Impact of Anti-CD4 Autoantibodies on Immune Reconstitution in People With Advanced HIV

Abstract

Background: Despite suppressive antiretroviral therapy (ART), 15%-30% of people with human immunodeficiency virus (HIV) experience a limited recovery of CD4 T cells. Although autoantibodies against the CD4 receptor have previously been identified in people with HIV (PWH), little is known about their longitudinal impact on CD4 T-cell reconstitution.

Methods: Anti-CD4 autoantibodies were evaluated by the fluid-phase luciferase immunoprecipitation systems immunoassay in ART-naive people with advanced HIV (CD4 count ≤100 cells/µL), PWH with CD4 count >200 cells/µL, long-term nonprogressors, people with idiopathic CD4 lymphopenia, people with autoimmune lymphoproliferative syndrome, and healthy volunteers without HIV. In the participants with advanced HIV, we assessed the association of anti-CD4 autoantibodies at ART initiation with CD4 recovery over a median follow-up of 192 weeks.

Results: Anti-CD4 autoantibodies were identified in 29% (61/210) of ART-naive participants with advanced HIV but were absent in people without HIV. Female PWH showed a 4-fold higher prevalence (P < .001) of anti-CD4 autoantibodies compared to males. After ART initiation, people with advanced HIV with anti-CD4 autoantibodies exhibited an overall slower rate of CD4 reconstitution (5.8 vs 6.6 cells/µL/month, P = .007) and lower week 192 CD4 count (268 vs 355 cells/µL, P = .037). Incidental, clinically indicated immunosuppressive therapy in these participants was associated with an improved rate of CD4 reconstitution (P = .0019) and higher week 192 CD4 count (551 vs 268 cells/µL, P = .019).

Conclusions: People with advanced HIV harboring anti-CD4 autoantibodies at ART initiation demonstrated a slower rate and extent of CD4 reconstitution after 4 years. Incidental immunosuppressive therapy was associated with increased CD4 counts in these participants.

Keywords: CD4 T cells; HIV; anti-CD4 autoantibodies; immune reconstitution; immunological nonresponse.

Graphical Abstract

This graphical abstract is also available at Tidbit: https://tidbitapp.io/institutional-portal/clinical-infectious-diseases/tidbits/impact-of-anti-cd4-autoantibodies-on-immune-reconstitution-in-people-with-advanced-hiv-4a548e51-4326-4edd-a70b-73333668e1cb/update

Figure 1.

Prevalence of anti-CD4 autoantibodies. Anti-CD4 autoantibodies were detected in 29% of people with advanced human immunodeficiency virus (HIV) (CD4 ≤100 cells/µL) beginning antiretroviral therapy (ART), 26% of people with HIV not on ART with CD4 >200 cells/µL (“PWH CD4 >200”), and 9% of long-term nonprogressors (LTNP, median CD4 count of 746 cells/µL and median HIV-1 plasma RNA <50 copies/mL). No evidence of anti-CD4 autoantibodies was found in patients with autoimmune lymphoproliferative syndrome (ALPS), patients with idiopathic CD4 lymphopenia (ICL), or healthy controls without HIV (HC). Geometric mean levels with geometric standard deviation are shown (relative light units [RLU]). The dotted line represents the cutoff values for determining seropositivity. *P < .05, **P < .01, ***P < .001, and ****P < .0001 when comparing autoantibody seropositivity rates between groups using Fisher exact tests.

Figure 2.

Mapping of autoantibody binding to different regions of the CD4 receptor. Relative autoantibody binding to different regions of the CD4 receptor determined by serological testing are summarized (−, negative, +; weak and +++; strong). The extracellular region of the CD4 receptor consisting of the D1–D4 protein domains showed strong immunoreactivity (A). A deletion mutant of CD4 containing only the D1 domain (B) was seronegative for CD4 autoantibody binding. Antibody binding was strongly retained with deletion mutants of CD4 maintaining the integrity of the D3–D4 domains (C), weakly with the D3 domain (D), and strongly with the D4 domain (E).

Figure 3.

Consolidated Standards of Reporting Trials (CONSORT) flow diagram of patients based on presence of anti-CD4 autoantibodies and based on treatment with prolonged immunosuppressive therapy. Prolonged immunosuppression was defined as cytotoxic chemotherapy, biologic immunotherapy, and/or receiving ≥20 mg/day prednisone (or equivalent) for ≥1 month (not including courses completed prior to starting antiretroviral therapy). Treatment failure was defined as plasma human immunodeficiency virus type 1 RNA >1000 copies/mL after initial suppression, ≥2 consecutive levels of >200 copies/mL, or lack of suppression by week 24. Abbreviations: ART, antiretroviral therapy; HIV, human immunodeficiency virus; LTFU, lost to follow-up (including voluntary withdrawal and death) prior to week 96.

Figure 4.

CD4 count at study timepoints (A) and time to immunologic response (B) by anti-CD4 autoantibody status. A, CD4 count was significantly lower at weeks 176–192 in those who were positive for anti-CD4 autoantibodies. B, Immunologic response was defined as CD4 count ≥350 cells/µL. Participants positive for anti-CD4 autoantibodies had decreased hazards of immunologic response, but this relationship was not statistically significant. *P < .05, **P < .01. Abbreviations: AAb, autoantibody; Ab, antibody; ART, antiretroviral therapy.

Figure 5.

CD4 count at study timepoints (A) and time to immunologic response (B) in patients with anti-CD4 autoantibodies by immunosuppressive therapy receipt. A, CD4 counts were significantly higher in participants who had received prolonged immunosuppressive therapy at weeks 128–192. B, Immunologic response was defined as CD4 count ≥350 cells/µL. Participants with anti-CD4 autoantibodies who received prolonged immunosuppression had an increased hazard of immunologic response, which was significant by log-rank but not in the multivariable Cox proportional hazards model. *P < .05, **P < .01, ***P < .001. Abbreviation: ART, antiretroviral therapy.