Intravoxel Incoherent Motion Magnetic Resonance Imaging of Oropharyngeal Cancer in Response to Chemoradiation Therapy

Abstract

To investigate the utility of intravoxel incoherent motion (IVIM) MRI as a response indictor in human receiving chemoradiation therapy for oropharyngeal cancer. Eight male patients with histologically documented stage II/III squamous cell carcinoma of the oropharynx were included in this prospective study. IVIM-MRI was carried out at a 3.0-T GE MRI scanner using laterally placed 6-element flex coils. Patients were scanned twice; the first scan was at baseline prior to the start of radiation therapy (RT) and the second scan was at mid-treatment after delivery 30-55 Gy of RT. All patients were scanned in supine position with the same RT immobilization devices including individualized thermoplastic head and shoulder mask, customized mold, and dental stent to reduce motion artifacts and to improve image registration in longitudinal scans. IVIM parameters (D, pure diffusion coefficient; f, perfusion fraction; D*, pseudodiffusion coefficient) were calculated on a pixel-by-pixel basis using a bi-exponential model implemented within ImageJ with ten b-values ranging from 0-800 s/mm2. A fitting threshold of R2 > 0.5 was applied to all parametric maps. Based on post gadolinium T1-weighted images, tumors were contoured on IVIM maps for the purpose of comparing changes of IVIM parameters with tumor response. Paired-samples Wilcoxon signed rank test was used to compare IVIM parameters for assessment of treatment response. The mean IVIM parameters in pre-RT tumors were: D = 8.94 ± 1.67 (× 10-4 mm2/s), f = 0.23 ± 0.11, D* = 20.52 ± 9.71 (× 10-3 mm2/s); while those in mid-RT tumors were: D = 15.43 ± 3.68 (× 10-4 mm2/s), f = 0.34 ± 0.20, D* = 30.63 ± 8.47 (× 10-3 mm2/s). D and f were statistically significantly higher in mid-RT tumors (P < 0.005 and P < 0.02, respectively), while changes of D* in tumor tissues during RT were considered to be not significant at this sample size (P < 0.053). The preliminary results of this study show IVIM parameters change during RT. In particular, D and f are robust, useful clinical markers with a high level of confidence. On the other hand, D* also show potential feasibility in assessing treatment response. We continue to accrue study patients and to collect long-term clinical outcomes to corroborate the reproducibility and clinical significance of these pilot findings.