N eoadjuvant chemoradiotheraphy (CRT) improves tumor down- staging, pathological complete response (pCR), and local control (1, 2). pCR rates of 13%–30% have been reported in phase II and phase III trials following 5-fluorouracil-based preoperative CRT (3, 4). Currently, management of patients with clinical complete response (cCR) remains controversial (5–8). A recent meta-analysis including 218 phase I/II or retrospective studies and 28 phase III trials of adjuvant CRT reported that T3 rectal cancer is associated with high local recurrence rates after nonsurgical treatment (9). In addition, similiar results were recently shown from a study using a “wait-and-see” policy after CRT (10). Accurate imaging methods are needed to evaluate CRT responses, and post-CRT magnetic resonance imaging (MRI) is frequently used for this purpose. However, the method has low accuracy in predicting the patho- logical stage of the tumor and can often overstage T1 and T2 tumors due to the limited capability of MRI to differentiate viable tumor, residual fi- brotic nontumoral tissue, and a desmoplastic reaction. Understaging of irradiated rectal cancer can affect treatment planning, including the surgi- cal strategy, and thus affects the tumor recurrence rate and prognosis (11). Diffusion-weighted (DW)-MRI is a functional imaging technique that yields qualitative and quantitative information and provides unique insights regarding tumor cellularity, integrity of cell membranes, and microcirculation. The motion of water molecules is more restricted in tissues with a high cellular density that are associated with numerous in- tact cell membranes (e.g., tumor tissue). Apparent diffusion coefficients (ADCs), which are quantitative expressions of diffusion characteristics of tissues, tend to decrease in diffusion restricted areas, whereas diffu- sion signal intensity (SI), which is the qualitative parameter of diffusion, increases in those areas (12). DW-MRI has been used in the diagnosis of rectal tumors since the 1990s. In recent years, researchers have used this technique to deter- mine the therapeutic efficacy of CRT in locally advanced rectal tumors (12, 13). However, a limited number of retrospective studies exists in the literature that have assessed DW-MRI for the determination of CR to neoadjuvant CRT (14–16). In this study, we prospectively examined whether DW-MRI can ac- curately predict CRs in patients with locally advanced rectal cancer by analyzing pre- and post-CRT changes in qualitative (SI) and quantitative (ADC) parameters. Materials and methods Patients Between May 2007 and April 2010, 30 patients (mean age, 54.3 years; range, 37–74 years; 8 females and 22 males) who had locally advanced Can diffusion-weighted MRI determine complete responders after neoadjuvant chemoradiation for locally advanced rectal cancer? Gülgün Engin, Rasul Sharifov, Zeynep Güral, Esra Kaytan Sağlam, Sezer Sağlam, Emre Balk, Oktar Asoğlu, Sümer Yamaner, Mine Güllüoğlu, Yersu Kapran, Sevda Özel PURPOSE We aimed to prospectively determine if analyzing pre- and post-chemoradiotheraphy (CRT) changes in the signal inten- sity (SI) and apperent diffusion coefficient (ADC) values from diffusion-weighted magnetic resonance imaging (DW-MRI) can accurately predict complete responders for locally ad- vanced rectal cancer. MATERIALS AND METHODS Thirty patients (mean age, 54.3 years) with locally advanced rectal cancer who underwent neoadjuvant CRT and subse- quent surgery were included in this study. All patients were evaluated pre- and post-CRT by standardized turbo spin echo and DW-MRI. Pre- and post-CRT tumor and normal rectal wall SI (which were gradually scored as very high, high, intermedi- ate, low, and no signal) and ADC values were recorded. RESULTS Tumor SIs were decreased in all of the patients that had a therapy response. However, complete tumor SI loss was only seen in two (22.2%) of nine patients with a pathological com- plete response, while it regressed to low and/or intermediate SI levels in the remaining seven patients (77.8%). Post-CRT ADC values of rectal tumors were significantly higher from the pre- CRT ADC values (P < 0.0001; Z=-9.39). However, post-CRT ADC values from the complete and partial/no response patient groups were not significantly different (P = 0.071; Z=-1.99). CONCLUSION In re-staging of rectal tumors by DW-MRI, an increase in ADC values and decrease in SIs can predict therapy response but cannot unequivocally determine a complete response. Key words: • colorectal carcinoma • neoadjuvant therapy • treatment outcome • diffusion magnetic resonance imaging • diagnostic imaging From the Oncology Institute (G.E.  gengin@istanbul.edu.tr, S.S.), Istanbul University, İstanbul, Turkey; the Departments of Radiology (R.S.), Radiation Oncology (Z.G., E.K.S.), General Surgery (E.B., O.A., S.Y.), Pathology (M.G., Y.K.), and Public Health (S.Ö.), Istanbul University Istanbul School of Medicine, İstanbul, Turkey. Received 14 February 2012; revision requested 13 March 2012; revision received 18 April 2012; accepted 1 May 2012. Published online 13 July 2012 DOI 10.4261/1305-3825.DIR.5755-12.1 Diagn Interv Radiol 2012; 18:574–581 © Turkish Society of Radiology 2012 ABDOMINAL IMAGING ORIGINAL ARTICLE 574