Diffusion MRI with Semi-Automated Segmentation Can Serve as a Restricted Predictive Biomarker of the Therapeutic Response of Liver Metastasis.

PURPOSE: To assess the value of semi-automated segmentation applied to diffusion MRI for predicting the therapeutic response of liver metastasis. METHODS: Conventional diffusion weighted magnetic resonance imaging (MRI) was performed using b-values of 0, 150, 300 and 450s/mm(2) at baseline and days 4, 11 and 39 following initiation of a new chemotherapy regimen in a pilot study with 18 women with 37 liver metastases from primary breast cancer. A semi-automated segmentation approach was used to identify liver metastases. Linear regression analysis was used to assess the relationship between baseline values of the apparent diffusion coefficient (ADC) and change in tumor size by day 39. RESULTS: A semi-automated segmentation scheme was critical for obtaining the most reliable ADC measurements. A statistically significant relationship between baseline ADC values and change in tumor size at day 39 was observed for minimally treated patients with metastatic liver lesions measuring 2-5cm in size (p=0.002), but not for heavily treated patients with the same tumor size range (p=0.29), or for tumors of smaller or larger sizes. ROC analysis identified a baseline threshold ADC value of 1.33µm(2)/ms as 75% sensitive and 83% specific for identifying non-responding metastases in minimally treated patients with 2-5cm liver lesions. CONCLUSION: Quantitative imaging can substantially benefit from a semi-automated segmentation scheme. Quantitative diffusion MRI results can be predictive of therapeutic outcome in selected patients with liver metastases, but not for all liver metastases, and therefore should be considered to be a restricted biomarker.

Monitoring the development of xenograft triple-negative breast cancer models using diffusion-weighted magnetic resonance imaging.

Evaluations of tumor growth rates and molecular biomarkers are traditionally used to assess new mouse models of human breast cancers. This study investigated the utility of diffusion weighted (DW)-magnetic resonance imaging (MRI) for evaluating cellular proliferation of new tumor models of triple-negative breast cancer, which may augment traditional analysis methods. Eleven human breast cancer cell lines were used to develop xenograft tumors in severe combined immunodeficient mice, with two of these cell lines exhibiting sufficient growth to be serially passaged. DW-MRI was performed to measure the distributions of the apparent diffusion coefficient (ADC) in these two tumor xenograft models, which showed a correlation with tumor growth rates and doubling times during each passage. The distributions of the ADC values were also correlated with expression of Ki67, a biomarker of cell proliferation, and hypoxia inducible factor (HIF)-1α and vascular endothelial growth factor receptor-2 (VEGFR2), which are essential proteins involved in regulating aerobic glycolysis and angiogenesis that support tumor cell proliferation. Although phosphatase and tensin homolog (PTEN) levels were different between the two xenograft models, AKT levels did not differ nor did they correlate with tumor growth. This last result demonstrates the complexity of signaling protein pathways and the difficulty in interpreting the effects of protein expression on tumor cell proliferation. In contrast, DW-MRI may be a more direct assessment of tumor growth and cancer cell proliferation.

Task-based evaluation of segmentation algorithms for diffusion-weighted MRI without using a gold standard.

In many studies, the estimation of the apparent diffusion coefficient (ADC) of lesions in visceral organs in diffusion-weighted (DW) magnetic resonance images requires an accurate lesion-segmentation algorithm. To evaluate these lesion-segmentation algorithms, region-overlap measures are used currently. However, the end task from the DW images is accurate ADC estimation, and the region-overlap measures do not evaluate the segmentation algorithms on this task. Moreover, these measures rely on the existence of gold-standard segmentation of the lesion, which is typically unavailable. In this paper, we study the problem of task-based evaluation of segmentation algorithms in DW imaging in the absence of a gold standard. We first show that using manual segmentations instead of gold-standard segmentations for this task-based evaluation is unreliable. We then propose a method to compare the segmentation algorithms that does not require gold-standard or manual segmentation results. The no-gold-standard method estimates the bias and the variance of the error between the true ADC values and the ADC values estimated using the automated segmentation algorithm. The method can be used to rank the segmentation algorithms on the basis of both the ensemble mean square error and precision. We also propose consistency checks for this evaluation technique.

Evaluating segmentation algorithms for diffusion-weighted MR images: a task-based approach.

Apparent Diffusion Coefficient (ADC) of lesions obtained from Diffusion Weighted Magnetic Resonance Imaging is an emerging biomarker for evaluating anti-cancer therapy response. To compute the lesion's ADC, accurate lesion segmentation must be performed. To quantitatively compare these lesion segmentation algorithms, standard methods are used currently. However, the end task from these images is accurate ADC estimation, and these standard methods don't evaluate the segmentation algorithms on this task-based measure. Moreover, standard methods rely on the highly unlikely scenario of there being perfectly manually segmented lesions. In this paper, we present two methods for quantitatively comparing segmentation algorithms on the above task-based measure; the first method compares them given good manual segmentations from a radiologist, the second compares them even in absence of good manual segmentations.

A Clustering Algorithm for Liver Lesion Segmentation of Diffusion-Weighted MR Images.

In diffusion-weighted magnetic resonance imaging, accurate segmentation of liver lesions in the diffusion-weighted images is required for computation of the apparent diffusion coefficient (ADC) of the lesion, the parameter that serves as an indicator of lesion response to therapy. However, the segmentation problem is challenging due to low SNR, fuzzy boundaries and speckle and motion artifacts. We propose a clustering algorithm that incorporates spatial information and a geometric constraint to solve this issue. We show that our algorithm provides improved accuracy compared to existing segmentation algorithms.

Regulation of the Warburg effect in early-passage breast cancer cells.

Malignancy in cancer is associated with aerobic glycolysis (Warburg effect) evidenced by increased trapping of [(18)F]deoxyglucose (FdG) in patients imaged by positron emission tomography (PET). [(18)F]deoxyglucose uptake correlates with glucose transporter (GLUT-1) expression, which can be regulated by hypoxia-inducible factor 1 alpha (HIF-1alpha). We have previously reported in established breast lines that HIF-1alpha levels in the presence of oxygen leads to the Warburg effect. However, glycolysis and GLUT-1 can also be induced independent of HIF-1alpha by other factors, such as c-Myc and phosphorylated Akt (pAkt). This study investigates HIF-1alpha, c-Myc, pAkt, and aerobic glycolysis in low-passage breast cancer cells under the assumption that these represent the in vivo condition better than established lines. Similar to in vivo FdG-PET or primary breast cancers, rates of glycolysis were diverse, being higher in cells expressing both c-Myc and HIF-1alpha and lower in cell lines low or negative in both transcription factors. No correlations were observed between glycolytic rates and pAkt levels. Two of 12 cell lines formed xenografts in mice. Both were positive for HIF-1alpha and phosphorylated c-Myc, and only one was positive for pAkt. Glycolysis was affected by pharmacological regulation of c-Myc and HIF-1alpha. These findings suggest that c-Myc and/or HIF-1alpha activities are both involved in the regulation of glycolysis in breast cancers.

Promise and progress for functional and molecular imaging of response to targeted therapies.

Biomarkers to predict or monitor therapy response are becoming essential components of drug developer's armamentaria. Molecular and functional imaging has particular promise as a biomarker for anticancer therapies because it is non-invasive, can be used longitudinally and provides information on the whole patient or tumor. Despite this promise, molecular or functional imaging endpoints are not routinely incorporated into clinical trial design. As the costs of clinical trials and drug development become prohibitively more expensive, the need for improved biomarkers has become imperative and thus, the relatively high cost of imaging is justified. Imaging endpoints, such as Diffusion-Weighted MRI, DCE-MRI and FDG-PET have the potential to make drug development more efficient at all phases, from discovery screening with in vivo pharmacodynamics in animal models through the phase III enrichment of the patient population for potential responders. This review focuses on the progress of imaging responses to new classes of anti-cancer therapies targeted against PI3 kinase/AKT, HIF-1alpha and VEGF. The ultimate promise of molecular and functional imaging is to theragnostically predict response prior to commencement of targeted therapy.

Menkes Copper ATPase (Atp7a) is a novel metal-responsive gene in rat duodenum, and immunoreactive protein is present on brush-border and basolateral membrane domains.

We previously noted strong induction of genes related to intestinal copper homeostasis (Menkes Copper ATPase (Atp7a) and metallothionein) in the duodenal epithelium of iron-deficient rats across several stages of postnatal development (Collins, J. F., Franck, C. A., Kowdley, K. V., and Ghishan, F. K. (2005) Am. J. Physiol., 288, G964-G971). We now report significant copper loading in the livers and intestines of iron-deficient rats. These findings are consistent with the hypothesis that there is increased intestinal copper transport during iron deficiency. We additionally found that hepatic Atp7b gene expression does not change with iron deficiency, suggesting that liver copper excretion is not altered. We have developed polyclonal antibodies against rat ATP7A, and we demonstrate the specificity of the immunogenic reaction. We show that the ATP7A protein is present on apical domains of duodenal enterocytes in control rats and on brush-border and basolateral membrane domains in iron-deprived rats. This localization is surprising, as previous in vitro studies have suggested that ATP7A traffics between the trans-Golgi network and the basolateral membrane. We further demonstrate that ATP7A protein levels are dramatically increased in brush-border and basolateral membrane vesicles isolated from iron-deficient rats. Other experiments show that iron refeeding partially corrects the hematological abnormalities seen in iron-deficient rats but that it does not ameliorate ATP7A protein induction, suggesting that Atp7a does not respond to intracellular iron levels. We conclude that ATP7A is involved in copper loading observed during iron deficiency and that increased intestinal copper transport is of physiological relevance, as copper plays important roles in overall body iron homeostasis.