Commentary: Classification and grading of pituitary tumors. Observations of two working neuropathologists

As part of the World Health Organization (WHO) ‘‘blue book’’ series, the recently updated endocrine volume includes a 38-page chapter on pituitary tumors [3], divided into concise multi-authored sections. New developments, remaining shortcomings, and future directions of research have been summarized by Al-Shraim and Asa in the current issue [1]. Overall, this newest ‘‘blue book’’ is superbly written, provides a wealth of useful information, and is sold at a readily affordable price, making it a bargain to anyone even remotely interested in pituitary pathology. As with other issues of the WHO series, current molecular data are well integrated with classic pathology and the entities are beautifully illustrated. The pituitary falls under the purview of both endocrine and neuropathology. However, given its intracranial location, the gland is biopsied by neurosurgeons. Therefore, despite crossing subspecialty boundaries, neuropathologists must be well acquainted with this area of endocrine pathology. In the current summary, we attempt to represent the working neuropathologist’s perspective. Pituitary adenomas can be classified on the basis of their gross/radiographic features (macrovs microadenoma, invasive vs noninvasive), hormone status (functional vs nonfunctional), cytology (acidophil, basophil, chromophobe), grade (typical, atypical, carcinoma), immunohistochemical/ultrastructural cell type (lactotroph, somatotroph, corticotroph, gonadotroph, thyrotroph), and molecular attributes (transcription factor, growth factor, and or tumor suppressor gene expression). Practicing neuropathologists are also aware of the wide morphological spectrum encountered routinely. Depending upon the histological pattern, particularly in nonfunctioning adenomas, one can consider diagnoses such as ependymoma, oligodendroglioma, meningioma, or even carcinoma, particularly if one is unaware of the sellar origin of the tissue sample. Despite the limitations of classifying adenomas by their tinctorial characteristics, as discussed in the introductory chapter of the WHO book [3], a more complete discussion of helpful morphological features is worthy of re-emphasis. For example, calcospherites (prolactinomas), eosinophilia or fibrous bodies (growth hormone adenomas), paranuclear vacuolation (acidophil stem cell adenoma), cytoplasmic basophilia or Crooke’s hyaline (corticotroph adenomas), pseudorosette formation (gonadotroph adenomas), and oncocytic change (gonadotroph or null cell adenomas) serve as clues to tumor subtype. Synaptophysin immunoreactivity was also emphasized as a more consistent broad spectrum neuroendocrine marker for pituitary adenoma than chromogranin [3]. This distinctive immunophenotypic feature is generally under appreciated. As common as pituitary adenomas are, it is surprising to see the current inter-institutional variation in diagnostic approach. This ranges from a single hematoxylin and eosin section and a final diagnosis of ‘‘pituitary adenoma’’ in virtually every case, to the ‘‘Cadillac approach’’ of a full pituitary hormone immunostain panel in nearly all instances and electron microscopy in selected cases. An interesting point highlighted in the review of Al-Shraim and Asa [1] is the histogenetic link between transcription factor expression in normal development and pituitary adenomas. Work in this area [4] has advanced our understanding of plurihormonal adenomas, such as the coexpression of growth hormone, prolactin and thyrotropin in the majority of acromegalyassociated adenomas, as well as the gonadotroph-like properties of so-called ‘‘null cell’’ adenomas. NonetheA. Perry (&) Division of Neuropathology, Washington University School of Medicine, Campus Box 8118, 660 S. Euclid Ave, 63110-1093 St. Louis, MO E-mail: aperry@wustl.edu Tel.: +1-314-3627426 Fax: +1-314-3624096