PerspectiveIn the Journals Plus

Stem cell source affects graft-versus-host disease occurrence for leukemia subtypes

Show Citation

March 7, 2018

Peripheral blood stem cell grafts appeared associated with increased risk for acute graft-versus-host disease compared with bone marrow among patients undergoing hematopoietic stem cell transplantation for acute myeloid leukemia and acute lymphoblastic leukemia, according to a study published in Cancer.

However, there were no observed differences in OS or other transplant-related outcomes, leaving unanswered the question of which graft source is superior for use in haploidentical transplantation.

“The role of stem cell source in the setting of related or unrelated donor transplant ... and a myeloablative conditioning regimen has been evaluated in randomized trials, showing an excess of chronic graft-versus-host disease with peripheral blood stem cells as the stem cell source, with no differences in disease-free and overall survival,” Annalisa Ruggeri, MD, PhD, of the Hôpital Saint Antoine in Paris, and colleagues wrote. “The number of unmanipulated haploidentical stem cell transplantations in adult patients with hematological malignancies such as acute myeloid leukemia and acute lymphoblastic leukemia is increasing constantly.”

To assess the effects of stem cell source, researchers studied 451 patients (median age, 45 years) in first or second complete remission at transplantation, who underwent haploidentical HSCT with posttransplantation cyclophosphamide reported to the European Society for Blood and Marrow Transplantation registry from 2010 to 2014.

As a stem cell source, 260 patients received bone marrow and 191 received peripheral blood stem cells.

Median follow-up was 22.8 months for bone marrow and 18.3 months for peripheral blood stem cells. Patients receiving bone marrow were more likely to undergo myeloablative conditioning (61% vs. 49%; P = .008).

Ninety-two percent of patients receiving bone marrow had engraftment compared with 95% of those receiving peripheral blood stem cells (P < .001). Time to engraftment was longer in the bone marrow group (18 days vs. 17 days; P < .001).

The primary endpoint of 2-year leukemia-free survival was 49% for bone marrow compared with 54% for peripheral blood stem cells. Researchers reported 2-year OS of 55% for bone marrow and 56% for peripheral blood stem cells.

Peripheral blood stem cells appeared associated with an increased risk for stage II to stage IV acute GVHD (HR = 2.1; 95% CI, 1.46-3) and stage III to stage IV GVHD (HR = 3.8; 95% CI, 1.7-8.2).

Researchers did not observe associations between stem cell source and chronic GVHD, relapse or nonrelapse mortality.

“With the available data, our study indicates that [among] patients with acute leukemia in complete remission 1 or complete remission 2 who underwent haploidentical transplantation with posttransplantation cyclophosphamide, the use of peripheral blood stem cells significantly increased the risk [for] acute GVHD, whereas survival outcomes were comparable,” the researchers wrote. “Importantly, with a follow-up of 2 years, chronic GVHD, which is a major contributor to long-term morbidity and mortality, is similar using [peripheral blood stem cells] or [bone marrow] grafts.” – by Cassie Homer

Disclosures: The authors report no relevant financial disclosures.

itj+ Perspective

Ibrahim Aldoss

Ibrahim Aldoss

The ideal source of graft for haploidentical stem cell transplant (haplo-SCT) with posttransplant cyclophosphamide remains undefined. The decision likely will be influenced by disease risk, conditioning intensity, and patient and donor characteristics.

In a matched-pair analysis that compared graft source in nonablative haplo-SCT, peripheral blood appeared associated with lower risk for relapse than bone marrow, which, in turn, translated to better DFS. However, there were no noticeable differences in other transplant-related outcomes.

In a larger registry study that utilized the Center for International Blood and Marrow Transplant Research database (n = 681) and included various hematologic malignancies and conditioning regimens, haplo-SCT using bone marrow graft appeared associated with lower risk for acute and chronic GVHD and higher risk for relapse compared with peripheral blood. However, neither OS nor nonrelapse mortality were affected by graft source. Notably, relapse risk appeared higher with bone marrow graft among patients transplanted for leukemia diagnosis but not for lymphoma.

More recently, Ruggeri and colleagues evaluated European Society for Blood and Marrow Transplantation registry data to compare haplo-SCT outcomes based on graft source with analysis restricted to patients with acute leukemia in first and second complete remission (n = 450).

The study demonstrated comparable leukemia-free survival, OS, nonrelapse mortality and relapse rate between the two cohorts. Recipients of peripheral blood grafts had higher risk for acute GVHD but no difference in chronic GVHD. Engraftment was faster — by 1 day — and higher (3% absolute difference) in the peripheral blood cohort. Nonetheless, the use of reduced-intensity conditioning for patients included in the study was associated with higher relapse rate and resulted in inferior OS and leukemia-free survival.

The authors concluded that, although survival outcomes are comparable between peripheral blood and bone marrow grafts following haplo-SCT, peripheral blood grafts lead to higher risk for acute GVHD.

This study supports the feasibility and activity of peripheral blood grafts in haplo-SCT and illustrates that outcomes are comparable to bone marrow overall; however, the two arms were imbalanced due to the higher frequency of reduced-intensity conditioning recipients in the peripheral blood cohort. This imbalance is unsurprising given the retrospective nature of the study, but it also may have influenced the overall outcomes. Therefore, the question regarding which graft source is superior for use in haplo-SCT remains unanswered and warrants a prospective study to address this in a more definitive manner. Considering the steadily increasing number of haplo-SCTs being performed, identifying which graft source is preferred for a particular patient subset is an important need. Only a carefully designed study that balances crucial patient, disease and transplant factors and includes prespecified subanalyses will yield a reliable answer.


Bashey A, et al. J Clin Oncol. 2017;doi:10.1200/JCO.2017.72.8428.

O’Donnell PV, et al. Bone Marrow Transplant. 2016;doi:10.1038/bmt.2016.215.

Ibrahim Aldoss, MD

HemOnc Today Next Gen Innovator
City of Hope

Disclosures: Aldoss reports no relevant financial disclosures.