Bayesian Hidden Markov Modeling of Array CGH Data

Genomic alterations have been linked to the development and progression of cancer. The technique of Comparative Genomic Hybridization (CGH) yields data consisting of fluorescence intensity ratios of test and reference DNA samples. The intensity ratios provide information about the number of copies in DNA. Practical issues such as the contamination of tumor cells in tissue specimens and normalization errors necessitate the use of statistics for learning about the genomic alterations from array-CGH data. As increasing amounts of array CGH data become available, there is a growing need for automated algorithms for characterizing genomic profiles. Specifically, there is a need for algorithms that can identify gains and losses in the number of copies based on statistical considerations, rather than merely detect trends in the data. We adopt a Bayesian approach, relying on the hidden Markov model to account for the inherent dependence in the intensity ratios. Posterior inferences are made about gains and losses in copy number. Localized amplifications (associated with oncogene mutations) and deletions (associated with mutations of tumor suppressors) are identified using posterior probabilities. Global trends such as extended regions of altered copy number are detected. Since the posterior distribution is analytically intractable, we implement a Metropolis-within-Gibbs algorithm for efficient simulation-based inference. Publicly available data on pancreatic adenocarcinoma, glioblastoma multiforme and breast cancer are analyzed, and comparisons are made with some widely-used algorithms to illustrate the reliability and success of the technique.

SURVIVAL PREDICTION FOR BRAIN TUMOR PATIENTS USING GENE EXPRESSION DATA

Brain tumor is one of the most aggressive types of cancer in humans, with an estimated median survival time of 12 months and only 4% of the patients surviving more than 5 years after disease diagnosis. Until recently, brain tumor prognosis has been based only on clinical information such as tumor grade and patient age, but there are reports indicating that molecular profiling of gliomas can reveal subgroups of patients with distinct survival rates. We hypothesize that coupling molecular profiling of brain tumors with clinical information might improve predictions of patient survival time and, consequently, better guide future treatment decisions. In order to evaluate this hypothesis, the general goal of this research is to build models for survival prediction of glioma patients using DNA molecular profiles (U133 Affymetrix gene expression microarrays) along with clinical information. First, a predictive Random Forest model is built for binary outcomes (i.e. short vs. long-term survival) and a small subset of genes whose expression values can be used to predict survival time is selected. Following, a new statistical methodology is developed for predicting time-to-death outcomes using Bayesian ensemble trees. Due to a large heterogeneity observed within prognostic classes obtained by the Random Forest model, prediction can be improved by relating time-to-death with gene expression profile directly. We propose a Bayesian ensemble model for survival prediction which is appropriate for high-dimensional data such as gene expression data. Our approach is based on the ensemble "sum-of-trees" model which is flexible to incorporate additive and interaction effects between genes. We specify a fully Bayesian hierarchical approach and illustrate our methodology for the CPH, Weibull, and AFT survival models. We overcome the lack of conjugacy using a latent variable formulation to model the covariate effects which decreases computation time for model fitting. Also, our proposed models provides a model-free way to select important predictive prognostic markers based on controlling false discovery rates. We compare the performance of our methods with baseline reference survival methods and apply our methodology to an unpublished data set of brain tumor survival times and gene expression data, selecting genes potentially related to the development of the disease under study. A closing discussion compares results obtained by Random Forest and Bayesian ensemble methods under the biological/clinical perspectives and highlights the statistical advantages and disadvantages of the new methodology in the context of DNA microarray data analysis.

Analyzing the Role of αvβ8 Integrin in Glioma-Induced Angiogenesis

In this study we aimed to determine the functional roles for αvβ8 integrin in astrocytoma-induced angiogenesis. These studies originate from our analyses of αvβ8 integrin in developmental brain angiogenesis. αv and β8 knockout (KO) mice develop brain-specific vascular phenotypes that resemble vascular pathologies observed in the malignant astrocytoma, glioblastoma multiforme (GBM). Indeed, a murine xenograft model of astrocytoma suggested a role for the integrin in glioma-induced angiogenesis. Primary mouse astroglia were cultured from wild type (WT) and β8 KO neonates and were immortalized (HPV:E6/E7) and transformed (HRas:G12V). WT and β8 KO transformed astroglia were intracranially injected into athymic mice. WT tumors displayed pathological features of grade III astrocytomas, whereas β8 KO tumors resembled grade IV GBMs. KO tumors contained widespread edema and hemorrhage as well as pathological angiogenesis, as assessed by quantitation of microvascular density and blood vessel morphology. Additionally, exogenous expression of β8 integrin in β8 KO transformed astroglia resolved the pathologies observed in KO tumors giving further credence to the idea that loss of αvβ8 integrin expression correlates with tumorigenic potential of oncogene-transformed astroglia. To compliment our mouse model, several established human glioma cell lines were characterized for expression of αvβ8 integrin protein. Some of the cell lines displayed low expression of αvβ8 integrin, whereas others showed high levels, as compared to non-malignant human astrocytes. Intracranial implantation of high and low β8 integrin-expressing human glioma cell lines resulted in tumors exhibiting similar phenotypes to those observed in the mouse model; low expressers were marked by vascular pathologies indicative of β8 KO mouse tumors. Upon overexpression of β8 integrin in a low β8 integrin-expressing human glioma cell line, angiogenic pathologies were largely resolved. Moreover, intracranially injected αvHI- and αvLO-sorted GBM stem cells (GSCs) resulted in significantly different tumor sizes, where those GSCs endogenously expressing low levels of αv integrin formed two to three fold larger tumors. Furthermore, lentiviral knockdown of β8 integrin in transformed human astrocytes formed tumors that strikingly recapitulated the characteristics of the murine β8-/- tumors, exhibiting a significant increase in microvascular density leading to decreased overall survival. A paracrine mechanism was discovered involving endothelial cell homeostatic control governed by canonical TGFβ signaling initiated by αvβ8 integrin’s role in the latent cytokine’s activation. Diminished TGFβ signaling in tumor-associated endothelial cells promoted increased angiogenesis and decreased overall survival as a result of αvβ8 integrin’s loss on the tumor cell. Collectively, these data suggest an important functional role for αvβ8 integrin in glioma angiogenesis.

FoxM1B regulates NEDD4-1 expression, leading to cellular transformation and full malignant phenotype in immortalized human astrocytes.

Our recent studies have shown that the FoxM1B transcription factor is overexpressed in human glioma tissues and that the level of its expression correlates directly with glioma grade. However, whether FoxM1B plays a role in the early development of glioma (i.e., in transformation) is unknown. In this study, we found that the FoxM1B molecule causes cellular transformation and tumor formation in normal human astrocytes (NHA) immortalized by p53 and pRB inhibition. Moreover, brain tumors that arose from intracranial injection of FoxM1B-expressing immortalized NHAs displayed glioblastoma multiforme (GBM) phenotypes, suggesting that FoxM1B overexpression in immortalized NHAs not only transforms the cells but also leads to GBM formation. Mechanistically, our results showed that overexpression of FoxM1B upregulated NEDD4-1, an E3 ligase that mediates the degradation and downregulation of phosphatase and tensin homologue (PTEN) in multiple cell lines. Decreased PTEN in turn resulted in the hyperactivation of Akt, which led to phosphorylation and cytoplasmic retention of FoxO3a. Blocking Akt activation with phosphoinositide 3-kinase/Akt inhibitors inhibited the FoxM1B-induced transformation of immortalized NHAs. Furthermore, overexpression of FoxM1B in immortalized NHAs increased the expression of survivin, cyclin D1, and cyclin E, which are important molecules for tumor growth. Collectively, these results indicate that overexpression of FoxM1B, in cooperation with p53 and pRB inhibition in NHA cells, promotes astrocyte transformation and GBM formation through multiple mechanisms.

Mechanisms of Ad-p53 induced apoptosis in human malignant glioma

Malignant brain tumors are one of the most challenging cancers affecting society today. In a recent survey, an estimated 17,000 annual cases were recorded with a staggering total of 13,300 deaths. A unique degree of heterogeneity typifies glial tumors and presents a challenge for solitary anti-neoplastic treatments. Tumors subsist as heterogeneous masses that progress through dysplasia to astrocytomas, mixed glioma and glioblastoma multiforme. Although traditional therapeutic approaches have provided increments of success, the median survival time remains 12 months. The urgency to improve upon current clinical protocols has encouraged alternative experimental strategies such as p53 adenoviral gene therapy (Ad-p53). This study addresses the efficacy of Ad-p53 for the treatment of glioma. Our model presents a tumor response that is unique among human cancers. Ad-p53 effectively induces apoptosis in mutant p53 expressing cells yet fails to do so in those with wildtype p53. In order to adopt Adp53 as a standard anti-cancer modality, we characterized the role of the tumor suppressor gene p53 in mediating apoptosis. We demonstrate that altering cellular p53 status through the introduction of a dominant negative mutant p53 (175H, 248W, 273H) sensitized cells to Ad-p53. We discovered that wild-type p53 expressing glioma cells retain the apoptotic machinery necessary to accomplish cell death, but have developed mechanisms that interfere with p53 signaling. Earlier studies have not addressed the mechanisms of Ad-p53 apoptosis nor the resistance exhibited by wild-type p53 glioma. To explain the divergent phenotypes, we identified apoptotic pathways activated and effectors of the response. We illustrated that modulation of the death receptor Fas/APO-1 is a principal means of Ad-p53 signaling that is impaired in wild-type p53 glioma. Moreover, the apoptotic response was found to be a multi-faceted process that engaged several caspases, most notably caspases -1, -3 and -8. Lastly, we assessed the ability of anti-apoptotic molecules Bcl-2 and CrmA to inhibit Ad-p53 apoptosis. These studies revealed that Ad-p53 is a powerful tool for inducing apoptosis that can be delayed but not inhibited by anti-apoptotic means. This work is critical for understanding the development of glioma and the phenotypic and genotypic alterations that account for tumor resistance. ^

Ligand-directed targeting of malignant gliomas

The dramatic poor survival of patients diagnosed with glioblastoma multiforme (GBM) is a reflection of the struggles that accompany traditional treatments. Thus, the development of molecular-based targeted therapies represents new windows for intervention. In this study, we hypothesized that we could select peptide-ligands that selectively target GBM based on the idea that the glioma microenvironment may induce or modify the expression of cell surface receptors that could be accessed by circulating peptides. To select the peptides we employed two distinct in vivo screenings. First, a random phage-displayed peptide library was injected into mice bearing intracranial tumors. Phage that bound to tumor were recovered and sequenced. We found that the tumor-derived phage CLSYKGRC, CNKVSTKC and CQSSREKC were recovered with the highest frequencies and used for subsequent targeting experiments. Second, the phage peptide library was injected into mice without tumors and phage were recovered from brain and sequenced. A phage-displayed peptide (CRTIGPSVC) with homology to transferrin (Tf) was selected and injected into brain tumor-bearing mice. Results showed that after 6 hours of circulation, the CLSYKGRC, CNKVSTKC and CQSSREKC-phage selectively targeted GBM vasculature. In contrast, Tf-like phage accumulated outside the tumor blood vessels in the cytoplasm of cells located within GBM, suggesting it was internalized in vivo. However, after short periods of circulation this phage was restricted to the tumor vasculature. Importantly, none of the selected phage targeted normal brain cells in animals bearing intracranial tumors. An affinity column coupled to the CNKVSTKC zpeptide was used to identify receptors from GBM. Using mass-spectrometry Vimentin, a marker of glial malignancy, was identified as a potential receptor. Other studies showed that the Tf-like phage bound selectively to Apo-Tf (iron free), with no binding to Holo-Tf (iron loaded) or to Tf receptor (TfR). However, the binding of Tf-like phage to glioma cells that express TfR increased in the presence of Apo-Tf. Thus, the Tf-like phage could indirectly target TfR using the endogenous Tf pathway. We propose that the novel peptides identified in this study could be conjugated to therapeutic or imaging agents for use GBM. ^

A mechanism of insulin-like growth factor binding protein 2-induced cell mobility in glioma

Insulin-like growth factor binding protein 2 (IGFBP2) is a protein known to be overexpressed in a majority of glioblastoma multiforme (GBM) tumors. While it is known the IGFBP2 is involved in promoting GBM tumor cell invasion, no mechanism exists for how the protein is involved in signal transduction pathways leading to enhanced cell invasion. ^ We follow up on preliminary microarray data on IGFBP2-overexpressing GBM cells and protein sequence analysis of IGFBP2 in generating the hypothesis that IGFBP2 interacts with integnn α5 in regulating cell mobility. Microarray data showing upregulation of integrin α5 by IGFBP2 is validated and evidence of protein-protein interaction between IGFBP2 and integrin α5 is found. The exact binding domain on IGFBP2 responsible for its interaction with integrin α5 is also determined, confirming our initial findings and reaffirming that the IGFBP2/integrin α5 interaction is specific. Disruption of this interaction resulted in attenuation of IGFBP2-enhanced cell mobility. Further, we found that cell mobility is only enhanced when IGFBP2 and integrin α5 are both overexpressed and able to interact with each other. ^ We also determined fibronectin to be a critical player in the activation of the IGFBP2/integrin α5 pathway. The activation of this pathway appears to be progressive and initiates once GBM cells have sufficiently established anchorage. ^

NHERF1 recruits the tumor suppressors PTEN and PHLPP to synergistically inhibit the PI-3K pathway

Glioblastoma multiforme is the most common form of brain cancer that presents patients with a poor prognosis that has remained unchanged over the past few decades. The tumor suppressor phosphatase PTEN antagonizes one of the major oncogenic pathways involved in the progression of glioblastoma, and is frequently deleted in this cancer type. Contrary to our expectations, we found that most glioblastoma cells expressing endogenous PTEN also harbor basal PI-3K/AKT activation mainly attributable to impaired PTEN membrane localization. This alteration correlated with a shift of the adaptor protein NHERF1, which contributes to PTEN membrane recruitment in normal cells, from the membrane to the cytoplasm. In cells expressing membrane-localized NHERF1, only simultaneous PTEN and NHERF1 depletion achieved AKT activation, suggesting the involvement of additional PI-3K/AKT suppressor regulated by NHERF1. We identified these novel interactors of NHERF1 as the PHLPP1 and PHLPP2 phosphatases. ^ NHERF1 directly interacted and recruited both PHLPP proteins to the membrane and, through both NHERF1 PDZ domains, assembled ternary complexes consisting of PTEN-NHERF1-PHLPP. Only simultaneous depletion of PTEN and PHLPP1 significantly activated AKT and increased proliferation in cells with membrane-localized NHERF1. Analysis of glioblastoma human tumors revealed frequent loss of membrane-localized NHERF1. On the other hand, targeting of NHERF1 to the membrane achieved suppression of AKT and cell proliferation. Our findings reveal a novel mechanism for PI-3K/AKT regulation by the synergistic cooperation between two important tumor suppressors, PTEN and PHLPP, via the scaffold protein NHERF1. ^

TAZ AS A REGULATOR OF MESENCHYMAL TRANSFORMATION AND CLINICAL AGGRESSIVENESS IN GLIOMAS

Glioblastoma multiforme (GBM) is an aggressive, high grade brain tumor. Microarray studies have shown a subset of GBMs with a mesenchymal gene signature. This subset is associated with poor clinical outcome and resistance to treatment. To establish the molecular drivers of this mesenchymal transition, we correlated transcription factor expression to the mesenchymal signature and identified transcriptional co-activator with PDZ-binding motif (TAZ) to be highly associated with the mesenchymal shift. High TAZ expression correlated with worse clinical outcome and higher grade. These data led to the hypothesis that TAZ is critical to the mesenchymal transition and aggressive clinical behavior seen in GBM. We investigated the expression of TAZ, its binding partner TEAD, and the mesenchymal marker FN1 in human gliomas. Western analyses demonstrated increased expression of TAZ, TEAD4, and FN1 in GBM relative to lower grade gliomas. We also identified CpG islands in the TAZ promoter that are methylated in most lower grade gliomas, but not in GBMs. TAZ-methylated glioma stem cell (GSC) lines treated with a demethylation agent showed an increase in mRNA and protein TAZ expression; therefore, methylation may be another novel way TAZ is regulated since TAZ is epigenetically silenced in tumors with a better clinical outcome. To further characterize the role of TAZ in gliomagenesis, we stably silenced or over-expressed TAZ in GSCs. Silencing of TAZ decreased invasion, self-renewal, mesenchymal protein expression, and tumor-initiating capacity. Over-expression of TAZ led to an increase in invasion, mesenchymal protein expression, mesenchymal differentiation, and tumor-initiating ability. These actions are dependent on TAZ interacting with TEAD since all these effects were abrogated with TAZ could not bind to TEAD. We also show that TAZ and TEAD directly bind to mesenchymal gene promoters. Thus, TAZ-TEAD interaction is critically important in the mesenchymal shift and in the aggressive clinical behavior of GBM. We identified TAZ as a regulator of the mesenchymal transition in gliomas. TAZ could be used as a biomarker to both estimate prognosis and stratify patients into clinically relevant subgroups. Since mesenchymal transition is correlated to tumor aggressiveness, strategies to target and inhibit TAZ-TEAD and the downstream gene targets may be warranted in alternative treatment.

The role of platelet -derived growth factor (PDGF) signaling in gliomagenesis

Gliomas are primary central nervous system (CNS) neoplasms that are believed to arise from astrocytes, oligodendrocytes or their precursors. Gliomas can be classified into two major histopathological groups: oligodendroglial and astroglial tumors. The most malignant of the astroglial tumors is glioblastoma multiforme (GBM). A great deal of genetic and epigenetic alterations have been implicated in gliomagenesis. In particular, PDGF signaling is frequently over-activated in a large number of human gliomas. In order to gain insights into the biology of gliomas, we manage to model human gliomas in mice using a somatic gene transfer approach—RCAS/TVA system. In our previous study, combined activation of AKT and RAS pathways gave rise to glioblastomas from CNS progenitors. In the present study, we demonstrate that in vivo autocrine PDGF stimulation induces oligodendrogliomas and mixed oligoastrocytomas from CNS progenitors and differentiated astrocytes respectively. In culture autocrine PDGF stimulation dedifferentiates astrocytes into progenitor-like cells and blockade of PDGF signaling reverses these phenotypic changes. Experimental disruption of cell cycle arrest pathway, such as Ink4a-Arf loss, is not required for the initiation of PDGF-induced gliomagenesis; instead, this mutation contributes to the tumor progression by enhancing tumor malignancy and shortening tumor latency. P53 deficiency does not promote the PDGF-induced gliomagenesis. In addition, 1p and 19q, often deleted in human oligodendrogliomas, remain intact in these PDGF-induced gliomas. Therefore, our studies suggest that autocrine PDGF stimulation alone may be sufficient to induce gliomagenesis. In contrast to transient stimulation in vitro, constitutive PDGF stimulation activates neither AKT nor RAS/MAPK pathways during gliomagenesis. This results in the formation of oligodendrogliomas, instead of glioblastomas. Sustained activation of the AKT pathway converts PDGF-induced oligodendrogliomas into astrocytomas. Our studies suggest that constitutive PDGF stimulation is not equivalent to transient PDGF stimulation, and that a transition between oligodendroglial and astroglial tumors in humans may be possible, depending on additional alterations. In summary, PDGF signaling plays a pivotal role in gliomagenesis in the mouse, and its hyperactivity is capable of contributing to both oligodendroglial and astroglial tumorigenesis. ^

Neutralizing monoclonal antibodies to hepatocyte growth factor/scatter factor (HGF/SF) display antitumor activity in animal models

The hepatocyte growth factor (HGF/SF) receptor, Met, regulates mitogenesis, motility, and morphogenesis in a cell type-dependent fashion. Activation of Met via autocrine, paracrine, or mutational mechanisms can lead to tumorigenesis and metastasis and numerous studies have linked inappropriate expression of this ligand-receptor pair to most types of human solid tumors. To prepare mAbs to human HGF/SF, mice were immunized with native and denatured preparations of the ligand. Recloned mAbs were tested in vitro for blocking activity against scattering and branching morphogenesis. Our results show that no single mAb was capable of neutralizing the in vitro activity of HGF/SF, and that the ligand possesses a minimum of three epitopes that must be blocked to prevent Met tyrosine kinase activation. In vivo, the neutralizing mAb combination inhibited s.c. growth in athymic nu/nu mice of tumors dependent on an autocrine Met-HGF/SF loop. Importantly, growth of human glioblastoma multiforme xenografts expressing Met and HGF/SF were markedly reduced in the presence of HGF/SF-neutralizing mAbs. These results suggest interrupting autocrine and/or paracrine Met-HGF/SF signaling in tumors dependent on this pathway is a possible intervention strategy.

Selective tumor kill of cerebral glioma by photodynamic therapy using a boronated porphyrin photosensitizer.

The prognosis for patients with the high-grade cerebral glioma glioblastoma multiforme is poor. The median survival for primary tumors is < 12 months, with most recurring at the site of the original tumor, indicating that a more aggressive local therapy is required to eradicate the unresectable "nests" of tumor cells invading into adjacent brain. Two adjuvant therapies with the potential to destroy these cells are porphyrin-sensitized photodynamic therapy (PDT) and boron-sensitized boron neutron capture therapy (BNCT). The ability of a boronated porphyrin, 2,4-(alpha, beta-dihydroxyethyl) deuteroporphyrin IX tetrakiscarborane carboxylate ester (BOPP), to act as a photosensitizing agent was investigated in vitro with the C6 rat glioma cell line and in vivo with C6 cells grown as an intracerebral tumor after implantation into Wistar rats. These studies determined the doses of BOPP and light required to achieve maximal cell kill in vitro and selective tumor kill in vivo. The data show that BOPP is more dose effective in vivo by a factor of 10 than the current clinically used photosensitizer hematoporphyrin derivative and suggest that BOPP may have potential as a dual PDT/BNCT sensitizer.

Identification of the human prostatic carcinoma oncogene PTI-1 by rapid expression cloning and differential RNA display.

Elucidating the relevant genomic changes mediating development and evolution of prostate cancer is paramount for effective diagnosis and therapy. A putative dominant-acting nude mouse prostatic carcinoma tumor-inducing gene, PTI-1, has been cloned that is expressed in patient-derived human prostatic carcinomas but not in benign prostatic hypertrophy or normal prostate tissue. PTI-1 was detected by cotransfecting human prostate carcinoma DNA into CREF-Trans 6 cells, inducing tumors in nude mice, and isolating genes displaying increased expression in tumor-derived cells by using differential RNA display (DD). Screening a human prostatic carcinoma (LNCaP) cDNA library with a 214-bp DNA fragment found by DD permitted the cloning of a full-length 2.0-kb PTI-1 cDNA. Sequence analysis indicates that PTI-1 is a gene containing a 630-bp 5' sequence and a 3' sequence homologous to a truncated and mutated form of human elongation factor 1 alpha. In vitro translation demonstrates that the PTI-1 cDNA encodes a predominant approximately 46-kDa protein. Probing Northern blots with a DNA fragment corresponding to the 5' region of PTI-1 identifies multiple PTI-1 transcripts in RNAs from human carcinoma cell lines derived from the prostate, lung, breast, and colon. In contrast, PTI-1 RNA is not detected in human melanoma, neuroblastoma, osteosarcoma, normal cerebellum, or glioblastoma multiforme cell lines. By using a pair of primers recognizing a 280-bp region within the 630-bp 5' PTI-1 sequence, reverse transcription-PCR detects PTI-1 expression in patient-derived prostate carcinomas but not in normal prostate or benign hypertrophic prostate tissue. In contrast, reverse transcription-PCR detects prostate-specific antigen expression in all of the prostate tissues. These results indicate that PTI-1 may be a member of a class of oncogenes that could affect protein translation and contribute to carcinoma development in human prostate and other tissues. The approaches used, rapid expression cloning with the CREF-Trans 6 system and the DD strategy, should prove widely applicable for identifying and cloning additional human oncogenes.

Current concepts of tumour metastasis

Background: Tumour metastasis remains the principal cause of treatment failure and poor prognosis in patients with cancer. Recent advances in our understanding of the biology of metastasis are providing novel potential targets for anti-cancer therapies. Aim: This paper reviews the current concepts in tumour metastasis. Methods: A review of Medline publications relating to the molecular biology and therapy of human tumour metastasis was conducted. Results and Discussion: Early metastasis models were based upon the premise of uninterrupted tumour growth, with the inevitable formation of distant metastases and eventual death of the patient. However, current research suggests that metastasis is an inefficient process governed by several rate-limiting steps, and that failure to negotiate these steps can lead to tumour dormancy. Successful metastatic tumour growth depends upon appropriate tumour-host microenvironment interactions and, ultimately, the development of vascularised metastases post-extravasation in the target organ. An understanding of the molecular mechanisms involved in this dynamic process will aid in the identification of therapeutic targets that may allow earlier diagnosis and more specific therapies for patients with metastasis.

Supportive care needs of people with brain tumours and their carers

Goals of work: The diagnosis and treatment of a brain tumour may result in long-term changes in a patient's functional and social abilities and/or in a greatly reduced life span. A qualitative investigation was conducted to examine the supportive care needs of patients with brain tumour and their carers. Materials and methods: Overall, 18 patients and 18 carers participated in focus groups or telephone interviews, following a structured interview guide to elicit supportive care services of importance to these patients and carers. Main results: Six major themes were identified using the framework analysis method, including needs for information and coping with uncertainty, practical support, support to return to pretreatment responsibilities or prepare for long-term care, support to deal with social isolation and organize respite care, support to overcome stigma/discrimination and support to discuss potentially reduced life expectancy. Conclusions: Five recommendations to improve service delivery include: assignment of a dedicated member of the care team or case manager; proactive dissemination of information, education and psychosocial support; access to objective assessment of neuropsychological functioning; facilitating easier access to welfare payments; and services facilitating communication about difficult illness-related topics. Provision of services along these recommendations could improve supportive care of brain tumour patients and their carers.