Successful combination of anti-CD33 antibody (gemtuzumab ozogamicin) and minimal conditioning before second allografting in recurrent acute myeloid leukaemia.

Abstract

Patients with acute myeloid leukaemia (AML) who have failed haematopoietic stem cell transplantation (HSCT) have a poor prognosis (Webb, 1999). The monoclonal anti-CD33 antibody, gemtuzumab ozogamicin (GO) (Mylotarg, CMA-676; Wyeth Laboratories, Philadelphia, PA, USA) is an emerging therapeutic option for patients with CD33-positive disease (Sievers et al, 2001). We report two patients with recurrent AML who received GO as part of their minimal conditioning before allogeneic HSCT as a second transplantation. The first patient, a 40-year-old woman, was diagnosed with AML [French–American–British (FAB) classification M2] with a normal karyotype in October 2000. First-line chemotherapy led to complete remission (CR) in February 2001. Treatment was accompanied by serious complications and as a result no consolidation therapy was performed. The patient's first relapse was treated with high-dose busulphan (16 mg/kg) and autologous HSCT. Because of further disease progression, allografting from a human leucocyte antigen-matched, unrelated donor was planned. The second patient was a 19-year-old woman who was diagnosed with AML (FAB M4) in April 2001. After achieving CR, the patient underwent myeloablative conditioning therapy (busulphan 16 mg/kg, cyclophosphamide 120 mg/kg) with an allogeneic bone marrow transplantation (BMT) from a matched sibling donor. Unfortunately, only 3 months later, a haematological relapse was diagnosed and a second allogeneic transplant with peripheral blood stem cells (PBSC) from the same donor was scheduled. As a second high-dose conditioning appeared inappropriate for both patients, they received 6 mg/m2 and 3 mg/m2 of mylotarg 21 and 14 d before allogeneic HSCT respectively. Minimal conditioning consisted of 30 mg/m2 fludarabine (on d −3 to −1) and 2 Gy total body irradiation on (d 0) (McSweeney et al, 2001). Granulocyte colony-stimulating factor (G-CSF)-mobilized PBSC were infused on d 0. Patient 1 received PBSC from a matched unrelated donor (5·3 × 106/kg CD34+ cells, 3·1 × 108/kg CD3+ cells). Patient 2 received a PBSC infusion from the original sibling (8·2 × 106/kg CD34+ cells, 4·3 × 108/kg CD3+ cells). Graft-versus-host disease (GVHD) prophylaxis consisted of cyclosporin A (CsA) and mycophenolate mofetil (on d 0–35) in patient 1, and no prophylactic immunosuppression was administered to patient 2. The maximum non-haematological toxicity observed was grade 2 stomatitis and hyperbilirubinaemia, occurring on d 7 and d 21 after HSCT in patient 1. Both patients achieved a fast and sustained haematological engraftment. Full donor chimaerism could be detected on d +26 in patient 1 and on d +17 in patient 2. Both patients developed acute GVHD grade II involving the skin and liver, as well as limited chronic GVHD. CR was documented on d +25 for patient 1 and on d +49 for patient 2. Both patients remained in CR at the time of writing, 10 and 14 months after the second HSCT with stable donor chimaerism. Figure 1 summarizes the clinical course and the subset chimaerism of patient 2. The clinical course and subset chimaerism in patient 2, with relapsed AML after a first allogeneic bone marrow transplantation (BMT), and who was successfully treated by gemtuzumab ozogamicin and a second allogeneic peripheral blood stem cell transplant (PBSCT) from the same donor. Preparation of cellular subsets and chimaerism analysis were performed as previously described (Thiede et al, 2001). The figure shows the percentage donor chimaerism of the patient against the markers CD4, CD8 and CD34 over a period of 350 d. The time of allogeneic BMT (allo-BMT), relapse, allogeneic PBSCT (allo-PBSCT), and the onset of acute and chronic GVHD are also shown. Sievers et al (2001) reported on 15 patients who underwent allogeneic (n = 10) or autologous (n = 5) HSCT as successful consolidation therapy several weeks after remission induction with GO. In contrast to the reported serious hepatotoxicity [common toxicity criteria (CTC) grade 3–4] in up to 23% of patients, we observed only mild hepatic side-effects in our patients (CTC grade 0–2). However, the cumulative dose of GO in our study was only 9 mg/m2, compared with 18 mg/m2 reported for the pivotal trial (Sievers et al, 2001). Despite poor prognostic factors, the patients in the present study achieved a sustained CR. The use of peripheral PBSC after failure of the first allogeneic BMT in patient 2 might be another cause of the higher antileukaemic efficacy of PBSC when compared with marrow. Furthermore, apoptosis induced by the administration of the anti-CD33 antibody may have induced a potent cytotoxic T-cell response (Selenko et al, 2002). In conclusion, gemtuzumab ozogamicin seems to be a feasible adjunct to minimal conditioning before allogeneic transplantation of haematopoietic stem cells from both related and unrelated donors. This study was supported in part by the Deutsche Krebshilfe (grant no. 70–2755 to CT/MB).