Introduction

At the 2006 meeting of the American Society of Clinical Oncology (ASCO) and the 2005 meeting of the American Society of Hematology last December, important results of clinical trials in indolent lymphoma were presented. For patients with low grade or follicular non-Hodgkin’s lymphoma, several large randomized trials have now suggested survival advantages associated with the inclusion of Rituxan® (rituximab) in initial or salvage therapy. In addition, clinical trials with radio-labeled antibodies continue to show encouraging results both as single agents, and following standard chemotherapy, especially in follicular NHL. Selected abstracts and presentations will be reviewed here.

Phase III Randomized Trials in Follicular NHL

At ASH 2005 and ASCO 2006, the results or updates of four important randomized clinical trials were presented. The trials comprised incorporating Rituxan into either the frontline treatment or relapsed treatment for patients with low-grade (predominately follicular) NHL.

R-CVP vs. CVP: ASH Abstract #350

First are the updated results of the European trial comparing CVP (cyclophosphamide, Oncovin®, prednisone) with R-CVP (Rituxan, CVP) as initial therapy in 321 patients with advanced stage follicular NHL. This study was originally presented and published by Marcus et al. and updated at ASH 2005 by Dr. Solal-Celigny, now with a 42-month median follow-up.[1],[2] The two randomized groups were well balanced for prognostic factors. The addition of Rituxan (375 mg/m2) to each of 8 cycles of CVP chemotherapy administered every 3 weeks was associated with improved outcome. This included higher response rates (81% vs. 57%, p < 0.0001), CR rate (41% vs. 10%, p < 0.0001), response duration (37.7 vs. 13.5 months, p < 0.0001), and time to tumor progression (33.6 vs. 14.5 months, p < 0.0001). In addition, the estimated 3-year survival rate was 89% for the R-CVP group vs. 81% for the CVP group. This achieved borderline statistical significance with a two sided p = 0.0553 test. It is important to note that these patients treated in this clinical trial had relatively poor prognostic factors. Using the FLIPI index, the majority of patients had high FLIPI score of 3-5, (poor prognosis ~50%) or FLIPI 2, (intermediate prognosis ~43%).[3] The benefit of Rituxan was observed in all patient groups and generally across all of the patient disease characteristics. While there were slightly more adverse events noted in the Rituxan containing arm, these were generally mild to moderate and did not lead to a higher rate of withdrawal from the study. Of note, the dose of cyclophosphamide used in the CVP for this trial was lower than often used in U.S. trials (750 mg/m2 compared to 1000 mg/m2). This, in addition to the high FLIPI risk scores, may have contributed to the relatively poor outcome of the chemotherapy only group.

One possible explanation, though unlikely, as to why there appears to be a survival advantage emerging for the Rituxan treated group in this European study may in part be due to limited access to Rituxan following relapse in the control group. Data on the use of salvage Rituxan, especially for patients in the non-Rituxan containing chemotherapy arm have not been presented. Nonetheless, it is clear that the use of up-front Rituxan in this large randomized study improved outcome and was associated with borderline statistical significant improvement in estimated 3-year survival.

CVP Followed by Rituxan vs. Observation: ASH Abstract #349

The second important trial in follicular NHL was presented by Dr. Horning et al. for the Eastern Cooperative Oncology Group and the Cancer and Leukemia Group B and was a follow-up analysis of the E1496 phase III trial of sequential Rituxan vs. observation for patients with SD/PR or CR to CVP therapy.[4] This study has been previously reported at ASCO 2004.

Patients with low-grade NHL were treated with 6-8 cycles of CVP chemotherapy (cyclophosphamide dose of 1,000 mg/m2) to best response. Patients with at least SD were randomized to observation vs. 4 weekly doses of Rituxan (375 mg/m2) every 6 months for up to 2 years. Patients were stratified based on histology and response to chemotherapy. The Rituxan started 4 weeks following the last cycle of CVP chemotherapy.

At ASH 2005, extended data was presented on the outcome of 237 patients with follicular NHL representing 78% of the 305 patients with responding disease eligible for randomization to Rituxan vs. observation. Compared to the European trial discussed earlier, this patient group had slightly better prognostic factors (~37% high FLIPI score). The two randomized groups were similar in their patient characteristics: FLIPI risk breakdown, disease burden and response to CVP chemotherapy. The median progression-free survival (PFS) from the time of randomization was 15 vs. 61 months, comparing observation vs. Rituxan, respectively. The PFS at 3 years from randomization was 36% vs. 62% (p = 0.0000003) and was statistically improved in patients with FLIPI 0-2, FLIPI 3-5, low or high tumor burden and minimal or gross residual disease.

The overall survival also favored the Rituxan treated group with a Log-rank one-sided p = 0.03 and a HR = 0.5 (0.3-1.1). The overall survival at 3 years from randomization was 92% vs. 83%, respectively. The entire group had excellent 3-year survival with 21 deaths in the observation group vs. 12 deaths in the Rituxan-treated group. While subgroup analyses had too few events to be statistically significant, patients with high tumor burden at study entry or gross residual disease following CVP had the most striking improvement in outcome.

Thus, sequential Rituxan following the initial treatment with CVP chemotherapy had a strong trend toward improved survival in these advanced stage follicular NHL patients. These results are somewhat surprising as they suggest that it is important to include the Rituxan early in the initial treatment of the disease rather than waiting for relapse. Although no data is provided, it is also more likely that in the United States, patients in the observation arm would receive Rituxan (and likely have a high rate of response) at the time of relapse/progression. However, it appears that this subsequent treatment was not able to offset the improvement in survival in the early sequential treatment arm. This observation, if confirmed by longer follow-up and detailed analysis of the study, will have important implications in how patients with follicular NHL should be initially treated.

R-CHOP vs. CHOP and Maintenance Rituxan vs. Observation—ASH Abstract #353

The third key trial in follicular NHL was the final analysis of the phase III EORTC #20981 trial presented by Van Oers et al. The study concerned the use of CHOP chemotherapy with or without Rituxan in induction followed by observation or maintenance Rituxan.[5]

This trial enrolled 465 patients with follicular NHL (Grade 1, II, III), stage III/IV at diagnosis, who had been previously treated with a maximum of two non-anthracycline-containing systemic regimens. It is important to note that in comparison to the first two studies discussed here, this trial was for patients who had previously been treated with chemotherapy but not Rituxan. Patients were initially randomized to induction therapy with R-CHOP vs. CHOP (every 21 days, maximum 6 cycles). In a second randomization, patients who had a PR or CR to the treatment were randomized to observation vs. Rituxan maintenance (375 mg/m2 every 3 months for 2 years or until relapse). In this study, the majority of patients had failed one prior regimen (82%), and about half of the patients had a PR as their best response to prior chemotherapy. The two groups were well balanced (R-CHOP vs. CHOP).

In the induction phase, the CR rate was superior in the R-CHOP arm (29.5 vs. 15.6%), the rate of progressive disease lower (2.6 vs. 9.5%), and the median time to tumor progression longer (33.1 vs. 20.2 months, p = .0003). The 3-year overall survival however was not statistically different, although favoring the R-CHOP arm (82.5% vs. 71.9%, p = 0.096).

For the secondary randomization, 334 patients were randomized to observation vs. maintenance and the progression-free survival was 51.6 months vs. 15 months favoring the maintenance arm. Interestingly and importantly, this advantage was observed not only in the subgroup analysis of patients who had received CHOP (42.2 vs. 11.6 months, p < 0.0001), but also in the subgroup of patients who had received R-CHOP induction (51.9 vs. 23.1 months, p = 0.004). A similar degree of improvement was also seen in patients who had either a CR or PR following induction therapy. After the second randomization, the OS was higher in the subgroup of patients randomized to maintenance Rituxan. While only subgroups were presented, the overall log-rank test following CHOP was p = 0.073, and following R-CHOP was p = 0.059. Thus, not only did Rituxan improve the outcome of salvage CHOP chemotherapy, the addition of maintenance Rituxan following R-CHOP or CHOP chemotherapy further improved progression free survival and overall survival.

Comment:  These studies suggest that Rituxan is improving outcome in patients with follicular NHL. At ASH 2005, Schulz et al. reported on a meta-analysis of five randomized clinical trials evaluating Rituxan with chemotherapy, including 994 randomized patients.[6] Overall survival was improved in the R-chemotherapy treated population with a HR of 0.61 (95% CI: 0.47-0.80). Together, the data from these trials suggest that there should be a significant change in the management of patients with follicular NHL. Unless enrolled on other clinical trials evaluating novel approaches, drugs or comparative trials, patients who require therapy should receive Rituxan as part of the induction therapy and/or should receive a prolonged course of Rituxan therapy. The ECOG 1496 trial shows that the 2-year course of sequential Rituxan for patients whose disease had responded to CVP was associated with a survival advantage (trend). As these patients were treated in the United States, it would be expected that one of the first treatments for patients randomized to the observation group upon relapse would be to receive either Rituxan or Rituxan in combination with a second course of chemotherapy. The fact that there was a survival advantage associated with the maintenance therapy suggests that this difference could not be recovered upon subsequent treatment. This goes against the general presumptions regarding the therapy of indolent NHL that it does not matter what treatment is given first or second as it will all be equal in the end. Complicating the story are the observations from the EORTC study demonstrating that Rituxan maintenance was beneficial following Rituxan containing chemotherapy (R-CHOP). While this was in relapsed patients, the temptation is to extend the observations from this trial to upfront therapy is compelling. While there is currently no data directly addressing this question (role of Rituxan maintenance following Rituxan-chemotherapy induction) there is no obvious reason why this would not be true in the upfront setting.

Current Issues

Additional questions that urgently still need to be addressed include the simple question: What is the best initial chemotherapy to be combined with Rituxan? These findings of survival advantages associated with the use of Rituxan make the previous findings that there was no particular advantage of one regimen over another in the long term more doubtful. Comparison of trials of CVP, CHOP, FND or MCP have all now demonstrated that the addition of Rituxan has been associated with improved outcome, including survival in several studies. While not reported in all trials, the FLIPI prognostic index may eventually allow comparison of results in specific prognostic subgroups. One preliminary observation that can be made is that the magnitude of benefit of Rituxan appears to be greater in the higher risk groups. Hopefully, large clinical trials using these regimens that are exploring the use of Rituxan maintenance following Rituxan-chemotherapy will provide answers in the coming years.

Radio-Immunotherapy With Anti-CD20 Antibodies (RIT)

The intergroup study led by SWOG (S0016) is ongoing and is comparing treatment of advanced stage follicular NHL with R-CHOP vs. CHOP followed by Bexxar® (131I tositumomab). The use of the RIT regimen is based on the outstanding results from a phase II study by Press et al. that was also updated at ASH 2005.[7] In this study the 4-year survival was 91% and the 5-year PFS was 70%. In a similar approach DeMonaco et al. reported on the treatment of 30 patients with follicular NHL with R-CHOP followed by Zevalin® (90Y-ibritumomab tiuxetan).[8] They observed a 36% CR rate after the R-CHOP that increased to 89% following the Zevalin. Toxicity was mainly hematologic as expected. It is not clear if either of the available RIT agents (Bexxar or Zevalin) would have an advantage over the other in these settings.

In the absence of stem cell support, RIT must be sequenced either before or following cytoreductive chemotherapy. However, when the RIT is used as an adjuvant, there is less tumor, which would allow less targeting of the RIT to the tumor and possibly less anti-tumor effect, but with continued hematologic toxicity. Zelenetez et al. reported on an interesting trial in 24 patients with newly diagnosed mantle cell NHL (<25% intrabecular bone space involvement with NHL) treated with upfront RIT, followed by CHOP chemotherapy.[9] Patients first received Bexxar, followed by CHOP. After the RIT, a response rate of 83% (46% CR/Cru) was observed, including a clearance of cells with the 11:14 translocation as detected in the peripheral blood by PCR in 46%. Unfortunately, following the CHOP chemotherapy given as consolidation the response rate was only 86% with CR/Cru in 67% and no further increase in clearance of molecular disease.

In an opposite approach, Smith et al. used 4 cycles of R-CHOP as induction therapy in 56 patients with newly diagnosed mantle cell NHL, followed 4-8 weeks later by RIT with Zevalin.[10]

After the R-CHOP, in 50 evaluable patients the overall response rate was 72% with 14% CR/CRu. After the RIT, the overall response rate was 84% with 45% CR and 15/37 patients with less than a CR/CRu to the R-CHOP chemotherapy improved their response following the RIT. At ASCO, Weigert et al. also reported on two European studies using Zevalin either upfront or following cytoreduction with chemotherapy for patients with relapsed mantle cell NHL.[11] The upfront RIT alone had a response rate of 33% (2/6) with a median duration of response of only 3.9 months. In contrast, chemo-induction induced 2 CR and 14 PR and following RIT 7/14 PR patients converted to CR. In this arm 13/16 (81%) remain in remission. The conclusion from this trial in the relapsed setting was that RIT was more effective as consolidation following chemotherapy.

Another approach has been to add standard dose RIT to myeloablative chemotherapy supported by stem cell transplant. Shimoni et al. used standard dose Zevalin on day 14 followed by the BEAM transplant regimen (d-6) in 20 patients with aggressive NHL.[12] There was a high response rate with 14 CR and 4 PR. Two patients died of transplant-related complications early and 2 died of late infections. At 12 months follow-up, the overall survival was 59%.

Several studies also explored the use of high-dose RIT with PBSC support. A study reported by Flinn et al. used high dose Zevalin.[13] Of the 22 patients with relapsed NHL enrolled in the study, 17 have been treated with doses up to 28 Gy. One patient died of toxicity, 10 had CR and 3 PR. Only four patients required admission to the hospital for treatment. Stem cells were given when radiation levels were less than 1 cGy/hr. Vanazzi et al. have explored the use of dose escalated Zevalin in doses of 0.8 to 1.5 mCi/kg with PBSC support given on day 13.[14] Fourteen patients with relapsed NHL have been treated with 6 CR to date. The treatment was well tolerated with the expected hematologic toxicity. While these preliminary studies are exciting, long-term follow-up for duration of remission and potential late toxicity is required.

Conclusion

Exciting data on new agents and maturing data from large randomized clinical trials in the treatment of patients with lymphoma have been reported at ASH and ASCO. By far, the greatest impact has been from the addition of Rituxan to standard chemotherapy for the treatment of both follicular and large B cell NHL, where it has become the standard of care. Newer approaches with novel antibodies, idiotype vaccines or second generation anti-CD20 mAbs are being evaluated with encouraging results, but require definitive clinical trials to establish their role in NHL therapy. RIT represents one of the most active single agents in the treatment of NHL and incorporation as a single agent or as consolidation following induction therapy appears very promising. Challenges as to how to best incorporate these new treatments to standard will become increasingly difficult as the number of new agents increases.

References