Quantitative analysis of lesion morphology and texture features for diagnostic prediction in breast MRI

RATIONALE AND OBJECTIVES

To investigate the feasibility using quantitative morphology/texture features of breast lesions for diagnostic prediction, and to explore the association of computerized features with lesion phenotype appearance on magnetic resonance imaging.

MATERIALS AND METHODS

Forty-three malignant/28 benign lesions were used in this study. A systematic approach from automated lesion segmentation, quantitative feature extraction, diagnostic feature selection using an artificial neural network (ANN), and lesion classification was carried out. Eight morphologic parameters and 10 gray level co-occurrence matrix texture features were obtained from each lesion. The diagnostic performance of selected features to differentiate between malignant and benign lesions was analyzed using receiver-operating characteristic analysis.

RESULTS

Six features were selected by an ANN using leave-one-out cross validation, including compactness, normalized radial length entropy, volume, gray level entropy, gray level sum average, and homogeneity. The area under the receiver-operating characteristic curve was 0.86. When dividing the database into half training and half validation set, a classifier of five features selected in the half training set achieved an area under the curve of 0.82 in the other half validation set. The selected morphology feature "compactness" was associated with shape and margin in the Breast Imaging Reporting and Data System lexicon, round shape and smooth margin for the benign lesions, and more irregular shape for the malignant lesions. The selected texture features were associated with homogeneous/heterogeneous patterns and the enhancement intensity. The malignant lesions had higher intensity and broader distribution on the enhancement histogram (more heterogeneous) compared to the benign lesions.

CONCLUSION

Quantitative analysis of morphology/texture features of breast lesions was feasible, and these features could be selected by an ANN to form a classifier for differential diagnosis. Establishing the link between computer-based features and visual descriptors defined in the BI-RADS lexicon will provide the foundation for the acceptance of quantitative diagnostic features in the development of computer-aided diagnosis.

Estrogen receptor and breast MR imaging features: a correlation study

PURPOSE

To compare the MRI features between estrogen receptor (ER) positive and negative breast cancers.

MATERIALS AND METHODS

Breast MRI of 90 consecutive patients confirmed with invasive ductal carcinoma (IDC), 51 ER positive and 39 ER negative, were analyzed. The tumor morphology and dynamic contrast-enhanced (DCE) kinetics were evaluated based on the American College of Radiology (ACR) Breast Imaging Reporting and Data System (BI-RADS) MRI lexicon and compared. Enlarged axillary lymph nodes on MRI and choline (Cho) detection using MR spectroscopy (MRS) were also analyzed and compared. For patients receiving axillary node dissection the pathological nodal status was also compared.

RESULTS

ER negative breast cancer had bigger tumors compared to ER positive cancer (3.6 +/- 2.0 cm vs. 1.8 +/- 1.3 cm, P < 0.00005). ER negative cancer was more likely to exhibit nonmass type enhancements compared to ER positive cancer (P < 0.005). Enlarged axillary lymph nodes were more frequently identified on MRI in ER negative compared to ER positive patients (P < 0.05). After excluding patients undergoing neoadjuvant chemotherapy, auxiliary lymph node status did not show significant difference between ER positive and ER negative cancer on MRI and pathology. ER negative cancer was more likely to show the malignant type enhancement kinetics (P = 0.15), rim enhancement (P = 0.15), and Cho detection on MRS (P = 0.23) compared to ER positive cancer, but it did not reach a level of statistical significance.

CONCLUSION

ER negative breast cancer was more aggressive, with larger tumor size, more non-mass-type enhancement lesions, and a higher percentage showing enlarged axillary nodes on MRI. These features might be related to its poorer cellular differentiation and/or a higher angiogenesis.

Quantitative correlation between (1)H MRS and dynamic contrast-enhanced MRI of human breast cancer

Proton magnetic resonance spectroscopy ((1)H MRS) and dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) provide functional information, including vascular volume, vascular permeability and choline (Cho) metabolism. In this study, we applied these two imaging modalities to quantitatively characterize 36 malignant breast lesions in 32 patients and analyzed the correlation between them. Cho concentration was quantified by single-voxel (1)H MRS using water as an internal reference. The measured Cho levels ranged from 0.32 to 10.47 mmol/kg, consistent with previously reported values. In 25 mass-type lesions, the Cho concentration was significantly correlated with tumor size (r=.69, P<.0002). In addition, the Cho level was found to be significantly higher in lesions presenting as mass-type lesions compared to non-mass-type diffuse enhancements (P=.035). The enhancement kinetics from tissues covered within each MRS voxel were measured and analyzed with a two-compartmental model to obtain pharmacokinetic parameters K(trans) and k(ep). A significant correlation was found between the Cho level and the pharmacokinetic parameter k(ep) (r=.62, P<.0001), indicating that tissues with a high Cho level have higher wash-out rates in DCE MRI. The results suggest a correlation between Cho metabolism and angiogenesis activity, which might be explained by the association of Cho with cell replication and angiogenesis required to support tumor growth.

Detection of choline signal in human breast lesions with chemical-shift imaging

PURPOSE

To investigate the application of MR spectroscopy using chemical-shift imaging (CSI) for characterizing human breast lesions at 1.5T, and to evaluate the diagnostic performance using ROC (receiver operating characteristics) analysis.

MATERIALS AND METHODS

Thirty-six patients (35-73 years old, mean 52), with 27 malignant and 9 benign lesions, underwent anatomical imaging, dynamic contrast-enhanced MR imaging, and CSI. The ROC analysis was performed and the cutoff point yielding the highest accuracy was found to be a choline (Cho) signal-to-noise ratio (SNR) >3.2.

RESULTS

The mean Cho SNR was 2.8 +/- 0.8 (range, 1.8-4.3) for the benign group and 5.9 +/- 3.4 (2.1-17.5) for the malignant group (P = 0.01). Based on the criterion of Cho SNR >3.2 as malignant, CSI correctly diagnosed 22 of 27 malignant lesions and 7 of 9 benign lesions, resulting in a sensitivity of 81%, specificity of 78%, and overall accuracy of 81%. If the criterion was set higher at Cho SNR >4.0 the specificity improved to 89% but sensitivity was lowered to 67%.

CONCLUSION

The ROC analysis presented in this work could be used to set an objective diagnostic criterion depending on preferred emphasis on sensitivity or specificity.

Magnetic resonance imaging features of fibrocystic change of the breast

PURPOSE

Studies specifically reporting MRI of fibrocystic change (FCC) of the breast are very few and its MRI features are not clearly known. The purpose of this study was to analyze the MRI features of FCC of the breast.

MATERIALS AND METHODS

Thirty-one patients with pathologically proven FCC of the breast were retrospectively reviewed. The MRI study was performed using a 1.5-T MR scanner with standard bilateral breast coil. The imaging protocol consisted of pre-contrast T1-weighed imaging and dynamic contrast-enhanced axial T1-weighed imaging. The MRI features were interpreted based on the morphologic and enhancement kinetic descriptors defined on ACR BIRADS-MRI lexicon.

RESULTS

FCC of the breast had a wide spectrum of morphologic and kinetic features on MRI. Two types of FCC were found, including a more diffuse type of nonmass lesion (12/31, 39%) showing benign enhancement kinetic pattern with medium wash-in in early phase (9/10, 90%) and a focal mass-type lesion (11/31, 35%) with enhancement kinetic usually showing rapid up-slope mimicking a breast cancer (8/11, 73%).

CONCLUSION

MRI is able to elaborate the diverse imaging features of FCC of the breast. Our result showed that FCC presenting as a focal mass-type lesion was usually overdiagnosed as malignancy. Understanding MRI of FCC is important to determine which cohort of patients should be followed up alone or receive aggressive management.

MRI measurements of tumor size and pharmacokinetic parameters as early predictors of response in breast cancer patients undergoing neoadjuvant anthracycline chemotherapy

PURPOSE

To investigate the value of using changes in three parameters (tumor size, transfer constant (K(trans)), and rate constant (k(ep))) obtained after the first treatment-cycle in predicting the final clinical response after two to four cycles of neoadjuvant anthracycline and cyclophosphamide (AC) chemotherapy.

MATERIALS AND METHODS

Early changes in the three parameters were measured in 29 patients with invasive breast cancer by MRI after one cycle of treatment. Changes were then assessed for their predictive value of final clinical response and compared among patients with four different response patterns, Group 1 = responder (R) after one cycle and also R after four cycles, Group 2 = nonresponder (NR) after one cycle, but eventual R after four cycles, Group 3 = NR after one cycle and still NR after four cycles, and Group 4 = NR after one cycle and determined as NR after two cycles, being switched to the taxane regimen.

RESULTS

Pearson's correlation analysis revealed significant correlation between early changes in tumor size and both pharmacokinetic parameters (r = 0.49 and P < 0.01 for K(trans), r = 0.66 and P < 0.001 for k(ep)). The areas under the receiver operating characteristic (ROC) curve differentiating between R (Groups 1+2) and NR (Groups 3+4) groups using changes in tumor size, K(trans), and k(ep) were 0.88 (standard error [SE] = 0.06, P < 0.0001), 0.63 (SE = 0.11, P = 0.11), and 0.77 (SE = 0.09, P = 0.001), respectively.

CONCLUSION

Early tumor size change in MRI after one cycle is better response predictor than that of either K(trans) or k(ep) in breast cancer undergoing neoadjuvant chemotherapy using an AC regimen.

The Predictive Value of Sentinel Lymph Node Biopsy in Locally Advanced Breast Cancer Patients who Have Undergone Neoadjuvant Chemotherapy

With the increasing usage of neoadjuvant chemotherapy (NAC) in locally advanced breast cancer (LABC), there is the need to investigate the routine axillary node dissections performed in this group of patients. Controversy exists about the utility of sentinel node biopsy (SNB), either before or after NAC. With the addition of trastuzumab in the treatment of Her2/neu-positive LABC patients, the validity of SNB in this subset population needs to be investigated. A retrospective study of 20 patients who underwent NAC for LABC was undertaken. The pathology of the axillary nodes, sentinel nodes, and primary tumor after neoadjuvant chemotherapy were examined. Twenty patients underwent NAC with doxorubicin and cyclophosphamide, followed sequentially by paclitaxel and carboplatin, with or without trastuzumab based on Her2/neu status. Post chemotherapy, 20 patients underwent mastectomy or lumpectomy with SNB with axillary node dissections. The overall accuracy of SNB was 95 per cent with a false-negative rate of 14 per cent (1/7). In Her2/neu-positive patients, overall accuracy was 100 per cent (8/8) and a false-negative rate of zero per cent. Sentinel node biopsy is a viable option in patients who have undergone NAC. Her2/neu-positive patients who had undergone NAC with trastuzumab had comparable accuracy for sentinel node biopsy in predicting axillary node status.

Predicting nodal status using dynamic contrast-enhanced magnetic resonance imaging in patients with locally advanced breast cancer undergoing neoadjuvant chemotherapy with and without sequential trastuzumab

HYPOTHESIS

Dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) is a reliable and accurate method for monitoring primary tumor response in the breast and can be used as a surrogate to predict final axillary nodal status.

DESIGN

Retrospective study (October 1, 2004, through February 28, 2006) of 46 patients with clinically staged locally advanced breast cancer.

SETTING

Comprehensive cancer center.

PATIENTS

Forty-six patients with locally advanced breast cancer.

INTERVENTIONS

Neoadjuvant chemotherapy (NAC), DCE-MRI, mastectomy and lumpectomy, and axillary lymph node dissection.

MAIN OUTCOME MEASURES

The DCE-MRI results and pathologic response of the breast and axillary lymph nodes.

RESULTS

Forty-six patients underwent NAC with doxorubicin hydrochloride and cyclophosphamide, followed by paclitaxel and carboplatin, with or without trastuzumab based on human epidermal growth factor receptor 2 (HER2/neu) status. Twenty-one patients (46%) had a complete pathologic response. For the HER2/neu-positive patients, the complete pathologic response rate was 70% (14/20). The accuracy, sensitivity, and specificity of the primary tumor response in predicting the axillary nodal status were 78%, 88%, and 72%, respectively. The accuracy, sensitivity, and specificity of the DCE-MRI-measured response in the primary tumor in predicting axillary nodal status were 74%, 62%, and 82%, respectively. For the HER2/neu-positive patients, the accuracy, sensitivity, and specificity improved to 80%, 75%, and 82%, respectively.

CONCLUSIONS

The results of DCE-MRI of the primary tumor can be predictive of axillary nodal status, especially in patients receiving trastuzumab who are HER2/neu positive. The HER2/neu-positive patients with a complete clinical response on DCE-MRI are highly unlikely to benefit from an axillary lymph node dissection. For HER2/neu-negative patients, sentinel lymph node sampling is warranted.

Behavioral, histological, and ex vivo magnetic resonance imaging assessment of graded contusion spinal cord injury in mice

This study characterized the Infinite Horizon (IH) Impactor for use in mouse models of contusion spinal cord injury (SCI), and investigated the feasibility and reliability of using magnetic resonance imaging (MRI) as a method to accurately measure lesion volume after mouse contusion SCI. Eight-week-old female C57Bl/6 mice received a mild (30 kilodyne), moderate (50 kilodyne), or severe (70 kilodyne) contusion injury at the T9 vertebral level. Uninjured control mice received a T9 laminectomy only. Functional recovery was assessed using the Basso, Beattie, Bresnahan (BBB) and Basso Mouse Scale (BMS) open-field locomotor rating scales. Next, 4% paraformaldehyde-perfused spinal cords were collected between the T6 and T12 spinal roots, and stored in phosphate-buffered saline (PBS) at 4 degrees C until MRI analysis. MRI lesion volumes were determined using T1-weighted images on a 7-Tesla MRI. Histology was performed on 20-microm polyester wax-embedded sections processed from the same spinal cords for stereological determination of fibronectin lesion volume and myelin basic protein spared white matter volume. Area of spared white matter at the epicenter was also analyzed. The results demonstrated that the IH Impactor produced precise, graded contusion SCI in mice. Lesion volumes were positively correlated with force of impact, and negatively correlated with spared white matter and functional recovery. Additionally, similar lesion volumes were detected using fibronectin staining and MRI analysis, although MRI may be more sensitive for milder injuries. These results give researchers more options in how to analyze spinal cord injuries in animal models.

Vascular volume and blood-brain barrier permeability measured by dynamic contrast enhanced MRI in hippocampus and cerebellum of patients with MCI and normal controls

PURPOSE

To measure the cerebrovascular volume and blood-brain barrier (BBB) permeability indices in hippocampus and cerebellum of patients with mild cognitive impairment (MCI) using dynamic contrast-enhanced MRI (DCE-MRI), and compare to that of normal controls.

MATERIALS AND METHODS

A total of 11 MCI subjects and 11 healthy elderly controls participated in this prospective study. DCE-MRI was performed to measure the contrast enhancement kinetics. The early enhancement percentage (at 50 seconds after injection) was defined as the vascular volume index, and the ratio between the four to five-minute enhancement relative to the 50-second enhancement was defined as the BBB permeability index.

RESULTS

The enhancement kinetics measured from hippocampus of MCI individuals demonstrated a lower magnitude and slower decay than healthy controls, suggesting that they had a smaller vascular volume (significant in the right side; P <0.001) and a higher BBB permeability (not reaching significance level). The vascular volume index was significantly correlated with naming ability (P 0.05).

CONCLUSION

These results suggest that changes in cerebrovasculature may occur in hippocampus of MCI. DCE-MRI may provide a noninvasive means to measure the subtle BBB leakage associated with the cerebrovascular pathology commonly found in Alzheimer's disease.

Regional quantification of white matter hyperintensity in normal aging, mild cognitive impairment, and Alzheimer's disease

BACKGROUND/AIMS

A quantitative method was applied to measure the volume of white matter hyperintensity (WMH) in different brain regions of subjects with Alzheimer's disease (AD), mild cognitive impairment (MCI) and normal healthy age-matched controls, and the relationship between regional WMH and age and cognitive function was investigated.

METHODS

Fifty-six subjects were included in this study, 27 AD, 15 MCI and 14 normal age-matched controls. A user-friendly software was developed for WMH quantification in frontal, temporal, and parieto-occipital lobes. Mini-Mental State Examination and cognitive scores in performing naming, language fluency, and memory tasks were obtained for correlation analysis.

RESULTS

AD patients had the greatest total WMH volume, followed by MCI, then controls. However, there was a large variation within each group, and the difference did not reach a significant level. There was a positive linear correlation between the total WMH (p = 0.031) and the frontal WMH (p = 0.006) vs. age. After age correction the Boston Naming Test scores were negatively correlated with the total WMH volume in the AD (p = 0.03) and the control (p = 0.03) groups, and with the frontal WMH in controls (p = 0.01).

CONCLUSION

We demonstrated a quantitative analysis method to measure regional WMH. Although WMH was not strongly associated with disease severity or cognition, it may provide a characteristic neuroimaging parameter in the study of AD development.

Regional pattern of increased water diffusivity in hippocampus and corpus callosum in mild cognitive impairment

BACKGROUND

Corpus callosum atrophy possibly indicates neuronal degeneration in association cortex in Alzheimer's disease (AD). Water diffusivity allows detecting physiological changes that probably occur earlier than structural shrinkage in cerebral regions during the development of AD.

OBJECTIVE

To simultaneously investigate the focal volumetric and early structural changes with apparent diffusion coefficient (ADC) of corpus callosum in mild cognitive impairment (MCI), and to explore the regional pattern of ADC increase in hippocampus and corpus callosum.

METHODS

Thirteen subjects with MCI and 13 healthy age-matched control subjects were studied. With three-dimensional volumetric spoiled gradient recalled echo MRI images, we measured the volume of hippocampus and corpus callosum. Using MRI-guided diffusion-weighted imaging analysis, we calculated ADCs from hippocampus and corpus callosum. Within the receiver-operation curve (ROC) statistical framework, areas under ROC curves from hippocampal and callosal ADCs were compared in differentiating between MCI and controls.

RESULTS

After normalization to intracranial volume, hippocampal, not callosal, volume was significantly reduced in MCI. ADC values were increased in hippocampus and corpus callosum in MCI. The extent of ADC increase was similar between hippocampus and corpus callosum.

CONCLUSION

Alterations in water diffusivity may precede corpus callosum atrophy during the development of MCI. Diffusion changes might occur simultaneously in allocortex and neocortex in MCI.

Mild Cognitive Impairment: Apparent Diffusion Coefficient in Regional Gray Matter and White Matter Structures

PURPOSE

To prospectively evaluate regional alterations in the apparent diffusion coefficient (ADC) of cortical gray and white matter and subcortical structures that are known to be involved in mild cognitive impairment (MCI).

MATERIALS AND METHODS

Magnetic resonance (MR) imaging was performed in 13 patients with MCI (nine men, four women; mean age, 74 years +/- 6 [standard deviation]) and 13 healthy elderly control subjects (seven men, six women; mean age, 75 years +/- 4). This study was approved by the institutional review board and was HIPAA compliant. Each subject gave informed consent. ADC was measured from manually drawn regions of interest (ROIs) of the hippocampus, parahippocampal gyrus, amygdala, corpus callosum, and anterior and posterior cingulate gyrus and from automatically defined frontal, parietal, occipital, and temporal lobes by using template masking. ROIs were outlined on anatomic images then mapped onto ADC maps by using coregistration transformation matrix. A skeleton-based region competition segmentation algorithm was used for segmentation of gray and white matter. The group difference in ADC values was assessed with independent-sample t tests. Pearson correlation analysis was used to examine the correlation of ADC values with age and memory test scores.

RESULTS

Higher ADCs were found in hippocampus, temporal lobe gray matter, and corpus callosum of patients with MCI compared with that of control subjects (P < .05). By pooling all subjects together, an elevated hippocampal ADC was significantly correlated with worse memory performance scores in 5-minute and 30-minute delayed word-list recall tasks (P < .05).

CONCLUSION

ADCs from gray and white matter of different brain regions can be analyzed by applying an automated template-masking method in conjunction with a skeleton-based region competition segmentation algorithm.

Comparison of choline and pharmacokinetic parameters in breast cancer measured by MR spectroscopic imaging and dynamic contrast enhanced MRI

Although an MRI scanner is a single stand-alone modality, different acquisition techniques may be applied to collect structural and functional information, including vascular or angiogenic properties measured by dynamic contrast enhanced MRI (DCE-MRI) and Choline (Cho) metabolism measured by proton MR spectroscopy imaging (MRSI). They may provide complementary information for a better characterization of neoplasm. In this study, we investigated the correlation between Choline measured by MRSI and vascular parameters measured by DCE-MRI in breast cancer. Fourteen patients with histologically proven invasive breast cancer were included. MRSI from a grid of 8 x 8 voxels within a selected slab from each lesion was performed. Each voxel was 1.0 x 1.0 x 1.2 cm3. Choline signal-to-noise ratio (SNR) was measured from each voxel showing an identifiable Choline peak. Corresponding DCE kinetics was measured from each voxel, and analyzed with a 2-compartmental model to obtain pharmacokinetic parameters Ktrans and k(ep). All parameters showed a wide variation within each lesion, and there were no consistent correlations between regional Cho and DCE parameters within the lesion of each individual patient. This finding might be attributed to the heterogeneous nature of breast cancer. The characteristic Cho and DCE-MRI parameters were obtained for each patient by averaging over all Cho-positive voxels. In these 14 patients there was a significant linear correlation between Cho with percent enhancement at 2 min after injection, SE%-2min (r = 0.75, p = 0.002), and pharmacokinetic parameters Ktrans (r = 0.74, p = 0.003), and k(ep) (r = 0.76, p = 0.002). The results suggested that overall there is a correlation between Choline metabolism and angiogenesis activity. Since Choline is associated with cell replication and angiogenesis is required to support tumor growth, this might explain the correlation between these two sets of measures among different lesions.

Quantification of Choline-containing Compounds in Malignant Breast Tumors by 1H MR Spectroscopy Using Water as an Internal Reference at 1.5 T

The quantification of choline-containing compounds (Cho) in breast tumors by proton MR spectroscopy (1H-MRS) has been of great interest because such compounds have been linked to malignancy. In this study, an internal reference method for the absolute quantification of Cho metabolite in malignant breast tumors was presented using a clinical 1.5 T scanner. We performed in vitro measurements to examine the accuracy of absolute quantification using four phantoms of known choline chloride concentrations. There was a high correlation between the calculated concentrations by MRS and the known concentrations (r2 > 0.98). We applied the technique to in vivo breast study conducted on 45 patients with biopsy-confirmed breast cancer. After T1 and T2 relaxation times were corrected, the Cho levels in this work had a range of 0.76-21.20 mmol/kg from 34 MR spectra of 32 patients with malignant breast lesions. This result was rather consistent with the previously published value (i.e., 1.38-10 mmol/kg, Bolan et al. in Magn Reson Med 50:1134-1143, 2003). Therefore, we conclude that the internal method using the fully relaxed water as a reference could be used for quantifying Cho metabolite accurately in breast cancer patients using a clinical 1.5 T scanner.

Application of an automated voxel-based morphometry technique to assess regional gray and white matter brain atrophy in a canine model of aging

In recent years, voxel-based morphometry (VBM) has emerged as a technique to examine regional brain changes associated with normal and pathological aging. Despite its popularity in studies of human aging, application of VBM to animal models of brain aging is rare. In the present study, VBM techniques were developed to validate earlier region of interest (ROI) measures of brain aging in the dog and to provide a more comprehensive analysis of local changes in a canine model of brain aging. Consistent with previous findings, frontal lobe atrophy increased with age, most notably in aged male dogs. Age-related gray matter reductions were also observed in parietal and temporal lobes, thalamus, cerebellum, and brainstem. Temporal lobe atrophy was particularly prominent in old females. A number of age-related changes in white matter not previously explored in the dog were also identified with VBM. Specifically, aged males exhibited greater decreases in the internal capsula and cranial nerve bundles compared to decreased volumes in the alveus of the hippocampus in old female dogs. Together, the present results indicate that application of VBM techniques in a canine model of aging yields more comprehensive information regarding topographical patterns of brain aging in male and female dogs than previously reported using traditional manual ROI methods.

Frontal lobe volume, function, and beta-amyloid pathology in a canine model of aging

Application of magnetic resonance imaging (MRI) techniques reveals that human brain aging varies across cortical regions. One area particularly sensitive to normal aging is the frontal lobes. In vitro neuropathological studies and behavioral measures in a canine model of aging previously suggested that the frontal lobes of the dog might be sensitive to aging. In the present study, MRI scans were acquired to compare age-related changes in frontal lobe volume with changes in executive functions and beta-amyloid pathology in the frontal cortex of beagle dogs aged 3 months to 15 years. Decreases in total brain volume appeared only in senior dogs (aged 12 years and older), whereas frontal lobe atrophy developed earlier, appearing in the old dogs (aged 8-11 years). Hippocampal volume also declined with age, but not occipital lobe volume past maturity. Reduced frontal lobe volume correlated with impaired performance on measures of executive function, including inhibitory control and complex working memory, and with increased beta-amyloid accumulation in the frontal cortex. Age-related hippocampal atrophy also correlated with complex working memory but not inhibitory control, whereas occipital lobe volume did not correlate with any cognitive measure. These findings are consistent with the frontal lobe theory of aging in humans, which suggests that the frontal lobes and functions subserved by this region are compromised early in aging.

A longitudinal study of brain morphometrics using serial magnetic resonance imaging analysis in a canine model of aging

Longitudinal changes in cortical atrophy, ventricular enlargement, and lesion development in serial MRI scans collected from 47 healthy dogs from 1999 (8-11 years old) to 2002 (11-14 years old) were studied. The first method involved manual region of interest volumetric analysis to examine changes in cerebral and ventricular volume during the three years. No change in cerebral volume was detected but ventricular volume increased significantly each year in 2000, 2001, and 2002. Increased ventricular volume parallels early studies of age-dependent ventricular enlargement in the brain of aging beagle dogs. The second method involved a visual analysis of co-registered serial MRIs for each subject. Consistent with the volumetric results, there was no visible change in cortical thickness indicating no cerebral atrophy, but a significant increase in ventricular size was noted. Visual examination also revealed a significant increase in number of dogs who developed aging lesions over the last 2 years in 2001 and 2002. Additionally, a disproportionate number of lesions were recorded in the frontal cortex and caudate nucleus compared to other brain regions. These lesion findings are consistent with other studies in the aging dog that suggest that the frontal lobes may be particularly vulnerable to age-related changes.

Effects of scopolamine challenge on regional cerebral blood volume. A pharmacological model to validate the use of contrast enhanced magnetic resonance imaging to assess cerebral blood volume in a canine model of aging

Cognitive impairment resulting from disruption of cholinergic function may occur through modulation of cerebrovascular volume (CBV). In the present study, dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) was used to examine cerebrovascular volume in young and old dogs during baseline and after administration of a cholinergic antagonist (scopolamine). In the first study, 24 animals (2-15 years of age) were given a baseline scan followed by a second scan after scopolamine administration (30 microg/kg). Gray matter rCBV was significantly higher than white matter rCBV during baseline and scopolamine administration. In the second study a subset of 7 dogs (4 young and 3 old) received scopolamine before anesthesia was induced for a second DSC-MRI scan. Consistent with the first study, gray matter rCBV was significantly higher than white matter rCBV. Scopolamine administered before anesthesia however, resulted in higher rCBV values compared to baseline in cerebral gray matter. Additionally, rCBVs were higher in young dogs at baseline in gray and white matter and marginally higher in gray matter when scopolamine was administered before anesthesia. These results indicate that in the dog, rCBV varies with brain compartment, decreases with age, and that DSC-MRI provides a measure of cerebrovascular function which may be related to age-dependent changes in cognition, brain structure, and neuropathology.

Increased blood clotting, microvascular density, and inflammation in eotaxin-secreting tumors implanted into mice

An important theme that is emerging in cancer research is the interaction between tumor cells and the host stroma. Because many types of human cancer are infiltrated by eosinophils that are believed to mediate an anti-tumor cytotoxic effect, we developed and studied a transfected B16 murine melanoma cell line that secretes high levels (510 pg/ml/100,000 cells/day) of eotaxin, a chemokine that recruits and activates primarily eosinophils. Here we report that there was increased inflammation (eosinophils, mast cells, mononuclear cells), blood clotting, and microvascular density within the tumors produced by subcutaneous implants of eotaxin-secreting tumor cells in 10 C57BL/6 compared to tumors produced by wild-type tumor cells. The extensive blood clotting in the eotaxin-transfected tumors was associated with significantly decreased blood flow to the tumors as measured by magnetic resonance imaging [(mean maximum signal enhancement of eotaxin-secreting tumors, 147 +/- 57 (n = 7) compared to 202 +/- 36 signal enhancement units (n = 8) for the wild-type melanoma cells; P = 0.04 by two-tailed, unpaired t-test]. Surprisingly, there was no significant difference between the growth rates or mean masses of the eotaxin-secreting tumors (750 +/- 280 mg, n = 10) and the wild-type tumors (780 +/- 290, n = 10) after 20 days of growth in vivo, despite the significantly slower growth rate in vitro of the eotaxin-secreting tumor cells. We conclude that eotaxin and the resultant tumor-infiltrating inflammatory cells are not likely to mediate a significant anti-tumor effect in vivo. Instead, elevated eotaxin is associated with increased inflammation, microvascular density, and blood clotting. Thus, eotaxin and eosinophils may play a more complex role in modulating the growth of tumors than the simple, anti-tumor cytotoxic effect that has been previously proposed.