Scientists have found that treating medulloblastomas in experimental mice using an antiviral drug and COX-2 inhibitor effectively slows tumor growth. The Karolinska Institutet-led team hit on the therapeutic approach following the observation that the majority of both primary medulloblastomas and medulloblastoma cell lines were persistently infected with human cytomegalovirus (HCMV). The virus induces COX-2 expression and production of the COX-2 PGE2, both of which are present at high levels in medulloblastomas and stimulate tumor cell proliferation.
Their results suggest that rather than representing an epiphenomenon in medulloblastoma progression, HCMV may play a pathogenic role, and represent a true therapeutic target for slowing tumor growth. Cecilia Söderberg-Nauclér, M.D., Ninib Baryawno, M.D., and collagues describe their findings in Journal of Clinical Investigation in a paper titled “Detection of human cytomegalovirus in medulloblastomas reveals a potential therapeutic target.”
HCMV DNA and proteins are found in a range of tumors, but the virus is not considered to be oncogenic, and its potential role in cancer appears to be oncomodulatory, the researchers report. A number of tumors, including medulloblastomas, in addition express COX-2, and inhibiting PGE2 synthesis using NSAIDS has been found to induce apoptosis in medulloblastoma cell lines and markedly inhibit the growth of medulloblastomas in vivo. Clinical trials have in addition suggested that COX-2 inhibition represents a promising anticancer approach.
A link between HCMV and oncogenesis can, however, be made in consideration of the fact that HCMV protein US28 induces COX-2 expression and results in STAT3 phosphorylation, which increases production of VEGF and IL-6, as well as tumor formation in vivo, the authors continue. Indeed, COX-2 inhibitors prevent viral production, while transgenic mice that express US28 develop tumors.
The researchers therefore investigated the prevalence of HCMV in medulloblastomas, and designed a series of studies to evaluate the connection between HCMV and COX-2 expression in tumors, and see whether blocking viral replication using antiviral and COX-2 inhibitors would affect tumor growth in vitro and in vivo.
Strikingly, 92% of 37 primary medulloblastomas examined expressed immediate-early (IE) HCMV protiens, and 73% expressed late proteins. In situ hybridization confirmed the presence of HCMV in primary medulloblastomas, and cytometric analysis demonstrated expression of HCMV IE and UL83 proteins in 90%–95% of the cells. All six available frozen tissue samples were also found to contain HCMV IE and other proteins, and HCMV protein-related DNA and RNA were identified in each of the eight human medulloblastoma cell lines tested.
Given the persistence of viral DNA, the researchers postulated that HCMV might reside in cancer stem cells. Staining medulloblastoma cell lines specifically for stem cell markers and HCMV proteins showed that 57% of the stem cells co-expressed HCMV IE proteins. Implantation of cells from one of these lines into experimental mice led to increased HCMV protein expression, and most of the tumor cells were positive for HCMV IE proteins.
Studies in frozen tissue sections from primary medulloblastomas and in cell lines confirmed that HCMV protein expression correlated with COX-2 expression in tumor cells, and that both were expressed in human medulloblastoma xenografts in mice. “These observations imply that HCMV may control the expression of COX-2 in medulloblastomas,” the authors remark.
To investigate this possibility, the team infected medulloblastoma cells with HCMV. This resulted in increased IE RNA expression within three hours, which continued over 6–72 hours. The late HCMV PP150 RNA also increased greatly by 24 hours post infection. “Interestingly, expression of COX-2 mRNA was already increased at one and three hours after infection in cells infected with both HCMV- and UV-treated virus inoculums, before induced IE expression was detected,” the researchers note. The COX-2 protein itself was also detectable at 3 hours after infection, while, significantly, US28 mRNA (US28 induces COX-2 production) was detectable within 15 minutes of infection. At three days post infection, HCMV had significantly increased PGE2 synthesis, which could be significantly reduced by treating the cells with ganciclovir, celecoxib, or both.
The presence of HCMV in the majority of medulloblastoma tumors, combined with evidence that viral mechanisms control COX-2 expression, prompted studies to test whether antiviral drugs and/or COX-2 inhibitors could slow tumor growth. The researchers used ganciclovir (which targets the DNA polymerase dring HCMV replication), and/or the COX-2 inhibitor celecoxib, which blocks HCMV replication by decreasing PGE2 levels.
In vitro studies on native medulloblastoma cells confirmed that administering increasing concentrations of ganciclovir did indeed reduce the clonogenic capacity of different cell lines. In contrast, the drug had no effect on the clonogenic capacity of cell lines derived from other cancers that didn’t harbor HCMV. Celecoxib also reduced the clonogenic capacity of two of the three medulloblastoma cell lines, while combining this drug with valganciclovir was effective in all three cell lines, and to a greater extent than either drug alone. Again, neither celecoxib or valganciclovir alone, or in combination, affected the survival of cancer cell lines that didn’t carry HCMV.
Further evaluation of the more long-term effects of drug treatment in one of the medulloblastoma cell lines showed that after two weeks of continuous treatment with ganciclovir, viral protein expression diminished substantially, but neither HCMV protein expression nor DNA was completely eliminated.
Clonogenic assays with HCMV-superinfected medulloblastoma cells further showed that treatment with increasing concentrations of ganciclovir or celecoxib significantly reduced tumorigenic capacity in a dose-dependent manner. The effect was boosted further by combining the two drugs, and resulted in an 81–97% reduction in growth. “Consistent with these observations, valganciclovir and celecoxib alone or in combination also reduced PGE2 production," the authors state.
A human medulloblastoma xenograft model was then established in NMRI nu/nu mice. Animals were given daily treatment for 12 days with oral valganciclovir or oral celecoxib, once their tumor volume had reached 0.1 mL. Encouragingly, each drug inhibited the growth of established medulloblastomas by approximately 40%, and using the two drugs in combination reduced tumor growth by 73%. Combined therapy also reduced the number of clusters positive for HCMV late protein in xenograft tumors by 80%, the team notes. Significantly, xenografts of two HCMV-negative tumor cell lines were unaffected by oral valganciclovir therapy.
The Karolinska team claims its findings imply there is a persistent source of HCMV in neuroprecursor cells, and indicate that the virus plays a pathogenic role in medulloblastoma. “Importantly, our findings also suggest HCMV as a novel therapeutic target for this tumor,” they stress. “We speculate that HCMV infection of a defined tumor precursor cell is a decisive early event in the disordered cerebellar development that leads to medulloblastoma through HCMV-specific effects on tumorigenesis, inflammation, and immune evasion strategies…Since COX-2 inhibitors efficiently prevent HCMV replication and reduce the growth of US28-expressing tumor cells, the ability of COX-2 inhibitors to protect against cancer in clinical trials may be mediated by control of HCMV replication.”
Targeting HCMV isn’t being suggested as a curative approach, however, professor Söderberg-Nauclér stresses. “The virus infection isn’t cured by the treatment, nor is the tumor, but the virus in the tumor decreases, which affects its growth. This therefore presents a new approach to treating tumors and could henceforth be used as a possible complementary therapy.”