Hedvig Hricak, MD, PhD

Methodologies of molecular imaging for oncology

The Molecular Imaging Laboratory in the Department of Radiology/Molecular Pharmacology and Chemistry Program is focused on the development of new reporter genetic systems for specific applications, such as sequential in-vivo imaging studies in cancer biology, cancer immunotherapy and radiation sciences. These reporter systems allow a) the development of multi-modality reporter genes, reporter tumor cell lines and animals for cross-breeding with transgenic models of cancer to image and study oncogenesis, as well as targeted drug response, in animals; b) the development of new approaches for repetitive in-vivo optical, nuclear and MR imaging of the fate of antigen-specific T-lymphocytes after their adoptive transfer for antitumor immunotherapy in prostate and colon cancers (done initially in animals and then in patients); and c) novel assessments of targeted drug therapy based on direct imaging of specific molecular targets pre- and post-treatment, with the goal of rapid translation into relevant clinical studies. In addition, the laboratory is conducting studies involving high-resolution magic angle spinning (HR-MAS) spectroscopy of prostate cancer tissue, the findings of which can be translated into clinical spectroscopy.

	[Enlarge Image] Figure 1 The paradigm of HSV1-TK Reporter Gene Imaging

The overall aim is to develop approaches for repetitive in vivo imaging of the fate of T lymphocytes after their adoptive transfer for antitumor vaccination. It is obvious that continuous whole body monitoring of the efficacy of many adoptive cell-based therapies can not be done by invasive methods (e.g., multiple biopsies) except for blood sampling.

In our opinion, noninvasive whole body imaging would significantly aid in the development and clinical implementation of cell-based therapeutic approaches by providing the means for noninvasive monitoring of the fate of the injected T cells over a long period of observation. Imaging could help to address several questions related to T cell migration and homing to the tumor target, their long-term viability, and their subsequent activation and cytolytic activity.

The ultimate aim of noninvasive imaging of the degree of T cell activation at tumor target is to assess cytolitic potential of tumor infiltrating T cells and predict the tumorolytic efficacy early on during therapy in clinical setting.

The paradigm of noninvasive reporter gene imaging involves the administration of a radiolabeled probe that is selectively bound or metabolized (e.g., phosphorylated) and trapped by interaction with the gene product (e.g., an enzyme) in the reporter gene-transduced cell. In this paradigm, the level of probe accumulation is proportional to the level of reporter gene product expressed.

We recently demonstrated the feasibility of serial PET imaging of antigen-specific T cells bearing an HSV1-tk reporter/suicide gene. In this work, we explore the feasibility of genetic labeling of T lymphocytes with novel reporter genes/probe combinations and novel multi-reporter gene constructs that would allow for repetitive noninvasive in vivo PET imaging of T cell trafficking and activation. We have developed several dual-reporter systems in which 1 reporter gene will be expressed constitutively as a "beacon" gene for imaging of the localization of adoptively transferred T cells; while the second reporter will be expressed as a "sensor" gene in a T cell- and activation-specific manner, thus allowing for visualization of activation and functional properties of T cells.

The proposed multi-reporter gene imaging paradigm will be implemented in the assessment and optimization of new anticancer T cell vaccines that utilize anti-CEA or anti-PSMA artificial T cell receptors in experimental therapies of colorectal and prostate carcinomas, respectively. Imaging will be used to monitor trafficking, localization, and activation of adoptively transferred T cells targeted against colon and prostate carcinomas before and during the co-stimulatory therapy with various cytokines.

	[Enlarge Image] Figure 2 In vivo PET imaging of hƒ´TK2 expression in transduced and wild-type U87 xenografts in mice at two hours after [124I]FIAU administration

An important requirement for a long-term genetic labeling of immune stem/progenitor cells in humans is that the reporter gene product should not be immunogenic in order to avoid the possibility of rejection of transduced cells. In preparation for potential clinical applications of reporter gene imaging, we have developed (and are in the process of developing) novel nonimmunogenic reporter genes that are of human origin for long-term repetitive imaging in patients. One of them is human truncated mitochondrial thymidine kinase type 2 (hDTK2), which is likely to be nonimmunogenic in humans; and could be readily translated into a clinical setting for PET imaging with [124I]FIAU or other novel nontoxic highly specific probes, such as 18F- or 11C-radiolabeled FEAU. hDTK2 may also be used as a suicide gene due to its ability to phosphorylate clinically used nucleotide analogs.

	[Enlarge Image] Figure 3. Multi-modality Imaging of GFP/Firefly Luciferase and RFP/Renilla Luciferase Reporter Genes Expression performed sequentially in the same living mouse in vivo. Structure of retroviral vectors containing GFPFLuc and RFPRLuc reporter genes (A,E); U87 cells expressing GFPFLuc (B) and RFPRLuc (F) reporter proteins; whole body bioluminescence imaging of GFPFLuc (C) and RFPRLuc (G) reporter gene expression in vivo; whole body fluorescence imaging of GFPFLuc (D) and RFPRLuc (H) reporter gene expression in vivo.

Multimodality imaging approaches allow for different imaging technologies to be combined during the course of the same study and harness the best features and utilities of each of the modality. The advantages of hybrid reporter genes that allow for multimodality imaging in vivo have been recognized as well. Initially, the advantages of multimodality imaging using herpes simplex virus type one thymidine kinase (HSV1-tk) for nuclear imaging and green fluorescent protein (GFP) for optical fluorescent imaging were demonstrated. It allows for optical microscopic and whole body fluorescence imaging, as well as PET imaging. The fusion of bioluminescence and fluorescence reporter genes has also been reported.

In a recent publication, we described the HSV1-tk/GFP/Firefly luciferase (TGL) triple reporter construct that preserves the functional activity of its subunits and allows for seamless transition from fluorescence microscopy and FACS to whole body bioluminescence imaging, nuclear (PET, SPECT, gamma camera) imaging, and back to in situ fluorescence image analysis. Currently we are developing several fusion reporter genes bearing functional components for fluorescent and bioluminescent imaging in the same animal. When validated, these approaches will have direct applications to various imaging studies where 2 molecular events need to be tracked simultaneously, including cell trafficking of 2 or more distinct cell populations, gene therapy vectors, and indirect monitoring of several endogenous genes through the use of these reporter genes. These fusion reporter genes will allow for a seamless transition from one imaging modality to the other and facilitate cross-interpretation of results obtained with various in vitro, in situ, and in vivo imaging modalities.