Commentary on Abstracts #1432, #4760, #950, #3154, #1429, #3157, #4780, #2479, #3167, #2478, and #791

The availability of a continually increasing menu of effective monoclonal antibodies has revolutionized the management of patients with indolent B-cell malignancies. The dilemma we now face is how to best incorporate these agents into our therapeutic paradigms. The most widely studied of the agents is rituximab (Rituxan), a chimeric anti-CD20 antibody. Rituximab induces responses in 50% of patients with relapsed or refractory follicular/low-grade non-Hodgkin’s lymphoma (NHL), with complete remissions in 6% (McLaughlin et al: J Clin Oncol 16:2825-2833, 1998). It is also being evaluated in aggressive lymphomas, Hodgkin’s disease, and nonmalignant disorders (eg, systemic lupus erythematosus, immune thrombocytopenic purpura).

Following rituximab therapy, B-cell numbers rapidly decrease to extremely low values. Antibody levels can be detected in the blood for approximately 6 months, and B cells begin to return to normal numbers at about 9 months; the median duration of response is about 1 year. Taken together, these observations have suggested that relapses may occur in association with a return of the B-cell count. Grillo-López et al (abstract #1432) explored this possibility but were unable to identify a clear association between B-cell recovery and response or serum levels of the drug; indeed, relapses occur while patients are still B-cell depleted, while some responses have lasted longer than 4 years despite normalization of B-cell numbers (abstract #4760).

Clearly, the most popular area of rituximab development has been combining the antibody with other drugs. A widely discussed and promising abstract at the ASH 2000 meeting was presented by Coiffier of the French GELA group (abstract #950). These investigators randomized 400 patients over the age of 60 years with diffuse large B-cell NHL to either CHOP (cyclophosphamide [Cytoxan, Neosar], doxorubicin HCl, vincristine [Oncovin], prednisone) or CHOP plus rituximab. The complete remission (CR) and CRu rate with the combined-modality arm was 76% vs 60%; the 12-month projected event-free survival of 69% and overall survival of 83% were significantly longer with the combined-modality approach than with CHOP alone (49% and 68%, respectively). Longer follow-up is clearly required to determine if the differences will persist.

An ongoing US Intergroup trial has already randomized more than 525 patients onto a similar study. There is also a secondary randomization following response to observation or further antibody therapy. Hopefully the results of the US study will confirm the French observations, leading to a change in our approach to these patients. It is important to note that these results cannot currently be extrapolated to younger patients with diffuse large B-cell NHL or to other histologies, such as follicular and low-grade NHL. The Southwest Oncology Group has an ongoing study comparing CHOP with CHOP plus rituximab with CHOP followed by tositumomab/iodine-131 tositumomab in patients with previously untreated follicular and low-grade NHL. A problem in the interpretation of the multiple trials is that not all of the various CHOP/rituximab regimens have similar designs (abstract #950; Czuczman et al: J Clin Oncol 17:268-276, 1999; Vose et al: 19:389-397, 2001). Whether one strategy is superior to the others is not known.

Fludarabine (Fludara) is one of the most active single chemotherapy agents for the treatment of patients with follicular NHL. Therefore, combining rituximab with this agent is a logical step. Czuczman et al (abstract #3154) published their experience with 39 patients with advanced-stage indolent NHL, two-thirds of whom were previously untreated. The response rate was 92%, with 67% CRs. This highly active regimen merits testing in randomized trials.

The group from M. D. Anderson has previously published impressive results with the FND regimen (fludarabine, mitoxantrone [Novantrone], dexamethasone) in relapsed and refractory patients with indolent NHL (McLaughlin et al: J Clin Oncol 14:1262-1268, 1996). The natural progression was to add rituximab to this regimen. At the recent ASH 2000 meeting, they presented their data on the molecular responses to this regimen (abstract #1429) in 134 patients with stage IV disease who received FND and rituximab either concurrently or sequentially. There were 95 patients who had baseline polymerase chain reaction (PCR) data, 74% of whom were positive for the bcl-2 gene rearrangement. They found a higher rate of molecular response at 6 months in the combination arm, but this difference was not notable at 12 months and there was not yet a significant difference in failure-free survival between the two arms. However, the relevance of PCR data remains uncertain. Patients in the rituximab pivotal trial became PCR negative despite persistence of lymphadenopathy. In addition, the reproducibility of PCR studies has come into question (Johnson et al: Ann Oncol 100:1349-1354, 1999).

Numerous other active regimens have been developed that incorporate rituximab with chemotherapy (abstracts #3157 and #4780); whether the antibody interacts more favorably with one agent or combination of agents is as yet unknown. However, this question will not be resolved by a never-ending series of small phase II studies.

Another promising area of development is combining rituximab with other biological agents. Interferon alpha (IFN) has been studied in patients with indolent NHL for many years; nevertheless, its role in the management of these patients remains controversial (Cheson et al: J Clin Oncol 18:2007-2009, 2000). Davis et al (Clin Cancer Res 6:2644-2652, 2000) first reported their results with a combination of rituximab and IFN in 38 patients with refractory or relapsed follicular or low-grade NHL. IFN was administered first for 12 weeks, with the antibody starting on week 5 for four infusions at the standard dose. The overall response rate of 45% with 11% CRs was comparable to results without IFN. However, the median response duration and time to progression of 22.3 and 25.2 months, respectively, appeared to be longer than had been previously observed with the antibody alone.

At the ASH 2000 meeting, Kimby and coworkers (abstract #2479) presented preliminary results from a randomized phase II trial in 125 patients, almost two-thirds of whom had not received prior therapy. They were first treated with four weekly infusions of rituximab; patients with a partial or minor response were randomized to receive either a second cycle of the antibody or IFN for 6 weeks. The response rate to the initial course of rituximab was 57%, with 12% CRs. The response rate after the second cycle of antibody was 76% (11% CR), compared with 100% (54% CR) in patients who received IFN. These data support previous publications suggesting that a second course of the antibody might augment the response rate or quality of the response (Hainsworth et al: Blood 95:3052-3056, 2000). These data require confirmation in a sufficiently powered phase III trial.

Because granulocyte colony-stimulating factor (G-CSF [Neupogen]) enhances granulocyte antibody-dependent cellular cytotoxicity, it has been studied in combination with rituximab, but with no apparent increase in response rate (abstract #3167). Interleukin-12 may also augment immune-mediated cytotoxicity from rituximab, with local release of lymphotoxic cytokines such as tumor necrosis factor. The combination of those two agents has shown activity in preliminary studies (abstract #2478). Another reasonable direction is to integrate rituximab into a vaccine protocol; however, care must be exercised in the design of such trials since the depletion of B cells prior to a vaccine may prevent the expression of the necessary humoral response to the vaccine.

Another potential use for rituximab has been as an in vivo purging agent in the setting of high-dose therapy with stem cell transplantation. Several such abstracts were submitted to the ASH meeting, each with small numbers of patients (abstract #791). The data suggest the potential for clinical and molecular benefit, but larger studies are needed to confirm this impression.