Alefacept promotes immunosuppression-free renal allograft survival in nonhuman primates via depletion of recipient memory T cells

Abstract

Renal allograft tolerance has been achieved in MHC-mismatched primates via nonmyeloablative conditioning beginning 6 days prior to planned kidney and donor bone marrow (DBM) transplantation. To extend the applicability of this approach to deceased donor transplantation, we recently developed a novel conditioning regimen, the “delayed protocol” in which DBM is transplanted several months after kidney transplantation. However, activation/expansion of donor-reactive CD8⁺ memory T cells (TMEM) occurring during the interval between kidney and DBM transplantation impaired tolerance induction using this strategy. In the current study, we tested whether, Alefacept, a fusion protein which targets LFA-3/CD2 interactions and selectively depletes CD2^highCD8⁺ effector memory T cells (TEM) could similarly induce long-term immunosuppression-free renal allograft survival but avoid the deleterious effects of anti-CD8 mAb treatment. We found that Alefacept significantly delayed the expansion of CD2^high cells including CD8⁺ TEM while sparing naïve CD8⁺ T and NK cells and achieved mixed chimerism and long-term immunosuppression-free renal allograft survival. In conclusion, elimination of CD2^high T cells represents a promising approach to prevent electively the expansion/activation of donor-reactive TEM and promotes tolerance induction via the delayed protocol mixed chimerism approach.

Introduction

Renal allograft tolerance has been successfully achieved in MHC-mismatched non-human primates (NHP) and humans through combined kidney and donor bone marrow transplantation (DBMT) performed after a 6-day pretransplant nonmyeloablative conditioning regimen (1–3). In this model, transient donor hematopoietic chimerism proved to be sufficient to induce long-term allograft survival in the absence of ongoing immunosuppression. To extend this approach to deceased donor transplantation, we recently developed a novel strategy, the “delayed tolerance induction” protocol, with which recipients initially undergo kidney transplantation (KTx) with conventional immunosuppression and then receive conditioning and DBMT 4 months later. In this setting, additional treatment with an anti-CD8 monoclonal antibody was found to be necessary to deplete alloreactive memory T cells (TMEM) activated during the interval between the kidney and DBM transplants. However, while overall and long-lasting depletion of CD8⁺ T and NK cells did promote chimerism and allograft tolerance (4, 5), this treatment was associated with a high incidence of viral infection and EBV related lymphoma in the allograft recipients (5). This observation stressed the need for more selective strategies designed to deplete or inactivate CD8⁺ TMEM cells while sparing naïve CD8⁺ T cells and NK cells and thereby preserving some immune competency.

LFA3-Ig (Alefacept) is a humanized chimeric fusion protein consisting of the extracellular CD2-binding portion of the human leukocyte function antigen-3 (LFA-3) adhesion molecule linked to the Fc (hinge CH2 and CH3 domains) portion of human IgG1. LFA3-Ig is known to deplete selectively CD8⁺ effector memory T cells (TEM) while sparing naïve CD8⁺ T cells (6). In the current study, we show that LFA3-Ig administration selectively depleted CD8⁺TEM while preserving CD8⁺ naïve T cells and NK cells and promoted mixed chimerism and long-term immunosuppression-free renal allograft survival in the delayed tolerance approach without the complications of opportunistic infections or lymphoma disorders.

Materials and methods

Animals

Cynomolgus monkeys that weighed 3 to 7 kg were used (Charles River Primates, Wilmington, MA). All cynomolgus monkeys (n = 21) received the same conditioning regimen with or without additional treatment with anti-CD8 mAbs or Alefacept (Supplemental Table 1 and Fig. 1). All surgical procedures and postoperative care of animals were performed in accordance with National Institute of Health guidelines for the care and use of primates and were approved by the Massachusetts General Hospital Subcommittee on Animal Research.

Figure 1.

All recipients underwent KTx with conventional triple drug immunosuppression followed by conditioning and DBMT after 4 months. Group A received a standard conditioning regimen, including low dose TBI, thymic irradiation, hATG and anti-CD154 mAb. Group B was administered humanized anti-CD8 mAb 5 mg/kg on Days 0 and 2 after DBMT in addition to the standard regimen. Group C underwent the standard conditioning regimen and received LFA3-IG (Alefacept) instead of anti-CD8 mAb. All recipients received Cyclosporin A for 28 days after DBMT.

Cynomolgus MHC genotyping

MHC characterization was performed as previously described (7,8). Briefly, genomic DNA was prepared from PBMC and splenocytes. Panels of seventeen microsatellite loci spanning ~5 Mb of the MHC region were amplified from the genomic DNA with fluorescent-labeled PCR primers and fragment size analysis was determined. The microsatellite haplotypes for each animal were converted to predicted MHC genotypes based on previous cloning and sequencing work with cynomolgus monkeys (7,8). All MHC typing of recipient/donor pairs in Groups A–C are shown in Supplemental Fig. 1.

Conditioning Regimens

All recipients initially underwent KTx alone with a conventional triple drug immunosuppressive regimen including tacrolimus (AstellasPharma Inc. Osaka, Japan), mycophenolate mofetil (Roche Inc., Nutley, NJ) and prednisone. Four months later, the recipients underwent conditioning and DBMT. The conditioning regimen consisted of low-dose total body irradiation (TBI, 1.5 GyX2) on days −6 and −5(relative to DBMT), thymic irradiation (TI, 7 Gy) on day-1, equine ATG (ATGAM, Pharmacia and Upjohn, Kalamazoo, MI, 50 mg/kg/day on days −2, −1 and 0) and anti-CD154 monoclonal antibody (anti-CD40L, Hu5C8, provided by K. Reimann DVM, University of Massachusetts, Worcester, MA, 20 mg/kg on Days 0 and +2, and 10 mg/kg on days 5, 7, 9, and 12) (Fig. 1).

Group A: Five recipients received only the standard conditioning regimen four months after KTx (Fig. 1, black). Group B: A course of humanized anti-CD8 mAb (cM-T807 provided by Centocor Inc. Horsham, PA) was added in 13 recipients at 5mg/kg on Days 0 and 2 after DBMT was added to the standard regimen (Fig. 1, green). Group C: three recipients received LFA3-Ig. (Alefacept 1mg/kg on days −1, 5, 12 and 19)(provided by AstellasPharma US, Inc, Northbrook, IL) in place of the anti-CD8 mAb (Fig. 1, blue). All recipients (except for one Group B recipient, which later died due to right atrial thrombosis), were pre-treated with intravenous Ketorolac (1 mg/kg on Days −1 and 0) in order to prevent anti-CD154 mAb-induced thrombosis, as previously described (2). Following DBMT, recipients in all groups were administered a short course of cyclosporine A (28 days) after which no further immunosuppression was administered.

Renal and bone marrow transplantation

Kidney transplantation (KTx) was performed as reported previously (7). The recipients also underwent unilateral native nephrectomy and ligation of the contralateral ureter on day 0. The remaining native (hydronephrotic) kidney was removed 60–80 days after transplantation. Bone marrow was harvested from the surviving original kidney donor’s iliac bones by multiple percutaneous aspirations. If the animal was sacrificed at the time of donor nephrectomy, bone marrow cells were also harvested from the vertebral bones after euthanasia and kept frozen until DBMT. The bone marrow cells (1.0 – 3.0 × 10⁸ mononuclear cells/kg) were infused intravenously.

Flow cytometric analyses, detection of chimerism and cell sorting

PBMCs, peripheral lymph nodes, spleen and bone marrow cells were labeled with a combination of the following mAbs: CD3 PerCP (SP 34-2), CD4 PerCP (L-200), CD8 PerCP (SK1), CD8 APC (SK1), CD95 FITC (DX2), CD95 APC (DX2), and CD28PE (CD28.2) (BD Pharmingen, San Jose, CA). Lymphocytes from the animals treated with anti-CD8 mAbs were stained with anti-CD8-PE (DK25, Dako, Inc., Carpenteria, CA). For chimerism analyses, we used an anti-MHC class I HLA mAb (H38, One Lambda, Inc, CA) reacting specifically with certain MHC class I antigens. The recipient and donor pairs were chosen based on their differential reactivity to this mAb. The fluorescence of the stained samples was analyzed using FACS Calibur and FACS Scan flow cytometers and Cell Quest Software (BD), or FlowJo software. For assessing memory T cell functions, fresh PBMCs were gated on lymphocytes and sorted into CD95⁻CD28⁺ naïve and CD95⁺CD28^low/high memory populations using a FACS Vantage cell sorter (BD Immunocytometry System). The purity of sorted cells was consistently > 95% as previously described (8).

Phenotype, localization and frequency of memory T cells in cynomolgus monkeys

The monoclonal antibodies anti-CD95 and anti-CD28 were used to detect naïve T cells (CD95−CD28+) as well as central (TCMs, CD95+CD28+) and effector (TEMs, CD95+CD28−) memory T cells as previously described (8, 9).

Measurement of T cell-mediated alloresponses by ELISPOT

ELISPOT plates (Millipore, Bedford, MA) were pre-coated with 5 μg/ml of capture antibodies against γIFN (Mabtech, Sweden) in PBS and stored overnight at 4°C. The responding cells separated from fresh PBMCs were co-cultured with an equal number of irradiated donor PBMCs as stimulating cells (1.5 × 10⁵ cells/well), or unstimulated in medium alone, or with PHA at 1 μg/ml (Sigma). After 44 hours incubation at 37°C, the plates were washed and biotinylated detection antibodies (Mabtech, Sweden) were added (4°C OVN). After 5 washes with PBS, streptavidin-horseradish-peroxidase conjugate in PBS BSA 0.5% (Dako, Glostrup, Denmark) was added for 2 hrs at room temperature, followed by 5 washes. Finally, 50 μl/well of 3,3,5,5-tetramethylbenzidine (TMB) liquid substrate (Sigma-Aldrich) was added and incubated for 30 min in the dark. The resulting spots were counted with an ELISPOT image analyzer (CTL Inc., Cleveland, OH), as described elsewhere (10).

Histological analyses

Protocol Renal biopsies were obtained every 2–4 months in recipients with stable function as well as whenever a rise in serum creatinine occurred. Tissue was processed for routine microscopy and a portion frozen for immunofluorescence staining. Other organs obtained surgically (lymph nodes, native kidney and spleen) were similarly processed. Following euthanasia of any monkey, complete autopsies were performed for histopathologic examination of the renal allograft, lymph nodes, heart, lung, liver, pancreas, thymus and skin.

Statistical analyses

We used a repeated measures analysis of variance to test whether the abundance of each cell subset was greater during the treatment period (days 0–50) than during the pre-transplant period. The data were log-transformed before analysis. In addition, we calculated the log-fold change for each cell and compared them in pairs using a paired t-test.

Results

Alefacept suppresses γIFN production of CD8⁺ T memory cells in vitro

To investigate the in vitro effects of LFA3-Ig on primary direct alloresponse by memory T cells, CD8⁺ and CD4⁺ memory T cells (TMEM; CD95+) and naïve T cells (CD95−) were isolated from naïve monkeys and were cultured with allogeneic irradiated stimulator cells (mixed lymphocyte reaction: MLR) and different concentrations of LFA3-Ig (0 μg/ml, 1 μg/ml, and 10 μg/ml) for 48 hours. The frequencies of γIFN-producing cells were evaluated using an ELISPOT assay (10). We observed a dose-dependent reduction of MLR reactivity with either whole PBMCs or CD8⁺ TMEMs (CD8⁺CD95⁺) in the presence of LFA3-Ig Alefacept (Fig. 2). In contrast, Alefacept had no significant effect on the direct alloresponse mediated by CD4⁺ TMEMs (CD4⁺CD95⁺) (Fig. 2) and by naïve T cells (CD3⁺CD95⁻) (Supplemental Fig. 2)

Figure 2.

γIFN response in MLR was determined by ELISPOT. A dose-dependent reduction of the MLR in the presence of LFA3-Ig was observed in either whole PBMCs or CD8⁺ TMEMs. In contrast, no significant effect on the direct alloresponse mediated by CD4⁺ TMEMs was observed.

Both anti-CD8mAb and Alefacept promote the development of donor hematopoietic chimerism

Four months after receiving MHC mismatched (Supplemental Fig. 1) kidney allografts, the recipients underwent nonmyeloablative conditioning (Fig. 1) and DBMT from the original kidney donor. Based upon the type of additional anti-memory T cell treatment, the recipients were divided into Groups A (no additional treatment), B (anti-CD8 mAb) and C (Alefacept) (Fig. 1 and Supplemental Table 1). Recipients in Groups B and C developed significantly greater levels of lymphoid and myeloid chimerism than those in Group A (p<0.0001 and p<0.0007, respectively) (Figs. 3 A–B). There was no statistically significant difference with regards to the magnitude and duration of donor hematopoietic chimerism between Groups B and C.

Figure 3.

(A and B) Comparable levels and duration of Lymphoid and Myeloid chimerism were detected in both Groups B and C in contrast with Group A where no multi-lineage chimerism was detected. (C) Profound depletion of NK cells was observed only in Group B. (D–F) No significant difference in depletion and recovery of CD4⁺ T cells among the three groups. (G and H) Group B showed a complete and long-lasting depletion of both CD8⁺ naïve and memory T cells. In Group C, although replenishment of CD8⁺ TMEM, including TEM and TCM was impaired as compared to Group A, it developed much faster than Group B. (I) Recovery of naïve CD8⁺ T cells was accelerated in Group C.

Selective prevention of CD8⁺ T memory cell expansion by Alefacept

The depletion and recovery of naïve (CD95⁻CD28⁺) and each TMEM subset, central memory T cells (TCM; CD28+CD95+), and effector memory T cells (TEM;CD28−CD95+)(8, 9) among CD4⁺ and CD8⁺ T cells isolated from monkey PBMCs were investigated (Fig. 3 D–I). In all groups, we observed a rapid depletion of all T cell subsets following the conditioning. In Group A, T cell recovery was initially detectable as early as d10 post-DBMT. These T cells were comprised of some naïve T cells and a majority of TMEM corresponding predominantly to CD8⁺ TEM (> 70% TMEM), which is consistent with homeostatic expansion of CD8⁺ TEM regularly observed after leukodepletion (Fig. 3 G–I) (11). At day 30 post-DBMT, the frequencies of CD8⁺ TEM in Group A recipients had returned to pre-DBMT values. In contrast, anti-CD8 mAb treatment (Group B) resulted in complete and long-lasting (> 30 days) depletion of both CD8⁺ naïve and memory T cells (Fig. 3 G–I). T-cell recovery in the Alefacept–treated recipients (Group C) proved to be most interesting. Recovery of naïve CD8⁺ T cells was most rapid in this group. (Group C vs. A p=0.038, Group C vs. B p<0.001) (Fig. 3I). On the other hand, replenishment of CD8⁺ TMEM, including TEM (Fig. 3G) and TCM (Fig. 3H) was significantly impaired as compared to that observed in the absence of Ab treatment, although this recovery was observed earlier than in monkeys treated with anti-CD8 mAbs (p<0.001 in both TEM and TCM) (Fig. 3 G–H). Finally, it is important to note that anti-CD8 mAb treatment but not Alefacept resulted in a profound and durable depletion of NK cells (p<0.0001) (Fig. 3C).

Alefacept treatment prolonged immunosuppression-free renal allograft survival without adverse effects associated with over-immunosuppression

None of the recipients in Group A developed multilineage mixed chimerism and 3 of 5 developed rejection (one Group A recipient survived for 703 days but with chronic rejection detectable on serial biopsies after day 130). Eleven of the thirteen recipients treated with anti-CD8 mAb (Group B) developed transient mixed chimerism. However, three subsequently died due to EBV related lymphoma or BK virus infection, and one died due to thrombogenic complications presumably associated with anti-CD154 mAb treatment. In addition, three lost their allograft due to rejection. Thus, only half of the recipients (6/13) successfully achieved long-term renal allograft survival (Supplemental Table 1 and Fig. 4A). In contrast, all three kidney transplants placed in monkeys treated with Alefacept (Group C) survived long-term (> 1 year) (Supplemental Table 1 and Fig. 4A) with no signs of rejection (Supplemental Fig. 3A) and no opportunistic infections or lymphoma disorders.

Figure 4.

A: All recipients in Group A failed to develop multilineage mixed chimerism and failed to achieve long-term survival with one exception, which survived for 703 days but with chronic rejection detectable after day 130. Eleven of the thirteen recipients in Group B developed transient mixed chimerism. However, three died due to EBV related lymphoma or BK virus infection and another died due to thrombogenic complications presumably associated with anti-CD154 mAb treatment. Three lost their allografts due to rejection. Approximately half of the recipients (6/13) achieved renal allograft tolerance in Group B. In contrast, all three recipients in Group C survived long-term (> 1 year after DBMT) with no signs of rejection and no opportunistic infections or lymphoma.

B: γIFN ELISPOT assays were performed before kidney, DBMT and multiple time points after DBMT to evaluate anti-donor or anti-third party responses of bulk PBMC, CD4 TMEM and CD8 TMEM. Although some anti-donor bulk PBMC responses were detectable in two of three recipients (M8110 and M2311), anti-donor CD4 and CD8 TMEM responses became undetectable after DBMT in all three recipients.

Donor specific TMEM unresponsiveness in all three recipients treated with Alefacept

We employed γIFN ELISPOT to evaluate anti-donor or anti-third party responses of bulk PBMC, CD4 TMEM and CD8 TMEM. Although some anti-donor bulk PBMC responses were detectable in two of three recipients (M8110 and M2311), anti-donor CD4 and CD8 TMEM responses became undetectable after DBMT in all three recipients (Fig. 4B).

Discussion

We and others have previously observed that the presence of preexisting TMEMs in primates constitutes a major barrier to induction of allograft tolerance (8, 10, 12). We, therefore, hypothesized that donor-specific TMEMs generated or reactivated after placement of kidney allografts would likely limit the success of subsequent attempt to induce tolerance. Evidence supporting this hypothesis was provided by our previously reported observations that myeloablative conditioning followed immediately by combined kidney and DBMT regularly achieved tolerance of the kidney allograft but the same protocol failed to induce tolerance if applied four months after placement of the kidney allograft whose survival was initially maintained with calcineurin inhibitors and steroids (delayed tolerance induction) (4). This prompted us to test whether deletion of alloreactive TMEM before DBMT could restore the capacity of the conditioning regimen to induce donor chimerism and long-term graft survival in the delayed tolerance protocol. Initially, we showed that pre-BMT treatment with anti-CD8 mAbs effectively prevented the expansion/activation of CD8⁺ TMEMs post-BMT and allowed mixed chimerism induction and long-term acceptance of kidney allografts that had been transplanted 4 months previously (4, 5). However, the resultant severe and long-lasting depletion of all CD8⁺ cells, including naïve and memory CD8⁺ T cells and NK cells, in these recipients was associated with a significant incidence of viral infections and lymphoma (5). Therefore, we sought alternative modalities to more selectively suppress CD8⁺ TMEMs. Alefacept has been reported to mediate cognate interactions between cells expressing human CD2 and CD16 to activate cells, e.g., increase extracellular signal-regulated kinase phosphorylation, up-regulate cell surface expression of the activation marker CD25, and induce release of Granzyme B (13–16). A previous study in NHP from Kirk’s laboratory had shown that Alefacept modulates the function of CD2 and depletes efficiently primate CD95⁺CD28⁻ TEM, leading to prolongation of kidney allograft survival when combined with lymphocyte costimulation blockade (6). As in that study, we have observed depletion of TEM but not naïve CD4⁺ and CD8⁺ T cells following Alefacept treatment. This property of Alefacept was shown to rely on the inverse relationship between the surface expression of CD2 and CD28 on T cells (6). Most importantly, we observed that this selective and transient depletion of CD8⁺ TEM promoted tolerance induction via mixed chimerism without clinically overt opportunistic infections and lymphoma disorders observed with anti-CD8 mAb treatment. However, recent NHP studies in kidney or islet transplantation have shown increased incidence of CMV reactivation by adding Alefacept (17, 18) and further studies will be necessary to conclude incidence of opportunistic infection by Alefacept.

The present study indicates that treatments using agents targeting CD2, such as Alefacept, may be used to avoid tolerance resistance mediated by TMEM in allosensitized monkeys. In support of this view are patients successfully tolerized of kidney allografts, using a mixed chimerism approach similar to that applied in monkeys, who had been treated with an anti-CD2 mAb (3).

Alefacept was originally approved for the treatment of patients with psoriasis (15, 16). However, the current manufacturer (AstellasPharma US Inc.) of clinically approved Alefacept (Amevive) recently decided to cease production and sales of Amevive. This decision was apparently not based upon safety concerns but was driven by business considerations. This is unfortunate since our study along with previous observations reported by Kirk’s laboratory clearly shows that this agent could be instrumental to successful tolerance of allogeneic transplants in clinical settings. Unfortunately, because of the company’s decision to discontinue further Alefacept research, we had to stop these studies after completing three recipients. Although the initial results are compelling, the efficacy and minimized toxicity of this strategy in our tolerance approach needs to be confirmed using alternative modalities that effectively suppress CD8 TMEMs.

Supplementary Material

Supp Fig S1-S3

Supplemental Figure 1. Panels of seventeen microsatellite loci were amplified and fragment size analysis was determined. The microsatellite haplotypes for each animal were converted to predicted MHC genotypes based on previous cloning and sequencing work with cynomolgus monkeys. A few recipient and donor pairs were haplo-matched while all the others displayed more than 4/8 CyLA (cynomolgus leukocyte antigen) mismatches.

Supplemental Figure 2. The frequencies of peripheral blood naïve T cells producing γIFN following a 48 hours’ exposure to medium alone or allogeneic stimulators (mixed lymphocyte reaction), with or without LFA3-Ig were measured by ELISPOT in two monkeys. The results are expressed as number of γIFN spots per million T cells.

Supplemental Figure 3. A: Renal allograft biopsy taken from M2211 on day 647 after donor bone marrow transplantation showing no evidence of acute or chronic rejection.

B: Renal allograft biopsy taken from M3505 in Group B on day 170. showing acute cellular rejection.

C, Majority of infiltrating cells in M3505 were CD4+CD3+ T cells (CD3 staining is not shown).

D; Infiltration of CD8+ T cell is rare in M3505 who was treated with anti-CD8 mAb.

E: Infiltrating T cells are positive for Ki67, indicating that T cells are proliferating in the allograft.

List of abbreviations

NHP

nonhuman primates

TMEM

memory T cells

TEM

effector memory T cells

TCM

central memory T cells

KTx

kidney transplantation

DBM

donor bone marrow

DBMT

donor bone marrow transplantation