Integrating Newer with Traditional Prognostic Factors in Evaluating Patients with CLL Receiving Frontline Chemoimmunotherapy.

Prognostic factors are important for evaluating patients (pts) with CLL, owing to heterogeneity in response to treatment and clinical course. The goal of this analysis was to evaluate the newer prognostic factors in relation to traditional ones in previously untreated pts receiving frontline chemoimmunotherapy. Traditional prognostic factors include age, Rai stage, absolute lymphocyte count (ALC), lymphocyte doubling time, and many others. Newer prognostic factors, include IgVH gene mutation status (MS), chromosome abnormalities by FISH, and leukemia cell expression of ZAP70. Finally, the clinical endpoint is important in considering and evaluating prognostic factors, since significant independent factors can be different for response to treatment, time to progression (TTP), and overall survival (OS).

We evaluate 488 previously untreated pts who had an NCI-WG ‘96 indication for treatment and received frontline chemoimmunotherapy (CIT). Treatment was with a fludarabinebased CIT regimen: FCR (n=300), FCMR (n=30), FCR3 (n=65), FCR+GM-CSF (n=45), or CFAR (n=48). The CR rate, TTP, and OS were not significantly different between the CIT regimens; median follow-up times varied. The overall estimated median follow-up time was 62mo, TTP was 76mo, and OS was not reached. An extensive list of pretreatment characteristics were evaluated in univariable analyses as traditional factors including age, Rai stage, beta-2 microglobulin (β2M), CD38, and others, as well as newer prognostic factors including chromosome abnormalities by FISH, IgVH MS and ZAP70 expression by immunohistochemistry. Clinical endpoints included complete remission (CR), TTP, and OS. Multiple factors were significant (p<.05) and varied according to clinical endpoint. Next, we evaluated these significant factors in multivariable (MV) analyses to identify the significant (p<.05) independent factors for each endpoint. The rank order importance for each independent factor identified in the MV model with current follow-up is shown (Table 1).

Table 1 Multivariable Model

Table 2 Significance of Factor in MV Model