Donor Cell-Derived Hematologic Neoplasms: Role of Germline Predisposition and Clonal Hematopoiesis

Introduction

Donor-cell derived hematologic neoplasms (DCHN) are estimated to account for up to 2% of relapses following allogeneic hematopoietic stem cell transplantation (allo-HSCT). However, reported series are scarce and with limited cases, and no systematic registers have been established, hindering the complete understanding of this poor prognosis event. Additionally, it is still unknown the real weight of two phenomena put forward as main potential causes: germline predisposition (GP) and clonal hematopoiesis of indeterminate potential (CHIP). The aim of this study was to estimate the incidence of DCHN in centers performing allo-HSCT in Spain and to depict their clinical and biological data.

Methods

We contacted transplant and molecular biology units from 37 centers conducting allo-HSCT in adults in the last ten years in Spain. Collected variables included clinical and laboratory characterization of the primary neoplasm (PN) and the DCHN, as well as descriptive data of the main peri-transplant events. If available, sequencing data from the primary neoplasm, the DCHN, and the donor was collected.

Results

We registered 18 patients diagnosed of a DCHN from 15 of the 37 centers contacted, becoming the largest series of assembled cases to date. Fourteen cases (77.8%) were diagnosed in the last five years. Primary neoplasm diagnoses showed a predominance of myeloid lineage origin (77.8%) including 11 acute myeloid leukemia (AML) and 3 myelodysplastic syndrome (MDS) cases. In addition, 2 Hodgkin lymphoma (HL), and 1 mantle cell lymphoma (MCL) patients developed a DCHN. In one case, the allo-HSCT was performed due to aplastic anemia (AA). As for the type of DCHN, the majority of cases also developed a myeloid malignancy after the allo-HSCT (83%): seven developed an AML (six from AML, one from HL), and eight MDS (four from AML, three from MDS, one from AA). The three remaining cases were: one mixed-phenotype acute leukemia (from an AML case), one cutaneous marginal zone B-cell lymphoma (from the MCL case), and one cutaneous T-cell lymphoma (from HL).

Notably, most cases had received reduced-intensity conditioning (70,6%). Regarding the type of donor, in 17 of the 18 cases, the donor was related, with a mean age of 36 years (range, 23-59). One donor met the criteria for poor mobilization of hematopoietic progenitors. The median latency period between the allo-HSCT and the DCHN diagnosis was 53 months (range, 16 months to 32 years). Nine patients (50%) died with the DCHN as the main cause. Of the 15 myeloid DCHN cases, 46,7% had abnormal cytogenetics, with 2 cases showing a del(20q).

As to the genomic studies, we had molecular information from ten cases by NGS. Of those, we identified 6 cases with a potentially pathogenic germline variant responsible for the DCHN (DDX41, n=2; CEBPA, n=2; one with SAMD9L and one with a KIT variant). CHIP (acquired mutation VAF≥2% in myeloid driver gene) was not detected in the four cases where NGS sequencing was available in donor DNA before infusion. Finally, nine of the ten patients with DCHN NGS showed the acquisition of pathogenic somatic variants in recognized myeloid neoplasm-associated genes.

Conclusions

From our observational and retrospective results, we conclude: i) the majority of diagnoses are grouped in the last 5 years, which may reflect increased awareness of the phenomenon and/or the proportional increase of familial (haploidentical) allo-HSCTs; ii) the younger age of the donors and the marked majority of cases with related donors in our series suggest a predominance of a germline cause over CHIP transmission, and iii) although limited, the sequencing data also seem to support the presence of pathogenic germline variants in the donor as a relevant factor in the development of DCHN.