Second Transplant Offers the Possibility of Long Term Graft-Versus-Host-Disease-Free Relapse-Free Survival

Consolidative allogeneic haematopoietic stem cell transplant (HSCT) is standard of care therapy for fit patients with various types of high risk haematological malignancy. Due to the complex interplay of disease biology and immune function, some patients will relapse following this potentially curative therapy. Historically, outcomes in patients relapsing following HSCT were dismal, with the majority either lacking further therapeutic options or unfit to receive them. Australian outcomes following second HSCT (HSCT2) for AML from 1994-2009 were previously reported, with a 3 year overall survival (OS) and disease free survival of 53% and 29% respectively (Bilmon et al, BMT 2014). Recent EBMT data on HSCT2 for AML/MDS reports a 3 year OS of 37% (Rodriguez-Arboli et al, TCT 2024), with similar outcomes for HSCT2 in ALL (Nagler et al, BJH 2019). With increasing novel salvage therapies, achievement of remission in post-HSCT relapse has increased eligibility for HSCT2. We assessed HSCT2 outcomes in the modern therapeutic landscape in NSW, Australia.

Methods

A retrospective analysis of adult patients who underwent HSCT2 from January 2009 to December 2021 for relapsed haematological malignancy at HSCT centres in NSW was performed. Patient characteristics and outcomes post HSCT and HSCT2 were collected via medical record audit. Kaplan-Meier methods were used to estimate disease free (DFS) and median OS (mOS). Multivariable Cox regression analysis including age at HSCT2, underlying disease, time to relapse and donor type was performed.

Results

Ninety-four eligible patients were included, with a median age of 48 years (20-74 years), 58.5% male. The median time between transplants was 2.1 years (92d – 26.7 years). The most common underlying malignancies were AML (n=40/94), followed by ALL (n=17/94, 6 Philadelphia [Ph] positive, 9 Ph negative and two CML blast phase) and MDS (n=14/94). Matched sibling or unrelated donors were used in 77% of HSCT2, and haploidentical donors in 20%. A different donor was used for HSCT2 in 71% of cases. The most common conditioning regimens were fludarabine with busulphan or melphalan in matched reduced intensity (RIC) HSCT, and cyclophosphamide with busulphan or total body irradiation (TBI) for matched myeloablative (MAC) regimens. Thymoglobulin was used in 75/94 patients for in vivo T cell depletion. Haploidentical RIC conditioning was with fludarabine/cyclophosphamide/TBI, with fludarabine/TBI for the one MAC haploidentical transplant, along with post-transplant cyclophosphamide for T cell depletion.

The two year OS and DFS were both 51%. The mOS following HSCT2 was 4.6 years (95%CI 4-6.3 years). In patients surviving beyond 2years, mOS was 5 years (95%CI 4-7 years) with a median follow up of 4.4 years (1-14 years). Two-year GVHD-free, relapse free survival (GRFS) was 32% (n=30/94). Relapse within 6 months of first HSCT was the only factor predictive of OS (HR 0.42 [95%CI 0.19-0.90], p = 0.027). Nonetheless, 36% (n=5/14) with early relapse were alive at last review with a median follow up of 3.2 years (2.9-10.9 years). Twelve patients had two MAC transplants, with a 66.7% 2 year OS, and 58% remain alive with a median follow up on 2.8 years (1-11 years). By univariable analysis, there was no survival difference at 2 years according to conditioning intensity or use of a different donor for HSCT2.

Overall, the leading cause of death was relapse (32%), which was uncommon beyond 2 years (5/48). The transplant related mortality rate was 21% at 2 years, including 4 patients who died of GVHD and 12 of infection. Three patients developed secondary malignancies, two arising in the first 6 months post HSCT2.

Conclusions

Our data suggests second allogeneic transplant can lead to durable remissions, with a 51% 2 year OS and a mOS of 4.6 years. Relapse within 6 months of first HSCT is a poor prognostic factor, however durable remission did still occur in a subset of such patients. The leading cause of death in the first two years post HSCT2 was relapse which may be augmented by the increasing adoption of maintenance therapy. Importantly, 62% of patients alive at 2 years had no significant GVHD. Transplant numbers are rising worldwide and second allogeneic transplants represent an area of growing demand. Second transplant should be considered as a viable option to prolong survival and potentially cure patients relapsing after first allogeneic stem cell transplant.