Haplo, we have a problem

In this issue of Blood, Ahci and colleagues describe a simple and robust method for identifying the phenomenon of leukemic HLA loss often associated with treatment failure of hematopoietic stem cell transplantation from haploidentical donors (haplo-HSCT).¹ 

Hematopoietic stem cell transplantation (HSCT) is potentially curative therapy for both acquired malignant and inherited nonmalignant hematologic disease.²  The lack of an appropriate donor has historically limited the applicability of HSCT, particularly for those not of white European ancestry for whom complete HLA matches are less likely to be found in unrelated donor and cord bank registries.³  The development of approaches to allow the safer use of HLA mismatched hematopoietic stem cell donors⁴  has greatly improved the feasibility of HSCT for all patients, with a marked increase in the number of haplo-HSCTs being performed over the past decade.⁵  Unfortunately for those with malignant disease, in particular the most common haplo-HSCT indication of acute leukemia, disease persistence and relapse after transplantation remain the most common reasons for transplant failure.⁶  The method described by Ahci et al should allow better understanding of the contribution of leukemic HLA loss to such failures.

Multiple factors associated with leukemic relapse after HSCT have been described, including measurable residual disease (MRD) burden,^7,8  conditioning intensity,⁹  and the genomic loss of the mismatched HLA haplotype in residual leukemia after haplo-HSCT.¹⁰  The relative contribution of these factors in the etiology of post-HSCT relapse remains undefined, but the former 2 components are easily quantifiable. Although multiple anecdotal reports of the latter phenomenon have since been described by many centers, no systematic large-scale survey of mismatched HLA stability after haplo-HSCT and association with leukemic relapse has been performed, likely in part because of the lack of an easy-to-use, reproducible, sensitive, and inexpensive measurement tool. The article by Ahci et al sought to address this deficiency by developing quantitative polymerase chain reaction (qPCR) tests specific for the most frequent HLA allele groups that can be used as an adjunct to conventional qPCR chimerism testing (ie, patient-specific non-HLA gene polymorphisms) to detect leukemic HLA loss. They show the ability to discriminate between classic relapse (ie, relapse of leukemia without loss of mismatched HLA) and leukemic HLA-loss relapse without the need to first purify relapsing leukemic blasts for genomic analysis.

This straightforward method requires only modest modification to the existing qPCR workflow for post-HSCT chimerism analysis at the time of leukemic relapse and opens up the possibility of a large multicenter census of the frequency of HLA loss at the time of leukemic relapse. The optimal treatment of HLA-loss leukemic relapse after haplo-HSCT has not been established. It is reasonable to suspect that donor lymphocyte infusion and other immunotherapeutic approaches may be less efficacious in this setting, but clinical trials are needed to test this hypothesis, and this rapid assessment of HLA loss may form the basis for trial inclusion criteria. Finally, it is likely that this qPCR method, along with qPCR assays to detect post-HSCT chimerism and acute leukemia MRD, will be superseded in the near future by next-generation sequencing incorporating error correction approaches. For the moment, however, this straightforward, easily adoptable and economical tool will allow quantification of the frequency of leukemic HLA loss at the time of the main problem of haplo-HSCT for acute leukemia, which is relapse.