Repotrectinib in ROS1 Fusion-Positive Non-Small-Cell Lung Cancer.

BACKGROUND: The early-generation ROS1 tyrosine kinase inhibitors (TKIs) that are approved for the treatment of ROS1 fusion-positive non-small-cell lung cancer (NSCLC) have antitumor activity, but resistance develops in tumors, and intracranial activity is suboptimal. Repotrectinib is a next-generation ROS1 TKI with preclinical activity against ROS1 fusion-positive cancers, including those with resistance mutations such as ROS1 G2032R. METHODS: In this registrational phase 1-2 trial, we assessed the efficacy and safety of repotrectinib in patients with advanced solid tumors, including ROS1 fusion-positive NSCLC. The primary efficacy end point in the phase 2 trial was confirmed objective response; efficacy analyses included patients from phase 1 and phase 2. Duration of response, progression-free survival, and safety were secondary end points in phase 2. RESULTS: On the basis of results from the phase 1 trial, the recommended phase 2 dose of repotrectinib was 160 mg daily for 14 days, followed by 160 mg twice daily. Response occurred in 56 of the 71 patients (79%; 95% confidence interval [CI], 68 to 88) with ROS1 fusion-positive NSCLC who had not previously received a ROS1 TKI; the median duration of response was 34.1 months (95% CI, 25.6 to could not be estimated), and median progression-free survival was 35.7 months (95% CI, 27.4 to could not be estimated). Response occurred in 21 of the 56 patients (38%; 95% CI, 25 to 52) with ROS1 fusion-positive NSCLC who had previously received one ROS1 TKI and had never received chemotherapy; the median duration of response was 14.8 months (95% CI, 7.6 to could not be estimated), and median progression-free survival was 9.0 months (95% CI, 6.8 to 19.6). Ten of the 17 patients (59%; 95% CI, 33 to 82) with the ROS1 G2032R mutation had a response. A total of 426 patients received the phase 2 dose; the most common treatment-related adverse events were dizziness (in 58% of the patients), dysgeusia (in 50%), and paresthesia (in 30%), and 3% discontinued repotrectinib owing to treatment-related adverse events. CONCLUSIONS: Repotrectinib had durable clinical activity in patients with ROS1 fusion-positive NSCLC, regardless of whether they had previously received a ROS1 TKI. Adverse events were mainly of low grade and compatible with long-term administration. (Funded by Turning Point Therapeutics, a wholly owned subsidiary of Bristol Myers Squibb; TRIDENT-1 ClinicalTrials.gov number, NCT03093116.).

North American ATLL has a distinct mutational and transcriptional profile and responds to epigenetic therapies.

Adult T-cell leukemia lymphoma (ATLL) is a rare T cell neoplasm that is endemic in Japanese, Caribbean, and Latin American populations. Most North American ATLL patients are of Caribbean descent and are characterized by high rates of chemo-refractory disease and worse prognosis compared with Japanese ATLL. To determine genomic differences between these 2 cohorts, we performed targeted exon sequencing on 30 North American ATLL patients and compared the results with the Japanese ATLL cases. Although the frequency of TP53 mutations was comparable, the mutation frequency in epigenetic and histone modifying genes (57%) was significantly higher, whereas the mutation frequency in JAK/STAT and T-cell receptor/NF-κB pathway genes was significantly lower. The most common type of epigenetic mutation is that affecting EP300 (20%). As a category, epigenetic mutations were associated with adverse prognosis. Dissimilarities with the Japanese cases were also revealed by RNA sequencing analysis of 9 primary patient samples. ATLL samples with a mutated EP300 gene have decreased total and acetyl p53 protein and a transcriptional signature reminiscent of p53-mutated cancers. Most importantly, decitabine has highly selective single-agent activity in the EP300-mutated ATLL samples, suggesting that decitabine treatment induces a synthetic lethal phenotype in EP300-mutated ATLL cells. In conclusion, we demonstrate that North American ATLL has a distinct genomic landscape that is characterized by frequent epigenetic mutations that are targetable preclinically with DNA methyltransferase inhibitors.

MUMAL2: Improving sensitivity in shotgun proteomics using cost sensitive artificial neural networks and a threshold selector algorithm.

BACKGROUND: This work presents a machine learning strategy to increase sensitivity in tandem mass spectrometry (MS/MS) data analysis for peptide/protein identification. MS/MS yields thousands of spectra in a single run which are then interpreted by software. Most of these computer programs use a protein database to match peptide sequences to the observed spectra. The peptide-spectrum matches (PSMs) must also be assessed by computational tools since manual evaluation is not practicable. The target-decoy database strategy is largely used for error estimation in PSM assessment. However, in general, that strategy does not account for sensitivity. RESULTS: In a previous study, we proposed the method MUMAL that applies an artificial neural network to effectively generate a model to classify PSMs using decoy hits with increased sensitivity. Nevertheless, the present approach shows that the sensitivity can be further improved with the use of a cost matrix associated with the learning algorithm. We also demonstrate that using a threshold selector algorithm for probability adjustment leads to more coherent probability values assigned to the PSMs. Our new approach, termed MUMAL2, provides a two-fold contribution to shotgun proteomics. First, the increase in the number of correctly interpreted spectra in the peptide level augments the chance of identifying more proteins. Second, the more appropriate PSM probability values that are produced by the threshold selector algorithm impact the protein inference stage performed by programs that take probabilities into account, such as ProteinProphet. Our experiments demonstrate that MUMAL2 reached around 15% of improvement in sensitivity compared to the best current method. Furthermore, the area under the ROC curve obtained was 0.93, demonstrating that the probabilities generated by our model are in fact appropriate. Finally, Venn diagrams comparing MUMAL2 with the best current method show that the number of exclusive peptides found by our method was nearly 4-fold higher, which directly impacts the proteome coverage. CONCLUSIONS: The inclusion of a cost matrix and a probability threshold selector algorithm to the learning task further improves the target-decoy database analysis for identifying peptides, which optimally contributes to the challenging task of protein level identification, resulting in a powerful computational tool for shotgun proteomics.

MUMAL: multivariate analysis in shotgun proteomics using machine learning techniques.

BACKGROUND: The shotgun strategy (liquid chromatography coupled with tandem mass spectrometry) is widely applied for identification of proteins in complex mixtures. This method gives rise to thousands of spectra in a single run, which are interpreted by computational tools. Such tools normally use a protein database from which peptide sequences are extracted for matching with experimentally derived mass spectral data. After the database search, the correctness of obtained peptide-spectrum matches (PSMs) needs to be evaluated also by algorithms, as a manual curation of these huge datasets would be impractical. The target-decoy database strategy is largely used to perform spectrum evaluation. Nonetheless, this method has been applied without considering sensitivity, i.e., only error estimation is taken into account. A recently proposed method termed MUDE treats the target-decoy analysis as an optimization problem, where sensitivity is maximized. This method demonstrates a significant increase in the retrieved number of PSMs for a fixed error rate. However, the MUDE model is constructed in such a way that linear decision boundaries are established to separate correct from incorrect PSMs. Besides, the described heuristic for solving the optimization problem has to be executed many times to achieve a significant augmentation in sensitivity. RESULTS: Here, we propose a new method, termed MUMAL, for PSM assessment that is based on machine learning techniques. Our method can establish nonlinear decision boundaries, leading to a higher chance to retrieve more true positives. Furthermore, we need few iterations to achieve high sensitivities, strikingly shortening the running time of the whole process. Experiments show that our method achieves a considerably higher number of PSMs compared with standard tools such as MUDE, PeptideProphet, and typical target-decoy approaches. CONCLUSION: Our approach not only enhances the computational performance, and thus the turn around time of MS-based experiments in proteomics, but also improves the information content with benefits of a higher proteome coverage. This improvement, for instance, increases the chance to identify important drug targets or biomarkers for drug development or molecular diagnostics.

QuACN: an R package for analyzing complex biological networks quantitatively.

MOTIVATION: Network-based representations of biological data have become an important way to analyze high-throughput data. To interpret the large amount of data that is produced by different high-throughput technologies, networks offer multifaceted aspects to analyze the data. As networks represent biological relationships within their structure, it turned out to be fruitful to analyze their topology. Therefore, we developed a freely available, open source R-package called Quantitative Analysis of Complex Networks (QuACN) to meet this challenge. QuACN contains different, information-theoretic and non-information-theoretic, topological network descriptors to analyze, classify and compare biological networks. AVAILABILITY: QuACN is freely available under LGPL via CRAN (http://cran.r-project.org/web/packages/QuACN/).

A network-based feature selection approach to identify metabolic signatures in disease.

The identification and interpretation of metabolic biomarkers is a challenging task. In this context, network-based approaches have become increasingly a key technology in systems biology allowing to capture complex interactions in biological systems. In this work, we introduce a novel network-based method to identify highly predictive biomarker candidates for disease. First, we infer two different types of networks: (i) correlation networks, and (ii) a new type of network called ratio networks. Based on these networks, we introduce scores to prioritize features using topological descriptors of the vertices. To evaluate our method we use an example dataset where quantitative targeted MS/MS analysis was applied to a total of 52 blood samples from 22 persons with obesity (BMI >30) and 30 healthy controls. Using our network-based feature selection approach we identified highly discriminating metabolites for obesity (F-score >0.85, accuracy >85%), some of which could be verified by the literature.

The liver toxicity biomarker study phase I: markers for the effects of tolcapone or entacapone.

The Liver Toxicity Biomarker Study is a systems toxicology approach to discover biomarkers that are indicative of a drug's potential to cause human idiosyncratic drug-induced liver injury. In phase I, the molecular effects in rat liver and blood plasma induced by tolcapone (a "toxic" drug) were compared with the molecular effects in the same tissues by dosing with entacapone (a "clean" drug, similar to tolcapone in chemical structure and primary pharmacological mechanism). Two durations of drug exposure, 3 and 28 days, were employed. Comprehensive molecular analysis of rat liver and plasma samples yielded marker analytes for various drug-vehicle or drug-drug comparisons. An important finding was that the marker analytes associated with tolcapone only partially overlapped with marker analytes associated with entacapone, despite the fact that both drugs have similar chemical structures and the same primary pharmacological mechanism of action. This result indicates that the molecular analyses employed in the study are detecting substantial "off-target" markers for the two drugs. An additional interesting finding was the modest overlap of the marker data sets for 3-day exposure and 28-day exposure, indicating that the molecular changes in liver and plasma caused by short- and long-term drug treatments do not share common characteristics.

Effects of pooling samples on the performance of classification algorithms: a comparative study.

A pooling design can be used as a powerful strategy to compensate for limited amounts of samples or high biological variation. In this paper, we perform a comparative study to model and quantify the effects of virtual pooling on the performance of the widely applied classifiers, support vector machines (SVMs), random forest (RF), k-nearest neighbors (k-NN), penalized logistic regression (PLR), and prediction analysis for microarrays (PAMs). We evaluate a variety of experimental designs using mock omics datasets with varying levels of pool sizes and considering effects from feature selection. Our results show that feature selection significantly improves classifier performance for non-pooled and pooled data. All investigated classifiers yield lower misclassification rates with smaller pool sizes. RF mainly outperforms other investigated algorithms, while accuracy levels are comparable among all the remaining ones. Guidelines are derived to identify an optimal pooling scheme for obtaining adequate predictive power and, hence, to motivate a study design that meets best experimental objectives and budgetary conditions, including time constraints.

Structural measures for network biology using QuACN.

BACKGROUND: Structural measures for networks have been extensively developed, but many of them have not yet demonstrated their sustainably. That means, it remains often unclear whether a particular measure is useful and feasible to solve a particular problem in network biology. Exemplarily, the classification of complex biological networks can be named, for which structural measures are used leading to a minimal classification error. Hence, there is a strong need to provide freely available software packages to calculate and demonstrate the appropriate usage of structural graph measures in network biology. RESULTS: Here, we discuss topological network descriptors that are implemented in the R-package QuACN and demonstrate their behavior and characteristics by applying them to a set of example graphs. Moreover, we show a representative application to illustrate their capabilities for classifying biological networks. In particular, we infer gene regulatory networks from microarray data and classify them by methods provided by QuACN. Note that QuACN is the first freely available software written in R containing a large number of structural graph measures. CONCLUSION: The R package QuACN is under ongoing development and we add promising groups of topological network descriptors continuously. The package can be used to answer intriguing research questions in network biology, e.g., classifying biological data or identifying meaningful biological features, by analyzing the topology of biological networks.

Molecular Characteristics of Repotrectinib That Enable Potent Inhibition of TRK Fusion Proteins and Resistant Mutations.

NTRK chromosomal rearrangements yield oncogenic TRK fusion proteins that are sensitive to TRK inhibitors (larotrectinib and entrectinib) but often mutate, limiting the durability of response for NTRK + patients. Next-generation inhibitors with compact macrocyclic structures (repotrectinib and selitrectinib) were designed to avoid resistance mutations. Head-to-head potency comparisons of TRK inhibitors and molecular characterization of binding interactions are incomplete, obscuring a detailed understanding of how molecular characteristics translate to potency. Larotrectinib, entrectinib, selitrectinib, and repotrectinib were characterized using cellular models of wild-type TRKA/B/C fusions and resistance mutant variants with a subset evaluated in xenograft tumor models. Crystal structures were determined for repotrectinib bound to TRKA (wild-type, solvent-front mutant). TKI-naïve and pretreated case studies are presented. Repotrectinib was the most potent inhibitor of wild-type TRKA/B/C fusions and was more potent than selitrectinib against all tested resistance mutations, underscoring the importance of distinct features of the macrocycle structures. Cocrystal structures of repotrectinib with wild-type TRKA and the TRKAG595R SFM variant elucidated how differences in macrocyclic inhibitor structure, binding orientation, and conformational flexibility affect potency and mutant selectivity. The SFM crystal structure revealed an unexpected intramolecular arginine sidechain interaction. Repotrectinib caused tumor regression in LMNA-NTRK1 xenograft models harboring GKM, SFM, xDFG, and GKM + SFM compound mutations. Durable responses were observed in TKI-naïve and -pretreated patients with NTRK + cancers treated with repotrectinib (NCT03093116). This comprehensive analysis of first- and second-generation TRK inhibitors informs the clinical utility, structural determinants of inhibitor potency, and design of new generations of macrocyclic inhibitors.

Sample size and statistical power considerations in high-dimensionality data settings: a comparative study of classification algorithms.

BACKGROUND: data generated using 'omics' technologies are characterized by high dimensionality, where the number of features measured per subject vastly exceeds the number of subjects in the study. In this paper, we consider issues relevant in the design of biomedical studies in which the goal is the discovery of a subset of features and an associated algorithm that can predict a binary outcome, such as disease status. We compare the performance of four commonly used classifiers (K-Nearest Neighbors, Prediction Analysis for Microarrays, Random Forests and Support Vector Machines) in high-dimensionality data settings. We evaluate the effects of varying levels of signal-to-noise ratio in the dataset, imbalance in class distribution and choice of metric for quantifying performance of the classifier. To guide study design, we present a summary of the key characteristics of 'omics' data profiled in several human or animal model experiments utilizing high-content mass spectrometry and multiplexed immunoassay based techniques. RESULTS: the analysis of data from seven 'omics' studies revealed that the average magnitude of effect size observed in human studies was markedly lower when compared to that in animal studies. The data measured in human studies were characterized by higher biological variation and the presence of outliers. The results from simulation studies indicated that the classifier Prediction Analysis for Microarrays (PAM) had the highest power when the class conditional feature distributions were Gaussian and outcome distributions were balanced. Random Forests was optimal when feature distributions were skewed and when class distributions were unbalanced. We provide a free open-source R statistical software library (MVpower) that implements the simulation strategy proposed in this paper. CONCLUSION: no single classifier had optimal performance under all settings. Simulation studies provide useful guidance for the design of biomedical studies involving high-dimensionality data.

MUDE: a new approach for optimizing sensitivity in the target-decoy search strategy for large-scale peptide/protein identification.

The target-decoy search strategy has been successfully applied in shotgun proteomics for validating peptide and protein identifications. If, on one hand, this method has proven to be very efficient for error estimation, on the other hand, little attention has been paid to the resulting sensitivity. Only two scores are normally used and thresholds are explored in a very simplistic way. In this work, a multivariate decoy analysis is described, where many quality parameters are considered. This analysis is treated in our approach as an optimization problem for sensitivity maximization. Furthermore, an efficient heuristic is proposed to solve this problem. Experiments comparing our method, termed MUDE (multivariate decoy database analysis), with traditional bivariate decoy analysis and with Peptide/ProteinProphet showed that our procedure significantly enhances the retrieved number of identifications when comparing the same false discovery rates. Particularly for phosphopeptide/protein identifications, we could demonstrate more than a two-fold increase in sensitivity compared with the Trans-Proteomic Pipeline tools.

Novel topological descriptors for analyzing biological networks.

BACKGROUND: Topological descriptors, other graph measures, and in a broader sense, graph-theoretical methods, have been proven as powerful tools to perform biological network analysis. However, the majority of the developed descriptors and graph-theoretical methods does not have the ability to take vertex- and edge-labels into account, e.g., atom- and bond-types when considering molecular graphs. Indeed, this feature is important to characterize biological networks more meaningfully instead of only considering pure topological information. RESULTS: In this paper, we put the emphasis on analyzing a special type of biological networks, namely bio-chemical structures. First, we derive entropic measures to calculate the information content of vertex- and edge-labeled graphs and investigate some useful properties thereof. Second, we apply the mentioned measures combined with other well-known descriptors to supervised machine learning methods for predicting Ames mutagenicity. Moreover, we investigate the influence of our topological descriptors - measures for only unlabeled vs. measures for labeled graphs - on the prediction performance of the underlying graph classification problem. CONCLUSIONS: Our study demonstrates that the application of entropic measures to molecules representing graphs is useful to characterize such structures meaningfully. For instance, we have found that if one extends the measures for determining the structural information content of unlabeled graphs to labeled graphs, the uniqueness of the resulting indices is higher. Because measures to structurally characterize labeled graphs are clearly underrepresented so far, the further development of such methods might be valuable and fruitful for solving problems within biological network analysis.

The impact of storage effects in biobanks on biomarker discovery in systems biology studies.

Sample handling and storage conditions in specimens frozen over long periods of time can severely impact marker levels. If laboratory technologies, practices and related protocols change over time, biomarker studies are potentially biased and report erroneous results. These issues and pitfalls are often overlooked in system biology studies using previously collected and stored materials, and are likely to be one notable cause for biomarker candidates failing to be validated. We present results from simulation studies quantifying the loss in statistical power to detect true biomarkers, due to diminishing concentration of analytes in samples subject to poor handling and storage conditions.

A new rule-based algorithm for identifying metabolic markers in prostate cancer using tandem mass spectrometry.

MOTIVATION: Prostate cancer is the most prevalent tumor in males and its incidence is expected to increase as the population ages. Prostate cancer is treatable by excision if detected at an early enough stage. The challenges of early diagnosis require the discovery of novel biomarkers and tools for prostate cancer management. RESULTS: We developed a novel feature selection algorithm termed as associative voting (AV) for identifying biomarker candidates in prostate cancer data measured via targeted metabolite profiling MS/MS analysis. We benchmarked our algorithm against two standard entropy-based and correlation-based feature selection methods [Information Gain (IG) and ReliefF (RF)] and observed that, on a variety of classification tasks in prostate cancer diagnosis, our algorithm identified subsets of biomarker candidates that are both smaller and show higher discriminatory power than the subsets identified by IG and RF. A literature study confirms that the highest ranked biomarker candidates identified by AV have independently been identified as important factors in prostate cancer development. AVAILABILITY: The algorithm can be downloaded from the following http://biomed.umit.at/page.cfm?pageid=516.

The liver toxicity biomarker study: phase I design and preliminary results.

Drug-induced liver injury (DILI) is the primary adverse event that results in withdrawal of drugs from the market and a frequent reason for the failure of drug candidates in development. The Liver Toxicity Biomarker Study (LTBS) is an innovative approach to investigate DILI because it compares molecular events produced in vivo by compound pairs that (a) are similar in structure and mechanism of action, (b) are associated with few or no signs of liver toxicity in preclinical studies, and (c) show marked differences in hepatotoxic potential. The LTBS is a collaborative preclinical research effort in molecular systems toxicology between the National Center for Toxicological Research and BG Medicine, Inc., and is supported by seven pharmaceutical companies and three technology providers. In phase I of the LTBS, entacapone and tolcapone were studied in rats to provide results and information that will form the foundation for the design and implementation of phase II. Molecular analysis of the rat liver and plasma samples combined with statistical analyses of the resulting datasets yielded marker analytes, illustrating the value of the broad-spectrum, molecular systems analysis approach to studying pharmacological or toxicological effects.

Stem cell mutations can be detected in myeloma patients years before onset of secondary leukemias.

Therapy-related acute myeloid leukemia and myelodysplastic syndromes (t-AML/t-MDS) are secondary hematologic malignancies associated with poor prognosis, warranting insights into their predisposing conditions and cells of origin. We identified patients with myeloma who developed t-AML/t-MDS and analyzed their stem and progenitor cells collected years before the onset of secondary disease. We demonstrate that aberrant stem cells with high CD123 expression can be detected long before the onset of overt leukemia. Rigorous sorting, followed by targeted sequencing, resulted in ultradeep functional depth of sequencing and revealed preexisting mutant hematopoietic stem cell (HSC) clones, mainly harboring TP53 mutations, that became the dominant population at the time of leukemic presentation. Taken together, these data show that HSCs can act as reservoirs for leukemia-initiating cells many years before the onset of myeloid leukemia.

Bioinformatics analysis of targeted metabolomics--uncovering old and new tales of diabetic mice under medication.

Metabolomics is a powerful tool for identifying both known and new disease-related perturbations in metabolic pathways. In preclinical drug testing, it has a high potential for early identification of drug off-target effects. Recent advances in high-precision high-throughput mass spectrometry have brought the metabolomic field to a point where quantitative, targeted, metabolomic measurements with ready-to-use kits allow for the automated in-house screening for hundreds of different metabolites in large sets of biological samples. Today, the field of metabolomics is, arguably, at a point where transcriptomics was about 5 yr ago. This being so, the field has a strong need for adapted bioinformatics tools and methods. In this paper we describe a systematic analysis of a targeted quantitative characterization of more than 800 metabolites in blood plasma samples from healthy and diabetic mice under rosiglitazone treatment. We show that known and new metabolic phenotypes of diabetes and medication can be recovered in a statistically objective manner. We find that concentrations of methylglutaryl carnitine are oppositely impacted by rosiglitazone treatment of both healthy and diabetic mice. Analyzing ratios between metabolite concentrations dramatically reduces the noise in the data set, allowing for the discovery of new potential biomarkers of diabetes, such as the N-hydroxyacyloylsphingosyl-phosphocholines SM(OH)28:0 and SM(OH)26:0. Using a hierarchical clustering technique on partial eta(2) values, we identify functionally related groups of metabolites, indicating a diabetes-related shift from lysophosphatidylcholine to phosphatidylcholine levels. The bioinformatics data analysis approach introduced here can be readily generalized to other drug testing scenarios and other medical disorders.

Distinct transcriptomic and exomic abnormalities within myelodysplastic syndrome marrow cells.

To provide biologic insights into mechanisms underlying myelodysplastic syndromes (MDS) we evaluated the CD34+ marrow cells transcriptome using high-throughput RNA sequencing (RNA-Seq). We demonstrated significant differential gene expression profiles (GEPs) between MDS and normal and identified 41 disease classifier genes. Additionally, two main clusters of GEPs distinguished patients based on their major clinical features, particularly between those whose disease remained stable versus patients who transformed into acute myeloid leukemia within 12 months. The genes whose expression was associated with disease outcome were involved in functional pathways and biologic processes highly relevant for MDS. Combined with exomic analysis we identified differential isoform usage of genes in MDS mutational subgroups, with consequent dysregulation of distinct biologic functions. This combination of clinical, transcriptomic and exomic findings provides valuable understanding of mechanisms underlying MDS and its progression to a more aggressive stage and also facilitates prognostic characterization of MDS patients.