Apparent diffusion coefficient for evaluating tumour response to neoadjuvant chemoradiation therapy for locally advanced rectal cancer

Abstract

To determine whether change in the apparent diffusion coefficient (ADC) before and after neoadjuvant chemoradiation therapy (CRT) for locally advanced rectal cancer (LARC) would more accurately predict pathological complete response (pCR) than analysing the pre- or post-CRT ADC individually. 76 patients with LARC (≥ T3 or lymph node-positive) underwent pre- and post-CRT, 1.5-T diffusion-weighted imaging. A blinded radiologist measured the tumour ADCs. The three measures were compared with regard to CR and non-CR. To calculate accuracy for determining pCR, receiver-operating characteristic curve (ROC) analysis was applied to the three measures. The post-CRT ADC (1.43 ± 0.10) (× 10(-3) mm(2)/s) and the percentage change (70.0 ± 23.5%) in the CR (n = 11) were significantly higher than those in the non-CR (n = 65) (1.14 ± 0.18, 30.2 ± 21.7%, respectively)(both, P < 0.0001). The pre-CRT ADC (0.85 ± 0.10) of the pCR did not differ significantly from that of the non-pCR (0.88 ± 0.14)(P = 0.4094). The post-CRT ADC showed comparable relative accuracy for determining pCR to that of the percentage change (P = 0.788). However, overall accuracy of the former (87%) was higher than that of the latter (75%) (P = 0.049). Analyzing post-CRT ADC alone can reliably differentiate pCR from non-pCR in LARC.