Response of locally advanced rectal cancer (LARC) to radiochemotherapy: DW-MRI and multiparametric PET/CT in correlation with histopathology

Abstract Aim To prospectively evaluate histological significance and predictive value of changes in apparent diffusion coefficient (ADC) and 18F-FDG PET/CT parameters in locally advanced rectal cancer (LARC) after neoadjuvant radiochemotherapy (RCT). Methods Twenty-one patients with untreated LARC underwent pre-RCT and post-RCT 18F-FDG PET/CT and diffusion-weighted magnetic resonance imaging (DW-MRI), followed by surgery. For both datasets, two readers measured the tumor SUVmax, SUVmean, MTV, TLG, ADCmin, ADCmean, and respective differences (∆SUVmax, ∆SUVmean, ∆MTV, ∆TLG, ∆ADCmin, ∆ADCmean) for the whole tumor. Tumor regression grade according to Mandard (TRGm), percentage of residual tumor cells and fibrosis were estimated by two pathologists in consensus. Relationship between parameters was assessed on stepwise multivariate regression analysis and ROC curve analysis to evaluate their performance and predict the treatment response. Results Eighteen LARCs were analyzed. SUVmax and SUVmean decreased from 21.3 ± 8.9 to 9.3 ± 5.5 g/mL, (p = 0.0002) and 12.3 ± 5.1 to 5.4 ± 3.1 g/mL, (p = 0.0002), respectively, after RCT, whereas ADCmin and ADCmean increased from 396 ± 269 to 573 ± 313×10–6 mm2/s (p = 0.014) and 1159 ± 212 to 1355 ± 194×10–6 mm2/s (p = 0.0008), respectively. TRGm and percentage of residual tumor cells independently correlated with post-RCT SUVmean (β = 0.73 and β = 0.76, p < 0.001) and post-RCT SUVmax (β = 0.72 and β = 0.78, p < 0.001), whereas percentage of fibrosis independently correlated with ∆ADCmean (β = 0.38, p = 0.008). Post-RCT, SUVmax and SUVmean performed well in predicting TRGm < 3 and residual tumor cells ≤ 20 %. ΔADCmean predicted fibrosis > 70 % well. Conclusion Post-RCT SUVmean, SUVmax and ∆ADCmean are complementary parameters for respectively evaluating residual tumor burden and amount of fibrosis in LARC. However, only SUV independently correlated with TRGm.