Background : 18Fluorodeoxyglucose-PET has been advocated as a powerful mean to evaluate response in aggressive non-Hodgkin’s lymphoma. Its relative merits compared to the widely accepted International Workshop Criteria (IWC) (Cheson, 1999. J Clin Oncol 17: 1244) have been investigated (Juweid, 2005 J Clin Oncol 23:4652), but need to be further confirmed.

Materials and methods : We studied 103 patients with diffuse large B-cell lymphoma after 4 cycles of inductive CHOP or CHOP-like chemotherapy, with (n=51) or without Rituximab (n=52) using the IWC criteria and PET. The patients where also evaluated with an early PET after 2 courses, the results of which did not influence treatment decisions. PET response criteria have already been published (

Haioun, Blood 2005; 106
). The interpretation of PET took into account the results of the early PET, in particular to define progressive disease.

Results : The baseline characteristics of the patients where as follows : Median age 53y (19–78); age >60 : 25 (24%); LDH > normal : 71 (69%); > 1 extranodal site : 53 (51%); poor performance status (>=2) : 37 (36%); stage III–IV: 84 (81%). The international prognostic index (IPI) score was 0–2 in 40 pts (39%) and 3–5 in 63 (61%). The comparative response evaluation using IWC criteria or the combination of IWC criteria and PET is shown in the following table.

Three years after completion of therapy, patients being in CR by PET at 4 cycles had an estimated event-free survival (EFS) of 83% if in CR/CRu by IWC criteria, and of 80% if in PR or SD by IWC. Patients not in CR by PET had an EFS of 51% when in CR/CRu by IWC criteria and of 8% when in PR, SD or PD by IWC criteria (p<0.0001 as compared to CR patients). The same differences were seen whether the patients had received Rituximab or not. In a multivariate Cox regression analysis, event free survival was adversely affected by a non-CR response after 4 cycles evaluated by PET (p<0.0001, relative risk (RR) = 8.3), but not by a poor IPI (3–5) (p = 0.6, RR = 0.81) or Rituximab-based treatment (p = 0.74, RR = 1.13).

Conclusion : We confirm the higher predictive power of PET as compared to IWC criteria. The CRu category disappears from evaluation criteria when integrating informations delivered by PET, and PR patients can be separated into CR versus PR patients. The predictive value of PET is seen in patients treated with, as well as without Rituximab.

IWC + PET
ResponseCRCRuPRSDPDTotal
* 4 pts showed progression between 2 and 4 cycles on the basis of PET** 2 pts had no PET at 4 cycles because of early progression 
IWC  
CR 23 24 
CRu 43 53 
PR 14 
SD 
PD 6* 
Total 76 13 10 101** 
IWC + PET
ResponseCRCRuPRSDPDTotal
* 4 pts showed progression between 2 and 4 cycles on the basis of PET** 2 pts had no PET at 4 cycles because of early progression 
IWC  
CR 23 24 
CRu 43 53 
PR 14 
SD 
PD 6* 
Total 76 13 10 101** 

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author

Sign in via your Institution