The National Cancer Institute estimated that 1,437,180 people would be diagnosed with cancer in 2008 and that 565,650 people would die of the disease. Cancer research is, of course, ongoing, and recent decades have seen great strides in our understanding of the disease’s pathology as well as in our approaches to diagnosis and treatment. Still, despite such significant advances, these statistics underscore the ever-present need to find innovative ways to fight cancer.Magnetic resonance imaging (MRI) offers tremendous potential for facilitating cancer screening and diagnosis, as well as for directing and monitoring treatment. Conventional MRI, which allows structural imaging of tumors, facilitates the detection and localization of primary cancer lesions. Investigators also have employed MRI for functional imaging, both alone and in combination with other modalities such as positron emission tomography (PET) and single-photon emission computed tomography (SPECT). By enabling researchers and clinicians to track the physiological changes underlying the formation and growth of tumors rather than restricting their examination to the tumors themselves, such strategies often open up entirely new opportunities for cancer diagnosis and treatment.Figure 1. For both diagnosis and treatment of breast cancer, researchers have explored the potential of the altered choline metabolism that accompanies the disease. The enzyme choline kinase is elevated in breast and other cancers and thus acts as an endogenous biomarker that can aid in diagnosis. Because the pathways that lead to altered choline metabolism are up-regulated in cancer cells but not in normal cells, understanding them could permit treatment that does not damage normal cells. Shown here are immunoblot assay results demonstrating reduced choline kinase expression levels in MCF-12A, MCF-7 and MDA-MB-231 cells following transfection with a small interfering RNA against the enzyme (siRNA-chk). Note that the expression levels in the nonmalignant MCF-12A cells are considerably lower than in the two breast cancer cell lines. Reprinted with courtesy of Cancer Research.Advancing diagnosis and treatmentDoctors have used conventional MRI as a supplemental tool for diagnosis of breast cancer since the early 1990s. Researchers are still divided, however, as to the clinical potential of the method. Some studies suggest that it is more successful than mammography and ultrasound, the current gold standards, in identifying tumors in the breast. Others, though, highlight MRI’s limitations – namely, its moderate specificity and its inability to detect calcifications. Aside from its diagnostic merits, the technique requires specialized equipment and highly trained technicians to read the scans, so its relatively high cost prevents its use for screening breast cancer in the general population.Dr. Zaver M. Bhujwalla of the Johns Hopkins University School of Medicine in Baltimore notes, however, that molecular and functional MRI (fMRI) are blazing new paths to clinical applicability. In an NMR in Biomedicine review of the field published online on Sept. 16, she and colleagues explained the current investigative focus: Whereas a decade ago breast cancer research concentrated mostly on genetic alterations, researchers today recognize that the progression, aggressiveness and response of the disease to treatment are influenced by a number of parameters, including the physiological microenvironment of the tumor, the extracellular matrix and a host of secreted factors and cytokines. MRI can provide molecular-functional characterization of tumors in the breast and thus can facilitate detection as well as advance drug discovery, development and delivery.Molecular-functional MRI also can contribute to diagnosis, and possibly treatment, by identifying targets that act across multiple levels. Bhujwalla and coworkers have, for example, reported on the altered choline metabolism that accompanies breast cancer. In this and other types of cancer, the enzyme that converts choline to phosphocholine – choline kinase – is elevated, thus serving as an endogenous biomarker of the disease. Researchers have, accordingly, developed a variety of choline-based methods for diagnosis, staging and therapy assessment in cancer patients.Understanding the altered choline metabolism also makes it possible to treat the disease without causing many of the detrimental side effects of other therapies. Most cytotoxic therapies damage normal cells as well as cancer cells, often resulting in diarrhea and nausea, for example. The ideal treatment, Bhujwalla and coworkers have maintained, would kill cancer cells while sparing normal tissue. This can be achieved by targeting pathways that are up-regulated in the former but not in the latter – as is the case with the pathways leading to the alterations in choline metabolism.Dr. Noriko Mori and colleagues, in a study published in the Dec. 1, 2007, issue of Cancer Research, reported that by downregulating choline kinase using, for example, a small interfering RNA against the enzyme, they could achieve significant reduction of cell proliferation in malignant human breast cancer cells but not in nonmalignant human mammary epithelial cells. Thus, the alterations in choline kinase that accompany cancer might suggest a safer, and possibly even more effective, therapy.Peering into the brainIn addition to advancing diagnosis and treatment of breast cancer, MRI, either as a stand-alone method or in conjunction with other technologies, can aid in addressing brain cancer. By itself, fMRI permits imaging of angiogenesis – the growth of new vasculature, and a marker of an aggressive lesion – by way of measuring blood flow. Thus, doctors can use fMRI in the clinic to characterize the aggressiveness of a tumor and to complement the structural information provided by conventional MRI, which helps to reveal the tumor’s size.Similarly, explained Dr. Bruce Rosen, Director of the Athinoula A. Martinos Center for Biomedical Imaging at Massachusetts General Hospital in Charlestown, clinicians can call on fMRI to monitor a tumor’s response to therapy. If the growth of new vasculature in the tumor slows, and especially if it reverses itself, doctors can consider a treatment effective. Functional MRI can contribute to treatment planning as well, in particular for patients being considered for surgical intervention. But it isn’t enough to circumscribe the mass to be resected; surgeons must be careful also to avoid healthy brain tissue and the brain’s functional areas – those devoted to motor function, for example, or memory. Inadvertently removing even a small section of one of these areas can result in deficits in the wake of the procedure.“Conventional MRI is very good at finding the tumor itself,” Rosen explained, “but by itself isn’t very good at saying, ‘This is the part of the brain the surgeon doesn’t want to cut.’” For this reason, doctors have turned to MRI techniques developed for brain mapping studies that identify regions responsible for specific functions such as memory. Carefully delineating such areas in the patient’s brain can contribute to the overall efficacy of surgical intervention by helping to prevent deficits.Producing anatomic road mapsIn addition to using functional MRI to delineate normal from diseased tissue in the brain and to track tumor growth, clinicians are using structural MRI to help localize metabolic information acquired with other modalities. For example, PET or SPECT might identify areas within the brain where glucose consumption is high (Aggressive tumors are “avid sugar-holics,” Rosen said), but these methods don’t provide the spatial resolution needed to pinpoint the locations. By producing high-resolution structural images of the brain, MRI can provide a sort of anatomic guide by which to determine the site of the metabolic activity.Figure 2. A recently developed PET imager fits inside an MRI scanner, allowing simultaneous acquisition of structural images with MRI and of metabolic activity with PET. This multimodal approach could help to improve diagnosis and treatment of brain tumors, for example.The challenge, Rosen explained, lies in mapping metabolic information onto the anatomy. Researchers and clinicians typically achieve this by fusing PET (or SPECT) and MRI images using specially developed software. However, because such images are acquired separately, the resulting overlay might suffer from a range of problems stemming from, for instance, differences in patient positioning.To address such issues, medical device companies have been working to develop prototype systems that combine two imaging modalities for synchronous data acquisition. One example of this hybrid technology is a PET imager that “lives inside” an MRI scanner. The first instrument in the US to combine PET/MR imaging was delivered to the Athinoula A. Martinos Center earlier this year. Called BrainPET, this three-dimensional PET scanner from Siemens Medical Solutions is designed to operate inside the bore of that company’s Trio 3T scanner.Figure 3. Using the integrated MR-PET imaging system, doctors and researchers can generate overlaid images of scans acquired simultaneously, helping to localize the metabolic activity within the high-resolution structural scan. Shown here are images of a patient with a malignant tumor, including PET (top row), MR (bottom row) and fused PET/MR images (middle row).The integrated system enables true simultaneous imaging of brain structure and metabolic and molecular processes, Rosen said, all within a 30- or 40-minute examination. It permits acquisition of fully spatially and temporally co-registered PET and MRI data, and it allows doctors and researchers to synergistically harness the advantages of each imaging modality for a more comprehensive view of tissue anatomy and physiology. For imaging tumor response to novel anti-angiogenic treatments, the BrainPET will enable simultaneous measurement of glucose metabolism (via PET) and tumor angiogenesis (via MRI).