Phase I Study of mTORC1/2 Inhibitor Sapanisertib (CB-228/TAK-228) in Combination with Metformin in Patients with mTOR/AKT/PI3K Pathway Alterations and Advanced Solid Malignancies

Abstract Background: Sapanisertib (CB-228/TAK-228) is a potent, selective ATP-competitive, dual inhibitor of mTORC1/2. Metformin is thought to inhibit the mTOR pathway through upstream activation of 5′-AMP-activated protein kinase (AMPK) suggesting combination therapy may enhance antitumor activity of sapanisertib. We report preliminary safety, tolerability, and efficacy from the dose-escalation study of sapanisertib in combination with metformin in patients with advanced solid tumors. Methods: Patients with advanced metastatic solid tumors resistant or refractory to standard treatment, with and without mTOR/AKT/PI3K pathway alterations, received sapanisertib 3 or 4 mg daily together with metformin once to three times daily (500–1,500 mg). All patients underwent 14-day titration period for metformin in cycle 1. Tumor measurements were performed following cycle 2 and subsequently every 8 weeks. Results: A total of 30 patients were enrolled across four cohorts (3 mg/500 mg; 3 mg/1,000 mg, 4 mg/1,000 mg; 4 mg/1,500 mg). 19 were female (63%), median age was 57 (range: 30–77), all were Eastern Cooperative Oncology Group performance status 1. Tumor types included sarcoma (6), breast (4), ovarian (4), head and neck (3), colorectal (2), lung (2), renal cell (2), endometrial (2), gastroesophageal junction (1), prostate (1), stomach (1), urachus (1), and cervical cancer (1). Median number of prior lines of therapy was 4. Most common genomic alterations included PIK3CA (27%), PTEN (17%), AKT1/2 (10%), mTOR (10%). Of 30 patients evaluable for response, 4 patients achieved partial response (PR); 15 patients achieved stable disease (SD) as best response. Disease control rate (PR+SD) was 63%. Of the responders in PR, 3 of 4 patients had documented PTEN mutations (3/5 patients enrolled with PTEN mutations had PR); 2 of 4 of patients in PR had comutations (patient with leiomyosarcoma had both PTEN and TSC; patient with breast cancer had both PTEN and STK11); 1 of 4 patients in PR had AKT and mTOR mutation; tumor types included leiomyosarcoma (n = 2), breast (n = 1), and endometrial cancer (n = 1). Most common treatment-emergent adverse events included nausea, anorexia, diarrhea, and rash. Grade (G) 3–5 treatment-related adverse events included hyperglycemia (4/30; 13%), fatigue (2/30; 7%), hypertriglyceridemia (1/30; 3%), rash (2/20; 7%), diarrhea (2/30; 7%), creatinine increase (1/30; 3%), acidosis (1/30; 3%). No dose-limiting toxicities (DLT) were reported in the 3 mg/500 mg cohort. One of 6 patient had DLT in the 3 mg/1,000 mg cohort (G3 diarrhea) and 2 of 11 patients had DLTs in the 4 mg/1,500 mg cohort (G3 fatigue, G3 rash). 4 mg/1,000 mg was defined as the MTD. Conclusions: The safety profile of mTORC1/2 inhibitor sapanisertib in combination with metformin was generally tolerable, with antitumor activity observed in patients with advanced malignancies harboring PTEN mutations and AKT/mTOR pathway alterations. Significance: Sapanisertib (CB-228/TAK-228) is a potent, selective ATP-competitive, next-generation dual inhibitor of mTORC1/2. Metformin is thought to inhibit the mTOR pathway through upstream activation of AMPK suggesting combination therapy may enhance antitumor activity of sapanisertib. This dose-escalation study of sapanisertib and metformin in advanced solid tumors and mTOR/AKT/PI3K pathway alterations, demonstrates safety, tolerability, and early clinical activity in advanced malignancies harboring PTEN mutations and AKT/mTOR pathway alterations. Clinical trial information: NCT03017833


Introduction
The PI3K/v-Akt Murine Thymoma Viral Oncogene (AKT)/mTOR pathway plays a diverse role in regulation of several cellular functions such as cellular growth, proliferation, and survival (1).This pathway is frequently dysregulated in various cancers, such endometrial, cervical, lung, prostate, skin, and breast cancer (2)(3)(4), and mTOR is a key node and central regulator in this pathway.mTOR exists in two physically and functionally distinct protein signaling complexes, one with raptor, which is sensitive to the mTOR inhibitor rapamycin, and the other with rictor, which is rapamycin insensitive (5).The first, mTOR complex 1 (mTORC1), phosphorylates 4EBP1 (the eukaryotic translation initiation factor 4E-binding protein 1) and p70S6 kinase and results in translation of proteins involved in cell-cycle progression.The second, mTOR complex 2 (mTORC2), has been shown to directly phosphorylate and activate the upstream kinase AKT at serine 473, which promotes proliferation and survival pathways (6).The rapalogs, such as everolimus, exert their inhibitory action predominantly via mTORC1, while their inhibitory effect on mTORC2 is limited or weak (7)(8)(9)(10).Consequently, continued signaling through significant pathway feedback loops, results in upregulation of AKT, which has the undesirable effect of accelerating cell proliferation, antagonizing the antiproliferative effect of mTOR inhibition (11).Thus, next-generation dual inhibitors of mTORC1 and mTORC2 have been developed, and preclinical models have supported the potency of this dual inhibition strategy (12)(13)(14)(15).
Metformin, a common antidiabetic medication has been repurposed as antineoplastic agent to enhance the effect of chemotherapy in certain cancers (18).The mechanisms attributed to antihyperglycemic effects of metformin are manifold, including activation of 5 -AMP-activated protein kinase (AMPK), decreased production of cyclic AMP, suppression of mitochondrial complex I of the electron transport chain, glycerophosphate dehydrogenase targeting, and gut microbiome alterations (19).Mechanistically, metformin inhibits the mTOR pathway through upstream activation of AMPK, which results in phosphorylation and activation of the tumor suppressor gene TSC2.This decreases AKT activation, and exerts an inhibitory effect on mTOR (20).Preclinically, metformin-induced activation of AMPK has been shown to disrupt cross-talk between insulin/IGF-1R and G protein-coupled receptors signaling in pancreatic cancer cells, (21) and inhibit cell proliferation, reduce colony formation, and inhibit MAP kinase, AKT, and mTOR in estrogen receptor-positive andnegative, as well as ERBB2-normal and -overexpressing breast cancer cell lines

Study Design
This was an open-label, single-center investigator-initiated phase I clinical trial that employed a 3 + 3 dose-escalation design and was conducted at The University of Texas MD Anderson Cancer Center (MDACC; Houston, TX) per Institutional Review Board guidelines.The primary endpoint was to evaluate the safety and tolerability of sapanisertib in combination with metformin, to determine MTD and dose-limiting toxicities (DLT) of the combination in patients with advanced cancers refractory to standard therapy.The secondary objective was to evaluate preliminary antitumor efficacy of this treatment per RECIST version 1.1 (RECIST v1.1; ref. 27).Dose escalation included four dose levels.During dose escalation, patients received one of four dosing schedules: sapanisertib 3 or 4 mg once daily in combination with metformin one to three times daily (500 mg/1,000 mg/1,000 mg/1,500 mg; Fig. 1).The dosing was based on data from previous phase I studies with this agent.Study INK128-001 was a phase I, open-label study of single-agent TAK-228 administered to patients diagnosed with advanced solid malignancies, including, but not limited to, colorectal, renal, endometrial, and urothelial tumors.Four dosing schedules were evaluated (once daily, every week, once daily × 3 days every week, and once daily × 5 days every week).The MTDs for the four schedules investigated in INK128-001 were determined to be 6 mg for once daily dosing, 16 mg for once daily × 3 days every week dosing, 10 mg for once daily × 5 days every week dosing, and 40 mg for every week dosing.On the basis of these data and pharmacokinetic data, it was decided to use the Q Day dosing schema in combination with metformin.
All patients underwent 14-day titration period for metformin in cycle 1 to limit side effects and commenced sapanisertib on cycle 1 day 15; thereafter, treatment was administered in 4-week cycles.Tumor measurements were performed following cycle 2 and subsequently every 8 weeks.Data cut-off date was March 31, 2021.
DLT was defined if the events occurred within the days 15-42 of the first cycle.Definitions for DLTs, as predetermined in the study protocol, were: any grade ≥3 non-hematologic toxicity, except for the following: grade 3 hyperglycemia lasting ≤14 days (all patients should have received optimal antiglycemic treatment, including insulin, as clinically indicated); grade 3 rash lasting ≤3 days (all patients should have received topical steroid treatment, oral antihistamines, and oral steroids, if indicated); inadequately treated grade 3 nausea and/or vomiting and grade 3 diarrhea (all patients should have received optimal antiemetic and/or antidiarrheal prophylaxis and/or treatment as indicated); grade 4 neutropenia lasting >7 days in the absence of growth factor support; any grade neutropenia of any duration accompanied by fever ≥38.5°C and/or systemic infection; any other ≥ grade 4 hematologic toxicity; inability to administer at least 75% of planned doses of TAK-228 within cycle 1, due to study drug-related

Patients
The study accrual period was from March 12, 2018 to 2020.Eligible patients included patients with metastatic or advanced solid tumors not amendable to standard therapy, an Eastern Cooperative Oncology Group (ECOG) performance status (PS) 0-1 and adequate hematologic, hepatic, and renal function.
Patients had to have the ability to swallow oral medications.Diabetic patients with well-controlled diabetes (i.e., normal range HbA1C ≤7%) were permitted and allowed to be on antidiabetic treatment other than metformin.Notable exclusion criteria included poorly controlled diabetes mellitus [defined as glycosylated hemoglobin (HbA1c) >7%], prior treatment with dual PI3K/mTOR inhibitors, TORC1/2 inhibitors or TORC1 inhibitors, significant cardiovascular or pulmonary disease, intercurrent uncontrolled illness, patients receiving corticosteroids (either intravenous or oral steroids, excluding inhalers or lowdose hormone replacement therapy) within 1 week before administration of the first dose of study drug, and malabsorption due to prior gastrointestinal (GI) surgery, GI disease (e.g., patients with enteric stomata were excluded).Patients with treated, stable brain metastases (defined as no evidence of disease progression for ≥3 months before the first dose of study drug, no hemorrhage after treatment, off-treatment with dexamethasone for 4 weeks before administration of the first dose of sapanisertib, no ongoing requirement for dexamethasone or antiepileptic drugs) were permitted.

Assessments
Patients who received at least one dose of any study drug were included in the safety population.DLT-evaluable population was defined as all patients who either experienced DLT during cycle 1 or completed treatment with at least 75% of the planned doses of any study drug in cycle 1 and had sufficient follow-up data to allow the investigators to determine whether DLT occurred.The responseevaluable population was defined as all patients who had measurable disease according to RECIST v1.1, who had received at least one dose of any study drug, and who have at least one available postbaseline response assessment as per RE-CIST v1.1.Response was assessed according to the RECIST v1.1 after every two treatment cycles.Adverse events (AE) were assessed using the NCI Common Terminology Criteria for Adverse Events, version 4.03.

Statistical Analysis
No formal hypotheses were tested, and analyses were descriptive and exploratory.Nonparametric correlations were determined with Spearman rank correlation coefficient.

Ethical Approval and Consent to Participate
The protocol was approved by the Institutional Review Board at The University of Texas MDACC (Houston, TX).The study was conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonization Good Clinical Practice guidelines.All patients provided written informed consent before enrollment.

DLTs and MTD Determination
Dose escalation, DLTs, and MTDs are summarized in Supplementary Table S1; dose was escalated to level 4. No DLTs were reported in the first cohort treated, dose level 1 (3 mg/500 mg).One patient (1/6) experienced DLT of grade 3 diarrhea in the second cohort, dose level 2 (sapanisertib 3 mg/ metformin 1,000 mg), which was managed supportively with no dose reductions.There were no DLTs in dose level 3. Two further patients (2/11) had DLTs in dose level 4 (sapanisertib 4 mg/metformin 1,500 mg cohort), one patient with grade 3 fatigue, which required two dose-level reductions for sapanisertib; one patient with grade 3 rash, which required reduced dose level on restarting medication.Both these patients were able to continue treatment after dose modification.Of 30 treated patients, 4 patients (13.3%) had dose changes with either drug.Three patients (10%) had dose reduction with sapanisertib.On the basis of these findings, sapanisertib 4 mg in combination with metformin 1,000 mg was defined as the MTD and recommended phase II dose.

Discussion
This open-label single-institution phase I trial studied the safety and tolerability of the sapanisertib and metformin in combination with patients with advanced refractory cancer and AKT/mTOR/PI3K alterations.The safety profile of mTORC1/2 inhibitor sapanisertib in combination with metformin was generally tolerable, and toxicities observed were attributable to mTOR inhibition consistent with previously published literature.In addition, the combination of sapanisertib 4 mg plus metformin 1,000 mg was defined as the MTD and recommended phase II dose.
We demonstrate strong preliminary antitumor activity using sapanisertib in combination with metformin in a number of dose levels in molecularly selected patients of a range of tumor types:  Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase.
mutations were more sensitive and responded better to PI3K inhibitors than those with a single mutation.It is possible that this "double mutation" in the same pathway is functional similarly in other tumors beyond breast cancer.
There are several limitations of this study, including the small number of patients treated, and the inclusion of many patients with comutations in addition to mTOR/AKT/PI3K pathway alterations, which may affect the underlying oncogenic driver pathway.Regardless, we show clinical activity in heavily pretreated patients which suggest that further study of sapanisertib in combination with metformin is warranted, particularly in patients with advanced malignancies harboring PTEN mutations and AKT/mTOR pathway alterations.This regimen could prove to be a highly effective treatment option, and a phase II study is warranted at the recommended phase II dose.In conclusion, sapanisertib had a manageable safety profile and the combination of sapanisertib and metformin was tolerable and exhibited preliminary therapeutic activity.

FIGURE 3
FIGURE 3Oncoplot showing mutations and co-occurring alterations in the patients enrolled in this clinical trial.

TABLE 1
Baseline patient demographics ECOG PS 1. Median number of prior lines of therapy was 4 and 70% of patients (n = 21) and more than three lines of therapy.

TABLE 2
Tumor response according to RECIST v1.1 (investigator assessment) in response-evaluable patients Two patients who had progression due to unequivocal progression in nontarget lesions did not have a tumor size measurement.Tumor response across cohorts is summarized in Table 2. Of the 24 evaluable patients, 19 patients (79%) had documented disease control [defined as stable disease (SD) or partial response (PR) per RECIST v1.1 of more than 16 weeks].Notably, 4 patients (17%) achieved PR as best tumor response; these responses were in patients

TABLE 3
AEs.Most common AEs are summarized by severity, severity based on NCI CTCAE criteria version 4.03