INTRODUCTION. Concomitant chemoradiotherapy (CRT) with cisplatin is a non-surgical method of choice for treatment of loco-regionally advanced head and neck squamous cell carcinoma (HNSCC). Radiologic imaging with capacity to predict treatment efficacy before or early during therapy would assisst in changing the therapy in case of non-responsive tumors.
The role of our study was to assess wheater MR diffusion and perfusion parameters, measured before treatment or their changes early during treatment could predict treatment response in patients with HNSCC, treated with cisplatin based CRT.
METHODS. MR diffusion and perfusion imaging was performed in 20 patients with loco-regionally advanced HNSCC before treatment, after 1 week (e.g. after 10 Gy) and after 4 weeks (e.g. after 40 Gy) of CRT, and 2,5 - 3 months after treatment completion. At each time point, apparent diffusion coefficient [ADC] was measured in the primary tumor with MR diffusion examination, as well as three MR perfusion parameters (volume transfer constant between extracellular extravascular space (EES) and blood plasma [Ktrans], EES volume per unit tissue volume [Ve] and blood plasma volume per unit tissue volume [Vp]). Baseline values and relative changes of all parameters after 10 Gy and 40 Gy (from baseline) were analysed by Firth logistic regression in order to determine predictive factors for treatment response. Spearman's rank correlation coefficient was used to asses the association between tumour volume and MR functional parametrs.
RESULTS. Relative change (from baseline) of Ktrans after 10 Gy was found to be a significant factor for prediction whether patients would response to treatment with CRT (p = 0,047; 95% confidence interval (CI), 0,99 – 0,87; median relative change of Ktrans after 10 Gy from baseline, 50,2 %; range, 25,1 % to 90,9 %). In patients with complete response to treatment, there was a borderline sginificant decrease of Ve after 10 Gy (p = 0,066; 26,9 %, from 76,7 % to 126,8 %). In addition, ADC after 40 Gy increased with a borderline significance (p = 0,079; 73,4 %; from 17,8 % to 121,6 %). None of the parameters measured before treatment appeared to be of statistical significance. Correlation canalysis showed no statistically significant association between tumour volume and measured MR parametrs at any time point during treatment.
CONCLUSIONS. Changes of Ktrans, Ve and ADC recognized early during treatment could potentionally help in prediction what would be a response of HNSCC to CRT treatment with cisplatin. MR diffusion and perfusion parametrs did not correlate with changes in volume of volume as measured with MR volumetry.