Brain tumors: recent advances in diagnosis and classification.

In this issue of the ARCHIVES, 4 significant areas of advancement in the clinical diagnosis and classification of brain tumors are highlighted: (1) newly recognized types of mixed glioneuronal tumors, with which the surgical pathologist should be conversant; (2) recently introduced immunohistochemical markers of practical utility in brain tumor diagnosis and prognosis; (3) the current status of our understanding of deletions of chromosomes 1p and 19q in the molecular diagnosis of oligodendroglioma; and (4) a review of state-of-the-art current and emerging neuroimaging techniques for central nervous system (CNS) tumors and other CNS diseases.

NEW ENTITIES

The unparalleled complexity of the CNS is mirrored by the ever-increasing diversity of recognized neoplastic entities that can afflict the organ. (1) In many instances, the emergence of "new" entities is a reflection of our increasing ability to parse large categories of tumors into more refined, clinically and biologically meaningful, subtypes for purposes of patient prognosis and/or treatment stratification. In this special section, Drs Edgar and Rosenblum provide an example in their description of newly recognized variants of mixed glioneuronal tumor. An awareness of and familiarity with the clinicopathologic features will permit the general surgical pathologist to recognize and accurately diagnose these tumors.

NEW IMMUNOHISTOCHEMICAL MARKERS FOR DIAGNOSIS AND PROGNOSIS

As in all subspecialty areas of surgical pathology, there is a constant stream of new immunohistochemical markers introduced into the literature on virtually a monthly basis. Initial excitement and enthusiasm for many promising markers is often tempered over time as the weight of additional multi-investigator, multi-institutional experience provides needed critical evaluation of the true utility, robustness, and specificity, or lack thereof, of new markers. Drs Takei, Bhattacharjee, Rivera, Dancer, and Powell provide an overview of several immunomarkers that have passed the initial test and stirred continuing interest, thus warranting additional attention. Two of these markers merit comment. The first is the use of antibodies directed against the INI1/hSNF5/SMARCB1/BAF47 gene product in the diagnosis of rhabdoid tumor (RT) and atypical teratoid/rhabdoid tumor (AT/RT). Deletion or mutation of INI1, the genetic signature of RT and AT/RT, is reflected by loss of the gene's protein product and the attendant loss of immunoreactivity. Thus, normal indigenous cellular constituents that serve as internal positive controls, such as endothelial cells and reactive inflammatory cells, as well as nonrhabdoid tumor mimics, such as medulloblastoma, exhibit positive nuclear immunoreactivity, whereas RT and AT/RT tumor cell nuclei are uniformly negative. The introduction of an anti-INI1 diagnostic antibody test provides an immunohistochemical surrogate for confirmatory gene deletion/mutation molecular testing. This is significant in light of the fact that routine immunohistochemistry is more widely available than specialized molecular testing, particularly in community hospitals and even more so globally. Thus, the INI/BAF47 story illustrates a paradigm in which clinically significant discoveries made at the bench, prominently including those flowing from basic molecular biologic and genomic investigation, are ultimately translated into practical, widely accessible clinical laboratory tests.

A second group of recently introduced clinically applicable immunoreagents warranting brief comment are the new proliferation markers, specifically mitotic figure immunostains. (2) These markers show promise for greatly expediting rapid quantitation, through manual microscopic evaluation or by automated instrumentation, of mitotic activity in a more accurate and objective manner. Potential clinical applications of mitotic figure immunostains are certainly not limited to neuro-oncologic pathology and will likely be of interest to all surgical pathologists who deal on a daily basis with tumors in which the degree of cellular proliferation, and in particular the mitotic index, has been demonstrated to be of clinical diagnostic or prognostic importance.

[FIGURE OMITTED]

MOLECULAR DIAGNOSIS IN ONCOLOGIC NEUROPATHOLOGY COMES OF AGE

In the field of CNS neoplasia, as in most areas of surgical pathology, the impact of molecular and genomic approaches on tumor diagnosis and classification is ascendant. (3) Examples within the sphere of neuro-oncology include the assessment of INI1 gene deletion/mutation in RTs discussed previously, O(6)-methyguanine-DNA methyl-transferase (MGMT) gene silencing in glioblastoma, (4) and, most prominently, deletion status evaluation of chromosomes 1p and 19q in oligodendroglial tumors. In this special section, Drs Aldape, Burger, and Perry provide AR-CHIVES readers with a concise review and update on this critical and rapidly evolving subject.

ADVANCES IN NEUROIMAGING--A CENTURY AFTER ROENTGEN 

Pathologists and radiologists share many commonalities, and our subspecialties have in many ways evolved in parallel. Contemporary neuroradiology, more than ever, provides us with a surrogate detailed look at the gross anatomy of the diseased brain that a century ago was possible only at the autopsy table. Roentgen's discovery of x-rays in the early 1900s made possible the first primitive visualization of brain tumors and other CNS mass lesions via plain skull radiographs, in which the anatomic location and relative size of a space-occupying mass could be indirectly surmised through observation of the degree and direction of displacement of a midline neuroanatomic landmark, the pineal gland, which, fortuitously for the imagers and clinicians of the era, undergoes progressive calcification with increasing age. (5) Subsequent advances in neuroimaging that followed include ventriculography in 1918, pneumoencephalography in 1919, and arteriography in 1927. (6) Pneumoencephalography was only supplanted as the prevailing neuroimaging technique in the 1950s and 1960s by cerebral angiography. The most salient advance, however, was the introduction of computerized axial tomography scanning in 1971, which was followed very quickly by nuclear magnetic resonance imaging (now magnetic resonance imaging) in the late 1970s and early 1980s.6 For those readers who may not have followed the progress in neuroimaging techniques beyond basic magnetic resonance imaging, the article in this special section by Drs Debnam, Ketonen, Hamberg, and Hunter provides a thoroughly contemporary review and update. Examination of magnetic resonance imaging fluid-attenuation inversion recovery sequences has already become a routine procedure, and fluid-attenuation inversion recovery is the sequence of choice for subtle, noncontrast-enhancing lesions, such as gliomatosis cerebri, in which its advantages include exquisite sensitivity and lack of the background ventricular and subarachnoid space cerebrospinal fluid signal "noise" seen in T2-weighted sequences. With respect to advanced techniques, proton magnetic resonance spectroscopy (MRS for short or simply "spect" in common parlance) provides an aid in shaping the differential diagnosis, as, for example, in many cases of radiation necrosis versus recurrent tumor, and also in "guiding the surgeon's hand" (whether real or prosthetic "stereotactic hand") toward the most likely area of highest grade tumor for biopsy. Beyond these 2 imaging techniques, there exists a growing armamentarium of additional noninvasive imaging modalities; for example, cerebral arteries and veins can be visualized without the need for administration of iodinated contrast material or radiation (magnetic resonance angiography and venography), and serial computed tomography reconstruction techniques permit the rendition of 3-dimensional images that can be rotated to provide the radiologist, surgeon, and pathologist with a detailed view of the relationship of the detailed bony features of the skull and spine with the vasculature and the anatomic alterations in them arising secondary to tumor or other pathologic processes. Debnam and colleagues elegantly describe and illustrate the advanced techniques that are now in current daily clinical usage. As for the future, more remarkable imaging methodologies are under active investigation, such as diffusion tensor imaging fiber tractography, which allows the noninvasive visualization of individual major white matter fiber tract pathways of the brain and the physical distortions and alterations in them produced by disease (Figure).

SUMMARY

Oncologic surgical neuropathology is a dynamic field. One message that emerges from the articles comprising this ARCHIVES special section is unambiguous: Advances in multiple aspects of brain tumor diagnosis and classification are coming steadily and quickly--stay tuned! It is hoped that the following articles convey the sense of enthusiasm and optimism that we, as clinicians and clinician-scientists, feel about the future of our efforts to understand and combat neoplastic disease of the CNS.

References

(1.) McLendon RE, Rosenblum MK, Bigner DD, eds. Russell & Rubinstein's Pathology of Tumors of the Nervous System. 7th ed. London, England: Hodder Arnold; 2006.

(2.) Tapia C, Kutzner H, Mentzel T, Savic S, Baumhoer D, Glatz K. Two mitosis-specific antibodies, MPM-2 and phospho-histone H3 (Ser28), allow rapid and precise determination of mitotic activity. Am J Surg Pathol. 2006;30:83-89.

(3.) Fuller GN. Molecular classifications. In: Barnett GH, ed. Current Clinical Oncology: High-Grade Gliomas: Diagnosis and Treatment. Totawa, NJ: Humana Press; 2006:37-42.

(4.) Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med. 2005;352:997-1003.

(5.) Fuller GN, Burger PC. Central nervous system. In: Mills SE, ed. Histology for Pathologists. 3rd ed. New York, NY: Raven Press; 2006:273-319.

(6.) Leeds NE, Kieffer SA. Evolution of diagnostic neuroradiology from 1904 to 1999. Radiology. 2000;217:309-318.

Accepted for publication October 12, 2006.

Gregory N. Fuller, MD, PhD

From the Department of Pathology, Section of Neuropathology, The University of Texas M. D. Anderson Cancer Center, Houston.

The author has no relevant financial interest in the products or companies described in this article.

Reprints: Gregory N. Fuller, MD, PhD, Department of Pathology, Section of Neuropathology, The University of Texas M. D. Anderson Cancer Center, Unit 085, 1515 Holcombe Blvd, Houston, TX 77030 (e-mail: [e-mail]).

Dr Fuller received his MD degree from Baylor College of Medicine and his PhD degree in neurochemistry from The University of Texas Graduate School of Biomedical Sciences in the Texas Medical Center. He completed his internship, residency, and neuropathology fellowship training at Duke University and joined the faculty of the Department of Pathology at The University of Texas M. D. Anderson Cancer Center in 1992. He is currently professor of pathology and chief of the Section of Neuropathology at M. D. Anderson and also serves as director of the Automated Image Acquisition and Tissue Microarray Core laboratories. Dr Fuller is actively engaged in patient care as chief of neuropathology and has an active research program centered on the identification of novel targets for brain tumor therapy and elucidation of the molecular pathology of the diffuse gliomas. He has published more than 200 peer-reviewed research papers and book chapters and is the coauthor of 2 textbooks, Practical Review of Neuropathology (Fuller and Goodman, 2001) and Genomic and Molecular Neuro-Oncology (Zhang and Fuller, 2004). Dr Fuller is an associate editor of the ARCHIVES.