Abstract
Background
Response criteria for 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) for thoracic malignancies include European Organization for Research and Treatment of Cancer (EORTC) criteria, Positron Emission tomography Response Criteria In Solid Tumors 1.0 (PERCIST), PeterMac Metabolic Visual Criteria and Deauville Criteria. It is unknown which criteria have the highest prognostic value in NSCLC.
Method
Between 2004 and 2016, three NSCLC prospective trials included patients treated with radical radiotherapy (RT) or chemoRT with baseline and post-treatment FDG-PET imaging. For each patient, the four FDG-PET response criteria were reported retrospectively and blinded to outcome. Responses to therapy were categorized as complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD) or progressive metabolic disease (PMD) and correlated with subsequent survival using Cox proportional hazard models, c-statistic, r2 and Akaike information criterion (AIC).
Result
Eighty-seven NSCLC patients underwent FDG-PET before and after radical RT (n=7) or chemoRT (n=80). Follow-up FDG-PET scans were performed at a median of 89 days (range 47-123 days) after RT. After a median follow-up of 49 months, median survival after PET response imaging was 28 months. Both qualitative response criteria (PeterMac and Deauville) showed perfect agreement (kappa = 1.0). Both semiquantitative criteria (EORTC and PERCIST) showed almost perfect agreement (kappa = 0.96). All four response criteria showed statistically significant associations with overall survival. The PeterMac and the Deauville criteria showed stronger survival associations (AIC=357.9) compared to EORTC (AIC=362.3) and PERCIST (AIC=362.6). The two qualitative criteria also performed better in the distinction between CMR and non-CMR (HR = 1.9, CI 1.0-3.4, p=0.047) versus EORTC (HR=1.2, CI 0.6-2.3, p=0.566) and PERCIST (HR 1.2, CI 0.6-2.3, p=0.548). Only 1, 4 and 6 patients had SMD in respectively PeterMac/Deauville, EORTC and PERCIST.
Conclusion
The visual PeterMac and Deauville criteria showed stronger predictive capacity than EORTC and PERCIST criteria, especially for distinguishing CMR from non-CMR.
Response criteria for 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) for thoracic malignancies include European Organization for Research and Treatment of Cancer (EORTC) criteria, Positron Emission tomography Response Criteria In Solid Tumors 1.0 (PERCIST), PeterMac Metabolic Visual Criteria and Deauville Criteria. It is unknown which criteria have the highest prognostic value in NSCLC.
Method
Between 2004 and 2016, three NSCLC prospective trials included patients treated with radical radiotherapy (RT) or chemoRT with baseline and post-treatment FDG-PET imaging. For each patient, the four FDG-PET response criteria were reported retrospectively and blinded to outcome. Responses to therapy were categorized as complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD) or progressive metabolic disease (PMD) and correlated with subsequent survival using Cox proportional hazard models, c-statistic, r2 and Akaike information criterion (AIC).
Result
Eighty-seven NSCLC patients underwent FDG-PET before and after radical RT (n=7) or chemoRT (n=80). Follow-up FDG-PET scans were performed at a median of 89 days (range 47-123 days) after RT. After a median follow-up of 49 months, median survival after PET response imaging was 28 months. Both qualitative response criteria (PeterMac and Deauville) showed perfect agreement (kappa = 1.0). Both semiquantitative criteria (EORTC and PERCIST) showed almost perfect agreement (kappa = 0.96). All four response criteria showed statistically significant associations with overall survival. The PeterMac and the Deauville criteria showed stronger survival associations (AIC=357.9) compared to EORTC (AIC=362.3) and PERCIST (AIC=362.6). The two qualitative criteria also performed better in the distinction between CMR and non-CMR (HR = 1.9, CI 1.0-3.4, p=0.047) versus EORTC (HR=1.2, CI 0.6-2.3, p=0.566) and PERCIST (HR 1.2, CI 0.6-2.3, p=0.548). Only 1, 4 and 6 patients had SMD in respectively PeterMac/Deauville, EORTC and PERCIST.
Conclusion
The visual PeterMac and Deauville criteria showed stronger predictive capacity than EORTC and PERCIST criteria, especially for distinguishing CMR from non-CMR.
| Original language | English |
|---|---|
| Pages (from-to) | s1875-s1876 |
| Number of pages | 2 |
| Journal | Journal of Thoracic Oncology |
| Volume | 12 |
| Issue number | 11, Supplement 2 |
| DOIs | |
| Publication status | Published - 01 Nov 2017 |
| Externally published | Yes |
