Development and validation of ACTE-MTB: A tool to systematically assess the maturity of molecular tumor boards.

Molecular tumor boards (MTBs) require specialized activities to leverage genomic data for therapeutic decision-making. Currently, there are no defined standards for implementing, executing, and tracking the impact of MTBs. This study describes the development and validation of ACTE-MTB, a tool to evaluate the maturity of an organization's MTB to identify specific areas that would benefit from process improvements and standardization. The ACTE-MTB maturity assessment tool is composed of 3 elements: 1) The ACTE-MTB maturity model; 2) a 59-question survey on MTB processes and challenges; and 3) a 5-level MTB maturity scoring algorithm. This tool was developed to measure MTB maturity in the categories of Access, Consultation, Technology, and Evidence (ACTE) and was tested on 20 MTBs spanning the United States, Europe, and Asia-Pacific regions. Validity testing revealed that the average maturity score was 3.3 out of 5 (+/- 0.1; range 2.0-4.3) with MTBs in academic institutions showing significantly higher overall maturity levels than in non-academic institutions (3.7 +/- 0.2 vs. 3.1 +/- 0.2; P = .018). While maturity scores for academic institutions were higher for Consultation, Technology, and Evidence domains, the maturity score for the Access domain did not significantly differ between the two groups, highlighting a disconnect between MTB operations and the downstream impact on ability to access testing and/or therapies. To our knowledge, ACTE-MTB is the first tool of its kind to enable structured, maturity assessment of MTBs in a universally-applicable manner. In the process of establishing construct validity of this tool, opportunities for further investigation and improvements were identified that address the key functional areas of MTBs that would likely benefit from standardization and best practice recommendations. We believe a unified approach to assessment of MTB maturity will help to identify areas for improvement at both the organizational and system level.

Development of a "meta-model" to address missing data, predict patient-specific cancer survival and provide a foundation for clinical decision support.

OBJECTIVE: Like most real-world data, electronic health record (EHR)-derived data from oncology patients typically exhibits wide interpatient variability in terms of available data elements. This interpatient variability leads to missing data and can present critical challenges in developing and implementing predictive models to underlie clinical decision support for patient-specific oncology care. Here, we sought to develop a novel ensemble approach to addressing missing data that we term the "meta-model" and apply the meta-model to patient-specific cancer prognosis. MATERIALS AND METHODS: Using real-world data, we developed a suite of individual random survival forest models to predict survival in patients with advanced lung cancer, colorectal cancer, and breast cancer. Individual models varied by the predictor data used. We combined models for each cancer type into a meta-model that predicted survival for each patient using a weighted mean of the individual models for which the patient had all requisite predictors. RESULTS: The meta-model significantly outperformed many of the individual models and performed similarly to the best performing individual models. Comparisons of the meta-model to a more traditional imputation-based method of addressing missing data supported the meta-model's utility. CONCLUSIONS: We developed a novel machine learning-based strategy to underlie clinical decision support and predict survival in cancer patients, despite missing data. The meta-model may more generally provide a tool for addressing missing data across a variety of clinical prediction problems. Moreover, the meta-model may address other challenges in clinical predictive modeling including model extensibility and integration of predictive algorithms trained across different institutions and datasets.

Reduced expression of ribosomal proteins relieves microRNA-mediated repression.

MicroRNAs (miRNAs) regulate physiological and pathological processes by inducing posttranscriptional repression of target messenger RNAs (mRNAs) via incompletely understood mechanisms. To discover factors required for human miRNA activity, we performed an RNAi screen using a reporter cell line of miRNA-mediated repression of translation initiation. We report that reduced expression of ribosomal protein genes (RPGs) dissociated miRNA complexes from target mRNAs, leading to increased polysome association, translation, and stability of miRNA-targeted mRNAs relative to untargeted mRNAs. RNA sequencing of polysomes indicated substantial overlap in sets of genes exhibiting increased or decreased polysomal association after Argonaute or RPG knockdowns, suggesting similarity in affected pathways. miRNA profiling of monosomes and polysomes demonstrated that miRNAs cosediment with ribosomes. RPG knockdowns decreased miRNAs in monosomes and increased their target mRNAs in polysomes. Our data show that most miRNAs repress translation and that the levels of RPGs modulate miRNA-mediated repression of translation initiation.

Not miR-ly small RNAs: big potential for microRNAs in therapy.

RNA interference (RNAi) describes a set of natural processes in which genes are silenced by small RNAs. RNAi has been widely used as an experimental tool that has recently become the focus of drug development efforts to treat a variety of diseases and disorders. Like all molecular therapies, in vivo delivery is the major hurdle to realizing therapeutic RNAi. Several strategies have been developed that increase small RNA half-life in the blood, facilitate transduction across biological membranes, and mediate cell-specific delivery. Importantly, these strategies permit targeting of mRNAs as well as microRNAs (miRNAs), a class of small RNAs encoded in the genome. miRNAs are required for multiple developmental and cellular processes. Dysfunction of miRNAs can result in a host of pathologies, suggesting that miRNAs are potential targets of therapy. Recent studies of miRNA function in immune-specific pathways indicate that specific miRNAs might be exploited as therapeutic targets to treat immune disorders, including autoimmunity, allergy, and hematopoietic cancers.

Chronic lymphocytic leukemia requires BCL2 to sequester prodeath BIM, explaining sensitivity to BCL2 antagonist ABT-737.

Antiapoptotic B cell leukemia/lymphoma 2 (BCL2) family proteins are expressed in many cancers, but the circumstances under which these proteins are necessary for tumor maintenance are poorly understood. We exploited a novel functional assay that uses BCL2 homology domain 3 (BH3) peptides to predict dependence on antiapoptotic proteins, a strategy we call BH3 profiling. BH3 profiling accurately predicts sensitivity to BCL2 antagonist ABT-737 in primary chronic lymphocytic leukemia (CLL) cells. BH3 profiling also accurately distinguishes myeloid cell leukemia sequence 1 (MCL1) from BCL2 dependence in myeloma cell lines. We show that the special sensitivity of CLL cells to BCL2 antagonism arises from the requirement that BCL2 tonically sequester proapoptotic BIM in CLL. ABT-737 displaced BIM from BCL2's BH3-binding pocket, allowing BIM to activate BAX, induce mitochondrial permeabilization, and rapidly commit the CLL cell to death. Our experiments demonstrate that BCL2 expression alone does not dictate sensitivity to ABT-737. Instead, BCL2 complexed to BIM is the critical target for ABT-737 in CLL. An important implication is that in cancer, BCL2 may not effectively buffer chemotherapy death signals if it is already sequestering proapoptotic BH3-only proteins. Indeed, activator BH3-only occupation of BCL2 may prime cancer cells for death, offering a potential explanation for the marked chemosensitivity of certain cancers that express abundant BCL2, such as CLL and follicular lymphoma.

Recapitulation of short RNA-directed translational gene silencing in vitro.

microRNAs (miRNAs) are a large class of endogenous short RNAs that repress gene expression. Many miRNAs are conserved throughout evolution, and dysregulation of miRNA pathways has been correlated with an increasing number of human diseases. In animals, miRNAs typically bind to the 3' untranslated region (3'UTR) of target mRNAs with imperfect sequence complementarity and repress translation. Despite their importance in regulating biological processes in numerous organisms, the mechanisms of miRNA function are largely unknown. Here, we report in vitro reactions for miRNA-directed translational gene silencing. These reactions faithfully recapitulate known in vivo hallmarks of mammalian miRNA function, including a requirement for a 5' phosphate and perfect complementarity to the mRNA target in the 5' seed region. Translational gene silencing by miRNAs in vitro requires target mRNAs to possess a 7-methyl G cap and a polyA tail, whereas increasing polyA tail length alone can increase miRNA silencing activity.

Activation of CREB/ATF sites by polyomavirus large T antigen.

Polyomavirus large T antigen (LT) has a direct role in viral replication and a profound effect on cell phenotype. It promotes cell cycle progression, immortalizes primary cells, blocks differentiation, and causes apoptosis. While much of large T function is related to its effects on tumor suppressors of the retinoblastoma susceptibility (Rb) gene family, we have previously shown that activation of the cyclin A promoter can occur through a non-Rb-dependent mechanism. Here we show that activation occurs via an ATF/CREB site. Investigation of the mechanism indicates that large T can synergize with CREB family members to activate transcription. Experiments with Gal4-CREB constructs show that synergy is independent of CREB phosphorylation by protein kinase A. Examination of synergy with Gal4-CREB deletion constructs indicates that large T acts on the constitutive activation domain of CREB. Large T can bind to CREB in vivo. Genetic analysis shows that the DNA-binding domain (residues 264 to 420) is sufficient to activate transcription when it is localized to the nucleus. Further analysis of the DNA-binding domain shows that while site-specific DNA binding is not required, non-site-specific DNA binding is important for the activation. Thus, CREB binding and DNA binding are both important for large T activation of CREB/ATF sites. In contrast to previous models where large T transactivation occurred indirectly, these results also suggest that large T can act directly at promoters to activate transcription.