Impact of Solvent-Accessible HLA Amino Acid Mismatches on Kidney Transplant Outcomes: A Multicenter Longitudinal Study

Impact of Solvent-Accessible HLA Amino Acid Mismatches on Kidney Transplant Outcomes: A Multicenter Longitudinal Study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Impact of Solvent-Accessible HLA Amino Acid Mismatches on Kidney Transplant Outcomes: A Multicenter Longitudinal Study Soufian Meziyerh, Suzanne Bezstarosti, Aleksandar Senev, Dennis van den Broek, and 16 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7115262/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Current alloimmune risk assessment in kidney transplantation relies on HLA antigen mismatches, lacking precision for individualized care. In contrast, solvent-accessible HLA amino acid (saAA) mismatches, empirically defined across all HLA loci, offer a stable and objective alternative. We conducted a multicenter longitudinal study evaluating saAA mismatches, calculated using HLA-EMMA, in relation to key transplant outcomes. Among 2,473 kidney transplant recipients (median follow-up 6.9 years), 36% reached a hierarchical composite endpoint of death-censored graft failure(DCGF), biopsy-proven acute rejection(BPAR), or de novo donor-specific antibody(dnDSA). Multivariable analyses showed total, class I, class II, and locus-specific saAA mismatch scores were significantly associated with the composite outcome, as well as DCGF, BPAR and dnDSA individually (all P<.001). These findings suggest that saAA mismatch load provides a more granular and reproducible measure of alloimmune risk compared to traditional antigen-level mismatches. Incorporating saAA mismatch analysis could enable personalized immunosuppression strategies and optimized transplant monitoring in clinical practice. Health sciences/Biomarkers/Predictive markers Biological sciences/Immunology/Transplant immunology Health sciences/Medical research/Outcomes research Health sciences/Biomarkers/Diagnostic markers Health sciences/Medical research/Epidemiology HLA Mismatch load analysis molecular mismatch epitope death-censored graft survival de novo (dn)DSA biopsy-proven acute rejection (BPAR) Introduction Long-term kidney allograft survival has not matched advancements in short-term outcomes over past decades. 1, 2 A major factor in late graft failure is alloimmune injury, characterized by de novo donor-specific HLA antibodies (dnDSA), antibody-mediated rejection (AMR), and T-cell-mediated rejection (TCMR). 3 Additionally, immunosuppression-related toxicities, including calcineurin inhibitor-related nephrotoxicity, malignancy, infection, cardiovascular, and metabolic complications further impair long-term patient and graft survival. 3, 4 Balancing immunosuppression and adverse events remains challenging in kidney transplantation with significant variability in clinical practice worldwide driven by limited evidence and the absence of reliable alloimmune risk stratification tools. 1, 5, 6 Current, KDIGO guidelines recommend using Human Leukocyte Antigens (HLA)-A, -B, and -DR antigen-level mismatches and uniform graft monitoring every three months, regardless of individual risk. 7 While antigen-level mismatches have shown prognostic value, 8, 9 they lack the granularity needed for personalized immunosuppression or monitoring strategies, such as tailored graft function, dnDSA, and/or donor-derived cell-free DNA (dd-cfDNA) assessments. Alloimmunity targets peptides presented in the HLA binding groove (recognized by T cells) and polymorphic amino acid configurations on HLA molecules, which serve as epitopes for B-cell and antibody recognition. Eplet mismatch analysis identifies clinically relevant epitopes, known as eplets, and has gained traction in alloimmune risk assessment. 10-15 However, eplets are theoretically defined, subject to continuous database updates, and suffer from overlap, limiting broad clinical application. Moreover, prior studies mainly focused on HLA-DR and -DQ loci, restricting comprehensive risk stratification. The HLA Epitope MisMatch Algorithm (HLA-EMMA) quantifies HLA mismatches at the solvent-accessible amino acid (saAA) level, providing a stable and objective alloimmune risk measure based on defined HLA sequences. 16 This first multicenter longitudinal study investigates the impact of HLA saAA mismatches on key kidney transplant outcomes, including a hierarchical composite endpoint of death-censored graft failure (DCGF), biopsy-proven acute rejection (BPAR), dnDSA development, and each outcome individually. Methods Study design and participants We included consecutive KTRs from \centers in Leiden, the Netherlands, and Leuven, Belgium. We collected pseudonymized data on donor and recipient characteristics, including sex, age, donor type (living or deceased), first or repeat transplantation, pre-transplant HLA DSA, cold ischemia time, HLA split-antigen mismatches, and saAA HLA mismatch load. Transplant outcomes recorded were DCGF, BPAR, and dnDSA development. Follow-up adhered to KDIGO guidelines. 17 The Dutch cohort comprised all adult single kidney transplant recipients between January 2005 and December 2019, with follow-up until July 2021 (Leiden University Medical Center IRB:W2020.031). The Belgian cohort included recipients between March 2004 and February 2013, with follow-up until September 2019 (NCT06505200; University Hospitals Leuven Ethics: S64006). All proceedings adhered to the Declaration of Helsinki. Combined or post-other organ transplants were excluded. All transplants had negative complement-dependent cytotoxicity (CDC) crossmatches. Baseline immunosuppression consisted primarily of tacrolimus, mycophenolic acid, and corticosteroids with basiliximab induction in Dutch KTRs and high-risk Belgian KTRs, while some Dutch high-risk KTRs received alemtuzumab. HLA genotyping and amino acid mismatch evaluation For the Dutch cohort, high resolution HLA data were obtained using two complementary approaches. First, we included donor-recipient pairs genotyped at second-field across 11 HLA loci using next-generation sequencing (NGS). DNA samples were processed with NGS kits from GenDx (Utrecht, The Netherlands), and sequenced on the Illumina NGS platform. Second, to generate high-resolution HLA data for the entire cohort, we applied an imputation algorithm developed by the NMDP Bioinformatics team. 18 We included the entire cohort in the imputation algorithm, including those with available second-field data, enabling quality assessment in a later phase. Low- and medium-resolution HLA data for HLA-A, -B, -C, -DRB1, and -DQB1 were imputed to second-field resolution using the most probable haplotypes and imputed genotypes, incorporating broad racial groups to enhance accuracy. This method produced HLA genotypes at second-field Antigen Recognition Domain (ARD) resolution (exons 2 and 3 for class I, and exon 2 for class II alleles) based on published US and European population haplotype frequencies. 19 As a result, we obtained translated second-field resolution data for HLA-A, -B, -C, -DRB1/3/4/5, and -DQB1. To assess imputation accuracy and quality, we compared and tested the agreement between the imputed high-resolution HLA data and actual NGS-derived second-field data (Supplementary Material Table S1). Whenever available, imputed alleles were replaced with actual NGS second-field alleles to minimize reliance on imputed data. HLA-DP imputation was excluded due to the absence of reliable quality assessment metrics. For the Belgian cohort, donor and recipient DNA samples were retrospectively genotyped at second-field resolution for 11 HLA loci using NGS. Half of the donor samples were genotyped using the MIA FORA NGS FLEX 11 HLA Typing Kit (Immucor, Norcross, GA) on the MiSeq sequencing instrument (Illumina, San Diego, CA); while the remaining donor samples and all recipient samples were genotyped at a high-resolution level (exon 2, 3, and 4 for class I and exon 2 and 3 for class II) using the HiSeq sequencing system (Illumina), as described previously. 20 Antigen & solvent-accessible amino acid mismatch evaluation for both cohorts Antigen mismatches (HLA-A, -B, -C, -DR, -DQ) were calculated at split level. saAA mismatches were assessed using high-resolution HLA genotypes (HLA-A,-B,-C,-DR1/3/4/5, and -DQB1) for class I and II loci with HLA-EMMA v1.06, reported per locus, class, and total. 16 saAA mismatch analysis was performed intralocus for HLA class I and II. Outcome definitions The primary endpoint was a hierarchical composite prioritizing DCGF, BPAR, and dnDSA formation. Secondary endpoints included each component individually. Death-censored graft failure (DCGF) DCGF was defined as permanent graft loss requiring dialysis or re-transplantation. Deaths without graft failure were censored. BPAR assessment and treatment of rejection episodes In the Dutch cohort, KTRs did not undergo routine protocol biopsies, except when included in a clinical trial. For-cause biopsies were performed in case of worsening serum creatinine and/or proteinuria without an evident alternative cause, and/or if dnDSA were detected during routine monitoring. All biopsy specimens were scored according to the Banff 2022 criteria. 23 For kidney biopsies performed between 2005-2011, we used the originally reported diagnosis from the histology report. For biopsies performed after 2011, a single pathologist (JK) retrospectively re-evaluated all specimens. Borderline changes were defined as foci of tubulitis (t>0) with minor interstitial inflammation (i1) or moderate to severe interstitial inflammation (i2 or i3) with mild (t1) tubulitis, as previously described. 22 AMR was diagnosed based on the three 2022 Banff criteria for either acute or chronic active AMR, though non-HLA antibodies and gene expression changes were not taken into account. 23 Chronic-active TCMR was not considered as a separate diagnostic category and was instead categorized as borderline rejection or acute TCMR. 22 In the Belgian cohort, KTRs underwent protocol and indication biopsies. All biopsy specimens were rescored according to the Banff 2022 criteria, based on prospectively collected individual lesion scores. 23 dnDSA detection Both cohorts underwent systematic HLA DSA monitoring pre- and post-transplant (Leiden: 6 months, Leuven: 3 months; then annually and at biopsies). First dnDSA detection defined dnDSA onset. All sera were screened using the Lifecodes LifeScreen Deluxe kit (Immucor), and for positive results, HLA antibody specificity was assessed using Lifecodes Single Antigen Bead kits (Immucor). Antibodies against HLA-A, -B, -C, -DRB1, -DRB345, -DQA1, and -DQB1, loci in the recipient sera were analyzed for donor specificity at the allelic level. A background-corrected median fluorescence intensity value of ≥500 for Leuven and ≥1000 for Leiden was used to define the presence of HLA DSA, as previously described. 21,22 Final DSA assignment considered donor-recipient HLA typing. Statistical analysis Analyses were conducted in R (version 1.4.2) and RStudio (Version 1.4.1717). Continuous variables were summarized as medians with interquartile ranges (IQR), means with standard deviations (SD) and 95% confidence intervals (CI), and categorical variables as frequencies and percentages. Multivariable Cox regression evaluated associations between saAA mismatch loads and outcomes, adjusting for donor/recipient sex, age, transplant center, donor type, prior transplants, cold ischemia time, and the presence of pre-transplant DSAs. Censoring occurred at last DSA monitoring, graft failure, or last clinical follow-up. Significance was set at p <0.05 with False Discovery Rate (FDR) controlled at 0.05 using the Benjamini-Hochberg procedure to control for multiple testing. We conducted multiple sensitivity analyses to evaluate the robustness of our findings. First, we limited the analysis to recipients with only NGS based high-resolution HLA typing to assess consistency across primary and secondary endpoints. For secondary endpoint analyses, we limited the assessment to Total, Class I and Class II saAA mismatches to maintain methodological rigor and power, and prevent uninterpretable, excessively broad confidence intervals. Second, we assessed the impact and magnitude of HRs for individual HLA loci by analyzing the median increase per HLA locus rather than applying a fixed increment, accounting for variations in HLA distribution. Third, to address potential residual confounding from pre-transplant DSAs, we repeated the analyses in KTRs without detectable pre-transplant DSAs, despite adjustment in the multivariable model. Fourth, we reanalyzed the association with BPAR after excluding rejections identified via protocol biopsies, focusing exclusively on for-cause biopsies. Fifth, we repeated analyses in a subgroup of KTRs after excluding all HLA-identical family donors. Full results are detailed in the manuscript text and Supplementary Tables. Results Cohort characteristics We included 2473 KTRs with a median follow-up of 6.9 years (IQR: 3.6-9.9) (Table 1). Of these, 937 (38%) received a living donor kidney, more frequently in Leiden than Leuven (57% vs 5%). Pre-formed DSA were present in 7% of recipients. The hierarchical composite endpoint occurred in 883 (35.7%) KTRs. dnDSA developed in 315 KTRs, predominantly targeting class II HLA (n=288; 91%), with HLA-DQ antibodies being the most frequent (n=188). BPAR occurred in 555 (22%) KTRs, DCGF in 335 (14%), and 549 (22%) patients died. High resolution HLA imputation and HLA-saAA mismatch scores Among 49.460 alleles (HLA-A, -B, -C, -DR, and -DQ), 22.907 (46%) were NGS-typed and 54% imputed. Quality assessment of the translation to high-resolution revealed accurate imputative results for all loci; HLA-A, -B, -C, -DRB1/3/4/5, and -DQB1 (Supplementary Material Table S1). Mean split-antigen mismatches were 1.0±0.7, 1.2±0.7, 1.1±0.7, 0.9±0.6, 0.8±0.6 for HLA-A, -B, -C, -DR, and -DQ, respectively (Table 1). Median saAA mismatches were 16 for Class I and 18 for Class II, with a strong correlation between total antigen and saAA mismatches (ρ=0.711,P<0.001) (Supplementary Figure S1). Median saAA mismatch scores and transplant outcomes KTRs experiencing the composide endpoint, BPAR, or dnDSA had significantly higher median total saAA mismatches compared to those without (P<0.001 for all; Supplementary Table S2). Notably, class II saAA mismatches were significantly higher in KTRs that experienced graft loss, BPAR and developed dnDSA. Cox regression analyses for saAA mismatch scores and kidney graft outcomes Hierarchical composite endpoint Total, class I, class II, and all locus-specific saAA mismatches were all significantly associated with the primary endpoint. For every ten saAA mismatches, HRs were 1.13 (95% CI: 1.09-1.16) for total, 1.15 (95% CI: 1.08-1.23) for class I, and 1.14 (95% CI: 1.10-1.19) for class II saAA (Table 2). Moreover, both class I and class II saAA mismatches were independent risk factors for the composite endpoint, with HRs of 1.12 (95% CI: 1.05-1.19), and 1.13 (95% CI: 1.09-1.18), respectively (Supplementary Table S3). DCGF Total, class I, class II, and all locus-specific saAA mismatch scores were significantly associated with DCGF (Table 3). For every ten saAA mismatches, HRs were 1.13 (95% CI: 1.07-1.19), 1.21 (95% CI: 1.09-1.33), and 1.11 (95% CI: 1.04-1.18) for total, class I, and class II, respectively. Locus-specific aHRs ranged from 1.16 (95% CI: 1.06-1.26) for HLA-DQ to 1.65 (95% CI: 1.27-2.14) for HLA-C (Table 3). Both class I and class II saAA mismatches were independent risk factors for DCGF, with HRs of 1.20 (95% CI: 1.08-1.32, p<.001) and 1.10 (95% CI: 1.04-1.17, p=0.002), respectively (Supplementary Table S4). BPAR All saAA mismatches were significantly associated with BPAR in multivariable analyses (Table 4). For every ten HLA saAA mismatches, HR was 1.12 (95% CI: 1.07-1.16), 1.18 (95% CI: 1.08-1.28), and 1.12 (95% CI: 1.06-1.17) for total, class I, and class II respectively. For the locus-specific saAA mismatches, HR ranged from 1.14 (95% CI: 1.06-1.22) for HLA-DQ to 1.58 (95% CI: 1.28-1.96) for HLA-B (Table 4). Both class I and class II saAA mismatches were independent risk factors for BPAR with HRs of 1.15 (95% CI: 1.06-1.25, p=0.001) and 1.10 (95% CI: 1.05-1.16, p<.001), respectively (Supplementary Table S5). dnDSA Total, class I, class II, and locus-specific saAA mismatches were significantly associated with dnDSA (Table 5a-5b). For every ten saAA mismatches, HR was 1.22 (95% CI: 1.16-1.29), 1.21 (95% CI: 1.10-1.33), and 1.30 (95% CI: 1.21-1.40) for total, class I, and class II respectively. For the locus-specific HLA, aR ranged from 1.26 (95% CI: 1.10-1.33) for HLA-A to 1.67 (95% CI: 1.31-2.13) for HLA-B. When investigating the impact of locus-specific saAA mismatches on locus-specific dnDSA, we observed comparable results with HR ranging from 1.90 (95% CI: 1.64-2.21) for HLA-DQ to 4.06 (95% CI: 2.08-7.96) for HLA-C (Table 5b). In multivariable models including antigen mismatches, saAA mismatches remained independently associated with locus-specific dnDSAs (Table 6). Thus, saAA mismatches impacted graft outcomes beyond antigenic mismatches. We identified both class I and class II saAA as independent risk factors for dnDSA, with HRs of 1.32 (95% CI: 1.20-1.45) and 1.18 (95% CI: 1.11-1.26), respectively (Supplementary Table S6). Furthermore, we also analyzed the impact of saAA mismatches within the context of one or two HLA antigen-level mismatches. In the presence of a single HLA antigen-level mismatch, locus-specific saAA mismatches across all HLA-loci significantly impacted locus-specific DSA formation. With two HLA antigen-level mismatches, locus-specific saAA mismatches had a significant effect on DR- and A-DSA, whereas effects on other HLA loci did not reach statistical significance – likely due to the lower number of events and reduced statistical power (Supplementary Material Table S7). Sensitivity analyses All findings remained consistent across all sensitivity analyses performed. NGS-only cohort Significant associations were observed between total, class I, class II, and locus-specific saAA mismatches and the hierarchical composite endpoint. HRs ranged from 1.09 to 1.60, with P-values between <0.001 and 0.029 for every ten saAA mismatches (Supplementary Tables S8-11). Median increase per locus As expected, results were comparable across loci. HRs for total, class I, and class II saAA mismatch scores were highest, reflecting their aggregation of all HLA loci (Supplementary Tables S12-16). As only the distribution of the saAA mismatch score was changed for this analysis, we did not report P-values as they are identical to the primary analysis. In patients without pre-transplant donor-specific antibodies The associations found between total, class I, class II, locus specific HLA-A, -B, -C, -DR, and -DQ and primary and secondary outcomes remained unchanged (Supplementary Tables S17-21). Exclusion of BPARs diagnosed in protocol biopsies BPAR associations remained unaffected (Supplementary Table S22). Exclusion of HLA-identical familial kidney donors. Results were stable, except saAA mismatch scores for HLA-A lost significance for BPAR (Supplementary Tables S23–27). Discussion In this longitudinal multicenter study, we demonstrate the impact of saAA mismatches, calculated using HLA-EMMA, on hierarchical kidney transplant outcomes. Leveraging two well-characterized, unique cohorts with extensive follow-up and a substantial number of events, we showed significant impact of saAA mismatches for class I and class II HLA on dnDSA (overall and locus-specific), BPAR, and DCGF beyond effects of traditional antigen-level HLA mismatches. Notable similarities and differences were observed between the cohorts from the Netherlands and Belgium. Both cohorts had comparable age and sex distributions among KTRs. However, donors in Leiden were older (53 vs. 48 years), with a higher proportion of living donors (57% vs 5% in Leuven), and with a higher proportion of kidneys donated from donors after circulatory death (58% vs. 17%). Fewer patients in Leiden had detectable pre-transplant DSAs (5% vs. 11%). The higher prevalence of living donor transplants in Leiden was associated with higher saAA mismatch scores across all HLA loci, reflecting reduced emphasis on matching in living donors. In addition, donor-recipient pairs in deceased donor transplantation (DBD and DCD) exhibited higher total saAA mismatch scores in Leiden, suggesting a higher immunological risk in this cohort. This may partly explain the higher incidence of de novo donor-specific antibodies (dnDSA) observed in Leiden compared to Leuven. Differences in DSA detection methods may also contribute to these findings. In Leiden, pre-transplant DSA were assessed using evolving techniques over time, including complement-dependent cytotoxicity (CDC) screening and crossmatches, flow cytometry crossmatches, and single antigen bead assays, with the latter applied to high-risk transplants after 2016. In contrast, the Leuven cohort utilized single antigen bead assays on biobanked sera for consistent pre-transplant DSA detection. These methodological differences could account for discrepancies in the reported prevalence of pre- and post-transplant DSAs. To account for center-specific differences, transplant center was included as a covariate in multivariable analyses. Interestingly, the incidence of BPAR was higher in Leuven, likely due to the use of protocol biopsies capturing subclinical rejection, which are not part of routine care in Leiden. Rates of DCGF between both cohorts were comparable at 2- and 5-years post-transplant, despite a higher percentage of living donor transplants in Leiden. These findings highlight the multifactorial nature of graft outcomes influenced by donor and recipient characteristics, the prevalence of pre-transplant DSAs, immunological risk, and variations in immunosuppressive therapy and drug exposure over time. Importantly, they emphasize that histocompatibility is only one determinant of alloimmune risk, which can be mitigated or modulated through tailored immunosuppressive strategies. 26, 27 In this study, we included saAA mismatch scores across all HLA-loci, excluding HLA-DP, to comprehensively assess all polymorphic residues potentially recognizable by B-cell receptors and DSA. We found that both class I and class II HLA are associated with dnDSA, BPAR, and DCGF. Notably, we observed a previously unreported hazard of locus-specific saAA mismatches for HLA class I (HLA-A, -B, -C) on locus-specific dnDSA. As expected, DQ-specific DSAs were most prevalent in our cohort, followed by DSAs targeting HLA-DR, -A, -B, and -C. These findings underscore the immunological relevance of both class I and class II HLA in driving alloimmune responses including development of dnDSA and incidence of BPAR, which in turn influence long-term graft survival. Consequently, our results challenge the traditional emphasis on HLA-A, -B, -DR antigen-level mismatches and advocate for a more refined, granular approach, which also considers saAA mismatches across other HLA-loci. This broader perspective may refine risk stratification and provide deeper insights into the complexity of alloimmunity. Supporting our findings, a prior study has demonstrated that some patients classified as low immunological risk based on traditional HLA-antigen mismatch may harbor a high epitope mismatch load, placing them at elevated immunological risk despite initial assessments using traditional antigenic mismatches. 11 Over the past decade, multiple studies have demonstrated that HLA-DR and HLA-DQ eplets outperform antigen-level HLA mismatches in predicting the risk of dnDSA, AMR, and TCMR, and that they modulate the association between tacrolimus exposure with adverse events from nonadherence. 11, 12, 26-32 These studies also highlight the superior predictive value of eplets over antigen mismatches for dnDSA formation. 12, 22, 31 Prior research primarily focused on HLA-DR and -DQ eplet mismatches due to the high prevalence DSAs against these loci. Yet, alloimmune responses are not restricted to these antigens alone as shown in this study. Additionally, earlier studies relied on theoretical eplet mismatches defined by HLA-Matchmaker, which is subject to continuous updates as eplet definitions changes. Notably, only a minority of eplets have gone through antibody-verification, making it difficult to assess their true immunogenic potential. 33, 34 These evolving definitions complicate reproducibility and limit clinical utility of eplet-based risk stratification for personalizing care in KTRs. 34 In contrast, HLA-EMMA provides an objective, standardized method for calculating saAA mismatches across all HLA-loci. 16, 35 This approach improves reproducibility and enhances clinical utility by predicting the relative solvent accessibility of HLA structures. Moreover, HLA-EMMA provides a user-friendly platform with ease of digital application. 36 However, it is important to acknowledge its limitations - while HLA-EMMA may have higher sensitivity, it could also include residues that are not clinically relevant, thereby negatively impacting specificity. Conversely, antibody-verified eplet analyses, while more specific, may exclude immunologically relevant eplets that have not yet been experimentally verified. Similarly, assessing overall eplet mismatches poses the same limitation as evaluating total saAA mismatches. 16 We intentionally refrained from using saAA mismatch scores to categorize patients into low-, medium-, or high-risk groups for alloimmunity. However, from a clinical perspective, a sensitive and quantitative marker remains valuable. A refined risk stratification tool could facilitate individualized tapering of immunosuppression and optimize screening protocols, particularly for patients with a low saAA mismatch score who may be at reduced immunological risk. The (kidney) transplant field urgently requires improved risk stratification tools for alloimmunity, as emphasized by Hariharan et al. and others. 1, 37 Currently, many centers reduce immunosuppression reactively in response to cumulative toxicity, often considering presumed immunosenescence and diminished immunogenicity over time. However, proactive immunosuppression minimization is rarely practiced, typically being reserved for HLA-identical living-related transplants. 5, 6 We propose that enhanced alloimmune risk stratification could significantly improve the personalization of both surveillance and maintenance immunosuppressive regimens, a concept supported by previous studies. 1, 26, 38 Patients with higher immunological risk (e.g., higher EMMA scores or younger age) might benefit from more frequent monitoring and sustained immunosuppression, while those at lower risk could achieve favorable outcomes with reduced immunosuppression. 38-40 This approach aligns with findings from recent studies utilizing HLA-DR/-DQ eplet mismatch analyses, which demonstrated that tailored immunosuppression and less intensive monitoring strategies could be implemented effectively based on specific thresholds. 11, 31, 39 These studies collectively advocate for a patient-centered approach based on molecular matching scores rather than adherence to center-specific protocols. 41 However, more prospective studies and randomized controlled trials (RCTs) will be essential for establishing clinical utility and implement or change KDIGO guidelines. This study has several limitations. First, while our findings suggest that alloimmune risk increases linearly with the number of saAA mismatches, this does not imply that all mismatches are equally immunogenic. In reality, the pathogenicity of specific amino acid polymorphisms varies, and HLA-EMMA does not fully account for these differences. 42-44 Additionally, HLA-EMMA does not integrate other critical factors influencing alloimmune risk, such as immunosenescence or intrapatient variability in immunosuppression. 45 Second, high-resolution HLA typing was unavailable for 54% of alleles, requiring imputation. However, internal validation demonstrated high concordance between imputed and actual high-resolution typing (Supplementary Material Table S1), with accuracy ranging from 86% to 92% for class I HLA. Given our novel findings on class I molecular mismatch load and kidney graft outcomes, this level of accuracy is a key strength. To further assess robustness, we performed a sensitivity analysis restricted to patients with only NGS-based typing, which yielded consistent findings across HLA-A, -B, -C, -DR, and -DQ, minimizing the risk of false-positives due to imputation (Supplementary Material Tables S24-27). Third, we lacked DQA1 data, which may have influenced the observed incidence of DQ-specific antibodies. This could have either inflated the apparent frequency of DQ antibodies or underestimated the true immunological burden. Nonetheless, given that our finding on HLA-DQ is supported by prior studies, 26-32, 39, 46 we believe this limitation has minimal impact. Future studies assessing clinical utility of HLA-EMMA, should definitely incorporate comprehensive DQA1 typing. Finally, we were unable to assess the potential impact of HLA-DP on kidney graft outcomes. This is noteworthy as a previous study demonstrated that HLA-DP DSA, while rare, represent a significant risk for AMR. 47 In conclusion, saAA mismatch burden in both class I and II HLA is associated with increased risk of dnDSA, BPAR, and graft loss. saAA mismatch analysis provides a more precise measure of alloimmune risk than traditional antigen-level mismatches and merits further prospective evaluation. This refined approach could improve alloimmune risk stratification, optimize surveillance strategies, and guide personalized immunosuppression to balance efficacy and toxicity in kidney transplant recipients. Declarations Disclosure statement In the last 3 years TvG has received lecture fees and study grants from Chiesi and Astellas, in addition to consulting fees from Roche Diagnostics, Thermo Fisher, Vitaeris, CSL Behring, Astellas and Aurinia Pharma. AdV received lecture and consulting fees from Sandoz, Chiesi, Astellas, Hansa, CSL Behring, Neovii, AstraZeneca, Sanofi, Takeda, and Novartis in the past years, all of which went to his employer LUMC. SM received lecture fees from Chiesi. JK received lecture fees from Chiesi, is consultant for Aiosyn BV, Hansa, Alentis Pharmaceuticals AG and Novartis AG. In all cases, reimbursements have been transferred to employer accounts, and none have been paid to personal bank accounts. The other authors of this manuscript have no conflicts of interest to disclose as described by Nature Medicine. Acknowledgements National Marrow Donor Program in the United States The contributions by Leuven were supported by Fonds Wetenschappelijk Onderzoek (FWO; Research Foundation – Flanders) and the Agentschap Innoveren en Ondernemen (VLAIO; Agency for Innovation and Entrepreneurship) by the TBM project grant IWT.150199, FWO research project grants G038024N and G087620N), and by a KU Leuven internal funding grant C2M/24/057. MC is a postdoctoral researcher of FWO (12D6423N), and MN is a senior clinical investigator of FWO (1842919N). Authors’ Contributions SM, SB, AS, MC, JK, DvdH, PvdB, HdF, GH, MN, SH, DR, AdV conceived and designed the work and played a key role in interpreting the results. All authors revised the manuscript and approved with the definitive version. Supplementary Materials See separate file References Hariharan S, Israni AK, Danovitch G. Long-Term Survival after Kidney Transplantation. New England Journal of Medicine 2021; 385: 729-743. Loupy A, Lefaucheur C. Antibody-Mediated Rejection of Solid-Organ Allografts. The New England journal of medicine 2018; 379: 1150-1160. Mayrdorfer M, Liefeldt L, Wu K , et al. Exploring the Complexity of Death-Censored Kidney Allograft Failure. Journal of the American Society of Nephrology 2021; 32: 1513-1526. Mayrdorfer M, Liefeldt L, Osmanodja B , et al. A single centre in-depth analysis of death with a functioning kidney graft and reasons for overall graft failure. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association 2023; 38: 1857-1866. Perez-Saez MJ, Montero N, Oliveras L , et al. Immunosuppression of HLA identical living-donor kidney transplant recipients: A systematic review. Transplantation reviews (Orlando, Fla) 2023; 37: 100787. Wojciechowski D, Wiseman A. Long-Term Immunosuppression Management: Opportunities and Uncertainties. Clinical journal of the American Society of Nephrology : CJASN 2021; 16: 1264-1271. Chadban SJ, Ahn C, Axelrod DA , et al. KDIGO Clinical Practice Guideline on the Evaluation and Management of Candidates for Kidney Transplantation. Transplantation 2020; 104: S11-s103. Susal C. Collaborative Transplant Study Website. 2022. Heidt S, Haasnoot GW, van Rood JJ , et al. Kidney allocation based on proven acceptable antigens results in superior graft survival in highly sensitized patients. Kidney international 2018; 93: 491-500. Tambur AR, Das R. Can We Use Eplets (or Molecular) Mismatch Load Analysis to Improve Organ Allocation? The Hope and the Hype. Transplantation 2023; 107: 605-615. Snanoudj R, Kamar N, Cassuto E , et al. Epitope load identifies kidney transplant recipients at risk of allosensitization following minimization of immunosuppression. Kidney international 2019; 95: 1471-1485. Wiebe C, Kosmoliaptsis V, Pochinco D , et al. A Comparison of HLA Molecular Mismatch Methods to Determine HLA Immunogenicity. Transplantation 2018; 102: 1338-1343. Cai J, Terasaki PI, Mao Q , et al. Development of nondonor-specific HLA-DR antibodies in allograft recipients is associated with shared epitopes with mismatched donor DR antigens. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2006; 6: 2947-2954. Tambur AR, Rosati J, Roitberg S , et al. Epitope analysis of HLA-DQ antigens: what does the antibody see? Transplantation 2014; 98: 157-166. Kramer CSM, Israeli M, Mulder A , et al. The long and winding road towards epitope matching in clinical transplantation. Transplant international : official journal of the European Society for Organ Transplantation 2019; 32: 16-24. Kramer CSM, Koster J, Haasnoot GW , et al. HLA-EMMA: A user-friendly tool to analyse HLA class I and class II compatibility on the amino acid level. Hla 2020; 96: 43-51. Kasiske BL, Zeier MG, Chapman JR , et al. KDIGO clinical practice guideline for the care of kidney transplant recipients: a summary. Kidney international 2010; 77: 299-311. Madbouly A, Gragert L, Freeman J , et al. Validation of statistical imputation of allele-level multilocus phased genotypes from ambiguous HLA assignments. Tissue Antigens 2014; 84: 285-292. Gragert L, Madbouly A, Freeman J , et al. Six-locus high resolution HLA haplotype frequencies derived from mixed-resolution DNA typing for the entire US donor registry. Human immunology 2013; 74: 1313-1320. Senev A, Lerut E, Van Sandt V , et al. Specificity, strength, and evolution of pretransplant donor-specific HLA antibodies determine outcome after kidney transplantation. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2019; 19: 3100-3113. Wisse BW, Kamburova EG, Joosten I , et al. Toward a Sensible Single-antigen Bead Cutoff Based on Kidney Graft Survival. Transplantation 2019; 103: 789-797. Senev A, Coemans M, Lerut E , et al. Eplet Mismatch Load and De Novo Occurrence of Donor-Specific Anti-HLA Antibodies, Rejection, and Graft Failure after Kidney Transplantation: An Observational Cohort Study. J Am Soc Nephrol 2020; 31: 2193-2204. Naesens M, Roufosse C, Haas M , et al. The Banff 2022 Kidney Meeting Report: Reappraisal of microvascular inflammation and the role of biopsy-based transcript diagnostics. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2024; 24: 338-349. Koshy P, Furian L, Nickerson P , et al. European Survey on Clinical Practice of Detecting and Treating T-Cell Mediated Kidney Transplant Rejection. Transplant international : official journal of the European Society for Organ Transplantation 2024; 37: 12283. van den Broek DAJ, Meziyerh S, Budde K , et al. The Clinical Utility of Post-Transplant Monitoring of Donor-Specific Antibodies in Stable Renal Transplant Recipients: A Consensus Report With Guideline Statements for Clinical Practice. Transplant international : official journal of the European Society for Organ Transplantation 2023; 36: 11321. Wiebe C, Rush DN, Nevins TE , et al. Class II Eplet Mismatch Modulates Tacrolimus Trough Levels Required to Prevent Donor-Specific Antibody Development. J Am Soc Nephrol 2017; 28: 3353-3362. Davis S, Wiebe C, Campbell K , et al. Adequate tacrolimus exposure modulates the impact of HLA class II molecular mismatch: a validation study in an American cohort. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2021; 21: 322-328. Wiebe C, Pochinco D, Blydt-Hansen TD , et al. Class II HLA epitope matching-A strategy to minimize de novo donor-specific antibody development and improve outcomes. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2013; 13: 3114-3122. Wiebe C, Gibson IW, Blydt-Hansen TD , et al. Rates and determinants of progression to graft failure in kidney allograft recipients with de novo donor-specific antibody. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2015; 15: 2921-2930. Wiebe C, Nickerson P. Strategic Use of Epitope Matching to Improve Outcomes. Transplantation 2016; 100: 2048-2052. Wiebe C, Kosmoliaptsis V, Pochinco D , et al. HLA-DR/DQ molecular mismatch: A prognostic biomarker for primary alloimmunity. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2019; 19: 1708-1719. Wiebe C, Nickerson PW. Human leukocyte antigen molecular mismatch to risk stratify kidney transplant recipients. Current opinion in organ transplantation 2020; 25: 8-14. Bezstarosti S, Bakker KH, Kramer CSM , et al. A Comprehensive Evaluation of the Antibody-Verified Status of Eplets Listed in the HLA Epitope Registry. Frontiers in immunology 2021; 12: 800946. Tambur AR, Das R. Can We Use Eplets (or Molecular) Mismatch Load Analysis to Improve Organ Allocation? The Hope and the Hype. Transplantation 2022. Duquesnoy RJ, Askar M. HLAMatchmaker: a molecularly based algorithm for histocompatibility determination. V. Eplet matching for HLA-DR, HLA-DQ, and HLA-DP. Human immunology 2007; 68: 12-25. Ladowski JM, Mullins H, Romine M , et al. Eplet mismatch scores and de novo donor-specific antibody development in simultaneous pancreas-kidney transplantation. Human immunology 2021; 82: 139-146. Ettenger R, Albrecht R, Alloway R , et al. Meeting report: FDA public meeting on patient-focused drug development and medication adherence in solid organ transplant patients. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2018; 18: 564-573. Hricik DE, Formica RN, Nickerson P , et al. Adverse Outcomes of Tacrolimus Withdrawal in Immune-Quiescent Kidney Transplant Recipients. J Am Soc Nephrol 2015; 26: 3114-3122. Wiebe C, Balshaw R, Gibson IW , et al. A rational approach to guide cost-effective de novo donor-specific antibody surveillance with tacrolimus immunosuppression. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2023; 23: 1882-1892. Matas AJ, Gaston RS. Moving Beyond Minimization Trials in Kidney Transplantation. J Am Soc Nephrol 2015; 26: 2898-2901. Axelrod DA, Naik AS, Schnitzler MA , et al. National Variation in Use of Immunosuppression for Kidney Transplantation: A Call for Evidence-Based Regimen Selection. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2016; 16: 2453-2462. Kosmoliaptsis V, Mallon DH, Chen Y , et al. Alloantibody Responses After Renal Transplant Failure Can Be Better Predicted by Donor-Recipient HLA Amino Acid Sequence and Physicochemical Disparities Than Conventional HLA Matching. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2016; 16: 2139-2147. Sapir-Pichhadze R, Zhang X, Ferradji A , et al. Epitopes as characterized by antibody-verified eplet mismatches determine risk of kidney transplant loss. Kidney international 2020; 97: 778-785. Mohammadhassanzadeh H, Oualkacha K, Zhang W , et al. On Path to Informing Hierarchy of Eplet Mismatches as Determinants of Kidney Transplant Loss. Kidney international reports 2021; 6: 1567-1579. van Gelder T. Within-patient variability in immunosuppressive drug exposure as a predictor for poor outcome after transplantation. Kidney international 2014; 85: 1267-1268. Wiebe C, Nevins TE, Robiner WN , et al. The Synergistic Effect of Class II HLA Epitope-Mismatch and Nonadherence on Acute Rejection and Graft Survival. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2015; 15: 2197-2202. Daniëls L, Claas FHJ, Kramer CSM , et al. The role of HLA-DP mismatches and donor specific HLA-DP antibodies in kidney transplantation: a case series. Transplant immunology 2021; 65: 101287. Tables Tables 1 to 6 are available in the Supplementary Files section. Additional Declarations There is NO Competing Interest. Supplementary Files 2025defEMMASupplementalMaterial.docx Supplemental Material belonging to the manuscript table1to6.docx Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7115262","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":485707858,"identity":"1586e8be-d4a9-44ca-8e4e-a1992002780d","order_by":0,"name":"Soufian Meziyerh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYNACAyCWYD54AMxhJ14LWwJECzPRNknwGBCnRb6BO/Hhj4LDcuazez4c5s2xYzAnpMXgAO9mAwmDw8Yyd85uOMy7LZnBspmQFgbebRIGBocTZ0jkgrQcYDA4TNBhvNt/JBgcrp8hkfOAOC0MB3i3MRwwOJwgIZHDQJwWg8O8myUbDNINZ0ikGRycuy2Zh7DD2ns3fvzxx1peQiL54YO32+zkDI43ENADCVJEKPEQ8gkM1BGrcBSMglEwCkYiAAB8Z0D6Y2RILwAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-7694-8381","institution":"Leiden University Medical Center","correspondingAuthor":true,"prefix":"","firstName":"Soufian","middleName":"","lastName":"Meziyerh","suffix":""},{"id":485707859,"identity":"38f31259-107d-4d9c-9f86-ed7c79b88f96","order_by":1,"name":"Suzanne Bezstarosti","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Suzanne","middleName":"","lastName":"Bezstarosti","suffix":""},{"id":485707860,"identity":"3666b65c-59cb-451b-b56e-73cfdb0a152d","order_by":2,"name":"Aleksandar Senev","email":"","orcid":"","institution":"Northwestern University","correspondingAuthor":false,"prefix":"","firstName":"Aleksandar","middleName":"","lastName":"Senev","suffix":""},{"id":485707861,"identity":"6b474fff-8a52-4ab0-b5bb-4593c9dcc11b","order_by":3,"name":"Dennis van den Broek","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Dennis","middleName":"van den","lastName":"Broek","suffix":""},{"id":485707862,"identity":"62bde156-30a8-466d-b8c3-773b982f51ed","order_by":4,"name":"Maarten Coemans","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Maarten","middleName":"","lastName":"Coemans","suffix":""},{"id":485707863,"identity":"fd987534-5319-4af6-a168-b08c0e84917c","order_by":5,"name":"Jesper Kers","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Jesper","middleName":"","lastName":"Kers","suffix":""},{"id":485707864,"identity":"5d07ef1b-c34b-46b1-8d10-92dd6e6ce32e","order_by":6,"name":"Teun van Gelder","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Teun","middleName":"van","lastName":"Gelder","suffix":""},{"id":485707865,"identity":"2bd36c60-4d8e-4b6b-9fa5-887c33b8bee6","order_by":7,"name":"Danny van der Helm","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Danny","middleName":"van der","lastName":"Helm","suffix":""},{"id":485707866,"identity":"0544e529-a3ca-43c0-857e-203a2b829e4f","order_by":8,"name":"Paul van der Boog","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Paul","middleName":"van der","lastName":"Boog","suffix":""},{"id":485707867,"identity":"0e3b1914-d493-4ddf-96d0-97151523dae2","order_by":9,"name":"Johan de Fijter","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Johan","middleName":"","lastName":"de Fijter","suffix":""},{"id":485707868,"identity":"2db5dc57-c3cf-4eb7-96b5-0c046cf64088","order_by":10,"name":"Abeer Madbouly","email":"","orcid":"","institution":"National Marrow Donor Program","correspondingAuthor":false,"prefix":"","firstName":"Abeer","middleName":"","lastName":"Madbouly","suffix":""},{"id":485707869,"identity":"2679e569-4831-4272-b714-000638dc83ae","order_by":11,"name":"Dirk Jan Moes","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Dirk","middleName":"Jan","lastName":"Moes","suffix":""},{"id":485707870,"identity":"174ca1a0-1367-480a-aed1-320659fd9d67","order_by":12,"name":"Cees van Kooten","email":"","orcid":"","institution":"LUMC","correspondingAuthor":false,"prefix":"","firstName":"Cees","middleName":"van","lastName":"Kooten","suffix":""},{"id":485707871,"identity":"41877292-e1e8-4bdb-a5f8-7bba0c45ed93","order_by":13,"name":"Gonca Karahan","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Gonca","middleName":"","lastName":"Karahan","suffix":""},{"id":485707872,"identity":"53bfacb1-31ca-4e0f-b5f5-6bc942e321ca","order_by":14,"name":"Geert Haasnoot","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Geert","middleName":"","lastName":"Haasnoot","suffix":""},{"id":485707873,"identity":"0cec1239-c628-4e79-9d71-0eb4ac565816","order_by":15,"name":"Marie-Paule Emonds","email":"","orcid":"https://orcid.org/0000-0002-2653-8656","institution":"Belgian Red Cross-Flanders","correspondingAuthor":false,"prefix":"","firstName":"Marie-Paule","middleName":"","lastName":"Emonds","suffix":""},{"id":485707874,"identity":"1bc0c010-f625-406a-b84c-5b48f871cf20","order_by":16,"name":"Maarten Naesens","email":"","orcid":"https://orcid.org/0000-0002-5625-0792","institution":"Department of Nephrology and Renal Transplantation, University Hospitals Leuven, Leuven, Belgium","correspondingAuthor":false,"prefix":"","firstName":"Maarten","middleName":"","lastName":"Naesens","suffix":""},{"id":485707875,"identity":"67314fcd-0fbd-4234-a39e-272abd8d2851","order_by":17,"name":"Sebastiaan Heidt","email":"","orcid":"https://orcid.org/0000-0002-6700-188X","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Sebastiaan","middleName":"","lastName":"Heidt","suffix":""},{"id":485707876,"identity":"f05524c7-2969-4a8e-b6d2-27dec3f58fb7","order_by":18,"name":"Dave Roelen","email":"","orcid":"","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Dave","middleName":"","lastName":"Roelen","suffix":""},{"id":485707877,"identity":"0b337a73-f761-4d2e-b19c-c0d88b01c670","order_by":19,"name":"Aiko de Vries","email":"","orcid":"https://orcid.org/0000-0002-9284-3595","institution":"Leiden University Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Aiko","middleName":"","lastName":"de Vries","suffix":""}],"badges":[],"createdAt":"2025-07-13 19:55:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7115262/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7115262/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87180845,"identity":"edac3de6-fb1f-41f7-b49e-3923789e5163","added_by":"auto","created_at":"2025-07-21 09:43:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1016160,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7115262/v1/b6143238-f0a1-4105-8daa-7b6ff7f0e867.pdf"},{"id":87179994,"identity":"a1c85191-ef39-45cd-900c-c17c3f26b4ed","added_by":"auto","created_at":"2025-07-21 09:35:36","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":185010,"visible":true,"origin":"","legend":"Supplemental Material belonging to the manuscript","description":"","filename":"2025defEMMASupplementalMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-7115262/v1/05135c770dd8c4b3a37da4ff.docx"},{"id":87179993,"identity":"308c35a2-2a4b-41a5-8bd0-2d0045c209cb","added_by":"auto","created_at":"2025-07-21 09:35:36","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":46919,"visible":true,"origin":"","legend":"","description":"","filename":"table1to6.docx","url":"https://assets-eu.researchsquare.com/files/rs-7115262/v1/03e42ef0fa450d930376fc51.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Impact of Solvent-Accessible HLA Amino Acid Mismatches on Kidney Transplant Outcomes: A Multicenter Longitudinal Study","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLong-term kidney allograft survival has not matched advancements in short-term outcomes over past decades.\u003csup\u003e1, 2\u003c/sup\u003e A major factor in late graft failure is alloimmune injury, characterized by \u003cem\u003ede novo\u003c/em\u003e donor-specific HLA antibodies (dnDSA), antibody-mediated rejection (AMR), and T-cell-mediated rejection (TCMR).\u003csup\u003e3\u003c/sup\u003e Additionally, immunosuppression-related toxicities, including calcineurin inhibitor-related nephrotoxicity, malignancy, infection, cardiovascular, and metabolic complications further impair long-term patient and graft survival.\u003csup\u003e3, 4\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBalancing immunosuppression and adverse events remains challenging in kidney transplantation with significant variability in clinical practice worldwide driven by limited evidence and the absence of reliable alloimmune risk stratification tools.\u003csup\u003e1, 5, 6\u003c/sup\u003e Current, KDIGO guidelines recommend using Human Leukocyte Antigens (HLA)-A, -B, and -DR antigen-level mismatches and uniform graft monitoring every three months, regardless of individual risk.\u003csup\u003e7\u003c/sup\u003e While antigen-level mismatches have shown prognostic value,\u003csup\u003e8, 9\u003c/sup\u003e they lack the granularity needed for personalized immunosuppression or monitoring strategies, such as tailored graft function, dnDSA, and/or donor-derived cell-free DNA (dd-cfDNA) assessments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAlloimmunity targets peptides presented in the HLA binding groove (recognized by T cells) and polymorphic amino acid configurations on HLA molecules, which serve as epitopes for B-cell and antibody recognition. Eplet mismatch analysis identifies clinically relevant epitopes, known as eplets, and has gained traction in alloimmune risk assessment.\u003csup\u003e10-15\u003c/sup\u003e However, eplets are theoretically defined, subject to continuous database updates, and suffer from overlap, limiting broad clinical application. Moreover, prior studies mainly focused on HLA-DR and -DQ loci, restricting comprehensive risk stratification. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe HLA Epitope MisMatch Algorithm (HLA-EMMA) quantifies HLA mismatches at the solvent-accessible amino acid \u0026nbsp;(saAA) level, providing a stable and objective alloimmune risk measure based on defined HLA sequences.\u003csup\u003e16\u003c/sup\u003e This first multicenter longitudinal study investigates the impact of HLA saAA mismatches on key kidney transplant outcomes, including a hierarchical composite endpoint of death-censored graft failure (DCGF), biopsy-proven acute rejection (BPAR), dnDSA development, and each outcome individually.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003e\u003cstrong\u003eStudy design and participants\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eWe included consecutive KTRs from \\centers in Leiden, the Netherlands, and Leuven, Belgium. We collected pseudonymized data on donor and recipient characteristics, including sex, age, donor type (living or deceased), first or repeat transplantation, pre-transplant HLA DSA, cold ischemia time, HLA split-antigen mismatches, and saAA HLA mismatch load. Transplant outcomes recorded were DCGF, BPAR, and dnDSA development. Follow-up adhered to KDIGO guidelines.\u003csup\u003e17\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Dutch cohort comprised all adult single kidney transplant recipients between January 2005 and December 2019, with follow-up until July 2021 (Leiden University Medical Center IRB:W2020.031).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Belgian cohort included recipients between March 2004 and February 2013, with follow-up until September 2019 (NCT06505200; University Hospitals Leuven Ethics: S64006). All proceedings adhered to the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003eCombined or post-other organ transplants were excluded. All transplants had negative complement-dependent cytotoxicity (CDC) crossmatches. Baseline immunosuppression consisted primarily of tacrolimus, mycophenolic acid, and corticosteroids with basiliximab induction in Dutch KTRs and high-risk Belgian KTRs, while some Dutch high-risk KTRs received alemtuzumab.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eHLA genotyping and amino acid mismatch evaluation\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eFor the Dutch cohort, high resolution HLA data were obtained using two complementary approaches. First, we included donor-recipient pairs genotyped at second-field across 11 HLA loci using next-generation sequencing (NGS). DNA samples were processed with NGS kits from GenDx (Utrecht, The Netherlands), and sequenced on the Illumina NGS platform. Second, to generate high-resolution HLA data for the entire cohort, we applied an imputation algorithm developed by the NMDP Bioinformatics team.\u003csup\u003e18\u003c/sup\u003e We included the entire cohort in the imputation algorithm, including those with available second-field data, enabling quality assessment in a later phase. Low- and medium-resolution HLA data for HLA-A, -B, -C, -DRB1, and -DQB1 were imputed to second-field resolution using the most probable haplotypes and imputed genotypes, incorporating broad racial groups to enhance accuracy. This method produced HLA genotypes at second-field Antigen Recognition Domain (ARD) resolution (exons 2 and 3 for class I, and exon 2 for class II alleles) based on published US and European population haplotype frequencies.\u003csup\u003e19\u003c/sup\u003e As a result, we obtained translated second-field resolution data for HLA-A, -B, -C, -DRB1/3/4/5, and -DQB1. To assess imputation accuracy and quality, we compared and tested the agreement between the imputed high-resolution HLA data and actual NGS-derived second-field data (Supplementary Material Table S1). Whenever available, imputed alleles were replaced with actual NGS second-field alleles to minimize reliance on imputed data. HLA-DP imputation was excluded due to the absence of reliable quality assessment metrics.\u003c/p\u003e\n\u003cp\u003eFor the Belgian cohort, donor and recipient DNA samples were retrospectively genotyped at second-field resolution for 11 HLA loci using NGS. Half of the donor samples were genotyped using the MIA FORA NGS FLEX 11 HLA Typing Kit (Immucor, Norcross, GA) on the MiSeq sequencing instrument (Illumina, San Diego, CA); while the remaining donor samples and all recipient samples were genotyped at a high-resolution level (exon 2, 3, and 4 for class I and exon 2 and 3 for class II) using the HiSeq sequencing system (Illumina), as described previously.\u003csup\u003e20\u003c/sup\u003e\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eAntigen \u0026amp; solvent-accessible amino acid mismatch evaluation for both cohorts\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eAntigen mismatches (HLA-A, -B, -C, -DR, -DQ) were calculated at split level. saAA mismatches were assessed using high-resolution HLA genotypes \u0026nbsp;(HLA-A,-B,-C,-DR1/3/4/5, and -DQB1) for class I and II loci with HLA-EMMA v1.06, reported per locus, class, and total.\u003csup\u003e16\u003c/sup\u003e saAA mismatch analysis was performed intralocus for HLA class I and II.\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eOutcome definitions\u0026nbsp;\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eThe primary endpoint was a hierarchical composite prioritizing DCGF, BPAR, and dnDSA formation. Secondary endpoints included each component individually.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eDeath-censored graft failure (DCGF)\u003c/h2\u003e\n\u003cp\u003eDCGF was defined as permanent graft loss requiring dialysis or re-transplantation. Deaths without graft failure were censored.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eBPAR assessment and treatment of rejection episodes\u003c/h2\u003e\n\u003cp\u003eIn the Dutch cohort, KTRs did not undergo routine protocol biopsies, except when included in a clinical trial.\u0026nbsp;For-cause biopsies were performed in case of worsening serum creatinine and/or proteinuria without an evident alternative cause, and/or if dnDSA were detected during routine monitoring. All biopsy specimens were scored according to the Banff 2022 criteria.\u003csup\u003e23\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor kidney biopsies performed between 2005-2011, we used the originally reported diagnosis from the histology report. For biopsies performed after 2011, a single pathologist (JK) retrospectively re-evaluated all specimens. Borderline changes were defined as foci of tubulitis (t\u0026gt;0) with minor interstitial inflammation (i1) or moderate to severe interstitial inflammation (i2 or i3) with mild (t1) tubulitis, as previously described.\u003csup\u003e22\u003c/sup\u003e AMR was diagnosed based on the three 2022 Banff criteria for either acute or chronic active AMR, though non-HLA antibodies and gene expression changes were not taken into account.\u003csup\u003e23\u003c/sup\u003e Chronic-active TCMR was not considered as a separate diagnostic category and was instead categorized as borderline rejection or acute TCMR.\u003csup\u003e22\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn the Belgian cohort, KTRs underwent protocol and indication biopsies. All biopsy specimens were rescored according to the Banff 2022 criteria, based on prospectively collected individual lesion scores.\u003csup\u003e23\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003ednDSA detection\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eBoth cohorts underwent systematic HLA DSA monitoring pre- and post-transplant (Leiden: 6 months, Leuven: 3 months; then annually and at biopsies). First dnDSA detection defined dnDSA onset. All sera were screened using the Lifecodes LifeScreen Deluxe kit (Immucor), and for positive results, HLA antibody specificity was assessed using Lifecodes Single Antigen Bead kits (Immucor). Antibodies against HLA-A, -B, -C, -DRB1, -DRB345, -DQA1, and -DQB1, loci in the recipient sera were analyzed for donor specificity at the allelic level. A background-corrected median fluorescence intensity value of \u0026ge;500 for Leuven and \u0026ge;1000 for Leiden was used to define the presence of HLA DSA, as previously described.\u003csup\u003e21,22\u003c/sup\u003e Final DSA assignment considered donor-recipient HLA typing.\u003c/p\u003e\n\u003ch2\u003eStatistical analysis\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eAnalyses were conducted in R (version 1.4.2) and RStudio (Version 1.4.1717).\u0026nbsp;Continuous variables were summarized as medians with interquartile ranges (IQR), means with standard deviations (SD) and 95% confidence intervals (CI), and categorical variables as frequencies and percentages.\u003c/p\u003e\n\u003cp\u003eMultivariable Cox regression evaluated associations between saAA mismatch loads and outcomes, adjusting for donor/recipient sex, age, transplant center, donor type, prior transplants, cold ischemia time, and the presence of pre-transplant DSAs. Censoring occurred at \u0026nbsp;last DSA monitoring, graft failure, or last clinical follow-up. Significance was set at p \u0026lt;0.05 with False Discovery Rate (FDR) controlled at 0.05 using the Benjamini-Hochberg procedure to control for multiple testing.\u003c/p\u003e\n\u003cp\u003eWe conducted multiple sensitivity analyses to evaluate the robustness of our findings. First, we limited the analysis to recipients with only NGS based high-resolution HLA typing to assess consistency across primary and secondary endpoints. For secondary endpoint analyses, we limited the assessment to Total, Class I and Class II saAA mismatches to maintain methodological rigor and power, and prevent uninterpretable, excessively broad confidence intervals. Second, we assessed the impact and magnitude of HRs for individual HLA loci by analyzing the median increase per HLA locus rather than applying a fixed increment, accounting for variations in HLA distribution. Third, to address potential residual confounding from pre-transplant DSAs, we repeated the analyses in KTRs without detectable pre-transplant DSAs, despite adjustment in the multivariable model. Fourth, we reanalyzed the association with BPAR after excluding rejections identified via protocol biopsies, focusing exclusively on for-cause biopsies. Fifth, we repeated analyses in a subgroup of KTRs after excluding all HLA-identical family donors. Full results are detailed in the manuscript text and Supplementary Tables.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003e\u003cstrong\u003eCohort characteristics\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eWe included 2473 KTRs with a median follow-up of 6.9 years (IQR: 3.6-9.9) (Table 1). Of these, 937 (38%) received a living donor kidney, more frequently in Leiden than Leuven (57% vs 5%). Pre-formed DSA were present in 7% of recipients. The hierarchical composite endpoint occurred in 883 (35.7%) KTRs. dnDSA developed in 315 KTRs, predominantly targeting class II HLA (n=288; 91%), with HLA-DQ antibodies being the most frequent (n=188). BPAR occurred in 555 (22%) KTRs, DCGF in 335 (14%), and 549 (22%) patients died.\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eHigh resolution HLA imputation and HLA-saAA mismatch scores\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eAmong 49.460 alleles (HLA-A, -B, -C, -DR, and -DQ), 22.907 (46%) were NGS-typed and 54% imputed. Quality assessment of the translation to high-resolution revealed accurate imputative results for all loci; HLA-A, -B, -C, -DRB1/3/4/5, and -DQB1 (Supplementary Material Table S1). Mean split-antigen mismatches were 1.0\u0026plusmn;0.7, 1.2\u0026plusmn;0.7, 1.1\u0026plusmn;0.7, 0.9\u0026plusmn;0.6, 0.8\u0026plusmn;0.6 for HLA-A, -B, -C, -DR, and -DQ, respectively (Table 1). Median saAA mismatches were 16 for Class I and 18 for Class II, with a strong correlation between total antigen and saAA mismatches (\u0026rho;=0.711,P\u0026lt;0.001) (Supplementary Figure S1).\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eMedian saAA mismatch scores and transplant outcomes\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eKTRs experiencing the composide endpoint, BPAR, or dnDSA had significantly higher median total saAA mismatches compared to those without (P\u0026lt;0.001 for all; Supplementary Table S2). Notably, class II saAA mismatches were significantly higher in KTRs that experienced graft loss, BPAR and developed dnDSA.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eCox regression analyses for saAA mismatch scores and kidney graft outcomes\u0026nbsp;\u003c/strong\u003e\u003c/h2\u003e\n\u003ch3\u003e\u003cstrong\u003eHierarchical composite endpoint\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eTotal, class I, class II, and all locus-specific saAA mismatches were all significantly associated with the primary endpoint. For every ten saAA mismatches, HRs were 1.13 (95% CI: 1.09-1.16) for total, 1.15 (95% CI: 1.08-1.23) for class I, and 1.14 (95% CI: 1.10-1.19) for class II saAA (Table 2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMoreover, both class I and class II saAA mismatches were independent risk factors for the composite endpoint, with HRs of 1.12 (95% CI: 1.05-1.19), and 1.13 (95% CI: 1.09-1.18), respectively (Supplementary Table S3).\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eDCGF\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eTotal, class I, class II, and all locus-specific saAA mismatch scores were significantly associated with DCGF (Table 3). For every ten saAA mismatches, HRs were 1.13 (95% CI: 1.07-1.19), 1.21 (95% CI: 1.09-1.33), and 1.11 (95% CI: 1.04-1.18) for total, class I, and class II, respectively. Locus-specific aHRs ranged from 1.16 (95% CI: 1.06-1.26) for HLA-DQ to 1.65 (95% CI: 1.27-2.14) for HLA-C (Table 3).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBoth class I and class II saAA mismatches were independent risk factors for DCGF, with HRs of 1.20 (95% CI: 1.08-1.32, p\u0026lt;.001) and 1.10 (95% CI: 1.04-1.17, p=0.002), respectively (Supplementary Table S4).\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eBPAR\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eAll saAA mismatches were significantly associated with BPAR in multivariable analyses (Table 4). For every ten HLA saAA mismatches, HR was 1.12 (95% CI: 1.07-1.16), 1.18 (95% CI: 1.08-1.28), and 1.12 (95% CI: 1.06-1.17) for total, class I, and class II respectively. For the locus-specific saAA mismatches, HR ranged from 1.14 (95% CI: 1.06-1.22) for HLA-DQ to 1.58 (95% CI: 1.28-1.96) for HLA-B (Table 4). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBoth class I and class II saAA mismatches were independent risk factors for BPAR with HRs of 1.15 (95% CI: 1.06-1.25, p=0.001) and 1.10 (95% CI: 1.05-1.16, p\u0026lt;.001), respectively (Supplementary Table S5).\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003ednDSA\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eTotal, class I, class II, and locus-specific saAA mismatches were significantly associated with dnDSA (Table 5a-5b). For every ten saAA mismatches, HR was 1.22 (95% CI: 1.16-1.29), 1.21 (95% CI: 1.10-1.33), and 1.30 (95% CI: 1.21-1.40) for total, class I, and class II respectively. For the locus-specific HLA, aR ranged from 1.26 (95% CI: 1.10-1.33) for HLA-A to 1.67 (95% CI: 1.31-2.13) for HLA-B. When investigating the impact of locus-specific saAA mismatches on locus-specific dnDSA, we observed comparable results with HR ranging from 1.90 (95% CI: 1.64-2.21) for HLA-DQ to 4.06 (95% CI: 2.08-7.96) for HLA-C (Table 5b).\u003c/p\u003e\n\u003cp\u003eIn multivariable models including antigen mismatches, saAA mismatches remained independently associated with locus-specific dnDSAs (Table 6). Thus, saAA mismatches impacted graft outcomes beyond antigenic mismatches.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe identified both class I and class II saAA as independent risk factors for dnDSA, with HRs of 1.32 (95% CI: 1.20-1.45) and 1.18 (95% CI: 1.11-1.26), respectively (Supplementary Table S6). Furthermore, we also analyzed the impact of saAA mismatches within the context of one or two HLA antigen-level mismatches. In the presence of a single HLA antigen-level mismatch, locus-specific saAA mismatches across all HLA-loci significantly impacted locus-specific DSA formation. With two HLA antigen-level mismatches, locus-specific saAA mismatches had a significant effect on DR- and A-DSA, whereas effects on other HLA loci did not reach statistical significance \u0026ndash; likely due to the lower number of events and reduced statistical power (Supplementary Material Table S7).\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eSensitivity analyses\u0026nbsp;\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eAll findings remained consistent across all sensitivity analyses performed.\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eNGS-only cohort\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eSignificant associations were observed between total, class I, class II, and locus-specific saAA mismatches and the hierarchical composite endpoint. HRs ranged from 1.09 to 1.60, with P-values between \u0026lt;0.001 and 0.029 for every ten saAA mismatches (Supplementary Tables S8-11).\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eMedian increase per locus\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eAs expected, results were comparable across loci. HRs for total, class I, and class II saAA mismatch scores were highest, reflecting their aggregation of all HLA loci (Supplementary Tables S12-16). As only the distribution of the saAA mismatch score was changed for this analysis, we did not report P-values as they are identical to the primary analysis.\u003c/p\u003e\n\u003ch3\u003e\u003cstrong\u003eIn patients without pre-transplant donor-specific antibodies\u003c/strong\u003e\u003c/h3\u003e\n\u003cp\u003eThe associations found between total, class I, class II, locus specific HLA-A, -B, -C, -DR, and -DQ and primary and secondary outcomes remained unchanged (Supplementary Tables S17-21).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eExclusion of BPARs diagnosed in protocol biopsies\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBPAR associations remained unaffected (Supplementary Table S22).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eExclusion of HLA-identical familial kidney donors.\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eResults were stable, except saAA mismatch scores for HLA-A lost significance for BPAR (Supplementary Tables S23\u0026ndash;27).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this longitudinal multicenter study, we demonstrate the impact of saAA mismatches, calculated using HLA-EMMA, on hierarchical kidney transplant outcomes. Leveraging two well-characterized, unique cohorts with extensive follow-up and a substantial number of events, we showed significant impact of saAA mismatches for class I and class II HLA on dnDSA (overall and locus-specific), BPAR, and DCGF beyond effects of traditional antigen-level HLA mismatches.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNotable similarities and differences were observed between the cohorts from the Netherlands and Belgium. Both cohorts had comparable age and sex distributions among KTRs. However, donors in Leiden were older (53 vs. 48 years), with a higher proportion of living donors (57% vs 5% in Leuven), and with a higher proportion of kidneys donated from donors after circulatory death (58% vs. 17%). Fewer patients in Leiden had detectable pre-transplant DSAs (5% vs. 11%). The higher prevalence of living donor transplants in Leiden was associated with higher saAA mismatch scores across all HLA loci, reflecting reduced emphasis on matching in living donors. In addition, donor-recipient pairs in deceased donor transplantation (DBD and DCD) exhibited higher total saAA mismatch scores in Leiden, suggesting a higher immunological risk in this cohort. This may partly explain the higher incidence of de novo donor-specific antibodies (dnDSA) observed in Leiden compared to Leuven. Differences in DSA detection methods may also contribute to these findings. In Leiden, pre-transplant DSA were assessed using evolving techniques over time, including complement-dependent cytotoxicity (CDC) screening and crossmatches, flow cytometry crossmatches, and single antigen bead assays, with the latter applied to high-risk transplants after 2016. In contrast, the Leuven cohort utilized single antigen bead assays on biobanked sera for consistent pre-transplant DSA detection. These methodological differences could account for discrepancies in the reported prevalence of pre- and post-transplant DSAs. To account for center-specific differences, transplant center was included as a covariate in multivariable analyses. Interestingly, the incidence of BPAR was higher in Leuven, likely due to the use of protocol biopsies capturing subclinical rejection, which are not part of routine care in Leiden. Rates of DCGF between both cohorts were comparable at 2- and 5-years post-transplant, despite a higher percentage of living donor transplants in Leiden. These findings highlight the multifactorial nature of graft outcomes influenced by donor and recipient characteristics, the prevalence of pre-transplant DSAs, immunological risk, and variations in immunosuppressive therapy and drug exposure over time. Importantly, they emphasize that histocompatibility is only one determinant of alloimmune risk, which can be mitigated or modulated through tailored immunosuppressive strategies.\u003csup\u003e26, 27\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, we included saAA mismatch scores across all HLA-loci, excluding HLA-DP, to comprehensively assess all polymorphic residues potentially recognizable by B-cell receptors and DSA. We found that both class I and class II HLA are associated with dnDSA, BPAR, and DCGF. Notably, we observed a previously unreported hazard of locus-specific saAA mismatches for HLA class I (HLA-A, -B, -C) on locus-specific dnDSA. As expected, DQ-specific DSAs were most prevalent in our cohort, followed by DSAs targeting HLA-DR, -A, -B, and -C. These findings underscore the immunological relevance of both class I and class II HLA in driving alloimmune responses including development of dnDSA and incidence of BPAR, which in turn influence long-term graft survival. Consequently, our results challenge the traditional emphasis on HLA-A, -B, -DR antigen-level mismatches and advocate for a more refined, granular approach, which also considers saAA mismatches across other HLA-loci. This broader perspective may refine risk stratification and provide deeper insights into the complexity of alloimmunity. Supporting our findings, a prior study has demonstrated that some patients classified as low immunological risk based on traditional HLA-antigen mismatch may harbor a high epitope mismatch load, placing them at elevated immunological risk despite initial assessments using traditional antigenic mismatches.\u003csup\u003e11\u003c/sup\u003e \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOver the past decade, multiple studies have demonstrated that HLA-DR and HLA-DQ eplets outperform antigen-level HLA mismatches in predicting the risk of dnDSA, AMR, and TCMR, and that they modulate the association between tacrolimus exposure with adverse events from nonadherence.\u003csup\u003e11, 12, 26-32\u003c/sup\u003e These studies also highlight the superior predictive value of eplets over antigen mismatches for dnDSA formation.\u003csup\u003e12, 22, 31\u003c/sup\u003e Prior research primarily focused on HLA-DR and -DQ eplet mismatches due to the high prevalence DSAs against these loci. Yet, alloimmune responses are not restricted to these antigens alone as shown in this study. Additionally, earlier studies relied on theoretical eplet mismatches defined by HLA-Matchmaker, which is subject to continuous updates as eplet definitions changes. Notably, only a minority of eplets have gone through antibody-verification, making it difficult to assess their true immunogenic potential.\u003csup\u003e33, 34\u003c/sup\u003e These evolving definitions complicate reproducibility and limit clinical utility of eplet-based risk stratification for personalizing care in KTRs.\u003csup\u003e34\u003c/sup\u003e In contrast, HLA-EMMA provides an objective, standardized method for calculating saAA mismatches across all HLA-loci.\u003csup\u003e16, 35\u003c/sup\u003e This approach improves reproducibility and enhances clinical utility by predicting the relative solvent accessibility of HLA structures. Moreover, HLA-EMMA provides a user-friendly platform with ease of digital application.\u003csup\u003e36\u003c/sup\u003e However, it is important to acknowledge its limitations - while HLA-EMMA may have higher sensitivity, it could also include residues that are not clinically relevant, thereby negatively impacting specificity. Conversely, antibody-verified eplet analyses, while more specific, may exclude immunologically relevant eplets that have not yet been experimentally verified. Similarly, assessing overall eplet mismatches poses the same limitation as evaluating total saAA mismatches.\u003csup\u003e16\u003c/sup\u003e We intentionally refrained from using saAA mismatch scores to categorize patients into low-, medium-, or high-risk groups for alloimmunity. However, from a clinical perspective, a sensitive and quantitative marker remains valuable. A refined risk stratification tool could facilitate individualized tapering of immunosuppression and optimize screening protocols, particularly for patients with a low saAA mismatch score who may be at reduced immunological risk.\u003c/p\u003e\n\u003cp\u003eThe (kidney) transplant field urgently requires improved risk stratification tools for alloimmunity, as emphasized by Hariharan et al. and others.\u003csup\u003e1, 37\u003c/sup\u003e Currently, many centers reduce immunosuppression reactively in response to cumulative toxicity, often considering presumed immunosenescence and diminished immunogenicity over time. However, proactive immunosuppression minimization is rarely practiced, typically being reserved for HLA-identical living-related transplants.\u003csup\u003e5, 6\u003c/sup\u003e We propose that enhanced alloimmune risk stratification could significantly improve the personalization of both surveillance and maintenance immunosuppressive regimens, a concept supported by previous studies.\u003csup\u003e1, 26, 38\u003c/sup\u003e Patients with higher immunological risk (e.g., higher EMMA scores or younger age) might benefit from more frequent monitoring and sustained immunosuppression, while those at lower risk could achieve favorable outcomes with reduced immunosuppression.\u003csup\u003e38-40\u003c/sup\u003e This approach aligns with findings from recent studies utilizing HLA-DR/-DQ eplet mismatch analyses, which demonstrated that tailored immunosuppression and less intensive monitoring strategies could be implemented effectively based on specific thresholds.\u003csup\u003e11, 31, 39\u003c/sup\u003e These studies collectively advocate for a patient-centered approach based on molecular matching scores rather than adherence to center-specific protocols.\u003csup\u003e41\u003c/sup\u003e However, more prospective studies and randomized controlled trials (RCTs) will be essential for establishing clinical utility and implement or change KDIGO guidelines.\u003c/p\u003e\n\u003cp\u003eThis study has several limitations. First, while our findings suggest that alloimmune risk increases linearly with the number of saAA mismatches, this does not imply that all mismatches are equally immunogenic. In reality, the pathogenicity of specific amino acid polymorphisms varies, and HLA-EMMA does not fully account for these differences.\u003csup\u003e42-44\u003c/sup\u003e Additionally, HLA-EMMA does not integrate other critical factors influencing alloimmune risk, such as immunosenescence or intrapatient variability in immunosuppression.\u003csup\u003e45\u003c/sup\u003e Second, high-resolution HLA typing was unavailable for 54% of alleles, requiring imputation. However, internal validation demonstrated high concordance between imputed and actual high-resolution typing (Supplementary Material Table S1), with accuracy ranging from 86% to 92% for class I HLA. Given our novel findings on class I molecular mismatch load and kidney graft outcomes, this level of accuracy is a key strength. To further assess robustness, we performed a sensitivity analysis restricted to patients with only NGS-based typing, which yielded consistent findings across HLA-A, -B, -C, -DR, and -DQ, minimizing the risk of false-positives due to imputation (Supplementary Material Tables S24-27). Third, we lacked DQA1 data, which may have influenced the observed incidence of DQ-specific antibodies. This could have either inflated the apparent frequency of DQ antibodies or underestimated the true immunological burden. Nonetheless, given that our finding on HLA-DQ is supported by prior studies, \u003csup\u003e26-32, 39, 46\u003c/sup\u003e we believe this limitation has minimal impact. Future studies assessing clinical utility of HLA-EMMA, should definitely incorporate comprehensive DQA1 typing. Finally, we were unable to assess the potential impact of HLA-DP on kidney graft outcomes. This is noteworthy as a previous study demonstrated that HLA-DP DSA, while rare, represent a significant risk for AMR.\u003csup\u003e47\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eIn conclusion, saAA mismatch burden in both class I and II HLA is associated with increased risk of dnDSA, BPAR, and graft loss. saAA mismatch analysis provides a more precise measure of alloimmune risk than traditional antigen-level mismatches and merits further prospective evaluation. This refined approach could improve alloimmune risk stratification, optimize surveillance strategies, and guide personalized immunosuppression to balance efficacy and toxicity in kidney transplant recipients.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eDisclosure statement\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eIn the last 3 years TvG has received lecture fees and study grants from Chiesi and Astellas, in addition to consulting fees from Roche Diagnostics, Thermo Fisher, Vitaeris, CSL Behring, Astellas and Aurinia Pharma. AdV received lecture and consulting fees from Sandoz, Chiesi, Astellas, Hansa, CSL Behring, Neovii, AstraZeneca, Sanofi, Takeda, and Novartis in the past years, all of which went to his employer LUMC. SM received lecture fees from Chiesi. JK received lecture fees from Chiesi, is consultant for Aiosyn BV, Hansa, Alentis Pharmaceuticals AG and Novartis AG. In all cases, reimbursements have been transferred to employer accounts, and none have been paid to personal bank accounts. The other authors of this manuscript have no conflicts of interest to disclose as described by Nature Medicine.\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eNational Marrow Donor Program in the United States\u003c/p\u003e\n\u003cp\u003eThe contributions by Leuven were supported by Fonds Wetenschappelijk Onderzoek (FWO; Research Foundation \u0026ndash; Flanders) and the Agentschap Innoveren en Ondernemen (VLAIO; Agency for Innovation and Entrepreneurship) by the TBM project grant IWT.150199, FWO research project grants G038024N and G087620N), and by a KU Leuven internal funding grant C2M/24/057. MC is a postdoctoral researcher of FWO (12D6423N), and MN is a senior clinical investigator of FWO (1842919N).\u003c/p\u003e\n\u003ch2\u003eAuthors\u0026rsquo; Contributions\u003c/h2\u003e\n\u003cp\u003eSM, SB, AS, MC, JK, DvdH, PvdB, HdF, GH, MN, SH, DR, AdV conceived and designed the work and played a key role in interpreting the results. All authors revised the manuscript and approved with the definitive version.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eSupplementary Materials\u003c/h2\u003e\n\u003cp\u003eSee separate file\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eHariharan S, Israni AK, Danovitch G. Long-Term Survival after Kidney Transplantation. \u003cem\u003eNew England Journal of Medicine\u003c/em\u003e 2021; \u003cstrong\u003e385: \u003c/strong\u003e729-743.\u003c/li\u003e\n\u003cli\u003eLoupy A, Lefaucheur C. Antibody-Mediated Rejection of Solid-Organ Allografts. \u003cem\u003eThe New England journal of medicine\u003c/em\u003e 2018; \u003cstrong\u003e379: \u003c/strong\u003e1150-1160.\u003c/li\u003e\n\u003cli\u003eMayrdorfer M, Liefeldt L, Wu K\u003cem\u003e, et al.\u003c/em\u003e Exploring the Complexity of Death-Censored Kidney Allograft Failure. \u003cem\u003eJournal of the American Society of Nephrology\u003c/em\u003e 2021; \u003cstrong\u003e32: \u003c/strong\u003e1513-1526.\u003c/li\u003e\n\u003cli\u003eMayrdorfer M, Liefeldt L, Osmanodja B\u003cem\u003e, et al.\u003c/em\u003e A single centre in-depth analysis of death with a functioning kidney graft and reasons for overall graft failure. \u003cem\u003eNephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association\u003c/em\u003e 2023; \u003cstrong\u003e38: \u003c/strong\u003e1857-1866. \u003c/li\u003e\n\u003cli\u003ePerez-Saez MJ, Montero N, Oliveras L\u003cem\u003e, et al.\u003c/em\u003e Immunosuppression of HLA identical living-donor kidney transplant recipients: A systematic review. \u003cem\u003eTransplantation reviews (Orlando, Fla)\u003c/em\u003e 2023; \u003cstrong\u003e37: \u003c/strong\u003e100787.\u003c/li\u003e\n\u003cli\u003eWojciechowski D, Wiseman A. Long-Term Immunosuppression Management: Opportunities and Uncertainties. \u003cem\u003eClinical journal of the American Society of Nephrology : CJASN\u003c/em\u003e 2021; \u003cstrong\u003e16: \u003c/strong\u003e1264-1271.\u003c/li\u003e\n\u003cli\u003eChadban SJ, Ahn C, Axelrod DA\u003cem\u003e, et al.\u003c/em\u003e KDIGO Clinical Practice Guideline on the Evaluation and Management of Candidates for Kidney Transplantation. \u003cem\u003eTransplantation\u003c/em\u003e 2020; \u003cstrong\u003e104: \u003c/strong\u003eS11-s103.\u003c/li\u003e\n\u003cli\u003eSusal C. Collaborative Transplant Study Website. 2022.\u003c/li\u003e\n\u003cli\u003eHeidt S, Haasnoot GW, van Rood JJ\u003cem\u003e, et al.\u003c/em\u003e Kidney allocation based on proven acceptable antigens results in superior graft survival in highly sensitized patients. \u003cem\u003eKidney international\u003c/em\u003e 2018; \u003cstrong\u003e93: \u003c/strong\u003e491-500.\u003c/li\u003e\n\u003cli\u003eTambur AR, Das R. Can We Use Eplets (or Molecular) Mismatch Load Analysis to Improve Organ Allocation? The Hope and the Hype. \u003cem\u003eTransplantation\u003c/em\u003e 2023; \u003cstrong\u003e107: \u003c/strong\u003e605-615.\u003c/li\u003e\n\u003cli\u003eSnanoudj R, Kamar N, Cassuto E\u003cem\u003e, et al.\u003c/em\u003e Epitope load identifies kidney transplant recipients at risk of allosensitization following minimization of immunosuppression. \u003cem\u003eKidney international\u003c/em\u003e 2019; \u003cstrong\u003e95: \u003c/strong\u003e1471-1485.\u003c/li\u003e\n\u003cli\u003eWiebe C, Kosmoliaptsis V, Pochinco D\u003cem\u003e, et al.\u003c/em\u003e A Comparison of HLA Molecular Mismatch Methods to Determine HLA Immunogenicity. \u003cem\u003eTransplantation\u003c/em\u003e 2018; \u003cstrong\u003e102: \u003c/strong\u003e1338-1343.\u003c/li\u003e\n\u003cli\u003eCai J, Terasaki PI, Mao Q\u003cem\u003e, et al.\u003c/em\u003e Development of nondonor-specific HLA-DR antibodies in allograft recipients is associated with shared epitopes with mismatched donor DR antigens. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2006; \u003cstrong\u003e6: \u003c/strong\u003e2947-2954.\u003c/li\u003e\n\u003cli\u003eTambur AR, Rosati J, Roitberg S\u003cem\u003e, et al.\u003c/em\u003e Epitope analysis of HLA-DQ antigens: what does the antibody see? \u003cem\u003eTransplantation\u003c/em\u003e 2014; \u003cstrong\u003e98: \u003c/strong\u003e157-166.\u003c/li\u003e\n\u003cli\u003eKramer CSM, Israeli M, Mulder A\u003cem\u003e, et al.\u003c/em\u003e The long and winding road towards epitope matching in clinical transplantation. \u003cem\u003eTransplant international : official journal of the European Society for Organ Transplantation\u003c/em\u003e 2019; \u003cstrong\u003e32: \u003c/strong\u003e16-24.\u003c/li\u003e\n\u003cli\u003eKramer CSM, Koster J, Haasnoot GW\u003cem\u003e, et al.\u003c/em\u003e HLA-EMMA: A user-friendly tool to analyse HLA class I and class II compatibility on the amino acid level. \u003cem\u003eHla\u003c/em\u003e 2020; \u003cstrong\u003e96: \u003c/strong\u003e43-51.\u003c/li\u003e\n\u003cli\u003eKasiske BL, Zeier MG, Chapman JR\u003cem\u003e, et al.\u003c/em\u003e KDIGO clinical practice guideline for the care of kidney transplant recipients: a summary. \u003cem\u003eKidney international\u003c/em\u003e 2010; \u003cstrong\u003e77: \u003c/strong\u003e299-311.\u003c/li\u003e\n\u003cli\u003eMadbouly A, Gragert L, Freeman J\u003cem\u003e, et al.\u003c/em\u003e Validation of statistical imputation of allele-level multilocus phased genotypes from ambiguous HLA assignments. \u003cem\u003eTissue Antigens\u003c/em\u003e 2014; \u003cstrong\u003e84: \u003c/strong\u003e285-292.\u003c/li\u003e\n\u003cli\u003eGragert L, Madbouly A, Freeman J\u003cem\u003e, et al.\u003c/em\u003e Six-locus high resolution HLA haplotype frequencies derived from mixed-resolution DNA typing for the entire US donor registry. \u003cem\u003eHuman immunology\u003c/em\u003e 2013; \u003cstrong\u003e74: \u003c/strong\u003e1313-1320.\u003c/li\u003e\n\u003cli\u003eSenev A, Lerut E, Van Sandt V\u003cem\u003e, et al.\u003c/em\u003e Specificity, strength, and evolution of pretransplant donor-specific HLA antibodies determine outcome after kidney transplantation. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2019; \u003cstrong\u003e19: \u003c/strong\u003e3100-3113. \u003c/li\u003e\n\u003cli\u003eWisse BW, Kamburova EG, Joosten I\u003cem\u003e, et al.\u003c/em\u003e Toward a Sensible Single-antigen Bead Cutoff Based on Kidney Graft Survival. \u003cem\u003eTransplantation\u003c/em\u003e 2019; \u003cstrong\u003e103: \u003c/strong\u003e789-797.\u003c/li\u003e\n\u003cli\u003eSenev A, Coemans M, Lerut E\u003cem\u003e, et al.\u003c/em\u003e Eplet Mismatch Load and De Novo Occurrence of Donor-Specific Anti-HLA Antibodies, Rejection, and Graft Failure after Kidney Transplantation: An Observational Cohort Study. \u003cem\u003eJ Am Soc Nephrol\u003c/em\u003e 2020; \u003cstrong\u003e31: \u003c/strong\u003e2193-2204.\u003c/li\u003e\n\u003cli\u003eNaesens M, Roufosse C, Haas M\u003cem\u003e, et al.\u003c/em\u003e The Banff 2022 Kidney Meeting Report: Reappraisal of microvascular inflammation and the role of biopsy-based transcript diagnostics. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2024; \u003cstrong\u003e24: \u003c/strong\u003e338-349.\u003c/li\u003e\n\u003cli\u003eKoshy P, Furian L, Nickerson P\u003cem\u003e, et al.\u003c/em\u003e European Survey on Clinical Practice of Detecting and Treating T-Cell Mediated Kidney Transplant Rejection. \u003cem\u003eTransplant international : official journal of the European Society for Organ Transplantation\u003c/em\u003e 2024; \u003cstrong\u003e37: \u003c/strong\u003e12283.\u003c/li\u003e\n\u003cli\u003evan den Broek DAJ, Meziyerh S, Budde K\u003cem\u003e, et al.\u003c/em\u003e The Clinical Utility of Post-Transplant Monitoring of Donor-Specific Antibodies in Stable Renal Transplant Recipients: A Consensus Report With Guideline Statements for Clinical Practice. \u003cem\u003eTransplant international : official journal of the European Society for Organ Transplantation\u003c/em\u003e 2023; \u003cstrong\u003e36: \u003c/strong\u003e11321.\u003c/li\u003e\n\u003cli\u003eWiebe C, Rush DN, Nevins TE\u003cem\u003e, et al.\u003c/em\u003e Class II Eplet Mismatch Modulates Tacrolimus Trough Levels Required to Prevent Donor-Specific Antibody Development. \u003cem\u003eJ Am Soc Nephrol\u003c/em\u003e 2017; \u003cstrong\u003e28: \u003c/strong\u003e3353-3362.\u003c/li\u003e\n\u003cli\u003eDavis S, Wiebe C, Campbell K\u003cem\u003e, et al.\u003c/em\u003e Adequate tacrolimus exposure modulates the impact of HLA class II molecular mismatch: a validation study in an American cohort. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2021; \u003cstrong\u003e21: \u003c/strong\u003e322-328.\u003c/li\u003e\n\u003cli\u003eWiebe C, Pochinco D, Blydt-Hansen TD\u003cem\u003e, et al.\u003c/em\u003e Class II HLA epitope matching-A strategy to minimize de novo donor-specific antibody development and improve outcomes. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2013; \u003cstrong\u003e13: \u003c/strong\u003e3114-3122.\u003c/li\u003e\n\u003cli\u003eWiebe C, Gibson IW, Blydt-Hansen TD\u003cem\u003e, et al.\u003c/em\u003e Rates and determinants of progression to graft failure in kidney allograft recipients with de novo donor-specific antibody. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2015; \u003cstrong\u003e15: \u003c/strong\u003e2921-2930.\u003c/li\u003e\n\u003cli\u003eWiebe C, Nickerson P. Strategic Use of Epitope Matching to Improve Outcomes. \u003cem\u003eTransplantation\u003c/em\u003e 2016; \u003cstrong\u003e100: \u003c/strong\u003e2048-2052.\u003c/li\u003e\n\u003cli\u003eWiebe C, Kosmoliaptsis V, Pochinco D\u003cem\u003e, et al.\u003c/em\u003e HLA-DR/DQ molecular mismatch: A prognostic biomarker for primary alloimmunity. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2019; \u003cstrong\u003e19: \u003c/strong\u003e1708-1719.\u003c/li\u003e\n\u003cli\u003eWiebe C, Nickerson PW. Human leukocyte antigen molecular mismatch to risk stratify kidney transplant recipients. \u003cem\u003eCurrent opinion in organ transplantation\u003c/em\u003e 2020; \u003cstrong\u003e25: \u003c/strong\u003e8-14.\u003c/li\u003e\n\u003cli\u003eBezstarosti S, Bakker KH, Kramer CSM\u003cem\u003e, et al.\u003c/em\u003e A Comprehensive Evaluation of the Antibody-Verified Status of Eplets Listed in the HLA Epitope Registry. \u003cem\u003eFrontiers in immunology\u003c/em\u003e 2021; \u003cstrong\u003e12: \u003c/strong\u003e800946.\u003c/li\u003e\n\u003cli\u003eTambur AR, Das R. Can We Use Eplets (or Molecular) Mismatch Load Analysis to Improve Organ Allocation? The Hope and the Hype. \u003cem\u003eTransplantation\u003c/em\u003e 2022. \u003c/li\u003e\n\u003cli\u003eDuquesnoy RJ, Askar M. HLAMatchmaker: a molecularly based algorithm for histocompatibility determination. V. Eplet matching for HLA-DR, HLA-DQ, and HLA-DP. \u003cem\u003eHuman immunology\u003c/em\u003e 2007; \u003cstrong\u003e68: \u003c/strong\u003e12-25.\u003c/li\u003e\n\u003cli\u003eLadowski JM, Mullins H, Romine M\u003cem\u003e, et al.\u003c/em\u003e Eplet mismatch scores and de novo donor-specific antibody development in simultaneous pancreas-kidney transplantation. \u003cem\u003eHuman immunology\u003c/em\u003e 2021; \u003cstrong\u003e82: \u003c/strong\u003e139-146.\u003c/li\u003e\n\u003cli\u003eEttenger R, Albrecht R, Alloway R\u003cem\u003e, et al.\u003c/em\u003e Meeting report: FDA public meeting on patient-focused drug development and medication adherence in solid organ transplant patients. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2018; \u003cstrong\u003e18: \u003c/strong\u003e564-573.\u003c/li\u003e\n\u003cli\u003eHricik DE, Formica RN, Nickerson P\u003cem\u003e, et al.\u003c/em\u003e Adverse Outcomes of Tacrolimus Withdrawal in Immune-Quiescent Kidney Transplant Recipients. \u003cem\u003eJ Am Soc Nephrol\u003c/em\u003e 2015; \u003cstrong\u003e26: \u003c/strong\u003e3114-3122.\u003c/li\u003e\n\u003cli\u003eWiebe C, Balshaw R, Gibson IW\u003cem\u003e, et al.\u003c/em\u003e A rational approach to guide cost-effective de novo donor-specific antibody surveillance with tacrolimus immunosuppression. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2023; \u003cstrong\u003e23: \u003c/strong\u003e1882-1892.\u003c/li\u003e\n\u003cli\u003eMatas AJ, Gaston RS. Moving Beyond Minimization Trials in Kidney Transplantation. \u003cem\u003eJ Am Soc Nephrol\u003c/em\u003e 2015; \u003cstrong\u003e26: \u003c/strong\u003e2898-2901.\u003c/li\u003e\n\u003cli\u003eAxelrod DA, Naik AS, Schnitzler MA\u003cem\u003e, et al.\u003c/em\u003e National Variation in Use of Immunosuppression for Kidney Transplantation: A Call for Evidence-Based Regimen Selection. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2016; \u003cstrong\u003e16: \u003c/strong\u003e2453-2462.\u003c/li\u003e\n\u003cli\u003eKosmoliaptsis V, Mallon DH, Chen Y\u003cem\u003e, et al.\u003c/em\u003e Alloantibody Responses After Renal Transplant Failure Can Be Better Predicted by Donor-Recipient HLA Amino Acid Sequence and Physicochemical Disparities Than Conventional HLA Matching. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2016; \u003cstrong\u003e16: \u003c/strong\u003e2139-2147.\u003c/li\u003e\n\u003cli\u003eSapir-Pichhadze R, Zhang X, Ferradji A\u003cem\u003e, et al.\u003c/em\u003e Epitopes as characterized by antibody-verified eplet mismatches determine risk of kidney transplant loss. \u003cem\u003eKidney international\u003c/em\u003e 2020; \u003cstrong\u003e97: \u003c/strong\u003e778-785.\u003c/li\u003e\n\u003cli\u003eMohammadhassanzadeh H, Oualkacha K, Zhang W\u003cem\u003e, et al.\u003c/em\u003e On Path to Informing Hierarchy of Eplet Mismatches as Determinants of Kidney Transplant Loss. \u003cem\u003eKidney international reports\u003c/em\u003e 2021; \u003cstrong\u003e6: \u003c/strong\u003e1567-1579.\u003c/li\u003e\n\u003cli\u003evan Gelder T. Within-patient variability in immunosuppressive drug exposure as a predictor for poor outcome after transplantation. \u003cem\u003eKidney international\u003c/em\u003e 2014; \u003cstrong\u003e85: \u003c/strong\u003e1267-1268.\u003c/li\u003e\n\u003cli\u003eWiebe C, Nevins TE, Robiner WN\u003cem\u003e, et al.\u003c/em\u003e The Synergistic Effect of Class II HLA Epitope-Mismatch and Nonadherence on Acute Rejection and Graft Survival. \u003cem\u003eAmerican journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons\u003c/em\u003e 2015; \u003cstrong\u003e15: \u003c/strong\u003e2197-2202.\u003c/li\u003e\n\u003cli\u003eDani\u0026euml;ls L, Claas FHJ, Kramer CSM\u003cem\u003e, et al.\u003c/em\u003e The role of HLA-DP mismatches and donor specific HLA-DP antibodies in kidney transplantation: a case series. \u003cem\u003eTransplant immunology\u003c/em\u003e 2021; \u003cstrong\u003e65: \u003c/strong\u003e101287. \u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 to 6 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"HLA, Mismatch load analysis, molecular mismatch, epitope, death-censored graft survival, de novo (dn)DSA, biopsy-proven acute rejection (BPAR)","lastPublishedDoi":"10.21203/rs.3.rs-7115262/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7115262/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Current alloimmune risk assessment in kidney transplantation relies on HLA antigen mismatches, lacking precision for individualized care. In contrast, solvent-accessible HLA amino acid (saAA) mismatches, empirically defined across all HLA loci, offer a stable and objective alternative. We conducted a multicenter longitudinal study evaluating saAA mismatches, calculated using HLA-EMMA, in relation to key transplant outcomes. Among 2,473 kidney transplant recipients (median follow-up 6.9 years), 36% reached a hierarchical composite endpoint of death-censored graft failure(DCGF), biopsy-proven acute rejection(BPAR), or de novo donor-specific antibody(dnDSA). Multivariable analyses showed total, class I, class II, and locus-specific saAA mismatch scores were significantly associated with the composite outcome, as well as DCGF, BPAR and dnDSA individually (all P\u003c.001). These findings suggest that saAA mismatch load provides a more granular and reproducible measure of alloimmune risk compared to traditional antigen-level mismatches. Incorporating saAA mismatch analysis could enable personalized immunosuppression strategies and optimized transplant monitoring in clinical practice.","manuscriptTitle":"Impact of Solvent-Accessible HLA Amino Acid Mismatches on Kidney Transplant Outcomes: A Multicenter Longitudinal Study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-21 09:35:31","doi":"10.21203/rs.3.rs-7115262/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"964a76dc-b868-4017-a6a9-2bcef6a816b0","owner":[],"postedDate":"July 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":51560464,"name":"Health sciences/Biomarkers/Predictive markers"},{"id":51560465,"name":"Biological sciences/Immunology/Transplant immunology"},{"id":51560466,"name":"Health sciences/Medical research/Outcomes research"},{"id":51560467,"name":"Health sciences/Biomarkers/Diagnostic markers"},{"id":51560468,"name":"Health sciences/Medical research/Epidemiology"}],"tags":[],"updatedAt":"2025-07-21T09:35:32+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-21 09:35:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7115262","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7115262","identity":"rs-7115262","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}