Head and neck squamous cell carcinoma: diagnostic performance of diffusion-weighted MR imaging for the prediction of treatment response

Radiology. 2013 Feb;266(2):531-8. doi: 10.1148/radiol.12120167. Epub 2012 Nov 14.

Abstract

Purpose: To determine the diagnostic performance of diffusion-weighted (DW) imaging for the prediction of treatment failure in primary head and neck squamous cell carcinoma (HNSCC).

Materials and methods: The study was approved by the local institutional ethics committee and conducted with informed written consent in patients with primary HNSCC treated with radiation therapy and chemotherapy. DW imaging of the primary tumor was performed before treatment in 37 patients and was repeated within 2 weeks of treatment in 30 patients. Histograms of apparent diffusion coefficients (ADCs) were analyzed, and mean ADC, kurtosis, skewness, and their respective percentage change were correlated for local failure and local control at 2 years by using the Student t test. Univariate and multivariate analyses of the ADC parameters, T stage, and tumor volume were performed by using logistic regression for prediction of local failure.

Results: Local failure occurred in 16 of 37 (43%) patients and local control occurred in 21 of 37 (57%) patients. Pretreatment ADC parameters showed no correlation with local failure. There was significant intratreatment increase in mean ADC and a decrease in skewness and kurtosis (P < .001, P < .001, P = .024, respectively) for the whole group of patients when compared with those before treatment. During treatment, primary tumors showed a significantly lower increase in percentage change of mean ADC, higher skewness, and higher kurtosis for local failure than for local control (P = .016, .015, and .040, respectively). These ADC parameters also were significant for predicting local failure with use of univariate but not multivariate analysis.

Conclusion: Early intratreatment DW imaging has the potential to allow prediction of treatment response at the primary site in patients with HNSCC.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Carcinoma, Squamous Cell / diagnosis*
  • Carcinoma, Squamous Cell / therapy*
  • Combined Modality Therapy
  • Diffusion Magnetic Resonance Imaging / methods*
  • Female
  • Head and Neck Neoplasms / diagnosis*
  • Head and Neck Neoplasms / therapy*
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Neoplasm Staging
  • Predictive Value of Tests
  • Proportional Hazards Models
  • ROC Curve
  • Sensitivity and Specificity
  • Treatment Failure
  • Treatment Outcome