Three patients with progressive multifocal leukoencephalopathy (PML) treated with highly active antiretroviral therapy (HAART) worsened clinically and radiologically. At the time of deterioration all three had reduced HIV viraemia and increased CD4 cell counts. Brain biopsy in all three disclosed PML and marked perivascular lymphoplasmacytic infiltration. We reviewed the slides of 28 brain biopsies diagnostic of PML. Inflammatory changes were observed in four out of nine patients on HAART and in one out of 19 patients not on HAART.

Objectives and experimental design Cerebella of young adults, elderly adults, and patients with Alzheimer's disease (AD) (with and without cerebellar amyloid deposits) were studied by Golgi staining and glial fibrillary acid protein (GFAP) immunocytochemical methods. Observations Three subtypes of Golgi epithelial cells and nine subtypes of stellate neuroglia (both normal and hypertrophic) were defined by their morphology, their GFAP‐reactivity, their specific location in the cortical layers, and their responses in senility and AD. The GFAP immunoreaction was subtype specific. In aged and AD cerebella, different morphological and GFAP‐immunoreactive subtype‐specific changes were observed: in the white matter, the subtypes were always GFAP‐immunopositive, but in the grey matter some astroglial subtypes showed a variable or no increase in GFAP staining. The astrocytes at the limits of the granule cell layer showed more and longer processes. Variations were seen in one or more folia, involving one or more subtypes and affecting different numbers of cells of each subtype. No clear differences were seen in glial reactivity between beta‐amyloid positive and β‐amyloid (Aβ) negative AD cerebella. No important relationships were found between Aβ deposits. In aged and AD cerebella, different subtypes expressed new proteins (APP, calretinin). Conclusions The existence of different glial subtypes in different locations suggests they have different functions. General and local variations in these subtypes suggest that both general and local induction factors must also exist. The responses of glial cells to as‐yet undefined stimuli might lead to general or local neuronal changes important in senility and the pathogenic course of AD. GLIA 2015;63:287–312

Direct and productive infection of neurons in vivo is still a matter of debate, although in vitro experiments have demonstrated that immature neuronal cells can be productively infected by various human immunodeficiency virus (HIV) strains. To address this controversy we have analyzed, using light microscopy and in situ hybridization (ISH), HIV-1 infected cells in brain tissue from four pediatric cases of HIV-1-associated encephalopathy (EP). HIV-1 RNA-expressing cells--therefore, actively infected cells--were detected by ISH in different amounts in all brain specimens from the four children. They mainly correspond to glial cells. However, in two of the four children, who had severe progressive EP, but not in the other two, who had the static form, HIV-1-infected neurons were clearly observed in the cortical brain samples. These results provide initial evidence that HIV-1 can actively infect neurons in vivo in children and show a cortical involvement of HIV brain infection in clear correlation with the clinical EP symptoms.

Squash cytology (SC) is a very useful procedure during neurosurgical intraoperative consultation (IOC), and it is especially recommended for the evaluation of soft tumors or tumors that are highly cellular (just the characteristics of pediatric central nervous system [CNS] tumors). The aim of this review is to familiarize pathologists with the range of cytomorphologic appearances that can occur during IOC for pediatric CNS tumors and with the diagnostic dilemmas and pitfalls encountered in this setting. This article is based on the medical literature and the authors' experience with a large series of cases accrued over a 12-year period at 3 institutions. SC is a specially recommended procedure in IOC for pediatric CNS tumors; it reveals the fine cellular details and background features in a manner not seen in corresponding frozen sections. Indeed, a differential diagnosis between histologically look-alike processes can be achieved with more confidence if SC is employed. Cancer (Cancer Cytopathol) 2015;123:331-46.

No abstract