Therapeutic HIV Vaccines and Broadly Neutralizing Antibodies

Therapeutic Vaccines and bNAbs Volume 27 Issue 4 December 2019/January 2020

Perspective Therapeutic HIV Vaccines and Broadly Neutralizing Antibodies

Therapeutic vaccines and broadly neutralizing antibodies (bNAbs) represent the MVA-B vaccine showed no substan- potential approaches to antiretroviral-free treatment of HIV. Although thera- tial effect on viral load rebound after peutic vaccines have been able to produce transient reductions in viral load ATI or on the viral reservoir with or with- during analytic treatment interruptions (ATIs), thus far none has been able out use of a latency reversal agent.3 to induce long-term remission. Pairing with latency reversal agents and Recent studies suggest that eliciting a immune modulators may improve vaccine efficacy. The bNAbs are investigated broad immune response may be asso- as a promising approach to achieving durable virologic control in the absence ciated with greater impact on viral re- of antiretroviral therapy. Combinations of antibodies are necessary for in- bound following ATI. For example, a trial creasing overall breadth and potency of coverage and preventing emergence examining a DC-HIV vaccine (dendritic of resistance. The next generation of antibodies includes engineered bispe- cells loaded with heat-inactivated autol- cific and trispecific antibodies that target 2 or 3 independent viral sites. This ogous HIV) showed that the vaccine in- article is based on a presentation by Magdalena E. Sobieszczyk, MD, MPH, duced broad immune responses and a at the International Antiviral Society-USA (IAS-USA) continuing education pro- substantial reduction in viral load during gram held in New York in March 2019. ATI. However that the effect was tran- sient4; in a separate study using a den- Keywords: HIV, therapeutic vaccines, broadly neutralizing antibodies, viral re- dritic cell platform, broader immune rebound, antibody targets, bNAbs, bispecific, trispecific. sponses correlated with better plasma viral load after ATI.7 And 2 trials investi- Therapeutic Vaccines and to augment host immune response to gating the ALVAC-HIV vaccine and Lipo- Antibodies kill infected CD4+ T cells and neutral- 6T showed that vaccine-induced CD4+ ize circulating virus in the absence and CD8+ T cell responses were associ- The rationale for investigating antiretro- of ART. ated with virologic control and delayed viral-free approaches to HIV treatment time to resumption of ART following is severalfold: Therapeutic HIV Vaccines ATI, compared with placebo.5,6 • It is impossible to eradicate HIV At a minimum, the goals of a 7 from latent viral reservoirs with 10 ● Vaccine Group therapeutic vaccine would be to antiretroviral therapy (ART) alone 106 simplify ART regimens and al- ●● 5 ● • It is important to have treatment 10 ● low for periodic analytic treat- ● ● ● ●● ● options including agents with 4 ● ● ● ● ● ment interruption (ATI). Optimal 10 ● ● ● ● ● ● ● ●● ● ● potential for less frequent dosing ● ● ● ● objectives would include the 3 ● ● ● ● ● 10 ● ● ● ● ●

(copies/mL) ● ● ● ● ●● ●● ●● ● ● ●

• There are gaps in ART delivery Plasma Viremia ●● ● ● ● ability to eliminate the need for 2 ● ● ● ● ● ● ● ● 10 ● • ART is associated with long-term ART either by eradicating the ● ● 1 ●●●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● adverse effects virus or by inducing host im- 10 0 4 8 120 160 200 240 280 320 • Adherence and retention in care mune responses capable of con- Days Since Discontinuation of ART remain a challenge trolling virus replication. However, in the many pla- The rationale for therapeutic HIV 107 Placebo Group vaccines and therapeutic use of anti- cebo-controlled studies thus far that have included interruption 106 HIV broadly neutralizing antibodies ● ● 5 ● of ART to measure therapeutic 10 ● ● ● ● ● (bNAbs) includes evidence from in- ● ● ● ● ● ● 4 ● vaccine efficacy, no therapeutic ● ● ● ● ● ● 10 ●● ● ● ● dividuals whose immune system nat- ● ●●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● vaccines have been successful ● ●●● ● 3 ● ● ● urally controls HIV without ART (ie, 10 ● ● ● ● (copies/mL) ● ● ● ● ● ● ● ● ● in achieving durable suppres- Plasma Viremia ● 2 ● ● ● ● long-term nonprogressors, elite con- 1- 6 10 ● ● ● ● ● ● sion of HIV viremia. For ex- ● ● ● trollers) that effective host-mediated ●●●● ● ● ● ● ● ● ● ●● ●● ●●●●● ●●● ● ●● ●●●● ● ample, a recently reported study 101 anti-HIV immunity is possible. This 0 4 8 120 160 200 240 280 320 raises the issue of whether it is possible showed that a DNA/rVSV therapeutic vaccine was unsuccess- Days Since Discontinuation of ART ful in achieving sustained sup- Dr Sobieszczyk is the Chief, Division of Infec- tious Diseases at New York-Presbyterian/ pression of virus after ART in- Figure 1. Absence of effect of DNA/rVSV therapeu- Columbia, and Associate Professor of Medi- terruption in individuals who tic vaccination compared with placebo on control of cine at Columbia University Irving Medical initiated ART early in infection HIV rebound following interruption of antiretroviral Center in New York, New York. (Figure 1).2 Similarly, a trial of therapy (ART). Adapted from Sneller et al.2

97 IAS–USA Topics in Antiviral Medicine

Despite such disappointments, the conserved regions of the HIV V1/V2 Glycan field is looking now at combining thera- envelope. Passive transfer of PGT141-45, PGDM1400, CAP256-VRC26 peutic vaccines with other agents, such bNAbs is being investigated for N332 Glycan Supersite as toll-like receptor 7 (TLR7) agonists treatment, eradication/cure, pre- PGT121, 10-1074 and latency reversal agents. For exam- vention, and to guide prevenz ple, a provocative study in SIV-infected vaccine design. CD4 Binding Site rhesus monkeys showed that use of the Much work has focused on VRC01, 3BNC117, therapeutic Ad26/MVA vaccine alone isolating bNAbs and identifying VRC07-523 induced broad cellular immune re- what targets on the HIV enve- sponses, but resulted in no clinically lope they attach to in order to significant decrease in viral load set- block viral entry into CD4 cells. point after ATI.8 However, with the A number of targets on the virus addition of a TLR7 agonist to the vac- have been identified as bNAb Figure 2. Examples of broadly neutralizing anti- cine, there was a 1.7510-log copies/mL attachment points that are cap- bodies (bNAbs) in development and their viral tar- reduction in viral load, a 2.5-fold delay able of neutralizing a wide range get sites. Adapted from Stewart-Jones et al.31 in viral rebound, and 33% of animals of HIV isolates, including the maintained undetectable viral load CD4 binding site, membrane proxi- In another study, the anti-CD4 bind- after ATI. mal external region (MPER), the gp41 ing site antibody VRCO1 was used Thus, although there have been no and gp120 interface, the V3 loop, and in HIV-infected patients who were not randomized controlled trials of thera- the V1/V2 loop (Figure 2).9 The fig- screened for VRC01 sensitivity.11 Pa- peutic vaccination that have induced tients underwent ATI and 3 to 8 infu- any remission after ATI, it is now pre- Optimizing bNAbs sions. The median time to plasma viral sumed that vaccines are needed that include identification rebound was 4 to 5.6 weeks, depending induce broad host immune responses on the number of infusions only slightly to recognize diverse escape viral vari- or engineering of more delaying rebound compared with his- ants after viral rebound. Furthermore, broad and potent torical controls. Viral rebound was poly- therapeutic vaccines are being paired clonal, despite high antibody levels in with potent latency reversal agents anti-bodies, modifying most patients and attributed to preex- (eg, vorinostat) or immune modulators Fc portion, to using isting resistance. A recently reported (TLR7 agonist) with the aim of poten- combinations of study of this antibody in Thai adults tially inducing evidence of remission. who initiated ART during acute HIV antibodies, and infection did not significantly delay viral rebound after ART discontinuation.12 Broadly Neutralizing Antibodies developing single In general, it has been shown that The long-established function of pas- antibodies with the antibodies have on average been sively administered bNAbs has been multiple targets able to reduce HIV viral load by ap- to inhibit viral entry into host cells by proximately 1.5 log10 copies/mL after blocking key binding sites on the viral ure shows some of the bNAbs in a single infusion of one antibody in envelope (neutralizing activity). More clinical development that target these patients with viremia.9,10,13,14 How- recent studies of these antibodies de- sites. ever, with ATI, viral rebound ultimately monstrate their capacity to engage the Several seminal studies have shown occurs and, at least in the monother- host immune system through Fc ef- antiviral activity with bNAb candidates. apy setting, viruses resistant to the fector functions to mediate killing of For example, passive infusion of the bNAbs emerge after several weeks, infected cells via antibody-dependent anti-CD4 binding site 3BNC117 anti- strengthening rationale for combina- cell-mediated cytotoxicity (ADCC), en- body showed significant reduction of tion bNAbs. Findings reported at the hancing immune responses against HIV virema10 and, in another study, sup- 2019 Conference on Retroviruses and and potentially clear latently infected pressed viral rebound among 13 ART- Opportunistic Infections (CROI) with cells, thereby potentially presenting a treated participants with 3BNC117- use of the PGT121 antibody that targets strategy for targeting the reservoir. sensitive viruses.10 Patients underwent the V3 loop bear noting, however. In A minority of HIV-infected individ- ATI and received 2 or 4 infusions of the viremic patients not on ART, response uals (5%-10%) develop the ability to antibody. The infusions were associ- in terms of virologic control depended neutralize various heterologous viruses ated with a viral rebound delay of an on baseline viral load. Those with a from different subtypes within 2 to 3 average of 8.4 weeks in all participants high baseline viral load before infusion years after infection. Very broad and versus 2.6 weeks in matched historical of the bNAb had an approximately potent neutralizing antibodies have controls. The rebound virus was found 1.7-log10 copy/mL drop, with rebound been isolated from these individuals. to have resistance to 3BNC117 with low resistant virus in most responders. These antibodies bind to relatively diversity in most patients. Patients with low baseline viral load ex-

98 Therapeutic Vaccines and bNAbs Volume 27 Issue 4 December 2019/January 2020

100 10E8 ■ ■ N6 are ongoing both for prevention and tive viremia the combination of 3 or VRC07-523 ■ 17-19 4E10 ■ VRC01■ treatment. 4 bNAbs may be required to cover the 80 ■ 3BNC117 ■ PG9 ■ PGDM1400 Other modifications to the Fc region swarm of viruses present. This also 10-1074 ■ may increase the effector functions raises the issue of the need to screen 60 ■ 8ANC195 PGT121 ■ ■ PGT128 of the antibody, potentially increasing for bNAb sensitivity prior to therapy ■ 2F5 ■ PGT151 CAP-256 ■ 40 ADCC or phagocytosis and potentially to potentially reduce the number of ■ b12 targeting the HIV reservoir. A report bNAbs required for treatment and to 20 at the 2019 CROI described an engi- amplify efficacy. ■ 2G12 ■ 35022 neered variant of the antibody PGT121 Finally, the next generation of bNAbs Breadth (% Neutralizing Strain) Breadth (% Neutralizing 0 that enhanced killing of HIV-infected will likely include engineered bispe- 10 1 0.1 0.01 CD4+ T-cells by natural killer cells.20 cific and trispecific antibodies, single Potency (IC at µg/mL) 80 Combining antibodies may have molecules that target multiple differ- particular promise for improving ther- ent epitopes on the virus or a cellular Figure 3. Examples of potency and breadth apeutic potential. In studies to date, receptor like CD4. Such an approach of effect of broadly neutralizing antibod- rebound viruses after bNAb treatment could potentially lower the likelihood ies (bNAbs) in development. IC80 = 80% did not demonstrate increased resis- of escape mutations and provide en- inhibitory concentration. Adapted from Gama and Koup.16 tance to other antibodies that target hanced neutralization potency and different envelope epitopes. Combina- breadth in a single molecule. An ex- hibited somewhat better viral suppres- tions of 2 or more bNAbs have been ample of a bispecific antibody in de- sion, with 2 experiencing suppression shown to result in more robust and velopment is 10E8.4/iMab, shown in for more than 6 months after a single sustained antiviral effects by increasing Figure 4.23,24 One arm of the antibody infusion of the antibody.15 overall breadth and potency of cover- binds an epitope in the membrane Steps toward optimizing bNAbs in- age and preventing the emergence of proximal external region (MPER) of clude identification or engineering of resistance. gp41; the other arm binds the HIV CD4 more broad and potent antibodies, A recently reported study exam- receptor molecule on T-helper cells. modifying Fc portion to, for example, ined the effects of 3 infusions of the The antibody entered first in human extend the half-life of the antibodies, combination of 2 antibodies 3BNC117 clinical testing in spring of 2019.25 using combinations of antibodies, and and 10-1074 during ATI in patients Trispecific antibodies are engineered developing single antibodies with mul- who were virologically suppressed on to bind to 3 sites on the HIV envelope. tiple targets through the engineering ART for more than 24 months who An example of one that recently en- of bispecific and trispecific antibodies. had been prescreened for sensitivity tered clinical testing in individuals with Greater potency and breadth of bNAbs to the antibodies. The median dura- HIV infection combines specificities of are needed to increase the proportion of viral suppression was 21 weeks 3 antibodies binding to the CD4 bind- tion of a viral strain neutralized by the for antibody-sensitive virus, a longer ing site, V1/V2 loop, and gp41 MPER.26 antibody and to neutralize a greater duration of suppression than has been Animal data presented at the 2019 proportion of all circulating strains of seen with single antibodies.21 A note- CROI showed that the antibody exerted the various viral subtypes. Figure 3 worthy finding is that 2 patients main- indicates the potency and breadth of tained virologic suppression long after effect of bNAbs that have been studied antibody levels had waned. This obser- to date.16 vation may suggest that the effector Clinical studies are exploring bNAbs function of the antibodies is acting to with extended half-lives. It has been augment host immune response, sup- shown that introduction of an LS muta- porting the investigation of whether tion—a 2 amino acid motif in the Fc bNAbs can be a component of cure strat- region of the antibody—increases the egies. Additionally, this combination affinity of that antibody for the neo- was also evaluated in viremic, treat- natal Fc receptor. This allows for re- ment naive individuals, and resulted in Figure 4. Example of an engineered bispe- circulation or recycling of the anti- an average HIV viral load decrease of cific antibody--10E8.4/iMab represented on body in the serum and extends its 2.05 log10 copies/mL in individuals with the left. See text for details. Adapted from 23 plasma half-life. Such a modification virus sensitive to both antibodies.22 Huang et al. can result in the ability to dose the anti- In short, antibody combinations • One arm of IgG binds an epitope in body less frequently. It has also been are likely to be necessary to increase the membrane proximal external region shown to increase the levels of the anti- overall breadth and potency of effect (MPER) of gp41 body at the mucosal surface in animal and to prevent the emergence of resis- • The other arm binds either the HIV-1 studies, which could play a role in pre- tance. The number of bNAbs required CD4 receptor on T-helper cell vention of acquisition of HIV. Studies for effective treatment may differ based Phase 1, first in human, clinical study for of the LS mutation antibody versions on the indication; for example, in ac- treatment and prevention is starting 2019

99 IAS–USA Topics in Antiviral Medicine

potent Fc effector functions, promising 14. Caskey M, Schoofs T, Gruell H, et al. Anti- References body 10-1074 suppresses viremia in HIV- for mediating ADCC and phagocytosis, 1–infected individuals. Nat Med. 2017; and potent suppression of viral repli- 23(2):185-191. 1. Tung FY, Tung JK, Pallikkuth S, Pahwa S, 15. Stephenson KE, Julg B, Ansel J, et al. Ther- cation in viremic simian/HIV-infected Fischl MA. A therapeutic HIV-1 vaccine 27,28 apeutic activity of PGT121 monoclonal animals. enhances anti-HIV-1 immune responses antibody in HIV-infected adults [Abstract Because current bNAbs are engi- in patients under highly active anti- 145LB]. Top Antivir Med. 2019;27(1s):54s. retroviral therapy. Vaccine. 2016;34(19): neered as human proteins, antidrug 2225-2232. 16. Gama L, Koup RA. New-generation high- potency and designer antibodies: role antibody (ADA) responses can be in- 2. Sneller MC, Justement JS, Gittens KR, et in HIV-1 treatment. Annu Rev Med. 2018; frequently observed, potentially in- al. A randomized controlled safety/effi- 69:409-419. cacy trial of therapeutic vaccination in terfering with activity and resulting HIV-infected individuals who initiated an- 17. Clinical Trials. First-in-human study of 10- in emergence of resistance to this im- tiretroviral therapy early in infection. Sci 1074-LS alone and in combination with Transl Med. 2017;9(419) 3BNC117-LS. https://clinicaltrials.gov/ct2/ mune therapy. In clinical studies to show/NCT03554408. Accessed on Janu- 3. Mothe B, Climent N, Plana M, et al. Safety ary 22, 2020. date, ADA responses have not been de- and immunogenicity of a modified vac- tected.29,30 cinia Ankara-based HIV-1 vaccine (MVA- 18. Clinical Trials. VRC 605: safety and phar- B) in HIV-1–infected patients alone or in macokinetics of a human monoclonal an- Recent advances in bNAb devel- combination with a drug to reactivate tibody, VRC-HIVMAB075-00-AB (VRC07- opment are exciting and might offer latent HIV-1. J Antimicrob Chemother. 2015; 523LS), administered intravenously or 70(6):1833-1842. subcutaneously to healthy adults. https:// alternative approaches to HIV therapy clinicaltrials.gov/ct2/show/NCT03015181. 4. Garcia F, Climent N, Assoumou L, et al. Accessed on January 22, 2020. or possibly cure. The role of bNAbs A therapeutic dendritic cell-based vac- in this area will be established by fur- cine for HIV-1 infection. J Infect Dis. 2 011; 19. Clinical Trials. Evaluating the safety and 203(4):473-478. serum concentrations of a human monother preclinical and clinical studies. As clonal antibody, VRC-HIVMAB075-00-AB these products move forward in devel- 5. Levy Y, Gahery-Segard H, Durier C, et al. (VRC07-523LS), administered in multiple Immunological and virological efficacy of doses and routes to healthy, HIV-unin- opment, however, and before they are a therapeutic immunization combined fected adults. https://clinicaltrials.gov/ct2/ widely adopted for the treatment and with interleukin-2 in chronically HIV- show/NCT03387150. Accessed on Janu- 1 infected patients. AIDS. 2005;19(3): ary 22, 2020. possible eradication, it is important 279-286. to consider their cost, frequency and 20. Thomsen ND, Balakrishnan M, Pace CS, 6. Levy Y, Durier C, Lascaux AS, et al. Sus- et al. GS-9722: first-in-class effector-en- practicality of dosing intravenously or tained control of viremia following thera- hanced broadly neutralizing antibody peutic immunization in chronically HIV- subcutaneously, their acceptability to for HIV cure [CROI Abstract 356]. In 1–infected individuals. AIDS. 2006;20(3): special Issue: Abstracts From the 2019 diverse populations. 405-413. Conference on Retroviruses and Oppor- 7. Levy Y, Thiebaut R, Montes M, et al. tunistic Infections. Top Antivir Med. 2019; Dendritic cell-based therapeutic vac- 27(1s):129s. Summary cine elicits polyfunctional HIV-specific T-cell immunity associated with control 21. Mendoza P, Gruell H, Nogueira L, et al. of viral load. Eur J Immunol. 2014;44(9): Combination therapy with anti-HIV-1 The bNAbs are a potentially promis- 2802-2810. antibodies maintains viral suppression. Nature. 2018;561(7724):479-484. ing approach toward durable control 8. Borducchi EN, Cabral C, Stephenson KE, 22. of HIV rebound in the absence of ART. et al. Ad26/MVA therapeutic vaccination Bar-On Y, Gruell H, Schoofs T, et al. Safety with TLR7 stimulation in SIV-infected rhe- and antiviral activity of combination HIV- They have been shown to be generally sus monkeys. Nature. 2016;540(7632): 1 broadly neutralizing antibodies in vi- safe and well tolerated in clinical stud- 284-287. remic individuals. Nat Med. 2018;24(11): 1701-1707. ies to date, and are also being actively 9. Scheid JF, Horwitz JA, Bar-On Y, et al. HIV-1 antibody 3BNC117 suppresses vi- 23. Huang Y, Yu J, Lanzi A, et al. Engineered pursued for prevention as passive pre- ral rebound in humans during treatment bispecific antibodies with exquisite HIV- vention and as a platform for design interruption. Nature. 2016;535(7613): 1–neutralizing activity. Cell. 2016;165(7): 1621-1631. of vaccines. To date, no randomized 556-560. 10. Caskey M, Klein F, Lorenzi JC, et al. 24. Padte NN, Yu J, Huang Y, Ho DD. Engi- controlled trials of therapeutic vaccina- Viraemia suppressed in HIV-1–infected neering multi-specific antibodies against tion have shown the ability to induce humans by broadly neutralizing anti- HIV-1. Retrovirology. 2018;15(1):60. body 3BNC117. Nature. 2015;522(7557): 25. Clinical Trials. 10E8.4/iMab bispecific long-term remission after ATIs. Combi- 487-491. antibody in HIV-uninfected and HIV- nation strategies of bNAbs, therapeutic 11. Bar KJ, Sneller MC, Harrison LJ, et al. Ef- infected adults. https://clinicaltrials.gov/ vaccines, and immunomodulators (eg, fect of HIV antibody VRC01 on viral re- ct2/show/NCT03875209. Accessed on bound after treatment interruption. N January 22, 2020. TLR7 agonist) may be needed to dimin- Engl J Med. 2016;375(21):2037-2050. 26. Clinical Trials. Pharmacokinetics of ish the HIV reservoir. 12. Crowell TA, Colby DJ, Pinyakorn S, et al. SAR441236. https://clinicaltrials.gov/ct2/ Safety and efficacy of VRC01 broadly show/NCT03705169. Accessed on Janu- ary 22, 2020. Presented by Dr Sobieszczyk in March 2019. neutralising antibodies in adults with acutely treated HIV (RV397): a phase 27. Xu L, Pegu A, Rao E, et al. Trispecific First draft prepared from transcripts by Mat- 2, randomised, double-blind, placebo- broadly neutralizing HIV antibodies medi- thew Stenger. Reviewed and edited by Dr controlled trial. Lancet HIV. 2019;6(5): ate potent SHIV protection in macaques. Sobieszczyk in $$$October$$$ 2019. e297-e306. Science. 2017;358(6359):85-90. 13. Lynch RM, Boritz E, Coates EE, et al. 28. Pegu A, Xu L, Demouth M, et al. Potent Virologic effects of broadly neutralizing antiviral activity of trispecific broadly Financial affiliations in the past 12 months: antibody VRC01 administration during neutralizing HIV antibodies [CROI Ab- Dr Sobieszczyk has no relevant financial affil- chronic HIV-1 infection. Sci Transl Med. stract 28LB]. In Special Issue: Abstracts iations to disclose 2015;7(319):319ra206. From the 2019 Conference on Retrovi-

100 Therapeutic Vaccines and bNAbs Volume 27 Issue 4 December 2019/January 2020

ruses and Opportunistic Infections. Top 30. Cohen YZ, Butler AL, Millard K, et al. Safe- 2016;165(4):813-826. doi:10. 1016/j.cell. Antivir Med. 2019;27(1s):12s. ty, pharmacokinetics, and immunogenic- 2016.04.010 29. Mayer KH, Seaton KE, Huang Y, et al. Safe- ity of the combination of the broadly neu- ty, pharmacokinetics, and immunological tralizing anti-HIV-1 antibodies 3BNC117 Top Antivir Med. 2019;27(4):97-101. activities of multiple intravenous or sub- and 10-1074 in healthy adults: A random- cutaneous doses of an anti-HIV monoclo- ized, phase 1 study. PLoS One. 2019;14(8): ©2019, IAS–USA. All rights reserved nal antibody, VRC01, administered to HIV- e0219142. uninfected adults: Results of a phase 1 31. Stewart-Jones GB, Soto C. Lemmin T. et al. randomized trial. PLoS Med. 2017;14(11): Trimeric HIV-1-env structures define gly- e1002435. can shields from clades A.B. and G. Cell.

101