放射学组学杂志

抽象的

Diffusion Weighted and Dynamic Contrast-Enhanced MRI as a Predictor of Treatment Response in Head and Neck Cancer to Chemoradiotherapy: The Role of Early Intratreatment Scanning

Garbajs M, Strojan P and Surlan-Popovic K

Background: In locoregionally advanced Head and Neck Squamous Cell Carcinoma (HNSCC), intra-treatment scanning early during the course of Concomitant Chemoradiotherapy (CRT) with advanced functional Magnetic Resonance Imaging (MRI) could potentially allow for modification of treatment to maximize the chance of favourable outcome. The aim of this prospective study was to assess the predictive value of parameters derived from Dynamic Contrast-Enhanced (DCE) and Diffusion Weighted (DW) MRI for treatment response early during the course of CRT in patients with HNSCC.
Methods: MRI scans were performed in 20 patients with locoregionally advanced HNSCC at baseline, after 10 Grays (Gy) and 40 Gy of CRT. DW and DCE derived parameters as well as the volumes were measured from primary tumours and lymph nodes. Tumour kinetic parameters (volume transfer constant (K_trans), extracellular extravascular volume fraction (v_e) and plasma volume fraction (V_p)) were assessed using the extended Tofts model. Relative changes in studied parameters from baseline to 10 Gy and 40 Gy were calculated. Factors predictive for treatment response were identified by the Firth logistic regression. Spearman’s rank correlation coefficient was used to investigate correlations among the parameters.
Results: Responders showed a significant decrease in K_trans after 10 Gy (median, -50.2 %; range, -25.1 to -90.9 %; P = 0.047). In addition, decreased v_e after 10 Gy (median, -26.9%; range -76.7 to -126.8%) and increased apparent diffusion coefficient (ADC) after 40 Gy (median, 73.4%; range, 17.8 - 121.6%) were borderline significant (P=0.066 and P=0.079, respectively). Positive correlation between K_trans and v_e after 10 Gy (r = 0.823, P <0.05) was noticed.
Conclusion: In HNSCC, early changes in DCE- and DWI-MRI derived parameters appear to predict tumour response before the actual morphologic changes occur.