Bispecific 10E8.4/iMab broadly neutralizing antibody in people with or without HIV-1: a partially randomized phase 1 trial

Broadly neutralizing antibodies (bnAbs) are a promising tool for HIV prevention and treatment. Here we conducted a first-in-human, phase 1 trial of the bispecific 10E8.4/iMab antibody, which consists of a 10E8.4 arm binding the HIV-1 envelope glycoprotein membrane-proximal external region and an ibalizumab (iMab) arm binding the human CD4 molecule. 10E8.4/iMab was administered intravenously (IV) or subcutaneously (SC). Safety/tolerability within 2 weeks of 10E8.4/iMab administration (primary outcome) and the pharmacokinetics (PK), antiviral activity, induction of anti-10E8.4/iMab antibodies, longitudinal CD4 + and CD8 + T cell counts and long-term safety (secondary outcomes) were evaluated. 54 participants living with HIV (PLWH) or without HIV (PLWoH) received 10E8.4/iMab or placebo. In arm 1, PLWoH received 10E8.4/iMab 0.3 mg kg −1 IV, 1 mg kg −1 SC, or 1 mg kg −1 IV ( n = 3 each). In arm 2, PLWoH received 10E8.4/iMab 3 mg kg −1 IV, 10 mg kg −1 IV or 30 mg kg −1 IV ( n = 6 each). In arms 3/3a, PLWH received 10E8.4/iMab 10 mg kg −1 IV ( n = 3) or 30 mg kg −1 IV ( n = 6). In arm 4, PLWoH were randomized to receive 10E8.4/iMab or placebo 2.5 mg kg −1 SC or 10 mg kg −1 SC ( n = 9 each). Participants in arms 1–3 were not randomized. No treatment-related serious adverse events (AEs) or AEs ≥ grade 3 were reported. The most common solicited AEs were tenderness (10/54, 18.5%), fatigue (18/54, 33.3%) and headache (12/54, 22.2%). Related grade 2 local and systemic solicited AEs occurred in one and six participants, respectively. Three of nine PLWH developed a generalized rash 8–12 days after infusion that resolved within 9–16 days. The primary objective of the study to evaluate the safety/tolerability of 10E8.4/iMab was met. These data support further study of 10E8.4/iMab to expand HIV treatment and prevention options. ClinicalTrials.gov: NCT03875209 .

Despite significant advances in human immunodeficiency virus (HIV) treatment and prevention over the past four decades, approximately 40.8 million people are living with HIV, and there are approximately 1.3 million new infections per year 1 . Given recent disruptions to global public health infrastructure, these numbers may increase 2 . Long-acting formulations of highly effective antiretrovirals for HIV treatment and prevention have become available in some settings 3 , but barriers to access persist, and preferences for long-acting products differ according to the population surveyed 4 . To end the HIV epidemic, it will be essential to expand the portfolio of long-acting options that are safe, tolerable and scalable so that people who stand to benefit can choose the option that works best for them 5 .

Broadly neutralizing antibodies (bnAbs) are a promising tool to control HIV viremia, reduce the size of the viral reservoir 6 , and prevent HIV acquisition 7 , 8 . Realizing the full potential of bnAbs will likely require combination therapy (for example, using multiple bnAbs that bind different targets) or bi- or tri-specific bnAb therapy (single molecules that bind multiple targets) 9 , 10 , 11 .

Immunotherapy with bispecific bnAbs is increasingly studied for a range of viral infections, as well as for oncologic and other non-oncologic indications 12 , 13 , 14 , 15 , 16 . The bispecific bnAb 10E8.4/iMab, one of the most potent and broad HIV bnAbs developed to date, neutralizes nearly all circulating HIV-1 strains tested and showed efficacy in preventing and treating HIV-1 in a humanized mouse model 17 , 18 . 10E8.4/iMab consists of one arm of 10E8.4, a monoclonal antibody that binds the HIV-1 glycoprotein 41 membrane-proximal external region, which is a highly conserved site on the viral envelope important for viral fusion (Extended Data Fig. 1 ) 19 , and one arm of ibalizumab (iMab), a humanized monoclonal antibody that binds human CD4 and thereby blocks HIV entry 20 . The linkage of these arms leads to marked synergistic enhancement of neutralization breadth and potency 17 . 10E8.4/iMab was additionally engineered to have an enhanced in vivo half-life and reduced Fc-effector functions, such as antibody-dependent cell-mediated cytotoxicity and complement-dependent cytotoxicity, to decrease potential IgG-mediated toxicity; iMab itself was developed as an IgG4 subclass monoclonal antibody to minimize Fc-effector functions 21 , 22 , 23 , 24 .

We conducted a phase 1 first-in-human trial to evaluate the safety, tolerability, pharmacokinetics (PK) and antiviral activities of the bispecific 10E8.4/iMab antibody given intravenously in people living with HIV (PLWH) and people living without HIV (PLWoH) and given subcutaneously in PLWoH.

Sixty-three participants were screened to determine eligibility. Eligible participants were 18–60 years of age and in good health. Eligibility for specific study arms and groups varied by HIV status and HIV-1 RNA level. PLWH were eligible for arm 3 group H (10 mg kg −1 IV) participation if their CD4 + T cell count was >350 cells mm − 3 and HIV-1 RNA was between 1,000 and 100,000 copies ml −1 in the context of being treatment-naïve, off treatment for at least 4 weeks in consultation with their primary provider or clinically stable and without recent changes to their treatment regimen. PLWH were eligible for arm 3a group I (30 mg kg −1 IV) participation if their CD4 + T cell count was >350 cells mm − 3 and HIV-1 RNA was undetectable on triple-combination antiretroviral therapy.

Fifty-four participants enrolled and received 10E8.4/iMab or placebo between 21 March 2019 and 1 October 2021. Arm 1 group A (0.3 mg kg −1 IV, PLWoH; n = 3) participants enrolled first. Arm 1 group B (1 mg kg −1 SC, PLWoH; n = 3) enrolled next. Arm 1 group C (1 mg kg −1 IV, PLWoH; n = 3) enrolled next. Thereafter, participants in arm 4 group J (2.5 mg kg −1 SC or placebo, PLWoH; n = 9) and arm 2 group D (3 mg kg −1 IV, PLWoH; n = 6) enrolled contemporaneously; if a participant was eligible for both groups and a group assignment was open for both, the participant could choose whether to enter arm 4 group J (and thereafter undergo randomization to 10E8.4/iMab or placebo) or arm 2 group D. Next, participants in arm 4 group K (10 mg kg −1 SC or placebo, PLWoH; n = 9) and arm 2 group E (10 mg kg −1 IV, PLWoH; n = 6) enrolled contemporaneously; similarly, if a participant was eligible for both groups and a group assignment was open for both, the participant could choose whether to enter arm 4 group K (and thereafter undergo randomization to 10E8.4/iMab or placebo) or arm 2 group E. Next, participants in arm 2 group F (30 mg kg −1 IV, PLWoH; n = 6) and arm 3 group H (10 mg kg −1 IV, PLWH with active viremia; n = 3) enrolled contemporaneously, with group assignment determined by HIV status. Finally, participants in arm 3a group I (30 mg kg −1 IV, PLWH with viral suppression; n = 6) enrolled (Fig. 1 ).

a , Consort diagram. b , Sequence of arm and group enrollments. Arm 1 group A enrolled first, followed by arm 1 group B and then arm 1 group C. Thereafter, arm 4 group J and arm 2 group D enrolled contemporaneously; if a participant was eligible for both groups, and if a group assignment was open for both groups, then the participant could choose whether to enter arm 4 group J (and thereafter undergo randomization to 10E8.4/iMab or placebo) or arm 2 group D. Next, arm 4 group K and arm 2 group E enrolled contemporaneously, with the same group assignment process as arm 4 group J/arm 2 group D. Next, arm 2 group F and arm 3 group H enrolled contemporaneously, with group assignment determined by HIV status. Finally, arm 3a group I enrolled.

There were eight protocol amendments (described in Methods ). The study opened under Protocol Amendment 2. Fifty-three protocol deviations were reported under the following categories: protocol procedure/assessment (33/53, 62.3%), follow-up visit schedule (14/53, 26.4%), eligibility/enrollment (3/53, 5.7%), Standard Operating Procedures for Good Clinical Practice guidelines (2/53, 3.8%) and investigational product administration/dosing (1/53, 1.9%). Among the eligibility/enrollment deviations, two participants had at least one screening lab that was outside the allowable screening window at enrollment, and one participant had an exclusionary calcium level at enrollment. The investigational product administration/dosing deviation was related to priming of 10E8.4/iMab with incorrect IV line tubing; however, the deviation was noted before the infusion began, and new product was obtained and administered to the participant via the correct tubing.

Forty-nine participants completed the study; five participants left early due to moving away ( n = 2) or loss to follow-up ( n = 3). One follow-up visit was missed, and two follow-up visits were performed out of window due to COVID-19-like symptoms or COVID-19 diagnosis.

Median age was 27.5 years, and 26 (48.1%) were assigned female at birth. Sex was not evenly distributed between groups; arms 1, 2 and 4, which enrolled PLWoH, included 50–67% female participants, whereas arms 3 and 3a, which enrolled PLWH, included no female participants. Half of participants identified as White (27/54, 50.0%), followed by Black or African American (9/54, 16.7%) and other/unknown/refused to specify (9/54, 16.7%), Asian (8/54, 14.8%) and multiracial (1/54, 1.8%). Twenty-one participants (38.9%) reported Latino or Hispanic ethnicity (Table 1 ). Among PLWH, median years from diagnosis was 2.9 (range 0.1–29.0), and nearly all antiretroviral treatment (ART) regimens included an integrase strand transfer inhibitor, except for one regimen that combined two nucleoside reverse transcriptase inhibitors with one non-nucleoside reverse transcriptase inhibitor. At enrollment, group H (viremic) participants agreed (in consultation with their primary provider) not to begin or change ART for 6 weeks after 10E8.4/iMab infusion, despite a clear explanation of HIV treatment guidelines. Study investigators discussed each PLWH with the participant’s HIV primary provider, with the understanding that ART could be started, restarted, or switched at any time if clinically indicated. All group H participants were on effective ART with HIV VL −1 at end of study.

The primary outcome was the rate of signs, symptoms and laboratory abnormalities, in addition to local and systemic solicited adverse events (AEs), within 2 weeks of 10E8.4/iMab administration in all study arms/groups. Secondary outcomes were the PK profile of 10E8.4/iMab (elimination half-life (t 1/2 ), clearance (CL/F), volume of distribution (Vz/F), area under the concentration-time curve (AUC) and decay curve in all study arms/groups); the decline in plasma HIV-1 RNA level by standard clinical assay after 10E8.4/iMab infusion in PLWH with viremia; the frequency and levels of induced anti-10E8.4/iMab antibodies in all study groups; the rate of signs, symptoms and laboratory abnormalities that occurred during study follow-up after 10E8.4/iMab infusion/injection in all study groups; and the absolute and relative CD4 + and CD8 + T cell counts after 10E8.4/iMab infusion. We reported the noncompartmental PK analysis by group for drug exposure (AUC), but because of the non-linear PK for 10E8.4/iMab, the remaining PK outcomes were reported using population modeling across groups by cohort, separated by PLWH versus PLWoH. All outcomes are presented here.

Among PLWoH, grade 1 or higher AEs (solicited or unsolicited signs, symptoms and laboratory abnormalities, related or unrelated) occurred within 2 weeks of 10E8.4/iMab administration at the following frequencies: 66.7% (2/3) among group A (0.3 mg kg −1 IV), 66.7% (2/3) among group B (1 mg kg −1 SC), 100% (3/3) among group C (1 mg kg −1 IV), 66.7% (4/6) among group D (3 mg kg −1 IV), 83.3% (5/6) among group E (10 mg kg −1 IV), 100% (6/6) among group F (30 mg kg −1 IV), 66.7% (4/6) among group J (2.5 mg kg −1 SC) and 100% (6/6) among group K (10 mg kg −1 SC). Among placebo recipients, 83.3% (5/6) reported a grade 1 or higher AE. Among PLWH, 66.7% (2/3) of participants in group H (10 mg kg −1 IV) and 83.3% (5/6) of participants in group I (30 mg kg −1 IV) reported a grade 1 or higher AE (Table 2 ).

Among PLWoH, grade 1 or higher related solicited AEs (local or systemic reactogenicity) occurred within 2 weeks of 10E8.4/iMab administration at the following frequencies: 33.3% (1/3) among group A, 66.7% (2/3) among group B, 33.3% (1/3) among group C, 50.0% (3/6) among group D, 66.7% (4/6) among group E, 66.7% (4/6) among group F, 33.3% (2/6) among group J and 100.0% (6/6) among group K. Among placebo recipients, 50% (3/6) reported a grade 1 or higher related solicited AE. Among PLWH, 66.7% (2/3) of participants in group H and 83.3% (5/6) of participants in group I reported a grade 1 or higher related solicited AE (Extended Data Table 1 ).

The most common local solicited AE (related or unrelated) was tenderness, which was reported by 18.5% (10/54) of study participants. The most common systemic solicited AEs (related or unrelated) were fatigue (33.3%, 18/54) and headache (22.2%, 12/54). All related solicited AEs were classified as grade 1 (mild) or grade 2 (moderate). Only one participant (group K, 10 mg kg −1 SC, randomized to 10E8.4/iMab) reported related grade 2 local symptoms (swelling, warmth, itching and erythema). Six participants reported related grade 2 systemic symptoms; of these, five were PLWH, including three PLWH who developed a generalized and occasionally pruritic grade 2 rash between days 8 and 12 following infusion. This rash developed in two of three individuals in group H and one of six participants in group I. One group H participant developed grade 2 headache, fever, fatigue and pruritus along with the rash. All rashes resolved with conservative medical management (range 9–16 days until resolution) (Extended Data Table 1 ). Additionally, three unrelated grade 3 (severe) solicited AEs (headache, myalgia and fatigue) occurred in a placebo recipient with influenza.

There were no acute infusion reactions following 10E8.4/iMab administration. No deaths, serious AEs, pregnancies or dose-limiting AEs were reported within 2 weeks of enrollment.

Among PLWoH, grade 1 or higher AEs (solicited or unsolicited signs, symptoms and laboratory abnormalities, related or unrelated) occurred during study follow-up (after 10E8.4/iMab administration) at the following frequencies: 66.7% (2/3) among group A, 100% (3/3) among group B, 100% (3/3) among group C, 83.3% (5/6) among group D, 100% (6/6) among group E, 100% (6/6) among group F, 66.7% (4/6) among group J and 100% (6/6) among group K. Among placebo recipients, 83.3% (5/6) reported a grade 1 or higher AE. Among PLWH, 100% (3/3) of participants in group H (10 mg kg −1 IV) and 100% (6/6) of participants in group I (30 mg kg −1 IV) reported a grade 1 or higher AE (Extended Data Table 2 ).

In total, 90 unsolicited AEs were reported by 37/54 (68.5%) participants over the study. No deaths, serious AEs, pregnancies or dose-limiting AEs were reported. One grade 4 (potentially life threatening) hypoglycemia AE was reported in a group C participant with a glucose level of 37 mg dl −1 on day 0; the blood sample was drawn before 10E8.4/iMab administration and resulted after the visit. The participant had not eaten before the visit but ate during the visit and was asymptomatic; their follow-up glucose levels during the study were normal.