In The Clinic

Radiating Change

Kevin Camphausen, M.D. (Photo: R. Baer)

Ever since his training at Harvard Medical School, Kevin Camphausen, M.D., knew that his career would combine translational research with patient care. While at Harvard, he trained with C. Norman Coleman, M.D. (now Head of the Radiation Research Program, Extramural NCI), and worked with Judah Folkman, M.D., a pioneer of research into angiogenesis and tumor formation that has resulted in a new class of cancer therapies. As head of the Imaging and Molecular Therapeutics Section in the Radiation Oncology Branch at CCR, Camphausen has a rare opportunity to forge the bench-to-bedside connections that are so vital to the progress of radiation oncology. The branch handles approximately 450 consults per year for many different cancers in adults as well as children. Although his team will do consults and referrals for every patient that comes through the door, Camphausen and his colleagues are limited to treating those that fall within the inclusion criteria of one of their clinical trial protocols. Thus, they cannot treat everyone, but each patient they do treat is also part of research that will lead to a brighter future for others diagnosed with their disease.

Radiation oncology is based on the principle that tumor tissue is more sensitive to radiation damage than normal tissue. Ionizing radiation damages DNA. The same mutations that cause cancerous cells to rapidly proliferate by forfeiting normal cellular checkpoints and DNA repair mechanisms make these cells more vulnerable to the molecular damage inflicted by radiation. In addition, the rapid divisions of cancerous cells cause DNA damage to accumulate at an increasing pace as it is passed on to daughter cells until the progeny are ultimately no longer viable.

Radiation oncology branched off from Radiology as a clinical specialization more than 40 years ago in order to foster its own unique blend of expertise. To treat patients effectively, we have to understand and manipulate both biology and physics. On the physics side, we use one set of technologies to identify and delineate the tumor within the body— computed tomography (CT) and magnetic resonance imaging (MRI)—and another set of technologies to irradiate it—linear accelerators (Linac). We must decide on the more esoteric parameters controlling the beam of radiation as well as contend with the more mundane but equally challenging issues of making sure the physical placement of the patient (and hence his tumor) in the beam is accurate to within millimeters.

On the biology side, we need to determine which types of tumors are best treated focally and which require wider radiation beams; we need to balance treatment between the different sensitivities of normal tissue and tumors. And the greatest potential for advances in radiation oncology lies in a better understanding of tumor biology and in discovering new agents to sensitize cancer cells to radiation.

For me, combining laboratory research, clinical research, and clinical care is the most satisfying way to bring about advances in radiation oncology that will extend and improve patients’ lives.

The greatest potential for advances in radiation oncology lies in a better understanding of tumor biology.

Glioblastoma Multiforme

Although the Radiation Oncology Branch (ROB) is involved in the treatment of a myriad of cancers, my own research focuses on brain cancers. Glioblastoma multiforme (GBM)—a cancerous proliferation of astrocytes, a type of "support cell" in the brain—is the most common brain cancer with 20,000 new cases diagnosed yearly. This cancer is the same type of cancer that Senator Edward Kennedy was diagnosed with last year. We typically see patients in their 40s and 50s who, after having no previous history of neurological disorders, suddenly experience a seizure or another acute symptom that prompts a physician to order an MRI. It is not uncommon for the GBM to have invaded a large portion of the brain by then.

With the vast amount of information about cancer now available online, most cancer patients tend to be fairly knowledgeable about their disease—my colleague down the hall who specializes in prostate cancer will have patients come in with a three-ring binder full of information that they have downloaded from the Internet about their disease and treatment options. But GBM patients— who may otherwise be in the prime of life with small children under their care—are often shell-shocked. There is not a lot of time between diagnosis and treatment, and the prognosis, unfortunately, is not very good for these patients. The standard of care treatment is a seven-week regimen of radiation therapy in combination with temozolomide, a drug that interferes with DNA replication. The average length of survival after diagnosis for these patients is 14 months, with about eight months after treatment until signs of disease progression emerge.

Images from a patient with glioblastoma multiforme (GBM). Magnetic resonance image of a large left frontal GBM taken one day before surgery (left). Tissue sample from the same tumor showing cellular proliferation and hemorrhage (right, large arrow). (Image: Adapted from Weil RJ, 2005; PLoS Med 3(1): e21. doi:10.1371/journal.pmed.0030031)

My laboratory has been looking for other drugs that might, in combination with radiation therapy, improve the odds for these patients. As a result of the work that we have done in cell and animal models, we are currently running a phase II clinical trial to augment the standard treatment of GBM with the addition of a drug called valproic acid, an inhibitor of the enzyme histone deacetylase (HDAC) (see "Balancing Silence: How a Cell’s Fate Is Determined," page 22).

From Cell Lines to Human Trials

HDACs are a class of enzymes that are involved in epigenetic regulation—they modify (deacetylate) the histone proteins that in turn interact with DNA to restrict or encourage gene expression. Altered HDAC activity has been seen in several cancers and appears to prevent the expression of tumor suppressor genes, so there has been a great deal of interest in developing HDAC inhibitors as cancer therapeutics.

Some work in the 1980s also indicated that HDAC inhibitors might enhance the sensitivity of tumor cells to radiation, but the available compounds at the time were not suitable for administration to people.

Another scientist here at the NIH, Philip Tofilon, Ph.D. (who has since moved to the H. Lee Moffitt Cancer Center in Florida), had the idea to revisit this therapeutic possibility with the newer generations of HDAC inhibitors that were being developed. In 2004, my team collaborated with Dr. Tofilon to publish research showing that an HDAC inhibitor, MS-275, enhanced the lethal effects of radiation on tumor cells. Unfortunately, we were not able to develop a collaboration with the company that makes MS-275 to continue this line of work, but we were encouraged enough by our results to jump at the suggestion from our colleague, Howard Fine, M.D., Chief of the Neuro-Oncology Branch at CCR, that another HDAC inhibitor valproic acid—might be an even better choice for enhancing radiation sensitivity. Valproic acid has long been used in the treatment of epilepsy, which means we know it is safe to use in people and will be transported across the barrier that restricts blood-borne molecules from entering the brain.

My laboratory has been looking for other drugs that might, in combination with radiation therapy, improve the odds for these patients.

So we went back and repeated our experiments with valproic acid instead of MS-275. In general, we go through a staged process of testing potential drugs in the laboratory. First, we perform what is known as a clonogenic survival— essentially, we irradiate tumor cells in a dish with or without the compound to see if it affects cell survival. Then we study the cellular mechanisms that might be responsible for the altered survival— regulation of the cell cycle and various cell death programs. Once we are confident that we have a strong result in cell lines, we move to testing animal models. Often, it is sufficient to introduce the cancer cell line of interest under the skin of a mouse and study the resulting tumor formation, but because there are special problems with drugs reaching the brain, my laboratory uses orthotopic models in which a glioblastoma cell line is implanted directly into the mouse brain.

Mouse models are, of course, only models. For example, GBMs in people are highly invasive, whereas they are not in our animal models. And in order to introduce human cancer cell lines into these mice, we need to genetically impair their immune systems so that they do not reject the grafts. However, strong data that the drug is crossing into the brain and affecting tumors in animal models are usually sufficient to start trials in people.

We are still enrolling patients in our clinical trial for the use of valproic acid to enhance radiation sensitivity in the standard of care regimen for GBM. One challenge that we face is purely practical—unlike many other courses of radiation treatment, the treatment for GBM is protracted. We put the patient on the treatment table every day for a seven-week course of radiation. Thus, it can be difficult to recruit patients who do not live in the vicinity of the NIH.