ORLANDO -Adding rituximab (Rituxan) to standard first-line
chemotherapy (CVP) improved outcomes for patients with advanced follicular
lymphoma, according to clinical trial results presented by Kevin Imrie, MD,
at the American Society of Clinical Oncology 41st Annual Meeting (abstract
6525). "The benefit in terms of time to progression is observed in all
prognostic groups,"said Dr. Imrie, director of postgraduate programs in
medicine, Toronto Sunnybrook Regional Cancer Centre, Ontario, Canada.
Dr. Imrie and his colleagues previously reported the study
outcomes in Blood (105:1417-1423, 2005). At ASCO, he reported an
analysis of the data according to baseline prognostic variables at a median
follow-up of 30 months.
The researchers randomized 321 patients with late-stage
follicular lymphoma who had received no prior treatment to either CVP or CVP
plus rituximab (R-CVP) for four cycles, every 3 weeks. Both groups received
cyclophosphamide 750 mg/m2 and vincristine 1.4 mg/m2 IV
on day 1, and prednisone 40 mg/m2 orally on days 1 through 5. The
intervention group of 162 patients also received rituximab 375 mg/m2
IV on day 1. After four cycles, patients were restaged. Those with stable or
progressive disease were taken off study. The remaining patients received an
additional four cycles of assigned treatment for a total of eight cycles.
Patients were well matched for age and other prognostic
factors. Almost half the patients in both arms had three or more risk factors
according to the Follicular Lymphoma International Prognostic Index (FLIPI).
More than one-quarter of patients had elevated lactate dehydrogenase (LDH)
levels. About 40% had bulky disease, and about one-third had B symptoms. Dr.
Imrie called the cohort an "adverse-risk group of patients."
"There is a clinically important and statistically
significant difference in response rate, time to treatment failure, duration of
response, disease-free survival, and time to new antilym-phoma treatment or
death all in favor of the R-CVP arm," Dr. Imrie said (see Table). "Only overall
survival didn't differ at this time."
Time to progression, relapse, or death, at a median
follow-up of 30 months, was 32 months in the patients on R-CVP vs 15 months in
the CVP cohort (P < .0001).
Analyzing data by FLIPI score, difference in time to
progression, relapse, or death was not significant in the low-risk patients (FLIPI
scores of 0 or 1). Patients with intermediate-risk disease (FLIPI scores of 2)
had a significant difference: 39 months in the R-CVP arm, compared with 17
months in the CVP cohort. In the largest group of patients (FLIPI scores of 3
to 5), the R-CVP cohort's time to progression, relapse, or death was 26 months
vs 12 months in the CVP group (P < .0001).
When examining the data by International Prognostic Index (IPI),
histology, B symptoms, and bulk, the researchers saw similar results in favor
of patients on the R-CVP arm. The pattern held for patients with hemoglobin
levels of greater than 12 g/dL. But in patients with a hemoglobin level of 12
g/dL or less, the researchers saw no difference between the treatment arms.
The investigators analyzed risk ratios for subgroups defined
by baseline variables, including indications for treatment, age, number of
extranodal sites, bone marrow involvement, LDH, beta2-microglobulin, B
symptoms, number of nodal sites, hemoglobin, IPI, and FLIPI.
"For each factor analyzed, the addition of rituximab
provided benefit, with risk ratios of less than 0.7 in all cases," Dr. Imrie
commented. "For only two categories-the group of patients with greater than one
extranodal disease and hemoglobin of 12 g/dL or less-did the confidence
intervals cross 1. And only in the case of hemoglobin was there an interaction
between treatment assignment and baseline variable on Cox regression analysis,
suggesting that the benefit was confined to patients whose hemoglobin was
greater than 12 g/dL."
The researchers also completed a multivariate analysis of
the prognostic factors, including composite FLIPI and IPI scores. In the team's
model using treatment regimen as a factor to predict outcome in all analyses,
only the FLIPI score added to the predictive power of treatment assignment.
When FLIPI as a composite score is not included in the model, only hemoglobin
level and number of nodal sites had additional predictive value on top of