Predictive value of (18)FDG PET-CT for tumour response in patients with locally advanced rectal cancer treated by preoperative chemoradiotherapy.

PURPOSE: Although (18)fluorine-2-deoxy-D-glucose positron emission tomography-computed tomography ((18)FDG PET-CT) is considered a reliable modality for determining tumour response after neoadjuvant chemoradiotherapy (CRT) in locally advanced rectal cancer (LARC), the role of (18)FDG PET-CT for predicting pathologic complete response (pCR) remains unclear. The aim of this study was to evaluate whether (18)FDG PET-CT can predict tumour response after CRT in patients with LARC, in terms of downstaging and pCR.

METHODS: Between March 2009 and February 2012, 151 patients with LARC treated with neoadjuvant CRT followed by radical surgery were reviewed retrospectively. Pre-CRT SUVmax (maximum standardized uptake value), post-CRT SUVmax, ΔSUVmax (difference between pre- and post-CRT SUVmax), and RI-SUV (response index) were measured before and after CRT. Univariate and multivariate analyses were used to analyse the association of PET-CT-related parameters and clinical variables, to assess downstaging and pCR.

RESULTS: Downstaging occurred in 48 patients (31.7 %) and pCR in 19 patients (12.5 %). Univariate and multivariate analysis revealed post-CRT SUVmax as a significant factor for prediction of downstaging, with sensitivity of 60.4 %, specificity of 65.0 %, and accuracy of 55.9 %, for a cutoff value of 3.70. Regarding pCR, post-CRT SUVmax was again found as a significant parameter by univariate and multivariate analysis, with sensitivity of 73.7 %, specificity of 63.7 %, and accuracy of 64.9 %, for a cutoff value of 3.55.

CONCLUSIONS: The results indicate that post-CRT SUVmax independently predicts downstaging and pCR. However, the predictive values of post-CRT SUVmax for tumour response after neoadjuvant CRT are too low in sensitivity and specificity to change the treatment plan for LARC.