Dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) as a predictor of response in head and neck squamous cell carcinoma (HNSCC)

Abstract

6007 Background: We wish to assess on a preliminary basis whether pretreatment DCE-MRI predicts response to chemoradiation in HNSCC. Methods: Tumor perfusion was assessed in 14 patients (median age, 56y; 13 M, 1 F; squamous cell cancer 13, poorly differentiated carcinoma 1; base of tongue 6; tonsil 6; larynx 1; nasopharynx 1) with nodal metastases (size >1 cc for DCE-MRI; N1, 3; N2a, 2; N2b, 4; N2c, 4; N3, 1; Stage III, 3; Stage IVa, 10; Stage IVb, 1) prior to platin-based concurrent chemoradiation therapy. Data were acquired on a 1.5 Tesla clinical scanner with a standard MRI protocol using a clinical contrast agent (Gd-DTPA). The two compartment model analysis measured the rate constants of the contrast agent transfer between the lesion and plasma compartments (kep) and elimination by the plasma (kel). Each patient was assigned a single MR imaging parameter of uptake slope and compartmental model (Akep) on the basis of the histogram analysis of all individually fit tumor voxels. Histogram analysis calculated the amplitude (a), width (s), and median (μ) of the distribution from the fitting procedure. Overall response was determined approximately 2–5 months post treatment. Results: DCE-MRI data showed a rapid rise in the time intensity curves for viable appearing tumor, whereas muscle showed minimal signal enhancement. The mean Akep value was higher for complete responders (CR/near CR, n = 7) than for incomplete responders (ICR, n = 7; 3 partial response, 4 no response) (12.29/min±4.09 vs 7.11/min±2.51 [P = 0.04]. The preliminary histogram analysis for the slope showed that the width and median were able to differentiate between CR and ICR, P=0.039 and P=0.013 respectively, while amplitude was not significant (P=0.150). Conclusions: These preliminary data suggest that DCE-MRI may prove to be a useful predictor of response in HNSCC patients, and thus may enhance prognostication, patient selection, and treatment outcomes. No significant financial relationships to disclose.