Apparent diffusion coefficient histogram shape analysis for monitoring early response in patients with advanced cervical cancers undergoing concurrent chemo-radiotherapy

BACKGROUND: To explore the role of apparent diffusion coefficient (ADC) histogram shape related parameters in early assessment of treatment response during the concurrent chemo-radiotherapy (CCRT) course of advanced cervical cancers. METHODS: This prospective study was approved by the local ethics committee and informed consent was obtained from all patients. Thirty-two patients with advanced cervical squamous cell carcinomas underwent diffusion weighted magnetic resonance imaging (b values, 0 and 800 s/mm(2)) before CCRT, at the end of 2nd and 4th week during CCRT and immediately after CCRT completion. Whole lesion ADC histogram analysis generated several histogram shape related parameters including skewness, kurtosis, s-sD(av), width, standard deviation, as well as first-order entropy and second-order entropies. The averaged ADC histograms of 32 patients were generated to visually observe dynamic changes of the histogram shape following CCRT. RESULTS: All parameters except width and standard deviation showed significant changes during CCRT (all P < 0.05), and their variation trends fell into four different patterns. Skewness and kurtosis both showed high early decline rate (43.10 %, 48.29 %) at the end of 2nd week of CCRT. All entropies kept decreasing significantly since 2 weeks after CCRT initiated. The shape of averaged ADC histogram also changed obviously following CCRT. CONCLUSIONS: ADC histogram shape analysis held the potential in monitoring early tumor response in patients with advanced cervical cancers undergoing CCRT.