Rationale and design of the CRAFT (Continuous ReAssessment with Flexible ExTension in Rare Malignancies) multicenter phase II trial.

BACKGROUND: Approvals of cancer therapeutics are primarily disease entity specific. Current molecular diagnostic approaches frequently identify actionable alterations in rare cancers or rare subtypes of common cancers for which the corresponding treatments are not approved and unavailable within clinical trials due to entity-related eligibility criteria. Access may be negotiated with health insurances. However, approval rates vary, and critical information required for a scientific evaluation of treatment-associated risks and benefits is not systematically collected. Thus clinical trials with optimized patient selection and comprehensive molecular characterization are essential for translating experimental treatments into standard care. PATIENTS AND METHODS: Continuous ReAssessment with Flexible ExTension in Rare Malignancies (CRAFT) is an open-label phase II trial for adults with pretreated, locally advanced, or metastatic solid tumors. Based on the evaluation by a molecular tumor board, patients are assigned to combinations of six molecularly targeted agents and a programmed death-ligand 1 (PD-L1) antagonist within seven study arms focusing on (i) BRAF V600 mutations; (ii) ERBB2 amplification and/or overexpression, activating ERBB2 mutations; (iii) ALK rearrangements, activating ALK mutations; (iv and v) activating PIK3CA and AKT mutations, other aberrations predicting increased PI3K-AKT pathway activity; (vi) aberrations predicting increased RAF-MEK-ERK pathway activity; (vii) high tumor mutational burden and other alterations predicting sensitivity to PD-L1 inhibition. The primary endpoint is the disease control rate (DCR) at week 16; secondary and exploratory endpoints include the progression-free survival ratio, overall survival, and patient-reported outcomes. Using Simon's optimal two-stage design, 14 patients are accrued for each study arm. If three or fewer patients achieve disease control, the study arm is stopped. Otherwise, 11 additional patients are accrued. If the DCR exceeds 7 of 25 patients, the null hypothesis is rejected for the respective study arm. CONCLUSIONS: CRAFT was activated in October 2021 and will recruit at 10 centers in Germany. TRIAL REGISTRATION NUMBERS: EudraCT: 2019-003192-18; ClinicalTrials.gov: NCT04551521.

Drug-based perturbation screen uncovers synergistic drug combinations in Burkitt lymphoma.

Burkitt lymphoma (BL) is a highly aggressive B-cell lymphoma associated with MYC translocation. Here, we describe drug response profiling of 42 blood cancer cell lines including 17 BL to 32 drugs targeting key cancer pathways and provide a systematic study of drug combinations in BL cell lines. Based on drug response, we identified cell line specific sensitivities, i.e. to venetoclax driven by BCL2 overexpression and partitioned subsets of BL driven by response to kinase inhibitors. In the combination screen, including BET, BTK and PI3K inhibitors, we identified synergistic combinations of PI3K and BTK inhibition with drugs targeting Akt, mTOR, BET and doxorubicin. A detailed comparison of PI3K and BTKi combinations identified subtle differences, in line with convergent pathway activity. Most synergistic combinations were identified for the BET inhibitor OTX015, which showed synergistic effects for 41% of combinations including inhibitors of PI3K/AKT/mTOR signalling. The strongest synergy was observed for the combination of the CDK 2/7/9 inhibitor SNS032 and OTX015. Our data provide a landscape of drug combination effects in BL and suggest that targeting CDK and BET could provide a novel vulnerability of BL.

Mutant KIT as imatinib-sensitive target in metastatic sinonasal carcinoma.

Background: Sinonasal carcinomas (SNCs) comprise various rare tumor types that are characterized by marked histologic diversity and largely unknown molecular profiles, yet share an overall poor prognosis owing to an aggressive clinical course and frequent late-stage diagnosis. The lack of effective systemic therapies for locally advanced or metastatic SNC poses a major challenge to therapeutic decision making for individual patients. We here aimed to identify actionable genetic alterations in a patient with metastatic SNC whose tumor, despite all diagnostic efforts, could not be assigned to any known SNC category and was refractory to multimodal therapy. Patients and methods: We used whole-exome and transcriptome sequencing to identify a KIT exon 11 mutation (c.1733_1735del, p.D579del) as potentially druggable target in this patient and carried out cancer hotspot panel sequencing to detect secondary resistance-conferring mutations in KIT. Furthermore, as a step towards clinical exploitation of the recently described signatures of mutational processes in cancer genomes, we established and applied a novel bioinformatics algorithm that enables supervised analysis of the mutational catalogs of individual tumors. Results: Molecularly guided treatment with imatinib in analogy to the management of gastrointestinal stromal tumor (GIST) resulted in a dramatic and durable response with remission of nearly all tumor manifestations, indicating a dominant driver function of mutant KIT in this tumor. KIT dependency was further validated by a secondary KIT exon 17 mutation (c.2459_2462delATTCinsG, p.D820_S821delinsG) that was detected upon tumor progression after 10 months of imatinib treatment and provided a rationale for salvage therapy with regorafenib, which has activity against KIT exon 11/17 mutant GIST. Conclusions: These observations highlight the potential of unbiased genomic profiling for uncovering the vulnerabilities of individual malignancies, particularly in rare and unclassifiable tumors, and underscore that KIT exon 11 mutations represent tractable therapeutic targets across different histologies.

Genome-wide association study of pathological gambling.

BACKGROUND: Pathological gambling is a behavioural addiction with negative economic, social, and psychological consequences. Identification of contributing genes and pathways may improve understanding of aetiology and facilitate therapy and prevention. Here, we report the first genome-wide association study of pathological gambling. Our aims were to identify pathways involved in pathological gambling, and examine whether there is a genetic overlap between pathological gambling and alcohol dependence. METHODS: Four hundred and forty-five individuals with a diagnosis of pathological gambling according to the Diagnostic and Statistical Manual of Mental Disorders were recruited in Germany, and 986 controls were drawn from a German general population sample. A genome-wide association study of pathological gambling comprising single marker, gene-based, and pathway analyses, was performed. Polygenic risk scores were generated using data from a German genome-wide association study of alcohol dependence. RESULTS: No genome-wide significant association with pathological gambling was found for single markers or genes. Pathways for Huntington's disease (P-value=6.63×10(-3)); 5'-adenosine monophosphate-activated protein kinase signalling (P-value=9.57×10(-3)); and apoptosis (P-value=1.75×10(-2)) were significant. Polygenic risk score analysis of the alcohol dependence dataset yielded a one-sided nominal significant P-value in subjects with pathological gambling, irrespective of comorbid alcohol dependence status. CONCLUSIONS: The present results accord with previous quantitative formal genetic studies which showed genetic overlap between non-substance- and substance-related addictions. Furthermore, pathway analysis suggests shared pathology between Huntington's disease and pathological gambling. This finding is consistent with previous imaging studies.

[Durable remission under dual HER2 blockade with Trastuzumab and Pertuzumab in a patient with metastatic gallbladder cancer]. / Dauerhafte Remission unter dualer HER2-Blockade mit Trastuzumab und Pertuzumab bei metastasiertem Gallenblasenkarzinom.

Gallbladder cancer represents a rare but dismal disease. The only curative option is complete surgical resection, though patients often develop recurrent disease. In patients with advanced biliary tract cancer, the combination of cisplatin and gemcitabine showed a benefit in overall survival compared to gemcitabine alone. However, there is no standardized second-line regimen after treatment failure. We report on a young patient with early recurrence of a gallbladder cancer with cutaneous and peritoneal metastases. Upon identification of an ERBB2 gene amplification within the NCT MASTER (Molecularly Aided Stratification for Tumor Eradication Research) exome sequencing program with resulting overexpression of HER2 in the tumors cells, the patient received a targeted therapy with the HER2 antibodies pertuzumab and trastuzumab in combination with nab-paclitaxel, which led to a durable remission for more than one year. This case report underlines the potential of molecularly aided personalized targeted therapy for patients with biliary tract cancer and the need for respective clinical trials.

Cooperation of BRAF(F595L) and mutant HRAS in histiocytic sarcoma provides new insights into oncogenic BRAF signaling.

Activating BRAF mutations, in particular V600E/K, drive many cancers and are considered mutually exclusive with mutant RAS, whereas inactivating BRAF mutations in the D(594)F(595)G(596) motif cooperate with RAS via paradoxical MEK/ERK activation. Due to the increasing use of comprehensive tumor genomic profiling, many non-V600 BRAF mutations are being detected whose functional consequences and therapeutic actionability are often unknown. We investigated an atypical BRAF mutation, F595L, which was identified along with mutant HRAS in histiocytic sarcoma and also occurs in epithelial cancers, melanoma and neuroblastoma, and determined its interaction with mutant RAS. Unlike other DFG motif mutants, BRAF(F595L) is a gain-of-function variant with intermediate activity that does not act paradoxically, but nevertheless cooperates with mutant RAS to promote oncogenic signaling, which is efficiently blocked by pan-RAF and MEK inhibitors. Mutation data from patients and cell lines show that BRAF(F595L), as well as other intermediate-activity BRAF mutations, frequently coincide with mutant RAS in various cancers. These data define a distinct class of activating BRAF mutations, extend the spectrum of patients with systemic histiocytoses and other malignancies who are candidates for therapeutic blockade of the RAF-MEK-ERK pathway and underscore the value of comprehensive genomic testing for uncovering the vulnerabilities of individual tumors.

Investigation of manic and euthymic episodes identifies state- and trait-specific gene expression and STAB1 as a new candidate gene for bipolar disorder.

Bipolar disorder (BD) is a highly heritable psychiatric disease characterized by recurrent episodes of mania and depression. To identify new BD genes and pathways, the present study employed a three-step approach. First, gene-expression profiles of BD patients were assessed during both a manic and an euthymic phase. These profiles were compared intra-individually and with the gene-expression profiles of controls. Second, those differentially expressed genes that were considered potential trait markers of BD were validated using data from the Psychiatric Genomics Consortiums' genome-wide association study (GWAS) of BD. Third, the implicated molecular mechanisms were investigated using pathway analytical methods. In the present patients, this novel approach identified: (i) sets of differentially expressed genes specific to mania and euthymia; and (ii) a set of differentially expressed genes that were common to both mood states. In the GWAS data integration analysis, one gene (STAB1) remained significant (P=1.9 × 10(-4)) after adjustment for multiple testing. STAB1 is located in close proximity to PBMR1 and the NEK4-ITIH1-ITIH3-ITIH4 region, which are the top findings from GWAS meta-analyses of mood disorder, and a combined BD and schizophrenia data set. Pathway analyses in the mania versus control comparison revealed three distinct clusters of pathways tagging molecular mechanisms implicated in BD, for example, energy metabolism, inflammation and the ubiquitin proteasome system. The present findings suggest that STAB1 is a new and highly promising candidate gene in this region. The combining of gene expression and GWAS data may provide valuable insights into the biological mechanisms of BD.

Hox-C9 activates the intrinsic pathway of apoptosis and is associated with spontaneous regression in neuroblastoma.

Neuroblastoma is an embryonal malignancy of the sympathetic nervous system. Spontaneous regression and differentiation of neuroblastoma is observed in a subset of patients, and has been suggested to represent delayed activation of physiologic molecular programs of fetal neuroblasts. Homeobox genes constitute an important family of transcription factors, which play a fundamental role in morphogenesis and cell differentiation during embryogenesis. In this study, we demonstrate that expression of the majority of the human HOX class I homeobox genes is significantly associated with clinical covariates in neuroblastoma using microarray expression data of 649 primary tumors. Moreover, a HOX gene expression-based classifier predicted neuroblastoma patient outcome independently of age, stage and MYCN amplification status. Among all HOX genes, HOXC9 expression was most prominently associated with favorable prognostic markers. Most notably, elevated HOXC9 expression was significantly associated with spontaneous regression in infant neuroblastoma. Re-expression of HOXC9 in three neuroblastoma cell lines led to a significant reduction in cell viability, and abrogated tumor growth almost completely in neuroblastoma xenografts. Neuroblastoma growth arrest was related to the induction of programmed cell death, as indicated by an increase in the sub-G1 fraction and translocation of phosphatidylserine to the outer membrane. Programmed cell death was associated with the release of cytochrome c from the mitochondria into the cytosol and activation of the intrinsic cascade of caspases, indicating that HOXC9 re-expression triggers the intrinsic apoptotic pathway. Collectively, our results show a strong prognostic impact of HOX gene expression in neuroblastoma, and may point towards a role of Hox-C9 in neuroblastoma spontaneous regression.

MicroRNA miR-885-5p targets CDK2 and MCM5, activates p53 and inhibits proliferation and survival.

Several microRNA (miRNA) loci are found within genomic regions frequently deleted in primary neuroblastoma, including miR-885-5p at 3p25.3. In this study, we demonstrate that miR-885-5p is downregulated on loss of 3p25.3 region in neuroblastoma. Experimentally enforced miR-885-5p expression in neuroblastoma cell lines inhibits proliferation triggering cell cycle arrest, senescence and/or apoptosis. miR-885-5p leads to the accumulation of p53 protein and activates the p53 pathway, resulting in upregulation of p53 targets. Enforced miR-885-5p expression consistently leads to downregulation of cyclin-dependent kinase (CDK2) and mini-chromosome maintenance protein (MCM5). Both genes are targeted by miR-885-5p via predicted binding sites within the 3'-untranslated regions (UTRs) of CDK2 and MCM5. Transcript profiling after miR-885-5p introduction in neuroblastoma cells reveals alterations in expression of multiple genes, including several p53 target genes and a number of factors involved in p53 pathway activity. Taken together, these data provide evidence that miR-885-5p has a tumor suppressive role in neuroblastoma interfering with cell cycle progression and cell survival.

The Forkhead factor FoxQ1 influences epithelial differentiation.

The Forkhead family of transcription factors comprises numerous members and is implicated in various cellular functions, including cell growth, apoptosis, migration, and differentiation. In this study, we identified the Forkhead factor FoxQ1 as increased in expression during TGF-ß1 induced changes in epithelial differentiation, suggesting functional roles of FoxQ1 for epithelial plasticity. The repression of FoxQ1 in mammary epithelial cells led to a change in cell morphology characterized by an increase in cell size, pronounced cell-cell contacts, and an increased expression of several junction proteins (e.g., E-cadherin). In addition, FoxQ1 knock-down cells revealed rearrangements in the actin-cytoskeleton and slowed down cell cycle G1-phase progression. Furthermore, repression of FoxQ1 enhanced the migratory capacity of coherent mammary epithelial cells. Gene expression profiling of NM18 cells indicated that FoxQ1 is a relevant downstream mediator of TGF-ß1-induced gene expression changes. This included the differential expression of transcription factors involved in epithelial plasticity, for example, Ets-1, Zeb1, and Zeb2. In summary, this study has elucidated the functional impact of FoxQ1 on epithelial differentiation.

[Molecular pathology of sarcomas. Update on the research group "Molecular Diagnosis of Sarcomas"]. / Molekularpathologie von Sarkomen. Erste Ergebnisse des Sarkomforschungsverbundes KoSar.

To establish precise diagnostic algorithms and standardised treatment of sarcomas in specialized centers, the interdisciplinary research group KoSar (sarcoma competence network) has been funded by German Cancer Aid. A sarcoma tissue repository and a diagnostic reference center have been set up, presently containing about 1000 accurately diagnosed sarcomas of different entities. Significant gene expression profiles for synovial sarcomas, leiomyosarcomas, myxoid liposarcomas and a small profile for myxofibrosarcomas as well as a new classification of angiosarcomas were defined. We systematically searched for activated signal transduction pathways in sarcoma cell lines and xenograft transplant models and candidate targets for molecular therapies were identified. Based on these results first clinical studies have been initiated by the German Interdisciplinary Sarcoma Study Group (GISG).

Comparison of performance of one-color and two-color gene-expression analyses in predicting clinical endpoints of neuroblastoma patients.

Microarray-based prediction of clinical endpoints may be performed using either a one-color approach reflecting mRNA abundance in absolute intensity values or a two-color approach yielding ratios of fluorescent intensities. In this study, as part of the MAQC-II project, we systematically compared the classification performance resulting from one- and two-color gene-expression profiles of 478 neuroblastoma samples. In total, 196 classification models were applied to these measurements to predict four clinical endpoints, and classification performances were compared in terms of accuracy, area under the curve, Matthews correlation coefficient and root mean-squared error. Whereas prediction performance varied with distinct clinical endpoints and classification models, equivalent performance metrics were observed for one- and two-color measurements in both internal and external validation. Furthermore, overlap of selected signature genes correlated inversely with endpoint prediction difficulty. In summary, our data strongly substantiate that the choice of platform is not a primary factor for successful gene expression based-prediction of clinical endpoints.

Integrated genomic profiling identifies two distinct molecular subtypes with divergent outcome in neuroblastoma with loss of chromosome 11q.

Imbalances in chromosome 11q occur in approximately 30% of primary neuroblastoma and are associated with poor outcome. It has been suggested that 11q loss constitutes a distinct clinico-genetic neuroblastoma subgroup by affecting expression levels of corresponding genes. This study analysed the relationship of 11q loss, clinical phenotype and global transcriptomic profiles in four clinico-genetic subgroups (11q alteration/favourable outcome, n=7; 11q alteration/unfavourable outcome, n=14; no 11q alteration/favourable outcome, n=81; no 11q alteration/unfavourable outcome, n=8; tumours with MYCN amplification and/or 1p loss were excluded). Unsupervised and supervised comparisons of gene expression profiles consistently showed significantly different mRNA patterns between favourable and unfavourable neuroblastomas, both in the subgroups with and without 11q loss. In contrast, favourable tumours with and without 11q loss showed highly similar transcriptomic profiles. Disproportionate downregulation of 11q genes was observed only in unfavourable tumours with 11q loss. The diverging molecular profiles were neither caused by considerable differences in the size of the deleted regions nor by differential methylation patterns of 11q genes. Together, this study shows that neuroblastoma with 11q loss comprises two biological subgroups that differ both in their clinical phenotype and gene expression patterns, indicating that 11q loss is not a primary determinant of neuroblastoma tumour behaviour.

The hematopoietic stem cell in chronic phase CML is characterized by a transcriptional profile resembling normal myeloid progenitor cells and reflecting loss of quiescence.

We found that composition of cell subsets within the CD34+ cell population is markedly altered in chronic phase (CP) chronic myeloid leukemia (CML). Specifically, proportions and absolute cell counts of common myeloid progenitors (CMP) and megakaryocyte-erythrocyte progenitors (MEP) are significantly greater in comparison to normal bone marrow whereas absolute numbers of hematopoietic stem cells (HSC) are equal. To understand the basis for this, we performed gene expression profiling (Affymetrix HU-133A 2.0) of the distinct CD34+ cell subsets from six patients with CP CML and five healthy donors. Euclidean distance analysis revealed a remarkable transcriptional similarity between the CML patients' HSC and normal progenitors, especially CMP. CP CML HSC were transcriptionally more similar to their progeny than normal HSC to theirs, suggesting a more mature phenotype. Hence, the greatest differences between CP CML patients and normal donors were apparent in HSC including downregulation of genes encoding adhesion molecules, transcription factors, regulators of stem-cell fate and inhibitors of cell proliferation in CP CML. Impaired adhesive and migratory capacities were functionally corroborated by fibronectin detachment analysis and transwell assays, respectively. Based on our findings we propose a loss of quiescence of the CML HSC on detachment from the niche leading to expansion of myeloid progenitors.

Expression of the tumour suppressor gene CADM1 is associated with favourable outcome and inhibits cell survival in neuroblastoma.

Cell adhesion molecule 1 (CADM1) is a putative tumour suppressor gene, which is downregulated in many solid tumours. In neuroblastoma, loss of CADM1 expression has recently been found in disseminated tumours with adverse outcome, prompting us to investigate its role in neuroblastoma tumour progression. Oligonucleotide-microarray analysis of 251 neuroblastoma specimens demonstrated that CADM1 downregulation is associated with unfavourable prognostic markers like disseminated stage 4, age >18 months, MYCN amplification and chromosome 11q alterations (P<0.001 each). Furthermore, low CADM1 expression was significantly correlated with unfavourable gene expression-based classification (P<0.001) and adverse patient outcome (P<0.001). Bisulphite sequencing and genetic analysis of 18 primary neuroblastomas suggested that neither haploinsufficiency nor hypermethylation is regularly involved in CADM1 gene silencing in neuroblastoma, which is in contrast to results obtained in other malignancies. In addition, no mutations disrupting the CADM1 reading frame were found in 25 primary neuroblastomas. Over-expression of CADM1 in neuroblastoma cells resulted in significant reduction of proliferation, viability and colony formation in soft agar. Collectively, our results suggest that downregulation of CADM1 tumour suppressor gene expression is a critical event in neuroblastoma pathogenesis resulting in tumour progression and unfavourable patient outcome.

Molecular signature of CD34(+) hematopoietic stem and progenitor cells of patients with CML in chronic phase.

In this study, we provide a molecular signature of highly enriched CD34+ cells from bone marrow of untreated patients with chronic myelogenous leukemia (CML) in chronic phase in comparison with normal CD34+ cells using microarrays covering 8746 genes. Expression data reflected several BCR-ABL-induced effects in primary CML progenitors, such as transcriptional activation of the classical mitogen-activated protein kinase pathway and the phosphoinositide-3 kinase/AKT pathway as well as downregulation of the proapoptotic gene IRF8. Moreover, novel transcriptional changes in comparison with normal CD34+ cells were identified. These include upregulation of genes involved in the transforming growth factorbeta pathway, fetal hemoglobin genes, leptin receptor, sorcin, tissue inhibitor of metalloproteinase 1, the neuroepithelial cell transforming gene 1 and downregulation of selenoprotein P. Additionally, genes associated with early hematopoietic stem cells (HSC) and leukemogenesis such as HoxA9 and MEIS1 were transcriptionally activated. Differential expression of differentiation-associated genes suggested an altered composition of the CD34+ cell population in CML. This was confirmed by subset analyses of chronic phase CML CD34+ cells showing an increase of the proportion of megakaryocyte-erythroid progenitors, whereas the proportion of HSC and granulocyte-macrophage progenitors was decreased in CML. In conclusion, our results give novel insights into the biology of CML and could provide the basis for identification of new therapeutic targets.

Gene expression signature of primary imatinib-resistant chronic myeloid leukemia patients.

Although the selective tyrosine kinase inhibitor imatinib is successfully used in the treatment of chronic myeloid leukemia (CML), inherent mechanisms confer primary resistance to leukemic patients. In order to search for potentially useful genes in predicting cytogenetic response, a retrospective gene expression study was performed. Leukocyte RNA isolated before imatinib from interferon-alpha-pretreated chronic phase CML patients (n=34) with or without major cytogenetic remission (< or =35% Philadelphia (Ph)+ metaphases) during the first year of treatment was comparatively analyzed using Affymetrix U133A chips. Using support vector machines for gene classification, an outcome-specific gene expression signature consisting of 128 genes was identified. Comparative expression data of specific genes point to changes in apoptosis (e.g. casp9, tumor necrosis factor receptor-associated protein 1, hras), DNA repair (msh3, ddb2), oxidative stress protection (glutathione synthetase, paraoxonase 2, vanin 1) and centrosomes (inhibitor of differentiation-1) within primary resistant patients. Independent statistical approaches and quantitative real-time reverse transcriptase-polymerase chain reaction studies support the clinical relevance of gene profiling. In conclusion, this study establishes a candidate predictor of imatinib resistance in interferon-alpha-pretreated CML patients to be subjected to future investigation in a larger independent patient cohort. The resulting expression signature point to involvement of BCR-ABL-independent mechanisms of resistance.

Gene expression profiling of CD34+ cells identifies a molecular signature of chronic myeloid leukemia blast crisis.

Despite recent success in the treatment of early-stage disease, blastic phase (BP) of chronic myeloid leukemia (CML) that is characterized by rapid expansion of therapy-refractory and differentiation-arrested blasts, remains a therapeutic challenge. The development of resistance upon continuous administration of imatinib mesylate is associated with poor prognosis pointing to the need for alternative therapeutic strategies and a better understanding of the molecular mechanisms underlying disease progression. To identify transcriptional signatures that may explain pathological characteristics and aggressive behavior of BP blasts, we performed comparative gene expression profiling on CD34+ Ph+ cells purified from patients with untreated newly diagnosed chronic phase CML (CP, n=11) and from patients in BP (n=9) using Affymetrix oligonucleotide arrays. Supervised microarray data analysis revealed 114 differentially expressed genes (P<10(-4)), 34 genes displaying more than two-fold transcriptional changes when comparing CP and BP groups. While 24 of these genes were downregulated, 10 genes, especially suppressor of cytokine signalling 2 (SOCS2), CAMPATH-1 antigen (CD52), and four human leukocyte antigen-related genes were strongly overexpressed in BP. Expression of selected genes was validated by real-time-polymerase chain reaction and flow cytometry. Our data suggest the existence of a common gene expression profile of CML-BP and provide new insight into the molecular phenotype of blasts associated with disease progression and high malignancy.

Microarray annotation and biological information on function.

OBJECTIVES: Many methods for statistical analysis of gene expression studies by DNA microarrays produce lists of genes as output. To understand gene lists in terms of traditional biology, e.g. which pathways may be affected, it is necessary to get appropriate annotations for the probes on an array. METHODS: Problems arise with the different sources that have been used by manufacturers to design microarray probes, and their association to biological entities like genes, transcripts and proteins. Function annotation is of crucial importance, and systems like Gene Ontology can be used for this purpose. It arranges annotation terms in a hierarchical manner and thus makes annotations in a gene list amenable to automated analysis. RESULTS: Several methods for analyses of gene function are described. The hierarchical nature of systems like Gene Ontology particularly suggests using methods from graph theory. CONCLUSIONS: The main problem in annotating microarray probes and inferring affected functional modules is the incompleteness and degree of error in current biological databases. Initial approaches to make use of functional annotation exist, but have to be extended, in particular with respect to estimating the statistical significance of results.