Training and validation of prognostic models in advanced stage follicular lymphoma. The patient cohort of the Primary Rituximab and Maintenance (PRIMA) trial was used to validate and compare the prognostic capacity of FLIPI and FLIPI-2. Because the PRIMA-PI was developed on the PRIMA cohort, its prognostic capacity cannot be compared with the other prognostic models on the PRIMA cohort because of overfitting. The combined FL2000 and Molecular Epidemiology Resource (MER) cohort was used to validate the prognostic capacity of the PRIMA-PI and to compare it with FLIPI. *The FLIPI-2 could not be assessed for the FL2000 + MER cohort.