The length of treatment of aggressive non-Hodgkin's lymphomas established according to the international prognostic index score: long-term results of the GISL LA03 study.

Objectives: To compare two different schedules of two different anthracycline-containing regimens, where length of treatment is modulated according to the international prognostic index (IPI) in patients with aggressive non-Hodgkin's Lymphoma (NHL).

Methods: In 1993 the Gruppo Italiano per lo Studio dei Linfomi (GISL) started a randomized 2 x 2 factorial phase III clinical trial for patients with newly diagnosed aggressive NHL comparing ProME(Epidoxorubicin)CE-CytaBOM (PE-C) to ProMI(Idarubicin)CE-CytaBOM (PI-C) and a fixed to a flexible treatment schedule where anthracycline dose was to be modulated according to observed hematological toxicity. Patients with low or low-intermediate IPI (IPI 0-2) and those with intermediate-high or high IPI (IPI 3-5) should receive six or eight courses, respectively. Involved-field radiotherapy was allowed for patients with initial bulky disease or with residual masses.

Results: Three hundred and fifty-six patients were registered into the study and randomized. Patients were well balanced among the four study arms in terms of clinical characteristics and prognostic factors. Three hundred and forty-five patients were available for evaluation of study endpoints. At the end of induction therapy complete remission rate was 61%, 5-year failure-free survival (FFS) rate was 40% and 5-year overall survival (OS) rate was 59%; no differences were observed according to treatment arms. Patients in the flexible arm received higher dose intensity of anthracycline (P < 0.001) with no apparent increase in toxicity. However, the flexible schedule was not superior to the fixed one. Patients with IPI 3-5 showed lower response rates (45% vs. 67%: P < 0.0001) and lower 5-year FFS (29% vs. 45%: P < 0.0001) compared to those with IPI 0-2.

Conclusions: six courses of fixed or flexible PE-C or PI-C can determine a promising success rate in patients with advanced aggressive NHL with IPI 0-2, whereas the same regimens are less effective in patients with IPI 3-5, even if two additional courses are delivered. For the latter group of patients innovative approaches are warranted.