Pan-Pacific Lymphoma Conference
Rituxan Review from January 2002, Maui, HI

The following is an overview summary of the most recent work done with Rituximab for NHL and CCL treatment. The paper cover the use of the monoclonal antibody alone and in combination with chemotherapy agents. Most of the information is well known to us but the summary is enlightening.

Optimizing Anti-CD20 Immunotherapy of Lymphomas and Chronic Lymphocytic Leukemia

David G. Maloney, MD, PhD, and Robert S. Mocharnuk, MD

Introduction

The island of Maui, Hawaii, served as the meeting place for researchers and clinicians who shared their experience with anti-CD20 immunotherapy at a symposium preceding the Pan Pacific Lymphoma Conference. The sessions focused on maximizing the therapeutic potential of the anti-CD20 antibody rituximab for the treatment of patients with non-Hodgkin's lymphoma (NHL) as well as chronic lymphocytic leukemia (CLL).

Single-Agent Immunotherapy of NHL A More Rational Approach to Dosing and Scheduling

Dr. David G. Maloney, Associate Member at the Fred Hutchinson Cancer Research Center in Seattle, Washington, acknowledged that rituximab has already been integrated into many clinical practices for the treatment of patients with NHL. He pointed out that rituximab has shown activity as a single agent, but this may represent only the proverbial iceberg tip of antibody therapy. Important questions that remain to be answered include the effect of rituximab on the natural course of lymphoma, especially when combined with or following standard chemotherapy. Recent clinical trials are providing interesting clinical results.

Dr. Maloney reviewed the data that led to the US Food and Drug Administration (FDA) approval of the use of rituximab in patients with NHL. The original phase 1 trial enrolled patients with relapsed low-grade NHL, who received a single dose of rituximab ranging from 10 mg/m2 to 500 mg/m2.[1] While dose-limiting toxicity was not reached at the 500 mg/m2 level, the infusion-related symptoms of fever, chills, and, more rarely, hypotension and/or dyspnea, resulted in prolonged infusion times that were not practical for outpatient administration. The protocol had predefined limits for infusion rates and allowed for the antibody treatment to be slowed or stopped in the presence of side effects. Phase 2 trials evaluated weekly dosing x 4 with doses of 125 mg/m2 to 375 mg/m2. Again, there was no dose-limiting toxicity observed at the higher doses used. Doses higher than 375 mg/m2 were not explored in this trial due to limited availability of the antibody.

FDA approval was given for use of the 375 mg/m2 weekly x 4 schedule, for treatment of patients with relapsed low-grade or follicular lymphoma, based on the pivotal trial results showing a 48% response rate among 166 patients.[2] Approval has also been given for expanding to 8 weekly doses based on a small phase 2 study. In this trial, a slightly higher response rate and longer time to progression were observed. However, the study was small and it was not a randomized comparison. Higher doses, including 500 mg/m2 each week x 8 have been used in patients with aggressive NHL, but they did not show higher activity and were associated with a higher rate of infusion-related toxicity compared with 375 mg/m2.[3]

The data obtained are more controversial in patients with CLL, a clinical setting where both dose and scheduling may need to be different. In fact, since there is a more rapid antibody elimination in these patients, trials have explored higher doses and more frequent antibody administrations. Ultimately, optimal dosing may require adjustments based on antibody pharmacokinetics, biologic effects (B-cell depletion) and clinical response.

Pharmacokinetics

Several factors are involved in the pharmacokinetics of antibody therapy. One variable among patients is the antigen burden. This is reflected in both the number of tumor cells vs normal B cells, as well as the density of CD20 molecules on the cell surface. This antigen sink will "sop up" antibody, resulting in lower serum antibody levels and more rapid clearance. Patients with rapid tumor growth will have increasing CD20 antigen burden, while patients with responding tumors will have decreasing antigen load, a process that can affect antibody levels.

Although CD20 does not undergo rapid modulation, all surface molecules turn over at some rate. There is evidence to suggest that CD20 does modulate at a slow rate and that some cells may release CD20 or cell fragments more readily than others, thus contributing to a more rapid clearance of the antibody from the serum. Lastly, the clearance of antibody from the serum, in part determined by antibody levels, is a complex process that may change owing to variable receptor expression patterns. In a limited number of studies, 2 patterns of pharmacokinetic clearance have been observed. Some patients tend to accumulate antibody during the 4 weekly infusions, with increasingly higher levels observed both immediately prior to subsequent infusions and increasingly higher peak serum levels following each infusion. In other patients, the antibody is rapidly cleared from the blood and may be almost completely eliminated before the next infusion.

Patients with accumulating antibody levels have a much higher response rate compared with patients who have a rapid clearance. However, other issues such as tumor type and tumor burden complicate this analysis further. Patients with small lymphocytic NHL are more likely to fall in the latter category and have had a low response rate; patients with follicular lymphoma generally have favorable pharmacokinetics and a higher response rate.

Clinical trials based on measured serum antibody levels should be able to answer the question of whether higher doses or more frequent antibody administration would result in higher response rates for patients with unfavorable pharmacokinetics. It is unlikely that repeated administration of standard dosing will affect response rate in nonresponding patients with poor pharmacokinetics. Conversely, repeated dosing of rituximab may prolong time to progression in responding patients.

Biologic Effects of Rituximab

Normal circulating B-cells are rapidly depleted from the peripheral blood following administration of rituximab. Recovery of these B-cells begins approximately 6 months afterward with return to the normal range in about 1 year. This is similar to the time to tumor progression that has been observed in patients with relapsed disease treated with rituximab. Retreatment with rituximab is one option that may prevent recovery of circulating B-cells and extend the duration of response.

Hainsworth and associates[4] have published data on the use of "maintenance" rituximab every 6 months following initial treatment of patients with low-grade, follicular, or small lymphocytic NHL. Eligible patients had achieved stable disease, a partial response (PR) or complete response (CR) following 4 weekly courses of rituximab. Among the 62 patients enrolled, 45% achieved either a PR or CR (6% CR) after the first course of therapy. Subsequent courses of treatment saw response rates rise to 65% with 27% CRs. Response rates after 3 or 4 courses were approximately 70%. At 2 years, 67% of responding patients remained in remission. These results appear superior to those observed in the original single-agent rituximab trials in relapsed patients, where the initial response rate was 48% with a time to progression in responding patients of approximately 12 months. Retreatment of responding patients at the time of relapse demonstrated a 40% response rate and increased duration of response when compared with their earlier response to rituximab.

Subset analysis has shown that even among patients with bulky disease, a 43% response rate was observed, which is again superior to outcomes seen in prior studies. Not surprisingly, duration of response was generally shorter at 8.1 months, compared with historic controls.

Similar to the clearance of B-cells from the peripheral blood, patients can be analyzed for evidence of cells with the t(14;18) translocation, which is present in most follicular lymphomas, using a polymerase chain reaction (PCR) specific for the translocation. Rituximab treatment results in the clearing of cells from the peripheral blood in 62% of patients and from the bone marrow in 56% of patients.

It remains to be proven whether the molecular clearance of cells from these compartments correlates with the degree and duration of response, although preliminary evidence seems to support this concept. Among patients with residual nodal disease who achieved PCR clearance, time to progression seems to be longer than in patients in whom PCR was positive or equivocal. It is possible that future studies will use molecular analysis for tumor cell detection as a surrogate measure for antibody dosing and/or antibody retreatment.

Are 8 Treatments Better Than 4?

Based upon the original phase 1 data that showed a linear relationship between antibody dose and area under the curve, with an increasing maximum concentration and longer antibody half-life, 4-week and 8-week treatment schedules have been used. In the 4-treatment trials, a more rapid antibody clearance was observed in nonresponders vs responders. Even in those patients achieving late responses, antibody clearance was more rapid than in early responders. Does this mean that nonresponding patients fail because they do not accumulate antibody, or does this simply serve as a marker for nonresponders?

A trial evaluating 8 weekly doses found that the antibody levels generally reach a plateau after 6 doses. However, some patients (generally nonresponders) continued to have poor pharmacokinetics. Conversely, responding patients had higher antibody levels that were sustained longer. The response rate was 57% among patients receiving 8 courses of rituximab, compared with the 48% response rate observed in the initial 4-treatment studies.[5] Median time to progression for responding patients was 19 months in the 8-treatment study vs 13 months in the standard treatment study. However, this was a small study and data evaluating the 4 vs 8 infusion protocol in a randomized setting have not been published yet.

Treating Earlier in the Course of the Disease

Perhaps using rituximab earlier in the course of disease will have more of an impact. The initial studies suggested that first relapse response rates were superior to the response rates seen after second or third relapses. The French Lymphoma Group addressed this issue by administering up-front rituximab to 49 treatment-naive patients as 4 weekly infusions.[6] An initial response rate of 73% was observed (26% CRs, 49% PRs), although this was later updated to 80% after additional follow-up.

Elimination of the t(14;18) translocation from the blood and bone marrow was recorded in 17 patients, 10 who had achieved CR; and at 1 year following completion of therapy, only 1 of these patients had progressed. Thirteen patients who responded to rituximab continued to express t(14;18), but none of these were complete responders. At 1 year, 8 of these 13 persistent PCR-positive patients had disease progression (62%). These data support the premise that molecular responses may correlate with clinical responses, and that molecular remissions may predict for longer time to tumor progression.

Rituximab in Combination Treatment of Lymphoma

Dr. Julie Vose, of the University of Nebraska, reviewed what is known about combining rituximab with cytotoxic chemotherapy in the treatment of NHL. In spite of many questions and controversies regarding combination therapy, it is clear that the clinical community has eagerly combined rituximab with conventional chemotherapy combinations. This has been based on the single-agent activity of rituximab, the nonoverlapping toxicities observed, the different mechanism of action, and the potential synergy with chemotherapeutic agents. A wide number of chemotherapy regimens have been combined with rituximab, with the cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) regimen leading the way.

Low-Grade NHL

Dr. Czuczman and colleagues[7] published the first data on combination therapy using CHOP combined with 6 doses of rituximab for the treatment of 38 patients with low-grade lymphoma, the majority (32) of which had received no prior treatment. A 100% response rate was observed with 55% CRs. Median time to progression had not been reached at 48 months. These results must be interpreted cautiously since the number of patients was small and since study entrants were not prospectively randomized. PCR analysis was also done on a handful of patients in this study. Eight patients who were PCR positive for Bcl-2 became negative by both blood and bone marrow assays following combination treatment.

Five of 7 patients remained PCR negative up to 38 months following completion of therapy. As observed previously, even patients who achieve only PRs with residual nodal disease were found to be PCR negative.

Rituximab has also been combined with other chemotherapy regimens besides CHOP. Czuczman and associates[8] treated an initial cohort of 10 indolent lymphoma patients with rituximab given between cycles of fludarabine. Six of 7 patients completed therapy, and all 6 patients achieved CRs. Three patients discontinued treatment secondary to myelosuppression. Among those completing therapy, a significant amount of neutropenic fevers were observed with some associated opportunistic infections, necessitating the use of growth factor support in most.

Because of this, the regimen was modified, with discontinuation of prophylactic antibiotics (secondary to concerns regarding drug-induced myelosuppression), and a 40% reduction in fludarabine from 5 consecutive doses per treatment cycle to 3. Twenty patients were treated with these modifications and 17 patients completed therapy. CRs were observed in 16, and only 1 of 3 patients discontinued therapy on account of myelosuppression. Fewer infectious complications occurred and significantly less growth factor support was required. A handful of patients did require dose reductions starting with cycles 3 or 4. An objective response rate of 93% was recorded, with 80% CRs (both confirmed and unconfirmed). Median duration of response was greater than 14 months with some patients in CR after more than 2 years. Nine of 9 patients screened for Bcl-2 in blood converted from positive to negative, while 6 of 7 became negative by bone marrow analysis.

Investigators at the MD Anderson Cancer Center in Houston, Texas, also studied rituximab in combination with fludarabine, mitoxantrone, and dexamethasone (FND) among 134 patients with low-grade NHL.[9] Patients were randomized to receive 8 cycles of either concurrent FND with rituximab or FND followed by rituximab. Interferon was also administered for 1 year following completion of treatment. PCR analysis of blood and bone marrow specimens was done every 6 months.

Baseline Bcl-2 rearrangement was found among 70 patients (74%) in blood or bone marrow at time of initial testing. At 12 months follow-up, 79% of specimens tested in the concurrent arm and 92% of the specimens tested in the sequential arms were seronegative for Bcl-2. Similarly, 73% of concurrent and 67% of sequential arm marrow specimens were negative for Bcl-2 at 12 months.

Aggressive NHL

Clinical data suggest that rituximab may be useful in the treatment of aggressive lymphomas. Most clinicians are familiar with the data published by Dr. Richard Fisher[10] showing equivalency between standard CHOP and 3 other more complicated treatment regimens, in which 5-year disease-free survival rates were 35% to 40%. Using these figures as historic benchmarks, a multicenter phase 2 trial administered CHOP plus rituximab to 33 patients with aggressive lymphoma.[11]

Patients received rituximab 375 mg/m2 on day 1 of each 3-week cycle, followed by CHOP on day 3. All patients received all 6 scheduled treatment cycles. Most of these patients had either diffuse large B-cell or follicular large-cell histology. Twenty patients (60%) had CR while 11 patients (33%) had PR. Initially, there was disappointment with these numbers, considering the equivalent historic response rates to CHOP therapy alone. Yet, it became clear after time that many of those patients deemed partial responders continued to respond and remained failure-free over a long period of time.

No differences in response were observed between patients younger or older than 60 years. Patients with higher International Prognostic Indicator (IPI) scores had a slightly lower response rate when compared with their low IPI counterparts, but this was not statistically significant, given the small number of patients studied. With a minimum of 2 years follow-up, over 90% of patients with an IPI score of 2 or less remained disease-free. For patients with an IPI score greater than 2, event-free survival at 2 years was still greater than 80%. While these results are intriguing, the small patient numbers argue for phase 3 testing before definitive conclusions can be drawn.

The most important trial to date is the randomized phase 3 trial in which elderly patients with diffuse large B-cell NHL were randomized to receive CHOP vs CHOP plus rituximab (given concurrently on day 1). This trial has been presented in preliminary form and was recently updated at the May American Society of Clinical Oncology meeting in San Francisco.[12] As reported by the Groupe d'Etude des Lymphomes de l'Adulte (GELA), 159 patients were treated with 8 cycles of CHOP while 169 patients received 8 cycles of CHOP plus rituximab. It should be noted that 60% of patients enrolled had IPI scores of 2 or more.

Consistent with previous historic data, the control CHOP arm recorded a complete response rate of 60%. There was a 75% complete response rate in the CHOP plus rituximab treatment arm. Median follow-up was only 12 months, but there was statistically superior event-free and overall survival in the combination arm.

The Eastern Cooperative Oncology Group trial E4494, which randomized patients to 8 cycles of CHOP vs 8 cycles of CHOP plus rituximab, has completed accrual of over 600 patients, and will be an important confirmatory study to the GELA data when analyzed. This study follows the original Czuczman format, in which rituximab was interspersed between CHOP treatments. Unlike the GELA trial, patients will also be randomized to 4 weekly doses of rituximab maintenance every 6 months for 2 years vs no maintenance therapy.

Clearly, combination therapy with rituximab for NHL has arrived, and, as indicated by several questions following Dr. Vose's formal presentation, its use as a radiosensitizer, as an adjunct to treatment in patients with mantle-cell lymphoma, and as a purging agent in stem cell transplant regimens continues to be explored.

Single-Agent Treatment of CLL

Rituximab as Single-Agent Treatment in CLL

Dr. Susan O'Brien, of the MD Anderson Cancer Center in Houston, Texas, concluded this symposium by focusing on the role of rituximab in CLL. As a reference point, the initial phase 1 and 2 data discussed previously included a small number of CLL patients, for whom response rates were particularly low. Possible explanations for this difference in response include less CD20 expression on cell surface membranes as well as more rapid antibody clearance among nonresponders. As discussed by Dr. Maloney, perhaps dosing and scheduling need to be customized for individuals with CLL histology. Based upon hematopathology data generated at the MD Anderson Cancer Center, there seems to be circulating CD20 antigen in CLL patients, resulting in extratumoral antibody binding. This has been confirmed by Western blot analysis as well as by enzyme-linked immunosorbent assay among 106 patients with CLL compared with 20 normal controls.[13] These data also appear to link higher levels of circulating CD20 expression with more advanced stage disease. What is not known is whether higher baselines of circulating CD20 are present in nonresponders.

While these hypotheses are being tested in ongoing prospective trials, Dr. O'Brien and colleagues shied away from standard doses of rituximab in their initial single-agent CLL trials, based, in large part, on the low response rates noted by other investigators. Given the lack of dose-limiting toxicity observed in the phase 1 studies with rituximab, Dr. O'Brien initiated phase 1 testing for CLL at an initial dose of 375 mg/m2, to be escalated on weeks 2, 3, and 4.[14] The highest dose given was 2250 mg/m2.

Forty of the 50 patients included had CLL, while the remaining patients had lymphomas with high levels of circulating cells and included marginal-zone lymphomas, mantle-cell lymphomas, and prolymphocytic leukemia. All patients except 1 with marginal-zone lymphoma had been previously treated with a median number of 2 prior regimens. Half of the CLL patients were refractory to fludarabine (ie, they failed to respond to their last fludarabine treatment). Mild grade 1 and 2 fevers and chills were observed in 48 of 50 patients (94%) treated with standard-dose rituximab. Six patients (12%) experienced grade 3 and 4 toxicity, including high fevers, chills, dyspnea, and hypoxia. Five patients became hypotensive and 1 patient became hypertensive. Only 1 of the 40 CLL patients experienced severe toxicity, whereas the other 5 cases were observed in the remaining histologic types.

While circulating CD20 may be responsible in part for this, it is also known that CD20 expression in mantle-cell and marginal-zone lymphoma is much more intense than in CLL. Maybe the combination of both circulating antigen and antigen overexpression accounts for the degree of toxicity experienced.

With dose escalation at week 2, only 3 patients experienced mild toxicity, even at doses of 1500 mg/m2. At the 2250 mg/m2 dose level, 8 of 12 patients experienced moderate (grade 1 and 2) toxicity, including fevers, chills, and malaise. Response rates ranged from 23% at the lowest dose delivered, to 80% at the maximum dose delivery, although little bone marrow clearance was observed. Lymph node size did not appear to predict for response, although prior resistance to fludarabine appeared to correlate with lower responses. A trend to lower response in more advanced stage was observed, but this did not reach statistical significance.

Median time to progression was approximately 8 months, shorter than the time to progression observed in the NHL trials. This trial confirmed that there was significant activity for rituximab in the treatment of CLL, and that responses appeared to be dose-related, although it failed to establish criteria for frequency of treatment or optimal dosing. Dr. John Byrd and colleagues,[15] at the Walter Reed Army Medical Center in Washington, DC, also initiated a dose-escalation trial in CLL, the results of which were recently published. Because of concerns regarding first dose toxicity, all patients received a limited dose of 100 mg of rituximab on day 1. This was followed by thrice weekly administration of 250 mg/m2 for the first 3 patients, with all remaining patients dose-escalated to a standard dose of 375 mg/m2 given 3 times weekly for 4 weeks.

Little toxicity was observed after the first dose, and this led to speculation that subsequent doses could be administered over a 1-hour time period, which they were. No increased toxicity was observed. Of the 33 patients enrolled, 31 had CLL and 2 had small lymphocytic lymphoma. Patients had received a median number of 2 prior treatments, although there were a handful of treatment-naive CLL patients.

The overall response rate observed in this study was 45%, with a trend to lower response rate in more advanced stage disease, and higher responses among the previously untreated (6 of 7 patients responded). A lower response rate was not observed among the fludarabine refractory patients, nor were there response differences based upon size of lymph nodes. An assay was developed to measure surface CD20 expression, but no correlation between response and expression was made.

The presence of first-dose reactions was also analyzed for response, and no differences were noted. Toxicity was still significant, even when a low dose of 100 mg of rituximab was given, but as observed previously in the MD Anderson data, no significant toxicity was observed during or beyond dose 3. Toxicity profiles did not appear to be influenced by patient age, elevation of initial lymphocyte count, or cell surface expression of CD20.

Combination Therapy for CLL

In an abstract presented during the last American Society of Hematology (ASH) meeting, an in vitro t(14;18) cell line was incubated with complement, then exposed to either single-agent rituximab, fludarabine, or both drugs.[16] Cell lysis was most dramatic when the 2 drugs were combined. Downregulation of CD55 was observed following fludarabine treatment. Since cell surface CD55 is known to protect against complement-mediated lysis, it was hypothesized that downregulation of CD55 by fludarabine potentially increased the efficacy of rituximab-induced, complement-mediated cytotoxicity. Independent of the intriguing science behind this abstract, a trial was launched at MD Anderson Cancer Center, inspired by the success of other combination therapy trials in other hematologic malignancies.

Building on a reference regimen of daily fludarabine 30 mg/m2 plus cyclophosphamide 300 mg/m2 for 3 days (FC), rituximab 375 mg/m2 was added. During the first cycle, rituximab was given on day 1, followed by 3 days of concurrent FC (FRC). In subsequent courses, all 3 drugs were started on day 1, and rituximab was increased to 500 mg/m2. Among the first 8 patients treated, 2 experienced severe tumor lysis, and 1 required temporary hemodialysis. The white blood cell count dropped precipitously but no other myelosuppression was observed. For these reasons, subsequent doses of fludarabine and cyclophosphamide were decreased to 25 mg/m2 and 250 mg/m2, respectively. Among the 59 chemotherapy-naive CLL patients thus treated, grade 3 and 4 toxicity was noted in 7% at the time of first infusion. Virtually no B-cells were detectable in the marrow after 3 cycles.

In an as yet unpublished update expanding upon a recently published ASH abstract,[17] 55 patients have completed 3 courses of treatment while 42 have completed 6. The complete remission rate stands at approximately 60%. Historically, the fludarabine plus cyclophosphamide combination has produced CRs in the 30% to 35% range, although conclusions regarding the superiority of FRC over FC are difficult to draw since there is no comparative internal control group in this study.

Among 12 patients achieving CRs, 6 became PCR-negative by bone marrow analysis. Duration of response cannot be determined at this short follow-up interval, particularly when one considers that the historic median time to progression following FC in treatment-naive patients is approximately 4 years. If remission duration continues to parallel response rate, then one can anticipate marked prolongation in those patients responding to FRC.

The Cancer and Leukemia Group B also conducted a trial in which patients were randomized to receive fludarabine plus rituximab concurrently vs fludarabine followed by rituximab.[18] Those patients randomized to receive concurrent treatment were given rituximab on days 1 and 4 during the first cycle, then on day 1 thereafter for a total of 6 cycles. Fludarabine was given in a 25 mg/m2 day 1-5 schedule. In the sequential arm, fludarabine was given for 6 cycles, followed by 4 weekly treatments of standard-dose rituximab.

Among the 104 patients enrolled, a higher incidence of neutropenia was observed in the concurrent treatment arm, although no differences were noted in the degree of anemia or thrombocytopenia. Response rates were approximately twice as high in the concurrent drug arm, compared with the fludarabine-only arm prior to administration of rituximab, echoing the MD Anderson findings seen with FRC.

When "consolidation" rituximab is factored in, the overall response rate for the concurrent treatment arm was still significantly higher, although there was no difference in time to progression. Again, these results must be interpreted cautiously due to short follow-up and small patient numbers.

Upregulation of cellular-surface CD20 expression may play a role in future treatment strategies, and studies are currently ongoing looking at various stimulatory agents, including interferon, interleukin-4, and granulocyte-macrophage colony stimulating factor (GM-CSF). Another ongoing trial has combined monoclonal antibody against the CD52 receptor (alemtuzumab) with rituximab, and is showing an initial response rate of approximately 40% across all patient types (chemotherapy refractory, chemotherapy nonrefractory, etc.).

Conclusions

Clearly, the use of monoclonal antibody therapy will continue to escalate as more becomes known about their specific mechanisms of action as well as their interactions with chemotherapy, radiation, novel agents, and other antibodies. Clinical trials are moving antibody therapy into the forefront of treatment of many hematologic and solid malignancies, and for the first time in a long time, it seems that these agents can significantly alter survival patterns that have resisted change with other treatment strategies. The next few years of investigation promise to be exciting ones.

References

1. Maloney DG, Liles TM, Czerwinski DK, et al. Phase I clinical trial using escalating single-dose infusion of chimeric antiCd20 monoclonal antibody in patients with recurrent B-cell lymphoma. Blood.1994;84:2457-2466.

2. McLaughlin P, Grillo-Lopez AJ, Link BK, et al. Rituximab chimeric therapy for relapsed indolent lymphoma: half of patients respond to a four-dose treatment program. J Clin Oncol. 1998;16:2825-2833.

3. Coiffier B, Haioun C, Ketterer N, et al. Rituximab for the treatment of patients with relapsing or refractory aggressive lymphoma: a multicenter phase II study. Blood. 1998;92:1927-1932.

4. Hainsworth JD, Burris H III, Morrissey L, et al. Rituximab as first-line and maintenance therapy for patients with indolent non-Hodgkin's lymphoma. Proc Am Soc Clin Oncol. 2001;20:294a. Abstract 1175.

5. Piro L, White CA, Grillo-Lopez AJ, et al. Extended rituximab (anti-CD20 monoclonal antibody) therapy for relapsed or refractory low-grade or follicular non-Hodgkin's lymphoma. Ann Oncol. 1999;19:655-661.

6. Solal-Celigny P. Rituximab as first-line monotherapy in low-grade follicular lymphoma with low tumor burden. Anticancer Drugs. 2001;12:S11-S14.

7. Czuczman MS, Grillo-Lopez AJ, White CA, et al. Treatment of patients with low-grade B-cell lymphoma with the combination of chimeric anti-CD20 monoclonal antibody and CHOP chemotherapy. J Clin Oncol. 1999;17:268-276.

8. Czuczman MS, Fallon A, Scarpace A, et al. Phase II study of rituximab in combination with fludarabine in patients with low-grade or follicular B-cell lymphoma. Blood. 2000;96:729a. Abstract 3154.

9. Cabanillas F, McLaughlin P, Hagemeister F, et al. Molecular responses with FND + Rituxan chemo-immunotherapy for stage IV indolent follicular non-Hodgkin's lymphoma. Blood. 2000;96:330a. Abstract 1429.

10. Fisher RI. Treatment of aggressive non-Hodgkin' lymphomas. Lessons from the past 10 years. Cancer. 1994;74:2657-2661.

11. Vose JM, Link BK, Grossbard ML, et al. Phase II study of rituximab in combination with CHOP chemotherapy in patients with previously untreated, aggressive non-Hodgkin's lymphoma. J Clin Oncol. 2001;19:389-397.

12. Coiffier B, Lepage E, Gaulard P, et al. Prognostic factors affecting the efficacy of rituximab plus CHOP therapy in elderly patients with diffuse large B-cell lymphoma. Results of a randomized GELA trial. Proceedings Am Soc Clin Oncol. 2001;20:283a. Abstract 1131.

13. Almasri NM, Duque RE, Iturraspe J, et al. Reduced expression of CD20 antigen as a characteristic marker for chronic lymphocytic leukemia. Am J Hematol. 1992;40:259-263.

14. O'Brien SM, Kantarjian H, Thomas DA, et al. Rituximab dose-escalation trial in chronic lymphocytic leukemia. J Clin Oncol. 2001;19:2165-2179.

15. Byrd JC, Murphy T, Howard RS, et al. Rituximab using a thrice weekly dosing schedule in B-cell chronic lymphocytic leukemia and small lymphocytic leukemia. J Clin Oncol. 2001;19:2153-2164.

16. Golay J, Xiao YM, Di Gaetano N, et al. Fludarabine synergises with anti-CD20 monoclonal antibody rituximab in complement mediated cell lysis. Blood. 2001;96:339a. Abstract 1463.

17. Keating MJ, O'Brien SM, Lerner S, et al. Combination chemo-antibody therapy with fludarabine, cyclophosphamide, and rituximab achieves a high CR rate in previously untreated chronic lymphocytic leukemia. Blood. 2000;96:514a. Abstract 2234.

18. Byrd JC, Peterson B, Park K, et al. Rituximab added to fludarabine improves response in previously untreated chronic lymphocytic leukemia: preliminary results from CALGB 9712. Proc Am Soc Clin Oncol. 2001;20:280a. Abstract 1116.