A Five-in-One First-in-Human Study To Assess Safety, Tolerability, and Pharmacokinetics of RO7049389, an Inhibitor of Hepatitis B Virus Capsid Assembly, after Single and Multiple Ascending Doses in Healthy Participants

RO7049389, an inhibitor of hepatitis B virus (HBV) capsid assembly, is being developed for the treatment of patients with chronic HBV infection. The objectives of this first-in-human study are to assess the safety, tolerability, pharmacokinetics (PK), food effect, inhibitory effect on CYP3A, and effect on QT of RO7049389 in healthy participants. Five components, single-ascending-dose (SAD) cohorts, multiple-ascending-dose (MAD) cohorts, food effect assessment, drug-drug interaction assessment, and concentration-QT analysis were integrated in one study (five-in-one).

ABSTRACT

RO7049389, an inhibitor of hepatitis B virus (HBV) capsid assembly, is being developed for the treatment of patients with chronic HBV infection. The objectives of this first-in-human study are to assess the safety, tolerability, pharmacokinetics (PK), food effect, inhibitory effect on CYP3A, and effect on QT of RO7049389 in healthy participants. Five components, single-ascending-dose (SAD) cohorts, multiple-ascending-dose (MAD) cohorts, food effect assessment, drug-drug interaction assessment, and concentration-QT analysis were integrated in one study (five-in-one). Participants randomly received a single dose of 150 to 2,500 mg RO7049389 or placebo in SAD cohorts (n = 41), or multiple doses of 200 to 800 mg RO7049389 or placebo in MAD cohorts (n = 42). A single doses of 450 mg RO7049389 was administered under fasted and fed condition. The microdose of midazolam was administered before and after multiple dosing of RO7049389. Safety and tolerability were monitored throughout the study. Serial blood and urine samples were collected for the PK analysis. RO7049389 was safe and well tolerated in healthy participants. Absorption and elimination of RO7049389 occurred rapidly in plasma with minimal recovery in urine. Greater than dose-proportional increases in plasma exposure were observed. Exposure of RO7049389 (450 mg) increased by ∼2-fold when administered with a high-fat meal. The inhibition effect of RO7049389 on CYP3A was weak (<20%). No effect on QT interval was observed at up to a single dose of 2,500 mg. RO7049389 displayed a favorable safety, tolerability and PK profile suitable for further clinical development. (This trial was registered at ClinicalTrials.gov with the identifier NCT02952924.)

INTRODUCTION

Hepatitis B virus (HBV) infection is a major cause of chronic liver diseases, which may lead to cirrhosis, liver failure, and hepatocellular carcinoma (HCC). Approximately 257 million people worldwide are living with chronic hepatitis B (CHB) infection (defined as hepatitis B surface antigen [HBsAg] positive). In 2015, HBV infection resulted in an estimated 887,000 deaths, mostly from complications including cirrhosis and HCC (1, 2). Currently, there are two classes of drugs available for the treatment of CHB: subcutaneously administered interferon alpha (IFN-α) preparations and orally administered nucleos(t)ide analogs (NUCs). Although both types of treatment can effectively induce the loss of hepatitis B e antigen (HBeAg) and result in the suppression of HBV DNA (virologic response) and normalization of liver transaminase enzymes (biochemical response), neither treatment achieves a high rate of functional cure, defined as sustained loss of HBsAg with or without seroconversion. In addition, IFN-based therapies are associated with many adverse effects, whereas NUCs frequently require prolonged or possibly lifelong therapy, especially for those who are HBeAg negative or with cirrhosis.

RO7049389 is a small molecule that is being developed as an orally administered solid dosage formulation for the treatment of CHB. RO7049389 binds to the HBV core protein and induces incorrect capsid assembly, suppression of viral replication, and ultimately the depletion of functional core protein. Hence, its antiviral activity derives from a mechanism distinct from currently approved drugs. In preclinical studies, RO7049389 has shown potent antiviral activity through the induction of HBV core protein misassembly and subsequent degradation. In addition, this antiviral activity displays a high degree of selectivity against HBV (3). However, preclinical studies indicated RO7049389 may cause time-dependent inhibition of CYP3A. Therefore, microdoses of a clinically validated probe substrate of CYP3A, midazolam, were used in this study to evaluate the drug-drug interaction (DDI) liability of RO7049389 due to its inhibition on CYP3A activity (4). Preclinical data flagged a potential risk of QT prolongation. Therefore, continuous 12-lead electrocardiogram (ECG) recordings (Holter monitor) were obtained to conduct the concentration-QT analysis to understand the effect on QT prolongation in clinic. RO7049389 was metabolized by CYP3A4 as major and uridine 5′-diphospho glucuronosyltransferase (UGT) 1A3 likely as minor. RO7049389 was a substrate of human multidrug resistance 1/permeability glycoprotein and of human liver organic anion transporting polypeptide 1B1 (OATP1B1) and OATP1B3. The unbound fractions of RO7049389 were 3.9 to 5.2% in humans. Based on these findings, a five-in-one first-in-human trial was performed to evaluate the safety, tolerability, PK, food effect, inhibitory effect on CYP3A, and effect on QT prolongation of RO7049389. Five components—single-ascending-dose (SAD) cohorts, multiple-ascending-dose (MAD) cohorts, food effect assessment, drug-drug interaction assessment, and concentration-QT analysis—were innovatively integrated in this study (five-in-one).

RESULTS

Participant characteristics.

The characteristics of the observed participants, including ethnicity and demographic data, are listed in Table 1. A total of 83 healthy male and female participants were enrolled into the study. Of the 83 healthy participants, 41 healthy participants received a single dose of RO7049389 (n = 30) or placebo (n = 11) in 6 SAD cohorts, as shown in Fig. 1. A further 42 healthy participants received repeat doses of RO7049389 (n = 32) or placebo (n = 10) for 14 days in 5 MAD cohorts as shown in Fig. 1. Two healthy participants (one from 200 mg twice-daily [BID] fed cohort, the other from 400 mg BID fed cohort) withdrew early for non-safety-related personal reasons and were replaced. A precautionary decision was made to terminate the 800 mg BID MAD cohort on day 3 pending a preliminary review of the newly available RO7049389 clinical metabolite data, which was disproportionately higher than that in animal species. The decision was a precautionary one and was not based on any newly identified safety concern.

TABLE 1.

Baseline demographics^a

Demographics	Healthy participants
	SAD (n = 41)	MAD (n = 42)
Median age in yr (range)	24 (18−54)	25 (19−51)
Male, no. (%)	40 (97.6)	42 (100)
Median (range)
Ht (cm)	180 (157–188)	178 (165–188)
Body wt (kg)	77 (54.5–99.2)	74 (54.8–91.9)
BMI	24.1 (19.8−29.9)	23.7 (18.4−30.4)
Race, no. (%)
Asian	5 (12)	7 (17)
White	33 (81)	29 (69)
Black or African-American	1 (2)	0
Native Hawaiian or other Pacific Islander	2 (5)	3 (7)
Multiple	0	1 (2)
Unknown	0	2 (5)

^aBMI, body mass index; SAD, single ascending doses; MAD, multiple ascending doses.

FIG 1.

Overview of study design.

The median (range) values for demographic characteristics of participants who participated in the SAD cohorts were 24 (18 to 54) years for age, 180 (157 to 188) cm for height, 77 (54.5 to 99.2) kg for body weight, and 24.1 (19.8 to 29.9) for body mass index (BMI). The values of participants who participated in MAD cohorts were 25 (19 to 51) years for age, 178 (165 to 188) cm for height, 74 (54.8 to 91.9) kg for body weight, and 23.7 (18.4 to 30.4) for BMI. The majority of participants were male (98.8% male) and white (74.7% white, 14.4% Asian, 1.2% black, and 9.6% other).

Safety.

RO7049389 were well tolerated in all healthy participants in this study. A total of 59 adverse events (AEs) were reported (Tables S1 and S2 in the supplemental material), with 44 AEs in 30 of 62 healthy participants who received RO7049389 and 15 AEs in 9 of 21 healthy participants who received the placebo. All AEs were of mild intensity, except for one case of upper respiratory tract infection that was moderate in intensity. All AEs have resolved. No dose dependency was observed in the incidence and intensity of AEs in both the SAD and MAD parts of the study. The most common AE was headache in both the SAD and MAD components (6/30 [20.0%] and 3/32 [9.4%], respectively). No severe or serious adverse events (SAEs) were observed in any cohorts, and no AEs leading to drug discontinuation were reported. There were no clinically significant changes in ECG parameters, vital signs, or laboratory safety test results.

PK of RO7049389.

The RO7049389 plasma PK profiles are shown in Fig. 2, and the PK parameters are listed in Table 2. The absorption of RO7049389 occurred rapidly after single-dose administration to healthy participants in SAD cohorts. Peak plasma concentrations were reached within 1.25 to 3.0 h under fasted conditions. A trend toward greater than dose-proportional increases in maximum plasma concentration (C_max) and area under the concentration-time curve from 0 h to infinity (AUC_0-∞) were observed after the administration of single doses ranging from 150 to 2,500 mg (see Table 2 and Fig. 2). RO7049389 was eliminated from plasma with apparent terminal half-life (t_1/2) ranging from 3.3 to 13.9 h. A limited amount of RO7049389 (<1% of the dose) was recovered in urine following single doses with a range of 150 to 2,500 mg (Table 2).

FIG 2.

Mean (±SD) plasma concentration (Conc)-time curves of RO7049389 after single doses of RO7049389 in healthy participants (semilog).

TABLE 2.

PK parameters for RO7049389 in healthy participants^e

Cohorts	Dose (n)	Day	Cmaxa (ng/ml)	AUCa,c (h ng/ml)	Tmaxb (h)	t1/2a (h)	fea (%)
SAD cohorts	150 mg (3), fasted		250 (33.4)	879 (7.5)	2.0 (2.0−2.0)	3.34 (35.3)
	450 mg (6), fasted		1,900 (79.7)	4,690 (53.2)	1.25 (1.0−2.0)	8.80 (81.1)	0.07 (68.2)
	450 mg (6), fed		4,610 (75.6)	10,900 (71.9)	3.0 (1.5−4.0)	4.20 (32.0)
	1,000 mg (9), fasted		8,990 (77.4)	24,400 (100.5)	1.5 (1.5−3.0)	8.50 (35.5)	0.19 (83.7)
	2,000 mg (6), fasted		14,700 (67.6)	39,400 (77.7)	2.0 (1.5−3.0)	13.9 (51.0)	0.18 (91.9)
	2,500 mg (6), fasted		24,800 (63.9)	134,000 (91.4)	3.0 (1.5−4.0)	12.4 (44.7)	0.29 (105.3)
MAD cohorts	200 mg (6), fasted	1	483 (42.5)	1,300 (31.9)	2.0 (1.5−3.0)	2.27 (24.1)	0.07 (74.3)
		14	735 (19.9)	1,810 (25.7)	1.5 (1.0−1.5)	5.79 (40.7)	0.12 (89.5)
	200 mg (6), fed	1d	840 (41.7)	1,820 (46.1)	2.0 (1.5−3.0)	2.44 (12.6)	0.07 (41.5)
		14	583 (41.3)	1,570 (27.5)	2.0 (1.5−4.0)	5.17 (26.8)	0.09 (60.7)
	400 mg (6), fed	1d	4,840 (53.7)	9,760 (55.7)	3.0 (1.5−4.0)	2.23 (23.2)	0.30 (53.3)
		14	2,720 (49.0)	6510 (29.0)	3.0 (1.5−4.0)	5.08 (10.5)	0.31 (53.8)
	600 mg (6), fed	1	10,600 (42.2)	27,000 (33.6)	3.0 (2.0−3.0)	1.53 (22.5)	0.45 (35.0)
		14	9,960 (32.7)	28,100 (32.5)	3.0 (3.0−4.0)	5.73 (28.9)	0.61 (26.3)
	800 mg (6), fed	1	14,500 (39.5)	32,500 (42.6)	3.0 (2.0−4.0)	1.70 (30.4)	0.60 (46.5)

^aPresented as the arithmetic mean (CV%).

^bPresented as the median (range).

^cAUC_∞ for SAD cohort; AUC_τ for MAD cohorts.

^dn = 7.

^eAUC_∞, area under the concentration-time curve from time 0 to infinity; AUC_τ, area under the concentration-time curve from time 0 to 12 h; BID, twice daily; C_max, maximum plasma concentration; T_max, time to maximum concentration; t_1/2, half-life; f_e, fraction of drug excreted unchanged in urine.

The C_max and AUC_∞ of RO7049389 (450 mg) increased by 2.4- and 2.2-fold, respectively, when administered with a high-fat meal with a slight delay in the time to maximum concentration (T_max). The RO7049389 plasma PK profiles with or without food are displayed in Fig. 3.

FIG 3.

Mean (±SD) plasma concentration (Conc)-time curves of RO7049389 after a single dose of 450 mg RO7049389 under fasted (gray line) and fed (black line) condition in healthy participants (linear).

The absorption of RO7049389 after oral administration of multiple doses to healthy participants was also rapid, with the T_max values of 1.0 to 3.0 h, 1.5 to 4.0 h, 1.5 to 4.0 h, and 2.0 to 4.0 h for 200 mg BID fasted or 200, 400, and 600 mg BID fed for 14 days, respectively. The arithmetic mean elimination t_1/2 of RO7049389 at steady state on day 14 was approximately 5.1 to 5.8 h. Limited accumulation of RO7049389 was observed after repeated dosing. AUC accumulation ratios of 1.45, 1.05, 0.88, and 1.09 were observed after 200 mg BID fasted or 200, 400, and 600 mg BID with a standard meal, respectively. A trend toward greater than dose-proportional increases in C_max and AUC_τ (the area under the concentration-time curve from time 0 to 12 h) was observed in the 200-, 400-, and 600-mg BID cohorts dosed with a standard meal. The RO7049389 plasma PK profiles are displayed in Fig. 4.

FIG 4.

Mean (±SD) plasma concentration (Conc)-time curves of RO7049389 after multiple doses of RO7049389 in healthy participants (semilog).

PK of midazolam.

The midazolam plasma PK profiles are shown in Fig. 5. A single 100-μg dose of midazolam was administered prior to and after 14 days of RO7049389 administration in all MAD cohorts to evaluate the perpetrator effect of RO7049389 on the activity of CYP3A. After 14 days, a 9, 15, 20, or 14% decrease in midazolam clearance was observed when CYP3A was given in combination with RO7049389 as 200 mg BID fasted, 200 mg BID fed, 400 mg BID fed, or 600 mg BID fed, respectively. The midazolam results suggest a very weak inhibition (<20%) of CYP3A by RO7049389. The clinical significance of this interaction is considered unlikely, but caution should be used with CYP3A substrates with a narrow therapeutic range when dosed in combination with RO7049389 in future clinical trials.

FIG 5.

Mean (±SD) plasma concentration (Conc)-time curves of midazolam before (gray line) and after (black line) multiple doses of RO7049389 in healthy participants (linear).

Concentration-QT analysis.

The relationship of change in QTcF (the QT interval corrected for heart rate by Fridericia’s method) from the time-matched baseline (i.e., ΔQTcF) and observed RO7049389 concentration was explored and can be seen in Fig. 6. For participants receiving placebo, the concentration value was assumed to 0. The empirical analysis indicated that there was no signal in the ΔQTcF-concentration relationship. The concentration-QTcF modeling will be published in a follow-up manuscript.

FIG 6.

Relationship of ΔQTcF (i.e., the change in QTcF from the time-matched baseline) and RO7049389 concentration. Linear (left) and semilog (right) data are shown.

DISCUSSION

This first-in-human study evaluated the safety, tolerability, PK, food effect, and inhibition effect on CYP3A of RO7049389, an inhibitor of HBV capsid assembly, administered to healthy participants. This study consisted of five components: SAD cohorts, MAD cohorts, food effect assessment, DDI assessment, and concentration-QTc analysis.

Starting dose selection is pharmacology driven and supported by nonclinical safety data. The total plasma C_average of 290 ng/ml was considered efficacious and projected to provide 2-log DNA reduction based on mouse models of chronic HBV infection. The proposed starting dose of 150 mg was predicted to provide a total plasma C_average of 55 ng/ml, which was around one-fifth of the total plasma C_average expected to be efficacious in humans. The 4-week good-laboratory-practice toxicology studies defined the NOAELs (no observed AE levels) as 250 and 350 mg/kg/day in rats and minipigs, respectively. These doses convert to human equivalent doses (HED) of 40 and 318 mg/kg, respectively. Using the lowest HED of 40 mg/kg from the rat NOAEL calculation and a safety factor of 10, the maximum recommended starting dose (MRSD) for a 60-kg individual was 240 mg. The starting dose of 150 mg proposed in this study was ∼1.6-fold lower than the MRSD. Therefore, a single dose of 150 mg was considered to be a safe starting dose for this first-in-human study.

The subsequent dose levels for SAD cohorts and the starting dose for MAD cohorts were not prespecified in this study, whereas the dose selection and escalation criteria were clearly defined. This study design allowed the flexibility on the dose selection for the following cohorts and minimized the protocol amendment. In the SAD part, after the fourth cohort (the 2,000-mg cohort), the dose-normalized exposure of the third cohort (1,000 mg, n = 3) was observed to be unexpectedly low compared to other cohorts. Considering the limited sample size of the third cohort and the large PK intersubject variability, the 1,000-mg cohort was repeated in the fifth cohort with a sample size of six active and two placebo subjects. In the MAD part, PK parameters in the fourth cohort (n = 6; 800 mg, BID fed) were only available to day 3 due to the decision to suspend dosing after day 3 pending review of newly available RO7049389 clinical metabolite data, which was disproportionately higher than in animal species. The decision was a precautionary one and was not based on any newly identified safety concern. Dosing was subsequently continued in the fifth MAD cohort (600 mg, BID fed) that was initiated after further DDI and safety evaluations of the human metabolites.

The solubility of RO7049389 in biorelevant media was low and pH dependent. To provide guidance with regard to food restriction for subsequent clinical trials, the effect of food on the PK of RO7049389 will be evaluated in one of the SAD cohorts. Because the exposure in humans was not known before this study, the dose for the food effect, 450 mg, was selected to be efficacious based on the emerging PK data of 150 mg (the first cohort of SAD part) and estimated efficacious target from nonclinical data.

A critical aspect of drug development is to predict clinically relevant DDIs. Based on in vitro data, RO7049389 is a weak CYP3A time-dependent inhibitor. At the early stage of drug development, it is generally challenging to choose a therapeutically relevant dose to assess drug interaction liability. The investigation of a wide range of RO7049389 dose levels would provide a comprehensive understanding on the dose versus CYP3A inhibition relationship and help to define the optimal dose range for phase II studies from the DDI perspective. It has been reported that the magnitudes of CYP3A inhibition detected by microdose of midazolam are similar to those by therapeutic dose of midazolam, suggesting that this unobstructive microdose approach can be applied to assess CYP3A activity as an alternative to the regular approach using the milligram dose level of midazolam (5). The dose selection for midazolam in this study (100 μg) is far lower than the therapeutic dose levels and considered a microdose according to ICH guidance (6) (ICH guideline M3). Giving 100 μg of midazolam to the study participant was not expected to affect the safety assessment of RO7049389 when given together on day 14.

RO7049389 was rapidly absorbed and eliminated in plasma. Therefore, it can quickly reach the steady state. Also, no accumulation was observed after multiple twice-daily administration. Exposure increased in a greater-than-proportional manner across the dose range. Possible reasons for this disproportionate increase in bioavailability at higher doses may include the saturation of the capacity of a metabolic route (CYP3A) and/or the saturation of the transporter (OATP1B).

Conclusions.

This first-in-human study of RO7049389 demonstrated that RO7049389 was well tolerated at a single dose of up to 2,500 mg, and multiple doses of up to 400 mg BID × 14 days and 800 mg BID × 3 days. The absorption and elimination of RO7049389 occurred rapidly in plasma with minimal recovery in urine. Greater than dose-proportional increases in plasma exposure were observed. Exposure of RO7049389 (450 mg) increased by ∼2-fold when it was administered with a high-fat meal. The inhibition effect of RO7049389 on CYP3A was weak (<20%). No effect on ΔQTcF was observed up to a single dose of 2,500 mg. These results show that RO7049389 displays a favorable safety, tolerability, and PK profile that make it suitable for further clinical development.

MATERIALS AND METHODS

The study was performed at the Auckland Clinical Studies, Auckland, New Zealand. The study was approved by the local ethics committee and was performed in accordance with the Declaration of Helsinki. Written informed consent was obtained from all study participants. The trial was registered at ClinicalTrials.gov with the identifier NCT02952924.

Participants.

Included in the study were healthy males and females aged 18 to 60 years, with negative drug screening tests and an agreement to refrain from any other drugs (including vitamins, herbal supplements, and over-the-counter or prescription medications) for the study duration. All participants were ascertained to be healthy on the basis of their medical history, physical examination, virology testing (hepatitis B and C and human immunodeficiency virus), and routine laboratory testing (liver and kidney function and hematology) before enrollment into the study. All male participants agreed to remain abstinent (refrain from heterosexual intercourse) or use contraceptive measures and to refrain from donating sperm. All female participants were either surgically sterile or postmenopausal for at least 1 year. Excluded from the study were participants with prior allergic drug reactions or clinically significant concomitant diseases or conditions.

Study design.

This was a randomized, sponsor-open, investigator-blinded, participant-blinded, placebo-controlled, adaptive single- and multiple-ascending-dose study designed to evaluate the safety, tolerability, PK, food effect, and inhibitory effect on CYP3A of RO7049389 following oral administration of single ascending doses (SAD cohorts) or multiple ascending doses (MAD cohorts) in healthy participants (Fig. 1). The effect of food on the PK of RO7049389 was evaluated within one of the SAD cohorts. Drug interaction with microdose midazolam was assessed in all MAD cohorts. Except the starting dose, no other doses were specified in the study design. However, the dose escalation and selection criteria were clearly defined in the protocol.

(i) Single-ascending-dose cohorts.

The first cohort in the SAD part of the study would receive a single dose of 150 mg. Doses for subsequent cohorts would be defined by an adaptive approach based on the safety and PK data from previously dosed healthy participants. Sentinel dosing of two participants (one active, one placebo) was used for the first cohort. After acceptable safety assessment of the first two participants by the investigator, and at least 24 h after the initial administration, three more participants (two active, one placebo) would be dosed. For subsequent SAD cohorts, four or eight (the sample size would increase to eight at dose levels at or above the projected efficacious exposure) healthy participants would be enrolled in a 3:1 ratio to receive RO7049389 and placebo. The doses in this part would be administered under fasted condition.

(ii) Food effect cohort.

The effect of food on the PK of RO7049389 would be evaluated within one of the SAD cohorts. The food effect cohort would be selected such that the expected exposure of RO7049389 will be efficacious, based on the safety and PK data from previous SAD cohorts, and estimated an efficacious target from the nonclinical data. After receiving one single dose of RO7049389 on day 1, eight healthy participants (six active, two placebo) were administered either a single dose same as day 1 of RO7049389 or placebo on day 16 with a high-fat and high-calorie breakfast. The fat content of the breakfast was approximately 50% of the total caloric content of the meal. The meal was high-calorie (approximately 800 to 1,000 calories).

(iii) Multiple-ascending-dose cohorts.

The first cohort in the MAD part of the study would start when exposure in the SAD dose level had reached 0.5-fold or higher of the predicted efficacious exposure. The starting dose of the MAD arm of the study would be chosen such that the expected steady-state exposure of RO7049389 would be 0.5-fold of the expected efficacious exposure or higher but less than or equal to the exposure that has been explored and well tolerated in previous SAD cohort(s). Up to five cohorts of eight healthy participants would be enrolled. In each MAD cohort, six healthy participants were randomized to receive RO7049389, and two were randomized to receive placebo. Healthy participants in the MAD cohorts would be dosed with RO7049389 for 13 days (once or twice daily, depending on the available PK and safety data). On day 14, only one dose in the morning would be given regardless of the dosing regimen. Potential drug interaction with microdose midazolam would be assessed in all MAD cohorts. A single oral dose of midazolam solution (100 μg) would be administered on days −1 and 14.

Dose escalation.

The decision to escalate to the next dose-level at a previously studied dose was based primarily on safety and tolerability data and secondarily on available PK data at the previous dosage level. A dose would not be escalated to a >3-fold increase in AUC from one cohort to the next. Dose escalation would continue according to the study design until the maximum tolerated dose (MTD) was identified or the systemic exposure reached an exposure ∼10-fold above the projected efficacious exposure or the exposure attained at NOAEL in the minipig. The MTD was defined by the incidence of dose-limiting events. Doses might be repeated, reduced, or escalated based on emerging safety, tolerability, and PK data at each dose level.

Safety assessment.

Vital signs, physical examination, ECGs, safety laboratory variables, and AEs were monitored throughout the study. Laboratory measures included hematology, clinical chemistry, coagulation, and urinalysis.

The QT interval was obtained from continuous 12-lead ECG recordings (Holter monitor) and corrected for heart rate by Fridericia’s method (QTcF). A central ECG laboratory extracted triplicate ECG measurements from the continuous recordings within ±10 min for PK time points (1, 2, 3, 4, 6, 12, and 24 h on day −1 and day 1 for all SAD and MAD cohorts and also day 14 for all MAD cohorts).

Pharmacokinetics assessment.

In SAD cohorts and the food effect cohort, RO7049389 plasma pharmacokinetic samples were collected predose and 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, 12, 24, 36, 48, 72, and 96 h postdose. In the MAD cohorts, RO7049389 plasma pharmacokinetic samples were collected predose and 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, and 12 h after the first dose (day 1); predose on days 2, 3, 4, 5, and 7; and predose and 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, 12, 24, 36, and 48 h after the last morning dose (day 14). Midazolam plasma pharmacokinetic samples were collected predose and 0.25, 0.5, 1, 2, 4, 6, 8, 10, 12, and 24 h postdose on both day −1 and day 14. In SAD cohorts, RO7049389 urine PK samples were collected at five intervals between 0 to 4 h, 4 to 8 h, 8 to 12 h, 12 to 24 h, and 24 to 48 h postdose. In MAD cohorts, RO7049389 urine PK samples were collected at four time intervals between 0 to 4 h, 4 to 8 h, 8 to 12 h, and 12 to 24 h postdose on days 1 and 14.

Plasma and urine concentrations of RO7049389 and plasma concentrations of midazolam were measured using specific and validated liquid chromatography-electrospray ionization-tandem mass spectrometry methods. The PK parameters were calculated from a noncompartment analysis using Phoenix software (WinNonlin models, v6.4; Pharsight Corp., Cary, NC).

Statistic methods.

Statistical summaries were descriptive in nature. The number of study participants to be randomized was chosen based on practical considerations and complies with standard safety review rules. With six study participants receiving active drug at a dose level, there is an 82% chance to observe at least one AE that has an incidence rate of 25% in the population. The geometric mean ratios and 90% confidences were calculated using SAS software, v9.4 (SAS Institute, Cary, NC).