A Tracer Kinetic Model for 18F-FHBG for Quantitating Herpes Simplex Virus Type 1 Thymidine Kinase Reporter Gene Expression in Living Animals Using PET
Leeta Alison Green, PhD1,3,
Khoi Nguyen, BS1,2,
Bijan Berenji, BS1,2,
Meera Iyer, PhD1,2,
Eileen Bauer, MS1,2,
Jorge R. Barrio, PhD2,
Mohammad Namavari, PhD2,
Nagichettiar Satyamurthy, PhD2 and
Sanjiv S. Gambhir, MD, PhD1,4
1 The Crump Institute for Molecular Imaging, David Geffen School of Medicine at UCLA, Los Angeles, California
2 Department of Molecular & Medical Pharmacology, The Divisions of Nuclear Medicine, David Geffen School of Medicine at UCLA, Los Angeles, California
3 Department of Biomathematics, David Geffen School of Medicine at UCLA, Los Angeles, California
4 Department of Radiology and Bio-X Program, Stanford University, Stanford, California
View larger version (13K):
[in a new window]
|
FIGURE 1. Structure of 18F-FHBG. Presence of fluorine atom renders molecule chiral.
|
|
View larger version (16K):
[in a new window]
|
FIGURE 2. Working kinetic models. Top model describes pharmacokinetics of 18F-FHBG in tissues expressing HSV1-sr39tk gene. First compartment represents 18F-FHBG in plasma and middle compartment represents unmetabolized 18F-FHBG intracellularly. Third compartment represents metabolized 18F-FHBG that is trapped intracellularly. Parameters K1 and k2 are related to transport, and parameter k3 is related to phosphorylation. k4 is related to dephosphorylation rates. Bottom model describes behavior of 18F-FHBG in control mice. In second compartment, metabolites of 18F-FHBG are lumped together with the free form. In practice, k3 is zero due to absence of HSV1-sr39TK and, therefore, 18F-FHBG remains unmetabolized.
|
|
View larger version (11K):
[in a new window]
|
FIGURE 3. Ratio of 18F-FHBG in plasma vs. whole blood. 18F-FHBG was injected into tail vein of 4 Swiss–Webster mice and blood samples were taken. Concentration of 18F-FHBG in plasma and whole blood was determined, and ratio of 18F-FHBG concentration in plasma vs. whole blood as function of time is shown. Each point represents average of 4 mice.
|
|
View larger version (21K):
[in a new window]
|
FIGURE 4. 18F-FHBG sampled LV blood time–activity curve was compared with partial-volume–corrected (0.81) heart microPET image blood time–activity curves. Units of LV sampled blood time–activity curve were calibrated to microPET to scanner units using calibration factor obtained from scanning a cylinder from microPET scanner (17). (A) Control mouse (scanned an additional 20-min frame). (B) Mouse infected with 1.0 x 109 pfu of HSV1-sr39tk adenovirus. Both sets of curves show excellent agreement between sample LV blood time–activity curve and partial-volume–corrected heart image time–activity curve. TAC = time–activity curve; pix = pixel.
|
|
View larger version (19K):
[in a new window]
|
FIGURE 5. (A) Two-compartment model fit for control mouse, which was scanned dynamically using microPET with 18F-FHBG. Each point represents scanner counts per pixel per minute at indicated time, observed from ROI drawn on left lobe of liver. Two-compartment fit was the most parsimonious for control mice. (B) Example of 2- and 3-compartment model fits for representative HSV1-sr39TK–positive mouse, which was scanned with 18F-FHBG 2 d after injection of 1 x 108 pfu of adenovirus carrying HSV1-sr39tk gene. Circles represent observed microPET data taken from ROIs drawn on left lobe of liver. Three-compartment fit clearly improves over 2-compartment fit (F = 136). comp = compartment; pix = pixel.
|
|
View larger version (27K):
[in a new window]
|
FIGURE 6. (A) Correlation between %ID/g liver and HSV1-sr39TK enzyme activity. Each circle represents data from 1 mouse; %ID/g was obtained from images 1 h after probe injection. HSV1-sr39TK enzyme activity was obtained from liver tissue samples. (B) Correlation between k3 from 3-compartment model and HSV1-sr39TK enzyme activity. Each circle represents data from 1 mouse. (C) Correlation between Ki and HSV1-sr39TK enzyme activity. Each circle represents data from 1 mouse. Ki was obtained 1 h after probe injection. (D) Correlation between normalized mRNA and HSV1-sr39TK enzyme activity. Each circle represents data from 1 mouse.
|
|
Copyright © 2004 by the Society of Nuclear Medicine.
