Elderly patients with aggressive non-Hodgkin's lymphoma treated with CHOP chemotherapy plus granulocyte-macrophage colony-stimulating factor: identification of two age subgroups with differing hematologic toxicity.

PURPOSE: Standard cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy repeated at 3-week intervals is difficult to deliver in elderly patients with non-Hodgkin's lymphoma (NHL). The use of hemopoietic growth factors may decrease the hematologic toxicity of chemotherapy and allow the delivery of full-dose CHOP.

PATIENTS AND METHODS: We conducted a phase II trial with the addition of granulocyte-macrophage colony-stimulating factor (GM-CSF) to CHOP chemotherapy in NHL patients older than 60 years of age. Twenty-six previously untreated patients were assessable; median age was 67 years (range, 61 to 84 years). CHOP included cyclophosphamide 750 mg/m2 intravenously day 1; doxorubicin 50 mg/m2 intravenously day 1; vincristine 1.4 mg/m2 (2 mg total dose) intravenously day 1; and prednisone 100 mg orally days 1 through 5. GM-CSF 5 microg/kg was administered subcutaneously on days 4 through 13. Cycles were repeated every 21 days for six cycles. Results were analyzed for the total group and for two age subgroups: 61 to 69 years (n = 15) and 70 years or older (n = 11).

RESULTS: Sixteen patients (62%) achieved a complete response (CR), four patients (15%) achieved a partial response (PR), and six patients (23%) did not respond to therapy. After a median follow-up of 41 months, the median progression-free and overall survival were 19 and 30 months, respectively. Twenty patients completed six cycles. One hundred thirty-eight of the 156 planned cycles were delivered (88%). The relative dose-intensity was 95%. The chemotherapy-induced toxicity was important. Absolute neutrophil count was less than 500/mL in 43% of the cycles, platelet nadir was less than 20,000/mL in 19%, and febrile neutropenia occurred in 21%. There were no grades 3 to 4 mucositis. Treatment-related death occurred in two patients, and was associated with neutropenic septic shock. The toxicity related to GM-CSF was mild hypotension after the cytokine was administered in 7% of cycles. When the results of the study were analyzed by age subgroups, we observed that whereas response and median survival were similar in patients aged 61 to 69 years or 70 years or older, there were significant differences in dose delivery and toxicity. Chemotherapy was delivered in 86 of 90 planned cycles in patients aged 61 to 69 years, but in only 52 of 72 planned cycles in patients aged 70 to 84 years (P = .00008). Absolute neutrophil count was less than 500/mL in 24% of cycles in patients aged 61 to 69 years and 73% of cycles in patients aged 70 years or older (P = .00001). The platelet nadir of less than 20,000/mL occurred in 5% of patients aged 61 to 69 years and in 42% of patients aged 70 years or older (P < .0001). Fever and neutropenia occurred in 8% of patients aged 61 to 69 years and in 42% of patients aged 70 years or older (P < .0001). Mucositis (grades 1 to 2) occurred in 21% of patients aged 61 to 69 years and in 42% of patients aged 70 years or older (P = .006).

CONCLUSION: CHOP chemotherapy plus GM-CSF is an active regimen in elderly patients with NHL. Despite cytokine support, the toxicity of the regimen is elevated. We have identified two age subgroups (61 to 69 and > or = 70 years) that do not differ in treatment efficacy but show large differences in treatment-related toxicity.