Complement Inhibition in Kidney Transplantation: Where Are We Now?

During transplantation, complement is involved at different time points. Relevant factors are the donor type (deceased or living donor), ischemia/reperfusion (cold and warm ischemia time) but also antigen mismatch, the occurrence of donor-specific antibodies and rejection events. Many components of the complement cascade are primarily formed in the liver, but can also be produced locally in response to a damaging stimulus [18]. The relevance of locally produced complement has been demonstrated in a mouse transplantation model using C3-deficient isografts showing only mild reperfusion injury compared with wildtype grafts when transplanted in a C3-positive recipient [19]. Gene expression analyses in biopsies taken before transplantation revealed significantly higher expression of various complement genes in kidneys from deceased donors compared with living donors [20]. Furthermore, complement factors are expressed at significantly higher levels in deceased donors at later time points after renal transplantation [20, 21], and correlate significantly with cold ischemia time [21].

The importance of complement activation in mediating ischemia/reperfusion (I/R)-induced tissue damage has been demonstrated in animal models using mice deficient for a particular complement factor or by using different complement inhibitors [22‐26]. In this regard, complement activation does not appear to be restricted to one pathway, as both inhibition of the alternative pathway by factor B deficiency [24] or anti-factor B antibodies [27], and C1-inhibitor therapy were successful in reducing I/R damage [28]. The C1-inhibitor SERPING1 is a serine esterase inhibitor that blocks C1s and C1r proteases of the classical pathway and MASP2 of the lectin pathway [29].There is an increasing body of evidence that the lectin pathway is of particular importance in mediating I/R injury [30‐33]. The pattern recognition molecule collectin-11 recognizes hypoxia-induced fucosylated ligands, which allow the formation of a complex with MASP2 and subsequent activation of the lectin pathway. Consequently, deficiency for collectin-11 but also treatment with l-fucose prevented I/R injury in mice [34, 35]. Delayed graft function (DGF) is a major consequence of a profound kidney injury mediated by different factors resulting from unstable hemodynamics, impaired homeostasis and circulating DAMPs from injured cells due to brain death, hypoxia or related to the primary disease. Reperfusion of the donor organ in the recipient exacerbates organ damage. Complement activation was shown to be observed in brain death donors before I/R as demonstrated by increased expression of complement factors in donor organs [36] and systemic complement activation [37]. Accordingly, sC5b-9 levels can be used as a sensitive marker to predict DGF [38] and donor treatment or ex vivo complement inhibition is thus a promising way to prevent the earliest effects of complement activation. This concept was successfully tested in a rat model of kidney transplantation [39] and was also used in the EMPIRIKAL study using the C3 inhibitor mirococept [40] (Table 1). Complement-mediated injury can be a direct consequence of the formation of the MAC or indirectly caused by enhancement of the inflammatory response by anaphylatoxins C3a and C5a. The anaphylatoxins are involved in leukocyte chemotaxis and activation as well as inducers of the production of pro-inflammatory mediators like cytokines and chemokines [41]. The C5a/C5aR1 axis has been shown to be critically involved in mediating I/R injury [42, 43]. Moreover, C5a/C5aR2 signaling in renal I/R is involved in activation of inflammatory cells but not in chemotaxis [44]. Thus, the C5a/C5aR axis represents a possible target for treatment of I/R injury.

Antibody-mediated rejection (ABMR) is the leading cause of subsequent kidney transplant failure [45], but efficient treatment options are lacking. Components of the complement system are involved in both regulation of the humoral response and ABMR-mediated allograft injury. In ABMR, HLA IgG alloantibodies produced by plasma cells bind to the donor antigens on graft microvasculature, leading to complement activation, margination and activation of inflammatory cells, and endothelial cell injury, sometimes with intimal arteritis. Donor-specific antibodies (DSA) play a key role in mediating ABMR pathology and are present at the time of transplantation in sensitized patients who have developed DSA due to a previous transplant, but may also be formed by the recipient later after transplantation. When DSAs bind to surface antigens, such as HLA antigens on endothelial cells, C1q recognize these immune complexes and can initiate the complement cascade via the classical pathway, ultimately leading to MAC formation and endothelial cell lysis [46]. This process is known as complement-dependent cytotoxicity. Additional to activation via the classic pathway, activation can also occur via the lectin pathway; in this case, sugar residues on IgM and IgG antibodies are recognized by mannose-binding lectin (MBL), for example, and initiate the activation cascade [47].

Not all antibodies have the ability to bind complement. In a study with more than 1000 transplanted patients on the presence of complement fixing antibodies it was shown that complement-binding donor-specific anti-HLA antibodies are helpful for the diagnosis and risk assessment of transplant rejection. The presence of C1q-fixing DSAs was associated with an increased rate of ABMR, a more severe graft injury phenotype with more extensive microvascular inflammation, and increased deposition of complement fraction C4d within graft capillaries [48]. Detection of complement-fixing DSAs allowed detection of ABMR also in C4d-negative cases [48]. Capillary C4d deposits have been established as a marker for ABMR and have been included in the Banff classification for ABMR diagnosis [49]. However, C4d deposits are not ABMR-specific and have been observed in other renal diseases [50], and the absence of C4d deposits is by no means an exclusion criterion for ABMR [51].

Complement is involved not only in antibody-mediated injury but also in the regulation of antibody production [52]. Binding of C3d-opsonized antigen from injured cells to CR2 on B cells promotes B-cell activation and antibody production by lowering the activation threshold [53]. Memory B-cell maintenance is achieved by binding of C3d-fixed antigens to CR2 on the surface of follicular dendritic cells [54]. B cells in the marginal zone can acquire intact major histocompatibility complexes from dendritic cells by complement-dependent trogocytosis for presentation to T cells [55]. Activated B cells in germinal centers receive co-stimulatory signals from T helper cells. In these stimulated germinal center B cells, the expression of complement regulators on the surface shifts, enabling activation of complement receptors on germinal center B cells, which is required for affinity maturation [56]. Positively selected B cells with high affinity can then differentiate into plasma cells that produce antibodies reactive with the donor HLA. Subsequent sublytic complement attack may stimulate endothelial cells to activate CD4+ and CD8+ T cells, promoting cellular and humoral rejection [57]. Anaphylatoxins C3a and C5a also play a role in ABMR by indirectly participating in the activation of B cells and polarization of T cells [58].

In clinical trials, two main inhibitors have been tested to prevent DGF after renal transplantation: (i) eculizumab, a recombinant humanized monoclonal antibody targeting C5, a key molecule of terminal complement activation, and (ii) C1-esterase inhibitors, also known as SERPING1, either purified from plasma or recombinantly produced and provided by different companies, inhibiting early activation of the classical and lectin-mediated pathway. An overview of all clinical trials investigating complement inhibition in early I/R injury to prevent DGF after transplantation is shown in Table 1.

Studies investigating the efficacy of eculizumab in preventing DGF initially used a single dose of eculizumab 1200 mg in a pilot study (ClinicalTrials.gov identifier: NCT01403389), which was administered before reperfusion. However, this study was stopped after an interim evaluation and the treatment was changed to another pilot study that used an additional eculizumab dose 12–24 h after transplantation (NCT0191934). Eculizumab did not significantly reduce the number of patients with DGF in either of these two pilot studies or in the PROTECT trial (NCT02145182), which included a total of 288 patients [73]. Therefore, the second pilot study was also terminated before planned enrollment. In contrast, pediatric kidney transplant patients who received eculizumab showed better early graft function, less arterial hyalinosis and chronic glomerulopathy in protocol biopsies taken at day 30, and after 1 and 3 years. However, four children in the eculizumab group lost their graft during a flu-like infection, while none of the children in the control group lost their graft (NCT01756508) [74].

Other studies have investigated the efficacy of a C1 inhibitor, which inhibits activation of the classical and lectin pathways, in preventing DGF (Table 1). At the moment, only one phase I/II study using the C1-estase inhibitor (n = 35) compared with placebo (n = 35) has been completed (NCT02134314). Similar to the C5 inhibitor studies, therapy was given directly on the day of transplantation and 24 hours after surgery. Regarding the primary outcome, the occurrence of DGF, defined as the need for dialysis within the first week after transplantation, no difference was observed between the groups (C1-esterase inhibitor 44% vs placebo 60%) [75]. However, the duration of dialysis was significantly shortened in the C1-esterase inhibitor group [75]. In the follow-up of this study, 3.5 years after transplantation, a significantly better eGFR was determined in the C1-esterase inhibitor group (56 mL/min per 1.73 m² vs 35 mL/min per 1.73 m²) [76].

Further studies with C1 esterase inhibitors to reduce the incidence of DGF with higher (NCT04696146, NCT02435732) and more numerous doses (NCT03791476) of the inhibitor are planned or currently have an unclear status (Table 1). Inhibition of the complement cascade at the level of C3 was investigated in another study perfusing the grafts with the CR1 analog mirococept ex vivo instead of the standard cold perfusion fluid (Soltran^®) (ISRCTN49958194) [77]. Primarily, seven study arms with different inhibitor concentrations were planned; however, the study was stopped after planned interim evaluation with the first dose of 10 mg (n = 53) versus placebo (n = 30), because the treatment did not prevent DGF. Instead of continuing the study, a re-dosing study was performed using pig kidneys to determine the saturation range for the inhibitor mirococept. The optimal dose for the pig kidney was determined to be 80 mg of mirococept, which is equivalent to a dose of 120 mg for the human kidney [40]. Ex vivo administration of mirococept at this dose was safe and feasible and provides the basis for future new studies on the treatment of DGF in deceased donor kidney transplants.

Similar to the studies focusing on prevention of DGF, mainly eculizumab and C1-esterase inhibitors have been used in clinical trials for the prevention of ABMR, but they were administered over a longer period of time and more frequently (Table 2). In addition, these studies primarily treated patients who were at particularly high risk of developing ABMR due to an unfavorable cross match or pre-existing donor-specific antibodies. In some cases, however, the study design was so specific that the studies were terminated because patients meeting the inclusion criteria were lacking (NCT01106027; NCT01095887; NCT03221842). A group of 26 highly sensitized recipients of living donor renal transplants, who received eculizumab post-transplant, was compared with a historical control group of 51 sensitized patients treated with a similar plasma exchange protocol without eculizumab to test the incidence of biopsy-proven ABMR in the first 3 months post-transplant. The incidence of ABMR was significantly lower in the eculizumab group compared with the control group (7.7% [2/26] vs 41.2% [21/51]; NCT00670774) [78]. In addition to decreased ABMR, chronic transplant glomerulopathy was detected after 1 year in 6.7% (1/15) of graft biopsies from eculizumab-treated recipients and in 35.7% (15/42) of control patients (p = 0.044) [78]. However, a later phase II, randomized, multicenter, open label, two-arm, parallel group study including 51 patients per arm at first glance showed no protective effect of eculizumab treatment for the primary endpoint of treatment failure rate and was therefore terminated. Only a reassessment of the study including grade I ABMR finally showed a slightly significant difference between the groups, indicating a potential benefit of eculizumab compared with standard operation of care to prevent active ABMR in sensitized recipients [79]. In a single-arm study of recipients who received a deceased donor transplant and had preformed donor-specific antibodies, the primary endpoint (a composite of biopsy‐proven grade II/III ABMR, graft loss, death, or loss to follow‐up, within 9 weeks post-transplant) was observed in 8.8% of eculizumab-treated patients and was thereby lower than expected for standard care (40%) [80].

While the studies described above have investigated whether ABMR after transplantation can be prevented by eculizumab, smaller studies have also investigated whether the course of already developed ABMR can be favorably influenced. Patients were treated with eculizumab immediately after diagnosis of biopsy-proven ABMR, followed by five additional weekly doses (n = 7) and compared with standard therapy (n = 4). Because eculizumab given as monotherapy did not significantly improve eGFR within 3 months, the study was terminated (NCT01895127) and a planned study of complement inhibition for treatment of subclinical ABMR was withdrawn (NCT02113891). However, in the terminated study, the timing of complement inhibitory therapy after transplantation was not determined, but may be important for success, as suggested by a retrospective observational study reporting effective treatment of ABMR in the first month after transplantation [81]. For the treatment of chronic complement-mediated renal injury after kidney transplantation, 11 patients were treated with slightly lower single doses of eculizumab (900 mg) for 6 months. Compared with the control group (n = 5), eGFR tended to be improved, but endothelial cell damage was not reduced (NCT01327573) [82].

In contrast to the studies with the C5 inhibitor eculizumab, of which many were designed to prevent ABMR, there is only one pilot study with a C1 esterase inhibitor (C1-INH). In this study with highly HLA-sensitized patients, none of the ten study participants developed DGF or ABMR in the C1-INH group (n = 10), while in the control group, four out of ten developed DGF and one developed ABMR. However, in further follow-up, two cases of ABMR occurred in each of the two groups [83]. The combination of standard therapy with antibody reduction and C1-INH may be useful for the prevention of ABMR, but further controlled studies are needed. Although a C1-INH trial to treat ABMR was withdrawn in 2012 due to reduced incidence of ABMR by recent improvements in clinical practice (NCT01035593), several new trials have still been initiated. The largest study with a total of 39 participants, 19 of whom received standard therapy for ABMR and 20 of whom also received C1-INH, showed in an interim evaluation no improvement in protection against transplant glomerulopathy and met the pre-specified criteria for futility (NCT02547220). Previously, a pilot study with the same C1-INH had also shown no differences in the primary endpoint, meaning no difference at day 20 after therapy initiation with respect to pathology or graft survival. However, no transplant glomerulopathy and a trend toward improved renal function in the C1-INH group was found at the 6-month biopsy (NCT01147302) [84]. In a first study with sutimlimab, which exclusively inhibits the classical pathway, five of eight C4d-positive recipients turned C4d negative in follow-up biopsies, while another two recipients showed a substantial decrease in C4d scores. There was, however, no change in renal inflammation, gene expression patterns, donor-specific antibody levels, or kidney function [85].

In the treatment of transplant patients, complement inhibitors are only used as monotherapy when given before transplantation (e.g. to avoid DGF). As with any other immunosuppressive therapy, it is a balancing act to, on the one hand, protect the graft from detrimental allogenic immune reactions and consequent graft loss, but, on the other hand, to avoid side effects, including primarily infections that also endanger the transplant and the patient. The complement system plays an important role in the opsonization of pathogens. One threatening side effect of eculizumab therapy is, therefore, the occurrence of meningococcal infections. As a result, appropriate vaccination is a critical prerequisite before treatment. However, vaccination of immunosuppressed patients is challenging and sometimes fails [86]. One study reported graft loss in four eculizumab-treated patients due to flu-like infection [74]. So far, drug-related serious adverse events associated with complement-inhibiting therapies have been observed only rarely, so at least the C1 and C5 inhibitors studied in several clinical trials can be considered safe. This is probably due to the fact that there are different activation pathways of the complement system and that the available complement inhibitors always block only one part of the cascade. Complement-inhibiting therapies aim to suppress the innate immune defense during transplantation and are used in addition to standard therapy including tacrolimus, mycophenolate mofetil and prednisolone, if appropriate, which suppress cell-mediated immune defense. Since complement factors also stimulate immune cells, it may be possible to lower the doses of standard therapy and thereby minimize side effects when combining them with complement inhibitors. Monitoring of complement activation in the plasma may help to determine an optimal and individual treatment of patients, allowing a tailored immune suppression [87].