Monitoring of Donor-specific Anti-HLA Antibodies and Management of Immunosuppression in Kidney Transplant Recipients: An Evidence-based Expert Paper

INTRODUCTION Following renal transplantation (RT), the occurrence of specific anti-HLA antibodies directed against one of the donor antigens (ie, donor-specific antibodies [DSAs])is associated with an increased risk of graft loss. This is one of the criteria, along with histological data derived from renal biopsy, employed for the diagnosis of antibody-mediated rejection (AMR) or humoral rejection (active or chronic). Therefore, these antibodies should be specifically searched for in patients with suspected rejection. However, the true significance of isolated DSAs in the absence of AMR is currently unknown. It has been postulated that this finding could anticipate AMR or indicate under immune suppression.1,2 Given the relevance of DSAs and the variability in their detection and monitoring,3 this expert consensus document has been drafted based on the best scientific evidence available and clinical experience. To this end, an in-depth analysis of both the literature and experience of Spanish RT units was performed beforehand. The article’s objective was to generate simple recommendations that are applicable to patients with DSAs, involving prevention, detection, and monitoring of DSAs, as well as the immunosuppressive treatment options should DSAs be detected. The article’s primary aim is to reduce the variability in clinical practice with respect to medium- and long-term DSA monitoring. Several expert documents had earlier been published about the immunological monitoring of kidney transplant patients.4,5 Although clinically relevant and practical, these reports have not provided sufficient information on the methodology used to attain their conclusions. In spite of its eminently clinical orientation, our report has been based on a critical review of all available literature. Although all types of antibodies can develop, our article is focused on DSAs against HLA antigens that are formed before, termed preformed DSAs, or after RT, termed de novo DSAs. MATERIALS AND METHODS Nominal group, systematic review, and Delphi techniques were used for the preparation of this document. First Nominal Group Meeting The nominal group consisted of a panel of expert nephrologists from the Spanish Group of Transplant Activities (GREAT) responsible for RT. This expert group from 20 transplanting hospitals in Spain agreed to participate in scientific meetings focused on new RT research projects. Beforehand, the document’s coordinators (M.C. and L.G.), in collaboration with the coordinator of the GREAT group (S.Z.), had prepared a discussion outline covering the document’s topics and sections. During the meeting, the panelists, guided by an expert methodologist, defined the document’s objectives, scope, and potential users in a first step, along with its various sections. Subsequently, different features and issues related to DSA monitoring and immune suppression management in RT patients were discussed for each section. This generated a document matrix upon which we continued our research work. We also identified areas wherein controversy may exist and decided to undertake a systematic literature review. Systematic Literature Review For each controversial aspect, a population, intervention, comparator, and outcome question were posed (Table 1), thereby generating the inclusion and exclusion criteria for each question to review. With the help of a Librarian, different search strategies were designed. The bibliographic databases screened from inception to February 2017 were as follows: Medline, Embase, and the Cochrane Library. MeSH and free text terms were used. We searched for articles that included patients with RT. Depending on the question, the studies were required to have analyzed different aspects of DSA development, in addition to the efficacy and safety of immunosuppressive treatments. Only studies with one of the following designs were considered: meta-analysis, systematic review, randomized clinical trial (RCT), and observational, prospective, and retrospective, studies involving >100 patients. For each review, 2 reviewers (Estíbaliz Loza, María Jesús García) independently analyzed the articles retrieved via the aforementioned search strategy in the different bibliographic databases. In the event of discrepancies, a third reviewer (Loreto Carmona) was called upon to resolve the conflict. The Jadad scale for RCTs6 and the Oxford scale for observational studies7 were employed to assess the methodological quality of the studies, with evidence and outcome tables created to summarize their main characteristics and results. Subsequently, a secondary search was performed focused on the bibliography of the included articles. The non-peer-reviewed literature was also reviewed, namely conferences and proceedings of interest over the past 2 y, along with documentation provided by the coordinators in case this was not already captured by the search strategies.TABLE 1.: PICO questions, base for the systematic literature reviewPreliminary Recommendations and Document While the SLR was being performed, the panel worked on the document’s matrix, adding editorial content to different sections. Once the SLR was finalized, a series of preliminary recommendations was generated, with an overall, though preliminary, document drafted. Second Nominal Group Meeting Before the meeting, the SLR results were sent to all panelists. At this meeting, the SLR results were presented and discussed, along with the preliminary recommendations. The overall draft document was then developed, including the definitive recommendations. Delphi The final recommendations were submitted to an online Delphi vote (2 rounds) to test the level of agreement. Participants were invited to vote on each recommendation according to one of the following categories: strongly disagree, basically disagree, basically agree, or fully agree. Agreement (A) was defined if at least 70% of the responses belonged to the full agreement category. Recommendations with A <70% were reevaluated and, if appropriate, reedited and voted on in a second round. It was, additionally, possible to add new recommendations after the first round. Final Document After the Delphi vote, the final document was created. With the assistance of the methodologist, each recommendation was assigned a level of evidence (LE) and grade of recommendation (GR), in line with the evidence-based medicine of the Oxford Centre for Evidence-Based Medicine.7 The A from the final Delphi vote was then added. The document was circulated among the experts for their final assessment and final comments. RESULTS The consensus paper has been divided into 2 main sections: monitoring of DSAs and actions to be taken in the presence of DSAs. Each section has been presented in recommendation format, each with the LE, GR, and A, followed by an explanation or clarification highlighting the available evidence. The Delphi results are shown in Table 2. At the end, 17 recommendations, some with several subsections, were accepted upon the second round.TABLE 2.: Recommendations in tabulated format, with LE and GR based on Oxford Centre of Evidence-based Medicine guidelines, agreement obtained from Delphi, and round in which final agreement was reachedMonitoring of DSAs R1. The patient’s immune risk should be stratified before deciding on the DSA monitoring regimen (LE 2a; GR B; A 70%) As DSA presence is associated with an increased risk of AMR and decreased transplant survival,8–10 it requires several detection, interpretation, and management actions. However, different DSA types confer different levels of risk—from the absence of damage to indolent conditions or active or chronic AMR with graft loss,11–14 therefore, these actions should be diverse and individually adapted to the characteristics of each patient. Allorecognition can generate antibodies to multiple nonself-epitope mismatches even on a single molecule. Traditional HLA matching based on typing by low-resolution antigen or high-resolution allele methods are used for risk stratification to minimize de novo DSA development (reference 10). An alternative approach uses a refined epitope-based matching method in which multiple potential immunogenic sites of the HLA molecule are evaluated. Individual HLA epitopes contribute collectively to the overall immunogenicity of the mismatch, with epitope load or specific immunogenic HLA epitopes driving the formation of de novo DSA.15 It is estimated that 2% to 35% of RT patients develop de novo DSA.10,16–43 The appearance of de novo DSAs over time is variable and dependent on several factors, such as HLA matching as noted, pretransplantation anti-HLA sensitization, adherence to treatment, intraindividual variability of immunosuppressant levels, cutoff of medium fluorescence intensity (MFI) to define DSAs, and others.10,16–27,44 DSAs may occur in the first year post-RT10,21–24 from the third month10 onwards or several years after RT.10,16–43 The overall mean time for their appearance is within 2 to 4 y post-RT, whereas the proportion of RT patients exhibiting DSAs increases with time.43 The most common anti-HLA antibodies are class II,10,43 mainly anti-DQ,22 followed by anti-DR and, very occasionally, DP. In the long-term, Class I and II antibodies can appear combined in rare occasions, whereas isolated DSAs directed against class I are very rare.10,43 The temporal relationship between DSA detection and AMR is poorly described.10,20,21 In some patients, AMR proves to be concurrent with DSA detection,10 but in others, AMR develops over time.10,20 The moment at which DSAs are detected significantly depends on the study design and practice protocols. Since DSAs have been included in the diagnostic criteria for AMR, they are investigated on many occasions at the time of AMR. When AMR occurs following de novo DSA detection, the occurrence time varies from months to several years, although this time period may be influenced by the therapeutic actions taken upon DSA detection. DSAs have been reported to be associated, in the long-term, with active or chronic AMR, the latter either with or without activity, and with graft loss.10 Based on the above, DSA monitoring should be adapted according to the patient’s stratum of immune risk, ranging from rather low to higher immune risk, as should immunosuppression management. Below, we have further outlined the immune risk levels, wherein the main risk factors are anti-HLA antibody presence pre-RT, transplantation, and acute rejection8,11,12,45: Low: No anti-HLA antibodies, no sensitizing events, and first RT.10,30,32,46,47 Medium: Anti-HLA antibodies (panel reactive antibodies, calculated by the most sensitive technique available,<80%) with previous sensitizing events but no DSAs, re-RT, DR-DQ incompatibility, or acute rejection. High: Anti-HLA antibodies (>80%) with DSA levels detected pre-RT.10,12,16–32 An international expert paper was recently published revealing that this classification, although necessary and useful in daily practice, has its limitations and may not be sufficiently sensitive in relation to the low-risk group.5 For this reason, the panel recommends to take this into consideration and be especially careful with patients classified at medium immunological risk. Note that research that is currently being carried out is specifically aimed to improve these models.5,11 R2. The panel recommends requesting at least 1 DSA determination between 3 and 12 mo post-RT in all patients (LE 4; GR D; A 95%) R3. For subsequent determinations In the presence of renal function stability in low-risk patients, the screening study on the Luminex platform should be performed at least every 12 to 24 mo (LE 5; GR D; A 70%). In the presence of renal function stability in medium- and high-risk patients, it is suggested to perform a study of an isolated or single antigen on the Luminex platform every 12 mo (LE 5; GR D; A 75%). Different consensus documents have stratified the risk and issued a series of recommendations on DSA monitoring.4,12 Given that in these articles, the cutoff points, though based on a certain evidence level, were likely to be rather arbitrary, each case must thus be individualized. Other factors, such as the patient’s clinical status and that of being potentially insensitive in low-risk patients, must be taken into account.5 R4. The monitoring periodicity should be adjusted according to the patient’s clinical situation (LE 4; GR D; A 85%) In all cases, the monitoring should be individualized in relation to other clinical circumstances and collected data. R5. Regardless of the risk level, proceeding to DSA assessment has been previously suggested in the following circumstances In cases where in a significant change in medication is considered (dose minimization/suspension/conversion) (LE 4; GR D; A 80%). In cases where in there is a significant variability in the calcineurin inhibitor drug levels (LE 4; GR D; A 70%). If poor adherence to immunosuppressive therapy is suspected (LE 4; GR D; A 75%). If graft dysfunction is detected (increased creatinine or proteinuria) (LE 4; GR D; A 80%). There are a number of clinical situations wherein DSA determination must be performed, regardless of the immune risk level or the timing. It is crucial to take this into account when deciding on a significant change in medication for any reason (eg, suspension), expected to be maintained over time, as well as in cases of relevant drug level variability, especially with tacrolimus. It is paramount to remember that this variability may be due to factors other than therapeutic compliance, such as diarrhea or interaction with other drugs, which must be ruled out first.10 Should renal function deteriorate, DSA must be determined, after first ruling out nonimmunological causes. In the context of a protocol biopsy, the findings may possibly lead to DSA determination, even in the absence of other findings warranting their determination or whether or not the protocol itself requires such a determination. The panel considers that patients should undergo at least 1 antibody determination with solid-phase techniques, preferably using the Luminex platform,48,49 along with studies using donor cells before RT. These would enable specific antibodies to each donor antigen to be studied, thereby producing quick results. This proves to be essential for the therapeutic decision-making process.8 These tests, either qualitative or semiquantitative in nature, could be, screening-, mixed-, or single-antigen type (this latter more expensive and sensitive). Concerning the most recommended technique, the panel suggested that any of them would be appropriate, whereas it was considered preferable (1) to perform before single antigen screening studies in patients with low immunological risk,48,49 and (2) direct single antigen tests in patients with medium and high immunological risk.12,32,36,38–40,50–58 The antibody detected can be against the HLA A, B, C, DRB1, B3, B4, B5, DQB1, DQA1, DPB1, or DPA1 antigens, and the assay proves to be more than the level of each antibody when deciding whether or not DSAs The of and or and antigens in isolated antigen or the of the in some cases, due to the of antigens, in the of the of the criteria used to define the of DSAs is the a level that may on the the or the the It has that this has Although the technique is it is currently not possible to a specific recommendation as to the level that the of DSAs. In this the fluorescence intensity is an to the but its has not of its although higher levels have been to with AMR or histological rejection this strategy several limitations (eg, it is not among and among of the This it currently to the risk according to the or With all this in the panel considers that the assay (ie, the immune or it it is more relevant to define the of Several have the of antibodies into or on whether the is or A of Spanish experts has that most have a cutoff of antibodies detected by Luminex between and To be more in the detection and clinical significance of DSAs, different areas are being of them is the of antibodies and their to damage and of an RT. such as the of DSAs to and the However, this is a very which there is Therefore, a higher LE is to recommendations. A that the of DSAs to on the Luminex platform is associated with graft and a higher of There is controversy as to whether the detection of these or post-RT DSAs has the The detection of is controversial of the available for their and the that a of them is In the event of de novo in a this finding must be (LE GR A 70%) When is detected post-RT, subsequent should be guided by the clinical situation (Table (eg, assess the intraindividual This in of the For in without sensitizing events in the directed against HLA have been as have been due to or factors that with the single antigen to be taken when de novo In the event of adherence must be (LE GR B; A At this the panel considers it to question the of treatment and adherence to preferably If with treatment is an can be to this by the medication in the if it can be and or by the variability of drug detection of post-RT, the panel monitoring of the (LE 5; GR D; A 95%) This is the risk of rejection or other The panel considers that monitoring should be individualized as in previous other and the of or some other detection of post-RT, the panel the as high risk (LE 5; GR D; A This following the recommendations previously for this of risk detection of post-RT, the panel recommends the to the immunosuppression (LE 4; GR D; A 70%) Different strategies can be If the panel a renal (LE 5; GR D; A is the technique used to is in an Given this the panel to that in cases in which a is by de novo DSA even if the proves to be for graft this not out that a rejection may develop in the new may be if of DSAs The panel considers that one of the of RT treatment in relation to DSAs is to their occurrence (LE 5; GR D; A This is the absence of a treatment that can or the of DSAs. strategy is to assess factors associated with DSA These may (1) in relation to (2) due to the of as these the of anti-HLA presence of antibodies other than number of HLA especially and or other sensitizing events or of acute to treatment, intraindividual variability of immunosuppression levels, of calcineurin inhibitor which may lead to calcineurin and the of of without different have the relevance of under immunosuppression as for DSA development and of the It is recommended to or at least the factors that may level variability as as GR B; A is an used in RT. It has been that level variability if variability with a may the risk of DSA and RT is low with high variability to an it to be the and undergo taken a treatment adherence and, with taken a its possible of variability may be the to or the of different Based on the available it has been in some that between may be Therefore, it is recommended that this be performed by an RT under monitoring of other factors that may drug level variability could that with or as well as have been reported for other such as As as possible the absence of clinical or of in with should be to DSA occurrence (LE 2a; GR B; A Several studies have that if is significantly DSAs may with an increased risk of graft In one of these studies, low levels the first year post-RT were associated with an increased risk of de novo DSAs and active humoral rejection at 1 y, as well as an increased risk of graft at Concerning specific levels, levels have been reported to be associated with an increased risk of de novo DSA and graft study that mean levels of and at 1 and 3 y, were associated with graft with or an article the results of 2 analyzed patients were randomized to either with and or with a of by without after The patients in the group without had an increased risk of de novo DSAs and AMR in relation to immunosuppressive For all the aforementioned the panel that (1) it is necessary to be very in the clinical monitoring of the immunosuppression in the RT and to to or other situations that a and could the occurrence of DSAs and, the (2) before or an immunosuppression it essential to this drug levels, and assess the of these if is necessary (eg, of direct or such as to an inhibitor of or with may be The panel recommends adherence to immunosuppressive therapy in kidney transplant for or the of that (LE 5; GR D; A 85%) As previously to immunosuppressive therapy is associated with the development of factors associated with have been such as events or treatment (eg, with a high number of To improve a series of actions have been including the of or strategies for treatment of patients (eg, the of immunosuppression or taken a management in patients should be based on the patient’s clinical (LE 5; GR D; A 85%) The panel considers it essential to each into account different clinical features renal function and results. In patients develop de novo DSA With and renal the panel not immunosuppression (LE 5; GR D; A 75%). With renal function and in the biopsy, the panel immunosuppression (LE 5; GR D; A 80%). the of a renal (eg, or the panel not to the immunosuppression and, in some cases, to even it (LE 5; GR D; A 70%). With acute the panel treatment with or or with or without LE GR A 70%). With or renal function the panel immunosuppressive therapy and (LE 5; GR D; A 80%). The panel following the recommendations that immunosuppressive therapy be in relation to the patient’s for used in RT (LE 5; GR D; A 75%). there is not evidence to define the best therapeutic among different an immunosuppression is available for RT patients, of either or including and or and or or and Concerning patients with a with active AMR and renal the available data were obtained from studies of low to Several have shown that in these patients, with or is likely to DSA levels and improve histological and renal function data. studies have analyzed the of in with other Although some have been it is to the of such in The of is In a previously published it was used in most cases in with other with an overall on DSAs and RT in some However, an including patients with as well as and was This reported that adding not at 1 In both there was a and significant in DSAs. Although at of patients not a significant in renal a the study was in both data in all patients, although to of patients AMR data at not significantly in either group, there were no and 2 were in each trial involving patients with chronic AMR and along with no at 1 y in DSA levels, or drug that has been in this is A trial a in in the group without data on DSA report no with in terms of graft and renal function were with and a that analyzed DSA patients with antibody-mediated rejection was In this at 2 y, no was between and in terms of graft DSA levels, or rejection in However, a number of were reported in the Note that no treatment was in this Concerning a analyzed patients with chronic kidney analysis a change in in patients with DSAs This report with a group that was to be with no change in DSA levels The suggested that this drug may kidney function in these data are also available on In a series of patients with chronic AMR, DSAs, and RT to and treatment, it was that using graft and at y and significantly decreased DSA levels, and renal function at 2 to a There is no evidence that DSAs are associated with a safety for used in RT. Therefore, the recommendations generated for using these drugs, as well as the in drug data To improve clinical practice in relation to RT and the of DSAs, it is essential to recommendations that in the management of these patients in different In years, interest in DSAs has specific in relation to immune rejection or graft is Therefore, the variability in their detection and monitoring this document to generate clinical recommendations, both simple and in daily practice, on the monitoring and management of patients with DSAs. Following an review of the evidence with a low LE for many recommendations, we would to the high level of agreement on all points, thereby our In these recommendations were generated and agreed upon by a group of Spanish RT The were to several The first of these the to the risk in Although there are no accepted criteria and many in this the panel considered this a very useful to the monitoring of However, this the panel also considered that based on the each case should be especially patients with risk. With to DSA determination, it was agreed that patients should have antibody determination by of a Luminex platform Although different cutoff have been to define DSA these are not well Therefore, the panel considered that the assay interpretation, that the immune or it more relevant for the Once de novo DSAs have developed, in this paper a of the to along with their This document has the relevance of from patients with de novo DSAs. It be that their systematic generated some when each case individually in clinical practice, it is possible that we may a case for such a may not be In the expert panel to even though the results may not indicate should not that such patients may be high risk. damage may develop and, therefore, should be in the document the to to DSA The panel agreed to the guidelines, many of which are related to the of certain and therapy This could potentially the development of DSAs. it was considered of for both DSA monitoring and management to each in the clinical situation and characteristics of each patient. immunosuppression management of a with DSAs, the expert panel some for to There is no one single treatment or a of that have more than the among different It is thus essential to each case into account the experience with the different treatments. we have an therapeutic for RT that may be for patients with de novo DSAs, especially in case of acute transplant In this document has a series of recommendations based on scientific evidence and expert We strongly that these recommendations may be relevant and useful for in their possibly other clinical practice for RT. The to the of the Spanish of the Spanish of and the Spanish for in