Commentary on Abstracts #404 and #3079

Leonard et al (abstract #404) described their experience with a new humanized anti-CD22 monoclonal antibody, epratuzumab. This antibody has previously been studied conjugated to both iodine-131 and yttrium-90 in the treatment of NHL. However, in the present trial, the naked antibody was used in a phase I setting with at least three indolent and three aggressive NHL patients per dose level to assess its activity in the absence of the isotope. Although the dose was escalated from 120 to 600 mg/m², no dose-limiting toxicities were encountered. Responses were noted beginning at the 240-mg/m² dose level, including CRs in patients with a follicular histology. Drug levels were detectable in the serum up to 3 months after treatment.Since this antibody targets a different antigen than rituximab, iodine-131 tositumomab, or ibritumomab tiuxetan, it might be useful combined not only with chemotherapy but also with other monoclonal antibodies.

Few effective drugs are available for patients with AML, especially those who do not respond to initial induction therapy. Other than stem-cell transplantation, no therapy is curative and new strategies are clearly needed. Leukemic blast cells from more than 90% of patients with AML express CD33, but this antigen is not expressed by normal hematopoietic stem cells. Initial in vivo studies with an unmodified murine anti-CD33 antibody in patients with AML demonstrated that the antibody bound to leukemia cells and that, following cell binding, the antigen-antibody complex was internalized. However, when patients with active leukemia were treated with the unmodified antibody, only transient decreases in peripheral blast counts were noted.

Sievers et al (Blood 93:3678-3684, 1999) initially published a study to determine whether they could selectively ablate CD33-positive cells in patients with AML using an immunoconjugate consisting of humanized anti-CD33 conjugated to the antitumor antibiotic calicheamicin. In this phase I trial involving 40 patients with relapsed or refractory disease, leukemic cells were eliminated from the blood and marrow of 20% of patients, with normalization of blood counts in 3 (8%) patients. The drug was well tolerated, with fevers and chills being the most common adverse effects, although myelosuppression was also reported.

At the 1999 ASH meeting, these same investigators (abstract #3079) presented exciting preliminary results from a phase II trial of the CMA-676 monoclonal antibody in 59 patients with AML in first relapse following an initial remission of at least 6 months; those with secondary AML or preexisting MDS were excluded. The response rate was 34% among patients in first relapse. Although the median survival was only 161 days, 12 of patients who achieved a remission remained relapse-free with a median follow-up of 238 days. Post-remission therapy included bone marrow transplantation (BMT; N = 12) or chemotherapy (N = 3). There was one BMT-related death, and two patients relapsed after allogeneic stem-cell transplantation. Of five patients who did not receive post-CR therapy, two remained free of disease at 4.5 and 3.5 months after the first dose of the antibody. The drug was well tolerated.

Several phase II and III clinical trials are attempting to determine the optimal strategy for the use of this antibody. An early interim analysis of one phase III study, involving patients with AML in first relapse, found that CMA-676 treatment eliminated blasts from peripheral blood and marrow in almost half of the patients. Antibody therapy was associated with much less toxicity than produced by standard chemotherapy. Other studies using radiolabeled anti-CD33 and anti-CD45 antibodies are exploring the possible role of these antibodies as part of the preparative regimen for BMT (Appelbaum: Semin Hematol 36:2-8, 1999). It is likely that CMA-676 and other antibodies will play major roles in the future treatment of AML.