FLIPI and F2 Index – Prognostic Indices in Advanced Stage Follicular Lymphoma

Background: The follicular lymphoma international prognostic index (FLIPI) has become a widely accepted tool for stratification of patients with follicular lymphoma according to their risk profiles (Solal-Celigny et al., Blood 2004). Its prognostic relevance has been proven in patients treated with standard chemotherapy as well as with immunochemotherapy (Buske et al., Blood 2006). Recently, a new prognostic index, the F2 index, has been proposed as possible alternative to the FLIPI (Federico et al., ICML-10, 2008). Methods: We used the collective of patients with advanced stage follicular lymphoma treated first-line within the recent trials of the GLSG to explore the prognostic relevance of the F2 prognostic index in comparison to the FLIPI. Outcome parameters were time to treatment failure (TTF) and overall survival (OS). Statistical methods included Kaplan-Meier plots to validate the prognostic indices and multiple Cox regression to explore the relevance of the prognostic factors.

Results: Of the 1562 patients, the F2 index could only be assessed in 758 patients (50%), of those 355 had had a treatment failure and 124 had died. The low number of evaluable patients was mainly due to the use of β₂-microglobulin as prognostic factor, which was missing in 45% of the patients. The FLIPI could be assessed in 1424 patients (93%), of those 697 had had a treatment failure and 255 had died. The F2 index classified 7% of the patients as low risk, 58% as intermediate risk, and 35% as high risk, in comparison to 14%, 41% and 45% according to the FLIPI. Both F2 and FLIPI separated three risk groups with regard to TTF (p<0.0001). However, the F2 index did not separate risk groups in the three FLIPI risk subgroups. In addition, in multiple regression analysis of the F2 and FLIPI prognostic factors, two of the three new F2 prognostic factors, namely bulk > 6 cm and elevated β₂-microglobulin showed no prognostic relevance for TTF. Furthermore, after backward elimination, all the FLIPI risk factors (age > 60 years, LDH > ULN, hemoglobin > 120 g/l and more than 4 involved nodal regions) remained independently significant together with bone marrow involvement, whereas bulk > 6cm and b₂-microglobulin were of no relevance. With regard to OS, the FLIPI separated three groups, whereas according to the F2 index low and intermediate risk groups were overlapping.

Conclusion: The F2 index did not show superior relevance to the FLIPI in our evaluation and should not replace the FLIPI as prognostic index in follicular lymphoma in general. However, a simplification of the FLIPI with regard to the assessment of the number of involved nodal areas and the incorporation of the performance status would be challenging.