Introduction

Emvododstat, also known as PTC299 (Fig. 1), is an orally bioavailable small molecule originally identified as an inhibitor of the translation of vascular endothelial growth factor A (VEGFA) mRNA (Cao et al., 2016) and was developed as an oncology agent for the treatment of solid tumors (Packer et al., 2015; Bender Ignacio et al., 2016; Weetall, et al., 2016). Later research revealed that emvododstat’s mechanism of action is due to its direct and potent inhibition of the dihydroorotate dehydrogenase (DHODH) enzyme, a rate-limiting enzyme in the de novo pyrimidine nucleotide synthesis (Cao et al., 2019). In vitro studies demonstrated that emvododstat is more potent against leukemic malignancies, including acute myeloid leukemia (AML), than against solid tumors (Cao et al., 2019; Branstrom et al., 2022). Emvododstat also showed broad-spectrum antiviral activity, and most importantly, emvododstat potently inhibited viral replication and suppressed induction of inflammatory cytokines in COVID-19 (SARS-CoV-2) cell-based assays (Luban et al., 2021). Based on these results, emvododstat has the potential to address unmet needs in certain cancers and RNA virus infection diseases, where cancer cells or viruses rely on the de novo biosynthesis of pyrimidine nucleotides for survival or rapid proliferation. Emvododstat is now under clinical development for the treatment of AML and for hospitalized patients with COVID-19 infection.

Fig. 1.

Structure of emvododstat and proposed metabolic pathways of emvododstat in human subjects following oral dose administration.

In vitro metabolism studies showed that O-demethylation followed by glucuronidation were the major metabolic pathways for emvododstat; multiple cytochrome P450s appeared to be involved in emvododstat metabolism. Emvododstat and O-desmethyl emvododstat were both inhibitors of CYP2D6 and BCRP transporter, but neither of them was a substrate for common efflux or uptake transporters investigated (Ma et al., 2022a). Following oral administration, emvododstat is bioavailable in mice, rats, dogs, and monkeys; the absorption is generally slow, and the plasma exposure of less pharmacologically active O-desmethyl emvododstat is lower than that of parent emvododstat in rodents but relatively higher in dogs and monkeys (Ma et al., 2022a). Following a single oral dose in rats and dogs, excretion of 14C-emvododstat–derived radioactivity was faster in dogs than in rats, whereas urinary excretion was minimal (<1% of dose) in both rats and dogs; emvododstat was the dominant radioactive component in rat plasma and feces and was also the dominant radioactive component in dog feces, whereas emvododstat and two of its metabolites (O-desmethyl emvododstat and M312) were the major circulating components in dog plasma (Ma et al., 2022b).

Emvododstat has been administered to healthy subjects, patients with solid tumors and AML, and hospitalized subjects with COVID-19 infection. Following a single oral dose of emvododstat ranging from 0.03 mg/kg to 3.0 mg/kg in healthy subjects, the mean time to maximum blood or plasma concentration (Tmax) was observed in a range of 3.5–5.3 hours. Although Cmax increased in a dose-proportional manner, the mean area under the plasma concentration-time curve (AUC) generally increased more than dose proportionally, with increasing dose from 0.03 to 3.0 mg/kg (Weetall et al., 2016). Plasma exposure of O-desmethyl emvododstat in human subjects was low, 1.3%–4.5% of emvododstat AUC after a single oral dose of emvododstat, but this ratio increased to 9.6%–38% after repeat 40 mg twice-daily, 80 mg twice-daily, and 100 mg twice-daily dosing for 28 days in human immunodeficiency virus–infected patients with Kaposi sarcoma (Bender Ignacio et al., 2016). Both emvododstat and O-desmethyl emvododstat had long plasma terminal elimination half-life (T1/2), and the T1/2 was dose dependent: longer T1/2 was observed with escalating dose (Weetall et al., 2016).

As advocated by the recent Food and Drug Administration (FDA) draft guidance for human radiolabeled mass balance studies, if the investigational drug and/or active metabolite(s) exhibit time-dependent pharmacokinetics, the subjects should receive a single radiolabeled dose of the drug after reaching steady state with nonradiolabeled doses, and the bioanalysis of the nonradiolabeled moieties at steady state should be conducted to help interpret the results because this approach only evaluates the clearance pathway of the radiolabeled drug (https://www.fda.gov/regulatory-information/search-fda-guidance-documents/clinical-pharmacology-considerations-human-radiolabeled-mass-balance-studies). For such a dosing regimen, however, quantification of nonradiolabeled moieties is not possible for unknown metabolites or if the reference standards of the known metabolites are not readily available. On the other hand, although the absorption, distribution, metabolism, and excretion of the total drug and metabolites (radiolabeled and nonlabeled) may have reached steady state following a single radioactivity dose preceded by repeat dosing with nonradiolabeled drug since the overall quantification is based on the radioactivity measurement derived from the last radioactive dose, and the unlabeled parent and metabolites already in the body are not accounted for, such dosing a regimen is still considered as a single radioactive dose. With the increasing application of accelerator mass spectrometry (AMS) technology in pharmaceutical industry, we believe that at the same total therapeutic dose, repeated daily microtracer radioactivity dosing followed by a traditional high-radioactivity dose is a practical and better approach to investigate the absorption, metabolism, and excretion (AME) properties of compounds with nonlinear pharmacokinetics (PK) or for which the metabolism is time dependent. AMS analysis of matrices (plasma, urine, and feces) from the last microtracer radioactivity dosing provides AME properties at or near steady state, whereas liquid scintillation counting of the samples from the traditional higher-radioactivity dose provides rich information, such as mass balance and metabolite formation and elimination kinetics, at a reasonable cost. Here, we report the AME properties of emvododstat in healthy human subjects at close to steady state by applying a dedicated study design that includes repeated daily microtracer radioactivity oral dosing followed by a high-radioactivity oral dose.

ResultsSafety Evaluation

All subjects received 112 mg emvododstat over 7 days (16 mg/day). On Days 1–6, the oral dose contained 0.3 µCi of 14C-emvododstat (1.8 µCi in total), and on Day 7, the oral dose contained 100 µCi of 14C-emvododstat. The administration of a daily oral dose of 16 mg 14C-emvododstat for 7 days to healthy male subjects was found to be safe and well tolerated. All treatment-emergent adverse events were of mild (grade 1) severity. All treatment-emergent adverse events recovered without sequelae. No deaths or other serious adverse events (SAEs) were reported. There were no findings of clinical relevance with respect to clinical laboratory parameters, vital signs, electrocardiograms, or physical examinations.

Plasma PK and Metabolite Profiles following Repeat Daily Oral 14C-Emvododstat Dose Administration on Day 6

For AMS analysis, quality control was performed by the triplicate analysis of the QC samples at two different concentrations in each run. The measured 14C/12C ratios of the QC samples did not deviate more than 15% from the true value, and the coefficient of variation was <15% in all runs.

Following once-daily oral dosing of 16 mg/0.3 µCi 14C-emvododstat for 6 days, TRA in Day-6 plasma increased from 0.14 µg equivalent (eq)/mL predose to a maximum of 0.30 µg eq/mL at 6 hours postdose. The TRA decreased to a value similar to the predose value by 24 hours postdose (0.16 µg eq/mL). The lower limit of quantification for plasma TRA was 0.002 µg eq/mL. The plasma TRA concentration-time curve of pooled plasma across seven subjects is depicted in Fig. 2, and the PK parameters are shown in Table 1.

Fig. 2.

Mean concentration-time curve of total radioactivity in plasma on Day 6 following a 16 mg/0.3 µCi or 11.1 kBq repeat daily oral dose of emvododstat and mean ± S.D. concentration-time curves of total radioactivity in blood and plasma following a 16 mg/100 µCi or 3.7 MBq oral dose of emvododstat on Day 7 in healthy male human subjects. Figure insert depicts 0–24-hour time profiles only. The Day-6 profile was generated from AMS analysis after six daily 0.3 µCi doses. Day 7 profile was generated from scintillation counting after six daily 0.3 µCi doses and a 100 µCi dose on Day 7. The total radioactivity after repeated 0.3 µCi daily dosing was too low to affect the scintillation counting results. Thus, the Day-7 profile was considered a single-dose profile.

TABLE 1

Pharmacokinetic parameters of 14C-emvododstat–derived radioactivity in 0–24-hour plasma on Day 6 following a 16 mg/0.3 µCi or 11.1 kBq repeat daily oral dose of emvododstat and in blood and plasma following a 16 mg/100 µCi or 3.7 MBq oral dose of emvododstat on Day 7 in healthy male human subjects

Following once-daily oral dosing of 16 mg/0.3 µCi 14C-emvododstat for 6 days, emvododstat was the dominant radioactive component, whereas O-desmethyl emvododstat glucuronide and M443 were the most abundant metabolites in 0–24-hour plasma after the last dose on Day 6. Emvododstat was not detectable, whereas M474 was the dominant metabolite in pooled 0–24-hour urine after the last dose on Day 6. Emvododstat and O-desmethyl emvododstat were the most abundant components in pooled 0–24-hour feces after the last dose on Day 6. The Day-6 plasma metabolite profile is shown in Fig. 3, PK parameters of emvododstat and its metabolites are shown in Table 2, and the metabolite distribution data in urine and feces are summarized in Table 3.

Fig. 3.

Representative HPLC radio-chromatograms of Day-6 0–24-hour AUC pooled plasma (A) following six daily 16 mg/0.3 µCi or 11.1 kBq oral doses of emvododstat and 1 hour (B), 3-hour (C), 6-hour (D), 12-hour (E), 48-hour (F), 144-hour (G), and 240-hour (H) pooled plasma, 0–240-hour pooled urine (I), and 0–240-hour pooled feces (J) following a 16 mg/100 µCi or 3.7 MBq oral dose of emvododstat on Day 7 in male healthy subjects.

TABLE 2

Pharmacokinetic parameters of total radioactivity, emvododstat, and its metabolites in pooled plasma following oral administration of 14C-emvododstat in healthy male human subjects

TABLE 3

Distribution of emvododstat and its metabolites in pooled Day-6 0–24-hour urine and feces following repeat once-daily oral dosing of 16 mg/0.3 µCi or 11.1 kBq 14C-emvododstat for 6 days and in pooled 0–240-hour urine and feces following a 16 mg/100 µCi or 3.7 MBq 14C-emvododstat dose on Day 7 in healthy male human subjects

Blood/Plasma PK, Mass Balance, and Metabolite Profiles following Oral 14C-Emvododstat Dose on Day 7

Following a single oral 16 mg/100 µCi 14C-emvododstat dose on Day 7, which was preceded by 6 days of daily oral dosing of 16 mg/0.3 µCi 14C-emvododstat, both whole-blood and plasma TRA increased from the first timepoint onwards to reach a mean maximum of 0.11 and 0.19 µg eq/mL, respectively, at 6 hours postdose, rapidly declined from 6 hours to 36 hours postdose, and decreased slowly thereafter. TRA in whole-blood and plasma was measurable at 240 hours postdose in all subjects. The mean blood:plasma TRA ratio was similar throughout the sampling period and ranged from 0.55 to 0.62. The plasma and blood TRA concentration-time curves are depicted in Fig. 2, and the PK parameters are shown in Table 1.

By 240 hours after the 16 mg/100 µCi 14C-emvododstat dose on Day 7, a mean of 6.0 and 19.9% of the administered dose was recovered in urine and feces, respectively. The mean amount of TRA excreted in urine in each 24-hour interval increased with time to reach a maximum mean excretion during the first follow-up collection interval (384–408 hours postdose). After this time, the mean radioactivity in urine decreased during each consecutive 24-hour collection period. Excretion of TRA in feces started within 24 hours postdose, reached a maximum during the 24–48-hour interval, and showed a slow decrease from the 120–144-hour interval onwards. This decrease continued until the last collection interval. The mean combined excretion in urine and feces was 0.5% of the dose on Days 59 to 60 and was 0.4% of the dose on Days 73 to 74, the last follow-up visit. Based on (log) linear excretion rate constants, the mean extrapolated excretion from 0 hours to infinity was calculated. The mass balance data are summarized in Table 4.

TABLE 4

Percentage of the administered dose recovered following a 16 mg/100 µCi or 3.7 MBq oral dose of 14C-emvododstat on Day 7 in healthy male human subjects

Following a single oral 16 mg/100 µCi 14C-emvododstat dose on Day 7, which was preceded by 6 days of daily oral dosing of 16 mg/0.3 µCi 14C-emvododstat, the plasma metabolite profiles were similar to those following the last 16 mg/0.3 µCi 14C-emvododstat dose on Day 6. The representative metabolite profiles are shown in Fig. 3, the plasma concentration-time curves of TRA and the most prominent emvododstat metabolites are shown in Fig. 4, PK parameters are shown in Table 2, and the metabolite distribution data in urine and feces are summarized in Table 3.

Fig. 4.

Concentration-time curves of total radioactivity (TRA), emvododstat and its prominent metabolites in pooled plasma following a 16 mg/100 µCi or 3.7 MBq oral dose of emvododstat on Day 7 in male healthy subjects.

Metabolite Identification

In addition to unchanged emvododstat, eight metabolites were identified or characterized in plasma, urine, or feces after oral administration of 14C-emvododstat.

Emvododstat and O-desmethyl emvododstat were identified by direct comparison of HPLC retention times and high-resolution mass spectral data with reference standards. Structures of other metabolites were proposed based on their high-resolution mass spectral data of chlorine isotopic patterns and fragmentation ions relative to emvododstat or O-desmethyl emvododstat. Proposed structures and metabolic pathways for the formation of the detected metabolites are presented in Fig. 1.

Emvododstat

The observed accurate mass for the protonated molecular ion (MH+) of emvododstat was at m/z 467.0923 (calculated 467.0924 with a formula of C25H21O3N2Cl2+). The characteristic product ion at m/z 359.0350 was due to the neutral loss of the anisole moiety (C7H8O) and at m/z 121.0649 (C8H9O+) was attributed to the 4-methoxyphenylmethylium ion.

O-Desmethyl Emvododstat

The observed accurate mass for MH+ of O-desmethyl emvododstat was at m/z 453.0765 (calculated 453.0767 with a formula of C24H19O3N2Cl2+). The characteristic product ion at m/z 359.0353 was due to the neutral loss of the phenol moiety (C6H6O) and at m/z 107.0494 (C7H7O+) was attributed to the 4-hydroxyphenylmethylium ion.

O-Desmethyl Emvododstat Glucuronide

The observed accurate mass for MH+ of O-desmethyl emvododstat glucuronide was at m/z 629.1086 (calculated 629.1088 with a formula of C30H27O9N2Cl2+), which is 176.0321 Da (C6H8O6) higher than O-desmethyl emvododstat, indicating a glucuronic acid conjugate of O-desmethyl emvododstat. The characteristic product ions at m/z 453.0770, 359.0350, and 107.0493 all agreed well with O-desmethyl emvododstat glucuronide.

M312

The observed accurate mass for MH+ of M312 was at m/z 313.1098 (calculated 313.1102 with a formula of C18H18ON2Cl+), which is 153.9822 Da (C7H3O2Cl) lower than emvododstat, indicating the neutral loss of the 4-chlorophenyl formate moiety due to the amide bond hydrolysis. The characteristic product ions at m/z 205.0530 (C11H10ClN2+) and 121.0648 also agreed well with the proposed structure.

M298

The observed accurate mass for MH+ of M298 was at m/z 299.0945 (calculated 299.0946 with a formula of C17H16ON2Cl), which is 153.9822 Da (C7H3O2Cl) lower than O-desmethyl emvododstat, or 14.0157 Da (CH2) lower than M312, indicating the neutral loss of the 4-chlorophenyl formate moiety due to the amide bond hydrolysis from O-desmethyl emvododstat or demethylation from M312. The characteristic product ions at m/z 205.0530 and 107.0492 also agreed well with the proposed structure.

M474

The observed accurate mass for MH+ of M474 was at m/z 475.1265 (calculated 475.1267 with a formula of C23H24O7N2Cl+), which is 176.0321 Da (C6H8O6) higher than M298, indicating a glucuronic acid conjugate of M298. The characteristic product ions at m/z 299.0946, 205.0530, and 107.0492 also agreed well with the proposed structure.

M324

The observed accurate mass for MH+ of M324 was at m/z 325.0736 (calculated 325.0738 with a formula of C18H14O2N2Cl+), which is 11.9636 Da (1 oxygen minus 4 hydrogens) higher than M312, indicating that M324 could be an oxidation and dehydrogenation metabolite of M312. The product ions at m/z 310.0503 of (C17H11O2N2Cl˙+), 282.0554 (C16H11ON2Cl˙+), and 253.0605 (C15H10N2Cl+) were observed.

M443

The observed accurate mass for MH+ of M443 was at m/z 444.1318 (calculated 444.1321 with a formula of C22H23O5N3Cl+). Compared with emvododstat, the characteristic product ions of M443 at m/z 357.0999 (C19H18N2O3Cl+) and 121.0648 (C8H9O+) suggest that the tetrahydro-2H-pyrido[3,4-b]indole-2-carboxylate and 4-methoxyphenyl moieties are intact, indicating the modification of the 4-chlorophenyl moiety of emvododstat. The observed product ions of M443 at m/z 106.0499 (C3H8NO3+), 88.0392 (C3H6NO2+), and 60.0444 (C2H6NO+) are characteristic of serine-derived ion species, indicating that M443 could be a serine conjugate product following the ester bond hydrolysis of emvododstat.

M482

The observed accurate mass for MH+ of M482 was at m/z 483.0875 (calculated 483.0873 with a formula of C25H21O4N2Cl2+), which is 15.9949 Da (one oxygen) higher than emvododstat, indicating that M482 was a mono-oxidation metabolite of emvododstat.

Discussion

Emvododstat is a lipophilic, neutral compound with low aqueous solubility. Following a single oral dose in male Long-Evans rats, distribution of 14C-emvododstat–derived radioactivity was extensive, with the endocrine, fatty, and secretory tissues containing the highest radioactivity (Ma et al., 2022b). Consistent with extensive distribution and retention in fatty tissues in male Long-Evans rats, excretion of 14C-emvododstat–derived radioactivity in intact male Sprague-Dawley rats was slow following a single oral dose, approximately 65% of the dose (0.3% in urine and 54.7% in feces, whereas 35.4% in carcass) within 7 days postdose in rats. In contrast, approximately 93% of dose was recovered by 8 days postdose, with most of the radioactivity excreted in 0–24-hour feces following a single oral dose in dogs. The faster excretion in dogs was most likely due to lower absorption as indicated by the observation that close to 80% of dose was excreted in 0–24-hour feces, and >94% of fecal radioactivity was attributed to unchanged emvododstat, with little contribution from metabolites (Ma et al., 2022b).

Following a single oral 16 mg/100 µCi 14C-emvododstat dose on Day 7 in human subjects in the current study, approximately 26% of the dose (6% in urine and 19.9% in feces) was recovered by 240 hours postdose. Additional 24-hour collection intervals were used to follow up the excretion until 1608 hours postdose. The total recovery as extrapolated to infinity in the current study was 94.7% (37.3% in urine and 56.6% in feces) (Table 4). Urinary excretion of 14C-emvododstat–derived radioactivity was low in both rats and dogs (<1% dose within 8 days post oral dose) but was much higher, 6% of the dose by 240 hours postdose and up to 37% of the dose extrapolated to infinity, in human subjects (Table 4).

Since the metabolism of emvododstat in humans appears to be time dependent (Bender Ignacio et al., 2016), a study design with a 6-day once-daily 16 mg/0.3 µCi oral dosing followed by a 16 mg/100 µCi oral dose was adopted to evaluate mass balance and emvododstat metabolism near the steady state. Similar metabolite profiles were observed between Day 6 and Day 7 in all matrices following repeat once-daily oral dosing of 16 mg/0.3 µCi 14C-emvododstat for 6 days or following a single oral 16 mg/100 µCi 14C-emvododstat dose on Day 7. Following the last 16 mg/0.3 µCi 14C-emvododstat oral dose on Day 6, emvododstat was the dominant component in plasma, accounting for 51.0% of TRA in the 0–24-hour plasma pool, whereas M443, O-desmethyl emvododstat glucuronide, and O-desmethyl emvododstat, accounting for 14.1%, 11.7%, and 9.6% of TRA, respectively, were the most abundant metabolites in 0–24-hour plasma after the last 16 mg/0.3 µCi dose on Day 6. Other metabolites were much less abundant, each accounting for less than 4% of the 0–24-hour plasma radioactivity (Table 2). Emvododstat was not detectable, whereas M474, accounting for 72.5% of TRA, was the dominant metabolite in 0–24-hour urine on Day 6. Emvododstat and O-desmethyl emvododstat were the most abundant components, accounting for 20.3% and 47.2% of TRA, respectively, in 0–24-hour feces on Day 6 (Table 3). Following a 16 mg/100 µCi 14C-emvododstat oral dose on Day 7, unchanged emvododstat was the most abundant circulating entity, accounting for 35.6% of AUC of 0–240-hour plasma TRA; M443 and O-desmethyl emvododstat glucuronide were the most abundant metabolites, accounting for 15.0% and 11.7% of plasma TRA, respectively; and O-desmethyl emvododstat, M474, M298, and M312 were less abundant metabolites, accounting for 8.61%, 7.41%, 4.66%, and 4.31% of AUC from time zero to 240 hours postdose TRA, respectively (Table 2). In 0–240-hour urine, emvododstat was not detectable, and M474 was the most abundant metabolite, accounting for 4.06% of the dose (Table 3). In 0–240-hour feces, emvododstat accounted for 1.63% of the dose, and O-desmethyl emvododstat was the most abundant metabolite, accounting for 8.97% of the dose; the other metabolites M298, M443, M312, and M324 were less abundant, accounting for 1.15%, 1.09%, 0.76%, and 0.50% of the dose, respectively (Table 3).

It should be noted that except for O-desmethyl emvododstat and its glucuronide, both peaking at 6 hours postdose, formation of all other metabolites was slow, with Tmax observed from 144 to 240 hours following a single oral 16 mg/100 µCi dose on Day 7 in human subjects (Table 2). Due to slow elimination of TRA as observed from 36 to 240 hours postdose, concentrations of these slowly forming metabolites were low and were therefore hard to accurately measure using the scintillation counting approach. For this end, it is advantageous that metabolites were enriched after a 6-day repeat daily microtracer radioactivity dosing and quantified using the most sensitive AMS detection technique in combination with the plasma AUC pooling strategy (Hop et al., 1998). For example, M443, M312, and M298 were not detectable or barely detectable by liquid scintillation counting within 24 hours after a single high-radioactivity dose on Day 7 but can be adequately detected by AMS in 0–24-hour plasma following repeat daily microtracer radioactivity doses for 6 days (Table 2). A follow-up high-radioactivity dose on Day 7 provided the desired mass balance data and, most importantly, the plasma metabolite kinetics, generating useful and complementary distribution results for the Day-6 plasma metabolite data.

As noted above, different radioactivity detection techniques were used to analyze samples following the Day-6 microtracer radioactivity dose and the Day-7 high-radioactivity dose: after the last microtracer radioactivity dosing on Day 6, one plasma AUC pool, one urine pool, and one feces pool from all subjects were profiled using AMS technology, whereas pooled plasma at individual timepoints and one pooled urine and one pooled feces from all subjects were analyzed using traditional scintillation counting techniques. Compared with the high-radioactivity dose (100 µCi) on Day 7, the microtracer dose (0.3 µCi/day) is negligible, and the resulting radioactivity after repeated dosing is too low to be detected by less sensitive scintillation counting technique, and thus the Day-7 dose is considered a single dose that the plasma radioactivity was lower in Day-7 plasma than in Day-6 plasma (Fig. 2). Although the profiles are qualitatively similar to each other for each matrix obtained after the last microtracer radioactivity dose on Day 6 and after the high-radioactivity dose on Day 7, we believe the differences of relative abundance between the two profiles are largely due to repeat dosing versus single dosing, but other factors, such as analytical method used and sample pooling strategy (AUC pool versus individual timepoint plasma pool and 0–24-hour pool for Day-6 samples versus 0–240-hour pool for samples after Day-7 dose) cannot be excluded. A head-to-head comparison using AMS analysis on diluted Day-7 samples might provide additional information in this regard.

In other clinical studies that used similar daily oral doses, both emvododstat and O-desmethyl emvododstat reached steady state in approximately 2 weeks; therefore, steady state may not have been reached after six daily doses in the current study. Due to the limitation of tedious sample processing and long analysis time in the AMS analysis, similar to plasma sample analysis, urine and fecal samples were analyzed using different methods; therefore, the differences between Day-6 and Day-7 profiles may also be contributed to by analytical methods as well as by dosing frequency. Therefore, additional head-to-head sample analysis is needed to get a clear answer. Nevertheless, the current study design clearly shows the added value of repeated microtracer radioactivity dosing for drugs with slow metabolism and disposition, where steady state metabolite profiles might be different from single-dose metabolite profiles. The new FDA mass balance guidance suggests that one could employ a single radiolabeled dose of the investigational drug after reaching steady state with nonradiolabeled doses for such drugs. Because this approach only evaluates the clearance pathway of the radiolabeled drug, bioanalysis of the nonradiolabeled moieties at steady state should be conducted to help interpret the results (https://www.fda.gov/regulatory-information/search-fda-guidance-documents/clinical-pharmacology-considerations-human-radiolabeled-mass-balance-studies). However, bioanalytical methods may not be available for all moieties, especially if such studies are conducted at the early stage as the FDA advocated in that same draft guidance. On the other hand, quantification of nonradiolabeled moieties is not possible if the metabolites are not ambiguously identified or if the reference standards of the known metabolites are not available. In such situations, the repeated microtracer radioactivity dose study design may be a better choice to allow for full metabolite profiling at or near steady state if the drug or metabolite exhibits time-dependent pharmacokinetics, without the need for nonradiolabeled metabolite synthesis and bioanalytical method development. In our opinion, such design may be considered for other investigational drugs that show slow metabolism or for which metabolite formation is time dependent.

Theoretically, the amount of drug input equals the amount of drug excretion during the same dosing period at true steady state. If one is able to dose the subject close to the true steady state, and obtain all needed data (e.g., plasma PK based on individual timepoints and mass balance, etc.) after the last microtracer radioactivity dosing using more sensitive but costly AMS analysis, then the high-radioactivity dose may be optional, and the only advantage of such a dose is to obtain excretion kinetics if the elimination of radioactivity is slow like in our current study. Obviously, more research is needed to optimize such study designs and to maximize outcomes with a reasonable use of resources.

Following a single oral dose of 14C-emvododstat in rats and dogs, emvododstat was the dominant radioactive component in rat plasma, whereas emvododstat and its two metabolites (O-desmethyl emvododstat and M312) were the major circulating components in dog plasma (Ma et al., 2022b). Although O-demethylation followed by glucuronidation are the common pathways in rats, dogs, and humans, the N-carbamoyl ester link was stable in rats but labile in dogs and humans: hydrolysis on the amide side led to the formation of M312 and subsequently M324, M298, and M474 in dogs and humans, whereas hydrolysis or transesterification on the ester side resulted in the formation of M443 in humans.

Liquid Chromatography/Mass Spectrometry (LC/MS) data indicate that M443 is a serine conjugate metabolite of emvododstat, possibly through the transesterification mechanism. Serine is a nonessential amino acid that is the precursor for cysteine, selenocysteine, tryptophan, glycine, and phospholipids and plays a role in the biosynthesis of purines and pyrimidines (Reitzer, 2009). Serine conjugates with xenobiotics (Steventon and Hutt, 2002; van Ravenzwaay et al., 2003; Aloysius et al., 2008) and endogenous serine conjugates with lipids such as N-arachidonoyl-serine (Milman et al., 2006) and N-stearoyl serine (Tan et al., 2010) have been reported in animal species, but the formation of serine conjugate metabolites of xenobiotics in humans seems to be rare. Nevertheless, additional research is needed to better understand the formation of M443 and the relevance of emvododstat metabolites to the efficacy and safety of emvododstat in humans.

In conclusion, since the metabolism and pharmacokinetics of emvododstat in humans is time dependent, the absorption, metabolism, and excretion of emvododstat were investigated by using a combination approach of microtracer radioactivity and high-radioactivity doses. The resolution of challenges due to slow metabolism and elimination of a lipophilic compound highlighted in this study was achieved by repeat daily microtracer radioactivity oral dosing followed by high-radioactivity oral dosing at therapeutically relevant doses. Such a design may provide an alternative approach to better understanding AME properties of investigational drugs that show slow metabolism or for which metabolite formation is time dependent.