Background
The outcome of chemotherapy in patients older than 40 years with acute lymphoblastic leukaemia is poor and myeloablative allogeneic haematopoietic stem-cell transplantation (HSCT) has a high transplant-related mortality (TRM) in this age cohort. The aim of this study was to assess the activity and safety of reduced-intensity conditioned allogeneic HSCT in this patient population.
Methods
This was a single-arm, prospective study within the UKALL14 trial done in 46 centres in the UK, which recruited patients to the transplantation substudy. Participants in UKALL14 had B-cell or T-cell acute lymphoblastic leukaemia, were aged 25–65 years (BCR-ABL1-negative) or 18–65 years (BCR-ABL1-positive), and for this subcohort had a fit, matched sibling donor or an 8 out of 8 allelic matched unrelated donor (HLA-A, HLA-B, HLA-C, and HLA-DR). On June 20, 2014, the protocol was amended to allow 7 out of 8 matched unrelated donors if the patient had high risk cytogenetics or was minimal residual disease (MRD)-positive after the second induction course. Patients were given fludarabine, melphalan, and alemtuzumab (FMA; intravenous fludarabine 30 mg/m2 on days –6 to –2, melphalan 140 mg/m2 on day –2, and alemtuzumab 30 mg on day –1 [sibling donor] and days –2 and –1 [unrelated donor]) before allogeneic HSCT (aged ≥41 years patient pathway). Donor lymphocyte infusions were given from 6 months for mixed chimerism or MRD. The primary endpoint was event-free survival and secondary and transplantation-specific endpoints included overall survival, relapse incidence, TRM, and acute and chronic graft-versus-host disease (GVHD). This study is registered with ClinicalTrials.gov, NCT01085617.
Findings
From Feb 22, 2011, to July 26, 2018, 249 patients (236 aged ≥41 years and 13 younger than 41 years) considered unfit for a myeloablative allograft received an FMA reduced-intensity conditioned HSCT. 138 (55%) patients were male and 111 (45%) were female. 88 (35%) participants received transplantations from a sibling donor and 160 (64%) received transplantations from unrelated donors. 211 (85%) participants had B-precursor acute lymphoblastic leukaemia. High-risk cytogenetics were present in 43 (22%) and another 63 (25%) participants were BCR-ABL1-positive. At median follow-up of 49 months (IQR 36–70), 4-year event-free survival was 46·8% (95% CI 40·1–53·2) and 4-year overall survival was 54·8% (48·0–61·2). 4-year cumulative incidence of relapse was 33·6% (27·9–40·2) and 4-year TRM was 19·6% (15·1–25·3). 27 (56%) of 48 patients with TRM had infection as the named cause of death. Seven (15%) of 48 patients had fungal infections, 13 (27%) patients had bacterial infections (six gram-negative), and 11 (23%) had viral infections (three cytomegalovirus and two Epstein-Barr virus). Acute GVHD grade 2–4 occurred in 29 (12%) of 247 patients and grade 3–4 occurred in 12 (5%) patients. Chronic GVHD incidence was 84 (37%) of 228 patients (50 [22%] had extensive chronic GVHD). 49 (30%) of 162 patients had detectable end-of-induction MRD, which portended worse outcomes with event-free survival (HR 2·40 [95% CI 1·46–3·93]) and time-to-relapse (HR 2·41 [1·29–4·48]).
Interpretation
FMA reduced-intensity conditioned allogeneic HSCT in older patients with acute lymphoblastic leukaemia in first complete remission provided good disease control with moderate GVHD, resulting in better-than-expected event-free survival and overall survival in this high-risk population. Strategies to reduce infection-related TRM will further improve outcomes.
Funding
Cancer Research UK.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)