First-in-human phase I dose escalation study of MK-8033 in patients with advanced solid tumors

Abstract

Background

C-Met, which is frequently activated in multiple cancers, has been implicated in tumor formation, progression, metastasis, angiogenesis, and resistance to multiple therapies. MK-8033 is a small-molecule inhibitor of c-Met that binds preferentially to the activated conformation, and has demonstrated anti-tumor activity in preclinical models. This first-in-human trial was performed to establish the safety and maximum tolerated dose (MTD), as well as preliminary pharmacokinetics (PK) and clinical activity.

Methods

Forty-seven patients were enrolled in three parts. The primary objective of Parts A and B was safety, whereas Part C evaluated the effect of proton-pump inhibitors on MK-8033 absorption. Dose escalation used an accelerated continual reassessment method, and dose-limiting toxicities (DLTs) were any treatment-related, first course nonhematologic grade ≥ 3 toxicity (except alopecia or inadequately treated nausea/vomiting/diarrhea), grade 4 hematologic toxicity (except grade 3 neutropenic fever and thrombocytopenia), or toxicity where treatment is held >3 weeks.

Results

Forty-six patients were treated across nine dose levels, and the MTD was 750 mg twice daily. DLTs were fatigue, nausea, vomiting, transaminitis, and hypokalemia. Most frequent toxicities were fatigue (28.3%), nausea (21.7%), and alopecia (19.6%), predominately grade ≤ 2. One patient with endometriod adenocarcinoma achieved a partial response and eight had stable disease. Median progression-free survival (PFS) was 57 days. Strikingly, the PFS for the one responder was 846 days. PK results showed that proton-pump inhibitors have no effect on MK-8033 absorption.

Conclusion

MK-8033 was well tolerated with no significant toxicity issues, albeit with limited clinical activity. Unfortunately, the company decided to discontinue further clinical development of MK-8033.

Introduction

Receptor tyrosine kinases (RTKs) function to regulate key cellular processes, including tissue development and homeostasis. C-Met, also known as hepatocyte growth factor receptor (HGFR), is a prototypical RTK that is largely expressed in epithelial/endothelial cells, and is essential for embryonic development, organogenesis, and wound healing [1–3]. The c-Met RTK family is structurally different than other RTK families and is the only high-affinity receptor for hepatocyte growth factor (HGF) [4, 5]. On binding with HGF, c-MET dimerizes, autophosphorylates, and activates downstream signaling pathways, including the MAPK and PI3K pathways [6]. Receptor amplification and mutation in the gene that encodes c-Met (i.e., MET) have been described in various malignancies [3], and aberrant c-Met signaling has been implicated in oncogenesis and tumor progression [3, 7]. Due to the vast array of information supporting a role of c-Met and HGF in the pathogenesis of several human cancers, c-Met has become an attractive oncogenic target [8].

The presence of genetic alterations of the MET gene, such as activating mutations and gene amplification, suggest that some tumors may be especially dependent on the MET oncogene. As c-Met overexpression is also associated with poor prognosis [9–14], it has been hypothesized that c-Met overexpression is required for the maintenance of the malignant state. Experiences with other small molecule inhibitors, including ABL inhibitors in chronic myeloid leukemia [15] and KIT inhibitors in gastrointestinal stromal tumors [16, 17], suggest that targeted agents can have a pronounced efficacy in the oncogene “addicted” context. Therefore, inhibitors of c-Met may demonstrate considerable efficacy if the maintenance of the tumor is dependent on c-Met activity, including those tumors with MET genetic alterations.

MK-8033 is a potent (IC50 = 1 nM) small-molecule inhibitor of the HGF/c-Met axis. In cell culture models, MK-8033 has broad anti-proliferative activity with sub-micromolar IC50 for proliferation in a number of cell lines, with the most pronounced activity in gastric and non-small cell lung cell lines. In GTL-16 gastric cancer cells that harbor a constitutively activated c-MET, MK-8033 potently inhibited autophosphorylation of c-Met in the C-terminal docking site, as well as inhibited downstream signaling events such as ERK1/2 and AKT [18]. In the A549 non-small cell lung carcinoma cell line, MK-8033 inhibited HGF-induced phosphorylation of c-Met, as well as phosphorylation of AKT and ERK1/2. In the GTL-16 xenograft, MK-8033 inhibited pathway activation and tumor growth at twice-daily doses of 100 mg/kg administered orally [19]. Additionally, this dose appears to be well-tolerated in mice, and achieved plasma concentrations exceeding the IC50 for target inhibition for 24 h/day. Based on the preclinical data that suggested efficacy of MK-8033, we designed this first-in-human study to evaluate the safety and tolerability of MK-8033 in patients with metastatic or locally advanced solid tumors.

Patients and methods

Patient eligibility criteria

All patients signed informed consent forms according to institutional guidelines. The trial protocol (NCT00559182) was approved by the independent ethics committee for each institution, and was conducted in accordance with the Declaration of Helsinki.

Eligible patients had histologically confirmed metastatic or locally advanced solid tumors for which no standard therapy existed. To be enrolled in Part B, patients must have demonstrated c-Met positivity by immunohistochemistry. Other requirements included: Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 1, age ≥ 18 years, absolute neutrophil count ≥1500/mcL, platelets ≥100,000/mcL, hemoglobin ≥10 g/dL, serum creatinine ≤1.5 × upper limit normal (ULN), bilirubin ≤ ULN or if > ULN, direct bilirubin must have been within normal range (i.e., Gilbert’s disease), alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphatase ≤2.5 × ULN or 5 × ULN if liver disease, and prothrombin time and partial thromboplastin ≤1.2 × ULN.

Exclusion criteria included: prior bisphosphonate therapy in the last six months; prior anti-cancer therapy within the preceding four weeks; significant history of cardiac disease; brain metastases unless stable for three months prior to study entry; primary central nervous system tumors; clinically significant ascites or pleural effusion; received therapy with a proton-pump inhibitor (PPI, only those patients in Part A and B), histamine₂-receptor antagonist, or antacid within one week of therapy; concurrent serious infection or other uncontrolled medical condition.

Study design

This was a multicenter, open-label, three part (Part A, B, and C) Phase I study that used an accelerated dose-escalation design. The primary objective of parts A and B was to evaluate the safety and tolerability of MK-8033 and determine the recommended phase II dose (RP2D) based on safety and pharmacokinetic (PK) parameters. The primary objective of part C was to evaluate the impact of omeprazole on MK-8033 PKs at steady state using a new acidified formulation that is designed to improve PK performance in PPI-treated patients. Anti-tumor activity was evaluated as an exploratory objective.

Study assessments

Patients were monitored for adverse events, and severity and attribution were recorded using the Common Terminology Criteria for Adverse Events (CTCAE) v3.0. Serum chemistries (including liver enzymes) and hematology parameters were measured weekly in the first cycle and on day 1 of subsequent cycles. Dose limiting toxicity (DLT) was defined as any first cycle grade 4 neutropenia, neutropenic fever, grade 3 or 4 thrombocytopenia, any grade 3 or 4 nonhematologic toxicity other than alopecia, inadequately treated diarrhea, nausea, or vomiting, grade 3 transaminitis lasting longer than 1 week or any grade 4 transaminitis, or therapy held for more than three weeks for any drug-related toxicity. Patients continued receiving study treatment until disease progression, withdrawal of patient consent, or unacceptable toxicity.

Safety and efficacy data were summarized for all patients that received at least one dose of study drug. Patients were assessed by radiographic evaluation at the end of Cycle 4, and every two cycles thereafter using The Response Evaluation Criteria in Solid Tumors (RECIST) version 1.0.

Plasma levels of MK-8033 were measured from PK blood samples obtained during the first course of therapy on Cycle 1 Day 1 at predose, 1, 1.5, 2, 3, 4, 6, 8, 12, and 24 h after the first dose of study medication (24 h collection was prior to first dose on Day 2). PK samples were also collected prior to dosing on Days 4 and 8. Repeated samples were also obtained on Day 14 of Cycle 1 at the same timepoints as day 1, with subsequent samples collected on Days 16 and 17. For patients in Part C, blood samples were collected at the same time points for up to 24 h post-dose on the last day of Period 1 (Day 8) and for up to 12 h post-dose on the last day of Period 2 (Day 15).

Accelerated dose escalation

In Parts A and B, sequential cohorts were treated using an accelerated dose escalation design (Supplemental Table 1). The starting dose was 50 mg administered twice daily (BID) in the first 14-day cycle followed by a 1-week drug holiday. After the 1-week drug holiday, patients continued receiving MK-8033 at the same dose level, for subsequent cycles of 14 days (Cycles 2 to 4) and 28 days (Cycle 5 and beyond). If treatment-related toxicities had resolved to Grade ≤ 2 at the end of the drug holiday, the patient proceeded to the next cycle. For all subsequent cycles, patients progressed to the next cycle of treatment without a drug holiday if treatment-related toxicities were Grade ≤ 2.

In Part A, doses were initially escalated by 100% and proceeded in single or expanded to 3 to 6 patient cohorts. Part B of the study began when either the first instance of a DLT occurs in any cycle, or the second instance of a drug-related Grade 2 toxicity occurs in any cycle. In Part B, doses were escalated by 40% in cohorts of 3 to 6 patients until a maximally tolerated dose (MTD) was determined. Dose escalation stopped when 2 of 3 or 2 of 6 patients (>30% in a single cohort) experienced a DLT in the first cycle. The initial MTD was the next lowest dose level at which no more than 1 of 6 patients experienced a first-cycle DLT. Once the MTD was identified, cohorts were expanded to 10 patients by enrolling 4 more patients.

After the MTD expansion was completed, as many as 8 more patients could be enrolled sequentially using the Time-to-Event Continual Reassessment Method (TITE-CRM), which is a technique used to find the dose (considered the RP2D) at or below the MTD that can be safely administered for at least three-cycles with a target DLT rate of 20%. Safety and tolerability obtained in all cycles from patients assigned at all dose levels was used to determine the RP2D using TITE-CRM. The RP2D was thus based on both acute and chronic toxicity data obtained from all patients enrolled.

Part C was a balanced crossover study design evaluating co-administration of 20 mg omeprazole administered two hours before MK-8033 at 770 mg BID or MK-8033 770 mg BID alone. Regardless of treatment assignment, patients received 8 days of dosing in Period 1 (the Day 8 PM dose of MK-8033 was not administered) and 7 days of dosing in Period 2. At the end of Period 2 patients were given the option to continue MK-8033.

Statistical design

This was an accelerated dose escalation phase I study with a cohort expansion at the MTD. If less than 2 of the 10 patients had a DLT, the upper bound of the 80% confidence interval for the DLT rate excluded 33%, rendering it an acceptable dose for further investigation. After determining the MTD, additional patients were enrolled sequentially based on a TITE-CRM [20]. The goal was to identify a dose at or below the MTD that could be safely used for at least three cycles, which would be the RP2D.

For the TITE-CRM, we assumed the dose response to be:

$$\varphi \left(d_j;\alpha \right)=d_j{e}^{\alpha },$$ (1)

where α is assumed to have a normal prior with mean 0 and variance 1.34 (suggested by the authors of [20]), and d_i is the initial estimate of the DLT rate at dose level i. It is assumed that up to 11 dose levels will be investigated in this study. Supplemental Table 1 provides the initial estimates for actual dose levels used in the study. Let y_j be an indicator of DLT (1/0: yes/no) and μ_j be the follow-up time within the observation window T. The likelihood function for α is:

$$L(\alpha )=\prod _{j=1}^n{\left\{w_j\varphi \left(d_j;\alpha \right)\right\}}^{y_i}{\left\{1-w_j\varphi \left(d_j;\alpha \right)\right\}}^{1-y_j},$$ (2)

where w_j = μ/T if y_j = 0, and w_j = 1 if y_j = 1. The target DLT for RP2D was 20%. Patients were enrolled to the dose level with an estimated DLT rate closest to 20%.

Results

Patient demographics

Between December of 2007 through June of 2010, a total of 47 patients were enrolled in the three-part study at five participating centers. Baseline demographics for all patients along with the number of patients enrolled at each dose level are summarized in Table 1. Forty-four patients were enrolled in Parts A and B, while three patients were enrolled in the crossover clinical pharmacology study (Part C). The median age was 65 (range: 30 to 87), and 24 patients were male. Forty patients were white, while the other seven were Asian. A range of primary malignancies was included in the study: colorectal cancer (n = 29); non-small cell lung cancer (NSCLC, n = 3); head & neck cancer, ovarian cancer, pancreatic cancer, or a tumor of unknown primary (each n = 2); and breast cancer, endocervical cancer, endometrial adenocarcinoma, endometrioid adenocarcinoma, esophageal cancer, transitional cell of the bladder, tcc of the ureter, hemangio pericytoma, or melanoma (each, n = 1). Among the 47 patients enrolled, 46 received at least one dose of MK-8033, and of those patients, 36 discontinued for progressive disease, eight for adverse events, and three withdrew consent after starting therapy.

Table 1.

Baseline demographics of all patients enrolled on study

	n	%
Patients in Population	47
Gender
M	24	51.1
F	23	48.9
Age (Years)
Median	65
Range	30–87
Race
Caucasian	40	85.1
Asian	7	14.9
Study Part and Dosages (BID)
Part A	44	93.6
50 mg	1	2.1
100 mg	1	2.1
Part B
200 mg	3	6.4
400 mg	4	8.5
550 mg	7	14.9
750 mg	18	38.3
1050 mg	6	12.8
1500 mg	3	6.4
Withdrew prior to dosing	1	2.1
Part C (770 mg)	3	6.4
Primary Malignancy
Colon Cancer	22	46.8
Rectal Cancer	5	10.6
Non-Small Cell Lung Cancer	3	6.4
Pancreatic Cancer	2	4.3
Head & Neck Cancer	2	4.3
Ovarian Cancer	2	4.3
Endometrial Adenocarcinoma	2	4.3
Breast Cancer	1	2.1
Endocervical Cancer	1	2.1
Esophageal Cancer	1	2.1
Hemangiopericytoma	1	2.1
Melanoma	1	2.1
Transitional Cell of The Bladder	1	2.1
TCC of the Ureter	1	2.1
Unknown Primary	2	4.3

Toxicity

The safety of MK-8033 was examined in 46 patients at 9 dose levels between the range of 100 mg/daily to 3000 mg/daily. Six patients had at least one Cycle 1 DLT: 550 mg (n = 1), 750 mg (n = 1), 1050 mg (n = 2) and 1500 mg (n = 2) (Table 2). The DLTs associated with MK-8033 were: grade 3 fatigue, grade 2 and 3 nausea, grade 2 and 3 vomiting, grade 3 elevated ALT, AST, blood amylase, and grade 4 hypokalaemia. From dose escalation, the MTD was determined to be dose level 6 (i.e., MK-8033 750 mg BID). An additional eight patients were then enrolled sequentially using the TITE-CRM, with the intent to determine the dose at or below the MTD that could be safely administered for at least three cycles. Therefore, based on the acute and chronic toxicity (i.e., toxicity evaluated in the MTD expansion cohort), the RP2D was dose level 6.

Table 2.

Summary of dose limiting toxicities

Treatment Group	MK-8033 Dose (BID)	DLT	No. of Subjects with a DLT	Number of DLTs*
				Grade 1	Grade 2	Grade 3	Grade 4
Dose Level 1 (n = 1)	50 mg	Total	0
Dose Level 2 (n = 1)	100 mg	Total	0
Dose Level 3 (n = 3)	200 mg	Total	0
Dose Level 4 (n = 4)	400 mg	Total	0
Dose Level 5 (n = 7)	550 mg	Total	1
		Fatigue	1	0	0	1	0
Dose Level 6 (n = 18)+	750 mg	Total	1
		Nausea	1	0	1	0	0
Dose Level 7 (n = 6)	1050 mg	Total	2
		Nausea	2	0	0	2	0
		Vomiting	2	0	1	1	0
Dose Level 8 (n = 3)	1500 mg	Total	2
		Elevated ALT	1	0	0	1	0
		Elevated AST	1	0	0	1	0
		Elevated Amylase	1	0	0	1	0
		Hypokalaemia	1	0	0	0	1
Part C - (n = 3)	750 mg	Total	0

BID: twice daily; DLT, Dose Limiting Toxicity; Toxicity Grades are based on CTCAE version 3.0

^*Only the highest reported grade for a given DLT is counted for the individual subject

⁺Dose determined to be the maximum tolerated dose (MTD) and the recommended phase 2 dose (RP2D)

Of all 46 patients treated, 31 (67%) patients experienced drug-related adverse events, and one experienced serious drug-related adverse events (i.e., vomiting and hypokalemia). Eight (17%) patients discontinued the study due to adverse events, but only two of those patients, one at dose level 5 and one at dose level 7, withdrew from the study due to unacceptable drug-related toxicity. Table 3 displays all drug-related adverse events experienced by at least three patients each, with the most common being fatigue (28.3%), nausea (21.7%), alopecia (19.6%), elevated alanine aminotransferase (17.4%), and anorexia (17.4%). Grade ≥ 3 toxicities related to MK-8033 were experienced in ten patients (Table 3), with the most common (4.3%) being nausea, lymphopenia, dyspepsia, and elevated levels of amylase and lipase. Five patients died during the study, however it was determined by the investigators to be unrelated to MK-8033.

Table 3.

Most common (frequency > 5%) treatment-related adverse events categorized by highest grade experienced per patient (n = 46)

Adverse Event	All Grades		Grade ≥ 3*
	n	%	n	%
Fatigue	13	28.3%	1	2.2%
Nausea	10	21.7%	2	4.3%
Alopecia	9	19.6%	0	–
Elevated ALT	8	17.4%	1	2.2%
Anorexia	8	17.4%	0	–
Vomiting	6	13.0%	1	2.2%
Elevated Creatinine	6	13.0%	0	–
Elevated AST	5	10.9%	1	2.2%
Lymphopenia	4	8.7%	2	4.3%
Dysgeusia	4	8.7%	0	–
Diarrhea	3	6.5%	0	–
Dyspepsia	3	6.5%	2	4.3%
Elevated Amylase	3	6.5%	2	4.3%
Elevated Lipase	3	6.5%	2	4.3%
Decreased Hemoglobin	3	6.5%	1	2.2%
Hypokalemia	3	6.5%	1	2.2%
Headache	3	6.5%	0	–

^*Other Grade > =3 events experienced by at least one patient were Prolonged Activated Partial Thromboplastin Time and Hypophagia

ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase

Preliminary pharmacokinetics

Average plasma concentration profiles over the 50 to 1500 mg BID dose range are shown in Fig. 1a with a linear scale and in Figure B with a semi-log scale. Peak concentrations on Day 1 and 14 occurred between 1 to 3 h across the dose range, albeit with large inter-subject variability across the dose levels (data not shown). The Day 14/Day 1 geometric mean accumulation ratios for the max concentration (C_max) ranged from 1.2 to 4.2 over the 200 to 1500 mg BID dose range. However, these results were not conclusive given the limited data and degree of inter-patient variability. Additionally, due to single patient cohorts and variability in the 200 to 1050 mg BID doses, dose proportionality could not be accurately assessed.

Fig. 1.

Mean plasma concentration profiles over the 50 to 1500 mg twice daily doses of MK-8033. Peak concentrations on Day 1 and 14 appeared to occur between 1 and 3 h across the dose range; however, this analysis was not conclusive due to the large inter-subject variability and the small number of samples at each time point (see Note). Note: For the 200 mg dose, n = 2 for Day 8 predose, Day 14 12 h, and Day 14 72 h time points; For the 400 mg dose, n = 3 for Day 4 predose and Day 14 12 h and 72 h, and n = 2 for Day 14 28 h time points. For the 550 mg dose, n = 6 for Day 1 2 h, Day 4 predose, and all Day 14 time points, and n = 5 for Day 14 24 h and 72 h time points. For the 750 mg dose, n = 3 for Day 1 3 h time point. For the 1050 mg dose, n = 4 for Day 8 and Day 14 time points. Lastly, for the 1500 mg dose, n = 1 for Day 14 time points

Individual MK-8033 steady state (Day 7 or 8) plasma concentration profiles for the two subjects who completed the crossover study in Part C are shown in Fig. 2. From these preliminary profiles, co-administration with the PPI omeprazole does not appear to strongly impact the steady state C_max, area under the curve between 0 and 12 h (AUC_0–12h), and the time to reach C_max (T_max).

Fig. 2.

Individual steady state plasma concentration profiles following administration of twice daily oral doses of 770 mg MK-8033 alone or in combination with one daily 20 mg omeprazole for the two patients who completed the cross-over study (inset = semilog scale). Based on these data, the proton-pump inhibitor omeprazole does not appear to strongly impact the steady state pharmacokinetics of MK-8033

Preliminary efficacy

One patient with endometrioid adenocarcinoma experienced a partial response (PR); eight patients (18.2%) with ovarian cancer (n = 1), hemangiopericytoma (n = 1), NSCLC (n = 2), pancreatic cancer (n = 1), colon cancer (n = 1), and head and neck cancer (n = 2) had stable disease (SD) for a disease control rate of 17.4%. Thirty-three patients (71.7%) had a best response of progressive disease (PD). The remaining four patients came off study due to clinical progression before a postbaseline disease assessment was performed. The median progression-free survival (PFS) was 57 days (95% confidence interval 42–115 days); the PFS for the patient with endometrioid adenocarcinoma was 846 days. The PFS rate at 6 weeks and 12 weeks was 80.4% and 15.2%, respectively.

Discussion

It is well known that the c-Met receptor, which is either overexpressed or mutated in multiple types of cancers, plays critical roles in tumor formation, progression, metastasis, and angiogenesis; thus, the development of therapeutics directed towards c-Met is an active area of investigation. MK-8033 is a potent small molecule inhibitor of the HGF/c-MET axis that showed broad anti-proliferative activity with sub-micromolar IC50 in multiple cancer cell lines. Thus, we performed this first-in-human, multiple center, phase I study to evaluate the RP2D, which was determined to be 750 mg BID based on dose limiting toxicities of nausea/vomiting, ALT/AST elevation, amylase elevation, and hypokalemia, and the ability to safely administer that dose for at least three cycles in an MTD expansion cohort. Generally, single-agent MK-8033 was well tolerated with a majority of the adverse events being grades 1 or2.Observedclinical efficacywas limited asonly one patient achieved a partial response and eight patients achieved a best response of stable disease. Strikingly, the patient with endometrioid adenocarcinoma who responded was progression free for 846 days.

A secondary objective of this study was to characterize the preliminary pharmacokinetics of single agent MK-8033, as well as the effect, if any, of PPIs on the plasma PK profile of a different formulation of MK-8033 when administered with and without omeprazole. Regardless of dose, T_max was similar and the max concentration appears to increase with dose (Fig. 1); although, dose proportionality could be not accurately assessed with the data available. Additionally, given the limited data and the degree of inter-patient variability, the geometric mean accumulation ratios could only be estimated with low confidence and should therefore be interpreted with caution. During development of MK-8033, it was hypothesized that the use of PPIs would limit drug activity; however, given the preliminary PK data from the two patients who completed the crossover study (Fig. 2), the use of the PPI omeprazole appeared to have no effect on the plasma PK profile of MK-8033.

At the time this trial was initiated, there were three other c-Met inhibitors (i.e., cabozantinib, tivantinib, and crizotinib) in development that had shown acceptable safety profiles and preliminary efficacy in early-stage trials. (Since crizotinib is currently marketed for patients with alterations in the anaplastic lymphoma kinase gene, the remaining discussion will compare MK-8033 to cabozantinib and tivantinib.) In regard to potency towards c-Met, MK-8033 (IC50 = 1 nM) has similar potency as cabozantinib (IC50 = 1.3 nM) but more potent than tivantinib (IC50 = 0.1 μM) [18, 21, 22]. However, the selectivity of tivantinib is higher as it only binds to c-Met, whereas both MK-8033 and cabozantinib target multiple tyrosine kinases [18, 21, 22].

The toxicity profile for MK-8033 was similar to what has been observed with both tivantinib and cabozantinib. Both fatigue and nausea were the most frequently observed adverse events with MK-8033 and tivantinib, while cabozantinib had diarrhea occurring more frequently than both fatigue and nausea [23, 24]. Additionally, several of the dose-limiting toxicities reported in previous phase I trials of tivantinib and cabozanitinib was also observed with MK-8033. However, the frequency and severity of hematological DLTs observed in one study of tivantinib were not observed in the current study [24].

Single-agent MK-8033 had limited clinical activity with only one patient achieving a response. Previous phase I trials of tivantinib reported similar clinical responses as MK-8033 but with more durable disease control rates [23, 24]. Based on this limited clinical activity of single agent tivantinib, combination regimens with chemotherapy and targeted agents have been explored in a variety of tumor types [25–27]. Even though cabozantinib monotherapy showed encouraging single agent activity [28, 29], especially in medullary thyroid cancer with a 49% response rate [30], trials investigating cabozantinib in combination with chemotherapy, radiation, targeted agents, and checkpoint inhibitors have also been investigated across multiple tumor types [31–34]. These regimens were reported as safe and tolerable with encouraging clinical activity, thus providing a strong rationale for exploring MK-8033 in combination.

Indeed, there are recent preclinical data to suggest that MK-8033 in combination with chemotherapy or radiation might be efficacious. In multiple ovarian cancer cell lines, MK-8033 synergized with chemotherapy (paclitaxel + carboplatin) to inhibit cell growth [35]. Another study in non-small cell lung cancer reported that MK-8033 radiosensitized high c-Met expressing cell lines but not low-c-Met expressing cell lines [36]. However, irradiation of a high and low c-Met expression cell line induced an increase in c-Met expression increased at 30 min after exposure, and subsequent targeted with MK-8033 followed by a second radiation dose reduced the clonogenic survival in both cell lines [36]. These data pooled together suggested that targeting c-Met could be an effective strategy to radiosensitize both non-small cell lung tumors with high c-Met expression or tumors with low c-Met expression that have acquired resistance to radiation.

In conclusion, the RP2D based on both acute and chronic toxicity was determined to be dose level 6, i.e. MK-8033 750 mg BID. Although data obtained from this study showed that MK-8033 appeared well-tolerated with a similar toxicity profile to other c-Met inhibitors that were in development when this study began, based on limited observed efficacy there was unfortunately a decision to discontinue the study and the clinical development of MK-8033 by Merck Oncology Franchise from a portfolio consideration.