Meta-Analysis Reveals Potential Role of Clonal Hematopoiesis on Allogeneic and Autologous SCT Outcomes - Episode 3

Donor Clonal Hematopoiesis May Predict Post-Transplant Relapse and Outcomes in Hematologic Malignancies

Screening for donor-derived clonal hematopoiesis may have prognostic value for outcomes following auto-HSCT in patients with hematologic malignancies.

Donor-derived clonal hematopoiesis was found to have a distinct, bimodal effect on disease relapse and clinical outcomes for patients with hematologic malignancies undergoing allogeneic hematopoietic stem cell transplantation (allo-HSCT) or autologous HSCT (auto-HSCT), according to findings from a comprehensive systemic review and meta-analysis published in Bone Marrow Transplantation.

Although donor-engrafted clonal hematopoiesis prior to allo-HSCT was associated with a significantly decreased risk of disease relapse (HR, 0.79, 95% CI, 0.67-0.93) with low heterogeneity (I² = 0%), it did not significantly affect overall survival (OS; HR, 0.91; 95% CI, 0.75-1.11), progression-free survival (PFS; HR, 0.94; 95% CI, 0.63-1.41), or non-relapse mortality (NRM) after allo-HSCT (HR, 1.06; 95% CI, 0.81-1.39). Sensitivity analyses confirmed the robustness of this finding, and no significant publication bias was detected. Moreover, donor clonal hematopoiesis did not increase the risk of developing acute or chronic graft-versus-host disease (GvHD).

Conversely, pre-transplantation clonal hematopoiesis in the auto-HSCT group was associated with worse outcomes across several efficacy parameters. The risk of inferior OS increased 1.30-fold (HR, 1.30; 95% CI, 1.16-1.46; P < .0001) with low heterogeneity across studies (I² = 0%; P = .8637). This patient population also experienced worse PFS outcomes (HR, 1.35; 95% CI, 1.18-1.54; P < .0001; I² = 8.59%, P = .4516) and a higher risk of developing therapy-related myeloid neoplasms than patients without clonal hematopoiesis undergoing auto-HSCT (HR, 4.85; 95% CI, 2.39-9.82; P < .0001; I² = 0%, P = 0.4347). Notably, results from the sensitivity and publication bias analyses were not shown due to the limited number of studies. These findings indicate that pre-transplant clonal hematopoiesis may be a significant negative prognostic factor in auto-HSCT.

“We [showed] a reduced risk for disease relapse after transplanting clonal hematopoiesis mutation–harboring cells in the setting of allo-HSCT but increased mortality after transplanting clonal hematopoiesis mutation–harboring cells in the setting of auto-HSCT,” study author Shyam A. Patel, MD, PhD, of the University of Massachusetts Chan Medical School in Worcester, and study coauthors, wrote in the journal article. “[Overall], our study suggests utility in screening for donor clonal hematopoiesis in auto-HSCT recipients. This is especially important when considering the possibility that clonal hematopoiesis clones might be transmitted from donor to recipient.”

Background and Statistical Methodology

Clonal hematopoiesis, defined as the clonal expansion of hematopoietic stem cells without evidence of myeloid neoplasms, is a known risk factor for progression to hematologic malignancies, including myelodysplastic neoplasms and acute myeloid leukemia. Clonal hematopoiesis is also linked to a multitude of non-hematologic conditions, signifying its impact on overall population health.

As the number of haploidentical transplants performed continues to rise, assessing the incidence and impact of factors such as donor age and clonal hematopoiesis is increasingly relevant. Conflicting studies report varying outcomes related to donor-derived clonal hematopoiesis, such as potential increased risk for GvHD; however, there is no clear consensus on OS or NRM. Moreover, some findings indicate that mutations associated with clonal hematopoiesis, particularly DNMT3A mutations, might decrease the risk of relapse but elevate the risk of chronic GvHD.

To quantify the effect of clonal hematopoiesis on patient outcomes in allo- and auto-HSCT, investigators performed both a meta-analysis and a meta-regression on 5 allo-HSCT studies involving donor-derived clonal hematopoiesis and 9 auto-HSCT studies involving pre-transplant clonal hematopoiesis. The study selection process included cohort and case-control studies that evaluated the effect of clonal hematopoiesis HSCT outcomes and reported or provided measurable hazard ratios (HRs). Studies were excluded if they shared data, were case reports, or were narrative reviews. Clonal hematopoiesis was defined as the presence of a somatic myeloid mutation with a variant allele frequency (VAF) of at least 0.005. Notably, this threshold is considered lower and more encompassing than the International Consensus Classification’s definition of clonal hematopoiesis of indeterminate potential (CHIP), which has a VAF cutoff of 0.02.

Among the 14 studies that met these inclusion criteria and were included in the final analysis, data were extracted from 3192 donor-recipient pairs for allo-HSCT and 2854 patients for auto-HSCT. Most studies originated from North America (80%), and most were retrospective cohort studies (4 for allo-HSCT; 8 for auto-HSCT). Reviewers conducted independent assessments, and the Newcastle-Ottawa Scale was used to assess study quality, with a median score of 7 (range, 5-8) across studies.

Subgroup Analyses

Subgroup analyses assessing the effects of different VAF cutoffs on clinical outcomes in the allo-HSCT group revealed that VAF was not a significant source of heterogeneity for most outcomes, such as GvHD, NRM, and OS. However, VAF thresholds were more likely to influence relapse risk, with lower VAF showing greater prognostic significance. Patients with a VAF of less than 0.02 had a significantly reduced risk of relapse (HR, 0.77; 95% CI, 0.63-0.94, P = 0.0119). In contrast, those with a VAF of at least 0.02 did not experience a higher risk of relapse (HR, 0.83; 95% CI, 0.48-1.43, P = 0.5086). No statistically significant effect of donor clonal hematopoiesis on neutrophil or platelet engraftment was observed.

Further subgroup analyses and meta-regression for pre-transplant clonal hematopoiesis in auto-HSCT explored OS outcomes across variables such as geographic area, VAF cutoff, and a primary diagnosis of lymphoma vs multiple myeloma. Geographic area was a potential source of heterogeneity, with OS remaining statistically significant in studies conducted in North America but not in Europe. Differences in VAF cutoffs and primary diagnoses were not major contributors to heterogeneity, as pooled HRs remained consistently greater than 1.

Meta-regression confirmed that there was no significant difference in outcomes when using traditional CHIP criteria (VAF ≥ 0.02) vs lower VAF thresholds (HR ratio for OS, 1.18; 95% CI, 0.90-1.55; P = 0.2406).

“An explanation for the observed effect in allo-HSCT may be related to immunological intolerance between donor and recipient, with consequent augmentation of graft-vs-leukemia effect (thus decreased risk for relapse),” the study authors proposed. “In auto-HSCT, on the other hand, there may be functional dominance of clonal hematopoietic–mutant hematopoietic stem cells (HSCs) compared with wild-type HSCs within the bone marrow microenvironment, as the former may successfully outcompete the latter for niche occupancy, leading to inferior outcomes.”

Study Limitations, Next Steps, and Unanswered Questions

The debate surrounding screening for clonal hematopoiesis centers around whether the potential benefits—such as selecting a donor with optimal genomic integrity and improving recipient outcomes—outweigh the potential drawbacks—which include the absence of management recommendations for pathogenic mutations, the high cost of screening, and possible delays in donor selection—for a given patient. The relationship between clonal hematopoiesis and allo-HSCT outcomes indicates a reduced risk of relapse in the presence of donor clonal hematopoiesis, highlighting its potential positive effect on hematologic malignancy recurrence. Longitudinal studies are needed to better understand the broader implications of donor clonal hematopoiesis on recipient health, particularly the risk of non-hematologic complications post-transplant.

The authors noted that key study limitations include the limited primary literature on donor clonal hematopoiesis and its clinical implications; the variability in clonal hematopoiesis (CH) detection methods across centers; the lack of standardized VAF cutoffs for clonal hematopoiesis; and the inclusion of both allo-HSCT and auto-HSCT, which involve different diseases and treatment regimens. Accordingly, this research would benefit from more large-scale, geographically diverse studies, including randomized phase 3 trials comparing HSCT outcomes from clonal hematopoietic–mutant vs wild-type donors.

Investigators noted that evaluation of pre-transplant clonal hematopoiesis may be used to identify higher-risk patients, potentially guiding alternative treatment options or more intensive surveillance. Mutations in DNA repair genes like PPM1D and TP53 have also been linked to worse outcomes in CH-positive patients, indicating a need for further studies to clarify mutation-specific risks and mitigation strategies.

“Our study merits further consideration of the exact risk that donor clonal hematopoiesis poses on a patient-by-patient basis, including genotype- specific risk stratification” the authors concluded. “Considerations for the future include comprehensive analysis of donor stem cell products for recurrent somatic mutations by next-generation sequencing. Such personalized approaches to donor selection may allow for risk mitigation.”

Reference

Xie Y, Kazakova V, Weeks LD, et al. Effects of donor-engrafted clonal hematopoiesis in allogeneic and autologous stem cell transplantation: a systematic review and meta-analysis. Bone Marrow Transplant. Published online August 25, 2024. doi:10.1038/s41409-024-02403-2