The pathology of primary CNS lymphoma is always one of diffuse histology since there are no lymph node structures within the central nervous system. Although we see all subtypes of non-Hodgkin’s lymphoma within the brain, the prognostic significant of the histologic subtypes do not appear to be as significant or as important as they are for the subtypes with systemic lymphomas. Although pure T-cell primary CNS lymphomas have been reported occasionally in the literature, the vast majority of primary CNS lymphomas tend to be of B-cell origin. Finally, Epstein-Barr virus can be found in almost 100% of primary CNS lymphoma cells in immunocompromised patients but interestingly, almost never found in the lymphoma cells in the immunocompetent patients, again suggesting that there is probably a difference in the natural history and the biology of CNS lymphoma in these two patient populations. When you actually look at the pathophysiology of primary CNS lymphoma we can see that the vast majority of patients present as focal lesions. Though other types of presentations are possible, including a more diffuse type of picture, uveal presentation and leptomeningeal or intramedullary presentation. No matter how this tumor presents however, this is always a diffusely infiltrative disease and is never a localized process. This tumor spreads along the CSF pathways, although interestingly even at times of advanced disease and at time of death of patients with widespread cranial spinal disease, we rarely see this tumor outside the central nervous system, probably less than 10% of the time. Just to give you an idea of the infiltrative and diffuse nature of this disease, this is a high-power section of the brain of a patient with a CNS lymphoma. You can see these lymphoma cells throughout the cerebral cortex and spreading along the pia mater and into the Virchow-Robin spaces of the brain. When you actually look at a whole mount of a brain of a patient who died of a CNS lymphoma you can see the diffuse infiltrative nature of this disease, really in every component of the brain.

In order to get a better idea about the natural history and the characteristics of this disease, we actually did a significant literature review where we reported the characteristics of every case of primary CNS lymphoma reported in the literature between 1980 and 1993 in patients who were both immunocompetent as well as AIDS patients. This group of patients encompassed almost 800 immunocompetent patients and over 300 HIV-positive patients. We saw several significant differences. The first one of course is the male to female ratio, with that being much higher obviously in the HIV population, as well as the fact that the HIV population tended to be younger when they developed the disease. Again, I think this reflects more of the epidemiology of AIDS, at least in the period of 1980 to 1993 than it necessarily represents a statement about the biology of primary CNS lymphoma. We also looked at the clinical presentation. One of the interesting differences that we saw is that patients with AIDS tended to present much more with mental status changes and seizures as opposed to patients who were immunocompetent tended to present much more with more focal types of symptoms, suggesting that the disease in patients with AIDS tended to be a much more diffuse presenting disease as opposed to patients who were immunocompetent, who presented more like patients with other types of primary CNS tumors.