Page 244 - 18F-FDG PET as biomarker in aggressive lymphoma; technical and clinical validation

Chapter 9
Importance of uniform PET criteria
During this research project we performed an individual patient data meta- analysis. As we collected scans and patient data from multiple studies we had to harmonize the data first. It was soon realized that, although most studies used the Deauville criteria for their analyses, these were not applied exactly the same between studies. Mostly this was due to an unclear definition for DS5 (uptake “markedly higher than liver uptake” or “2-3 times above the maximum SUV in the liver”). We solved this by re-analyzing all scans with Deauville 5 in a semi- quantitative way by assigning a Deauville 5 if the lesional SUVmax exceeded 3 times the liver SUVmax and/or in case of new lymphoma lesions. It would be helpful for future analyses and comparisons between studies to adopt a clear and similar criteria definition integrating visual and semi-quantitative assessment.
Strengths and weakness of the current research
By building a strong consortium we were able to perform the individual patient data meta-analyses. One of the strengths of this consortium is that all patient data and scans are stored for future analyses. Future research proposals should be approved by the consortium partners before their data is used in new analyses. The consortium partners should have confidence in the correct use of their valuable data, of course. Therefore, legal policy documents were signed before entering the PETRA consortium. For 2 potential consortium partners we were not able to collaborate due to legal and logistic issues (France) and legal issues (USA). Finally, we could include 8 studies in the meta-analyses. Still we experienced a low number of patients with an interim [18F]FDG PET after 1 and 3 cycles, respectively. Besides that, we did not have enough contrast in treatment strategies to investigate the effect of differences in treatment on PET prediction and -interpretation.
Current clinical guidelines and future perspectives
Role of interim [18F]FDG PET in current clinical guidelines
The current thesis provides an overview of the growing evidence and validation of interim [18F]FDG PET in DLBCL in the recent years. The most recent Dutch DLBCL guideline, published by HOVON in September 2021, recommends to perform an interim [18F]FDG PET after 2 cycles of R-CHOP instead of a CT scan in cases where this could affect treatment de-escalation or escalation choices [19].
242