腦腫瘤的分子分類(lèi)指導(dǎo)個(gè)性化治療

高 華 趙 鵬 張亞卓*

(1.北京市神經(jīng)外科研究所細(xì)胞生物室中樞神經(jīng)損傷修復(fù)北京市重點(diǎn)實(shí)驗(yàn)室首都醫(yī)科大學(xué)，北京100050;2.首都醫(yī)科大學(xué)附屬北京天壇醫(yī)院神經(jīng)外科，北京100050)

Brain tumors are the leading cause of cancer death in children，the second most common cause of cancer in young adults，and account for a high proportion of deaths in older adults［1］.The most widely used histological system of brain tumor typing is the classification scheme of the World Health Organization(WHO)，in which tumors are classified according to histological features characteristic of the assumed cell of origin［2］.No variable predicts prognosis more precisely，and classification is also the basis on which clinicians make critical therapeutic recommendations to their individual patients:neuro-oncologists apply therapies in a relatively uniform way for all patients with a given tumor type.In the future，as specific therapies become based on individual biological alterations within tumors，precise classification will assume even greater importance to guide these distinct treatments.

1 Gliomas

Gliomas are histologically heterogeneous tumors often diagnosed from small tissue specimens，which may be prone to sampling error unless obtained using imaging techniques to guide biopsy or sampling during resection.Early tumor classifications relied on comparing tumor features with those of normal tissue.The WHO classification is based on histological features including cell clarity，mitotic activity，nuclear atypia，vascularity and necrosis.However，irrespective of the classification system used，the small sample size and subjective criteria means that for many gliomas diagnosis may be difficult and subject to inter-and intraobservervariation［3-4］.

1.1 Molecular biology of gliomas

The most widely accepted being，perhaps，the classification in which they are divided into primary or de no-vo and secondary or progressive.De novo gliomas are more common;they are observed in individuals older than 50 years，appear in a matter of weeks or months and，at the molecular level，do not harbor mutations in the p53 gene but do have amplification of the epidermal growth factor receptor(EGFR)gene［5］.Secondary glioblastomas are diagnosed in younger patients and，in some cases，there is evidence of progression from benign astrocytomas.At the molecular level，these secondary glioblastomas harbor mutations in the p53 gene.

Recently，Phillips et al.［6］proposed another classification of gliomas.These authors identified the molecular signature of 3 subsets of gliomas that were termed proneural，proliferative，and mesenchymal.In the proneural tumors，chromosome gains and losses are not evident，the phosphatase and tensin homologue(PTEN)gene is intact，and EGFR is not amplified;however，the notch signaling pathway is activated.

1.2 Predictive molecular diagnostics of gliomas

High-throughput genomic techniques will accelerate predictive individualized care.Screening for gene polymorphisms and loss of heterozygosity by single nucleotide polymorphism microarrays;analysing chromosomal gains and losses by comparative genomic-hybridization arrays;determining global patterns of methylation，acetylation and alternative splicing on microarrays;and identifying characteristic proteomic profiles，all will probably play a part in the new molecular diagnostics.Gene-expression profiling will probably be central to this effort.

The high frequency of signalling-pathway alterations in brain tumors，such as chronic phosphoinositide3-kinase pathway activation in malignant gliomas and chronic SHH signalling in medulloblastomas，enablesidentification of potential therapeutic targets for small molecule inhibitors.High-throughput screens of small-molecule inhibitors have generated many pathway/oncogene-specific drugs that could potentially be used for the treatment of patients with cancer.Morphologically identical tumors can be distinct in their mutational patterns，signalling-pathway alterations and gene-expression profiles，and，most importantly，in their response to arange of therapies.

In modern clinical neuro-oncology，histopathological diagnosis affects therapeutic decisions and prognostic estimation more than any other variable.Among high-grade gliomas，histologically classic glioblastomas and anaplastic oligodendrogliomas follow markedly different clinical courses.Unfortunately，many malignant gliomas are diagnostically challenging;these nonclassic lesions are difficult to classify by histological features，generating considerable interobserver variability and limited diagnostic reproducibility.The resulting tentative pathological diagnoses create significant clinical confusion.Gene expression profiling，coupled with class prediction methodology，prediction model was capable of classifying high-grade，nonclassic glial tumors objectively and reproducibly.Moreover，the model provided a more accurate predictor of prognosisin these nonclassic lesions than did pathological classification.Class prediction models，based on defined molecular profiles，classify diagnostically challenging malignant gliomas in a manner that better correlates with clinical outcome than does standard pathology［7］.

1.3 Distinctive molecular features and their prognostic values in gliomas

The vastmajority(93%)of low-grade diffuse gliomas carry at least one of the following genetic alterations:IDH1 mutation，IDH2 mutation，TP53 mutation，and 1p/19q loss.IDH1 mutations are a significant prognostic marker of favorable outcome in patients with glioblastoma and anaplastic glioma.The predictive role of IDH1 mutation in WHO grade II glioma remains to be established.Loss of 1p/19q is a well-recognized predictive marker in oligodendrogliomas.TP53 mutation is a significant prognostic marker for shorter survival in patients with low-grade diffuse glioma［8］.

EGFR(7p12)amplification is a hallmark of glioblastoma(GBM)，specifically primary tumors.About 50%of EGFR-amplified GBM express a ligand-independent truncated mutant variant，EGFRⅧ.The clinical importance of EGFR-activating mutations in GBM is not currently understood.The subsequent strong and persistent activation of downstream PI3K signaling provides advantages for cell survival，proliferation，and motility.Activation of the PI3K pathway is significantly linked to increasing tumor grade，decreased levels of apoptosis，and adverse outcome in human gliomas.

The DNA-repair enzyme methyl guanine-DNA methyl transferase(MGMT)gene，located on 10q26，has been a subject of interest owing to its association with response to alkylating drugs.About 40%of primary GBM and over 70%of secondary GBM display epigenetic MGMT silencing［9］

1.4 Changing treatment regimens based on biomarker assessment

Given the complex network of mechanisms and mol-ecules involved，one approach would be multitargetedtherapy，simultaneously aimed at different pathway constituents.The assessment of IDH，1p and 19q codeletions，and MGMT status could be built into a management algorithm for patients with anaplastic gliomas and glioblastoma［10］.Such algorithms might change quickly as new data and concepts emerge and might need to be adapted according to institutional preferences.Importantly，the decision for specific treatments must consider several issues such as patient preference，tumor location，target volume of radiotherapy，and potential comorbidities that might increase the risk of toxicity from therapy.

1.5 Glioma stem cells

The discovery of a highly tumorigenic subpopulation of stem-like cells embedded within fresh surgical isolates of malignant gliomas lent support to a new paradigm in cancer biology—the cancer stem cell hypothesis.At the same time，these“glioma stem cells”seemed to resolve a long-standing conundrumon the cell of origin for primary cancers of the brain.However，cancer stem cells are predicted to be difficult targets for cancer therapeutics because① they cycle slowly，②express high levels of drug export proteins，and③they may not express or may not depend upon the oncoproteins that are targeted by the new generation of“smart drugs，”such as Gleevec and Iressa.Collectively，these observations and considerations(and speculations)provide a fresh rationale for tumor resistance to therapy and suggest that a new class of agents should be designed to specifically target cancer stem cells［11］.The operational definition of glioma stem cell is a tumor subpopulation that can self-renew in culture，perpetuate a tumor in orthotopic transplant in vivo，and generate diversified neuron-like and glia-like postmitotic progeny in vivo or in vitro.In the case of malignant glioma，there are at least three neural cell types that could，in principle，serve as cell of origin for glioma stem cells.These are① mature“dedifferentiated”glia，②“restricted”neural progenitors that are normally unipotent，and ③ pluripotent neural progenitors［12］.Mutations originally induced in neural stem cells lie dormant and only trigger malignant transformation following differentiation into oligodendrocyte precursor cells［13］.

Do glioma stem cells arise from developmentally stalled neural progenitors such as the type B cell?Prediction 1:neural progenitors and brain tumor stem cells are driven by common signaling pathways;Prediction 2:signaling pathways that constrain the growth of normal progenitor cells are suppressed in brain cancers;Prediction 3:neural progenitors are competent for transformation by mutations found in human brain tumors;Prediction 4:brain tumors will initiate and/or cluster near the germinal centers of the brain;Prediction 5:genes that govern replication competence of neural progenitors will serve as‘gatekeepers’for development of CNS cancers.

2 Pituitary adenomas

Pituitary adenomas are categorized based on primary cell origin and type of hormone secreted.In the diagnostic approach to a suspected pituitary adenoma，it is important to evaluate complete pituitary function，because hypopituitarism is common.Despite their high prevalence in the general population，these tumors are invariably benign and exhibit features of differentiated pituitary cell function as well as premature proliferative arrest.Therapy for pituitary adenomas depends on the specific type of tumor，and should be managed with a team approach to include endocrinology and neurosurgery when indicated.

2.1 Clinical features

Pituitary adenomas present clinically in three ways:syndromes of hormone hypersecretion or deficiency;neurologic manifestations from mass effect of an expanding gland;or an incidental finding on imaging done for an unrelated issue.

2.2 The biomarker of invasion

Pituitary adenomas exhibit a wide range of behaviors.The prediction of aggressive or malignant behavior in pituitary adenomas remains challenging.The World Health Organization classification of endocrine tumors suggests that invasion of the surrounding structures，size at presentation，an elevated mitotic index，a Ki-67 labeling index higher than 3%，and extensive p53 expression are indicators of aggressive behavior.Nevertheless，Ki-67 and p53 labeling index evaluation is subject to interobserver variability，and their cutoff values are controversial［14］.

Fibroblast growth factors(FGFs)and their receptors(FGFRs)are a family of ligands and receptors that regulate development，growth，differentiation，migration，and angiogenesis.Pituitary adenomas have altered FGFR subtype and isoform expression.The normal human pituitary expresses mRNAs for FGFR1，2，and 3，including both Ig-like extracellular domains as well astransmembrane and kinase domains;by contrast，tran-scripts of only the first Ig-like domain of FGFR4 were found in the normal gland.Pituitary adenomas show two major alterations:loss of FGFR2，with resultant upregulation of MAGEA3［15］，and an N-terminally truncated cytoplasmic isoform of FGFR4，known as pituitary tumor derived(ptd-FGFR4).

Matrix metalloproteinases(MMPs)are a family of single-chain zinc-containing proteolytic enzymes that regulate the extracellular matrix in both physiological and pathological conditions including neoplasia.Eight different classes that encompass at least 24 functional types of MMPs have been described［16］.Altas et al.［17］investigated the impact of this polymorphism in pituitary adenomas and demonstrated that 90%of invasive pituitary adenomas occur in patients who are homozygous for this MMP1 SNP.MMP9 expression is significantly higher in invasive pituitary adenomas.Moreover，there maybe a correlation between activation of protein kinase C(PKC)and increased levels of MMP9，which can be antagonized by PKC inhibitors［18］.

Pituitary tumor transforming gene(PTTG)is a member of the securin family，which regulates sister chromatid separation during mitosis.Filippella et al.reported that a PTTG/Ki-67 score higher than 2.9%predicts a biologically aggressive behavior in pituitary adenomas.PTTG expression is also high in malignant rat-PRL-producing tumors.In addition，nuclear Ki-67(identified with the MIB-1 antibody)is a marker of cell division that is usually counted to determine a proliferation index in neoplasms［19］.

Sparsely granulated adenomas，which are by definition aggressive GH-producing adenomas，are thought to lack the high levels of cAMP that predict a response to somatostatin analogs;instead，they have altered STAT signaling that in some cases has been attributed to a somatic GHR mutation resulting in a histidine toleucine substitution in the extracellular domain in exon 4，codon 49.Altered GHR signaling is associated with a morphological change resulting in the formation of paranuclear keratin aggresomes，also known as‘fibrous bodies’.

2.3 Drug resistance

The majority of prolactinomas can be managed medically with dopamine agonists(DA).DA approved for use in the United States are bromocriptine(Parlodel)and cabergoline.The most common adverse effects of DA are nausea，vomiting，and fatigue.Medical management of growth hormone-and adrenal cortex hormone(ACTH)secreting tumors is less effective than for prolactinomas，and surgery via transsphenoidal resection is the preferred treatment.

Although prolactinomas can be effectively treated with DA，about 20%of patients develop dopamine resistance or tumor recurrence after surgery.Combining microarray data derived from human prolactinoma and pituitaries of estrogen-treated ACI rat，145 concordantly expressed genes，including E2F1，Myc，Igf1，and CEBPD，were identified.Gene set enrichment analysis revealed that 278 curated pathways and 59 gene sets of transcription factors were enriched in estrogen-treated ACI rats，suggesting a critical role for Myc，E2F1，CEBPD，and Sp1 in this rat prolactinoma［20］.DA resistant prolactinomas have a reduced density of D2Rs using different methods.Our group found the variant of C1orf170，DPCR1， DSPP， KRTAP10-3，MUC4，MX2，POTEF，PRB3，PRG4 and RP1L1 identificated by whole-exome sequencing.And low levels of PRB3 mRNA may contribute in some unknown way to promoting drug resistance and tumor recurrence of prolactinomas［21］.In addition，estrogen receptor(ER)antagonists are an alternative for treating patients with bromocriptine-resistant prolactinomas(BCRP).

Somatostatin analogs(SA)are widely used in acromegaly，either as first-line or adjuvant treatment after surgery.Generally，the response to SA takes into account both control of GH and IGF-I excess，with consequent improvement of clinical symptoms directly related to GH and IGF-I excess，and tumor shrinkage.Predictors of response are patients'gender，age，initial GH and IGF-I levels，and tumor mass，as well as adequate expression of somatostatin receptor types 2 and 5，those with the highest affinity for octreotide and lanreotide.The response to SA also depends on treatment duration and dosage of the drug used，so that a definition of resistance based on short-term treatments using low doses of long-acting SA is limited.Current data suggest that response to SA should be analyzed by taking together the effects on GH and IGF-I and those on tumor mass because only the lack of both responses might be considered as a poor response or resistance［22］.

2.4 Pathway analyses

Several signaling pathways and networks were found to be significantly associated with a pituitary adenoma，and include mitochondria dysfunction，oxidative stress，cell-cycle dysregulation，and the MAPK-signaling system through 2DGE-arrayed adenoma and control proteome images［23］.

The relative mRNA levels of the 17 genes determined by microarray analysis shows a comparison of these results with the RT-qPCR analysis.Down-regulation of all genes involved in the extracellular matrix(ECM)receptor interaction pathway in all pituitary adenomas when compared with normal pituitary glands，includingdown-regulation ofCOL4A6，COL6A1 and COL6A2 genes.Genes involved in TGF-β signaling pathway are down-regulated in five pituitary adenoma subtypes，suggesting a role for this pathway in the pathogenesis of pituitary adenomas［24］.And pathway analysis showed that the P53 and GnRH signaling pathways may play an important role in tumorigenesis of prolactinomas.The p53 and Notch signaling pathways may play an important role in tumorigenesis and progression of PHPAs，and ECM-receptor interactions likely play a role in the inhibition of invasion and metastasis in these tumors［24］.

2.5 Gene and familial pituitary adenomas

Familial pituitary adenomas occur in the classical syndromes of multiple endocrine neoplasial(MEN1)in Familial Isolated Pituitary Adenomas(FIPA).Approximately 20%-40%of FIPA families have been shown to harboran AIP gene mutation.In some families and also rarely in sporadic tumors germline mutations of a gene located on chromosome11q13 known as the aryl hydrocarbon receptor interacting protein have been found［25］.The clinical features of these familial acromegaly families were clearly distinguishable from the MEN1 syndrome:patients had primarily somatotroph adenomas，which only occurs in 25%of MEN1-related pituitary adenomas，no parathyroid or pancreas disease were identified and the penetrance was considerably lower than in MEN1.

It is now possible to imagine a day in the not-toodistant future when serum biomarkers and molecular imaging probes that will be used for screening or early detection.Tumors will undergo global systems biologic analysis(or analysis of a subset of markers)to identify pathway alterations that point to the most beneficial therapy or combination of therapies.Responses to therapy will be quantitatively and reproducibly monitored in a minimally invasive fashion using molecular-imaging probes and/or serum biomarkers to detect the biological effect of drugs on their intended target gene or pathway.

3 Reference

［1］ Legler J M，GloecklerRies L A，Smith M A，et al.RESPONSE:re:brain and other central nervous system cancers:recent trends in incidence and mortality［J］.J Natl Cancer Inst，2000，92(1):77A-78A.

［2］ Takeshima H，Sawamura Y，Gilbert M R，et al.Application of advances in molecular biology to the treatment of brain tumors［J］.Curr Oncol Rep，2000，2(5):425-433.

［3］ Giannini C，Scheithauer B W，Weaver A L，et al.Oligodendrogliomas:reproducibility and prognostic value of histologic diagnosis and grading［J］.J Neuropathol Exp Neurol，2001，60(3):248-262.

［4］ Hlaihel C，Boussel L，Cochet H，et al.Dose and image quality comparison between prospectively gated axial and retrospectively gated helical coronary CT angiography［J］.Br J Radiol，2011，84(997):51-57.

［5］ Louis D N，Ohgaki H，Wiestler O D，et al.The 2007 WHO classification of tumours of the central nervous system［J］.Acta Neuropathol，2007，114(2):97-109.

［6］ Phillips H S，Kharbanda S，Chen R，et al.Molecular subclasses of high-grade glioma predict prognosis，delineate a pattern of disease progression，and resemble stages in neurogenesis［J］.Cancer Cell，2006，9(3):157-173.

［7］ Nutt C L，Mani D R，Betensky R A，et al.Gene expression-based classification of malignant gliomas correlates better with survival than histological classification［J］.Cancer Res，2003，63(7):1602-1607.

［8］ Hartmann C，Hentschel B，Tatagiba M，et al.Molecular markers in low-grade gliomas:predictive or prognostic?［J］.Clin Cancer Res，2011，17(13):4588-4599.

［9］ Masui K，Cloughesy T F，Mischel P S.Molecular pathology in adult high-grade gliomas:from molecular diagnostics to target therapies［J］.Neuropathol Appl Neurobiol，2012，38(3):271-291.

［10］Weller M，Pfister S M，Wick W，et al.Molecular neuro-oncology in clinical practice:a new horizon［J］.Lancet Oncol，2013，14(9):e370-e379.

［11］Al-Hajj M，Becker M W，Wicha M，et al.Therapeutic implications of cancer stem cells［J］.Curr Opin Genet Dev，2004，14(1):43-47.

［12］ Stiles C D，Rowitch D H.Glioma stem cells:a midterm exam［J］.Neuron，2008，58(6):832-846.

［13］Sukhdeo K，Hambardzumyan D，Rich J N.Glioma development:where did it all go wrong? ［J］.Cell，2011，146(2):187-188.

［14］ Trouillas J.In search of a prognostic classification of endocrine pituitary tumors［J］.Endocr Pathol，2014，25(2):124-132.

［15］Li Z B，Niu G，Wang H，et al.Imaging of urokinase-type plasminogen activator receptor expression using a Cu-64-labeled linear peptide antagonist by microPET［J］.Clin Cancer Res，2008，14(15):4758-4766.

［16］ Arakaki P A，Marques M R，Santos M C.MMP-1 polymorphism and its relationship to pathological processes［J］.J Biosci，2009，34(2):313-320.

［18］Hussaini I M，Trotter C，Zhao Y，et al.Matrix metalloproteinase-9 is differentially expressed in nonfunctioning invasive and noninvasive pituitary adenomas and increases invasion in human pituitary adenoma cell line［J］.Am J Pathol，2007，170(1):356-365.

［19］ Mete O，Ezzat S，Asa S L.Biomarkers of aggressive pituitary adenomas［J］.J Mol Endocrinol，2012，49(2):R69-R78.

［20］ Tong Y，Zheng Y，Zhou J，et al.Genomic characterization of human and rat prolactinomas［J］.Endocrinology，2012，153(8):3679-3691.

［21］Wang F，Gao H，Li C，et al.Low levels of PRB3 mRNA are associated with dopamine-agonist resistance and tumor recurrence in prolactinomas［J］.J Neurooncol，2014，116(1):83-88.

［22］ David A，Hwa V，Metherell L A，et al.Evidence for a continuum of genetic，phenotypic，and biochemical abnormalities in children with growth hormone insensitivity［J］.Endocr Rev，2011，32(4):472-497.

［23］ Zhan X，Desiderio D M.Signaling pathway networks mined from human pituitary adenoma proteomics data［J］.BMC Med Genomics，2010，3:13.

［24］ Jiang Z，Gui S，Zhang Y.Analysis of differential gene expression using fiber-optic bead array and pathway analyses in pituitary adenomas［J］.J Clin Neurosci，2011，18(10):1386-1391.

［25］ Ozfirat Z，Korbonits M.AIP gene and familial isolated pituitary adenomas［J］.Mol Cell Endocrinol，2010，326(1-2):71-79.