The low-grade NHLs remain incurable with conventional therapies. A variety of new, unique agents are being evaluated, which, hopefully, will improve these results. The purine analogs, particularly fludarabine (Fludara), have shown impressive activity in these patients and have become part of the standard treatment armamentarium. In patients who have not responded to an alkylating agent, fludarabine achieves a 12% to 15% complete remission rate with an overall response rate of about 50%. When used as initial therapy, fludarabine produces a complete remission in almost 40% of patients and an overall response rate of 60% to 70% (Solal-Céligny: J Clin Oncol 14:514-519, 1996).
The low-grade NHLs remain incurable with conventional therapies. A variety of new, unique agents are being evaluated, which, hopefully, will improve these results. The purine analogs, particularly fludarabine (Fludara), have shown impressive activity in these patients and have become part of the standard treatment armamentarium. In patients who have not responded to an alkylating agent, fludarabine achieves a 12% to 15% complete remission rate with an overall response rate of about 50%. When used as initial therapy, fludarabine produces a complete remission in almost 40% of patients and an overall response rate of 60% to 70% (Solal-Céligny: J Clin Oncol 14:514-519, 1996).
Hagenbeek et al (abstract #1294) compared single-agent fludarabine with monthly cyclophosphamide, vincristine, and prednisone (CVP) in previously untreated patients with advanced-stage low-grade NHL. The overall and complete response rates were significantly higher in the fludarabine-treated group, with a trend toward a prolongation of time to progression. Fludarabine was associated with more myelotoxicity, but with no difference in the frequency of severe infections.
Thus, advantages of the nucleoside analogs include higher response rates, less cumulative myelosuppression, and a lower likelihood of secondary malignancies. These data provide additional support for fludarabine as an initial treatment option for patients with indolent NHL.
Nevertheless, relapse occurs uniformly in patients with low-grade NHL, and a number of combination regimens have been developed to improve on the activity of single agents. Fludarabine interferes with DNA repair followingexposure of cells to a number of drugs, including topoisomerase II inhibitors, such as mitoxantrone (Novantrone), and alkylating agents.
One of the most promising regimens is the combination of fludarabine, Novantrone, and dexamethasone (FND). In a series of 51 patients, McLaughlin and coworkers at the M. D. Anderson Cancer Center reported a complete remission rate of 47% and a partial remission rate of 47%, with the complete remissions lasting a median of 21 months (McLaughlin et al: J Clin Oncol 14:1262-1268, 1996). The Southwest Oncology Group (SWOG) recently completed a phase II trial of a fludarabine-mitoxantrone regimen as initial therapy for low-grade NHL, the results of which will be published in the near future.
Foussard et al (abstract #1295) randomized 68 patients with at least one adverse feature to either FM (fludarabine plus mitoxantrone) or CHEP (cyclophosphamide, doxorubicin, Eldisine, and prednisone). Response and failure rates significantly favored FM. This small study suggests that the FM regimen is a reasonable initial treatment option for patients with low-grade NHL.
Rummel et al (abstract #1705) developed a similar regimen of CdA with mitoxantrone, which they used in 35 patients, 20 of whom were untreated. The response rate of 85%, including a 42% complete remission rate, is respectable; however, the study is difficult to interpret since duration of response is not provided.
Eastern Cooperative Oncology Group (ECOG) investigators (Hochster et al: Proc Am Soc Clin Oncol 17:17a [abstract 66], 1998) recently published the updated results of their phase II trial of the combination of fludarabine and cyclophosphamide in 27 previously untreated patients; the response rate was 100%, including an 89% rate of complete remissions. These results stimulated the development of an important ongoing ECOG phase III trial in which this fludarabine-cyclophosphamide regimen is being compared to cyclophosphamide, vincristine, and prednisone (CVP); responders undergo a secondary randomization to rituximab (Rituxan) or no additional therapy.
At the ASH meeting, Flinn et al (abstract #1706) reported their results with another variation of fludarabine and cyclophosphamide in previously untreated low-grade NHL. They treated 19 patients with follicular lymphoma, 11 with mantle cell NHL, and 8 with other histologies with cyclophosphamide(600 mg/m² on day 1) and fludarabine (20 mg/m² on days 1-5), with G-CSF support. Prophylactic antimicrobial therapy was necessary. Responses were noted in 95% of the follicular NHL, including a 63% complete remission rate; complete remissions also occurred in 36% of mantle cell lymphoma and 50% of the other histologies. Response duration data were not provided.
These studies support a clear, and appropriate, movement away from alkylator-based regimens and toward fludarabine combinations.