Abstract 3432

Poster Board III-320

There are no head-to-head randomized controlled trials (RCTs) assessing the efficacy of available treatments for naive patients with chronic lymphocytic leukemia (CLL). Bayesian mixed treatment comparison (MTC) meta-analyses provide a means of indirectly estimating the relative treatment effect of one intervention to another.

Methods

A systematic review identified RCTs that reported or summarized efficacy data for any of the treatments of interest in treatment-naive CLL patients. The primary endpoint was progression-free survival (PFS) and secondary endpoints included event-free survival (EFS), overall survival (OS), disease-free survival (DFS), duration of remission, overall response (OR), time to new CLL treatment or death, and rates of molecular complete and partial remission. Data were extracted using a standard electronic data form and study quality assessed using the Jadad checklist for RCTs. Extracted data were analyzed in a Bayesian MTC using a fixed-effects model. A Cox regression model was assumed for time to event outcomes, and the log hazards were summarized across RCTs and interpreted as medians based on the PFS curve for FCR presented in the CLL-8 study report.

Results

The systematic review identified 683 articles of which eight met pre-specified inclusion criteria on review of the title, abstract, or full-text. Although study design and reporting were often limited quality, with only one RCT reporting being blinded, the RCTs were sufficiently similar in design, patient inclusion criteria and characteristics, and outcome reporting to conduct a MTC (Table 1).

Table 1:

Characteristics of RCTs included in the mixed treatment comparison

RCTInterventionsPatients (N)Median Age (range)Disease stage (%)*
CLL-8, Hallek 2008 FCR vs FC 810 60 (43-77) A (5), B (64), C (31) 
Catovsky 2007 FC vs F 777 65 (35-86) A (25), B (45), C (30) 
Rai 2000 F vs C 509 63 (32-83) Rai 0-II : 61. Rai III-IV :39 
Eichhorst 2006 FC vs F 362 57 (30-65) A (10), B (55), C (35) 
Knauf 2007 B vs C 305 64 (n.s.) B (70), C (30) 
Hillmen 2007 A vs C 294 60 (35-86) Rai 0-II : 66. Rai III-IV : 34 
Flinn 2007 FC vs F 278 61(33-86) Rai 0-II : 56. Rai III-IV : 44 
Eichhorst 2007 F vs C 206 70(64-80) A (15), B (47), C (38) 
RCTInterventionsPatients (N)Median Age (range)Disease stage (%)*
CLL-8, Hallek 2008 FCR vs FC 810 60 (43-77) A (5), B (64), C (31) 
Catovsky 2007 FC vs F 777 65 (35-86) A (25), B (45), C (30) 
Rai 2000 F vs C 509 63 (32-83) Rai 0-II : 61. Rai III-IV :39 
Eichhorst 2006 FC vs F 362 57 (30-65) A (10), B (55), C (35) 
Knauf 2007 B vs C 305 64 (n.s.) B (70), C (30) 
Hillmen 2007 A vs C 294 60 (35-86) Rai 0-II : 66. Rai III-IV : 34 
Flinn 2007 FC vs F 278 61(33-86) Rai 0-II : 56. Rai III-IV : 44 
Eichhorst 2007 F vs C 206 70(64-80) A (15), B (47), C (38) 

Key: FCR = fludarabine, cyclophosmphamide, and rituximab; F = fludarabine; A = alemtuzumab; C = chlorambucil; FC = fludarabine and chlorambucil (FC); B = bendamustine.

*

Binet stage unless otherwise stated.

Sufficient data were available to compare PFS, complete remission (CR) and OR although only small patient numbers were reported for some treatments and outcomes. The MTC suggested FCR to be the best therapy for treatment-naïve CLL when considering PFS and OR endpoints compared to all other treatments (Table 2). FCR significantly prolonged PFS compared to all other treatments with hazard ratios between 0.24 (0.17-0.34, chlorambucil) and 0.56 (0.43-0.72, FC).

Table 2:

Outcomes for FCR versus other treatments

Relative efficacy of FCR versus comparator treatment regimen (95% CI)
Progression-Free SurvivalComplete RemissionOverall Response
Chlorambucil 0.24 (0.17-0.34) 30.3 (17.5-53.4) 14.2 (8.3-24.7) 
Bendamustine NR 1.5 (0.3-5.4) 4.2 (2.0-8.9) 
Fludarabine 0.28 (0.20-0.38) 10.1 (6.3-16.4) 5.8 (3.5-9.8) 
Alemtuzumab 0.41 (0.26-0.66) 1.8 (0.4-6.0) 3.5 (1.6-7.5) 
FC 0.56 (0.43-0.72) 2.7 (2.0-3.8) 2.3 (1.6-3.4) 
Relative efficacy of FCR versus comparator treatment regimen (95% CI)
Progression-Free SurvivalComplete RemissionOverall Response
Chlorambucil 0.24 (0.17-0.34) 30.3 (17.5-53.4) 14.2 (8.3-24.7) 
Bendamustine NR 1.5 (0.3-5.4) 4.2 (2.0-8.9) 
Fludarabine 0.28 (0.20-0.38) 10.1 (6.3-16.4) 5.8 (3.5-9.8) 
Alemtuzumab 0.41 (0.26-0.66) 1.8 (0.4-6.0) 3.5 (1.6-7.5) 
FC 0.56 (0.43-0.72) 2.7 (2.0-3.8) 2.3 (1.6-3.4) 

PFS are provided as hazard ratios, CR and OR are median odds ratios. CI – credible interval NR – hazard ratio not reported

Median PFS was prolonged on FCR by 20 and 8 months when compared to chlorambucil and FC, respectively. The probability of CR was significantly higher for FCR compared to chlorambucil, fludarabine and FC, but data were insufficient to draw conclusions against alemtuzumab or bendamustine. FCR also significantly increased OR when compared to all other treatments.

Conclusion

This meta-analysis of available RCT data demonstrates that the addition of rituximab to FC prolongs PFS and significantly increases OR in comparison to all other treatments. FCR was also shown to increaseCR with respect to chlorambucil, fludarabine and FC. These findings are based on data from eight RCTs. Such analyses will be strengthened by the inclusion of additional RCT data that may become available in the future.

Disclosures

Janssens:Roche: Honoraria. Foà:Roche: Advisory board member, Speakers Bureau; Bayer: Advisory board member, Speakers Bureau; GlaxoSmithKline: Advisory board member, Speakers Bureau; Mundipharma: Advisory board member, Speakers Bureau; Celgene: Advisory board member, Speakers Bureau. Keating:Genzyme: Honoraria, Speakers Bureau; Genetech: Honoraria, Speakers Bureau. Ouwens:MAPI: Employment. Tatt:Roche: Employment. Carr:Roche: Employment.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution