Allogeneic hematopoietic stem cell transplantation in thalassemia major: results of a reduced-toxicity conditioning regimen based on the use of treosulfan

Allogeneic hematopoietic stem cell transplantation (HSCT) remains the only potentially curative treatment able to render patients with thalassemia major (TM) transfusion independent.^1-5  However, the clinical outcome after HSCT in children with TM who belong to class 3 of the Pesaro classification and in adults with poor performance status and/or organ dysfunction is still unsatisfactory because of the high risk of life-threatening complications or graft failure.^1-5 

Treosulfan-based conditioning regimens have been shown initially to be safe and effective in adults with hematologic malignancies not eligible for conventional preparations.^6-9  These results have been confirmed recently in pediatric patients with high-risk hematologic malignancies receiving a treosulfan-based preparative regimen before second or third HSCT.¹⁰  Moreover, a treosulfan-based conditioning regimen proved to be valuable in children with primary immunodeficiencies.¹¹ 

We reported recently preliminary, encouraging results on the use of a thiotepa/treosulfan/fludarabine myeloablative regimen in a cohort of 20 patients with TM. In particular, we found a high probability of cure of the disease in the absence of major transplantation-related complications in adults and in patients with poor performance status.¹²  In the present study, we report the final results on the safety and efficacy of this regimen in a large cohort of TM patients.

Sixty patients with TM (median age, 7 years; range, 1-37) received transplantations between November 2005 and August 2011 in 3 Italian centers (Pavia, Roma, and Cagliari). The trial was approved by the institutional review boards (approval no. 2005-005182-11) of all participating institutions. Written informed consent was obtained from all patients or from their parents/legal guardians in accordance with the Declaration of Helsinki.

Details on patient, donor, and transplantation characteristics are reported in Table 1. Before transplantation, pediatric patients were assigned to 1 of the 3 classes of risk using the Pesaro classification.¹  Among 48 children, 27 were assigned to class 1, 17 to class 2, and 4 to class 3; the remaining 12 patients were adults. Twenty patients were transplanted from an HLA-identical sibling (matched family donor [MFD]) and the remaining 40 from an unrelated donor (UD). In the MFD cohort, the donor was always an HLA-identical sibling, and 11 patients received cord blood cells. In all patients receiving transplantations from a UD, high-resolution molecular typing was performed to characterize HLA class I and II loci (ie, loci A, B, C, DRB1, and DQB1); only donors fully matched or with a single class I allelic disparity were selected. In UD recipients, an autologous rescue of BM cells was harvested and cryopreserved to be used in case of graft failure.

All patients received the same conditioning regimen, which included IV thiotepa (8 mg/kg on day −7), treosulfan (14 g/m²/d from days −6 to −4), and fludarabine (40 mg/m²/d for 4 consecutive days from days −6 to −3). GVHD prophylaxis varied according to the stem cell source and type of donor (see Table 1 for details). In all patients, cyclosporin A was given for 12 months after transplantation. Pretransplantation antithymocyte globulin (Fresenius; 10 mg/kg/d for 3 consecutive days from days −5 to −3) was administered in all recipients of UD-HSCT. BM was used as stem cell source in 47 patients, 11 patients received MFD cord blood–derived HSCs, and the remaining 2 patients were given transplantations with peripheral blood–mobilized HSCs.

Acute and chronic GVHD were diagnosed and graded according to the Seattle criteria.¹³  Patients surviving more than 14 and 100 days after transplantation were evaluated for acute and chronic GVHD, respectively.

Analysis used December 31, 2011 as the report date at which the centers locked data on patient outcome. Patients were censored at time of death or last follow-up. Probability of overall survival and thalassemia-free survival (TFS) were estimated by the Kaplan-Meier product-limit method and expressed as a percentage with 95% confidence interval (CI). For calculation of TFS, data on patients were recorded at time of death, graft failure, or last follow-up. Probabilities of acute and chronic GVHD, graft failure, and transplantation-related mortality were calculated as cumulative incidence curves, to adjust the analysis for competing risks. P < .05 was considered statistically significant.

All patients engrafted except one, who died on day +11; the median time to neutrophil and platelet recovery was 20 days (range, 11-30 and 11-36, respectively). Five patients (all of whom were given BM cells) experienced secondary graft failure at a median of 9 months (range, 1.5-18) after HSCT; the cumulative incidence of graft failure was 9% (95% CI, 3%-19%; Figure 1A). Only 1 of these 5 patients belonged to risk class 3; all of the others had been allocated to class 1. A total of 3 and 2 of these 5 patients received the allograft from either a UD or an MFD, respectively. Four patients were successfully retransplanted from the same donor (an MFD or a UD in 2 cases each) using a preparative regimen combining busulfan/thiotepa/fludarabine and are currently alive and transfusion independent without any sign of GVHD.

Eight of the 59 patients at risk experienced grade II-IV acute GVHD, which was severe in 4 cases; the cumulative incidence of grade II-IV and grade III-IV acute GVHD was 14% (95% CI, 6%-24%) and 7% (95% CI, 2%-15%), respectively. Only 1 child of the 56 patients at risk developed chronic GVHD (of limited extension); the cumulative incidence was 2% (95% CI, 0%-8%).

Four patients died of transplantation-related complications (all within 4 months after HSCT): 3 patients due to severe acute GVHD unresponsive to several lines of immunosuppressive therapy and 1 patient due to viral pneumonia. All patients who died had received transplantations from a UD and belonged to risk class 2 or 3. The cumulative incidence of transplantation-related mortality was 7% (95% CI, 3%-18%; Figure 1A). No case of veno-occlusive disease was recorded. With a median follow-up of 36 months (range, 4-73), the 5-year overall survival probability is 93% (95% CI, 83%-97%; Figure 1B), with no difference between MFD and UD recipients (data not shown). The probability of TFS after first HSCT for the whole cohort is 84% (95% CI, 71%-91%); it was 87% (95% CI, 55%-97%) and 82% (95% CI, 66%-91%) for patients receiving either MFD or UD HSCT, respectively (P = .43; Figure 1C). Because 4 of the 5 patients experiencing secondary graft failure were rescued by a second allograft, 55 patients are currently alive and transfusion independent with sustained donor engraftment. We did not observe any difference in terms of outcome between patients belonging to class 1 or 2 of the Pesaro classification and class 3 adult patients, the probability of TFS being 85% (95% CI, 69%-93%) and 81% (95% CI, 52%-94%; P = .55̧ Figure 1D), respectively.

These results confirm in a large cohort of TM patients the safety and efficacy of the thiotepa/treosulfan/fludarabine combination as a preparative regimen. This regimen was also able to abolish the difference usually observed among patients of different classes of risk.^1-5  The outcome of class 3 adult patients was comparable with that reported by Sodani et al in class 3 children receiving transplantations from an MFD using a novel approach aimed at enhancing both immune suppression and eradication of the thalassemic erythropoiesis.¹⁴  High-resolution typing and the stringent criteria of donor compatibility contributed to the excellent probability of TFS observed in our patients receiving transplantations from UDs. We propose this preparation as a suitable and appropriate option for minimizing the risk of life-threatening complications in adult poor-performance status TM patients, although it is also of value for patients with good prognostic characteristics. The incidence and severity of late effects after this regimen is a subject for future investigation. Randomized trials comparing treosulfan-based and busulfan-based regimens are warranted.

This study was supported in part by grants from Istituto Superiore di Sanità (National Program on Stem Cells), Ministero dell'Istruzione, dell'Università e della Ricerca, Progetti di Rilevante Interesse Nazionale, Associazione Italiana per la Ricerca sul Cancro (IG9062 to M.E.B.), and by the special grant “5 × 1000” from Associazione Italiana per la Ricerca sul Cancro (to F.L.).

Contribution: M.E.B. performed the transplantations, collected and analyzed the data, and wrote the manuscript; G.G., A.B., D.P., E.P., A.V., G.C., A.M. and R.M.P. performed the transplantations, collected and analyzed the data; M.Z. performed the transplantations; B.C. performed the statistical analysis; G.L.N. performed the transplantations, collected and analyzed the data, and contributed to the design of the study; and F.L. designed the study, performed the transplantations, analyzed the data, and contributed to the final writing of the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Maria Ester Bernardo, MD, PhD, Dipartimento di Ematologia ed Oncologia Pediatrica, IRCCS Ospedale Pediatrico Bambino Gesù, P le S Onofrio, 00165 Roma, Italy; e-mail: mariaester.bernardo@opbg.net.