Transformation of follicular lymphoma (FL) is a morphological diagnosis based on the demonstration of diffuse large B-cell lymphoma (DLBCL) in a patient who has been diagnosed as having follicular lymphoma, either consecutively or concurrently. This is often regarded as having an adverse effect on prognosis compared to de novoDLBCL. However, a pathological diagnosis of transformed FL may be made in a range of quite different clinical settings and the aim of this study was to determine the extent to which this determined the prognostic significance of transformation.

The study was based on an established UK population cohort (www.hmrn.org), which since 2004 has tracked all patients diagnosed with haematological malignancies in a population of 3.6M. Within the study region, all tissue biopsies are handled by a single specialist haematopathology laboratory (www.hmds.info) and are investigated using morphology, immunocytochemistry, flow cytometry and molecular diagnostics according to standard protocols and WHO criteria. Clinical and treatment information is systemically collected for all patients and survival data is acquired through links with national data sources. Survival comparisons described below were adjusted for age and gender.

This study presents data on 1787 patients with de novo DLBCL and 724 newly presenting patients with FL. 121 patients with DLBCL had evidence of FL at the time of diagnosis either in the lymph node (88) or in the staging bone marrow (33). These patients were slightly younger than those with DLBCL without evidence of FL (65 years v 70.6 years) but had similar IPI scores, and no significant differences by overall survival following standard R-CHOP chemotherapy (p=0.16 & p=0.9 respectively). 91 patients (12.6%) of patients presenting with FL were subsequently diagnosed with DLBCL; 29 of these had received no chemotherapy and 7 were treated with local radiotherapy before the diagnosis of transformation, and the outcome of this group was identical to that of de novo DLBCL (p=0.6). 145 patients with FL previously treated with chemotherapy (the majority received R-CVP) relapsed; 55 of these patients relapsed with transformed lymphoma and the outcome for this group was poor compared to de novo DLBCL (Figure 1,
Figure 1

Overall Survival: Relapsed FL versus de novo DLBCL

Figure 1

Overall Survival: Relapsed FL versus de novo DLBCL

Close modal
, median survival 2 years v 6 years p<0.0003). The outcome was significantly worse than those relapsing with FL (median survival 8 years p=0.0002) (Figure 2 
Figure 2

Overall Survival: Relapsed versus Transformed to DLBCL

Figure 2

Overall Survival: Relapsed versus Transformed to DLBCL

Close modal
). The median time to relapse for patients with transformed disease was 2.1 years after chemotherapy and the comparable time for those with relapsed FL was 6 months.

Morphological evidence of transformed FL, either concurrently or consecutively is not by itself an adverse prognostic factor when patients are treated with standard DLBCL chemotherapy. This result is consistent with previously reported studies that suggested that the presence of t(14;18) as a sole genetic abnormality had little effect on outcome in patients with DLBCL. However, when transformation occurs following prior exposure to chemotherapy the prognosis is poor. This group constitutes 8% of all FL patients, with most of these events occurring within 3 years of FL presentation. Further consideration is needed on how this group of patients should be managed.

Disclosures:

Jack:Roche /Genentech: Research Funding. Off Label Use: Rituximab in Burkitts Lymphoma. Patmore:Roche: Consultancy, Honoraria.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution