Enormous advances have been achieved in recent years in our understanding of the molecular basis of cancers, particularly those affecting the nervous system. Many of these advances have been propelled by sequencing of the human genome of all major cancers, greater availability of high-throughput analytic techniques, and large, publically available databases such as that created by The Cancer Genome Atlas (TCGA). The TCGA was pioneered in the study of glioblastoma and aims to provide a comprehensive, multidimensional picture of all major human cancers. More recently, the whole spectrum of diffuse gliomas (astrocytomas and oligodendrogliomas grades II–III) were also studied under the umbrella of “lower grade glioma.” Many of these studies have generated increased complexity and vast amounts of data which, although comprehensive, make it challenging to identify key markers for molecular diagnosis and therapeutic targeting. A major challenge resides in incorporating this molecular data in the context of tissue-based techniques that have resided in the realm of neuropathology, including routine histologic examination and immunohistochemistry. In the current text, we describe the status of biomarkers applicable to the pathology of neoplastic disorders of the brain. First, an overview on the current status of biomarker analysis in brain tumors will be provided, and a discussion on how major scientific discoveries are reshaping the current practice of neoplastic neuropathology. Descriptions of the current techniques available for the clinical and experimental evaluation of biomarkers in brain neoplasms are presented, including the classic techniques of immunohistochemistry and in situ hybridization which have provided enormous assistance in diagnosis and prognostication of brain tumors for over a decade now. High-throughput molecular techniques, array-based methods, methylation profiling, next-generation sequencing, and practical gene panels are finding increasing applications in neuropathology practice as robust biomarker tests. As no single technique provides a complete picture of neoplasms in the particular patient, incorporation of multiple biomarkers in the development of molecular subgroups with biologic and therapeutic relevance are discussed, an approach increasingly applied to the study of brain cancer.

As important as these techniques are, optimal preservation of tissue balancing clinical and research needs is an equally important topic that is covered on a section on biobanking. Using tissue obtained in clinical settings also raises important ethical considerations, including tissue ownership and incorporation of patients and their families in decision making. These issues and institutional guidelines are covered. Finally, the major categories of neoplastic disorders involving the nervous system are discussed, with emphasis on diagnostic, prognostic, and predictive biomarkers that are in current use in the pathologic evaluation of brain tumors. Biomarkers resulting from major scientific breakthroughs and that have withstood the test of time in neuropathology practice will represent the primary focus, including IDH1 mutations (diffuse gliomas), 1p19 co-deletions (oligodendroglial tumors), MGMT promoter methylation (glioblastoma), BRAF alterations (pediatric gliomas), and molecular subgrouping (medulloblastoma).

Broadly speaking, biomarkers used in neoplastic neuropathology and oncology may be classified into three different types: diagnostic, prognostic, and predictive [1]. Specific biomarkers of course may be placed in more than one category. 1p19q co-deletion, for example, is one of the most robust biomarkers in neoplastic neuropathology and is used in routine diagnosis (i.e., diagnostic biomarker), since it provides strong evidence for oligodendroglioma in the evaluation of adult gliomas. It also identifies the subset of adult diffuse gliomas with the best outcome at the present time (i.e., prognostic biomarker) and also identifies a subgroup of diffuse gliomas with increased response to Procarbazine, CCNV, Vincristine (PCV) chemotherapy and irradiation (i.e., predictive) [2, 3].

New biomarkers are proposed on a daily basis on the consistently evolving scientific literature. However, only a handful stand the test of time and become standard of care despite ever increasing requests from the clinical neuro-oncology community. Rating scales taking into account the relative value of prognostic biomarkers based on their diagnostic, prognostic, and predictive characteristics using end points such as their impact on survival, quality, and cost of care have been developed [4, 5]. The National Comprehensive Cancer Network (NCCN) gathered a task force to assess the utility of selected markers in six malignancies, including brain cancer [6]. In brief, for a specific biomarker to justify inclusion into daily care of oncology patients, strict validation must demonstrate analytic validity and clinical utility [6]. Ranking by strength of evidence may be formulated taking into account these variables: category 1 (high-level evidence, uniform NCCN consensus), category 2A (lower-level evidence, uniform NCCN consensus), category 2B (lower-level evidence, NCCN consensus), and category 3 (any level of evidence, but no NCCN consensus)[6].

Guidelines for discovery studies of new biomarkers are critical in this regard, given the wide variety of methodologies used. For example, Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) were outlined following NCI-EORTC recommendations [7]. These guidelines recommend the inclusion of key information in these studies to assess the significance of biomarker reporting in the scientific literature, including attention to population studied and analytical and statistical methods.