T-Cell Depleted Allografts Lessen GVHD in MDS

Treatment with T-cell depleted transplantation was associated with a lower incidence of acute graft versus host disease (GVHD) and a very low incidence of chronic GVHD compared with unmodified allografts in patients with advanced myelodysplastic syndrome.

Roni Tamari, MD

Treatment with T-cell depleted transplantation was associated with a lower incidence of acute graft versus host disease (GVHD) and a very low incidence of chronic GVHD compared with unmodified allografts in patients with advanced myelodysplastic syndrome (MDS), according to findings presented at the 2015 International MDS Symposium.

“By multivariate analysis, we had similar relapsed-free and overall survival between the treatment two groups,” the lead study author Roni Tamari, MD, said during her presentation. The two groups also demonstrated similar efficacy.

“However, despite having over 25% of patients in the T-cell depleted group with a mismatched donor, acute GHVD at day 100 was significantly lower in the T-cell depleted group compared with the unmodified group, with a much greater difference in chronic GVHD of 3.5% compared with 59%.”

In the retrospective study, 60 patients received T-cell depleted transplants at Memorial Sloan Kettering Cancer Center (MSK) along with myeloablative conditioning (MAC) and antithymocyte globulin (ATG). In the comparator arm, 129 patients were treated with unmodified transplants at MD Anderson Cancer Center, 87 of whom received MAC and 42 received reduced intensity conditioning. The majority of patients in this arm (n = 113) received tacrolimus and mini-dose methotrexate for GVHD prophylaxis.

T cells were depleted from peripheral blood stem cell allografts using CD34+ cell selection with the Isolex system followed by sheep red blood cells (sRBC) rosetting (n = 30) and CliniMACS (n = 25). In 5 patients, T-cell depletion was completed on bone marrow grafts by sequential soybean lectin agglutination and sRBC-rosette depletion. These patients did not receive further GVDH prophylaxis.

It was determined by univariate analysis that non-relapse mortality rates were similar in both groups. Overall, the 1-year relapse rates were 8.5% versus 31% and the 3-year rates were 15.5% compared with 39.4%, with the T-cell depleted and unmodified allografts, respectively.

Given the retrospective dual institution patient selection, there was a clear dichotomy in patient characteristics between the two arms. There were more patients in the unmodified versus modified arms with very poor cytogenetics at diagnosis (36.4% vs 13.6%, respectively) and blast counts at transplant between 10% and 19% (25.2% vs 0%).

In those with only good/intermediate risk cytogenetics and less than 10% blasts at transplant, the relapse incidence was similar between the two arms. The 3-year cumulative incidence (CI) of relapse in the T-cell depleted arm was 7.9% compared with 18% in the unmodified group (P = .185).

Other parameters demonstrated discordance between the two groups. In the unmodified arm, 33.3% of patients had therapy-related MDS compared with 11.7% in the T cell depleted arm. Fourteen patients (23.3%) in the T-cell depleted arm had mismatched unrelated donors compared with 0 in the unmodified arm.

"Based on these disease characteristics, of course the univariate analysis favored the T-cell depleted group for relapse, survival, and relapse-free survival," said Tamari, hematologist/oncologist in the Bone Marrow Transplant Service at MSK. "We went on to do a multivariate analysis accounting for type of transplant, disease etiology, blast cells at transplant, and cytogenetic risk."

In multivariate regression models, no significant differences were identified between the two treatment groups for relapse-free survival (RFS) and overall survival (OS). The hazard ratio (HR) for RFS was 1.44 (P = .128) and the HR for OS was 1.35 (P = .236).

“In this analysis, the groups were much more balanced,” Tamari said. “The only predictors for RFS and OS were high-risk cytogenetics, which indicated poorer outcomes in both arms.”

The 100-day CI of grade II-IV acute GVHD in the T-cell depleted group was 12.8% compared with 35.8% in the unmodified group (P = .031). Rates of grade III/IV acute GVHD were similar between the two groups at 100 days (P = .546). The 3-year CI of chronic GVHD was just 3.5% in the T-cell depleted arm compared with 59% in the unmodified group (P <.001).

However, advantages in GVHD were countered by higher rates of infection in the T-cell depleted group. The most common causes of death in the T-cell depleted group were infection (32%) and relapse (28%). In the unmodified group, the most common causes of death were relapse (55%), GVHD (20%), and infections (13%).

Findings from the study suggest that T-cell depleted transplant could have significant advantages over traditional unmodified allografts, in terms of quality of life. Results from the dual institution study are being confirmed in a larger prospective study, Tamari noted.

"The primary endpoint of this study will be a composite of RFS and GVHD-free survival, which looks at whether patients are alive without GVHD," Tamari added. "We did not do that analysis in this study, but since the patients had similar survival with less GVHD this endpoint would have been significantly better in patients who received T-cell depleted transplant."

The 3-arm phase III study known as BMT CTN 1301 or PROGRESS II is exploring unmodified bone marrow grafts with tacrolimus plus methotrexate (control) in comparison with bone marrow graft followed by posttransplant cyclophosphamide or CD34-selected T-cell depleted peripheral blood stem cell grafts. The study, which began in January 2015, plans to enroll 345 patients with MDS or acute leukemia (NCT02345850).

Tamari R, Oran B, Hilden P, et al. Outcomes in patients with advanced myelodysplastic syndrome receiving ex-vivo T-cell depleted or unmodified allografts: comparison of results at two institutions. Leuk Res. 2015;39:1s (suppl; abstr 16).


View more from the 2015 MDS Symposium