In-depth virological and immunological characterization of HIV-1 cure after CCR5Δ32/Δ32 allogeneic hematopoietic stem cell transplantation

Abstract

Despite scientific evidence originating from two patients published to date that CCR5Δ32/Δ32 hematopoietic stem cell transplantation (HSCT) can cure human immunodeficiency virus type 1 (HIV-1), the knowledge of immunological and virological correlates of cure is limited. Here we characterize a case of long-term HIV-1 remission of a 53-year-old male who was carefully monitored for more than 9 years after allogeneic CCR5Δ32/Δ32 HSCT performed for acute myeloid leukemia. Despite sporadic traces of HIV-1 DNA detected by droplet digital PCR and in situ hybridization assays in peripheral T cell subsets and tissue-derived samples, repeated ex vivo quantitative and in vivo outgrowth assays in humanized mice did not reveal replication-competent virus. Low levels of immune activation and waning HIV-1-specific humoral and cellular immune responses indicated a lack of ongoing antigen production. Four years after analytical treatment interruption, the absence of a viral rebound and the lack of immunological correlates of HIV-1 antigen persistence are strong evidence for HIV-1 cure after CCR5Δ32/Δ32 HSCT.

Main

Human immunodeficiency virus type 1 (HIV-1) persists in the body during antiretroviral therapy (ART) in latently infected CD4⁺ T cells, but allogeneic hematopoietic stem cell transplantation (HSCT) has been shown to substantially reduce the viral reservoir^1,2. However, some of the reservoir-harboring immune cells are extremely long-lived³, partially resistant to chemotherapy regimens used during HSCT procedures and can cause viral rebound on analytical treatment interruption (ATI)^4,5. Notably, both cases of successful HIV-1 cure published so far—the ‘London patient’ (IciStem no. 36) and the ‘Berlin patient’—received a CCR5Δ32/Δ32 allograft^6,7 conferring extended resistance to HIV-1 due to the absence of surface expression of the CCR5 coreceptor.

In this study, we provide a detailed longitudinal virological and in-depth immunological analysis of the peripheral blood and tissue compartments of a 53-year-old male patient (IciStem no. 19)⁸, alive and in good health 117 months after CCR5Δ32/Δ32 allogeneic HSCT and 48 months after ATI. The patient was diagnosed to be HIV-1 clade B positive in January 2008 and presented with a CD4⁺ T cell count of 964 cells per μl and an HIV-1 plasma viral load of 12,850 copies per ml (Centers for Disease Control and Prevention A1, which was no indication for initiation of ART according to the national guidelines at that time). In October 2010, an ART regimen with tenofovir disoproxil fumarate (TDF), emtricitabine (FTC) and darunavir and ritonavir (DRV/r) was initiated (503 CD4⁺ T cells per μl and 35,303 HIV-1 copies per ml), resulting in a continuously suppressed plasma viral load (Fig. 1a). In January 2011, the patient was diagnosed with acute myeloid leukemia (AML) M2 according to the French–American–British classification, which carried an inversion of chromosome 16 at p13q22, resulting in the CBFB–MYH11 fusion protein. The patient achieved hematological complete remission after chemotherapy, which included idarubicin, cytarabine, etoposide induction therapy and three high-dose cytarabine (HiDAC) consolidation cycles. To avoid drug–drug interactions, DRV/r was switched to raltegravir in March 2011.

In September 2012, the patient experienced an AML relapse but achieved a second complete remission after treatment with the A-HAM chemotherapy regimen (retinoic acid, HiDAC, mitoxantrone) and a second cycle of HiDAC. A systematic search identified a 10/10 HLA-matched (no mismatches in HLA-A, HLA-B, HLA-C, HLA-DR and HLA-DQ loci) unrelated female stem cell donor with a homozygous CCR5Δ32 mutation (Extended Data Table 1). After reduced-intensity conditioning with fludarabine, treosulfan and anti-thymocyte globulin, 8.74 × 10⁶ unmodified CD34⁺ peripheral blood stem cells per kg of body weight were transplanted in February 2013. Immunosuppressive therapy consisted of cyclosporine and mycophenolate mofetil and was later changed to tacrolimus monotherapy. In June 2013, the patient experienced a second AML relapse. He achieved molecular oncological remission a third time after eight cycles with 5-azacytidine and four donor lymphocyte infusions (DLIs; 1 × 10⁶, 10 × 10⁶ and twice 50 × 10⁶ donor T cells per kg of body weight). Thirty-four days after HSCT, full donor chimerism was established and retained except for a short period during the second relapse at months 3 and 4 after HSCT (Extended Data Fig. 1a). In 2014, elevated liver enzymes caused discussion about possible hepatic graft versus host disease (GvHD), but a liver biopsy in May 2015 was interpreted as drug-induced liver injury. In July 2014, the patient also experienced reactivation of cytomegalovirus (CMV) (duodenal ulcer) and herpes simplex virus 2 (genital ulcers and cerebral vasculitis) and human herpesvirus 8 and Epstein–Barr virus (viremia) but recovered after specific antiviral treatment of the CMV and herpes simplex virus 2 infections. After DLI, mild chronic GvHD of the eyes with bilateral keratoconjunctivitis sicca developed that persists until today. ART was continued throughout and proviral HIV-1 DNA and HIV-1 RNA remained undetectable despite intensified testing in clinical routine assays (Fig. 1a). However, multiple assessments of the HIV-1 viral reservoir in the peripheral blood and lymphoid and gut tissue before and after ATI revealed sporadic HIV-1 DNA traces at several time points, with a higher frequency compared to HIV-1-negative donors and no-template controls (Extended Data Table 2). Although rare, residual HIV-1 DNA and HIV-1 RNA were also detected by in situ hybridization (DNAscope and RNAscope assays) from histological sections of inguinal lymph node tissue from month 51 and some gut biopsies from month 77 (Fig. 1b); the number of HIV-1 RNA⁺ cells (2.61 ± 0.13 HIV-1 RNA⁺ cells per 10⁵ cells) and HIV-1 DNA⁺ cells (5.08 ± 1.74 HIV-1 DNA⁺ cells per 10⁵ cells) were only modestly above the limit of detection established for the assay. Importantly, neither HIV-1 p24, HIV-1 RNA or HIV-1 DNA was detectable in peripheral blood mononuclear cells (PBMCs) by repeated cell culture-based quantitative viral outgrowth assay nor by intact proviral DNA assay (Extended Data Table 2). Negative in vivo outgrowth assays using two different humanized mouse models confirmed the absence of replication-competent virus in the tested samples (Extended Data Fig. 2). Despite repeated in-depth reservoir assessments to determine whether or not the virus persisted and thus ensure the safest possible approach for ATI, the presence of residual replication-competent virus could not be completely ruled out^7,9,10 and ATI ultimately remained the only way to prove HIV-1 cure^8,9.

ART was discontinued 69 months after HSCT in November 2018, after careful consideration, and no antiretroviral agents were detected in four plasma samples collected after ATI (Fig. 1a). On cessation of ART, the patient remained without any clinical or laboratory signs of an acute retroviral syndrome. No rebound of plasma HIV-1 RNA occurred during ATI after 48 months in the absence of ART (Fig. 1a).

Extended immunological profiling before and after ATI demonstrated stable CD4⁺ T cell counts, absence of CCR5 expression (Extended Data Fig. 1c) and an immune status comparable to previous reports of people living with HIV (PLWH) after HSCT (with reduced naive T cell frequencies, elevated terminally differentiated effector memory T (T_EMRA) cell frequencies and elevated frequencies of CD56⁻ natural killer (NK) cells)⁸ (Extended Data Fig. 3). The activation levels of the patient’s peripheral blood NK cells and cytotoxic CD8⁺ T cells after ATI were within the range observed in HIV-1-negative controls (Fig. 2a,b; see also previously published data in ref. ⁸). Moreover, the immune cell composition and, in particular, the CD4⁺ T cell density was normal in the investigated lymphoid tissue at the time of sampling (month 51 after HSCT) and in gut tissue samples obtained after ATI (month 77 after HSCT; Extended Data Fig. 4a). Additionally, there was no evidence of elevated inflammation in the lymphoid and gut tissue (measured by MX1 expression) or gut barrier damage (measured by MPO expression) by immunohistochemical staining (Extended Data Fig. 4b,c).

In month 39 after HSCT, HIV-1-specific CD8⁺ T cells were weakly detected on stimulation with overlapping peptide pools spanning HIV-1 Gag, Pol and Nef by intracellular cytokine staining (Fig. 2c; see also previously published data in ref. ⁸). However, the frequencies of HIV-1-specific T cells were substantially lower than previously observed for other PLWH^8,10, further declined below threshold levels while still on ART and did not increase after ATI (Fig. 2c). Likewise, no significant T cell responses toward a set of 37 HIV-1 peptides known to be restricted by the patient’s major histocompatibility complex (MHC) class I molecules were detected before or after ATI as measured ex vivo by interferon-γ (IFNγ) enzyme-linked immunosorbent spot (ELISpot) (Supplementary Table 1) or MHC class I tetramer enrichment of an HLA-A*02-restricted reverse transcriptase epitope (RT-YV9) (Extended Data Fig. 5a). However, stimulation of PBMCs with the HLA-A*02-restricted RT-YV9 peptide and the gag-p6 peptide Gag07-121 elicited pronounced T cell responses in vitro after discontinuation of tacrolimus in month 56 that eventually decreased and were not detected in the most recent samples (Extended Data Fig. 5b). These short-term in vitro-expanded HIV-1-specific T cells were donor-derived because they carried the CCR5Δ32/Δ32 deletion (Extended Data Fig. 1b) suggesting antigen presentation to donor T cells in the peri-transplant period, despite continued ART⁸. In contrast, CMV-specific CD8⁺ T cells were strongly and consistently detected (Extended Data Fig. 5c), indicating that the waning of HIV-1-specific CD8⁺ T cells was probably due to the absence of stimulation by HIV antigens.

Immunoblot analyses of the HIV-1-specific antibody response paralleled the progressive loss of cellular HIV-1 reactivity: anti-gp120 and anti-gp160 antibodies showed the longest persistence (Fig. 2d), as reported previously^2,7. At month 39 after HSCT, the peripheral blood HIV-1-specific antibody levels were below the cutoff for PLWH and comparable to HIV-1-negative individuals (Extended Data Fig. 5d). Because the maintenance of virus-specific responses is dependent on antigen exposure in chronic infection¹¹, the extremely weak and waning HIV-1-specific T cell response and the decreasing levels of specific antibodies suggest that the HIV-1 reservoir capable of antigen production has been extremely depleted by the HSCT procedure and/or the graft versus HIV effect^1,2.

Despite traces of HIV-1 DNA, there was no evidence of viral rebound or immunological correlates of antigen exposure. It is unclear whether the residual HIV-1 DNA signals stem exclusively from defective viral fragments or from an infinitesimally small pool of intact proviruses because proof for the absence of residual replication-competent virus was limited by the number of cells obtained and restricted accessibility of some anatomical compartments known to harbor the HIV-1 reservoir^9,10. Therefore, it might be decisive that the patient was transplanted with a CCR5Δ32/Δ32 graft and that only an extremely low proportion (0.14%) of the proviral sequences retrieved before HSCT was characterized as possibly CXCR4-tropic, while potentially intact proviruses of the predominant R5-tropic population would not be able to propagate in this protective setting because of the absence of CCR5⁺ target cells⁴ (Extended Data Table 3).

The detailed observational characterization of single cases of HIV-1 cure after HSCT gives important insights but is anecdotal by nature and lacks the power of controlled prospective studies. Therefore, certain aspects of the cases published to date are of uncertain relevance for HIV-1 cure^6,7, for example, underlying hematological malignancy, conditioning regimens or degree of GvHD as well as donor–recipient sex mismatch or the timing of ATI. However, two similarities are crucial for the outcome: all three patients achieved long-term remission after CCR5Δ32/Δ32 HSCT and harbored predominant R5-tropic virus strains. While CCR5Δ32/Δ32 HSCT cannot prevent the rebound of X4-tropic viruses⁵, this modification of the host’s immune system is a key component to prevent reservoir reseeding in PLWH with R5-tropic viruses requiring HSCT. Reservoir reduction during conditioning chemotherapy and immune-mediated HIV-1 clearance by donor cells (the ‘graft versus HIV effect’), and unspecific decay of latently infected cells driven by alloreactivity (GvHD; potentially further enhanced by DLI), might be additional, potential components that together led to the cure of HIV-1. However, it remains elusive to what extent reservoir depletion is necessary and contributes to HIV-1 cure in this and other cases^2,6,7,8. One limitation of our study was the scarcity of samples available for detailed immunological and virological analysis directly before and after the HSCT.

Although HSCT using donors with a CCR5Δ32/Δ32 mutation is neither a low-risk nor an easily scalable procedure, its relevance to cure strategies is highlighted by recent reports of successful long-term HIV-1 remission after CCR5Δ32/Δ32-HSCT^7,12,13. Expansion of this approach to introduce the CCR5Δ32 mutation into wild-type stem cell grafts using gene therapy in combination with new reservoir reduction strategies may hold the promise of an HIV-1 cure outside of life-threatening hematological malignancies. This third case of HIV-1 cure after allogeneic CCR5Δ32/Δ32 HSCT provides detailed information on the virological and immunological signature before and after ATI and generates valuable insights that will hopefully guide future cure strategies.

Methods

Ethics

The described individual (male, 53 years old as of 2022) was enrolled as patient no. 19 in the IciStem program at University Hospital Düsseldorf. ATI and examinations of the viral reservoir were performed after consultation of the ethics board of the Medical Faculty of the Heinrich Heine University Düsseldorf (official statement from 29 July 2016). Written informed consent was obtained from the patient for ATI and additionally performed immunological and virological studies (ethics board of the Medical Faculty of the Heinrich Heine University Düsseldorf no. 4261) in accordance with CAse REport (CARE) guidelines and the 2013 Declaration of Helsinki principles.

As controls for in situ hybridization assays and immunohistochemistry, a single lymph node and gut biopsies of one HIV-1-positive control (male, 56 years old) were collected in the ACS cohort, an ongoing, prospective multicenter acute HIV-1 infection cohort in Belgium, coordinated at the HIV Cure Research Center of Ghent University Hospital (ClinicalTrials.gov ID: NCT03449706) and approved by the ethics committee of Ghent University Hospital (no. BC-00812). The collection of rectum biopsies of one HIV-1-negative volunteer (male, 76 years old) undergoing screening colonoscopy at the HIV Cure Research Center at Ghent University Hospital was approved by the ethics committee of Ghent University Hospital (nos. BC-00812 and BC-11798). The lymph node of one HIV-1-negative individual (female, 41 years old) was provided by Knight BioLibrary (institutional review board-approved, no. IRB00004918) under full ethical approval by the Oregon Health & Science University institutional review board. Rectal tissue of one HIV-1-positive individual (male, 30 years old) for MPO staining was obtained from a study reported previously¹⁴….