We agree with Michaelis et al and sympathize with their frustration that the role of cytomegalovirus (CMV) in glioblastoma has remained controversial and that the field would benefit from a collaborative interdisciplinary approach. For these reasons, we have previously organized several international meetings of experts in virology, neuro-oncology, and pathology to evaluate all of the data in a nonbiased and critical way and try to come up with a consensus on the role of CMV in malignant glioma. Nevertheless, the field continues to languish in a futile cycle of positive and negative reports that leave the scientific community understandably skeptical. It should be possible in this era to prove unequivocally that CMV nucleic acids and proteins are present in glioblastoma cells in vivo. Indeed, several publications have done this with various methodologies. Unfortunately, each new “negative” study seems to question previous works that appeared definitive.
Perhaps it is time that we in the neuro-oncology community clearly outline specific questions related to CMV and glioblastoma multiforme which, if answered, will universally be agreed upon as definitive. Such agreed-upon results will then serve as an objective platform from which the field can move forward to answer other important questions such as “Is CMV an important target for immunotherapy?” and “Does CMV infection initiate and/or promote gliomagenesis?” Perhaps the neuro-oncology community could support a meeting to come up with such criteria?
In the meantime, we should not hold up any possible clinical trials that may allow antiviral strategies to be utilized for glioblastoma given the possible impact these could have on patient survival. Indeed, while they are very small studies, 2 of the most promising studies and all of the literature for glioblastoma are based on Valcyte and dendritic cell immunotherapy to CMV pp65.