This phase I, first-in-human, open-label, dose-escalation trial was conducted in patients with metastatic or locally advanced unresectable solid tumors. M4112 was given orally, two times per day in 28-day cycles, with three patients per cohort. The starting dose of 100 mg two times per day was determined from preclinical studies and pharmacokinetic/pharmacodynamic modeling of the predicted pharmacologic active dose range in humans. Patients continued M4112 treatment until disease progression, unacceptable toxicity, or withdrawal of consent.

The primary objective was to determine the safety and tolerability or, if observed, the maximum tolerated dose (MTD), and to define the recommended phase II dose (RP2D) of M4112 monotherapy in patients with solid tumors. The secondary objectives included M4112 pharmacokinetic characterization, evaluation of preliminary clinical activity (via Response Evaluation Criteria In Solid Tumors (RECIST) version 1.1), and assessment of potential M4112 concentration-related QT prolongation. Exploratory objectives included evaluation of the effect of M4112 on kynurenine and tryptophan in the plasma and on IDO1 activity in ex vivo stimulated whole blood from M4112-treated patients.

Adults with histologically or cytologically advanced or metastatic solid tumors refractory to standard therapies or tumors for which no effective treatment was available were eligible. Other key inclusion criteria were Eastern Cooperative Oncology Group performance status of 0 or 1, and an adequate hematologic and hepatic function. Key exclusion criteria included concomitant treatment with strong inhibitors or inducers of cytochrome P450 3A4. Full eligibility criteria are shown in the online supplementary information.

To assess the safety of M4112 (primary endpoint), adverse events (AEs), dose-limiting toxicities (DLT), physical examination findings, vital sign measurements, ECG results, and changes in clinical laboratory parameters (chemistry, hematology, and coagulation) were recorded. Safety parameters were assessed from the first dose to the end of treatment. AEs were graded and classified by the investigator according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03. The MTD was based on the incidence of DLTs during cycle 1 (days 1–28). DLTs were generally defined as any grade ≥3 AEs or immune-related AEs for non-hematologic toxicities and grade ≥4 AEs for hematologic toxicities during cycle 1 at any dose level.

Blood samples for pharmacokinetic analysis were collected on days 1 and 15 of cycle 1 (predose and 0.5, 1, 2, 3, 4, 6, and 8 hours postdose), day 8 of cycle 1 (predose), and on day 1 of cycle 2 (predose and 2 hours postdose) (online supplementary figure 1). Plasma concentrations were determined by LC-MS/MS. The pharmacokinetics of M4112 were characterized by non-compartmental analyses (Phoenix WinNonlin version 6.4, Certara USA, Princeton, New Jersey, USA) for calculation of maximum observed concentration (C_max), time to C_max (t_max), and area under the drug concentration–time curve from 0 to 8 hours postdose (AUC_0–8). Accumulation (R_acc) was calculated for day 15/day 1 for C_max and AUC_0–8 as well as for dose-normalized values (C_max/dose and AUC_0–8/dose). Predose concentrations (C_pre and C_pre/dose) were determined for day 1, day 8, and day 15, and for day 1 of cycle 2. Time-matched corrected QT values were extracted for planned concentration-QT analysis.

To evaluate the pharmacodynamic activity of M4112, plasma samples were collected on day 1 (predose and 1, 2, 4, and 6 hours postdose), day 8 (predose), day 15 (predose and 1, 2, 4, and 6 hours postdose), and day 1 of cycle 2 (predose and 2 hours postdose) (online supplementary figure 1). Concentrations of kynurenine and tryptophan were measured by LC-MS/MS and kynurenine:tryptophan ratios were calculated accordingly. In addition, to assess the activity of M4112 against IDO1, blood samples were drawn directly into TruCulture blood collection tubes (Myriad RBM, Austin, Texas, USA) and preloaded with 100 ng/mL lipopolysaccharide (LPS) to stimulate IDO1; whole blood samples were collected on day 1 (predose and 1, 2, 4, and 6 hours postdose), day 8 (predose), day 15 (predose and 1, 2, 4, and 6 hours postdose), and day 1 of cycle 2 (predose and 2 hours postdose). Tubes were incubated at 37°C for 24 hours before measuring kynurenine levels by LC-MS/MS. A predose blood sample was also collected into tubes without preloaded LPS as background. Inhibition of IDO1 was calculated by measurement of the decrease in kynurenine level compared with the predose level. All patients who received treatment with M4112 were included in the safety, pharmacokinetic, and pharmacodynamic analysis sets.

CT or MRI scans were performed at baseline and within 7 days prior to day 1 of cycles 3, 5, and 7, and every three cycles thereafter to assess tumor response. Objective response, best overall response, disease control rate, and progression-free survival were assessed by the investigator according to RECIST version 1.1.27

Decisions on dose escalation were made by the Safety Monitoring Committee based on recommendations from a Bayesian two-parameter logistic regression model. The model was updated with the number of evaluable patients and DLTs observed after completion of the DLT period for each cohort. Recommendations on the next dose level were based on the posterior distribution of toxicity. The primary analysis assessed the number and percentage of DLTs, performed on the DLT analysis set by dose level. The MTD was defined using a Bayesian two-parameter logistic regression model with a target toxicity of 30%. Descriptive statistics were used for baseline characteristics, safety assessments, pharmacokinetic and pharmacodynamic data, and efficacy endpoints. The 95% exact Clopper-Pearson CI was used for binary efficacy endpoints. The Kaplan-Meier method was used for survival endpoints. Dose proportionality of pharmacokinetic parameters was evaluated statistically using a power model based on the original parameters (ln[pharmacokinetic parameter]=α+β×ln[dose]).

To assess the level of target engagement in vivo by M4112, kynurenine and tryptophan levels were measured in the plasma, liver, and tumors of mice. In mice with MC38 colon tumors expressing IDO1, the concentration of M4112 in the tumor peaked 1–2 hours post-treatment (figure 1A). By 24 hours post-treatment, the concentration of M4112 in the tumor decreased to baseline levels. A similar pharmacokinetic profile was also observed in the liver and the plasma. However, exposure to M4112 was approximately 10-fold higher in the liver compared with both the plasma and the tumor. In MC38 tumor-bearing mice, M4112 at all three dose levels (30, 75, or 200 mg/kg) significantly decreased the kynurenine:tryptophan ratio in the tumor 4 hours after M4112 treatment compared with vehicle (p<0.001) (figure 1B). A dose-dependent reduction in the kynurenine:tryptophan ratio was also often observed in the liver and plasma of mice that received M4112. In mice with IDO1-expressing CT26-KSA tumors, M4112 significantly decreased the kynurenine:tryptophan ratio in the liver after 2 hours compared with vehicle (p=0.0266) and MSC2574260 (p=0.0327) (figure 1C). Although exposure to MSC2574260 (the selective IDO1 inhibitor) was significantly higher in the liver compared with exposure to M4112 (p=0.0009), MSC2574260 failed to significantly modulate the kynurenine:tryptophan ratio.

Fifteen patients with advanced solid tumors were enrolled (table 1). Most patients were women (86.7%) and the median age was 57.0 years (range 37–77 years). All patients had received chemotherapy and eight (53.3%) patients had received radiotherapy. Patients (N=15) received one of the following doses of M4112: 100, 200, 400, 600, or 800 mg two times per day orally (three patients per cohort).

All 15 patients were included in the safety analyses. The median duration of M4112 exposure was 16.0 weeks (range 2.7–44.0 weeks). During cycle 1 (DLT assessment period), one patient in the 800 mg two times per day arm experienced a DLT (grade 3 allergic dermatitis). Another patient dosed at 800 mg two times per day developed a grade 2 QT prolongation, resulting in a dose reduction; this event did not meet the DLT criteria. Following the DLT in the 800 mg two times per day arm, the SMC recommended reducing the dose level of M4112 to 600 mg two times per day for the next patient cohort (n=3); no DLTs were observed at this dose level. Two patients in the 600 mg two times per day dose group had dose reductions due to AEs (one grade 3 diarrhea and one grade 3 rash after the DLT period). There were no deaths reported during the dose-escalation phase. The Bayesian model estimated the median probability of a DLT occurring at 800 mg to be 20.5%, with a 95% credible interval of 3.1%–62.2%. Given that the estimate was clearly below the target value of 30% and that the credible interval was quite wide, no MTD was defined. The RP2D was also not established.

The most common treatment-emergent AEs (TEAEs) that occurred in ≥2 patients (at any grade) are presented in table 2. The most frequently reported TEAEs were fatigue (33.3%), nausea (26.7%), and vomiting (26.7%). Three patients, one from each of the 400, 600, and 800 mg two times per day cohorts, experienced a serious TEAE during the trial; these included grade 3 disease progression (unrelated to treatment), grade 2 ascites (unrelated to treatment) and grade 3 allergic dermatitis (related to treatment). Two (13.3%) patients discontinued M4112 treatment due to TEAEs: one patient in the 400 mg two times per day cohort discontinued due to disease progression; and the patient on 800 mg two times per day who experienced a DLT discontinued due to allergic dermatitis.

Pharmacokinetic evaluations were based on all 15 patients who received M4112 treatment. A summary of the mean pharmacokinetic measurements is presented in table 3. The pharmacokinetics of M4112 were characterized by a t_max of approximately 1–3 hours across all dose levels on both days 1 and 15. The mean M4112 plasma concentration curves on cycle 1 day 1 and day 15 are shown in figure 2. The mean exposure of M4112 increased in a greater than dose-proportional manner from 100 to 600 mg two times per day (slope estimate of 1.46–1.67) on day 15 (figure 2B). M4112 exposure for the 800 mg two times per day cohort was initially lower than that of the other dose groups (cycle 1, day 1). However, by day 15, exposure for the 800 mg two times per day dose group increased in the one patient that continued with this regimen. There was no appreciable accumulation or evidence of autoinduction following two times per day dosing over the 100–600 mg dose range.

Following M4112 administration, a rapid onset of IDO1 inhibition in ex vivo stimulated blood was observed across all doses tested (figure 3A). Maximum inhibition was achieved 1–4 hours post-M4112 administration. There was a trend toward dose-dependent IDO1 inhibition at steady state. Approximately 90% of IDO1 activity was inhibited throughout the dosing interval at 800 mg two times per day.

A small yet consistent decrease in plasma kynurenine levels (12%–35%) was observed on day 1 of cycle 1 (up to 6 hours post-M4112 treatment) after the first dose was administered for all dose levels investigated (figure 3B). Plasma tryptophan levels increased on day 1 of cycle 1 following doses of 200 mg two times per day and above. The increase in plasma tryptophan levels was largely responsible for the decrease in the kynurenine:tryptophan ratio observed after 6 hours. However, on day 15 of cycle 1, kynurenine levels had returned to at least baseline levels (figure 3C). Based on these findings, it was decided to discontinue the clinical study due to the insufficient pharmacodynamic effect observed.

The best overall response was stable disease in 9 of 15 (60.0%) patients, and the median progression-free survival was 3.7 months. There were no objective responses. Figure 4 shows individual patient responses to treatment. There was no indication of a relationship between the dose of M4112 and disease control or duration of progression-free survival.