Ibritumomab tiuxetan (Zevalin) is a murine IgG directed against CD20 and conjugated to yttrium-90. The basic antibody is the murine rituximab. The yttrium-90 isotope was selected because it has a number of properties that are considered to be more favorable than those of iodine-131. These include the fact that ibritumomab tiuxetan is a pure beta-emitter, with higher energy and a longer path length. Ibritumomab tiuxetan has been reported to induce responses in 67% of patients with intermediate- and high-grade NHLs and 82% of those with low-grade NHL who had not been treated previously with rituximab (Witzig et al: J Clin Oncol 17:3793-3803, 1999).
Although we are fortunate in having several active monoclonal antibodies available for our patients, a number of important issues still need to be addressed. For example, it is unclear whether a conjugated or an unconjugated antibody is more effective, which groups of patients are most likely to benefit from each type, or whether these agents are still effective when used sequentially. The optimal isotope is also a subject of controversy.
Witzig et al (abstract #2805) are attempting to resolve the question of the relative efficacy of naked antibodies vs radioimmunoconjugates. They presented the interim results of a randomized trial comparing rituximab with ibritumomab tiuxetan. The 143 patients were comparable with respect to age, bone marrow involvement, prior therapy, LDH level, performance status, and other features. An interim analysis of data from the first 90 patients showed that the overall response rate with ibritumomab tiuxetan of 80% was significantly higher than the 44% rate with rituximab. However, it is unclear whether the higher response rate will translate into longer responses or prolonged survival. Obviously, data from the completed study are needed.
Rituximab is commonly used in relapsed and refractory follicular NHL, and is attractive because it is relatively easy to administer even in the outpatient setting. However, patients uniformly relapse, and one possible role for the radioimmunoconjugates may be as salvage therapy after failure of naked antibody therapy.
Gordon et al (abstract #396) conducted a phase I/II trial of ibritumomab tiuxetan in patients considered to be refractory to rituximab (defined by a lack of response or a time to progression of less than 6 months). An interim analysis of the first 26 patients in this trial noted an overall response rate of 46%. The sequence of antibodies was not associated with major untoward toxicity: transient grade 4 thrombocytopenia and neutropenia occurred in 8% and 23% of patients, respectively.
In the expanded access database of results with iodine-131 tositumomab in rituximab failures (Gregory et al, abstract #397), iodine-131 tositumomab was given safely to the first 30 patients, with only two new cases of HAMA. No response data were provided.
One of the major complications of radioimmunoconjugate therapy is myelosuppression. For example, in the pivotal trial of iodine-131 tositu-momab, the median number of days to recovery of platelets and neutrophils was 70 to 80 days. In the phase I/II trial with iodine-131 tositumomab (Witzig et al: J Clin Oncol 17:3793-3803, 1999), the median time to recovery of nadir counts was 14 days. This complication has limited eligibility for clinical trials to patients with adequate bone marrow reserves.
Witzig et al (abstract #400) studied the possibility that administration of reduced doses of iodine-131 tositumomab may be effective in NHL without worsening preexisting thrombocytopenia. Pretreatment platelet counts were found to predict the severity of myelotoxicity. In the phase I/II portion of the study, the maximum tolerated dose was 0.4 mCi/kg in patients with normal blood counts prior to therapy but was 0.3 mCi/kg in patients with mild pretreatment thrombocytopenia. The present phase II study expanded the experience using 0.3 mCi/kg in 24 patients with < 25% bone marrow involvement and a platelet count of 100,000/µL to 149,000/µL. The response rate was 68%, including 23% CRs. More importantly, the median platelet nadir was 34,000/µL, and the incidence of grade 4 thrombocytopenia was only 15%.