Molecular medical researchers no longer have to fly blind as they develop ways to navigate therapeutic genes or stem cells toward disease sites. By harnessing the natural behavior of viruses and cells, molecular imaging illuminates the path of therapeutic genes and cells and serves as the guiding light in human clinical trials. The technology is influencing basic research now and is making the transition to human trials that could have sweeping implication for radiological practice in the future.
The importance of reporter gene imaging stems from MI's incredible demands on image resolution. Ideally, MI should enable a radiologist to directly observe genes as they switch on or off in response to genetically based therapies. In reality, in vivo imaging does not generate enough resolution to delineate individual cells, much less the activity of nanometric structures inside their nuclei. Reporter gene imaging involves a clever set of strategies most frequently using PET, SPECT, or optical imaging to indirectly observe intracellular or genomic events.
Researchers are beginning to appreciate the insights possible with this new technology, according to Dr. Ronald G. Blasberg, director of the Center for Multidisciplinary In Vivo Molecular Imaging at Memorial Sloan-Kettering Cancer Center.
"What you basically have is the people doing gene therapy agreeing to link their gene therapy protocols to reporter gene imaging," Blasberg said.
Reporter gene imaging almost always involves two components: a targeted reporter gene capable of penetrating and occupying cells that exhibit specific gene expressions, and a reporter probe that carries a radioisotope and can accumulate in great quantities only in those cells occupied by the reporter gene. The PET or PET/CT presentation of reporter gene imaging is similar to that of standard FDG-PET. Cells expressing the desired genetic behavior glow orange on the screen. Hyperenhancing stem cells carrying the reporter gene and accumulated probes can be observed as they arrive and begin to proliferate in targeted tissue.
One of the more well-known reporter genes is herpes simplex virus type 1 thymidine kinase (HSV1-tk). After being fused into a construct with a constitutive promoter, HSV1-tk is packaged into a vector, usually an adenovirus (see illustration, page 2). As in many recent studies, the viral vector is injected at a target site such as a tumor. The adenovirus transduces the reporter gene construct into cells, which transcribe the reporter genes into messenger RNA (mRNA) in the nucleus. Once translated, HSV1-tk specifically phosphorylates thymidine analogs, such as the drugs ganciclovir and penciclovir, and traps them in the cell. As thymidine analogs, they can interfere with DNA replication and halt cell division. The use of HSV1-tk to stop cancerous growths has thus earned the name "suicide gene therapy."
With positron-emitting fluorine isotopes attached, low concentrations of these thymidine analogs also reveal their location in the body on PET. Other varieties of reporter probes were designed to abet phosphorylation and/or uptake into the cell. These include 2«-deoxy-2«-[F-18]fluoro-5-fluoro-1-b-D-arabinofuranosyluracil (FFAU) and 9-(4-[F-18]-fluoro-3-hydroxymethyl-butyl)guanine (FHBG). Iodine isotopes such as I-124 also emit positrons and can be incorporated into fluorodeoxyuridine derivatives to produce the probe 5-[I-124]-iodo-2«-fluoro-2«-deoxy-1-b-D-arabinofuranosyluracil (FIAU), which the HSV1-tk also phosphorylates.
"Originally, these radiolabeled nucleosides weren't developed for the purposes of suicide gene therapy but for imaging viral infection in humans," said Dr. Peter S. Conti, director of the PET Imaging Science Center at the University of Southern California. "But when gene therapy hit the presses, it was immediately obvious that a radioactively labeled version of ganciclovir could also be used as a reporter probe."
For years, Conti focused on developing tracers that would allow him to spot viral infections that tend to lay hidden in the body. This approach would be a far better option than, for example, the risky procedure of taking a biopsy of the brain in search of viral encephalitis. To study the infection of herpes simplex virus, Conti and others in the field attached a radioactive tracer to drugs such as ganciclovir, which was commonly used to treat viral infection.
Many systems are now available. Nearly all rely on PET, SPECT, or optical cameras such as cooled charge-coupled devices. Optical imaging is a valuable tool in the preclinical world of cells and small animals, which are small enough to allow detection of fluorescent and bioluminescent light. PET, SPECT, and gamma cameras can detect radionuclide-labeled probes anywhere in the human body, which is too large for optical signals to be detected. Although compact optical breast imagers have been developed, researchers disagree about whether optical images will find a place in clinical practice.
"We use a lot of optically based systems for monitoring small animals, but it doesn't translate well to the clinic," said Harvey Herschman, Ph.D., vice chair of molecular and medical pharmacology at the University of California, Los Angeles.
"For clinical applications, we're strongest at imaging radionuclides with PET or SPECT."
Reporter genes for optical imaging include green and red fluorescent proteins, firefly luciferase, and sea pansy Renilla luciferase. The sophistication of cameras has extended the use of optical imaging, used for decades to study gene expression in cultured cells, to small animals. Fluorescence and bioluminescence are easy to use, have a low price tag, generate images with low background, and do not require radioactive probes. These advantages keep optical imaging popular. It is the preferred modality in preclinical studies, and researchers are developing fiber-optic scopes packaged into catheters and endoscopes to reach the inner organs and cavities of the human body.
"No single imaging modality can address all the questions that we can answer with molecular imaging. Each of the correlating reporters has its strengths and weaknesses," said Dr. David Piwnica-Worms, director of the Molecular Imaging Center at Washington University in St. Louis.
TREAT AND TELL
Integrated into human clinical trials, the reporter genes are proving valuable in providing essential feedback. Most are fused into a gene construct with the therapeutic gene, which shares the same promoter. The expression of the reporter gene thus reflects the expression of the therapeutic gene. But because of the additional regulatory requirements for adding the reporter gene, most existing gene therapy protocols use viral thymidine kinase as the suicide gene and as a reporter gene, Blasberg said.
That was the strategy for Dr. Sanjiv "Sam" Gambhir, director of the Molecular Imaging Program at Stanford University (MIPS) and his colleague, Dr. Ivan Penuelas at the University of Navarra in Spain. Their study, scheduled to be published in the Journal of Gastroenterology, provides details of their pioneering efforts in conducting the first human clinical trial of adenoviral gene therapy with reporter gene imaging using FHBG and PET/CT. The results showed active gene expression at sites of the tumor in seven patients with hepatocellular carcinoma who received injections of the viral vector. By comparing images taken on days two and nine, the researchers found that ganciclovir treatment caused the tumors to shrink.
Gambhir is also planning a study that will use reporter gene imaging to follow gene therapy for rheumatoid arthritis.
Researchers at Henry Ford Health System in Detroit plan to employ imaging in gene therapy for prostate cancer. Their study will use an adenovirus to deliver a gene construct that includes the suicide genes cytosine deaminase and HSV1-tk, which leave cancerous cells more vulnerable to the effects of radiation therapy. The gene construct, also carrying the reporter gene NIS, will be injected into the prostate.
"It works great in dogs, but does it work in people?" asked Steve Brown, director of preclinical studies and a staff scientist in the department of radiation oncology at Henry Ford.
At the University of Alabama at Birmingham, ovarian cancer and head and neck cancer will be the targets for two human clinical trials slated to begin this year, said Kurt R. Zinn, Ph.D., codirector of molecular imaging. The studies will use adenoviruses, modified to increase their infectivity, to deliver the HSV1-tk suicide gene to kill ovarian tumor cells or the cytosine deaminase gene to fight head and neck cancer. Both studies, led by Dr. David Curiel, gene therapy director, and Dr. Ronald Alvarez, head of gynecologic oncology, will track gene expression using the somatostatin receptor hSSTr2 reporter gene.
"This is an evolutionary process," Zinn said. "We make improvements, and then we learn something else and improve it more."
While positive results are inspiring researchers, they are also facing a barrage of obstacles. Most challenging is the interference of the immune system, which is programmed to neutralize foreign particles such as adenoviruses, other viral vectors, nonviral vectors, therapeutic and reporter proteins, and gene constructs and transcripts.
In Gambhir's trial, the expression of HSV1-tk lasted only a few weeks. A second injection of the gene construct, which was packaged into the same adenovirus, failed to result in the type of robust gene expression previously seen.
"It's probably because of the immune system's response to the virus," Gambhir said. "In cancer, it's a catch-22: You want long-term gene expression and a good immune response, but it's hard to get both."
Researchers, however, would be hard-pressed to give up on viral vectors.
"Viruses will be the vehicles that we'll be using for the near future," Herschman said. "Viruses are naturally built to target cells and bring in information. At least for now, we're not going to be better than nature at designing cell-targeting agents."
Gambhir and many others are studying lentiviruses, which can integrate into the cellular genome and provide gene expression that lasts for months. The long-term expression opens the possibility of creating an effective therapy with just a single administration of virus. As members of the retrovirus family, however, lentiviruses will be intensely scrutinized before they are declared safe for human trials. Gambhir is working with lentiviruses containing a reporter gene that is exclusively expressed in the prostate.
"I think the outlook is positive as we prove their safety more and more, and as we get more experience," he said.
Another promising strategy may save investigators from having to wrestle with viruses. If the patient's own cells are used to deliver therapeutic genes, the immune system won't see the delivery vehicle as a foreign object. And by inserting a reporter gene, cellular immunotherapists can finally gather the information they need to improve the protocols.
"There are so many unknowns in this process," Piwnica-Worms said. "The opportunity to directly visualize and assess the migration of the cells in animal models is where imaging reporter genes are having a tremendous impact."
That's what oncologists at the City of Hope National Medical Center are counting on. The institution in Duarte, CA, is collaborating with Gambhir to treat glioblastoma with the aid of reporter gene imaging. Researchers will take patients' T cells and transduce them with the gene encoding the particular antigen on the tumor and a reporter gene.
"You harvest your own cells, teach them how to do a better job of killing the tumor, and reinject them, and they kill the tumor," Gambhir said. "So now when we have the cells ex vivo, we will also inject a reporter gene. We're marrying the best of both worlds."
REPORTERS TAKE HEART
Cardiologists are also embracing reporter genes in their study and treatment of heart disease.
Over the years, efforts to stimulate new blood vessel growth in the heart by administering the gene for vascular endothelial growth factor (VEGF) have been hampered by the lack of a reliable way to monitor expression of VEGF, said Dr. Joseph C. Wu, a cardiologist at the MIPS. To solve the problem, Wu has been working on ways to introduce molecular imaging. He recently completed a study in which he injected rats with adenoviruses carrying both VEGF and the reporter gene HSV1-tk. While PET imaging allowed Wu to monitor and quantify the activity of VEGF, such expression was limited to two weeks because of the immune system's response to the adenovirus.
To circumvent this problem, Wu has begun using the rat's own embryonic stem cells to deliver VEGF along with a reporter gene. His studies in rats have laid the groundwork for monitoring the activity of transplanted cells in rats and understanding their survival patterns. Wu transduced both firefly luciferase and HSV1-tk reporter genes into rat embryonic cardiomyoblasts, which he then injected into the rats' hearts. Optical imaging and microPET revealed that the cells remained in the heart myocardium and expressed both reporter genes. But because these cells lasted for only two weeks, Wu's lab is investigating ways to prolong cell survival.
"The advantage of molecular imaging is that you can transduce these cells with multiple reporter genes," he said. "With optical bioluminescence, you can do high-throughput studies with small animals and figure out which cell types are best in vivo, and then translate it into clinical applications using PET."
Gene and cell therapists face the additional challenge of gaining acceptance from patients and the FDA. Researchers in the field are constantly guarding against the potential for gene constructs to travel to sites outside the target area and to integrate into the genome of stem cells, Brown said.
Still, the FDA's regulatory hurdles can be overwhelming, Conti said. Approaching the approval of therapeutic gene constructs and their respective probes as drugs, the federal agency has slowed the path to human trials. Unlike pharmaceutical companies, universities, which are usually the source of such innovative agents, don't have the wherewithal to complete the tedious obstacle course of FDA approval.
But as the protocols mature and professional societies push the issue, the path may widen, said Conti, who is president-elect of the Society of Nuclear Medicine.
"There's momentum in the right direction in reducing the regulatory requirements," he said. "Within a year or two, there will be a much clearer and well-defined path, and people will move quickly into human studies."
That doesn't mean that pharmacies will be stocking gene or cell therapy kits any time soon, however.
"In reality, we're a long way from thinking about the routine use of diagnostic imaging genes to look at function or for therapy," Piwnica-Worms said. "While promising work is under way for human application, it's way in the future."