In this retrospective study, we indicated that aggressive treatment for CAMR patients was associated with better graft survival. However, aggressive treatment group also had higher incidence of adverse events and a reduced adverse event free survival. The factors independently associated with graft loss were proteinuria and supportive treatment.
Currently, there are no approved treatments for CAMR. Billing et al. reported a prospective pilot study on an antihumoral therapy consisting of IVIG and rituximab in 20 paediatric renal transplant recipients with CAMR. Over an observation period of 2 years, IVIG and rituximab significantly reduced or stabilized the progressive loss of transplant function (3, 10) However, the subgroup with transplant glomerulopathy (TG) was associated with a poorer response. Another study conducted by Bachelet et al. showed IVIG with rituximab treatment for severe TG in CAMR did not change the natural history of TG (4). Recently, a multicenter, prospective, randomized double-blind clinical trial for evaluation the efficacy and safety of IVIG with rituximab also revealed no difference between the treatment and placebo groups in eGFR decline, increase of proteinuria, and MFI of the immunodominant DSA. The author considered the presence of TG as an inclusion criteria (mean cg score in the treatment group: 2.3 ± 0.8), and this may be the reason of a poor response in this study (11). In fact, there was evidence that the combination of IVIG and rituximab appeared to be beneficial in patients with high levels of microvascular injury, for example biopsies with g ≥ 2 and/or (g + ptc) ≥ 4 (12). On the contrary, patients with low microvascular injury scores appeared less likely to benefit from antihumoral therapy.
Bortezomib had also been evaluated in patients with CAMR. Clinical experience of bortezomib in transplantation had shown variable results among patients with different disease states and populations. Recently, a randomized, placebo-controlled trial (the Bortezomib in Late Antibody-Mediated Kidney Transplant Rejection [BORTEJECT] Trial) investigated whether two cycles of bortezomib (each cycle: 1.3 mg/m2 intravenously on days 1, 4, 8, and 11) prevent GFR decline by halting the progression of late DSA–positive ABMR. However, no signiﬁcant differences between bortezomib- and placebo-treated groups in median measured GFR at 24 months , 2-year graft survival , urinary protein concentration, and DSA levels (5) . Advanced tissue injury and higher proportion of preformed DSAs (19 of 21 patients) might be a possible explanation. On the other hand, HLA antibodies produced by long-lived plasma cells (LLPCs) are more refractory to proteasome inhibitor therapy. LLPC resistance and immunologic compensatory mechanisms may also play a role for treatment failure (13).
In our study, CAMR were diagnosed at relatively early stage (median cg score: 1.0, ci+ct: 2.0) with compared to previous studies (mean cg score: 2.0; ci+ct :3.5 in a recent clinical trial (11) and mean cg score: 2.2; ci+ct score: 2.8 in a previous retrospective study (4)). Furthermore, the microvascular injury was prominent (median [g + ptc] score: 3.5 ). The above characteristics made our patients more likely to respond to antihumoral therapy. The graft survival was significantly better in the aggressive treatment group with compared to supportive treatment group. Supportive treatment was a predictor of graft loss in univariate analysis (HR 2.77, 95% CI [1.19-6.41], P = 0.017). After adjustment of proteinuria, creatinine, and cg score, supportive treatment was still an independent risk factor of graft loss (HR 2.86, 95%CI [1.05-7.77], P = 0.038). A subgroup analysis revealed that aggressive treatment for CAMR resulted in better graft survival in patients with proteinuria < 1.73 g/d but not in patients with proteinuria ≥ 1.73g/d. Our study highlight the importance of aggressive treatment in CAMR under the circumstance of earlier stage and higher degree of microvascular injury.
Rituximab, IVIG, and bortezomib treatment are not reimbursed by national insurance in Taiwan and should be self-paid. Therefore, some patients received methylprednisolone pulse therapy with plasmapheresis only. These patients had graft survival between aggressive treatment and supportive treatment group (supplement Fig 1). Previously, Redfield et al. had reported a retrospective study for outcome of CAMR (14). The author divided their patients into three groups: steroid/IVIG with rituximab or antithymocyte globulin, steroid/IVIG alone or in combination, and no treatment. The most aggressive treatment group had the best graft survival, which was in line with our study. However, the graft biopsies of CAMR in Redfield’s series had relatively advanced disease (median cg of 2 and proteinuria > 1g). Therefore, the graft survival in our study was better (overall median graft survival 5.4 years vs. 1.9 years).
Following aggressive treatment of CAMR, adverse event is an important issue. The most frequently prescribed antihumoral agent for our patients was rituximab, followed by IVIG, bortezomib, and ATG. A retrospective study published by Kamar et al. revealed that, following rituximab treatment, 9.1% of kidney transplant patients died due to infectious disease (15). This result was similar to that in our aggressive treatment group (10.16%). The most common adverse events in our patients were CMV disease, urinary tract infection, bacterial / PCP pneumonia, and infectious diarrhea.
CMV infection had been reported to be associated with Rituximab, ATG, and bortezomib treatment in renal transplant and myeloma patients (16-18). ABO-incompatible kidney transplant recipients who received rituximab had higher incidence of CMV disease (16). Studies revealed an increased frequency of CMV disease associated with ATG treatment, probably due to the release of TNF-α after ATG administration, which may stimulate cellular nuclear factor кB and viral replication via binding to the promoter region of the CMV immediate-early antigen gene (17). Furthermore, several studies indicated that bortezomib treatment is associated with higher risk of viral infection, including CMV (18-20). Basler M et al. demonstrated reduced cytotoxic T cell response and impaired viral clearance in bortezomib treated mice (19).
PCP is a major cause of morbidity and mortality in patients receiving immunosuppressant therapies. The risk factors for the development of PCP in kidney transplant patients include: overall load of immunosuppressive therapy, higher donor age, higher recipient age, lymphopenia, previous CMV infection, or treatment used for episodes of graft rejection (21) .
Focusing on the mortality cases in our aggressive treatment group, 3 of 6 mortality cases died of CMV disease, including 1 CMV pneumonia and 2 CMV colitis. Adequate valganciclovir prophylaxis may reduce mortality rate by 50 percent in our patients who received aggressive treatment.
Kamar et al. demonstrated that the median duration between last rituximab and first infection episode in kidney transplant recipients was about 5 months (15), which was in accordance with our study (6 months). They found the independent predictive factors for infection-induced death were the combined use of rituximab and ATG given for induction or anti-rejection therapy, recipient age, and bacterial and fungal infections. Our previous policy about prophylactic antibiotics (valganciclovir and trimethoprim-sulfamethoxazole) was to give these 2 agents for just one month after aggressive treatment. In this regards, it is reasonable to recommend that valganciclovir and trimethoprim-sulfamethoxazole prophylaxis should be given for at least 5~6 months after aggressive anti-rejection therapy.
Despite the significantly higher rate of adverse events in aggressive treatment group, there was no significant difference in patient survival (Figure 2), implying that the patients still could have a reasonable chance of survival if these complications can be treated judiciously.
There are limitations in our study. Firstly, this is a retrospective study. There were no rules for treatment of CAMR in our cohort, and the need for treatment was determined by each clinical physician. Second, DSA was performed in limited patients, thus the role of DSA can not be obtained exactly from this study. This is because determination of HLA antibody by LuminexR method is expensive in Taiwan and is not affordable to many patients. To compensate this shortcoming, annual PRA are usually performed, which is reimbursed by national health insurance.