Comparison of metabolic variables in FDG PET/CT in rectal cancer patients who achieved a complete or incomplete response following neoadjuvant chemoradiotherapy.

OBJECTIVE

To evaluate qualitative and quantitative metabolic parameters on 18 F-fluorodeoxyglucose positron emission tomography/computed tomography [FDG PET/CT] for differentiating complete from incomplete response in patients with locally advanced rectal cancer after neoadjuvant chemoradiotherapy.

MATERIALS AND

METHODS

This retrospective single-center study included 74 patients with locally advanced rectal cancer who underwent FDG PET/CT before and after neoadjuvant chemoradiotherapy. Patients were classified as good responders [complete clinical response, n = 34] or poor responders [n = 40]. Visual response was assessed using a five-point visual response score. Quantitative parameters included SUVmax, SUVmean, metabolic tumor volume, and total lesion glycolysis, using a fixed 41% threshold and best visual fit segmentation. Group comparisons, receiver operating characteristic analysis, and logistic regression were performed.

RESULTS

VRS grade 4 [no uptake post-treatment] demonstrated high specificity [95%] and positive predictive value [90.9%] for identifying good responders, with a sensitivity of 58.8%. Two false positives were observed [carcinoma in situ and mucinous tumor]. Significant quantitative differentiators included post-treatment SUVmax41% [p < 0.0001], SUVmaxBVF [p < 0.0001], SUVmean41% [p < 0.0001], SUVmeanBVF [p < 0.0001], post-treatment TLGBVF [p = 0.0164], ΔSUVmax [p = 0.0306], and ΔSUVmean41% [p = 0.0072]. Post-treatment SUVmax showed the highest accuracy, with an area under the curve [AUC] of 0.917.

CONCLUSIONS

The present study suggests that FDG-PET/CT provides valuable complementary information for response assessment after nCRT in rectal cancer and may assist multidisciplinary decision-making regarding organ-preserving strategies. Absence of residual uptake on post-treatment imaging and post-treatment SUVmax demonstrated high diagnostic accuracy for identifying complete responders.