Prognostic Indexes in Follicular Lymphoma: A Comparison of Different Prognostic Systems

Abstract & Commentary

By Stuart M. Lichtman, MD, FACP, Associate Attending, Memorial Sloan-Kettering Cancer Center, Commack, NY. Dr. Lichtman reports no financial relationship to this field of study.

Synopsis: In this series, all 3 indexes, IPI, ILI, and FLIPI, were useful to classify FL patients into differentiated risk groups, although the FLIPI identified a larger proportion of high-risk patients than the IPI and ILI.

Source: Perea G, et al. Prognostic indexes in follicular lymphoma: a comparison of different prognostic systems. Ann Oncol. 2005;16:1508-1513.

Follicular lymphoma (FL) is most commonly seen in middle-aged patients and accounts for ~30% of newly diagnosed non-Hodgkin lymphomas (NHL). Despite the prolonged median survival time, ~10 years, progression-free survival (PFS) and overall survival (OS) are poor in some patients. Several attempts to build up a prognostic index that is useful to make risk-adapted treatment recommendations have been made. The International Prognostic Index (IPI) has been successfully applied to patients with FL, but it seems to have a limited discriminating power as most patients are in the favorable- or the intermediate-risk groups.1 In the last few years 2 specific prognostic scores have been proposed: the Italian Lymphoma Intergroup Index (ILI)2 and, more recently, by an international group, the Follicular Lymphoma International Prognostic Index (FLIPI).3 The IPI and ILI indexes have been applied in FL patients with different success. The aim of this study was to apply the 3 prognostic indexes in a large group of patients with FL and to try to determine the relative merits of each of them.

Four hundred and sixty-five patients with a histologically confirmed diagnosis of FL grade I or II according to the WHO classification, consecutively diagnosed at 5 hospitals from Barcelona between January 1976 and December 2001, were initially considered for this study. Data needed to calculate all 3 indexes were finally obtained from 411 patients. The median follow-up of surviving patients was 73 months (range, 6-292). Patients received varying first-line treatments: 18 patients (4%) did not receive any treatment, since a watch-and-wait policy was adopted; 36 patients (10%) were treated with radiotherapy and/or surgery alone; 58 patients (14%) with a single alkylating agent (cyclophosphamide or chlorambucil); 51 patients (12%) with a combination chemotherapy regimen without an anthracycline (basically cyclophosphamide, vincristine and prednisone); 21 patients (5%) were treated with fludarabine combinations; and 227 patients (55%) with a chemotherapy regimen with an anthracycline (CHOP/CNOP). Response to treatment was assessed within 3 months after therapy was completed. In general patients were evaluated every 3 months during the first year, every 4 months during the second year, every 6 months during the next 3 years, and every year thereafter. Response after treatment was available in most patients: 190 (49%) achieved a complete response (CR) with initial therapy and 149 (39%) a partial response (PR), whereas 45 patients (12%) failed to respond to treatment.

The IPI was calculated according to the International Non-Hodgkin’s Lymphoma Prognostic Factors Project. The variables used were age (< 60 vs > 60 years), performance status (Eastern Cooperative Oncology Group performance status 0 or 1 vs > 2), Ann Arbor stage (I-II vs III-IV), extranodal involvement (< 2 vs 2 or more sites) and serum lactate dehydrogenase (LDH) level (normal vs high). Three risk groups were defined by IPI: score 0-1, low-risk; score 2, intermediate-risk; score > 3, high-risk. The high-intermediate and high-risk groups were joined to form a single high-risk group for comparisons with the other indexes.

The ILI index was calculated as detailed by the Italian Lymphoma Intergroup. Six variables were used to construct this index, 3 of them were also included in IPI (age, extranodal involvement and LDH level). The other 3 variables considered were presence of B-symptoms, male sex and erythrocyte sedimentation rate > 30 mm/h. Depending on the number of adverse prognostic factors (0-1, 2, or > 3), patients were classified into low-, intermediate- or high-risk groups.

The FLIPI was calculated according to the Follicular Lymphoma International Prognostic Project. The variables used to classify patients according to the FLIPI index were age > 60, advanced stage (III-IV), increased serum LDH, hemoglobin level < 12 g/dL and nodal involvement 5 or more sites). Three risk groups were considered: score 0-1, low risk; score 2, intermediate risk; and score > 3, high risk.


This report compares 3 prognostic indexes for FL (ie, the IPI, ILI and FLIPI) in an attempt to determine the merits of each one of these prognostic models. The IPI, initially designed for use in aggressive lymphomas, is easily applicable in clinical practice, and is also valuable in low-grade lymphomas. A major setback with the IPI is that only a small percentage (~8-11%) of patients with FL are included in the high-risk group. The ILI and FLIPI indexes, specifically designed for FL, also include variables that are easy to calculate, and separate patients into different risk groups. Regarding the FLIPI, patients from the current series were distributed in 3 different survival groups, with an OS probability at 5 and 10 years very similar to those previously reported, although in this series a higher proportion of patients were included in the high-risk group (38% vs 27%). The FLIPI seems to classify a larger number of patients into the high-risk group than IPI and ILI, even when only younger patients (> 60 years old) are considered: 14%, 16% and 20% of patients were included in the high-risk group after applying IPI, ILI and FLIPI indexes, respectively. According to these results, all 3 systems are useful to distribute FL patients into different risk groups, although the ILI index was especially valuable to separate high-risk patients, because in contrast with the other 2 indexes, differences in survival between low-risk and intermediate-risk patients were less significant. Of note, median survival of patients in the high-risk group is quite long (30-34 months) whatever the prognostic system employed. Because of this, Perea and colleagues argue that prognostic indexes for FL, as currently devised, are not useful for making treatment decisions.

The prognostic assessment of patients with FL could be improved by using other variables. In addition, the prognostic impact of genomic aberrations in patients with FL has also been investigated and a negative impact has been described for deletions of 6q. The whole-genome microarray analysis of gene expression has also been applied in FL and a survival predictor model constructed according to the gene expression signatures derived from non-malignant immune cells presents in the tumor at diagnosis.4 These biological findings may contribute significantly to risk assessment in patients with FL.

In conclusion, all 3 prognostic systems investigated (IPI, ILI, and FLIPI) were useful to identify patients with FL and different survival probabilities. In this series, however, the FLIPI identified a larger number of patients in the high-risk category. Future research should aim at improving the prognostic assessment of patients with FL by combining clinical variables with recently discovered biological variables.


1. Lopez-Guillermo A, et al. Applicability of the International Index for aggressive lymphomas to patients with low-grade lymphoma. J Clin Oncol. 1994;12:1343-1348.

2. Federico M, et al. Prognosis of follicular lymphoma: a predictive model based on a retrospective analysis of 987 cases. Intergruppo Italiano Linfomi. Blood. 2000;95:783-789.

3. Solal-Celigny P, et al. Follicular lymphoma international prognostic index. Blood. 2004;104:1258-1265.

4. Dave SS, et al. Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells. N Engl J Med. 2004;351:2159-2169.