The majority of studies have enrolled patients after 6 months post-transplantation. Few early conversions, i.e., before 6 months, have been reported. Initially, KT recipients were switched early to belatacept in the context of severe renal dysfunction. Two studies assessed the results in patients with very low eGFR (8 ± 12 mL/min/1.73 m² (n = 25 patients) and 16 ± 12 mL/min/1.73 m² (n = 20 patients) after a median time of 71 [15–161] and 42 [18–74] days post-KT respectively [6, 7]. The benefit in terms of eGFR at 1-year post conversion ranged between +16.6 mL/min/1.73 m [2] and +38.2 mL/min/1.73 m [2]. In the first study (Le Meur et al.), 48% of patients had a baseline eGFR <15 mL/min/1.73 m [2] and 29.1% were on dialysis. At 1 year, only 3 patients were still on dialysis. Graft and patient survival at 1 year were 83.3% and 96% respectively. In the second study (Wojciechowski et al.), 75% of patients required dialysis post-KT and before conversion. At 1 year post conversion, graft survival was 95% and no patient was still on dialysis. Patient survival was 100% at 1 year.

Some other non-randomized studies reported results on stable transplant patients in larger numbers of patients (60–453 patients) [8–12]. eGFR at the time of conversion in these patients ranged between 19.4 to 27 mL/min/1.73 m². The gain of eGFR at 1 year post conversion ranged from +14.4 mL/min/1.73 m² to +18.6 mL/min/1.73 m². Graft and patient survival at 1 year were 83.3% and 97.2% (Bertrand et al.) and 100% and 90.9% (Moein et al.) respectively.

In a randomized controlled study, Tawhari et al. assessed the impact of early belatacept conversion (3 months) in 27 KT recipients with stable renal function (mean eGFR at conversion was 68.5 ± 18 mL/min/1.73 m [2]) [13]. Nine patients received belatacept with MMF, 8 received belatacept with low-dose tacrolimus, and 10 had no belatacept conversion. The evolution of eGFR at 2 years was +8.8 mL/min/1.73 m [2] in the belatacept plus low-dose tacrolimus patients, −0.38 mL/min/1.73 m² in the tacrolimus plus MMF group and −6.60 mL/min/1.73 m² in the belatacept plus MMF group. The rate of graft and patient survival was not different between groups (96.3% and 92.5% respectively at 2 years).

Overall, in the early post-KT period, conversion to belatacept appears to be associated with an important improvement in renal function, with acceptable graft and patient survival. The gain in GFR appears to be even higher in patients with delayed graft function in the very early phase. Further randomized studies are required to confirm the optimal use of belatacept during this period.

The renal function benefit of late conversion protocols has been demonstrated in several studies. In a randomized phase II trial of 173 patients comparing belatacept conversion to CNI at 19–20 months post-KT [14], the increase in eGFR at 36 months was +8.9 mL/min/1.73 m² in the belatacept group compared to +1.1 mL/min/1.73 m² in the CNI group (p = 0.01) [15, 16].

Budde et al. conducted a prospective randomized controlled study of belatacept conversion (n = 223) versus CNI maintenance (n = 223) at 6 months post-KT [17]. At 24 months, patient and graft survival (>97%)were similar in the 2 groups. At 24 months, the mean eGFR gain was +5.2 mL/min/1.73 m² in the belatacept group and −1.9 mL/min/1.73 m² in the CNI group (delta 7 mL/min/1.73 m²).

A recent retrospective study by Divard et al. compared 243 kidney recipients with a propensity-matched cohort of patients on a CNI-based regimen [18]. The median time to conversion was 1 year in the belatacept group, and the follow-up was 7 years. Graft survival was higher (78%) at last follow-up in the belatacept group versus 63% in the CNI group. The eGFR at 7 years was higher in the belatacept group, 26 mL/min/1.73 m² versus 20.2 mL/min/1.73 m² in the CNI group. Interestingly, a retrospective study evaluated the effect of conversion to belatacept in patients with severe vascular lesions (cv ≥ 2) and poor kidney function (eGFR between 25 and 27 mL/min/1.73 m²). The conversion to belatacept (n = 69) was found to be associated with a better graft survival at 3 years (84%) compared to patients who remained on CNI (n = 70, 65.1%) [19]. Fewer de novo DSA (7.4% versus 23.4%) but more opportunistic infections (OPIs) (7.6/100 person-years versus 1.0/100 person-years) were noted, while the rate of rejection and patient survival were similar.

Finally, the majority of patients switched to belatacept do not appear to have corticosteroids in their immunosuppressive treatments. A recent study compared 199 late-switched patients to belatacept without reintroduction of steroids versus 313 patients on concomitant steroids at the time of conversion [20]. The absence of steroids was not associated with an increased risk of PBAR or worse graft survival while the use of steroids was independently associated with worse patient survival.

To address the increased rates of AR associated with standard belatacept regimens compared to CNI-treated patients, a combined strategy with short-term tacrolimus use in addition to belatacept has emerged in KT recipients. A cohort analysis of 50,244 patients including 417 patients receiving belatacept plus tacrolimus, 458 receiving belatacept, and 49,369 receiving tacrolimus has shown that the rate of AR was similar in tacrolimus and tacrolimus plus belatacept-based regimens and lower than in the belatacept regimen alone [43]. In contrast, eGFR and NODAT were higher and lower, respectively in the tacrolimus plus belatacept-treated patients than in the tacrolimus-treated patients. Results from a non-randomized study compared the modified belatacept-tacrolimus regimen (n = 87) with standard belatacept (n = 97) and tacrolimus treatments (n = 205) [44]. Patients also received Basiliximab induction, MMF, and corticosteroids. In the modified regimen, tacrolimus was administered for 3 months before tapering. At 3 months, the AR rates were similar for belatacept-tacrolimus (15%) and tacrolimus (17%), but nearly twice as high for belatacept (38%). However, the AR rate at 12 months for belatacept-tacrolimus (33%) was between that of tacrolimus (20.5%) and belatacept (50.5%). The rates of Banff grade IIB or III AR were 5%, 4%, and 13%, respectively. Despite higher AR rates, graft and patient survival at 3 years were similar between groups. To overcome the relapsed rate of AR, the tacrolimus exposure was extended to 9 months before being tapered within 2 months [44]. The 12-month AR rate for belatacept-extended tacrolimus was lower than in the historical tacrolimus cohort (16% vs. 20.5%), with 4% of patients experiencing Banff grade IIB or III AR. Over 3 years, the mean estimated GFR was higher for both belatacept-tacrolimus regimens than for standard tacrolimus treatment. Viremia rates for CMV and BK virus were similar between regimens suggesting that a belatacept-based regimen with transient tacrolimus use may yield AR rates comparable to those of standard CNI-based regimens without increasing infectious risks. Moreover, in a recent retrospective study analyzing the risk of de novo DSA based on the donor-recipient eplet mismatch showed that the risk was lower in the group of patients that received belatacept plus a transient exposure to tacrolimuns (n = 294) compared to the patients that received a tacrolimus-based regiment (n = 294) (hazard ratio [HR] = 0.4). The rate of antibody mediated rejection and acute rejection were also lower (HR = 0.2 and 0.45 respectively) [31].

Mammalian target of rapamycin (mTOR) is a protein kinase that has a central role in the regulation of cell metabolism, immune function, proliferation and migration. Sirolimus and everolimus are 2 mTOR inhibitors (mTORi) approved for the prevention of organ rejection in transplant recipients. The combination of belatacept with mTORi is an interesting association, allowing to remove CNI-related nephrotoxicity and adding the potential benefits of mTORi, such as antitumor and potential anti-CMV activity [45].

A randomized controlled study conducted by Ferguson et al. compared the evolution of belatacept de novo associated with MMF (33 patients), with sirolimus (26 patients) and with a standard group receiving tacrolimus with MMF (30 patients) [46]. At 1 year, the rate of BPAR was 4% in the mTORi group, and the mean eGFR was 61.8 mL/min/1.73 m². The safety profile, along with patient and graft survival was similar between groups. The recovery, post-antithymoglobulin injection, of peripheral blood CD4⁺, CD8⁺, memory CD4+ and regulatory T cells was also similar between the different groups.

In 2014, Kirk et al. assessed the outcome of 20 KT recipients from non-HLA identical living donors who received alemtuzumab induction therapy followed by de novo belatacept and sirolimus [47]. Patients were randomized to receive or not receive unfractionated donor bone marrow. Three patients were switched to MMF because of sirolimus-related side effects. At 1 year, no clinical or histological rejection occurred and the mean eGFR was 89 ± 3.5 mL/min/1.73 m². Safety was also excellent with no admissions for infection or malignancy. Interestingly, 10 patients reduced their immunosuppressive therapy and seven of these experienced no rejection on belatacept monotherapy. Safety was good: five patients had spontaneously resolving EBV viremia and 1 patient had a CMV viremia that resolved after increasing the prophylaxis dose.

From a cellular point of view, memory T cells may lose the expression of CD28, and thus escape the effect of belatacept and are implicated in the high rate of rejection in de novo studies reported above. After induction with a depleting agent, there is a marked increase in effector memory and terminally differentiated effector memory cells CD28-CD57⁺CD8⁺ T cells. In vitro and in vivo studies have shown that mTORi are able to suppress the expansion and the differentiation of these cells and thus reduce the risk of belatacept-resistant rejection [48, 49]. These cells have been shown to be more frequent in patients with belatacept-resistant rejection with increased expression of the mTOR pathway [22]. In CD4⁺CD57⁺ T cells, the mTOR pathway was not downregulated in belatacept-resistant cells as compared to belatacept-sensitive cells [49]. Taken together, these data suggest an interesting additional effect of mTORi in targeting belatacept-resistant CD8⁺ and CD4⁺ T cells.

The association of belatacept and mTORi may also be considered in post-KT conversion from the belatacept-MMF regimen to the belatacept-mTORi regimen. Very recently, Del Bello et al. reported their experience in 35 patients who were switched from MMF to mTORi in combination with belatacept [50]. They showed a lower incidence of CMV DNAemia in this group (incidence of 0.035/month of exposure) as compared to a propensity-matched cohort of belatacept–MMF treated patients (incidence of 0.072/month of exposure).

The use of a depleting agent may be beneficial in combination with the use of belatacept de novo to prevent belatacept-resistant rejections, as it reduces the rate of rejection when associated with CNI [51]. However, in clinical practice, antithymoglobulin failed to prevent these rejections [52]. In a mouse model, Muckenhuber et al. showed that antithymoglobulins induce an important pro-inflammatory cytokine release, including IL-6, and that blocking IL-6 in addition to a de novo belatacept regimen prevents the occurrence of belatacept-resistant rejection and prolongs graft survival [53]. This combination promoted intragraft immune regulation and increased regulatory T cells within the graft.

Additionally, Herr et al. showed that the CD4⁺CD57+PD1-memory T cell population, associated with belatacept-resistant rejection, had more IRF7 transcript (associated with Interferon-α (IFN-α) and IL-6 regulation) [54]. Inhibition of IL-6, along with type I IFN-α, reduced the proliferation of these belatacept resistant cells.

In de novo studies, belatacept is associated with a lower rate of de novo DSA occurrence. Non-human studies have also shown the effect of belatacept in impairing the class switching of B cells. In a situation of high risk of immunization patients returning to dialysis, several teams continue immunosuppressive therapy for variable periods of time to prevent sensitization that impairs access to another transplantation, despite the associated increased risks of toxicity and infection [55]. To reduce sensitization, Badell et al. tested in a randomized study the efficacy of using belatacept in this setting in 60 patients, compared to immunosuppressive discontinuation in 7 patients. They found that belatacept reduced the incidence of de novo DSA and prolonged its onset, with a comparable safety profile [56].

For patients who are already sensitized, several strategies have been proposed to reduce or eliminate anti-HLA antibodies. The majority of strategies target B cells or long-lived plasma cells. Rituximab, which mainly targets B cells has failed to demonstrate significant efficacy. Proteasome inhibitors are effective in targeting antibody-producing cells but a rebound of antibodies is often seen [57]. The association of belatacept in this setting may be of interest because of its effect on germinal centers and since long-lived plasma cells re-express CD28 [58–60]. In non-human sensitized primate models, this strategy was effective in preventing DSA rebound as compared to standard immunosuppression with tacrolimus and MMF [61–64]. The “dual targeting” combination of belatacept and proteasome inhibitor on germinal centers was tested to desensitize 4 highly-sensitized heart transplant candidates and in antibody-mediated rejection post KT [65, 66]. This strategy was able to reduce anti-HLA antibodies and DSA. After discontinuation of proteasome inhibitors, belatacept was able to prevent antibody rebound in the majority of patients. Circulating cell analysis showed a reduction in naïve and memory B cells and of T follicular helper cells.