Send to

Choose Destination
Radiology. 2016 Apr;279(1):158-66. doi: 10.1148/radiol.2015150400. Epub 2015 Oct 27.

Association of Apparent Diffusion Coefficient with Disease Recurrence in Patients with Locally Advanced Cervical Cancer Treated with Radical Chemotherapy and Radiation Therapy.

Author information

From the Radiation Medicine Program, Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada (A.G., M.M., A.F., J.X., N.B., W.L., L.M., W.F., K.H.); Departments of Radiation Oncology (A.G., M.M., A.F., J.X., N.B., W.L., L.M., W.F., K.H.) and Biostatistics (H.J.), University of Toronto, 610 University Ave, Toronto, ON, Canada M5G 2M9; and Department of Medical Imaging, University Health Network, Toronto, ON, Canada (J.H., U.R.).



To investigate whether volumetrically derived apparent diffusion coefficient (ADC) from pretreatment diffusion-weighted (DW) magnetic resonance (MR) imaging is associated with disease recurrence in women with locally advanced cervical cancer treated with chemotherapy and radiation therapy.


An ethics board-approved, retrospective study was conducted in 85 women with stage IB-IVA cervical cancer treated with chemo- and radiation therapy in 2009-2013. All patients underwent MR imaging for staging, including T2-weighted and DW MR imaging series, by using a 1.5- or 3.0-T imager. The mean, median, 75th, 90th, and 95th percentile ADCs (ADCmean, ADC50, ADC75, ADC90, and ADC95, respectively) of all voxels that comprised each tumor were extracted and normalized to the mean urine ADC (nADCmean, nADC50, nADC75, nADC90, and nADC95, respectively) to reduce variability. The primary outcome was disease-free survival (DFS). Uni- and multivariable Cox regression analyses were used to evaluate the association of ADC parameters and relevant clinical variables with DFS.


Of the 85 women included, 62 were free of disease at last follow-up. Median follow-up was 37 months (range, 5-68 months). Significant variables at univariable analysis included T2-weighted derived tumor diameter, para-aortic nodal involvement, advanced stage, ADC90 and ADC95, nADC75, nADC90, and nADC95. Normalized parameters were more highly associated (hazard ratio per 0.01 increase in normalized ADC, 0.91-0.94; P < .04). Because nADC75, nADC90, and nADC95 were highly correlated, only nADC95 (which had the lowest P value) was included in multivariable analysis. At multivariable analysis, absolute and normalized ADC95 remained associated with DFS (hazard ratio, 0.90-0.98; P < .05).


The volumetric ADC95 may be a useful imaging metric to predict treatment failure in patients with locally advanced cervical cancer treated with chemo- and radiation therapy.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center