TBI based VS chemotherapy based myeloablative conditioning in T cell replete haploidentical transplant Free

Introduction: Limited data compare total body irradiation (TBI)-based versus chemotherapy-based myeloablative conditioning (MAC) in haploidentical transplants. An EBMT study showed similar major transplant outcomes, but TBI was linked to a higher risk of severe chronic GVHD in multivariable analysis.

Methods: We combined two CIBMTR datasets: one comparing myeloablative vs. reduced-intensity T-cell–replete haploidentical transplantation (PMID: 31582392), and the other evaluating optimal donor sources for African American patients with hematologic malignancies (PMID: 32649981). Both datasets included patients underwent haploidentical transplantation between 2008 and 2016. We excluded patients who received nonmyeloablative or reduced-intensity conditioning, and those received umbilical cord grafts. All patients received post-transplant cyclophosphamide, a calcineurin inhibitor, and mycophenolate mofetil for GVHD prophylaxis. Patients received ≥12 Gy TBI were classified as TBI-based MAC; all others receiving MAC were classified as chemotherapy-based (CT)-MAC. We calculated the cumulative incidence (CI) of relapse, non-relapse mortality (NRM), and grade II–IV acute GVHD up to 40 months post-transplant. CI of relapse was estimated using death as a competing risk, and CI of NRM was calculated using relapse as a competing risk. CI of acute and chronic GVHD was calculated using death as the competing risk.

Both groups were matched on key confounders using inverse probability of treatment weighting (IPTW), including age, sex, KPS (>90 vs. ≤90), disease type (AML vs. ALL), disease risk (low, intermediate, high/very high), disease status at transplant (CR vs. refractory), transplant year (2008–2013 vs. 2013–2016), donor CMV status, HCT-CI (0, 1, 2, ≥3), and graft source (peripheral blood vs. bone marrow). All analyses were performed in R version 4.4.0.

Results: A total of 629 patients were included. The median age was 45 years (range 18–70). Of these, 353 (56%) received TBI-based MAC and 276 (44%) received CT-based MAC. A total of 319 patients (52%) had a KPS ≥90, and 329 (52%) were male. Compared to the CT-based group, patients received TBI-based MAC were more likely to receive peripheral blood grafts (76% vs. 58%, P<0.01), have acute lymphoblastic leukemia (43% vs. 13%, P<0.01), and received transplant between 2013–2016 (89% vs. 77%, P<0.01). There were no significant differences between groups in disease risk, KPS, HCT-CI score, or disease status at transplant. Median follow-up was 24.6 months (95% CI, 24.3–25.1) overall, 24.4 months (95% CI, 24.1–25.0) in the TBI group, and 24.7 months (95% CI, 24.3–34.0) in the CT group.

The CI of relapse was 38% in the TBI group versus 33% in the CT group (P=0.73). NRM was 19% for TBI versus 24% for CT (P=0.13). The incidence of grade II–IV acute GVHD and chronic GVHD were significantly higher in the TBI group VS CT group (36% vs. 21%, P<0.01 and 34% VS 25%, P value=0.03 respectively).

After propensity score matching using IPTW, there was no significant difference in overall survival (OS) or relapse-free survival (RFS) between the TBI and CT groups. Median OS was not reached (NR) for the TBI group (95% CI, 23.3–NR) and 27.8 months for the CT group (95% CI, 17.4–60.0; P=0.15). Median RFS was NR for both groups, with 95% CI of 29.9–NR for TBI and 43.5–NR for CT (P=0.7). Subgroup analyses by disease risk (high/very high) and disease type (AML vs. ALL) also showed no differences in OS or RFS. Similarly, there were no significant differences in GVHD outcomes after matching. Median time to grade II–IV acute GVHD was 33.5 months (95% CI, 15.6–NR) in the TBI group versus 31.5 months (95% CI, 16.4–NR) in the CT group (Weighted P value=0.36). Median time to chronic GVHD was 15.8 months (95% CI, 9.87–NR) in the TBI group versus NR (95% CI, 10.5–NR) in the CT group (Weighted P value=0.95).

Conclusion: In this combined CIBMTR analysis of patients undergoing T-cell–replete haploidentical transplantation, there were no significant differences in OS, RFS, or NRM between TBI-based and chemotherapy-based MAC. TBI-based MAC was associated with a higher incidence of grade II–IV acute GVHD and chronic GVHD. The observed increase in acute and chronic GVHD with TBI may be partially driven by a higher use of peripheral blood grafts in this group. Overall, both conditioning strategies appear to offer comparable long-term outcomes, with graft source and GVHD risk requiring careful consideration in regimen selection.