Early Results from LRF CLL4: A UK Multicenter Randomized Trial.

On behalf of the UK NCRI CLL Trials Group and the Adult Leukaemia Working Party. The CLL4 trial opened in 2/99 and closed in 10/04. 783 previously untreated patients were randomized and 6 were excluded. 12% of patients were entered from outside the UK (Argentina, Ireland, Italy, New Zealand and Russia). Median follow-up is 21 months. The male:female ratio was 2.8:1 and the distribution by Binet stages was A progressive 25%, B 45% and C 30%. One third of cases were aged <60 years, one third 70 years or over. Patients were randomized between chlorambucil (10mg/m² x 7 days)(n=387) and fludarabine alone (FDR)(n=194) or in combination with cyclophosphamide (FC) (n=196). FDR and FC were given intravenously (IV) in the first 2 years (FDR 25mg/m² x 5 days and FDR 25mg/m² plus cyclophosphamide 250mg/m² for 3 days). Early in 2001 oral FDR became available and the protocol was modified to allow its use: FDR 40mg/m² x 5 days and FDR 24mg/m² plus cyclo 150mg/m², both for 5 days. Differences between the IV and oral formulations will be presented elsewhere in this meeting (Hillmen et al). Responses were assessed by bone marrow biopsies. Responses to treatment are summarized in the table below in 661 patients with available data. Within each age group treatment affected response (CR+nodPR v PR/NR/PD: p=0.0003 for <60 years, p<0.0002 for 60–69 years and p=0.002 for ≥70 years) with highest responses with FC and lowest with chlorambucil. The toxicities profile showed more neutropenia with FC (55%) than FDR (40%) and chlorambucil (29%) and more hospitalisations with FC and FDR. However, and as reported in ASH 2004, there were more hemolytic anemias after chlorambucil (13%) than with FDR (10%) and FC (4%). There was more nausea/vomiting and alopecia with FC than with the other regimens. Progression free survival (PFS) up to 10/04 showed fewer events with FC (O/E 0.5) than FDR (O/E 1.1) and chlorambucil (O/E 1.3); FDR+FC v Chl p<0.00005; FC v FDR p<0.0005. PFS at 3 years is chlorambucil 23%, FDR 31% and FC 62%. Analysis of survival thus far showed no differences between the three regimens. We conclude that the combination of FC is associated with a greater quality of responses, more myelotoxicity and less hemolytic anemia than FDR and chlorambucil and it could become the new standard regimen for CLL upon which future regimes will build. CLL4 has generated a wealth of prognostic information: FISH, VH mutations, expression of CD38 and ZAP-70 which will allow a better characterization of risk groups and their response to first-line therapy. Some of these analyses will be presented in this meeting (Oscier et al).