Arterial input function for quantitative dynamic contrast-enhanced MRI to diagnose prostate cancer

Area of Science:

• Radiology and Imaging Science
• Oncology and Urology Research
• Biomedical Engineering and Medical Physics

Background:

• Distinguishing prostate cancer (PCa) from benign lesions in the transition zone (TZ) and peripheral zone (PZ) is crucial for accurate diagnosis and treatment.
• Quantitative dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) offers potential for characterizing prostate tissue, but its efficacy depends on reliable arterial input function (AIF) determination.
• Standardization of AIF methods and perfusion parameters is needed to optimize PCa detection using DCE-MRI.

Purpose of the Study:

• To evaluate the ability of quantitative DCE-MRI to differentiate between prostate cancer (PCa) and benign lesions in the transition zone (TZ) and peripheral zone (PZ).
• To compare different methods for arterial input function (AIF) determination, including manual (AIFm), automated (AIFa), and population-based (AIFp) approaches.
• To identify the optimal quantitative perfusion parameters and AIF method for accurate PCa detection.

Main Methods:

• Retrospective analysis of DCE-MRI data from 50 consecutive patients with PCa who underwent multiparametric MRI.
• Application of three distinct AIF determination methods: manual region of interest (AIFm), automated algorithm (AIFa), and population-based (AIFp).
• Analysis of quantitative perfusion parameters (Ktrans, ve, kep) derived from the Tofts model in PCa, PZ, and TZ tissues across the different AIF methods.

Main Results:

• Ktrans and kep values were significantly higher in PCa compared to benign tissue, irrespective of the AIF method used.
• In the peripheral zone (PZ), both Ktrans and kep effectively differentiated PCa from benign tissue (P < .001).
• In the transition zone (TZ), only kep using the population-based AIF (AIFp) showed a significant difference (P = .039), while automated AIF (AIFa) demonstrated higher correlation with manual AIF (AIFm) for parameter estimation.

Conclusions:

• Ktrans and kep parameters can reliably differentiate PCa from benign PZ tissue, independent of the AIF determination method.
• The automated AIF (AIFa) method is considered the most feasible for routine clinical practice due to its efficiency and correlation with manual methods.
• Quantitative perfusion parameters showed limited success in differentiating PCa within the transition zone (TZ), highlighting a need for further research in this area.