Hans Jürgen Machulla

Simultaneous PET-MRI: a new approach for functional and morphological imaging

Noninvasive imaging at the molecular level is an emerging field in biomedical research. This paper introduces a new technology synergizing two leading imaging methodologies: positron emission tomography (PET) and magnetic resonance imaging (MRI). Although the value of PET lies in its high-sensitivity tracking of biomarkers in vivo, it lacks resolving morphology. MRI has lower sensitivity, but produces high soft-tissue contrast and provides spectroscopic information and functional MRI (fMRI). We have developed a three-dimensional animal PET scanner that is built into a 7-T MRI. Our evaluations show that both modalities preserve their functionality, even when operated isochronously. With this combined imaging system, we simultaneously acquired functional and morphological PET-MRI data from living mice. PET-MRI provides a powerful tool for studying biology and pathology in preclinical research and has great potential for clinical applications. Combining fMRI and spectroscopy with PET paves the way for a new perspective in molecular imaging.

PET imaging of prostate cancer xenografts with a highly specific antibody against the prostate-specific membrane antigen.

Prostate-specific membrane antigen (PSMA), a transmembrane glycoprotein, is highly expressed by virtually all prostate cancers and is currently the focus of several diagnostic and therapeutic strategies. We have previously reported on the generation of several monoclonal antibodies (mAb) and antibody fragments that recognize and bind with high affinity to the extracellular domain of cell-adherent PSMA. This article reports the in vivo behavior and tumor uptake of the radiolabeled anti-PSMA mAb 3/A12 and its potential as a tracer for PET. Methods: The mAb 3/A12 was conjugated with the chelating agent 1,4,7,10-tetraazacyclododecane-N,N′,N″,N′″-tetraacetic acid (DOTA) and radiolabeled with 64Cu. Severe combined immunodeficient mice bearing PSMA-positive C4-2 prostate carcinoma xenografts were used for small-animal PET imaging. Mice with PSMA-negative DU 145 tumors served as controls. For PET studies, each animal received 20–30 μg of radiolabeled mAb corresponding to an activity of 7.6–11.5 MBq. Imaging was performed 3, 24, and 48 h after injection. After the last scan, the mice were sacrificed and tracer in vivo biodistribution was measured by γ-counting. Results: Binding of the mAb 3/A12 on PSMA-expressing C4-2 cells was only minimally influenced by DOTA conjugation. The labeling efficiency using 64Cu and DOTA-3/A12 was 95.3% ± 0.3%. The specific activity after 64Cu labeling was between 327 and 567 MBq/mg. After tracer injection, static small-animal PET images of mice with PSMA-positive tumors revealed a tumor-to-background ratio of 3.3 ± 1.3 at 3 h, 7.8 ± 1.4 at 24 h, and 9.6 ± 2.7 at 48 h. In contrast, no significant tracer uptake occurred in the PSMA-negative DU 145 tumors. These results were confirmed by direct counting of tissues after the final imaging. Conclusion: Because of the high and specific uptake of 64Cu-labeled mAb 3/A12 in PSMA-positive tumors, this ligand represents an excellent candidate for prostate cancer imaging and potentially for radioimmunotherapy.

Central serotonin transporter levels are associated with stress hormone response and anxiety.

RationaleNegative mood states are characterized by both stress hormone dysregulation and serotonergic dysfunction, reflected by altered thalamic serotonin transporter (5-HTT) levels. However, so far, no study examined the individual association between cortisol response and cerebral in vivo 5-HTT levels in patients suffering from negative mood states.ObjectiveThe objective of this cross-sectional study was to assess the interrelation of cortisol response, thalamic 5-HTT levels, and anxiety in healthy subjects and two previously published samples of patients with unipolar major depression (UMD) and obsessive–compulsive disorder (OCD), controlling for age, gender, 5-HTT genotype, smoking, and seasonality.MethodsRegional 5-HTT levels and cortisol response to dexamethasone-corticotropin (Dex-CRH) challenge were assessed in consecutive samples of medication-free patients suffering from UMD (N = 10) and OCD (N = 10), and 20 healthy volunteers. The intervention used was combined Dex-CRH test and [11C]DASB positron emission tomography. The main outcome measures were: 5-HTT binding potential (BPND) in a predefined thalamic ROI, cortisol response defined as the maximum cortisol increase in the combined Dex-CRH-test, and state of anxiety from the state-trait-anxiety inventory.ResultsReduced thalamic 5-HTT BPND was associated with increased cortisol response (r = −0.35, p < 0.05; in patients: r = −0.53, p < 0.01) and with increased state anxiety (r = −0.46, p < 0.01), surviving correction for age, gender, 5-HTT genotype, smoking, and seasonality (p < 0.05). The 5-HTT genotype, on the contrary, was not significantly associated with cortisol response (p = 0.19) or negative mood (p = 0.23).ConclusionThe association between stress hormone response, thalamic 5-HTT levels, and anxiety in patients suffering from negative mood states suggests an interaction between two major mechanisms implicated in negative mood states in humans.

Dependency of the [18f]fluoromisonidazole uptake on oxygen delivery and tissue oxygenation in the porcine liver

We have previously shown that the accumulation of fluorine-18-labeled fluoromisonidazole ([(18)F]FMISO) is inversely correlated to tissue oxygenation, allowing the quantification of porcine liver tissue hypoxia in vivo. We determined the activity from administered [(18)F]FMISO in relation to the hepatic oxygen availability and the partial pressure of oxygen in tissue (tPO(2)) to define a critical oxygen delivery on a regional basis. [(18)F]FMISO was injected 2 h after onset of regional liver hypoxia due to arterial occlusion of branches of the hepatic artery in 10 domestic pigs. During the experimental procedure the fractional concentration of inspired oxygen (FiO(2)) was set to 0.67 in group A ( N=5) and to 0.21 in group B ( N=5) animals. Immediately before sacrifice, the tPO(2) was determined in normal flow and flow-impaired liver segments. The standardized uptake values (SUV) for [(18)F]FMISO was calculated from 659 single tissue samples obtained 3 h after injection of approximately 10 MBq/kg body weight [(18)F]FMISO and was compared with the regional total hepatic oxygen delivery (DO(2)) calculated from the regional arterial and portal venous flow (based on (141)Ce- and (99m)Tc-microspheres measurements) and the oxygen content of the arterial and portal venous blood. In 121 tPO(2)-measured liver tissue samples, the mean DO(2) was significantly decreased in occluded liver tissue samples [group A: 0.063 (0.044-0.089); group B: 0.046 (0.032-0.066)] compared to normal flow segments [group A: 0.177 (0.124-0.252); group B: 0.179 (0.128-0.25) mL x min(-1) x g(-1); geometric mean (95% confidence limits); p < 0.01 in group A and p < 0.001 in group B]. The tPO(2) of occluded segments [group A: 5.1 (3.2-8.1); group B: 3.9 (2.4-6.2) mm Hg] was significantly decreased compared to normal flow segments [group A: 20.2 (12.6-32.5); group B: 22.4 (14.3-35.2) mm Hg; p < 0.01 in group A and p < 0.001 in group B]. Three hours after [(18)F]FMISO administration, the mean [(18)F]FMISO SUV determined in tPO(2)-measured occluded segments was significantly higher [group A: 4.08 (3.12-5.34), group B: 5.43 (4.14-7.13)] compared to normal liver tissue [group A: 1.57 (1.2-2.06), group B: 1.5 (1.16-1.93); p < 0.001 for both groups]. The [(18)F]FMISO SUV allowed prediction of the tPO(2) with satisfying accuracy in hypoxic regions using the exponential regression curve [[(18)F]FMISO=1.05+6.7((-0.117 tPO(2))); r(2)=0.75; p < 0.001]. In addition, regardless of ventilation conditions, a significant exponential relationship between the DO(2) and the [(18)F]FMISO SUV was found ( r(2)=0.39, p < 0.001). Our results suggest that the reduction of the oxygen delivery below the critical range of 0.1-0.11 mL x min(-1) x g(-1) regularly causes liver tissue hypoxia. The severity of hypoxia is reflected by the [(18)F]FMISO accumulation and allows the in vivo estimation of the tPO(2) in hypoxic regions.

Intratumoral Spatial Distribution of Hypoxia and Angiogenesis Assessed by 18F-FAZA and 125I-Gluco-RGD Autoradiography

The hypoxia-inducible factor-1α (HIF-1α) activates angiogenesis in response to cellular hypoxia, suggesting a spatial correlation between angiogenesis and tissue hypoxia. Methods: Using digital autoradiography of coinjected 18F-labeled azomycin arabinoside (8F-FAZA) (assessing regional hypoxia) and a glycosylated RGD-containing peptide (125I-3-iodo-dTyr4-cyclo(-Arg-Gly-Asp-dTyr-Lys(SAA)-), or 125I-Gluco-RGD) (assessing angiogenesis via binding to αvβ3 integrin receptors on endothelial cells) performed on 22 EMT6 tumor xenografts, we investigated the intratumoral spatial distribution of these tracers. We applied a Bayesian bivariate image analysis using the mean tumor-to-muscle ratio as a discriminator, resulting in 4 groups: FAZA high/RGD high (Q1), FAZA low/RGD high (Q2), FAZA low/RGD low (Q3), and FAZA high/RGD low (Q4). In an additional 18 xenografts, the immunohistochemically derived HIF-1α protein distribution was compared with 18F-FAZA autoradiography. Animals were divided into groups breathing either room air or carbogen (95% oxygen, 5% CO2) for 4 h until sacrifice. Results: Under room air conditions, roughly 60% of the tumor surface displayed a spatial coupling of 18F-FAZA and 125I-Gluco-RGD uptake: either high (Q1) or low (Q3) uptake for both tracers, with Q1 indicating spatial association of hypoxia and angiogenesis and Q3 indicating adequate oxygenation without active angiogenesis. However, the remaining approximately 40% of the tumor surface showed discordant 18F-FAZA and 125I-Gluco-RGD uptake, indicating that hypoxia and angiogenesis are not necessarily spatially linked to each other and highlighting substantial intratumoral heterogeneity of the 18F-FAZA and 125I-Gluco-RGD uptake. Although carbogen breathing conditions significantly decreased the mean 18F-FAZA tumor-to-muscle ratio, no significant changes were observed for 125I-Gluco-RGD, indicating that an acute increase in tumor oxygenation did not influence αvβ3 integrin receptor expression. The HIF-1α–positive (HIFpos) tumor cell fraction was not significantly influenced by breathing conditions and covered between 0% and 35% of the total tumor section surface. However, the HIFpos tumor section surface was much smaller than the tumor section surface of increased 18F-FAZA uptake, suggesting that both markers are identifying distinctly different biologic processes associated with hypoxia. Conclusion: The study revealed a substantial spatial discordance of the 18F-FAZA and 125I-Gluco-RGD tumor distribution suggesting that hypoxia and angiogenesis are not necessarily spatially linked in malignancies. These results may prove essential in developing advanced targeted systemic chemotherapeutic approaches (such as combinations of hypoxia-activated cytotoxins and antiangiogenic drugs) for hypoxic tumors.

A priori identifiability of a one-compartment model with two input functions for liver blood flow measurements

An extended dual-input Kety-Schmidt model can be applied to positron emission tomography data for the quantification of local arterial (f(a)) and local portal-venous blood flow (f(p)) in the liver by freely diffusible tracers (e.g., [15O]H2O). We investigated the a priori identifiability of the three-parameter model (f(a), f(p) and distribution volume (Vd)) under ideal (noise-free) conditions. The results indicate that the full identifiability of the model depends on the form of the portal-venous input function (c(p)(t)), which is assumed to be a sum of m exponentials convolved with the arterial input function (c(a)(t)). When m>or=2, all three-model parameters are uniquely identifiable. For m=1 identifiability of f(p) fails if c(p)(t) coincides with tissue concentration (q(t)/Vd), which occurs if c(p)(t) is generated from an intestinal compartment with transit time Vd/f(a). Any portal input, f(p) c(p)(t), is balanced by the portal contribution, f(p) q(t)/Vd, to the liver efflux, leaving q(t) unchanged by f(p) and only f(a) and Vd are a priori uniquely identifiable. An extension to this condition of unidentifiability is obtained if we leave the assumption of a generating intestinal compartment system and allow for an arbitrary proportionality constant between c(p)(t) and q(t). In this case, only f(a) remains a priori uniquely identifiable. These findings provide important insights into the behaviour and identifiability of the model applied to the unique liver environment.

Imaging of Apoptosis: The Need to Distinguish Tracer Uptake Rate from Regional Contribution of Blood Flow

In oncology, PET is known to contribute substantially to localization of tumors and metastases and to control of therapy. One particularly attractive application is the assay of apoptosis induced by chemotherapy or radiation. Various tracers, such as 99mTcannexin-V and 18F-ML-10, are under discussion and in application. At Stanford University, the group of Jianghong Rao reported the unique concept of a radiotracer that reaches its final form within the apoptotic cells (1). This happens in a series of 3 reaction steps: the 18F-labeled precursor, 18F-C-SNAT, is selectively cleaved by caspase-3/7 between the D and X of the DEVD-X peptide sequence (where X 5 any amino acid and DEVD 5 Asp-Glu-Val-Asp) and the disulfide is reduced to a thiol by intracellular glutathione; the released amino and thiol groups react with the cyano group on the hydroxyquinoline at the other end of the molecule via an intramolecular reaction; and the resulting cyclized compound (cyclic18F-C-SNAT) is highly lipophilic and thus remains trapped within the apoptotic cell and apoptotic bodies.

CycloSal-dRFIB, a Thymidine Mimetic, Thymidine Kinase by-Pass Nucleoside Prodrug: Radioiodination, in vitro Cellular Uptake and Biodistribution in Murine Models

Cyclo-(3-methylsaligenyl)-5-O-[1-(2,4-difluoro-5-[125I]iodophenyl)-2-deoxy-β-D-ribofuranosyl]phosphate (cycloSal- dRF[125I]IB) was radioiodinated with sodium [125I]iodide via copper-catalyzed isotope exchange in 48% radiochemical yield. cycloSal-dRF[125I]IB was found to be incorporated into the cytoplasmic nucleic acid and mitochondrial fractions of murine KBALB and K-STK (engineered to express HSV-1 thymidine kinase) cells in cell culture. Uptake was greater than that for either the corresponding nucleoside dRF[125I]IB or [125I]IUdR. These in vitro studies support a mechanism of metabolic activation to the free nucleotide, thereby effecting TK-bypass. Pharmacokinetic studies in rats reflect a complex interplay of tissue depot effects, hepatobiliary recycling, and metabolism. Biodistribution studies in tumor- bearing mice provide further evidence for lipophilic depot effects and hepatobiliary recirculation, with no evidence for active (metabolic) accumulation in any tissue.