Vyriad, Harpoon, ADC, Amgen, Pfizer, Millennium, Imclone, Acerta Pharma, Rgenix, Bellicum, Gossamer Bio, Arcus Bio, Seattle Genetics, TempestTx, Shattuck Labs, Synthorx, Inc., Revolution Medicines, Inc., Bicycle Therapeutics, Zymeworks, Relay Therapeutics, NGM Biopharma, Scholar Rock, Stemcentrx, Beigene, CALGB, Cyteir Therapeutics, Foundation Bio, Innate Pharma, Morphotex, OncXerna, NuMab, AtlasMedx, Treadwell Therapeutics, IGM Biosciences, Hutchinson MediPharma, MabSpace, REPARE Therapeutics, NeoImmune Tech, Regeneron, PureTech Health, Phoenix Bio, Cyteir, Molecular Partners, Innate, Torque, Tizona, Janssen, Tolero, Amgen, Moderna Therapeutics, Continuum Clinical, Agios, Pfizer, Samsung Bioepios, and Fusion Therapeutics during the conduct of the study, as well as other support from Gilead, Genentech/Roche, BMS, Five Prime, Lilly, Merck, MedImmune, Celgene, Taiho, Macrogenics, GSK, Novartis, OncoMed, LEAP, TG Therapeutics, AstraZeneca, Boehringer Ingelheim, Daiichi Sankyo, Bayer, Incyte, Apexigen, Array, Sanofi, ARMO, Ipsen, Merrimack, Oncogenex, FORMA, Arch Oncology, Prelude Therapeutics, Phoenix Bio, Cyteir, Molecular Partners, Innate, Torque, Tizona, Janssen, Tolero, Amgen, Moderna Therapeutics, Seattle Genetics, Beigene, Continuum Clinical, Tanabe Research Laboratories, Agios, Bicycle Therapeutics, Relay Therapeutics, Evelo, Pfizer, Samsung Bioepios, and Fusion Therapeutics outside the submitted work. A.C. Fan reports grants from Calithera during the conduct of the study. A.C. Fan also reports other support from Molecular Decisions Inc.; grants from Earli Inc., Vortex Inc., and Filtricine; and personal fees from Dendreon and Verily outside the submitted work. B.A. Gartrell reports other support from Janssen, Pfizer, and Aveo outside the submitted work. O.B. Goodman Jr reports personal fees from Exelixis outside the submitted work. A.R. Kaleblasty reports personal fees and other support from Exelixis, AstraZeneca, Bayer, Pfizer, Novartis, Genentech/Roche, BMS, EMD Serono, Gilead Sciences, Janssen, Eisai, and Seattle Genetics/Astellas, as well as personal fees from Sanofi, Amgen, Myovant Sciences, and Aveo outside the submitted work. Y. Zakharia reports other support from Bristol Myers Squibb, Amgen, Janssen, Eisai, Exelixis, Castle Bioscience, AstraZeneca, Array, Pfizer, Clovis, EMD Serono, and Myovant (advisory board); grant/research support from Pfizer (institution clinical trial support); Exelixis, Eisai, and Janssen Research and Development (DSMC); and Pfizer and Novartis (consultant honorarium). K. Orford reports personal fees from Calithera Biosciences during the conduct of the study, as well as personal fees from Calithera Biosciences outside the submitted work. N.M. Tannir reports grants from Calithera Bioscience during the conduct of the study; N.M. Tannir also reports grants from Bristol Myers Squibb, Nektar Therapeutics, Arrowhead Pharmaceuticals, Eisai, and Novartis, as well as personal fees from Oncorena, Bristol Myers Squibb, Pfizer, Nektar Therapeutics, Exelixis, Eisai Medical Research, Lilly, Novartis, Ipsen, and Merck Sharp & Dohme outside the submitted work. No disclosures were reported by the other authors..

C.-H. Lee: Conceptualization, investigation, writing–original draft, writing–review and editing. R. Motzer: Conceptualization, investigation, writing–original draft, writing–review and editing. H. Emamekhoo: Investigation, writing–review and editing. M. Matrana: Investigation, writing–review and editing. I. Percent: Investigation, writing–review and editing. J.J. Hsieh: Investigation, writing–review and editing. A. Hussain: Investigation, writing–review and editing. U. Vaishampayan: Investigation, writing–review and editing. S. Liu: Investigation, writing–review and editing. S. McCune: Investigation, writing–review and editing. V. Patel: Investigation, writing–review and editing. M. Shaheen: Investigation, writing–review and editing. J. Bendell: Investigation, writing–review and editing. A.C. Fan: Investigation, writing–review and editing. B.A. Gartrell: Investigation, writing–review and editing. O.B. Goodman Jr: Investigation, writing–review and editing. P.G. Nikolinakos: Investigation, writing–review and editing. A.R. Kalebasty: Investigation, writing–review and editing. Y. Zakharia: Investigation, writing–review and editing. Z. Zhang: Investigation, writing–review and editingmp;. H. Parmar: Formal analysis, funding acquisition, writing–review and editing. L. Akella: Data curation, formal analysis, funding acquisition, validation, writing–review and editing. K. Orford: Conceptualization, supervision, funding acquisition, writing–review and editing. N.M. Tannir: Conceptualization, investigation, writing–review and editing.

We thank the investigators and site staff and patients and their families for their participation in the study. Editorial support was provided by Ingrid Koo, PhD, and funded by Calithera Biosciences, Inc. This study was funded by Calithera Biosciences, Inc. Patients treated at MSKCC were supported in part by MSKCC Support Grant/Core Grant (P30 CA008748).

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