Human Safety, Tolerability, and Pharmacokinetics of Molnupiravir, a Novel Broad-Spectrum Oral Antiviral Agent with Activity against SARS-CoV-2

Molnupiravir (EIDD-2801/MK-4482), the prodrug of the active antiviral ribonucleoside analog β-d-N4-hydroxycytidine (NHC; EIDD-1931), has activity against a number of RNA viruses, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV), and seasonal and pandemic influenza viruses. Single and multiple doses of molnupiravir were evaluated in this first-in-human, phase 1, randomized, double-blind, placebo-controlled study in healthy volunteers, which included evaluation of the effect of food on pharmacokinetics.

models (2)(3)(4). Molnupiravir is quickly cleaved in plasma to EIDD-1931, which after distribution into various tissues, is converted to its corresponding 59-triphosphate by host kinases (Fig. 1) (4). EIDD-1931 59-triphosphate is a competitive alternative substrate for the virally encoded RNA-dependent RNA polymerase, and upon incorporation into nascent chain viral RNA, it induces an antiviral effect via viral error catastrophe (4,5), a concept that is predicated on increasing the viral mutation rate beyond a biologically tolerable threshold, resulting in impairment of viral fitness and leading to viral extinction. Molnupiravir has demonstrated in vitro activity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in human airway epithelial cell cultures (2). Prophylactic and therapeutic administration of molnupiravir to mice infected with severe acute respiratory syndrome coronavirus (SARS-CoV) or Middle East respiratory syndrome coronavirus (MERS-CoV) improved pulmonary function, and reduced virus titer and body weight loss. In the ferret model of influenza, treatment of pandemic influenza A virus with molnupiravir resulted in reduced viral shedding and inflammatory cellular infiltrates in nasal lavages, with a normal humoral antiviral response (3).
Here, we report the results of a first-in-human, phase 1, randomized, double-blind, placebo-controlled study to determine the safety, tolerability, and pharmacokinetics of single and multiple ascending oral doses of molnupiravir in healthy subjects. A randomized, open-label, crossover evaluation in the fed (high fat) and fasted states was also conducted to assess the effect of food on the pharmacokinetics of single doses of molnupiravir.

RESULTS
Eligible subjects were randomized in a 3:1 ratio to either study drug or placebo in the single-and multiple-ascending-dose parts of the study. Each cohort comprised 8 subjects, of whom 6 were administered molnupiravir and 2 were administered placebo. Single oral doses of 50 to 1,600 mg molnupiravir or placebo were administered in the single-ascending-dose part, and twice-daily (BID) doses of 50 to 800 mg molnupiravir or placebo were administered for 5.5 days in the multiple-ascending-dose part. Subjects were followed for 14 days following completion of dosing for assessments of safety, tolerability, and pharmacokinetics. Subjects in the food effect evaluation were randomized in a 1:1 ratio to either 200 mg molnupiravir (anticipated to be a therapeutic dose based on preclinical studies) in the fed state followed by 200 mg molnupiravir in the fasted state, or vice versa, with a 14-day washout period between doses. A capsule formulation was used in all parts of the study, with the exception of single ascending doses of #800 mg, where an oral solution formulation was used. An oral solution of the study drug in the single ascending dose part of the study allowed the greatest flexibility in initial dose escalation, based on safety and level of exposure. A capsule formulation of the study drug was used in the multiple-ascending-dose part of the study and for the food effect evaluation because this is the intended formulation for use in the clinical setting.
Disposition. Sixty-four subjects received a single dose of between 50 and 1,600 mg molnupiravir or placebo, 55 subjects received between 50 and 800 mg molnupiravir or FIG 1 Molnupiravir is rapidly converted in the plasma to EIDD-1931 (NHC), which after distribution into various tissues is converted by host kinases into EIDD-1931 59-triphosphate, the active antiviral agent.
placebo BID for 5.5 days, and 10 subjects received a single dose of 200 mg molnupiravir in the fed state followed by a single dose of 200 mg molnupiravir in the fasted state after a washout period of 14 days, or vice versa. Additionally, 1 subject in the multipleascending-dose part received 800 mg molnupiravir BID for 3 days, but was discontinued from dosing by the investigator on day 4. All subjects, including the subject who discontinued dosing, completed the protocol-specified study procedures and assessments.
Demography. Subjects were aged between 19 and 60 years, with a mean body mass index between 24.4 and 25.4 kg/m 2 ( Table 1). The majority of the subjects were male Caucasians. There were no other notable differences in subject demography between cohorts, except for age, where the mean age was higher in the food effect evaluation cohort, the 50-mg molnupiravir single-dose cohort, and the 100-mg molnupiravir multiple-dose cohort (data not shown).
Tolerability. Adverse events were graded using the Division of Microbiology and Infectious Diseases (DMID) toxicity grading (6).
(i) Single ascending doses. Overall, 37.5% of subjects reported an adverse event ( Table 2). There were no apparent dose-related trends, with a greater proportion of subjects reporting adverse events following administration of placebo (43.8%) than following administration of molnupiravir (35.4%). Only 1 moderate adverse event (headache; grade 2) was reported following administration of molnupiravir, which occurred at the 400-mg dose level. One subject also reported moderate adverse events (nausea and headache; grade 2) following administration of placebo. No severe (grade 3) adverse events were reported. The most frequently reported adverse event was headache, which was reported by 18.8% of subjects who were administered placebo and 12.5% of subjects who were administered molnupiravir.
(ii) Multiple ascending doses. Overall, 44.6% of subjects reported an adverse event (Table 3). There were no apparent dose-related trends, with a greater proportion of subjects reporting adverse events following administration of placebo (50.0%) than following administration of molnupiravir (42.9%). With the exception of 1 subject who reported moderate (grade 2) events of oropharyngeal pain, pain in extremity, and influenza-like illness, all adverse events were mild (grade 1) in severity. The most frequently reported adverse event was diarrhea, which was reported by 7.1% of subjects who were administered molnupiravir and 7.1% of subjects who were administered placebo. One subject discontinued study drug administration on day 4 because of an adverse event of mild, truncal, maculopapular, pruritic rash following multiple BID doses of 800 mg molnupiravir, which was considered by the investigator to be related to the study drug. Following discontinuation, the subject was administered potent topical steroid treatment and antihistamines, and pruritis and rash had both resolved within 18 days.
(iii) Food effect evaluation. Three subjects in the food effect evaluation each reported 1 adverse event, all of which were mild (grade 1) in severity.
There were no serious adverse events reported across the entire study and there were no trends of increased frequency or severity of adverse events with higher doses of molnupiravir.
Safety. There were no clinically significant findings or dose-related trends in clinical laboratory, vital signs, and electrocardiogram data. There were no clinically significant changes in hematological parameters seen in this study. Dose escalations were discontinued before a maximum tolerated dose was reached because plasma exposures (as assessed by area under the plasma concentration versus time curve [AUC] and Preferred terms reported by more than 1 subject (n) c Diarrhea  1  1  1  1  Back pain  2  1  Headache  1  2  Somnolence 2 1 a n, number of subjects with an adverse event; nE, number of adverse events. b BID, twice daily; N, number of subjects. c Subjects who had more than 1 occurrence of the same preferred term were counted only once.
Painter et al. Antimicrobial Agents and Chemotherapy maximum observed concentration [C max] ) that were expected to be efficacious based on scaling from animal models of seasonal and pandemic influenza were exceeded (3).
Pharmacokinetics. (i) Single ascending doses. Concentrations of molnupiravir were generally below the limit of quantification (BLQ) at doses up to 800 mg, with the exception of the 0.25-h time point after doses of 600 and 800 mg, where concentrations were quantifiable in 5 and 4 subjects, respectively, and the 0.5-h time point after a dose of 800 mg, where concentrations were quantifiable in all subjects. At doses of 1,200 and 1,600 mg, concentrations of molnupiravir were quantifiable at 1 or more time points between 0.25 and 1.5 h postdose in all subjects. Molnupiravir pharmacokinetic parameters were not calculable for doses of #400 mg; however, at doses of $600 mg, C max , time of C max (t max ), and time of last quantifiable concentration were reported. Following administration of between 600 and 1,600 mg molnupiravir, mean C max values were up to 13.2 ng/ml, and values of median t max were between 0.25 and 0.75 h (data not shown). It should be noted that molnupiravir concentrations represented only approximately 0.2% of EIDD-1931 concentrations and t max of molnupiravir occurred at the first sampling time point for the 600-mg dose level, and therefore C max may have been underestimated. At doses of $800 mg, trace amounts of molnupiravir were detected in the urine, which represented approximately 0.002% of the dose (data not shown).
Following oral administration of molnupiravir at doses up to 800 mg, EIDD-1931 appeared rapidly in plasma, with a median t max of 1.00 h postdose in all dose cohorts, after which plasma concentrations declined in an essentially monophasic manner with geometric mean terminal elimination half-lives (t 1/2 ) of between 0.910 and 1.29 h postdose (Table 4 and Fig. 2). However, at doses of 1,200 and 1,600 mg, median t max was delayed, with median t max occurring at 1.75 and 1.50 h, respectively. Plasma concentrations at doses of 1,200 and 1,600 mg were quantifiable in at least 1 subject until 24 h postdose, and these profiles showed biphasic elimination with a second slower elimination phase where mean t 1/2 was longer, with values of 1.81 and 4.59 h, respectively.
The plasma concentration-time profiles were generally well defined, with the percentage of area under the plasma concentration-time curve from time o0extrapolated to infinity (AUC inf ) that was extrapolated being ,10% for all subjects. When assessed using a power model [ln(parameter) = intercept 1 slope Â ln(dose) 1 random error], mean C max increased in a dose-proportional manner, with the 90% confidence interval containing unity. Similarly, mean AUC inf increased in an approximately dose-proportional manner; however, the lower bound of the 90% confidence interval was slightly above unity ( Table 5).
The amount of EIDD-1931 excreted in urine from time zero to 24 h postdose (Ae 0-24 ) increased supraproportionally with dose, and there was a similar trend for apparent clearance (CL R ) to increase. Between 0.820% (at the 50-mg dose level) and 6.70% (at the 1,600-mg dose level) of the dose was excreted in urine as EIDD-1931, and the majority of the total amount was generally excreted within the first 4 h postdose.
(ii) Multiple ascending doses. Concentrations of molnupiravir were generally BLQ at doses of #400 mg BID, and pharmacokinetic parameters were not calculable. Concentrations of molnupiravir were quantifiable in 4 subjects at either 0.5 or 1 h postdose on day 1 and in 3 subjects at 0.5 h postdose on day 6 at the 600-mg BID dose level. At the 800-mg dose level, concentrations of molnupiravir were quantifiable from all except 1 subject at 0.5 h postdose on days 1 and 6, but at no other time points, consistent with single ascending doses.
Following oral administration of molnupiravir, EIDD-1931 appeared rapidly in plasma, with a median t max in all dose cohorts of between 1.00 and 1.75 h postdose across both days 1 and 6 ( Table 6 and Fig. 3). For all dose levels, plasma concentrations declined in an essentially monophasic manner on day 1, with mean t 1/2 ranging from 0.918 to 1.18 h. Similarly, plasma concentrations declined in an essentially monophasic   manner on day 6 for subjects at dose levels of #200 mg BID and for the majority of subjects at the 300-and 400-mg BID dose levels. In contrast, for 1 subject at each of the 300-and 400-mg dose levels and for all subjects at the 600-and 800-mg BID dose levels, biphasic elimination was observed, and there was the emergence of a second, slower elimination phase on day 6, which was reflected in an increase in the mean t 1/2 with increasing dose at doses of $200 mg. Of note, at the 600-mg BID dose level, the lack of a clearly defined terminal elimination phase confounded the evaluation of t 1/2 for the majority of subjects. At the 800-mg BID dose level, the mean t 1/2 was 7.08 h. There was no evidence of accumulation, with the geometric mean accumulation ratios based on area under the plasma concentration-time curve during a dosing interval (AUC t ) and C max values between 0.938 and 1.16 and between 0.843 and 1.10, respectively, across all dose levels.
On day 1, when assessed using the power model, mean C max and AUC inf increased in an approximately dose-proportional manner. However, the upper bound of the 90% confidence interval for C max was slightly below unity and the lower bound of the 90% confidence interval for AUC inf was slightly above unity (Table 7). On day 6, mean C max increased in a dose-proportional manner, with the 90% confidence interval containing unity. Similarly, mean AUC t increased in an approximately dose-proportional manner; however, the lower bound of the 90% confidence interval was slightly above unity (Table 7). AUC inf on day 1 for the multiple-dose cohorts, where a capsule formulation was administered, was similar to those for the corresponding single-dose cohorts where a     solution formulation was administered, with geometric mean ratios of between 0.91 and 1.09. Geometric mean C max was slightly lower following dosing with the capsule formulation, with geometric mean ratios of between 0.76 and 1.00, and a trend to smaller ratios at higher doses. Median t max occurred up to 0.75 h later following administration of the capsule formulation, with the difference being greatest at doses $600 mg BID. Thus, it appears that the extent of absorption is similar for the solution and capsule formulations, but the rate of absorption appears to be slightly slower for the capsules. However, these data should be interpreted with caution because this was not a crossover study. Between 0.854% and 3.61% of the dose was excreted in urine as EIDD-1931 on both days 1 and 6, and, similar to single doses, the majority was excreted in the first 4 h postdose (Table 6). There was no consistent dose-related trend in the percentage of dose administered recovered in urine during a dosing interval (Fe 0-t ) or CL R at doses of #200 mg BID. However, there was a trend for Fe 0-t and CL R to increase with dose at doses .200 mg BID, with a 4-fold increase in dose from 200 to 800 mg BID resulting in a 16-fold increase in the amount of the dose recovered in urine during a dosing interval (Ae 0-t ) on day 1 and an 11-fold increase in Ae 0-t on day 6.
(iii) Food effect. Concentrations of molnupiravir were generally BLQ and pharmacokinetic parameters were not calculable. Concentrations of EIDD-1931 were quantifiable at 0.25 h postdose for 2 subjects in the fasted state, but no subjects in the fed state. The first quantifiable concentrations in the fed state were between 0.5 and 1.5 h postdose.
Following administration of 200 mg molnupiravir in the fed state, t max of EIDD-1931 occurred later, with a median of 3.00 h postdose versus 1.00 h postdose (Table 8 and Fig. 4). Generally, the slower absorption and later t max in the fed state was reflected in lower C max ; however, one subject had similar profiles for both treatments (data not shown). Mean C max was approximately 36% lower in the fed state compared to the fasted state, and this difference was statistically significant; however, exposure (assessed by AUC inf ) was similar for both fed and fasted states and demonstrated that the rate, but not the extent, of absorption was lower in the fed state. Following C max , concentrations of EIDD-1931 declined in an essentially monophasic manner in both   the fed and fasted state and remained quantifiable until between 9 and 15 h postdose in the fed state and between 6 and 9 h in the fasted state. The mean t 1/2 was similar between fed and fasted treatments, with values of 1.09 and 0.977 h, respectively. Urine pharmacokinetic parameters were similar to those reported for single ascending doses.

DISCUSSION
There is a significant need for an antiviral drug against coronaviruses with pandemic potential that is generally safe and well tolerated and can be easily administered in the outpatient setting. The oral route of administration of molnupiravir makes it appropriate and convenient for administration to outpatients.
Molnupiravir is well absorbed and the appearance of the parent ribonucleoside analog, EIDD-1931, in plasma demonstrates linear, dose-proportional pharmacokinetics when administered between doses of 50 and 1,600 mg. Although the rate of absorption was slower in the fed state, with lower values of t max and C max and a longer duration of measurable exposure, the extent of absorption (as assessed by AUC inf ) was similar for both fed and fasted states. Therefore, the administration of molnupiravir with food is unlikely to have an effect on therapeutic exposure. There was no evidence in humans of the capacity-limited uptake observed in pharmacokinetic studies conducted in mice (4). Exposures in the multiple-ascending dose part of the study, which utilized doses between 50 and 800 mg administered BID, could be extrapolated from the exposures observed in the single-ascending dose part of the study. The t 1/2 of EIDD-1931 is dose dependent, and ranges between 0.907 and 7.08 h. However, it should be noted that the decision to utilize BID dosing in the multiple ascending dose part of the study was based on t 1/2 values for the active antiviral agent, EIDD-1931 59-triphosphate, determined in cell culture and in lung tissue from animal model studies (4,7). In these studies, the intracellular half-life of EIDD-1931 59-triphosphate ranged from 3 h in Huh-7 cells, a human liver cell line, up to 6.6 h in murine lung tissue.
Molnupiravir was well tolerated at doses of 50 to 800 mg administered BID for 5.5 days and at single doses up to 1,600 mg. The most frequently observed adverse event was headache in the single-ascending dose part and diarrhea in the multipleascending dose part. A greater number of placebo-treated subjects reported headaches in the single-ascending dose part (18.8% placebo versus 12.5% molnupiravir) and the same number of placebo-treated subjects reported diarrhea in the multipleascending dose part (7.1%) as molnupiravir-administered subjects. One subject discontinued dosing in this study because of rash. No subjects experienced serious adverse events. The absence of clinically significant findings or dose-related trends in clinical laboratory, vital signs, and electrocardiography, taken with the tolerability findings,

Safety and Pharmacokinetics of Molnupiravir
Antimicrobial Agents and Chemotherapy indicate that molnupiravir was generally safe at the dose levels and duration tested in this study cohort. Homogeneity in body mass index and other enrollment restrictions that minimize metabolic differences among subjects (for example, diabetes and gastrointestinal disease, etc.) may have reduced variability in pharmacokinetic parameters. While most participants were male and Caucasian because of enrollment criteria and geographic constraints of the study site, there are no known additional metabolism concerns that are sex-or race-based. Data from this study support advancement of molnupiravir into phase 2 studies in subjects with susceptible RNA-viral diseases, including COVID-19. The ability of molnupiravir to potentially treat highly pathogenic respiratory RNA virus infections has been demonstrated in ferret models of disease. In ferrets, molnupiravir was highly effective at treating seasonal and pandemic influenza infections and in blocking SARS-CoV-2 transmission. The lowest fully efficacious doses were 2.3 mg/kg of body weight against seasonal influenza and 7 mg/kg of body weight against pandemic influenza (3). A therapeutic dose of 5 mg/kg was completely effective in blocking SARS-CoV-2 transmission from infected animals that shed virus to cohoused, uninfected animals (8). The doses that were effective in ferrets yielded plasma exposure levels ranging between 466 and 1,418 ng/ml for C max and between 1,474 and 4,487 ng · h/ml for AUC. Based on both C max and AUC levels achieved in the phase 1 study (Table 4), adequate plasma exposure to be effective in treating seasonal and pandemic influenza and in blocking the transmission of SARS-CoV-2 in animal models is reached in humans at doses between 200 and 800 mg.
Molnupiravir was also highly effective when administered prophylactically and therapeutically in mouse models of SARS-CoV-2 and MERS-CoV. Doses of 50, 150, and 500 mg/kg BID were effective in significantly reducing lung viral loads and in improving pulmonary function (2). Scaling between efficacious molnupiravir exposures in mouse models of coronavirus infection and in humans is significantly more difficult than in ferrets. High plasma levels of EIDD-1931 are required to achieve efficacious exposures to EIDD-1931 59-triphosphate in murine lung tissue (4) because of two key factors. First, mice have between 45% and 94% higher plasma uridine levels than humans (9)(10)(11). Uptake of EIDD-1931 from the plasma into tissue appears to be mediated by sodium-dependent concentrative transporters that show a high affinity for uridine. High plasma levels of uridine can effectively outcompete the uptake and distribution of EIDD-1931 from the plasma into key tissues in the pathogenesis of disease. Second, levels of uridine in mouse tissues are often 10-fold higher than the levels in plasma (9). Consequently, uridine can also compete with EIDD-1931 for phosphorylation. Therefore, the levels of uridine seen in mouse plasma and tissue can result in significantly higher concentrations of drug being required to produce efficacious levels of active 59-triphosphate at the site of drug action. Nonetheless, exposures (based on AUC levels) that were associated with therapeutic effects in mice were achieved in humans.
In conclusion, molnupiravir is well absorbed after oral administration and absorption is minimally affected by food intake. Plasma exposure to EIDD-1931 is essentially dose proportional across the range of doses tested, and based on scaling from animal models of influenza and pathogenic coronavirus infections, these plasma exposures may be adequate to provide prophylactic and therapeutic benefit in the treatment of these respiratory infections. No accumulation was observed in the multiple-ascendingdose part of this study. Very little molnupiravir or EIDD-1931 was detected in urine, despite the fact that nucleoside analogs as well as natural nucleosides are in general actively secreted by the kidney. This may be the result of metabolism of EIDD-1931 to cytidine and uridine. Molnupiravir was well tolerated in this study, and no subjects experienced serious adverse events.
Molnupiravir, having demonstrated good tolerability and dose-proportional pharmacokinetics following administration to healthy volunteers at clinically relevant doses is well positioned to be evaluated for clinical efficacy and safety in large-scale COVID-19 studies.