Valcyte and Cytomegalovirus (CMV) in the treatment of Glioblastoma



Charles S. Cobbs, MD
The Gregory Foltz, MD Endowed Director
Ben & Catherine Ivy Center for
Advanced Brain Tumor Treatment
Swedish Neuroscience Specialists


Today we are going to talk about the antiviral drug Valcyte. Valcyte is an orally available form of the IV drug ganciclovir, that has been used in thousands of patients worldwide for decades.  Over the last 12 years, my laboratory has had an interest in the potential role of the virus cytomegalovirus (CMV) and brain tumors. In 2002 we published the first study in the journal Cancer Research showing that there was a very high incidence of CMV infection in glioblastoma, close to 100% (Cancer Res. 2002 Jun 15;62(12):3347-50) .  For years these findings did not resonate in the neuro-oncology community because no one could confirm them – until my colleagues at Duke University and at the Karolinska Institute in Stockholm, Sweden learn how to do the highly sensitive immunostaining technique we had developed, and then they subsequently confirmed our findings.

Even before our colleagues at Duke University and the Karolinska Institute confirmed the findings, we began to wonder if the virus was driving the tumor and if strategies to attack the virus might help create an antitumor response. Essentially there were two concepts that we considered. First, my colleagues at Duke who were very interested in immunotherapy considered the possibility that CMV-infected tumor cells would create a unique situation in which the immune system could attack those virally infected cells (ie., tumor cells) without harming normal uninfected cells.

Researchers at Duke have spent the last 8 years investigating this concept, and their preliminary studies suggest that this approach may have very good efficacy. These studies may lead to a completely revolutionary concept in terms of glioblastoma therapy – that reactivation of a dormant immune response to CMV could create a robust antitumor immune response.

The second anti-CMV approach that we considered as early as in 2004 was the possibility that an antiviral drug may decrease the aggressiveness of the tumor. We had reason to believe that CMV infection of tumor cells might not only promote the genetic mutations within the tumor but also promote properties of the tumor such as increased blood growth, increased immunosuppression, and increase invasiveness. Therefore at least in theory, the antiviral approach should be able to diminish some of these aggressive features.  At the time, I was at the University of Alabama at Birmingham (UAB) and was performing my research under an NCI SPORE grant for brain tumor research. We wrote a protocol at that time to try to use the commonly available oral anti-CMV drug Valcyte for the treatment of glioblastoma patients.

One of the problems we encountered early on was that this drug was made by Roche (now Genentech, a subsidiary of Roche), but they were unable to offer the drug for our clinical trials. Since Roche was not able to provide the drug for the study, we were unable to fund a trial in the United States given the high cost of the drug. However, Dr. Soderberg –Naucler in Stockholm, Sweden approached her medical center and found that the Swedish government was able to provide the drug for the studies.

In 2006, Dr. Soderberg –Naucler initiated a study at the Karolinska Institute using Valcyte in addition to radiation and temozolomide in patients with newly diagnosed glioblastoma who had undergone at least a 90% resection. This study only had about 48 patient’s total and it was randomized to placebo or Valcyte. The primary endpoint of the study was to determine whether or not there was a reduction in tumor volume at 6 months based on MRI imaging. Unfortunately, this endpoint was a poor indicator of overall efficacy. After 6 months, there was a trend in the patients receiving Valcyte suggesting smaller tumor size.  After 6 months, when patient’s found out that they were on the placebo arm and they heard the others patients were doing well with no toxicities from the Valcyte, many of the placebo patient switched over to the treatment arm of the study. The study had to be evaluated ultimately based on the primary endpoint of a 6 month tumor volume and the secondary endpoint of overall survival. Based on the “intention to treat” analysis, the study was published in the International Journal of Oncology in 2013 without a statistically significant improvement in survival. That is due to the fact that the patients had to be analyzed based on the original group they were placed in.

However, after the patients were unblinded, and after the initial improvements in patient survival were identified, the Karolinska group allowed almost all patients with newly diagnosed glioblastoma at their institution to start taking Valcyte within a month of their diagnosis. They continued to follow these patients and then retrospectively categorize them based on how long they had been on the Valcyte. Of note, this type of retrospective analysis raises all kinds of red flags in terms of determining any statistically significant findings. Nevertheless, they did follow a group of about 25 patients who were treated with Valcyte after diagnosis of glioblastoma. As they reported in the September 2013 issue of New England Journal of Medicine, this small group of patients had a 90% two-year survival and a mean overall survival of 56.4 months. The likelihood that this was due to some random chance appears to be unlikely.

Based on my own prescription off label of Valcyte to many of my patients while I was practicing in California, and the findings from the Swedish group, my suspicion is that there is significant improvement of survival in patients with glioblastoma who take Valcyte. This suspicion will require validation in Phase II studies.  The toxicities so far have been very low with occasional slight drops in white blood count among these patients.

The significant problem we have encountered since these results have been released to the public is that most neuro-oncologists don’t believe that these results can be considered worthy of changing standard practice, and I agree. Secondly, the cost of the drug is prohibitive for many patients and their insurance companies are often not willing to allow the patient to take Valcyte off label since it is not standard of care for glioblastoma.

To confuse matters even further, we are often asked to do a blood test for CMV in patients. It appears that glioblastoma patient’s immune systems are defective, and we have documented cases of patients whose blood tests are negative while the tumors are positive. We did develop a very sensitive immunohistochemical tests for CMV from primary glioblastoma paraffin embedded specimens when I was at UAB, but I have not yet been able to get this up and going as a reference lab since moving to Seattle. There are government guidelines that restrict me from being able to offer testing of patient’s tumors without having specific regulatory certifications outside of my institution.

Finally, it may be that Valcyte works not only against CMV infection but against tumor cells themselves. Valcyte is a drug that is essentially a modified nucleoside analog.  Nucleosides are the building blocks of DNA, and we know these as the “T, C, G, and A”  in the DNA strand.  Valcyte is a modified version of the “G” (guanine) in DNA. The drug works by inserting itself in the dividing DNA chain and it is thought that it does so much more effectively in the viral DNA than in the human DNA. This is thought to be the basis for its antiviral effect since this will cause chain termination of DNA synthesis. Obviously, terminating DNA synthesis will have an effect not only on viral replication but also on replicating tumor cells. There has also been recent literature that suggests that Valcyte or its parent drug ganciclovir is an extremely potent inhibitor of inflammation in the brain outside of its antiviral effects (J Exp Med. 2014 Feb 10;211(2):189-98). Since inflammation is an integral part of tumor biology, and inflammation and brain tumors may promote their biology, it’s possible that Valcyte has off target effects that inhibit tumor growth that we don’t even understand yet.

Ultimately, for patients the critical question is whether this drug helps them, whether it’s safe, and if they can get it at a reasonable cost. Despite my efforts over the last decade to start clinical trials to clarify these issues in the best possible way, I was unable to convince the pharmaceutical industry to provide the drug for the studies. I am happy that other researchers have recently been told that they may be able to do the study with Roche / Genentech offering their drug.

Here are some of the frequent questions we often hear from patients with respect to Valcyte. I’m sorry I do not have great answers for all of the questions but hopefully these answers will be helpful.

1. Do I need to have CMV testing of my tumor to be prescribed Valcyte?

This is a good question. In theory, you would think that the tumor has to be infected with CMV for any antiviral drug to benefit the patient. However, research in my laboratory in the last couple of years suggests that Valcyte may have both an antiviral effect and a true antitumor effect regardless of the viral infection. Thus, since we think most tumors are infected with CMV, it’s probably not even important to get a CMV test of your tumor since the clinical results and patient suggests that it works regardless.

2. What if my neuro- oncologist doesn’t want to prescribe Valcyte?

This is a difficult question. The neuro- oncologist should really be the quarterback taking care of the brain tumor patient. Any physician can prescribe any medication under their discretion. This is called off label use of the medication. If you can convince your neuro- oncologist that you’re interested in receiving Valcyte based on preliminary findings as demonstrated in the New England Journal of Medicine letter in 2013 (N Engl J Med. 2013 Sep 5;369(10):985-6.), then that may be all you need for them to prescribe you the medication. Other patients have sought out other physicians such as their primary care physicians for this prescription.

3.  Valcyte can cost thousands of dollars. What if my insurance doesn’t cover the cost?

This is also difficult question. Until appropriately designed clinical trials convince the FDA and the insurance industry that this drug is beneficial for brain tumor patients, it is unlikely that insurance companies will eagerly provide the drug for patients. In some countries, compassionate use programs are available through the government. Some patients seem to have also been able to get generic forms of Valcyte from Canada and Italy that may be less expensive.

4.     What are the risks of Valcyte?

In the Swedish patients who receive Valcyte in addition to radiation and temozolomide for glioblastoma, there were no significant severe side effects. The generic side effects of Valcyte include diarrhea, dizziness, drowsiness, and potentially important side effect of decreased white blood cell count. The white blood cells that are usually affected by Valcyte are not the same as those that are typically affected by temozolomide so there are not usually synergistic side effects. Platelet levels may also affected by Valcyte. Routine monitoring of blood counts is important for patients on Valcyte. Any one taking Valcyte should be monitored by their physician or side effects.

5.  What is the recommended dosing for Valcyte?

The study in Sweden used a loading dose followed by a maintenance dose. The loading dose was 900 mg orally twice a day for three weeks, followed by 900 mg a day. Valcyte tablets come in 450 mg tablets so that would be 2 tablets in the morning and in the evening for 3 weeks and then 2 tablets in the morning thereafter every day.

6.  When will we know whether or not Valcyte works for glioblastoma definitively?

I am pleased that a group of investigators working through the Radiation Ttherapy Oncology Group (RTOG) has apparently had success in their interaction with Roche/Genentech. I heard from officials at Roche/Genentech that they will support a large prospective randomized trial for newly diagnosed glioblastoma. It is hard to get more information at this point on this potential trial and I am hopeful that it will occur in the next year or so.

7.  Does Valcyte work for recurrent and newly diagnosed glioblastoma?

It’s difficult to answer this without any really good data. My suspicion based on laboratory work in my own group suggests that the drug may have a beneficial interaction with radiation treatment. That being said, it may still work for recurrent tumors as well.

8.  What are the next steps in terms of antiviral therapy for glioblastoma?

I have recently moved to Seattle and now I am the Director of the Ben and Catherine Ivy Center for Advanced Brain Tumor Treatment at the Swedish Neuroscience Institute. Among the many responsibilities I have here, I plan to do the following within the next year:

a. Develop a sensitive, standardized immunohistochemical test and become a reference lab to test tumor specimens for CMV infection

b. Initiate clinical trials for testing Valcyte in glioblastoma patients alongside other institutions throughout the country.

c. Continue research into the mechanism of action of antiviral drugs. My group has recently begun to investigate the role of other antiviral drugs in addition to Valcyte that may have efficacy in brain tumors.  We recently published a paper in the journal of Clinical Cancer Research showing that one such compound has the most potent antitumor effects ever tested in the UCSF mouse model except for possibly temozolomide (Clin Cancer Res. 2013 Dec 1;19(23):6473-83). We hope to have derivatives of this compound in human clinical trials in the next few years.

d. I am currently funded by the NINDS and NCI (at the NIH) under research grants (RO1 grants), to study the mechanisms of CMV infection in cancer biology and potential antiviral drugs in glioblastoma. I am working with an international group of investigators to continue to explore and push this research forward. I am organizing a meeting in Cambridge, Massachusetts to bring all of these investigators together to develop new increased collaborations in order to further our understanding of the role of CMV in cancer and cancer therapy.

e.  I am actively seeking funding from philanthropy and other organizations to foster this orphan area of research, which I believe may have monumental return on investment.  If you have any questions about this research please contact me through the Ivy Center for Advanced Brain Tumor Treatment in Seattle, WA at 206-320-2300.   If you would like to assist in funding this initiative, please email Davida Pennington at the Ivy Center at

Thank you. Sincerely,

Charles S. Cobbs, MD


Disclaimer: This is a personal blog. Any views or opinions represented in this blog are personal and belong solely to the blog author and do not represent those of people, institutions or organizations that the owner may or may not be associated with in professional or personal capacity, unless explicitly stated.  All content provided on this blog is for informational purposes only. The owner of this blog makes no representations as to the accuracy or completeness of any information on this site or found by following any link on this site. The owner will not be liable for any errors or omissions in this information nor for the availability of this information. The owner will not be liable for any losses, injuries, or damages from the display or use of this information. These terms and conditions of use are subject to change at anytime and without notice.

All content provided on this blog is for informational purposes only. The owner of this blog makes no representations as to the accuracy or completeness of any information on this site or found by following any link on this site. The owner will not be liable for any errors or omissions in this information nor for the availability of this information. The owner will not be liable for any losses, injuries, or damages from the display or use of this information.