We welcome the publication of a score (Follicular Lymphoma International Prognostic Index [FLIPI]) assisting in the choice of treatment for patients with newly diagnosed follicular lymphoma (FL). 1 As the authors state, treatment options for this disease range from “watch and wait” to allogeneic bone marrow transplantation, and the selection of the best available option will be favored by a realistic estimation of the expected survival for each particular patient. The 5 independent prognostic factors retained in the FLIPI reflect important characteristics of the patient (age), of the disease extension and clinical aggressiveness (stage and number of nodal sites involved, lactose dehydrogenase [LDH]), and of the tumorhost interaction (hemoglobin level). Surprisingly enough, histology grade, a disease characteristic considered by many clinicians to be of paramount importance for prognosis and choice of treatment, is missing from the index and was not even significant in the univarate analysis.

The absence of this parameter from the index and the omission of any discussion about this fact in the paper could suggest that histologic grade is indeed not relevant to patient survival and should therefore not be taken as an important information. We believe that this misunderstanding results from a bias that is not sufficiently covered in the article and that should be clarified. It was long recognized that an important proportion of large cells (centroblasts) at histologic examination (an observation referred to as grade 3 in the current World Health Organization [WHO] classification) does confer to FL a worse prognosis, despite apparently favorable clinical prognostic features. 2, 3 Nevertheless, the outcome of these cases can be similar to the forms with less centroblasts when they are treated with an anthracycline-containing regimen. 4, 5 This information was known in 19856 (the year of the start of data collection for the FLIPI) and, although details on treatment are not given in the paper, it is probable that the majority of grade 3 FLs of this data set were treated with anthracyclines, thus influencing the prognosis.