MSBI Thesis & Abstracts 2012-2013

Madhav Agrawal
Advisor: Dr. Youngho Seo, PhD
Thesis Title: 18f-FDG PET/CT to measure glucose metabolism of various breast cancer cell linesin vivo
Introduction: Breast cancer is prevalent health issue in women. Currently PET scanning is used as a noninvasive means to provide physiological information the uptake of glucose and its metabolism. This is performed by using the 18F-FDG tracer, which is a glucose analog. Currently, in vitro experiments are executed to understand the metabolism and glucose uptake of breast cancer cell lines in order to develop novel therapeutics. However, the in vitro glucose uptake data on breast cancer cell lines contradicts our speculation of how the trend should look, which is based off our understanding of cell line's characteristics in human models. The aim of our study is to ensure if glucose uptake in vitro correlates with in vivo glucose uptake (FDG tracer).

Methods: 7 breast cancer cell lines were bilaterally implanted with the same cell line in the mammary fat pad of immunodeficient mice (SCID and NSG). The mice were monitored for volume growth and were imaged by microPET/CT when the tumors reached about 200 mm3. The mice were injected with about 200 &mgr;Ci/0.1 ml concentrations.

Results and Discussions: Small correlation (R2 = 0.08394) was between the comparisons of glucose uptake in vitro and FDG uptake in vivo. Imaging was seen as inconsistent for two different cell lines. No correlation (R2 = 0.00666) was observed in the comparison between volume (in all instance) and their corresponding FDG uptake. Volume growth was an issue with all the animals and it is speculated that the implantation procedure is not optimal. Low correlation (R2 = 0.0352) is observed in the overall comparison between FDG uptakes and volume growth rate. A comparison was done to clarify if %ID/gmax can be used instead of SUVmax, which resulted in a strong correlation (as expected). Further studies are needed due to small sample size.

Conclusion: At the end of the study, it was observed that glucose uptake in vitro does not correlate with FDG uptake in vivo. This has a strong implication that in vitro studies for glucose uptake might not be translated to in vivo uptake.

Naeim Bahrami
Advisor: Dr. Peder Larson, PhD
Thesis Title: Modeling the 13-C Hyperpolarized Pyruvate and Urea Concentration Kinetics With Multiband RF Excitation MRI in Prostate Cancer
The accurate detection and characterization of cancerous tissue is still a major problem for the clinical management of individual prostate cancer patients and for monitoring their response to therapy. rho1 (TR times to times points per second over T1 ratio) of urea, pyruvate, lactate, and alanine, also the amount urea and pyruvate perfusion, and conversion constant between pyruvate to lactate(Kpl) and pyruvate to alanine(Kpa) are important parameters in different organs including cancerous and healthy tissues. rho 1 of urea in kidneys, prostate cancerous tissues, and liver are measured 0.13(1/s), 0.15(1/s), and 0.075(1/s), respectively and rho 1 of pyruvate in kidneys, liver cancer and healthy part of liver is 0.08(1/s), 0.13(1/s), and 0.064(1/s), respectively with TR=250ms. Kpl in cancerous tissues are more than 0.44(1/s) which is significantly higher than Kpl of metabolites in healthy tissues (Kpl = 0.028(1/s)) with p value less than 0.001. This Kpl is proportional to the lactate signal to pyruvate signal ratio with Correlation Coefficient=0.95. High perfusion amount of the accumulation of pyruvate, lactate, and alanine in compare to urea perfusion has been seen in cancerous tissues (liver cancer and prostate cancer) significantly (p<0.001) less than in healthy tissues.

Rupinder Chandi
Advisor: Dr. Youngho Seo, PhD
Thesis Title: In vivoImaging Assessment of Radionucliotide Therapy of Neuroblastoma Using Small Molecule PET Prove Targeting hNET
Purpose: Neuroblastoma is an embryonic tumor of the peripheral sympathetic nervous system. Most of neuroblastoma tumors express norepinephrine transporter (NET), which makes metaiodobenzylguanidine (mIBG), an analogue of norepinephrine, an ideal tumor specific agent for therapy, when labeled with I-131. Although drug therapies of neuroblastoma are performed regularly, there is lack of accurate quantitative assessment of tumor response to the therapy. This study was carried out with the objectives: the development of a novel 18F-labeled small molecule-imaging agent that targets the human NET (hNET) receptor, the establishment of a quantitative methodology to assess the therapeutic effect of [131I]mIBG and the establishment of murine xenograft model with a human neuroblastoma cell line that overexpresses hNET for better mIBG uptake would fill an unmet need to further develop treatment for this disease.

Methods: In order to enhance mIBG uptake for therapy, luciferase-expressing and hNET-transduced NB1691 (NB1691-luc-hNET) cells were implanted subcutaneously and in the renal capsule of murine xenograft models. Once the tumors reached a defined volume, the mice were injected with the [18F]RP-109, a NET imaging agent and scanned on the microPET/CT instrument. The xenografts were also imaged using bioluminescence to assess the viability of the cells in vivo. The mice were then treated with [[131I]mIBG using an established protocol. Three weeks after the [[131I]mIBG treatment, they were re-imaged on the microPET/CT using [18F]RP-109 as well as by bioluminescence. The PET images were then correlated with the bioluminescence scans as well as tumor volume measurements to assess the utility of the probe for monitoring therapy response of the tumors.

Results: The novel probe, [18F]RP-109, was successfully prepared and evaluated in neuroblastoma, but no tumor uptake was observed. The treatment of neuroblastoma with [131I]mIBG therapy showed a better response with higher dose of [[131I]mIBG. The therapeutic data obtained from the bioluminescence imaging helps to assess the effectiveness of [131I]mIBG treatment of neuroblastoma model and evaluate the functional status of tumors.

Conclusion: The NET probe was successfully radiolabelled with fluorine-18 and it did not show any visible uptake in the tumors that overexpress hNET. The [[131I]MIBG therapy study in mouse models using bioluminescence and tumor volume measurement as follow-up showed the effective therapy at 2 mCi level, which was consistent with the estimated tumor dosimetry study performed previously. 

Dharshan Chandramoham
Advisor: Dr. Xiaojuan Li, PhD
Thesis Title: Advanced Quantitative MR Imaging for Detecting Early Degeneration of Cartilage
Magnetic Resonance Imaging (MRI) of cartilage offers unique insights into biochemical and microstructural processes of degeneration that might be useful in characterizing the early stages of Osteoarthritis. Quantitative MR imaging sequences enable us to map magnetic relaxation parameters of the tissue, including T1ρ and T2 , which are highly sensitive to microscopic changes in cartilage. Different imaging strategies and mathematical models have been used to quantify T1ρ. In this study we compare two strategies for T1ρ imaging: the SPGR-based MAPSS sequence, and the FSE-based CUBE sequence. The sequences were evaluated using simulation, phantom scans, and in vivo human scans. The CUBE sequence appears to outperform the MAPSS sequence in terms of SNR efficiency and repeatability on the basis of phantom and human studies. The CUBE sequence was then used to evaluate the feasibility of performing multi-exponential T1ρ quantification in a 63 TSL scan of a porcine knee specimen. The multi-exponential T1ρ procedure was able to identify two spin populations, even when the number of TSLs used to fit the model was reduced to 16. This data suggests the feasibility of performing multi-component T1ρ imaging in vivo, especially if combined with advanced accelerating techniques in the future. The results from this study need to be confirmed by larger scale studies.

YiShin Chang
Advisor: Dr. Pratik Mukherjee, PhD
Thesis Title: Differing Patterns of Brain Connectivity in Autism Spectrum Disorder vs Sensory Processing Disorder
Sensory Processing Disorders (SPD) affect 5-16% of children in the US, and have been shown to impair intellectual and social development due to disrupted processing and integration of sensory input. SPD exists both comorbidly with other neurpathologies such as autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD), but often also exists in isolation. In our previous work, tract-based spatial statistics (TBSS) were used with diffusion tensor imaging (DTI) data of SPD subjects to demonstrate decreased white matter integrity in parieto-occipital white matter tracts.

Autism Spectrum Disorders (ASD) are typically clinically characterized by impaired communication, social interaction, and behavioral flexibility. However, the majority of individuals with ASD also exhibit sensory processing differences. It is valuable to characterize the similarities and differences in sensory processing deficits between sensory processing deficits within SPD and ASD for differential diagnosis and potentially for treatment planning.

We validate and expand upon previous white matter studies of ASD and our previous results in SPD by taking a tractographical approach to examining white matter tracts that we expect to be commonly and differentially affected in SPD vs ASD. Diffusion tensor imaging data was acquired in 16 boys with SPD, 15 boys with ASD, and 23 typically developing boys. Probabilistic fiber tractography was then used with this data to delineate white matter tracts of interest, and white matter structural integrity was assessed for each of these tracts using average values of fractional anisotropy, mean diffusivity, radial diffusivity, and axial diffusivity within each tract. Significant decreases in FA were found for both the SPD and ASD cohorts within the parieto-occipital tracts as compared with the control cohort, with the SPD cohort demonstrating surprisingly and pervasively more affected FAs. Significant decreases in FA were found for the ASD cohort, but not the SPD cohort, within the temporal white matter tracts.

These findings validate abnormal white matter as a biological basis for SPD, demonstrate differential white matter pathology between ASD and SPD, and suggest that tracts contributing to social and emotional processing (and implicated in ASD) are independent of sensory processing.

Lindsay Conner
Advisor: Dr. Norbert Schuff, PhD
Thesis Title: The Association of Amyloid Count With a Decrease in the Complexity of the BOLD Signal Observed in Alzheimer's Disease
Alzheimer's disease (AD) is a prevalent degenerative disease that is both idiopathic and highly debilitating, calling for early detection and intervention before extensive irreversible brain damage occurs. Here, a novel functional complexity measure based on information theory was applied to mark a disease-related reduction in stochastic low frequency (0.01-0.08 Hz) BOLD signal fluctuations at rest. As AD progresses, BOLD signal complexity is expected to decrease in affected regions, alongside an overall increase in brain amyloid burden.

In this study, a retrospective analysis was performed on 18F-AV-45 PET and rs-fMRI data from 65 subjects (30 males, 35 females; mean age ± SD: 74 ± 7.4 years), across four clinical groups (Control, Early Mild Cognitive Impairment (EMCI), MCI, and AD). The regional BOLD signal complexity measures, or transient information (TI) values, were determined from the change in uncertainty of BOLD pattern prediction over time (i.e. block entropy growth rate).

The main findings in this study included an expected regional reduction in functional complexity with increasing amyloid burden, as well as an unexpected, albeit non-significant, global increase in functional complexity with increased amyloid burden and disease progression. One third (27 out of 82) of the cortical and subcortical grey matter regions presented a significant (>90% CI) effect of brain amyloid load on BOLD signal complexity, including regions associated with disease-related dysfunction in memory, language processing, attention, behavior, somatosensory functions, and motor functions. With disease diagnosis taken into account, only the EMCI group indicated a decrease in global BOLD signal complexity with increased brain amyloid load.

Analysis of rs-BOLD signal complexity has the potential to provide a more accurate representation of disease state than current amyloid plaque or structural measures, as well as identify the regions altered by disease pathology. Based on the reduced signal complexity reported in regions previously linked to disease pathology (precuneus/posterior cingulate, lateral temporal lobe, and frontal regions), further study of this novel metric is advised. With validation, the BOLD complexity analysis metric can be attuned to use as a cognitive biomarker in the clinical setting, potentially improving disease diagnosis, treatment monitoring, and evaluation of future treatment options.

Charvi Shetty
Advisor: Dr. Alastair Martin, PhD
Thesis Title:Developing Methods to Aid Edge Detection in a Micro-Computed Tomography Based Subcutaneous Versus Visceral Fat Segmentation Algorithm
Micro-computed tomography can be used to provide a precisein-vivoassessment of adipose tissue quantity and distribution, including information on subcutaneous and visceral fat volume in mouse models. This study aims to develop methods to aid edge detection in order to eventually segment out the visceral and subcutaneous fat compartments automatically. The algorithm detailed in this paper optimizes steps in the Canny edge detection method and utilizes low-pass filtering and gradient edge detection. Ten mice (weight range: 19.96--57.66 g) were tested with micro-CT scans to verify the utility of this algorithm. The algorithm demonstrated stability despite the broad range of body weights and adiposity. Comparisons of the data between unfiltered versus filtered mice volumes suggest that this algorithm can be used to effectively increase edge strength for use in separating visceral and subcutaneous fat compartments. The eventual application of this method would be to assess metabolic disease risk, such as those associated with central obesity including diabetes, hypertension, and heart disease.

Caleb Tam
Advisor: Dr. Sunita Ho, PhD
Thesis Title:Quantitative Comparison of Bone Metabolism Using Fluorescence and X-Ray Microscopy Techniques
Introduction: X-ray computed tomography has been proposed as a non-invasive, high-throughput, 3-D alternative to conventional bone histology, but a direct comparison between these two modalities has not yet been established. In this study, we specifically assess the ability of μCT (micro-computed tomography) to produce metrics of osseous tissue formation using dynamic histomorphometry as a standard for comparison.

Methods: Rat dento-alveolar complexes were implemented as preclinical models to examine formation events in bone, dentin, and cementum. A total of 8 specimens were processed (4 control, 4 ligated), with alternating tetracycline hydrochloride/alizarin red fluorochromes administered over 11 weeks. On week 12, rats were sacrificed, with hemi-maxilla dissected, embedded in Stycast®® epoxy, sectioned/ground to ~100μm, and imaged via optical and μCT microscopy. Fluorescence images were collected in 3 channels: brightfield, mCherry, Sapphire, at 20× magnification, while μCT radiographs and tomograms were collected at 90/30 kVp, 20/40×, and single tile/mosaic sizes. Images were taken by tooth-location-tissue, then analyzed via FIJI-ImageJ (line, plot profiles) and Microsoft Excel (linear regressions). Final correlates for dynamic histomorphometry and μCT were mineral apposition rate (MAR) defined as inter-label distance (μm)/time between injections (weeks), and mineral density rate (MDR) defined as change in image gray values (I.a.u.)/distance (μm), respectively.

Results: Cementum had the highest MAR (113 ± 16 μm/week), but lowest MDR (0.03 ± 0.01 I.a.u./μm), while dentin had lowest MAR (34 ± 5 μm/week), but highest MDR (0.09 ± 0.02 I.a.u./μm), with bone being intermediate for both (MAR = 86 ± 20 μm/week, MDR = 0.04 +,m; 0.01 I.a.u./μm). MAR vs. MDR revealed the tissue that extended fastest in time simultaneously mineralized the least in space, and vice versa, 1/MAR vs. MDR showed growth and mineralization are positively correlated when normalized to distance, and MAR×MDR calculations produced constant mineral density rates in time (MDRT), suggesting mineralization is a centralized process.

Conclusion: Our study finds μCT and dynamic histomorphometry provide complementary information regarding bone turnover activity, and that tissue apposition dictates mineral distribution, not mineralization rate. Continued research is needed to ascertain whether this relationship holds true for other regions within the skeletal system.

Shuyu Tang
Advisor: Dr. Peder Larson, PhD
Thesis Title: Development of a Novel 2D RF Pulse Sequence to Achieve An Improved Localization of Hyperpolarized 13-C Imaging
Hyperpolarization of metabolically active compounds labeled with 13C has been recently utilized for imaging metabolic processes in vivo. This work focused on developing a new 2D radiofrequency (RF) pulse for tracking a 13C-labeled pyruvate bolus injection in vivo. This pulse was designed for a clinical scanner, which allows for rapid translation to human studies. To meet clinical needs, we examined the flexibility of our design, and specifically the ability to shift the pulse location. The off-resonance response due to inhomogeneity, [1-13C]alanine and [1-13C]lactate have also been examined. Experimental results have validated the profile of our designed 2D RF pulse and off-resonance signals. To prepare for in vivo studies, our proposed pulse has been incorporated into a hyperpolarized 13C imaging sequence.

Alyssa Zhu
Advisor: Dr. Roland Henry, PhD
Thesis Title: Optimization of Brain Segmentation in Multiple Sclerosis Patients
Multiple sclerosis is an idiopathic, autoimmune disease that affects the central nervous system (CNS). Imaging studies have shown that gray matter volume, rather than whole brain or white matter volume, acts as the best imaging biomarker for MS progression. Previous studies were performed via cross-sectional analysis of each time point and then interrogating the difference between values. Because of variability inherent in software tools, the population of cross-sectional analysis studies is dependent on the segmentation program being utilized with smaller standard deviations allowing for smaller subject populations, particularly when the tissue volume difference being studied is small in comparison those standard deviations. Longitudinal analysis aims to minimize that variability and give more accurate segmentation results. Segmentation in MS is also plagued by the presence of white matter lesions, whose T1 hypointensities can result in the tissue being misclassified as gray matter. Two longitudinal programs that have been validated for healthy controls and patients with Alzheimer's disease---aBEAT and FreeSurfer---were explored by retrospectively analyzing 7 sets of longitudinal data both cross-sectionally and longitudinally. A comparison between programs revealed that FreeSurfer produced more accurate both segmentation and anatomical parcellation results. Quantitative analysis of gray matter volumes also showed FreeSurfer to be superior to aBEAT with FreeSurfer's cross-sectional processing yielding the smoothest transition from time point to time point. The investigation into cortical thicknesses obtained by FreeSurfer, on the other hand, yielded slightly conflicting results between R2 values and observed longitudinal trends. Further analysis of both longitudinal processing and lesion segmentation is required to evaluate the usefulness of currently available longitudinal processing programs and to avoid the need for manual segmentation respectively. Currently, cross-sectional segmentation is the optimal method for longitudinal brain volume analysis as longitudinal segmentation programs have proven inferior rather than superior to their cross-sectional counterparts.