Cardiac allograft tolerance can be achieved in nonhuman primates by donor bone marrow and kidney cotransplantation

Abstract

Long-term, immunosuppression-free allograft survival has been induced in human and nonhuman primate (NHP) kidney recipients after nonmyeloablative conditioning and donor bone marrow transplantation (DBMT), resulting in transient mixed hematopoietic chimerism. However, the same strategy has consistently failed in NHP heart transplant recipients. Here, we investigated whether long-term heart allograft survival could be achieved by cotransplanting kidneys from the same donor. Cynomolgus monkeys were transplanted with heart allografts alone or heart and kidney allografts from the same major histocompatibility complex (MHC)-mismatched donors. All animals except one received DBMT, either at the same time or after a 2- to 4-month delay, plus short-term costimulation blockade and calcineurin inhibitor treatment. Long-term, immunosuppression-free heart allograft survival was consistently achieved in heart/kidney, but not heart-alone, recipients. This was not associated with greater donor/recipient histocompatibility or altered lymphoid cell reconstitution after conditioning. The maintenance of tolerance after heart/kidney transplantation was associated with the presence of forkhead box P3 (Foxp3⁺) regulatory T cell (T_reg)-rich organized lymphoid structures in kidneys but not hearts. Substituting high-dose erythropoietin treatment for kidney transplantation was unsuccessful, suggesting that it was not the sole mechanism of action. RNA sequencing analysis revealed that gene expression in hearts from tolerant recipients closely resembled that in hearts from chronically immunosuppressed recipients but differed markedly from rejecting allografts and naïve hearts. A version of this protocol may be able to induce tolerance in patients with end-stage heart and kidney failure who require combined heart and kidney transplantation.

Fig. 1.. Protocol schematics and allograft survival.

(A) S-Protocol recipients underwent simultaneous DBMT and heart-alone (n = 6), kidney-alone [n = 8, previously published (11)], or heart/kidney transplantation (n = 9) 6 days after conditioning commenced. (B) Allograft survival rates in recipients undergoing the S-Protocol. Survival curves are indicated as follows: heart alone in dark blue, heart/kidney in orange, kidney alone in teal, and heart alone with EPO in red. Kidney alone data were published previously (11). (C) D-Protocol recipients underwent conditioning and DBMT 2 to 4 months after heart-alone (n = 4), kidney-alone [n = 13, previously published (12)], or heart/kidney (n = 4) transplantation. The dotted line signifies the commencement of conditioning. (D) Allograft survival rates in recipients undergoing the D-Protocol. Survival curves are indicated as follows: heart alone in dark blue, heart/kidney in orange, and kidney alone in teal. Kidney-alone data were published previously (12). (E). Heart-alone recipients undergoing the D-Protocol required the addition of IL-6R mAb and ATGAM at the time of organ transplantation to prevent rejection during the delay period. For (B) and (D), recipients rejecting their heart allografts after donor nephrectomy or donor skin transplantation were censored at the time of the secondary procedure, as were two recipients euthanized for PTLD or CMV pneumonia with rejection-free allografts (see Table 1). P values in (B) and (D) were calculated by log-rank test.

Fig. 2.. Chimerism in Mauritius-origin recipients.

(A) Degree and duration of peripheral donor lymphocytes, monocytes, and granulocytes over time in heart-alone recipients (M1201, M1701, M2001, M2811, M10319, and M1320) or heart/kidney recipients (M1111, M1811, M4311, M2110, M4709, M4810, and M1511) undergoing S-Protocol and heart-alone recipients (M3113, M3413, and M5212) or heart/kidney recipients (M4212, M4710, M5610, and M8314) undergoing the D-Protocol. Chimerism data in kidney-alone recipients were previously published (16). Maximal lymphoid chimerism (P < 0.003) (B) and myeloid chimerism (P < 0.27) (C) in heart-alone (n = 9) versus heart/kidney recipients (n = 12) undergoing the S- and D-Protocols. (D) Duration of lymphoid chimerism in heart-alone versus heart/kidney recipients undergoing the S- and D-Protocols (P < 0.001). P values in (B) and (C) were calculated by Mann-Whitney U test and in (D) by the log-rank test.

Fig. 3.. DSA and anti–donor T cell responses.

(A) Peripheral DSA is represented as the fold change in MFI over pretransplant/naïve baseline levels in heart-alone and heart/kidney recipients undergoing the S- and D-Protocols. An MFI over 2× pretransplant is considered significant. Values obtained after allonephrectomy or challenge skin grafting were excluded. (B) IFN-γ ELISPOT assays. The frequencies of recipient T cells capable of mounting a direct alloresponse were measured before and at different time points after DBMT in recipients of heart/kidney allografts that experienced ACR (M1511) or long-term allograft survival (M4710, M4810, and M5610). Recipient peripheral blood T cells were cultured alone (white bars), with irradiated donor PBMCs (black bars) or third-party PBMCs (gray bars), and the frequency of cells producing IFN-γ cytokine was measured by ELISPOT. The results are expressed as the number of IFN-γ spots per 150 × 10³ cells ± SD (technical replicates). P values were calculated using two-way ANOVA followed by Tukey’s post hoc test. ns, not significant.

Fig. 4.. Histology of challenge skin grafts on long-term S- and D-Protocol heart/kidney recipients.

(A) Day 160 after skin transplantation on M4311. Skin grafting was performed 651 days after DBMT and 217 days after allonephrectomy. Donor skin was accepted indefinitely, whereas three third-party skin grants were acutely rejected in 8, 8, and 12 days. (B) Day 40 after skin transplantation on M4212. Skin grafting was performed 511 days after BMT and 197 days after allonephrectomy. Donor skin survived 46 days, whereas three third-party skin grafts were rejected in 9, 9, and 23 days. (C) Day 27 after skin transplantation on M1111. Skin grafting was performed 631 days after DBMT and 120 days after native kidney explantation. Donor skin survived 58 days, whereas two third-party skin grafts were rejected in 8 and 27 days. (D) Day 21 after skin transplant on M8919. Skin grafting was performed 807 days after DBMT and 273 days after allonephrectomy. The donor skin graft survived 86 days, whereas a third-party skin graft was rejected in 21 days. (E) Day 28 after skin grafting on M1811. Skin transplantation was performed 497 days after DBMT. Donor skin was rejected on day 16, whereas three third-party skin grafts were rejected in 8, 16, and 16 days.

Fig. 5.. Volcano plots of NanoString quantitative mRNA expression.

Heart samples are shown on the left and kidney samples are shown on the right. (A and B) Volcano plots comparing gene expression in rejecting versus tolerant samples. (C to F) Volcano plots comparing gene expression in immunosuppressed versus tolerant samples [(C) and (D)] and naïve versus tolerant samples [(E) and (F)]. In all graphs, genes expressed more highly in tolerant samples are marked in red, and the top 20 most significantly differentially expressed genes are labeled even if they do not meet the criteria.