A Curve-Fitting Approach to Estimate the Arterial Plasma Input Function for the Assessment of Glucose Metabolic Rate and Response to Treatment

Journal of Nuclear Medicine

Wim J.G. Oyen¹  Eric P. Visser¹  Lioe-Fee de Geus-Oei¹  Dennis Vriens¹ 

[1] Department of Nuclear Medicine, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands Department of Nuclear Medicine, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands Department of Nuclear Medicine, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands

关键词: ¹⁸F-FDG;    positron emission tomography;    chemotherapy;    therapy monitoring;    input function;    theoretical models;    pharmacokinetics;    body fluid compartments;    least-squares analysis;    tissue distribution;   

DOI  :  10.2967/jnumed.109.065243

学科分类：医学（综合）

来源: Society of Nuclear Medicine

【 摘 要 】

For the quantification of dynamic 18F-FDG PET studies, the arterial plasma time–activity concentration curve (APTAC) needs to be available. This can be obtained using serial sampling of arterial blood or an image-derived input function (IDIF). Arterial sampling is invasive and often not feasible in practice; IDIFs are biased because of partial-volume effects and cannot be used when no large arterial blood pool is in the field of view. We propose a mathematic function, consisting of an initial linear rising activity concentration followed by a triexponential decay, to describe the APTAC. This function was fitted to 80 oncologic patients and verified for 40 different oncologic patients by area-under-the-curve (AUC) comparison, Patlak glucose metabolic rate (MRglc) estimation, and therapy response monitoring (ΔMRglc). The proposed function was compared with the gold standard (serial arterial sampling) and the IDIF. Methods: To determine the free parameters of the function, plasma time–activity curves based on arterial samples in 80 patients were fitted after normalization for administered activity (AA) and initial distribution volume (iDV) of 18F-FDG. The medians of these free parameters were used for the model. In 40 other patients (20 baseline and 20 follow-up dynamic 18F-FDG PET scans), this model was validated. The population-based curve, individually calibrated by AA and iDV (APTACAA/iDV), by 1 late arterial sample (APTAC1sample), and by the individual IDIF (APTACIDIF), was compared with the gold standard of serial arterial sampling (APTACsampled) using the AUC. Additionally, these 3 methods of APTAC determination were evaluated with Patlak MRglc estimation and with ΔMRglc for therapy effects using serial sampling as the gold standard. Results: Excellent individual fits to the function were derived with significantly different decay constants (P < 0.001). Correlations between AUC from APTACAA/iDV, APTAC1sample, and APTACIDIF with the gold standard (APTACsampled) were 0.880, 0.994, and 0.856, respectively. For MRglc, these correlations were 0.963, 0.994, and 0.966, respectively. In response monitoring, these correlations were 0.947, 0.982, and 0.949, respectively. Additional scaling by 1 late arterial sample showed a significant improvement (P < 0.001). Conclusion: The fitted input function calibrated for AA and iDV performed similarly to IDIF. Performance improved significantly using 1 late arterial sample. The proposed model can be used when an IDIF is not available or when serial arterial sampling is not feasible.

【 授权许可】

Unknown