Comparison of 18F-FDG PET/CT and DW-MRI in assessment of neoadjuvant radiochemotherapy response in locally advanced rectal cancer patients

Abstract Objective Our aim is to evaluate if different metabolic parameters obtained by 18F-FDG PET/CT and diffusion weighted magnetic resonance imaging (DW-MRI) can aid in neoadjuvant radiochemotherapy (RCT) response assessment in locally advanced rectal cancer (LARC) patients. Methods Out of 20 LARC patients, who were planned to receive neoadjuvant RCT, 19 were included in this prospective study. Patients had 18F-FDG PET/CT and DW-MRI at initial staging, interim (2 weeks after onset of RCT) and after completion of RCT (post-therapy). Standardized uptake value (SUV) parameters (SUVmax, SUVmean, SUVpeak, SULpeak), metabolic tumor volume (MTV) and tumor lesion glycolysis (TLG) detected on PET images and apparent diffusion coefficient (ADC) values (for b = 400 and b = 1000 s/mm2) obtained from DW-MRI were recorded. Postoperative tumor regression grade (TRG) was used as gold-standard, except for 2 patients who were under complete remission with non-operative management 19 months post-therapy and scored as responders. Results On interim PET/CT, no significant difference was found among PET parameters between responders and non-responders, whereas post-therapy SUVmax, SUVpeak, MTV, SULpeak, TLG (p = 0.02, p = 0.014, p = 0.025, p = 0.007, p = 0.02, respectively) and initial MTV (p = 0.034) were significantly lower in responders. ADC response index (RI) was higher in responders (interim p = 0.026; post-therapy: p = 0.018) and ROC analysis revealed that a threshold of ADC RI > 41.6% for interim MRI and >44.6% for post-therapy MRI had sensitivity and specificity of 75.0% and 90.9%, respectively. Conclusions While interim 18F-FDG PET/CT failed to predict therapy response during RCT, post-therapy PET could accurately differentiate responders. DW-MRI was found to be more promising in interim detection of RCT response.