AUTHORS
Joseph Elie Maakaron, MD, Melhem Solh, MD2, Scott R Solomon, MD3, Lizamarie Bachier-Rodriguez, MD4, H Kent Holland5, Hong De Sa, MD6, Katelin Jackson7, Xu Zhang, PhD8, Asad Bashey, MD, PhD9
AFFILIATIONS
1 Northside Hospital, Atlanta, Georgia, United States
2 Northside Hospital, Atlanta, Georgia, United States
3 Northside Hospital, Atlanta, Georgia, United States
4 Northside Hospital, Atlanta, Georgia, United States
5 Northside Hospital, Atlanta, Georgia, United States
6 Northside Hospital, Atlanta, Georgia, United States
7 Northside Hospital, Atlanta, Georgia, United States
8 Northside Hospital, Atlanta, Georgia, United States
9 Northside Hospital, Atlanta, Georgia, United States
ABSTRACT
Relapse remains the most common reason for failure of allogeneic hematopoietic cell transplant (HCT). A second allogeneic HCT (HCT2) remains a viable consolidation strategy. Long-term outcomes after HCT2 have historically been guarded, but they may be improving, in part through the increasing ability to cross HLA barriers with alternative donors. We analyzed 120 consecutive adults who underwent HCT2 at our center between 2000 and 2024. Patients with more than 2 allografts were excluded; those with 2 allografts plus a prior autograft were retained. Overall survival (OS) was measured from the date of HCT2, with survivors censored on April 1, 2026. Relapse/progression and non-relapse mortality (NRM) were analyzed as competing risks (Aalen-Johansen). Associations were tested by univariate log-rank, then a multivariable Cox model with forward stepwise selection. Acute and chronic GVHD were tested as time-dependent covariates. Disease risk was classified by the Disease Risk Index (DRI). Transplants were grouped into 3 eras (E1 2000-2012, E2 2013-2018, E3 2019-2024).
Three patients with 3 allografts were excluded; 8 (6.7%) had a prior autograft before their two allografts. Median age at HCT2 was 49.5 years (IQR 39.0-59.0; range 19-76). Males accounted for 50.8% (61/120). Race was White in 93 (77.5%), Black/AA in 22 (18.3%), and Asian in 4 (3.3%); 10 (8.3%) were Hispanic. Indications for HCT2 were relapse/progression in 85 (70.8%), graft failure in 24 (20.0%), new malignancy in 5 (4.2%), declining chimerism in 2 (1.7%), persistent measurable residual disease in 1 (0.8%), and unknown/combination in 3 (2.5%). Median interval between allografts was 510 days (IQR 194-929), with 48 (40.0%) within 1 year and 72 (60.0%) beyond 1 year. DRI was low/intermediate in 55 (45.8%), high/very-high in 57 (47.5%), and unknown in 8 (6.7%). KPS was 90-100 in 21 (17.5%), 50-80 in 90 (75.0%), and unknown in 9 (7.5%). HCT-CI was 0-2 in 48 (40.0%), >=3 in 68 (56.7%), and unknown in 4 (3.3%). Donors were haploidentical in 77 (64.2%), matched unrelated (MUD) in 23 (19.2%), and matched related (MRD) in 20 (16.7%). Graft sources were PBSC in 108 (90.0%), bone marrow in 11 (9.2%), and cord blood in 1 (0.8%). HLA was fully matched in 37 (30.8%) and mismatched in 82 (68.3%). Conditioning was nonmyeloablative in 73 (60.8%), reduced-intensity in 35 (29.2%), and myeloablative in 11 (9.2%). GVHD prophylaxis was post-transplant cyclophosphamide (PTCy)-based in 74 (61.7%), calcineurin-based without PTCy in 40 (33.3%), and other in 6 (5.0%). Donor median age was 38.0 years (IQR 27.0-49.8); donor sex was male in 69 (57.5%). Recipients were CMV-positive in 96 (80.0%). Any acute GVHD occurred in 68 (56.7%), grade II-IV in 54 (45.0%), and grade III-IV in 26 (21.7%); moderate-severe chronic GVHD occurred in 41 (34.2%). Relapse/progression after HCT2 occurred in 48 (40.0%). Cause of death was relapse/progression in 48/84 (57.1%) and non-relapse in 36/84 (42.9%). Median potential follow-up (reverse Kaplan-Meier) was 7.6 years, with a median survivor follow-up of 79.6 months. Median OS from HCT2 was 14.1 months, with 1-, 3-, and 5-year OS of 52.5% (95% CI 42.4-60.2), 33.2% (24.9-41.7), and 32.1% (22.7-39.4). Cumulative incidence of relapse/progression at 1, 3, and 5 years was 34.2%, 44.2%, and 45.5%; NRM was 20.0%, 26.8%, and 26.8%. On univariate analysis, only DRI was associated with OS (p=0.005); non-significant factors included era (p=0.093), interval <=1 vs >1 year (0.124), HCT-CI (0.281), KPS (0.147), conditioning intensity (0.156), donor type (0.955), HLA match (0.930), indication (0.851), and age (0.438). On multivariable Cox regression, high/very-high vs low/intermediate DRI carried a HR of 1.90 (95% CI 1.21-2.98, p=0.005; n=112). On univariate analysis, later eras trended toward improved OS relative to E1 (E2 HR 0.84, 95% CI 0.50-1.39, p=0.50; E3 HR 0.56, 95% CI 0.33-0.95, p=0.033); after adjustment for DRI this attenuated and was no longer significant (E2 HR 0.96, 0.57-1.62, p=0.89; E3 HR 0.67, 0.38-1.16, p=0.15).
Second allogeneic transplant remains a viable consolidative strategy with potential for long-term survival, even in a high-risk, comorbid population. DRI was the strongest determinant of OS in our cohort, although outcomes may be improving over time.
DISCLOSURES
Joseph Elie Maakaron, MD: Sanofi - Honoraria; Incyte - Honoraria; J&J - Honoraria; Kite - Honoraria; Autolus - Honoraria; BMS - Honoraria; Dren Bio - Research Funding; NMDP - Research Funding; Incyte - Research Funding; BMS - Research Funding; Autolus - Consultancy (Includes expert testimony); BMS - Consultancy (Includes expert testimony); Sanofi - Consultancy (Includes expert testimony); J&J - Consultancy (Includes expert testimony); Kite - Consultancy (Includes expert testimony)