Impact of the HLA Immunopeptidome on Survival of Leukemia Patients After Unrelated Donor Transplantation

Human Leukocyte Antigen Mismatching and Survival in Contemporary Hematopoietic Cell Transplantation for Hematologic Malignancies

PURPOSE

Human leukocyte antigen (HLA) mismatching can reduce survival of patients with blood cancer after hematopoietic cell transplantation (HCT). How recent advances in HCT practice, in particular graft-versus-host disease (GVHD) prophylaxis by post-transplantation cyclophosphamide (PTCy), influence HLA risk associations is unknown.

PATIENTS AND METHODS

The study included 17,292 unrelated HCTs with 6-locus high-resolution HLA typing, performed mainly for acute leukemia or related myeloid neoplasms between 2016 and 2020, including 1,523 transplants with PTCy. HLA risk associations were evaluated by multivariable Cox regression models, with overall survival (OS) as primary end point.

RESULTS

OS was lower in HLA mismatched compared with fully matched transplants (hazard ratio [HR], 1.23 [99% CI, 1.14 to 1.33]; P < .001). This was driven by class I HLA-A, HLA-B, HLA-C (HR, 1.29 [99% CI, 1.19 to 1.41]; P < .001) but not class II HLA-DRB1 and HLA-DQB1 (HR, 1.07 [99% CI, 0.93 to 1.23]; P = .19). Class I antigen-level mismatches were associated with worse OS than allele-level mismatches (HR, 1.36 [99% CI, 1.24 to 1.49]; P < .001), as were class I graft-versus-host peptide-binding motif (PBM) mismatches compared with matches (HR, 1.42 [99% CI, 1.28 to 1.59]; P < .001). The use of PTCy improved GVHD, relapse-free survival compared with conventional prophylaxis in HLA-matched transplants (HR, 0.77 [0.66 to 0.9]; P < .001). HLA mismatching increased mortality in PTCy transplants (HR, 1.32 [1.04 to 1.68]; P = .003) similarly as in non-PTCy transplants (interaction P = .43).

CONCLUSION

Class I but not class II HLA mismatches, especially at the antigen and PBM level, are associated with inferior survival in contemporary unrelated HCT. These effects are not significantly different between non-PTCy compared with PTCy transplants. Optimized HLA matching should still be considered in modern HCT.

Donor selection in allogeneic stem cell transplantation

PURPOSE OF THE REVIEW

Recent progress in human leukocyte antigen (HLA) characterization, increased accrual of unrelated donors and cord blood units, and a new platform for haploidentical transplantation have resulted in the widespread availability of donors for allogeneic hematopoietic stem cell transplantation.

RECENT FINDINGS

Advances in HLA typing have identified an increasing number of loci and alleles that are crucial for successful transplantation. Newer HLA A, B, C, DRB1, and DQB1 alleles, DPB1 mismatches, and HLA B leader sequence matching are incorporated into donor selection algorithms. Donor selection is highly relevant because of recently published conflicting studies using different donor types. These studies are largely retrospective and compare patients with different diseases and stages, conditioning regimens, graft versus host disease (GVHD) prophylaxis, and time periods. A broad consensus indicates that the best donor is an available matched sibling, followed by a matched unrelated donor, and then alternative donors such as haploidentical, mismatched unrelated, and cord blood units. This consensus is being challenged by other factors, such as donor age, patient condition, urgency of transplantation, and costs involved.

SUMMARY

In this review, we will analyze the unique characteristics of each donor type, the HLA and non HLA factors that affect donor choices, and the outstanding comparative outcome studies of different donor usage in hematologic malignancies.

Haploidentical Versus Mismatched Unrelated Donor Hematopoietic Cell Transplantation: HLA Factors and Donor Age Considerations

HLA-mismatched unrelated donors and haploidentical related donors are suitable stem cell sources for hematopoietic cell transplantation (HCT) when patients lack HLA-matched donors. Clinical outcome after mismatched HCT is influenced by HLA factors including the similarity of peptide-binding motifs (PBMs) between the patient and unrelated donor, and of the HLA-B leader in unrelated and haploidentical donors. Whether these factors can aid in the selection between mismatched unrelated and haploidentical donors is not known. To address this question, we investigated outcomes between the two donor types defined by matching for the PBM and leader peptide. We compared PBM-matched (n = 614) and mismatched (n = 958) MMUDs with calcineurin-inhibitor-based prophylaxis to four haploidentical groups that received post-transplant cyclophosphamide (PTCy)-based prophylaxis. The haploidentical groups were B-leader matched/DRB1-mismatched (n = 722), B-leader matched/DRB1-matched (n = 154), B-leader mismatched/DRB1-mismatched (n = 493), and B-leader mismatched/DRB1-matched (n = 63). Multivariate analysis showed that the B-leader matched/DRB1-mismatched haploidentical group had the best overall survival (OS) compared to the PBM-matched MMUD, while other haploidentical groups had comparable OS. The PBM-mismatched MMUD showed the poorest outcomes, similar to the B-leader mismatched/DRB1-matched haploidentical group. Among non-HLA factors, donor age was the most significant predictor of OS. These results suggest that a B-leader matched/DRB1 mismatched haploidentical donor might be the preferred choice among donors of similar age. If such a donor is not available, the youngest donor from either PBM-matched unrelated or other haploidentical groups could be a beneficial choice. These findings need validation with both donor groups receiving PTCy-based graft-versus-host disease prophylaxis.

Effect of peptide-binding motif on survival of HLA-haploidentical transplantation with post-transplant cyclophosphamide

Peptide-binding motif (PBM) model, a hierarchical clustering of HLA class I based on their binding specificity, was developed to predict immunopeptidome divergence. The effect of PBM mismatches on outcomes is unknown in HLA-haploidentical haematopoietic cell transplantation with post-transplant cyclophosphamide (PTCy-haplo). We therefore conducted a retrospective study using national registry data in PTCy-haplo. Overall, 1352 patients were included in the study. PBM-A bidirectional mismatch was associated with an increased risk of overall mortality in multivariable analysis (hazard ratio, 1.26; 95% confidence interval, 1.06 to 1.50; p = 0.010). None of relapse, non-relapse mortality (NRM) and graft-versus-host disease showed significant differences according to PBM-A bidirectional mismatch status in the entire cohort. The impact of PBM-A bidirectional mismatch on overall survival (OS) was preserved within the HLA-A genotype bidirectional mismatch population, and their lower OS stemmed from higher relapse rate in this population. The worse OS due to high NRM with PBM-A bidirectional mismatch was prominent in lymphoid malignancies receiving reduced-intensity conditioning. The PBM model may predict outcomes more accurately than HLA genotype mismatches. In conclusion, this study demonstrated that the presence of PBM-A bidirectional mismatch elevated the risk of mortality of PTCy-haplo. Avoiding PBM-A bidirectional mismatch might achieve better outcomes in PTCy-haplo.

Younger unrelated donors may be preferable over HLA match in the PTCy era: a study from the ALWP of the EBMT

ABSTRACT

There is a paucity of information on how to select the most appropriate unrelated donor (UD) in hematopoietic stem cell transplantation (HSCT) using posttransplant cyclophosphamide (PTCy). We retrospectively analyzed the characteristics of 10/10 matched UDs (MUDs) and 9/10 mismatched UDs (MMUDs) that may affect transplant outcomes in patients with acute myeloid leukemia (AML) in first or second complete remission (CR1 or CR2). The primary end point was leukemia-free survival (LFS). Overall, 1011 patients were included with a median age of 54 years (range, 18-77). Donors had a median age of 29 years (range, 18-64); 304 (30%) were females, of which 150 (15% of the whole group) were donors to male recipients, and 621 (61%) were MUDs; 522 (52%) had negative cytomegalovirus (CMV-neg) serostatus, of which 189 (19%) were used for CMV-neg recipients. Donor age older than 30 years had a negative impact on relapse (hazard ratio [HR], 1.38; 95% confidence interval [CI], 1.06-1.8), LFS (HR, 1.4; 95% CI, 1.12-1.74), overall survival (HR 1.45; 95% CI, 1.14-1.85) and graft-versus-host disease (GVHD) free, relapse-free survival (HR, 1.29; 95% CI, 1.07-1.56). In addition, CMV-neg donors for CMV-neg recipients were associated with improved LFS (HR, 0.74; 95% CI, 0.55-0.99). The use of MMUD and female donors for male recipients did not significantly impact any transplant outcomes. For patients undergoing HSCT from a UD with PTCy for AML, donor age <30 years significantly improves survival. In this context, donor age might be prioritized over HLA match considerations. In addition, CMV-neg donors are preferable for CMV-neg recipients. However, further research is needed to validate and refine these recommendations.

HLA typing: A review of methodologies and clinical impact on haematopoietic cell transplantation

The importance of the HLA gene system in haematopoietic cell transplant outcomes was established early on and advances in both fields have led to ever increasing success of this clinical therapy. In large part, improvements in the understanding of HLA have been driven by the advancement in typing technologies. Each iteration of typing technology has improved the resolution of HLA typing, and often enabled the identification of polymorphism within the HLA loci. The discovery of the enormous amount of variation in the HLA genes, and the need to be able to characterise this for clinical HLA typing, has often resulted in a move away from one typing method to another more suited to typing of this complexity. Today, the gold standard for HLA typing are methods that can produce definitive HLA typing results.

HLA-DM and HLA-DO interplay for the peptide editing of HLA class II in healthy tissues and leukemia

HLA class II antigen presentation is modulated by the activity of the peptide editor HLA-DM and its antagonist HLA-DO, with their interplay controlling the peptide repertoires presented by normal and malignant cells. The role of these molecules in allogeneic hematopoietic cell transplantation (alloHCT) is poorly investigated. Balanced expression of HLA-DM and HLA-DO can influence the presentation of leukemia-associated antigens and peptides targeted by alloreactive T cells, therefore affecting both anti-leukemia immunity and the potential onset of Graft versus Host Disease. We leveraged on a large collection of bulk and single cell RNA sequencing data, available at different repositories, to comprehensively review the level and distribution of HLA-DM and HLA-DO in different cell types and tissues of the human body. The resulting expression atlas will help future investigations aiming to dissect the dual role of HLA class II peptide editing in alloHCT, and their potential impact on its clinical outcome.

SOHO State of the Art Updates and Next Questions | Current Status and Future Directions of Donor Selection

The landscape of HLA matching in hematopoietic cell transplantation (HCT) is continuously advancing, introducing more nuanced criteria beyond traditional 10/10 HLA-A, -B, -C, and -DRB1 allele matching. For 10/10 matched donors, prioritizing a donor with a "core" permissive HLA-DPB1 mismatch is recommended over "noncore" permissive mismatches, with nonpermissive mismatches being the least prefered. In the one-antigen mismatched setting (7/8 HLA-matched), HLA-C matching, particularly avoiding high-expression mismatches at residues 116 or 77/80, is preferred over HLA-A or HLA-B mismatches. HLA B-leader matching is beneficial in both one-antigen mismatched and haploidentical HCT. Additionally, specific HLA mismatches in haploidentical HCT, such as DRB1 mismatches with DQB1 matches and DPB1 nonpermissive mismatches are linked to better outcomes. Among non-HLA factors, evidence consistently underscores the pivotal impact of donor age on overall survival. For HLA-mismatched transplants, including haploidentical HCT, avoidance of donors against whom the recipient has preformed donor-specific antibodies is paramount. Selecting a cytomegalovirus (CMV) seronegative donor is important particularly for CMV-negative recipients; however, more research is needed in the letermovir prophylaxis era. The impact of ABO-matching on transplant outcomes is debatable. Other unanswered questions include defining "younger" donors and establishing hierarchy in donor selection based on factors like CMV status, ABO compatibility, or sex-mismatch, to name a few. Future research addressing these issues will refine donor selection algorithms and improve transplant success. In conclusion, selecting a donor for HCT requires multifaceted considerations, integrating evolving HLA-matching criteria and non-HLA factors, to optimize HCT outcomes in this rapidly advancing field.

Combined effect of unrelated donor age and HLA peptide-binding motif match status on HCT outcomes

ABSTRACT

An HLA-mismatched unrelated donor who is class I peptide-binding motif (PBM)-matched is preferred over a PBM-mismatched donor. We hypothesized that using a younger donor (aged ≤35 years vs >35 years) could compensate for the inferior overall survival (OS) associated with PBM mismatches. We compared 6 groups: HLA-matched/younger donor (n = 10 531), HLA-matched/older donor (n = 3572), PBM-matched/younger donor (n = 357), PBM-matched/older donor (n = 257), PBM-mismatched/younger donor (n = 616), and PBM-mismatched/older donor (n = 339) in patients undergoing transplantation with conventional graft-versus-host disease prophylaxis. In multivariate analysis, HLA-matched/younger donors were associated with superior OS relative to any other group. Pairwise comparisons showed that donor age significantly impacted OS in both HLA-matched and HLA-mismatched groups. Moreover, younger donors appeared to negate the detrimental effect of PBM mismatching: the PBM-matched/younger donor group had similar OS as the HLA-matched/older donor group and the PBM-mismatched/younger donor group had similar OS as the PBM-matched/older donor group. Our study suggests that older unrelated donor age and PBM mismatching confer similarly adverse effects on OS and the impacts are additive, a finding which may widen the "acceptable" donor pool. The best OS is observed with HLA-matched/younger donors and the worst with PBM-mismatched/older donors. These findings should be validated with other data sets and with posttransplantation cyclophosphamide-based prophylaxis.

HLA peptide-binding pocket diversity modulates immunological complications after cord blood transplant in acute leukaemia

Pocket motifs and their amino acid positions of HLA molecules are known to govern antigen presentation to effector cells. Our objective was to analyse their influence on the risk of graft-versus-host disease (GVHD) and relapse after umbilical cord blood transplant (UCBT). The transplant characteristics of 849 patients with acute leukaemia were obtained from the Eurocord/EBMT database. Higher acute (a) GVHD was associated with homozygosity of UCB HLA-C amino acid positions 77 and 80 (NN/KK) (p = 0.008). Severe aGVHD was associated with HLA-A pocket B YSAVMENVHY motif (p = 0.002) and NN and RR genotypes of the HLA-C amino acid positions 77 and 156 (p = 0.006 and p = 0.002). Such risk was also increased in case of recipient and UCB mismatches in P4 (p < 0.0001) and P9 (p = 0.003) pockets of HLA-DQB1 alleles. For chronic GVHD, the pocket B YYAVMEISNY motif of the HLA-B*15:01 allele and the absence of mismatch between recipient and UCB in the P6 pocket of HLA-DRB1 were associated with a lower risk (p = 0.0007 and p = 0.0004). In relapse, both UCB pocket B YFAVMENVHY belonging to HLA-A*32:01 and recipient pocket B YDSVGENYQY motif of the HLA-C*07:01 allele were associated with higher risk (p = 0.0026 and p = 0.015). We provide clues on HLA-mediated cellular interactions and their role in the development of GVHD and relapse.