Warning! OUTDATED BROWSER DETECTED!   Please update your browser immediately for a better experience on this website. Learn More
  • Look Ahead: Molecular Imaging

    Vikas Kundra, MD, PhD, discusses the evolution and potential of molecular imaging. By Vikas Kundra, MD, PhD

    March 1, 2018

    Molecular imaging is bringing novel approaches to diagnose and monitor disease. It has already proven successful in clinical practice. New advances will enable molecular characterization and should enable evaluation of therapy not only at the level of detection and response, but also to assess delivery and mechanistically to assess whether a therapy is affecting its intended target. This opens up fresh and exciting avenues for imaging.


    Multimodal imaging of ovarian cancer. DM-dual-Gd-ICG liposomal nanoparticles, which have high MR relaxivity due to gadolinium chelates on their surface and within them, and can be used for near infrared fluorescence (NIR) due to incorporated indocyanine green, were injected intravenously two days earlier. Intraperitoneal human ovarian tumor xenografts (dotted lines or arrows) enhanced by MR (left, axial view) and demonstrated NIR signal (left bottom, coronal view). This should enable presurgical planning by MR and NIR-guided surgical resection. K, kidneys; I intestines.

    What is Molecular Imaging?

    Broadly, molecular imaging may be defined as non- or minimally-invasive assessment of biologic and pathologic processes based on molecular or functional analysis. This may include both molecularly targeted and non-targeted imaging. Although there may be some debate regarding the exact definition, the practical clinical outcome is adding analysis of molecular/functional alteration to anatomic changes in order to improve differential diagnosis and to monitor therapy.


    Imaging of gene expression. Mouse tumors made to express a human somatostatin receptor type 2 (WT)-based reporter or a signaling deficient variant (SD) were imaged after injection of a radiolabeled somatostatin analogue. A reporter that does not perturb the cell such as by initiating signal transduction is desirable. The preferred signaling deficient version was imaged similar to wild type but was muted in initiating cellular signaling. Vec, negative control tumor without reporter expression; expected radiotracer excretion is seen via the kidneys (K).

    Non-Targeted Imaging

    An example of non-targeted molecular imaging performed clinically is diffusion weighted MRI, which is increasingly playing a role in detecting tumors and evaluating response. It is based on assessing molecular motion by MRI. Diffusion weighting depends on the b-value, which reflects the strength and timing of the magnetic gradients. At mid b-values, tissue-level motion can be evaluated and highly cellular structures such as tumors can be detected. Newer techniques include studying faster motion using low b-values, such as afforded by intravoxel incoherent motion (IVIM) for evaluating parameters such as D and f akin to tissue diffusion coefficient and perfusion fraction, as well as studying slower motion using high b-values to limit T2 “shine-through.” These techniques are being evaluated as to whether they enable improved detection and/or response evaluation.

    Molecular-targeted Imaging

    An example of targeted molecular imaging is 18F-fluorodeoxyglucose (FDG). The beta particle emission by the 18F label enables imaging by positron emission tomography (PET). FDG or fluorodeoxyglucose mimics glucose and enters the cell through glucose (GLUT) transporters and then becomes phosphorylated. The phosphorylated form is not a good substrate for the next enzyme in glycolysis and its negative charge keeps it from crossing the cell membrane; thus, it becomes entrapped in the cell and accumulates. Inflammatory cells have high metabolic rates and 18F-FDG PET imaging is used for infection/inflammation imaging. As outlined in the Warburg effect, cancer cells tend to utilize the less efficient anaerobic pathway to generate energy more than do normal cells, and thus require more input glucose, the primary fuel source. 18F-FDG PET imaging has been used to image a variety of, but not all, tumors and their response to therapy.

    With the advent of hyperpolarization imaging, it has become possible to perform biochemistry in vivo, including assessing late events in glycolysis. Hyperpolarization can increase the signal from atoms in molecules 10,000- to 100,000-fold, but lasts only seconds to minutes for the great majority. Due to the tremendous signal gain, not only can the input molecule be imaged, but also its metabolic products. For example, a key inflection point in glycolysis is pyruvate which sits at the decision point of whether to undergo anaerobic respiration and produce lactate or undergo aerobic respiration. In tumors, the rate of pyruvate to lactate conversion tends to be higher than in normal cells. In animal models there has been a suggestion that the rate of conversion may be associated with the degree of tumor dedifferentiation or stage and that it may be able to detect early response to therapy even when 18F-FDG PET imaging is not informative. Because multiple metabolites and even molecularly targeted agents may be hyperpolarized, this technology has promise to affect a variety of imaging schema.

    Molecularly-targeted imaging has been in place in nuclear medicine departments for some time. One example is peptide-based imaging of somatostatin receptors. 111In-octreotide-based imaging has been a nuclear medicine workhorse. The gamma particle emission by the 111In label enables imaging by single photon emission computed tomography (SPECT). Octreotide mimics somatostatin and binds to somatostatin receptors including types 2 and 5. Newer labels include 68Ga and 64Cu to enable PET for higher resolution imaging. Newer peptides include octreotate for more selectivity for somatostatin receptor type 2. These advances and advances in scanner design and fusion systems such as PET-MR will enable improved detection and localization of molecular signatures. Combining functional and anatomic information can enable quantification of receptors as shown in animal models.

    Promising New Techniques

    Oncology is beginning to define tumors beyond just the site of the organ of origin to also include alterations in gene expression for which molecular imaging may make significant contributions. Fusion imaging can help guide biopsy. This is particularly important for obtaining tissue for understanding genomic, metabolomic and proteomic alterations in lesions before and after targeted therapy. This information aids in understanding if the intended target of the therapy was indeed altered and whether this resulted in the expected clinical effect. Biopsy is impractical for whole subject evaluation, whereas, molecularly targeted imaging should enable evaluation of pathologic heterogeneity, such as in a primary tumor and among it and metastases in order to select appropriate targeted therapies. These may include immunotherapies in the future. New tracers are becoming available/approved, such as for prostate cancer imaging and for imaging somatostatin receptors, which should increase reimbursement. Such tracers should benefit from new hybrid technologies such as PET/MR since anatomy and pathology in many important target organs such as in the pelvis, liver, and brain can be better delineated by MR than CT. In addition, improved hardware and software such as time of flight imaging and solid state detectors for PET and faster gradients for MR are resulting in faster scanners with improved image quality resulting in increased subject throughput and improved lesion detection as well as interpretation.

    Tau and amyloid beta plaque-imaging agents were recently approved for imaging of neurodegenerative diseases such as Alzheimer’s disease. Such techniques have the potential for early detection even before there are overt clinical signs and for monitoring efficacy of preventive and therapeutic strategies, assuming that these or developing imaging agents visualize a relevant surrogate of clinical disease. In this case, the degree of tau or amyloid beta deposition is presumed to reflect the degree of Alzheimers disease and has been proposed to be used to assess the efficacy of disease interventions.

    Gene Therapy Applications

    New therapeutic techniques require new imaging approaches. Gene therapy and cellular therapy have found recent success but are limited by an inability to visually localize and quantify gene expression or cellular localization without biopsy. This may be approached using reporter imaging where a gene for imaging is inserted into the gene therapy vector. Commonly, this insert results in the production of a protein that can itself be imaged or can bind to an imaging agent. Most commonly the imaging agent is a radiopharmaceutical since nuclear medicine techniques have the greatest sensitivity, but MRI may follow in the future. It should be possible to insert an appropriately designed gene for imaging into a variety of gene therapy vectors so that several vectors or cell types can be imaged. In addition, it should be possible to link the gene for imaging to a therapeutic gene so that imaging can help measure expression of the therapeutic gene. Ideally, the reporter imaging gene system will enable imaging in patients, is small in order to fit into vectors, is non-immunogenic, and does not depend on its function for imaging so as not to disturb normal cell function. These reporters are becoming available, such as signaling deficient reporters based upon somatostatin receptor type 2. This type of imaging would further enable gene and cellular therapies.

    Team Approach

    Molecular imaging requires teams of experts such as those skilled in biology, chemistry, physics and instrumentation. It requires development from bench to bedside and expertise in commercialization to bring the innovations to patients. Molecular imaging has already been established in the clinic and new approaches suggest potential for improving current imaging and strong potential to develop new approaches for meeting future needs.

    Vikas Kundra, MD, PhD, is professor and director of molecular imaging in the Department of Radiology, University of Texas MD Anderson Cancer Center with a joint appointment in the Department of Cancer Systems Imaging. He received his MD and PhD from Harvard University. He trained at Harvard Medical School’s Brigham and Women’s Hospital. He is a fellow of the Society of Body Computed Tomography-Magnetic Resonance Imaging and Distinguished Investigator of the Academy of Radiology Research. Dr. Kundra practices as a clinical radiologist focused on body imaging primarily using CT and MRI. He has authored multiple clinical and basic/translational science papers and secured grants from federal sources including the National Institutes for Health, the Department of Defense and the National Science Foundation.