Conditioning with Thiotepa-busulfan-fludarabine Improves Outcomes for Patients with MF Undergoing ASCT

December 28, 2020
Darcy Lewis

When patients with myelofibrosis who required hematopoietic stem cell transplant received a conditioning regimen of thiotepa, busulfan, and fludarabine along with 2 two alkylating agents, the subsequent transplants were nearly all successful.

When patients with myelofibrosis (MF) who required hematopoietic stem cell transplant (HSCT) received a conditioning regimen of thiotepa, busulfan, and fludarabine (TBF) along with 2 two alkylating agents, the subsequent transplants were nearly all successful. These findings are from a single-center, retrospective analysis published recently in Leukemia & Lymphoma.

Twenty-nine consecutive patients with MF were enrolled at a single institution. The results show that of the patients enrolled, all but 2 patients successfully engrafted following the TBF conditioning regimen and no patients relapsed.

About half the patients (n = 15, 52%) were classified as high risk. Peripheral blood stem cells were the graft source in 27 patients. Donor types were as follows: HLA-matched related (n = 5), matched unrelated (n = 16), mismatched unrelated (n = 1), and haploidentical (n = 7).

About one-fifth of patients (21%) experienced grade II–IV acute graft-versus-host disease (GVHD; 95% Confidence Interval [CI], 10–42) by day 100. At 3 years, the cumulative incidence of chronic GVHD was 39% (95% CI, 23–65). Median follow-up was 39 months (range 14–60). At 3 years, overall survival (OS) was 69% (95% CI, 50–83).

“Our findings show that TBF is a valid and safe conditioning regimen providing good disease control and a promising OS,” wrote the authors, led by Mara Memoli, MD, of Saint Antoine Hospital, Paris, France, where the study took place. “Overall, our results indicate that the combination of two alkylating agents and the use of thiotepa in the conditioning regimen provide [a] high chance of achieving full donor chimerism engraftment, good disease control and promising OS in MF patients undergoing HSCT.

The TBF conditioning regimen varied slightly as the hospital’s policy evolved. The first 3 patients received a total dose of 10 mg/kg of thiotepa (5 mg/kg on days -7 and -6). This dose was reduced to 5 mg/kg (at day -6) in all remaining patients per revised hospital policy. All patients received a total dose of 150 mg/m2 of fludarabine (30 mg/m2 from day -5 to day -1). The total dose of intravenous busulfan was 6.4 mg/kg (on days -4 and -3) or 9.6 mg/kg (3.2 mg/kg/day on days -4, -3, and -2) based on the patient’s age, comorbidities and performance status. All patients received cyclosporine A and mycophenolate as GVHD prophylaxis.

OS was the study’s primary endpoint. Secondary endpoints included the cumulative incidences of acute and chronic GVHD neutrophil and platelet engraftment, non-relapse mortality (NRM), and disease-free survival (DFS).

Patients’ median age was 56 years (range 42–70) and patients were divided nearly equally among those who had primary MF (n = 14, 48%) and secondary MF (n = 15, 52%). Among these, 6 had previously had polycythemia vera and 9 had essential thrombocythemia.

More than half the patients (n = 18, 62%) had a JAK2 mutation, including 2 with an associated CALR mutation. Six patients (21%) had a single CALR mutation, while 1 patient (3%) had MPL mutation and 4 patients (14%) were triple negative. Eight patients (27%) underwent next-generation sequencing. Among the 3 patients with CALR mutation, all were also ASXL1-mutated and 2 had an associated EZH2 mutation. Among the 3 JAK2-mutated patients, 1 patient had SF3B1 co-mutation. In the 2 triple negative patients according to standard molecular biology, 1 had BCOR1 mutation, and 1 patient did not present with any mutation.

Nineteen patients had splenomegaly; 16 of them received ruxolitinib (Jakafi). Among these, 11 received treatment for more than three months. Two tolerated ruxolitinib but took it for less than 3 months due to the start of the conditioning regimen. Three patients had to stop treatment for severe thrombocytopenia before day 30. Most patients had evidence of disease progression or symptom persistence before HSCT but patients were not required to have a clinical response to ruxolitinib to progress to HSCT.

Among the patients who engrafted, the median time to neutrophil recovery was 14 days (range 9–34). The median times to achieve platelet engraftment of at least 20 x 109/L was 17 days (range 1–65) and 19 days (range, 8-185) to reach at least 50 x 109/L. The cumulative incidence of neutrophil engraftment at 30 days was 96% (95% CI, 80–99). The 2 patients who did not engraft died during aplasia, one at day 19 from infection and the other one at day 12 from multi-organ failure.

Among the patients who engrafted, all except two had a full donor chimerism at day 90. Three patients needed and received a successful CD34+-selected stem cell boost from their original donor, according to hospital center policy. Nine patients received the thrombopoietin mimetic agent eltrombopag.

The authors observed a trend toward better OS among patients who underwent transplant before the median time period of 11 months (<11 months 86%; > 11 months 53%; P = 0.05). They also observed lower OS among patients who received ruxolitinib before transplant compared to patients who did not (50% versus 92%, P = 0.02). Among the one-third of patients who died (n = 9), the cause of death was infection in 5 and GVHD and multi-organ failure in 2 each.

Memoli et al noted that the main limitations of their study were the sample size and its retrospective design. “However, this is the largest study on [the] TBF conditioning regimen before allo-HSCT for MF,” they wrote. “Further prospective studies of larger scale are needed to assess the use of [the] TBF conditioning regimen in specific patient settings.”


Memolia M, Paviglianitia A, Malarda F, et al. Thiotepa-busulfan-fludarabine as a conditioning regimen for patients with myelofibrosis undergoing allogeneic hematopoietic transplantation: a single center experience. Leuk. Lymphoma. Published online October 5, 2020. DOI: 10.1080/10428194.2020.1827246.