Early Prediction of Treatment Outcome: Functional MR Imaging and Standard Clinical Prognostic Factors in Advanced Cervical Cancer
NA Mayr,WT Yuh,JZ Wang,JF Montebello,S Nag,J Grecula,DH Wu,D Jarjoura,J Heverhagen,N Gupta,SM Edwards,MV Knopp
DOI: https://doi.org/10.1016/j.ijrobp.2005.07.167
2005-01-01
Abstract:Purpose/Objective: To evaluate dynamic contrast enhanced (DCE) MR parameters and correlate them with clinical prognostic factors, stage, nodal status and histology, for the early prediction of radiation therapy outcome in cervical cancer.Materials/Methods: Sixty-two cervical cancer patients stages IB2-IV treated with radiation/chemotherapy (RT/CT) underwent serial DCE MRI before (1st MRI), early (2nd MRI at 20–25 Gy/2 weeks) and at midpoint (3rd MRI at 40–50 Gy) of RT. Median follow up was 3.5 (0.9–6.3) years. MR parameters derived from the DCE curve, pixel histogram analysis of plateau signal intensity (SI) and lowest percentiles of poor-DCE tumor pixel populations within the entire tumor region, were correlated with local control and disease-free survival. The DCE parameters′ value as independent predictors was assessed by comparison with stage, histology and nodal status, using multivariate analysis.Results: DCE MR parameters predicted, as early as 2 weeks after RT start, local tumor control and disease-free survival (Figure). MR-based residual tumor volume and lowest 10th percentile of SI (SI 10) at 2 weeks were superior independent outcome predictors compared to stage, nodal status and histology (Table). The addition of the DCE MR parameters to clinical prognostic factors increased the sensitivity and specificity of the clinical factors from 71% and 62% to 100% and 76% for local recurrence, and from 83% and 68% to 100% and 82% disease-free survival. In low-risk patients (stage I–II, negative lymph nodes, n=33), SI 10<1.73 predicted local failure and death of disease with a sensitivity of 100% and specificity of 77%.Conclusions: This data suggests that microcirculation-based DCE MR parameters independently predict treatment outcome as early as 2 weeks after RT/CT start - allowing early modification of treatment to improve outcome. The predictive value of gold-standard clinical prognostic factors can be further enhanced by the addition of DCE-MR parameters.Tabled 1MR Parameters as Independent Predictors for Local Control and Disease-Free SurvivalKaplan-Meier analysis of disease-free survival based on lowest 10th SI percentile (≤1.73 vs. >1.73)View Large Image Figure ViewerDownload (PPT) Purpose/Objective: To evaluate dynamic contrast enhanced (DCE) MR parameters and correlate them with clinical prognostic factors, stage, nodal status and histology, for the early prediction of radiation therapy outcome in cervical cancer. Materials/Methods: Sixty-two cervical cancer patients stages IB2-IV treated with radiation/chemotherapy (RT/CT) underwent serial DCE MRI before (1st MRI), early (2nd MRI at 20–25 Gy/2 weeks) and at midpoint (3rd MRI at 40–50 Gy) of RT. Median follow up was 3.5 (0.9–6.3) years. MR parameters derived from the DCE curve, pixel histogram analysis of plateau signal intensity (SI) and lowest percentiles of poor-DCE tumor pixel populations within the entire tumor region, were correlated with local control and disease-free survival. The DCE parameters′ value as independent predictors was assessed by comparison with stage, histology and nodal status, using multivariate analysis. Results: DCE MR parameters predicted, as early as 2 weeks after RT start, local tumor control and disease-free survival (Figure). MR-based residual tumor volume and lowest 10th percentile of SI (SI 10) at 2 weeks were superior independent outcome predictors compared to stage, nodal status and histology (Table). The addition of the DCE MR parameters to clinical prognostic factors increased the sensitivity and specificity of the clinical factors from 71% and 62% to 100% and 76% for local recurrence, and from 83% and 68% to 100% and 82% disease-free survival. In low-risk patients (stage I–II, negative lymph nodes, n=33), SI 10<1.73 predicted local failure and death of disease with a sensitivity of 100% and specificity of 77%. Conclusions: This data suggests that microcirculation-based DCE MR parameters independently predict treatment outcome as early as 2 weeks after RT/CT start - allowing early modification of treatment to improve outcome. The predictive value of gold-standard clinical prognostic factors can be further enhanced by the addition of DCE-MR parameters.