Brain Tumor Research
By Charles Cobbs, M.D., neurosurgeon
Since returning to the San Francisco Bay Area in 2005, I have been fortunate to build a brain tumor research program at California Pacific Medical Center Research Institute (CPMCRI). My neurosurgical colleague at CPMC, Brian Andrews, M.D., had already begun a brain tumor clinical tumor board program that facilitated our growing interest in this area. Since the time of my neurosurgery residency at UC San Francisco (UCSF) in the 1990s, I had become interested in the most aggressive human brain tumors, called gliomas. Among these, glioblastoma multiforme (GBM) is traditionally associated with a survival of less than two years for most patients.
After training at UCSF, I returned to join the Division of Neurosurgery faculty at UAB Medical School, in my home town of Birmingham, Alabama. There I received research awards and grants to study the associated inflammatory pathways that were known to accelerate progression of GBM in patients. During this time, I diverted some time to investigate a risky concept that brain tumors might not arise spontaneously from DNA mutations, rather that these tumors may be associated with a chronic infection and inflammation in the brain, which could gradually lead to a GBM developing over years. This concept was not even considered a reasonable possibility in the brain tumor community in 2000, the year I initiated the project. But many aspects of GBM patients suggested that their tumors could have a viral involvement. GBMs were more common in upper-middleclass individuals, as is the case of Hodgkin’s lymphoma, which is associated with Epstein-Barr virus (EBV). Furthermore, GBM patients were known to possess profound abnormalities in their immune function, but the relevance of this aspect of their cancer had never been clearly defined. Based on a hunch, we focused our efforts on potential viruses that could chronically infect brain cells in immunosuppressed patients, and this led me to focus on one particular virus called cytomegalovirus (CMV).
CMV is a member of the herpes virus family, akin to herpes simplex virus (HSV), EBV, herpes zoster virus (HSV), and Kaposi’s sarcoma virus (KSV). Both EBV and KSV were known at the time to cause cancers, so it seemed reasonable that CMV might possess the capacity to cause or promote GBM. By enlisting the assistance of a superb technician who possessed the necessary skill to detect very low levels of CMV infection in brain tumors, our group determined that nearly 100 percent of GBM were infected with CMV, and we published findings in the journal Cancer Research in 2002 (1).
Unfortunately, as with many discoveries that seemingly come from “left field,” this publication appeared to remain unnoticed for several years. Then a couple of research groups attempted to confirm the findings without using the extra-sensitive detection techniques that we had developed, and their efforts were unsuccessful, thus casting doubt on the validity of our findings.
Undaunted, we were able to pursue a group at the National Cancer Institute (NCI) to support our work in this area through the NCI Special Projects in Oncology Research Excellence (SPORE) mechanism, and this funding facilitated confirmation of the presence of CMV in GBM by other well-established groups, including those at Duke University Medical Center and Karolinska Institute in Stockholm, Sweden.
The Duke group, headed by Drs. John Sampson and Duane Mitchell, not only confirmed our initial observations of CMV in more than 95 percent of GBM but also determined that the presence of the virus in the tumor cells might represent a unique target for an anticancer immune response(2).
Cancer immunotherapy is an emerging area of great interest in the cancer research community, and the concept of inoculating a cancer patient with a vaccine that will cause his/her immune system to target the tumor cells has always held great promise. The difficulty with the concept stems from the paradox that tumor cells are a person’s own cells, so the question is what target can a cancer vaccine use that will kill tumor cells but not other cells in the patient’s body. With the discovery that CMV infection appears to occur in the GBM cells specifically, and not in the adjacent normal brain cells, Drs. Sampson and Mitchell hypothesized that GBM patients had lost their immunity to CMV, but if they could boost a GBM patient’s immune system so that it could “see” the CMV infection in the tumor, then the ensuing immune response may be able to target and eradicate infected tumor cells. Clinical trials were initiated in 2006 at Duke to do just this. Duke is now the leading single-center Phase I and Phase II trials of GBM vaccines that target CMV, and the preliminary results are very encouraging(3).
At Karolinska Institute, another of our colleagues, Dr. Cecilia Soderberg-Naucler, was attempting to see whether use of an oral antiviral drug, valganciclovir, taken once a day in addition to the standard treatment of GBM—chemotherapy and radiation therapy—might be able to thwart the progression of the tumor. Dr. Soderberg-Naucler’s study has not been published, but the preliminary data suggest that patients who received the antiviral drug have had statistically increased survival compared to those receiving placebo.
Our research group has made progress in understanding the role the virus plays in GBM biology and how more novel antiviral strategies might be used to target the virus in GBM. Dr. Liliana Soroceanu, who heads the laboratory efforts for our research at CPMCRI, has made this research progress possible. A native of Romania, Dr. Soroceanu joined CPMCRI in 2006 after finishing a postdoctoral fellowship at Genentech.
Within months of joining our group, Dr. Soroceanu discovered that a key cell receptor in the development of GBM, called PDGFR, was the key human cellular receptor for CMV, and when the virus attaches to a cell, such as a tumor cell, this receptor is turned on and drives cancer-causing pathways. This seminal finding was published in the journal Nature (4). Further work in the lab has led to the concept that certain CMV genes, when expressed in human GBM cells, not only can drive the cells to grow faster but can cause dramatic alteration of normal growth-promoting and growth-arresting signals. Worldwide, investigations are leading to the novel concept that CMV is a tumor promoter in GBM.
Most recently, this consensus was reached at a small international meeting organized by our group and funded by the Accelerate Brain Cancer Cure (ABC2) and National Brain Tumor Society (NBTS) in Washington, D.C. in March 2011. At this meeting, a panel of international experts in virology and oncology presented recent findings in the area of CMV and GBM, and wrote a meeting summary documenting the state of the art in the field.
Our current efforts are aimed at initiating the first U.S. clinical trial using a novel, highly active antiviral drug against CMV for GBM therapy. We are planning a Phase I trial of the novel drug CMX001 for GBM. CMX001 is an orally available, highly potent antiviral drug that can penetrate brain tumor tissues. We plan to perform clinical trials to determine whether use of this drug is safe in the setting of radiation and chemotherapy in primary and recurrent GBM.
- Cobbs, C.S., et al. Human cytomegalovirus infection and expression in human malignant glioma. Cancer Res 62, 3347-3350. (2002).
- Mitchell, D.A., et al. Sensitive detection of human cytomegalovirus in tumors and peripheral blood of patients diagnosed with glioblastoma. Neuro Oncol 10,10-18 (2008).
- Sampson, J.H. & Mitchell, D.A. Is cytomegalovirus a therapeutic target in glioblastoma? Clin Cancer Res 17, 4619-4621.
- Soroceanu, L., Akhavan, A. & Cobbs, C.S. Plateletderived growth factor-alpha receptor activation is required for human cytomegalovirus infection. Nature 455, 391-395 (2008).