Gem-(R)CHOP Versus (R)CHOP: A Randomized Phase II Study of Gemcitabine Combined with (R)CHOP in Untreated Aggressive Non- Hodgkin’s Lymphoma - EORTC Lymphoma Group Protocol 20021

Despite recent advances many patients with aggressive NHL succumb to their disease and improvements in front-line chemotherapy are still needed. Gemcitabine is active in lymphoma. We embarked on a randomized phase II trial to determine the efficacy, feasibility and toxicity of the addition of gemcitabine to standard front-line chemotherapy. The trial consisted of two parts. The first part was designed to determine the maximal tolerated dose of gemcitabine and the second to determine the response rate, feasibility and toxicity of the treatment. Patients, 18–70 years old, with stage II–IV previously untreated B-large cell, follicular grade 3, anaplastic large cell or peripheral T cell lymphoma were eligible. They were randomized to receive either 8 cycles of standard CHOP given every 3 weeks or the same treatment combined with gemcitabine (Gem-CHOP). In the second part of the trial patients with B-NHL also received 8 cycles of rituximab (R). The starting dose of gemcitabine was 800 mg/m² on days 1 and 8. Dose escalation to 1000 and 1250 mg/m² and reduction to 800 mg/m² on day 1 only were foreseen. Twenty-five patients, 7 with T and 18 with B-NHL, were enrolled in the trial; 15 in the dose-finding part and 10 in the second part before early closure due to low accrual. Twelve received Gem-(R) CHOP and 13 (R)CHOP. Maximal tolerated dose of gemcitabine was 800 mg/m² given on days 1 and 8; dose limiting toxicity was hematologic. Five patients (42%) treated with Gem-(R)CHOP achieved CR in comparison to 10 (77%) treated with (R)CHOP. Three out of 4 patients with T-NHL responded to Gem-CHOP and 2 out of 3 to CHOP. Median time to treatment failure was 1.49 years for the Gem-(R)CHOP arm and 3.13 years for the (R)CHOP arm. Four patients receiving Gem-(R)CHOP had serious pulmonary toxicity, as compared to none receiving (R)CHOP. One patient died of severe pneumonitis. The apparently immunologically mediated pulmonary toxicity prompted us to add in the second part of the trial a three-day course of prednisone starting on day 8, together with the second gemcitabine dose. After this change, no additional cases of pulmonary toxicity were seen. The combination of gemcitabine with standard (R)CHOP results in significant but acceptable hematologic toxicity. Maximal tolerated dose of gemcitabine is 800 mg/m² given on days 1 and 8 every 3 weeks. In this small group of patients, addition of gemcitabine did not seem to improve outcomes. The use of gemcitabine in previously untreated patients with aggressive NHL results occasionally in severe, potentially fatal, pulmonary toxicity. Therefore, further exploration of this combination does not seem to be warranted.