Brain Tumor Imaging: Can Molecular Imaging Fill the Gaps?
New integrated imaging for brain tumors offers insight into tumor metabolism
Brain tumor imaging needs an upgrade. Most advanced MRI technologies have existed for more than a decade, and FDG-PET is less than ideal for imaging brain tumors. These options may not provide the necessary imaging power, especially as new treatment options emerge. This issue, we’re delving into the state of brain tumor imaging and are examining the potential of integrated technologies, new molecular modalities and tracers.
MRI imaging at a glance
Fondazione Istituto Neurologico Carlo Besta in Milan, Italy, like several other medical centers around the globe, employs advanced MR methods for state-of-the-art brain tumor imaging. Neuroradiologists routinely use perfusion and spectroscopy to characterize lesions, especially low-grade gliomas. The goal, says Alberto Bizzi, MD, of the department of neuroradiology, is to learn more about the metabolism and physiology of the tumor to better characterize the biology of the tumor and provide a prognosis. Later in the course of the disease, advanced MR imaging methods also fulfill a second purpose and measure therapeutic response. “There are currently many new drug options for brain tumors, but none is perfect. Preclinical research indicates that the combination of spectroscopy and perfusion can detect hypoxic areas. By measuring tumor response, MR perfusion and MR spectroscopy inform the decision to continue or switch drugs,” explains Bizzi.
Surgical patients are often referred for functional MRI studies. fMRI helps neuroradiologists identify eloquent regions of the brain (that enable language, motion, senses and memory) and their relationship to the tumor, providing neurosurgeons a map of cortical areas to spare. Recent research comparing fMRI to surgical intraoperative electrocortical stimulation in awake patients has validated the imaging study, demonstrating sensitivity and specificity in the 80 to 85 percent range, says Bizzi. Another pre-surgical option is MR tractography with diffusion tensor imaging. Using this type of evaluation, neuroradiologists map white matter bundles involved in motor control, speech and attention to provide additional information about which subcortical regions have to be spared during surgery.
Current imaging methods, however, do not provide a complete solution. MR spectroscopy and perfusion provide an estimate of a tumor’s cellular metabolism, but do not examine the tumor’s DNA and molecular biology. That is, the MR studies provide information about growth rate, but do not answer questions about the tumor cell’s DNA stability.
Molecular imaging may be used to identify proteins and receptors associated with various types of brain tumors. In turn, physicians could use the information to predict the most effective drug for a specific tumor.
PET/CT possibilities
PET/CT applications for brain tumor imaging are still relatively new but could deliver critical advances. Siemens Medical Solutions Biograph mCT is an important step forward, says Hartmuth Kolb, vice president of Siemens Medical Solutions Biomarker Research. It expands the strength of PET—the ability to visualize molecular events in a living organism—by combining PET with state-of-the-art CT. Another emerging technology is MR/PET, which represents another leap forward by further wedding structure, anatomy and function to enable physicians to better characterize brain lesions.
Other advances will come on the biomarker front, says Kolb. While FDG offers a good first option, it is not very specific. Brain imaging with FDG is problematic as the brain consumes a high amount of glucose, making it difficult to differentiate malignancies from the normal brain tissue. New tracers could increase specificity. 18F-FLT (fluorothymidine), for example, measures tumor proliferation and could play a role in monitoring therapeutic effectiveness. Another biomarker under development—18F-MISO (misonidazole)—measures oxygen saturation. This metric is critical as hypoxic cells tend to be more resistant to radiation therapy. Data about hypoxia could be used to fine-tune radiation therapy planning. That is, the radiation therapist could increase the entire radiation dose or increase the dose to the hypoxic areas, edging radiation therapy closer to the goal of personalized medicine.
MRI research at a glance
For more than 16 years, researchers at University of Michigan Medical School Department of Radiology in Ann Arbor, Mich., have been investigating diffusion-weighted MRI (DW-MRI) to track the diffusion value of water molecules in brain tumors following treatment. Researchers use MR images to produce a functional diffusion map (fDM), which may prove to be an effective and efficient way to provide an early prognostic indicator of a patient’s outcome, says Brian D. Ross, PhD, professor of radiology and biological chemistry. The technique may be used to measure early changes in the tumor and assess therapeutic response.
The principle behind fDM is fairly simple. DW-MRI can be used to quantify how rapidly water molecules move during a scan. Water molecules don’t move or diffuse very quickly in areas crowded with tumor cells. During effective chemotherapy, cell death eliminates the cellular barriers that impede water molecule movement. As the barriers are removed, the diffusion value rises, reflecting a less restricted environment that can be detected and quantified using DW-MRI.
fDM may improve current prognostic protocols, which employ conventional MRI imaging to provide an anatomical assessment of the tumor by looking at volume, morphology or size. One downside of conventional approaches is that it takes time for anatomical changes to occur, so physicians must wait up to 10 weeks to assess response. Specifically, the Macdonald criteria assesses radiologic response via MRI and correlates with patient survival, but the measurement takes place 10 weeks after the start of therapy. In contrast, fDM can be employed during treatment delivery at three weeks after the therapeutic-initiation. Recently, Ross and his colleagues published results of a 60-patient study in the Journal of Clinical Oncology comparing the Macdonald criteria to fDM.
The University of Michigan research showed that fDM can be employed three weeks from the start of treatment to stratify patients into responsive or non-responsive groups. Researchers concluded fDM offers predictive value equal to conventional neurological imaging of brain tumors, but at a much earlier time. In the future, researchers hope that fDM is further validated and could be employed as an early imaging treatment response biomarker that would allow for individualization of patient treatment.
Diffusion-weighted MRI has potential advantages for application in multi-center clinical trials, says Ross. For example, DW-MRI is a quantitative biophysical measure that can be obtained using any clinical strength scanner to produce the apparent diffusion maps needed for fDM analysis. Moreover, because diffusion is not related to perfusion, patients do not require contrast injection. And finally, DW-MRI requires very little scanning time as it only requires signaling averaging for about 20 seconds to several minutes.
Better brain tumor imaging
Researchers are working on multiple fronts to address critical brain tumor imaging needs. MRI is expanding and new techniques like fDM show promise in providing an early non-invasive mechanism for assessing therapeutic response. And as molecular modalities such as mCT and MR/PET enter the clinical realm, new and improved biomarkers will follow, improving the diagnostic information available to physicians. Armed with more accurate information about a tumor’s molecular activities, physicians can select the most appropriate therapeutic course.
 |
MRI imaging at a glance
Fondazione Istituto Neurologico Carlo Besta in Milan, Italy, like several other medical centers around the globe, employs advanced MR methods for state-of-the-art brain tumor imaging. Neuroradiologists routinely use perfusion and spectroscopy to characterize lesions, especially low-grade gliomas. The goal, says Alberto Bizzi, MD, of the department of neuroradiology, is to learn more about the metabolism and physiology of the tumor to better characterize the biology of the tumor and provide a prognosis. Later in the course of the disease, advanced MR imaging methods also fulfill a second purpose and measure therapeutic response. “There are currently many new drug options for brain tumors, but none is perfect. Preclinical research indicates that the combination of spectroscopy and perfusion can detect hypoxic areas. By measuring tumor response, MR perfusion and MR spectroscopy inform the decision to continue or switch drugs,” explains Bizzi.
Surgical patients are often referred for functional MRI studies. fMRI helps neuroradiologists identify eloquent regions of the brain (that enable language, motion, senses and memory) and their relationship to the tumor, providing neurosurgeons a map of cortical areas to spare. Recent research comparing fMRI to surgical intraoperative electrocortical stimulation in awake patients has validated the imaging study, demonstrating sensitivity and specificity in the 80 to 85 percent range, says Bizzi. Another pre-surgical option is MR tractography with diffusion tensor imaging. Using this type of evaluation, neuroradiologists map white matter bundles involved in motor control, speech and attention to provide additional information about which subcortical regions have to be spared during surgery.
Current imaging methods, however, do not provide a complete solution. MR spectroscopy and perfusion provide an estimate of a tumor’s cellular metabolism, but do not examine the tumor’s DNA and molecular biology. That is, the MR studies provide information about growth rate, but do not answer questions about the tumor cell’s DNA stability.
Molecular imaging may be used to identify proteins and receptors associated with various types of brain tumors. In turn, physicians could use the information to predict the most effective drug for a specific tumor.
PET/CT possibilities
PET/CT applications for brain tumor imaging are still relatively new but could deliver critical advances. Siemens Medical Solutions Biograph mCT is an important step forward, says Hartmuth Kolb, vice president of Siemens Medical Solutions Biomarker Research. It expands the strength of PET—the ability to visualize molecular events in a living organism—by combining PET with state-of-the-art CT. Another emerging technology is MR/PET, which represents another leap forward by further wedding structure, anatomy and function to enable physicians to better characterize brain lesions.
Other advances will come on the biomarker front, says Kolb. While FDG offers a good first option, it is not very specific. Brain imaging with FDG is problematic as the brain consumes a high amount of glucose, making it difficult to differentiate malignancies from the normal brain tissue. New tracers could increase specificity. 18F-FLT (fluorothymidine), for example, measures tumor proliferation and could play a role in monitoring therapeutic effectiveness. Another biomarker under development—18F-MISO (misonidazole)—measures oxygen saturation. This metric is critical as hypoxic cells tend to be more resistant to radiation therapy. Data about hypoxia could be used to fine-tune radiation therapy planning. That is, the radiation therapist could increase the entire radiation dose or increase the dose to the hypoxic areas, edging radiation therapy closer to the goal of personalized medicine.
MRI research at a glance
For more than 16 years, researchers at University of Michigan Medical School Department of Radiology in Ann Arbor, Mich., have been investigating diffusion-weighted MRI (DW-MRI) to track the diffusion value of water molecules in brain tumors following treatment. Researchers use MR images to produce a functional diffusion map (fDM), which may prove to be an effective and efficient way to provide an early prognostic indicator of a patient’s outcome, says Brian D. Ross, PhD, professor of radiology and biological chemistry. The technique may be used to measure early changes in the tumor and assess therapeutic response.
The principle behind fDM is fairly simple. DW-MRI can be used to quantify how rapidly water molecules move during a scan. Water molecules don’t move or diffuse very quickly in areas crowded with tumor cells. During effective chemotherapy, cell death eliminates the cellular barriers that impede water molecule movement. As the barriers are removed, the diffusion value rises, reflecting a less restricted environment that can be detected and quantified using DW-MRI.
fDM may improve current prognostic protocols, which employ conventional MRI imaging to provide an anatomical assessment of the tumor by looking at volume, morphology or size. One downside of conventional approaches is that it takes time for anatomical changes to occur, so physicians must wait up to 10 weeks to assess response. Specifically, the Macdonald criteria assesses radiologic response via MRI and correlates with patient survival, but the measurement takes place 10 weeks after the start of therapy. In contrast, fDM can be employed during treatment delivery at three weeks after the therapeutic-initiation. Recently, Ross and his colleagues published results of a 60-patient study in the Journal of Clinical Oncology comparing the Macdonald criteria to fDM.
The University of Michigan research showed that fDM can be employed three weeks from the start of treatment to stratify patients into responsive or non-responsive groups. Researchers concluded fDM offers predictive value equal to conventional neurological imaging of brain tumors, but at a much earlier time. In the future, researchers hope that fDM is further validated and could be employed as an early imaging treatment response biomarker that would allow for individualization of patient treatment.
Diffusion-weighted MRI has potential advantages for application in multi-center clinical trials, says Ross. For example, DW-MRI is a quantitative biophysical measure that can be obtained using any clinical strength scanner to produce the apparent diffusion maps needed for fDM analysis. Moreover, because diffusion is not related to perfusion, patients do not require contrast injection. And finally, DW-MRI requires very little scanning time as it only requires signaling averaging for about 20 seconds to several minutes.
Better brain tumor imaging
Researchers are working on multiple fronts to address critical brain tumor imaging needs. MRI is expanding and new techniques like fDM show promise in providing an early non-invasive mechanism for assessing therapeutic response. And as molecular modalities such as mCT and MR/PET enter the clinical realm, new and improved biomarkers will follow, improving the diagnostic information available to physicians. Armed with more accurate information about a tumor’s molecular activities, physicians can select the most appropriate therapeutic course.
| Next-generation ‘Smart Probes’ Differentiating Tumor Genetics |
| There are multiple routes to advance clinical brain tumor imaging. As imaging technologies become smarter and more integrated, new imaging probes offer a way to exploit the potential of molecular imaging solutions such as PET/CT and MR/PET. Thomas Meade, PhD, professor of chemistry at Northwestern University in Chicago, leads research focused on the study of molecular imaging of in vivo gene expression and the design of “smart probes” that incorporate novel functionality for MR imaging and electronic biosensors for the detection of DNA and proteins. His research provides the foundation for new and more sensitive methods for early detection of diseases, including gliomas. MII: Can you outline some of challenges in brain tumor imaging as they relate to MR probes? Meade: Currently, MRI contrast agents provide only anatomical information. Typically, agents provide contrast based on the higher perfused tissue characteristic of tumor cells. Current agents do not, however, provide physiological information to guide therapy or the prognostic course of a disease. For example, MR agents cannot differentiate necrotic from apoptotic cells post treatment. Since the agents cannot distinguish dead tumor cells from new tumor tissue, an invasive approach—surgery—may be required to make a determination. MII: What is the goal of your research? Meade: We are working to develop “bioactivated or responsive probes” to assist diagnostic and therapeutic decision-making and, ultimately, the prognostic course. These probes capitalize on the capabilities of molecular imaging technologies like PET/CT and MR/PET. Ultimately, we want to combine imaging and probes to provide physicians with information about the anatomy and physiology of specific tumors. The clinical end goal is personalized medicine. “Smart probes” will help drive personalized medicine by differentiating the genetic makeup of tumors, which, in turn, guides diagnostic and therapeutic decision-making. MII: Can you outline some of the advances in probe development? Meade: In 1997, we developed the first bioactivated MR probes that were modulated by the presence or absence of enzymes or secondary messengers. These conditional probes are not detectable until they are conditionally activated by a selected biochemical event such as gene expression. A familiar example of a conditional or bioactive probe for PET is 2-FDG, which lights up on PET scans in the presence of malignant cells. This class of agents has expanded greatly with a number of laboratories developing probes that can detect pH, oxygen and temperature. A second class of probes is focused on development of multi-modal contrast agents, or probes that can report in two modalities simultaneously. These agents are detectable by two modalities like MRI and optical imaging, PET/CT and PET/MR. Co-registration of acquired images from PET and MR probes as part of the same agent exploits the potential of both modalities. PET provides superior sensitivity of the agent, and MRI offers high spatial and temporal resolution. This is a powerful combination. (Multi-modal probes have not yet been clinically tested.) MII: What are the next steps? Meade: Our research is focused on expanding the library of bioactivated MR probes for enzymes and messengers such as calcium and zinc. We are combining these two classes by attaching bio-activated agents to multi-modal probes on nanoparticles. As before, the probe remains undetectable by either imaging modality until exposed to a specific event, which turns on the probe. The probe reports the presence of enzymes linked with genes expressed by malignant cells. MII: How might your research translate into clinical practice? Meade: One potential application centers on glioblastoma patients, where the survival rate is quite low. Patients typically undergo surgical procedures and radiation followed by chemotherapy. Successful therapy eradicates malignant tissue, resulting in necrotic cells. Unfortunately, necrotic cells resemble cancer cells; both are highly perfused. Traditional contrast agents cannot differentiate new tumor cells from necrotic cells. The challenge is to develop an agent that can distinguish necrosis from new tumor cells. A conditionally activated agent that turns on in the presence of an active tumor cells, but not in necrotic cells, could help pinpoint new tumor cells. The next generation of agents will be challenged to differentiate the specific genetic makeup of various tumors. A three-pronged design approach that targets an agent to a given cell line, and is activated in the presence of selected biological event, while being detectable by two modalities would be a considerable leap forward. |