A phase I, first‐in‐human study investigating the safety, tolerability, and pharmacokinetics of the maturation inhibitor GSK3739937

Abstract

We report the safety and pharmacokinetic properties of the HIV‐1 maturation inhibitor GSK3739937 (GSK'937) in healthy participants. This was a phase I, first‐in‐human, double‐blind, randomized, placebo‐controlled, single‐ (part 1) and multiple‐ (part 2) dose escalation study with an additional open‐label relative bioavailability and food effect study (part 3). Participants received oral ascending single doses (10–800 mg) in part 1, up to 18 once‐daily 25‐ to 100‐mg or 3 once‐weekly 500‐mg doses in part 2, and single 100‐mg doses as powder‐in‐bottle or tablet (in fed and fasted states) formulations in part 3. Primary and secondary objectives were safety and pharmacokinetic assessments, respectively. Ninety‐one participants were enrolled; 38 reported 81 total adverse events (AEs). All AEs in participants receiving GSK'937 were grade 1 or 2 and resolved during the study. Most drug‐related AEs were gastrointestinal (14/17, 82%). The terminal phase half‐life of GSK'937 was ~3 days for all doses following single and repeat dosing. Geometric mean maximum concentration and total drug exposures exhibited dose‐proportional increases during part 1. Accumulation in exposure following repeat dosing was 6‐ to 7‐fold with daily dosing and ~1.7‐fold after weekly treatment, as expected due to the long half‐life. Bioavailability of GSK'937 after a meal was 1.35‐ to 1.40‐fold greater as a tablet versus powder‐in‐bottle and >2‐fold higher in fed versus fasted states when provided as a tablet. No unexpected or dose‐limiting safety events occurred. Pharmacokinetic parameters of long half‐life and accumulation of exposure following repeat dosing suggest the potential for weekly oral dosing. ClinicalTrials.gov identifier: NCT04493684.

For most doses in the single‐dose administration study and all doses in the repeat‐dose administration study, the HIV‐1 maturation inhibitor GSK3739937 remained detectable in the plasma of healthy participants for at least 2 weeks after the last dose. Due to this generally long half‐life, repeated daily oral dosing resulted in dose‐proportional accumulation of GSK3739937; however, more limited accumulation was observed after once‐weekly dosing. Overall, the data support further development of GSK3739937 as an oral therapy administered once per week.

Abbreviations

AE

adverse event

ART

antiretroviral therapy

ARV

antiretroviral

AUC

area under the concentration‐time curve

AUC_0‐∞

AUC from time 0 (pre‐dose) extrapolated to infinity

AUC_0‐t

AUC from time 0 to last quantifiable concentration

C _max

maximum observed concentration

C _trough

minimum observed concentration of the dosing interval

GI

gastrointestinal

MAD

multiple ascending dose

MDSD

medical device site dermatitis

MED

minimally effective dose

MI

maturation inhibitor

MRSD

maximum recommended starting dose

NNRTI

non‐nucleoside reverse transcriptase inhibitor

PiB

powder in bottle

PK

pharmacokinetic

QD

once daily

QW

once weekly

R_oAUC_0‐trough

accumulation ratio of AUC_0‐trough

R_o C _max

accumulation ratio of C _max

R_o C _trough

accumulation ratio of C _trough

SAD

single ascending dose

SAE

serious adverse event

t _1/2

half‐life

t _max

time to C _max

1. INTRODUCTION

As of 2021, there were 38.4 million people living with HIV, an increase of 1.5 million from 2020. ¹ Fortunately, therapeutic advances have turned this once‐terminal diagnosis into a chronic but manageable condition. ² Lifelong treatment comes with its own set of challenges, however. Chronic exposure to antiretrovirals (ARVs) may be associated with several long‐term comorbidities (eg, chronic kidney disease, diabetes mellitus) and further complicated by the development of age‐associated comorbidities and adverse drug–drug interactions with treatments for those non‐HIV conditions. ³ Minimizing the long‐term effects of ARVs has been a reason for the development of 2‐drug regimens (instead of 3‐ or 4‐drug regimens) that partner an integrase strand transfer inhibitor with a nucleoside reverse transcriptase inhibitor or non‐nucleoside reverse transcriptase inhibitor (NNRTI). ⁴ The durability of available 2‐drug regimens has been well established ⁵ , ⁶ ; however, currently established 2‐drug regimens are not recommended for use in people with previous virologic failure ⁷ , ⁸ due to the risk of pre‐existing viral resistance to ARVs in the regimens. ⁹ , ¹⁰ Thus, exploring development of ARVs with novel mechanisms of action could provide an alternative partner for integrase inhibitors that would expand the use of oral or long‐acting injectable 2‐drug regimens to individuals with NNRTI or nucleoside reverse transcriptase inhibitor resistance.

HIV‐1 maturation presents a promising therapeutic target for the development of new ARVs. ¹¹ During maturation, the HIV‐1 protease cleaves the structural polyproteins Gag and Gag‐Pro‐Pol, inducing morphological changes within the virion. ¹² These events enable the virus to fuse with target cells ¹³ and successfully reverse transcribe its genome. ¹⁴ , ¹⁵ Even minor disruptions in Gag processing can have disproportionately large effects on viral infectivity. ¹⁴ , ¹⁵ , ¹⁶ The first‐in‐class HIV‐1 maturation inhibitor (MI) bevirimat disrupted cleavage of spacer peptide 1 from the C‐terminus of the capsid region of Gag and had potent antiviral activity, ¹⁷ demonstrating proof‐of‐concept of this novel ARV mechanism of action. Although bevirimat had favorable safety and efficacy profiles during early‐phase clinical trials, ¹⁸ its development was ultimately halted due to a high frequency of naturally occurring viral resistance mutations. ¹¹ Subsequent investigational MIs with inhibitory activity against a wider variety of Gag sequences have included GSK3532795 (BMS‐955176), ¹⁹ GSK2838232, ²⁰ and GSK3640254. ²¹

GSK3739937 (GSK'937), also known as VH3739937, is a new HIV‐1 MI with a similar chemical structure and profile to GSK3640254 that likewise interferes with capsid/spacer peptide 1 cleavage. Like GSK3640254, it has low nanomolar potency in vitro, exhibits significantly improved pan‐genotypic coverage and potency against Gag polymorphisms relative to prior MIs, and demonstrates a high barrier to the emergence of resistant viruses (except for A364V). ²² Unlike other developmental HIV‐1 MIs, which have pharmacokinetic (PK) profiles consistent with once‐daily dosing schedules, ²⁰ , ²¹ , ²³ GSK'937 is projected to have a dosing schedule less frequent than once‐daily. In this report, we investigate the safety, tolerability, and PK of oral GSK'937 in a phase I clinical study of healthy participants.

2. METHODS

This phase I study (ClinicalTrials.gov identifier, NCT04493684) consisted of 3 parts (Figure 1). Parts 1 and 2 included single‐ascending dose (SAD) and multiple‐ascending dose (MAD) escalation cohorts, respectively, designed to evaluate the safety, tolerability, and PK of orally administered GSK'937. Part 3 compared the bioavailability of powder‐in‐bottle (PiB) and tablet formulations as well as the effect of food on tablet bioavailability. Parts 1 and 2 were double‐blind (sponsor‐unblinded), randomized, and placebo‐controlled; part 3 was open‐label. Enrollment occurred at 1 site in the United States between July 2020 and August 2021. The study protocol, informed consent, and other information that required pre‐approval were reviewed and approved by the investigational center's institutional review board. All activities conformed to the ethical principles outlined in the Declaration of Helsinki.

FIGURE 1.

Study design for the single ascending dose (part 1), multiple ascending dose (part 2), and relative bioavailability and food effect (part 3) parts of NCT04493684 evaluating safety and PK in healthy participants. (A) Part 1 comprised a cross‐over design between 2 cohorts of 9 people for treatment with single 10‐ to 640‐mg doses. Part 2 included 4 independent cohorts of 10 participants each receiving escalating QD doses for up to 18 days or three 500‐mg QW doses. Part 3 comprised 1 cohort of 17 individuals randomly assigned to 1 of 3 treatment sequences with the dosing strategies of PiB with a moderate‐fat meal, tablet with a moderate‐fat meal, and tablet fasted. (B) Diagram of the possible treatment sequences for participants in cohorts 1, 2, and 7. GSK'937, GSK3739937; MAD, multiple ascending dose; PiB, powder in bottle. QD, once daily; QW, once weekly; PK, pharmacokinetics; RBA + FE, relative bioavailability and food effect; SAD, single ascending dose; SEQ, sequence. ^aEach individual was planned to receive GSK'937 in 2 of 3 treatment periods and the placebo once, with 6 participants receiving GSK'937 and 3 receiving the placebo each period. ^bEstimated minimally effective dose. ^cThe entirety of cohort 1 was replaced when the 800‐mg dose was added to avoid over‐representation of participants in different treatment groups. ^dEach group included 7 participants receiving GSK'937 and 3 receiving a placebo.

2.1. Study participants

Participants were adults aged 18–55 years with a body mass index of 18.5–32.0 kg/m². Negative polymerase chain reaction tests for COVID‐19 were required at screening and admission. Female participants were excluded if pregnant, breastfeeding, or of child‐bearing potential. During the study, participants who discontinued for non‐safety reasons were replaced at the discretion of the sponsor in consultation with the investigator. Written informed consent was obtained from each participant before the performance of any study‐specific procedures.

2.2. Study design

2.2.1. Single ascending dose (part 1)

Part 1 included 2 cohorts of 9 individuals each (cohorts 1 and 2, Figure 1). For the first treatment period, cohort 1 participants were randomized 6:3 to receive GSK'937 10 mg or placebo. This dose was lower than the maximum recommended starting dose (MRSD) estimated from the dose ranging and good laboratory practice toxicology studies in rats, in which a no observed adverse effect level was determined (10 mg/kg/day; MRSD = 11.3 mg), and in dogs, in which a lowest observed adverse effect level was determined (3 mg/kg; MRSD = 11.7 mg). ²⁴ Cohort 2 participants were randomized 6:3 to receive GSK'937 30 mg or placebo in the second treatment period. Subsequent dose increases continued alternating between cohorts 1 and 2 for a total of 6 periods, though each cohort was re‐randomized every period so each participant received the placebo once. Dose escalations did not exceed 3‐fold up to the estimated minimally effective dose (MED; 80 mg) and did not exceed 2‐fold thereafter. The initial MED was estimated from a PK profile predicted with allometric scaling and steady‐state plasma drug concentration‐mean residence time models generated with the toxicokinetic data from animal studies. The highest dose was anticipated to be 640 mg; however, the exposure at this dose was lower than expected. Therefore, a seventh period with an 800‐mg dose was added. For GSK'937 800 mg, all 9 participants of cohort 1 were new study participants and randomized 6:3 to receive the drug or placebo. New individuals were enrolled to avoid over‐representation of participants in different SAD treatment groups dosed and evaluated before this 800‐mg group. All doses were administered immediately following a moderate‐fat breakfast (600 calories, 30% fat). The protocol originally specified a 96‐h follow‐up period but was amended to a 360‐h follow‐up period after the half‐life (t _1/2) of GSK'937 was calculated from the results of Cohort 1 (t _1/2 = 67–97 h). This extended observation period was continued in later parts of the study.

2.2.2. Multiple ascending dose (part 2)

Part 2 comprised 40 participants randomized into 4 cohorts of 10 (cohorts 3–6, Figure 1). Beginning with cohort 3, 7 participants were randomly assigned to receive GSK'937 25 mg once daily (QD) for 14 days, and 3 were assigned placebo. The 25‐mg dose represented an updated estimate for the MED utilizing data from part 1. Dose escalation continued through cohorts 4 and 5, with the dosing period for cohort 5 lasting 18 days instead of 14. The 2‐fold dose escalations between cohorts 3, 4, and 5 were decided by the available safety and PK data. Cohort 6, which was added to the study after t _1/2 measurements made in part 1, received 3 once‐weekly (QW) doses of GSK'937 500 mg or placebo (7:3). The 500‐mg QW dose was determined from simulations based on data acquired in parts 1 and 2. Participants received all doses after a moderate‐fat meal.

2.2.3. Relative bioavailability and food effect (part 3)

Part 3 was an open‐label, single‐dose, 3‐period cross‐over study. Seventeen participants were randomly assigned to 1 of 3 treatment sequences with each treatment period separated by a ≥24‐day washout (cohort 7, Figure 1). All treatments were 100 mg, an amount chosen based on available PiB PK data and projected clinical doses of 25 mg QD or 500 mg QW. Participants fasted overnight for at least 9.5 h before dosing and again for at least 4 h after. Those in the fed group received a moderate‐fat meal 25 min before dosing.

2.3. Study assessments

2.3.1. Single and multiple ascending dose (parts 1 and 2)

The primary objective of parts 1 and 2 was to determine the safety and tolerability of GSK'937. Safety assessments included monitoring adverse events (AEs); serious adverse events (SAEs); hematology, clinical chemistry, and urinalysis laboratory analyses; vital signs; electrocardiograms (ECGs); and suicidal ideation. All AE data were tabulated using the Medical Dictionary for Regulatory Activities preferred terms and summarized descriptively according to GSK's Integrated Data Standards Library.

The secondary objective of parts 1 and 2 was obtaining GSK'937 PK profiles, which included evaluating GSK'937 accumulation after single, QD, and QW administrations. Pharmacokinetic parameters included maximum observed concentration (C _max), time to C _max (t _max), area under the concentration‐time curve (AUC) from time 0 (pre‐dose) extrapolated to infinity (AUC_0‐∞), AUC from time 0 to last quantifiable concentration (AUC_0‐t ), and t_1/2. In repeated dose conditions, minimum observed concentration of the dosing interval (C _trough) and accumulation ratios of AUC_0‐trough (R_oAUC_0‐trough), C _max (R_o C _max), and C _trough (R_o C _trough) were also assessed.

2.3.2. Relative bioavailability and food effect (part 3)

The 2 primary objectives in part 3 were (1) to assess the safety and tolerability of GSK'937 after single oral administration in healthy participants under fasted or fed conditions and (2) to evaluate the relative bioavailability of GSK'937 when administered as a tablet versus PiB. A secondary objective was measurement of the PK profile of the tablet formulation of GSK'937 under fasted and fed conditions. Assessments included the safety and PK parameters listed in parts 1 and 2.

2.4. Data analyses

No formal statistical hypotheses were tested. Sample size was based on feasibility. No formal calculation of power for parts 1 and 2 was performed. For part 3, 18 participants were enrolled to ensure at least 12 participants completed the study. Safety endpoints were summarized descriptively and not subjected to formal statistical analyses. Plasma concentration‐time data were calculated with standard non‐compartmental methods using Phoenix WinNonlin Professional (Pharsight Corporation). For each PK parameter, summary statistics were calculated by treatment group. Dose proportionality of selected single‐ and repeat‐dose PK parameters was assessed by the power model. Accumulation ratios in part 2 were calculated as Day 14, 15, or 18 to Day 1.

For assessment of relative bioavailability in part 3, the log_e‐transformed PK parameters were analyzed using separate mixed‐effect models with a fixed‐effect term for fed versus fasted and tablet versus PiB. Participants were treated as a random effect in each model. Point estimates and their associated 90% CIs were calculated for the difference in AUC from 0 to 24 (AUC_0‐24), AUC_0‐∞, and C _max. The point estimates and their associated 90% CIs were back‐transformed to provide point estimates and 90% CIs in PK parameter values on the original scale.

3. RESULTS

3.1. Participant disposition and demographics

Ninety‐one individuals participated in the study. Most participants were male (87/91, 96%; Table 1) and Black/African American (54/91, 59%). Across cohorts, the mean age ranged from 29.0 to 41.0 years and the mean body mass index ranged from 25.2 to 27.5 kg/m². Seven participants withdrew from part 1 (4 from cohort 1, 3 from cohort 2) and were replaced. No withdrawals were treatment‐related, instead stemming from participant decisions (n = 3), physician decisions (n = 2; failed baseline clinical labs and positive test result for trypanosomiasis), a protocol deviation (n = 1), and COVID‐19 infection (n = 1). In parts 2 and 3, 1 of 40 and 2 of 17 participants withdrew due to a family emergency and physician decisions (low hemoglobin and non‐compliance with follow‐up visits), respectively, and were not replaced.

TABLE 1.

Demographics and baseline characteristics.

Parameter	Ascending dose studies (Parts 1 and 2)							Part 3
	SAD	MAD						RBA + FE
	Cohorts 1–2	Placebo		Cohort 3	Cohort 4	Cohort 5	Cohort 6	Cohort 7
	10–800 mg (N = 34) a	QD (N = 9) b	QW (N = 3)	25 mg QD (N = 7)	50 mg QD (N = 7)	100 mg QD (N = 7)	500 mg QW (N = 7)	100 mg (N = 17)
Age, y, mean (SD)	39.1 (9.2)	40.2 (8.2)	29.0 (6.0)	41.0 (7.5)	35.1 (9.2)	41.0 (11.3)	36.9 (10.3)	34.9 (11.4)
BMI, kg/m2, mean (SD)	27.3 (2.9)	25.2 (2.4)	27.2 (5.2)	27.5 (2.5)	26.1 (4.1)	25.4 (4.1)	26.9 (2.0)	25.5 (2.7)
Gender, n (%)
Male	32 (94)	8 (89)	3 (100)	7 (100)	7 (100)	6 (86)	7 (100)	17 (100)
Female	2 (6)	1 (11)	0	0	0	1 (14)	0	0
Ethnicity, n (%)
Hispanic/Latino	6 (18)	3 (33)	0	3 (43)	3 (43)	1 (14)	3 (43)	4 (24)
Not Hispanic/Latino	28 (82)	6 (67)	3 (100)	4 (57)	4 (57)	6 (86)	4 (57)	13 (76)
Race, n (%)
Black	21 (62)	6 (67)	2 (67)	2 (29)	4 (57)	3 (43)	4 (57)	12 (71)
White	11 (32)	3 (33)	0	3 (43)	3 (43)	4 (57)	3 (43)	5 (29)
Asian	0	0	1 (33)	1 (14)	0	0	0	0
Mixed race	2 (6)	0	0	1 (14)	0	0	0	0

Abbreviations: BMI, body mass index; MAD, multiple ascending dose; NR, not reported; QD, once daily; QW, once weekly; RBA + FE, relative bioavailability and food effect; SAD, single ascending dose; SD, standard deviation.

^a Seven participants withdrew from part 1 and were replaced. The replaced participants were assigned to the same treatment sequence starting where the previous participant was prematurely discontinued. Previously administered doses were not repeated.

^b Aggregate of placebo recipients from cohorts 3–5.

3.2. Safety

3.2.1. SAD

A total of 36 AEs occurred among 17 of 34 (50%) participants. Most AEs were of grade 1 severity (28/36, 78%; Table 2) or grade 2 (7/36, 19%). Fourteen AEs (gastrointestinal [GI] disorders, n = 12; headache, n = 2) were treatment‐related. In general, GI disorders (Table S1), specifically diarrhea (n = 5), constipation (n = 4), abdominal pain (n = 3), and nausea (n = 2) were the most frequent AEs. Medical device site dermatitis (MDSD; caused by ECG leads) was the next most common AE, occurring in 7 participants. Except for diarrhea, no study treatment‐related or dose‐related pattern was observed for any of the AEs. All grade 1 and 2 AEs resolved within a timespan ranging from a few minutes to days. No deaths or SAEs occurred. An asymptomatic COVID‐19 infection accounted for the only AE leading to discontinuation. A single grade 4 AE occurred in the placebo group of the 800‐mg treatment period during the outpatient follow‐up in which the participant exhibited a grade 4 increase in transaminases that likely resulted from alcohol consumption. The increase in transaminases was considered unrelated to study treatment and lowered in severity during the remainder of the follow‐up period to grade 1. No other clinically meaningful changes were observed in laboratory chemistries, hematology parameters, liver function parameters, renal function parameters, vital signs, ECG parameters, or suicidality.

TABLE 2.

Adverse events by severity and incidence by preferred term ^a .

AE severity or preferred term, n occurrences (% of cohort) b	SAD c							MAD d						RBA + FE e
	PBO	10 mg	30 mg	80 mg	320 mg	640 mg	800 mg	PBO QD	PBO QW	25 mg QD	50 mg QD	100 mg QD	500 mg QW	PiB fed	Tablet fed	Tablet fasted
All‐cause	7 (33)	3 (17)	5 (50)	4 (50) f	5 (33)	8 (67)	4 (33)	8 (33)	0	0	3 (29)	8 (71)	7 (57)	2 (13)	7 (24)	7 (40)
Grade 2	2 (10)	0	1 (17)	0	1 (17)	2 (17)	1 (17)	0	0	0	1 (14)	0	0	0	3 (12)	2 (13)
Grade 3/4	1 (5) g	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
Drug‐related	1 (5) h	3 (17)	1 (17)	1 (17)	4 (17)	2 (17)	2 (33)	0	0	0	0	1 (14)	3 (43)	0	0	0
Grade 2	0	0	1 (17)	0	0	2 (17)	1 (17)	0	0	0	0	0	0	0	0	0
Grade 3/4	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
MDSD	1 (5)	0	1 (17)	0	0	4 (67)	1 (17)	1 (11)	0	0	1 (14)	1 (14)	1 (14)	0	0	0
Headache	0	1 (17)	0	1 (17)	0	0	0	0	0	0	1 (14)	0	0	0	2 (12)	2 (13)
Constipation	1 (5)	0	2 (33)	0	0	1 (17)	0	0	0	0	0	0	0	0	0	0
Diarrhea	1 (5)	0	0	0	1 (17)	1 (17)	2 (33)	0	0	0	0	0	3 (43)	0	0	0
Abdominal pain	0	1 (17)	0	0	1 (17)	1 (17)	0	0	0	0	0	0	1 (14)	0	0	0

Abbreviations: AE, adverse event; MAD, multiple ascending dose; MDSD, medical device site dermatitis (due to electrocardiogram leads); PiB, powder in bottle; PBO, placebo; QD, once daily; QW, once weekly; RBA + FE, relative bioavailability and food effect; SAD, single ascending dose.

^a Listed AEs must have occurred in ≥4 participants.

^b Some participants reported more than 1 AE and/or more than 1 instance of the same AE.

^c N = 21 for PBO, N = 6 for all doses.

^d N = 9 for PBO QD, N = 3 for PBO QW, N = 7 for all doses.

^e N = 15 for PiB fed, N = 17 for tablet fed, N = 15 for tablet fasted.

^f One grade 1 AE (COVID‐19 diagnosis) resulted in the participant's discontinuation from the study.

^g Elevated alanine transaminase.

^h Diarrhea.

3.2.2. MAD

Fourteen of 40 (35%) participants reported a total of 29 AEs, all grade 1 except for a single grade 2 headache in the 50‐mg group (Table S1). No participants receiving the estimated MED (25 mg) reported any AEs, and there were no deaths or SAEs. The only AEs observed in more than 1 participant were MDSD (4/14, 29%), diarrhea (3/14, 21%), and vessel puncture site bruise (2/14, 14%). All 3 instances of diarrhea were among recipients of GSK'937 500‐mg QW dose and were considered treatment‐related (Table 2). All reported AEs resolved during the study without intervention. No significant changes were measured in laboratory analyses, clinical parameters, or suicidality.

3.2.3. Relative bioavailability and food effect

Sixteen AEs were reported by 7 of the 17 (41%) participants, of which only headache (n = 4 among 3 participants) and skin abrasion (n = 2) occurred in more than 1 participant (Table S1). None were considered drug‐ or treatment‐related (Table 2). No deaths or SAEs occurred, all AEs were grade 1 or 2 in severity, and all AEs resolved. The 5 grade 2 AEs were headache (n = 4) and a skin laceration. No significant changes were measured in laboratory analyses, clinical parameters, or suicidality.

3.3. Pharmacokinetic results

3.3.1. SAD

Median t _max was 7–8 h for all doses except the 800‐mg dose, which had a t _max of 18 h (Table 3). Geometric mean AUC_0‐t demonstrated a dose‐proportional increase in exposure (slope, 1.000 [90% CI, 0.926, 1.07]). Although the 640‐mg dose caused a slight deviation from proportionality in C _max (slope, 0.914 [90% CI, 0.855, 0.966]; Figure 2), inter‐patient variability was highest for the 640‐mg treatment period (52–82 %CVb). GSK'937 concentrations were quantifiable up to 96 h after the 10‐mg dose, 264 h in 5 of 6 participants after the 30‐mg dose, and 360 h in participants of all other doses except for 1 individual in the 80‐mg cohort. There were no dose‐dependent trends for the estimated t_1/2 of GSK'937, which ranged from 67.4 to 96.8 h.

TABLE 3.

Pharmacokinetic data for single ascending dose administration of GSK'937.

Dose	n a	t max, h median (range)	C max, ng/mL GM (%CVb)	AUC0‐t , h·ng/mL GM (%CVb)	t 1/2 GM (%CVb)
10 mg	6	6.98 (6.0–8.0)	39.2 (41.6)	2486 (46.4)	ND
30 mg	6	8.00 (5.0–12.1)	127 (22.9)	11 441 (31.5)	75.9 (7.4)
80 mg	6	7.98 (5.0–24.1)	310 (21.0)	33 650 (42.5)	79.1 (14.4)
160 mg	6	7.95 (8.0–24.0)	529 (13.5)	64 938 (18.4)	91.8 (13.6)
320 mg	6	6.99 (6.0–8.0)	1014 (30.7)	90 730 (25.9)	96.8 (22.7)
640 mg	6	7.00 (5.0–24.2)	1458 (52.0)	158 068 (70.0)	82.6 (9.0)
800 mg	6	18.08 (12.1–24.1)	2863 (22.2)	293 637 (33.6)	67.4 (9.3)

Abbreviations: AUC_0‐t , area under the concentration‐time curve from time 0 to last quantifiable concentration; C _max, maximum concentration; CVb, between‐participant coefficient of variation; GM, geometric mean; GSK'937, GSK3739937; ND, not determined; t _max, time to C _max; t _1/2, apparent terminal phase half‐life.

^a n for t_1/2 is ND, 5, 4, 5, 5, 3, and 6 in order of 10 to 800 mg.

FIGURE 2.

Mean plasma concentration of GSK'937 in plasma during the single ascending dose part. (A) Linear and (B) semi‐logarithmic scale concentration‐time curves. The dashed line in B represents the lower limit of quantification (10 ng/mL). (C, D) Correlation plots showing the relationship between C _max and dose or AUC_0‐t and dose, respectively. AUC_0‐t , area under the concentration‐time curve from time 0 to last quantifiable concentration; C _max, maximum observed concentration; GSK'937, GSK3739937.

3.3.2. MAD

Among the 25‐ to 100‐mg doses, both C _max and AUC_0‐t increased proportionately to dose on Day 1 (Figure 3A, Table 4). This dose proportionality continued with repeated daily dosing, though the magnitude of the exposure difference between doses increased over time due to GSK'937 accumulation (Figure 3B). Six‐ to 7‐fold accumulation was observed in the QD groups. Additionally, QW dosing with GSK'937 500 mg resulted in only ~1.7‐fold accumulation. Decay rates of exposure were similar regardless of administration interval (Figure 3C). Median t _max ranged from 7 to 9 h across all time points and doses. GSK'937 t_1/2 ranged from 72.2 to 85.8 h and was still detectable in all part 2 participants 360 h after their final dose.

FIGURE 3.

Mean plasma concentration of GSK'937 in linear and semi‐logarithmic scale during the multiple ascending dose part. (A) Mean concentrations of GSK'937 during the first 24 h after the initial dose. (B) Mean C _trough by day. For QD dosing, values were determined from plasma collected immediately before subsequent doses. For QW dosing, values were collected in 24‐h intervals and immediately before subsequent doses when applicable. (C) Mean concentrations of GSK'937 at receipt of the last dose (0 h) until the end of the observation period (336 h). Dashed line in each lower panel represents the lower limit of quantification (10 ng/mL). C _trough, minimum observed concentration of the dosing interval; GSK'937, GSK3739937; QD, once daily; QW, once weekly.

TABLE 4.

Pharmacokinetic data for multiple ascending dose administration of GSK'937.

Dose	n	t max, h median (range)	C max, ng/mL GM (%CVb)	C trough, a ng/mL GM (%CVb)	AUC0‐t , b h·ng/mL GM (%CVb)	t1/2, h GM (%CVb)
25 mg QD
Day 1	7	9.00 (6.0–16.0)	132 (23.3)	89.3 (26.9)	1888 (24.6)	NA
Day 14	7	6.95 (0.0–9.0)	766 (16.0)	588 (22.5)	14 592 (12.6)	85.8 (25.9)
Accumulation ratio	7	ND	5.80 (21.5)	6.58 (31.4)	7.73 (25.7)	NA
50 mg QD
Day 1	7	8.00 (7.0–23.8)	232 (30.0)	189 (41.1)	3559 (26.3)	NA
Day 14	7	7.03 (3.0–9.0)	1149 (15.1)	990 (13.2)	24 888 (14.4)	72.2 (28.4)
Accumulation ratio	7	ND	4.95 (22.8)	5.25 (35.7)	6.99 (20.5)	NA
100 mg QD
Day 1	7	9.00 (6.0–23.8)	375 (50.2)	272 (75.3)	5894 (54.8)	NA
Day 14	6 d	7.49 (6.0–24.0)	2682 (21.0)	2415 (26.9)	55 316 (22.3)	NA
Day 18	6 d	9.01 (8.1–24.1)	2748 (21.1)	2458 (21.2)	58 418 (21.0)	78.5 (26.7)
Accumulation ratio c	6	ND	7.33 (43.0)	9.39 (72.5)	10.0 (48.5)	NA
500 mg QW
Day 1	7	9.88 (8.0–48.1)	1107 (12.8)	251 (36.3)	96 435 (14.1)	NA
Day 8	7	10.02 (5.1–16.0)	2318 (24.2)	498 (42.5)	191 605 (22.3)	NA
Day 15	7	10.00 (9.0–12.0)	1856 (24.1)	455 (46.5)	169 693 (31.9)	74.5 (19.1)
Accumulation ratio	7	ND	1.68 (26.9)	1.82 (33.8)	1.76 (27.0)	NA

Abbreviations: AUC_0‐t , area under the concentration‐time curve from time 0 to the end of the dosing interval; C _max, maximum concentration; C _trough, minimum observed concentration of the dosing interval; CVb, between‐participant coefficient of variation; GM, geometric mean; NA, not applicable; ND, not determined; QD, once daily; QW, once weekly; t _max, time to maximum concentration; t _1/2, half‐life.

^a C _trough measurement taken at the end of the dosing interval, 24 h for QD and 168 h for QW.

^b For QD doses t = 24 h, for QW t = 168 h.

^c Day 18/Day 1.

^d n = 7 for t _max.

3.3.3. Relative bioavailability and food effect

Variability in PK parameters was low for both PiB and tablet fed conditions. The geometric mean C _max was 390 and 525 ng/mL following a single oral dose of 100 mg in PiB and tablet formulations, respectively (Figure 4, Table S2), which corresponded to an approximately 37% increase in C _max when GSK'937 was provided as a tablet versus PiB (ratio of tablet fed to PiB [90% CI], 1.372 [1.19, 1.58]). Similarly, AUC_0‐24 and AUC_0‐∞ suggested the tablet formulation increased bioavailability by 38% (1.381 [1.22, 1.56]) and 35% (1.351 [1.22, 1.50]), respectively.

FIGURE 4.

Mean plasma concentrations by time (h) following single doses observed during the relative bioavailability and food effect parts in (A) linear and (B) semi‐logarithmic scales. Dashed line in (B) represents the lower limit of quantification (10 ng/mL). PiB, powder in bottle.

The effect of food on GSK'937 exposure was larger than the effect of formulation. Pharmacokinetic exposures differed by more than 2‐fold in the fed and fasted states, with bioavailability being greater when administration of the tablet formulation occurred immediately after a meal (Figure 4, Table S2). The ratio of C _max, AUC_0‐∞, and AUC_0‐24 geometric means was 2.31 (90% CI, 1.67–3.19), 2.10 (1.63–2.72), and 2.46 (1.82–3.32), respectively. Notably, variability in exposure was much higher between individuals when fasting: AUC_0‐∞ %CVb was 23.6 in the fed state and 64.0 in the fasted state, AUC_0‐24 %CVb was 33.9 and 91.0, and C _max %CVb was 31.8 and 93.5. Median t _max in the fasted state was earlier at 6.0 h than the fed formulations (tablet fed, 8.4 h; PiB fed, 9.0 h), though still within the range established in part 1 (5–24 h, Table 3). Considerable variability was also seen in t _max of fasted participants, however, with 1 participant requiring 47.7 h to reach C _max. There was no impact of food on t _1/2.

4. DISCUSSION

Antiretrovirals with novel mechanisms of action, improved tolerability, and alternative dosing strategies are needed to support people living with HIV; as such, maturation of the HIV‐1 virion presents an attractive target for therapeutic development. ¹¹ Here, we investigated the safety, tolerability, and PK parameters of the HIV‐1 MI GSK'937 in healthy individuals. Similar to the structurally related GSK'3640254, ²¹ GSK'937 was well‐tolerated and did not cause any significant changes to clinical laboratory analyses. GSK'937 was also slowly metabolized, with a t _1/2 of approximately 3 days, suggesting a possible QW dosing strategy, distinguishing GSK'937 from previous developmental MIs that have half‐lives supportive of daily dosing. ²⁰ , ²¹ , ²³ Higher plasma concentrations were achieved when GSK'937 was administered as a tablet compared to PiB and when delivered with a meal instead of after fasting. These results will guide dosing strategies in future clinical studies.

In the single and multiple ascending dose parts, most AEs were GI disorders (diarrhea, constipation, abdominal pain) and general disorders such as MDSD from ECG leads. Of the 65 total AEs, only 17 (26%) were attributable to GSK'937. These included GI disorders (14/17, 82%), headache (2/17, 12%), and dry skin (1/17, 6%). Gastrointestinal disorders were more common among GSK'937 recipients (13/64, 20%) than placebo recipients (3/33, 9%); however, sample sizes were too small to determine whether there was a significant relationship between GI disorders and GSK'937. The impact of GSK'937 on GI symptoms will be closely monitored in future clinical studies given other similar HIV‐1 MIs have demonstrated GI side effects. ²⁵ Nevertheless, no treatment‐related AEs were above grade 2 in severity and all AEs resolved a few minutes or days after their onset. Part 3 had 16 reported AEs with headache more frequent than GI effects, and none were study‐related. Across all parts of the study, there were no SAEs; deaths; clinically significant changes in hematology parameters, liver function parameters, renal function parameters, ECG parameters, or vital signs; or suicidality, suggesting no unexpected safety signals or major safety concerns.

A clear relationship between dose and systemic exposure was evident in parts 1 and 2. In the SAD (part 1), there were proportional increases in C _max and AUC_0‐t across all doses from 10 to 800 mg. This differed from GSK3640254 and GSK3532795, for which plateaus in absorption were observed at higher doses. ²¹ , ²³ Dose proportionality in GSK'937 exposure was further maintained after repeated daily administration, with accumulation of the drug following successive doses. Indeed, accumulation was greater in the 100‐mg QD treatment group (R_o C _max and R_oAUC_0‐trough, 7.33‐ and 10‐fold) than both the 25‐mg QD (5.8‐ and 7.73‐fold) and 50‐mg QD (4.95‐ and 6.99‐fold) groups. These high accumulation ratios may result from the generally long t _1/2 of GSK'937 (67–97 h).

Investigations of GSK'937 bioavailability determined the tablet to be superior to the PiB formulation under fed conditions. Median t _max for both the tablet and PiB formulations under fed conditions was 8–9 h, but approximately 37% more GSK'937 was measured after tablet administration. In contrast, treatment of individuals who were fasting with the tablet formulation led to an earlier median t _max of 6 h and a more than 2‐fold lower C _max than delivered by the tablet under fed conditions. Drug concentrations in the plasma were similar between fed and fasted individuals for the first 6 h but peaked at 6 h for fasting participants while continuing to increase in fed participants.

Although these data demonstrate the safety and PK of GSK'937 in the study participants, there were some limitations to the study. Sample sizes in the SAD and MAD portions of the study, for example, were appropriately small for a phase I study, but prohibited the testing of any statistical hypotheses. Additionally, almost all participants were men, leaving the possibility that the results might not be representative of women and therefore not generalizable to the total population.

In summary, GSK'937 administered to healthy adult participants after both single and multiple doses resulted in no unexpected safety signals or tolerability concerns. No SAEs or deaths were reported, and AEs were mainly of grade 1 severity and were fully resolved by the end of study. Data in this study describing dose proportionality, food effect, and long t_1/2 will be important in the design of future clinical studies. Overall, these data support the further clinical development of GSK'937 as a weekly administered oral treatment.

AUTHOR CONTRIBUTIONS

PDB, TJG, and MG conceived the study. PDB, YZ, LK, TJG, VB, BW, and MG designed the study. PDB, YZ, and BW acquired the data. PDB, YZ, TJG, and VB analyzed the data. PDB, YZ, LK, TJG, VB, and CF interpreted the data. PDB, YZ, LK, VB, and CF drafted the manuscript. PDB, YZ, LK, TJG, VB, CF, and BW critically revised the manuscript for important intellectual content. PDB, YZ, LK, TJG, VB, CF, BW, and MG provided final approval of the manuscript for publication.

DISCLOSURES

All authors are employees of ViiV Healthcare or GSK and may hold stock in GSK.

ETHICS STATEMENT

All protocols and consent forms were reviewed and approved by WCG Institutional Review Board (Santee, CA, USA), in accordance with the International Council for Harmonization and applicable country‐specific requirements. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. All participants provided written informed consent before the performance of any study‐specific procedures.

Supporting information

Table S1.

Table S2.

Benn PD, Zhang Y, Kahl L, et al. A phase I, first‐in‐human study investigating the safety, tolerability, and pharmacokinetics of the maturation inhibitor GSK3739937 . Pharmacol Res Perspect. 2023;11:e01093. doi: 10.1002/prp2.1093

DATA AVAILABILITY STATEMENT

Anonymized individual participant data and study documents will be readied for later availability at www.clinicalstudydatarequest.com.