Early and Risk-Adapted Therapy with Fludarabine in High-Risk Binet Stage A CLL Patients Prolongs Progression Free Survival but Not Overall Survival: Results of the CLL1 Protocol of the German CLL Study Group (GCLLSG).

Background: Most patients (pts) with CLL present at early stages of their disease. This group is not homogenous with regard to its prognosis, and some pts show progression within 12 months after diagnosis. So far it is not clear if these pts benefit from early risk-adapted therapy. The CLL1 protocol, a randomized phase III trial of the GCLLSG, was conducted to assess if the early use of fludarabine prolongs progression free survival (PFS) in high risk (HR) CLL pts. High risk was defined by an elevated thymidine kinase (>7 U/L) or beta-2-microglobulin (>3,5 mg/L) and short lymphocyte doubling time (<12 months) or diffuse bone marrow infiltration. Combining the biological and clinical factors, at least one parameter of each group needed to be positive to qualify for the high risk cohort.

Patients: From 09/1997 to 12/2004 877 previously untreated Binet A pts were enrolled. 788 pts could be stratified, 471 pts (68,3%) to the low risk (LR), 218 pts (31,6%) to the HR cohort for disease progression. 99 pts were not eligible due to violation of inclusion/exclusion criteria. 104 pts (HRF) were randomized to treatment with fludarabine (fludarabine IV 25 mg/m²/d, day 1–5, every 28 days), 114 pts (HRWW) to “watch and wait”. 30 pts were not eligible for PFS and OS. Analyses on PFS and OS on 188 pts are presented here. Response was defined by NCI-WG criteria (Cheson et al., 1996), progression was defined by slightly modified NCI-WG criteria with regard to the early stage.

Results: The median observation time was 45.3 months (95% CI, 3.2–96 mo) for HRF and 40.4 months (95% CI, 6.9–92 mo) for HRWW. The median age in the HRF arm was 61 vs. 62 yrs in the HRWW arm. The HRWW arm contained significantly more male pts (p=0.03). All other parameters used for risk stratification and other characteristics like age, performance state, comorbidity, B symptoms, absolute lymphocyte count, hemoglobine, platelets, lymphadenopathy, splenomegaly or hepatomegaly were not significantly different at time of randomization. Response to fludarabine could be evaluated in 70 pts. The overall response rate was 91%, 11 pts had complete remission (16%), 5 pts had nodular partial remission (PR) (7%), 48 pts (69%) had PR, 3 pts (4%) had stable disease and 3 pts (4%) had progressive disease. Main toxicities (grade 3 or grade 4 CTC) were leukocytopenia and infections. Infections were not significantly higher in the treatment arm (p=0.12). Dose reduction was necessary in 20% of pts due to toxicity. 20 deaths were reported in the HRF arm, 12 deaths were reported in the HRWW arm (p=0.47).

Discussion: On an intent-to-treat analysis PFS was significantly longer in the HRF arm in comparison to the HRWW arm (24.2 vs. 15.9 mo, p=0.03). Though the median OS was not yet reached, no significant difference between both groups in OS was assessed (p= 0.47).

Conclusion: CLL pts of Binet stage A but with risk features have a median time to progression of 15.9 months. Fludarabine prolongs the PFS, but not the OS in this group of patients.