Bayesian data integration and variable selection for pan-cancer survival prediction using protein expression data

Accurate prognostic prediction using molecular information is a challenging area of research, which is essential to develop precision medicine. In this paper, we develop translational models to identify major actionable proteins that are associated with clinical outcomes, like the survival time of patients. There are considerable statistical and computational challenges due to the large dimension of the problems. Furthermore, data are available for different tumor types; hence data integration for various tumors is desirable. Having censored survival outcomes escalates one more level of complexity in the inferential procedure. We develop Bayesian hierarchical survival models, which accommodate all the challenges mentioned here. We use the hierarchical Bayesian accelerated failure time model for survival regression. Furthermore, we assume sparse horseshoe prior distribution for the regression coefficients to identify the major proteomic drivers. We borrow strength across tumor groups by introducing a correlation structure among the prior distributions. The proposed methods have been used to analyze data from the recently curated “The Cancer Proteome Atlas” (TCPA), which contains reverse-phase protein arrays–based high-quality protein expression data as well as detailed clinical annotation, including survival times. Our simulation and the TCPA data analysis illustrate the efficacy of the proposed integrative model, which links different tumors with the correlated prior structures.     

Integrating multi-platform genomic data using hierarchical Bayesian relevance vector machines

Background

Recent advances in genome technologies and the subsequent collection of genomic information at various molecular resolutions hold promise to accelerate the discovery of new therapeutic targets. A critical step in achieving these goals is to develop efficient clinical prediction models that integrate these diverse sources of high-throughput data. This step is challenging due to the presence of high-dimensionality and complex interactions in the data. For predicting relevant clinical outcomes, we propose a flexible statistical machine learning approach that acknowledges and models the interaction between platform-specific measurements through nonlinear kernel machines and borrows information within and between platforms through a hierarchical Bayesian framework. Our model has parameters with direct interpretations in terms of the effects of platforms and data interactions within and across platforms. The parameter estimation algorithm in our model uses a computationally efficient variational Bayes approach that scales well to large high-throughput datasets.

Results

We apply our methods of integrating gene/mRNA expression and microRNA profiles for predicting patient survival times to The Cancer Genome Atlas (TCGA) based glioblastoma multiforme (GBM) dataset. In terms of prediction accuracy, we show that our non-linear and interaction-based integrative methods perform better than linear alternatives and non-integrative methods that do not account for interactions between the platforms. We also find several prognostic mRNAs and microRNAs that are related to tumor invasion and are known to drive tumor metastasis and severe inflammatory response in GBM. In addition, our analysis reveals several interesting mRNA and microRNA interactions that have known implications in the etiology of GBM.

Conclusions

Our approach gains its flexibility and power by modeling the non-linear interaction structures between and within the platforms. Our framework is a useful tool for biomedical researchers, since clinical prediction using multi-platform genomic information is an important step towards personalized treatment of many cancers. We have a freely available software at: http://odin.mdacc.tmc.edu/~vbaladan.

     

Phase I clinical trial of systemically administered TUSC2(FUS1)-nanoparticles mediating functional gene transfer in humans

Background

Tumor suppressor gene TUSC2/FUS1 (TUSC2) is frequently inactivated early in lung cancer development. TUSC2 mediates apoptosis in cancer cells but not normal cells by upregulation of the intrinsic apoptotic pathway. No drug strategies currently exist targeting loss-of–function genetic abnormalities. We report the first in-human systemic gene therapy clinical trial of tumor suppressor gene TUSC2.

Methods

Patients with recurrent and/or metastatic lung cancer previously treated with platinum-based chemotherapy were treated with escalating doses of intravenous N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium chloride (DOTAP):cholesterol nanoparticles encapsulating a TUSC2 expression plasmid (DOTAP:chol-TUSC2) every 3 weeks.

Results

Thirty-one patients were treated at 6 dose levels (range 0.01 to 0.09 milligrams per kilogram). The MTD was determined to be 0.06 mg/kg. Five patients achieved stable disease (2.6–10.8 months, including 2 minor responses). One patient had a metabolic response on positron emission tomography (PET) imaging. RT-PCR analysis detected TUSC2 plasmid expression in 7 of 8 post-treatment tumor specimens but not in pretreatment specimens and peripheral blood lymphocyte controls. Proximity ligation assay, performed on paired biopsies from 3 patients, demonstrated low background TUSC2 protein staining in pretreatment tissues compared with intense (10–25 fold increase) TUSC2 protein staining in post-treatment tissues. RT-PCR gene expression profiling analysis of apoptotic pathway genes in two patients with high post-treatment levels of TUSC2 mRNA and protein showed significant post-treatment changes in the intrinsic apoptotic pathway. Twenty-nine genes of the 82 tested in the apoptosis array were identified by Igenuity Pathway Analysis to be significantly altered post-treatment in both patients (Pearson correlation coefficient 0.519; p<0.01).

Conclusions

DOTAP:chol-TUSC2 can be safely administered intravenously in lung cancer patients and results in uptake of the gene by human primary and metastatic tumors, transgene and gene product expression, specific alterations in TUSC2-regulated pathways, and anti-tumor effects (to our knowledge for the first time for systemic DOTAP:cholesterol nanoparticle gene therapy).

Trial Registration

ClinicalTrials.gov {"type":"clinical-trial","attrs":{"text":"NCT00059605","term_id":"NCT00059605"}}NCT00059605     

Selective genomic copy number imbalances and probability of recurrence in early-stage breast cancer

A number of studies of copy number imbalances (CNIs) in breast tumors support associations between individual CNIs and patient outcomes. However, no pattern or signature of CNIs has emerged for clinical use. We determined copy number (CN) gains and losses using high-density molecular inversion probe (MIP) arrays for 971 stage I/II breast tumors and applied a boosting strategy to fit hazards models for CN and recurrence, treating chromosomal segments in a dose-specific fashion (-1 [loss], 0 [no change] and +1 [gain]). The concordance index (C-Index) was used to compare prognostic accuracy between a training (n = 728) and test (n = 243) set and across models. Twelve novel prognostic CNIs were identified: losses at 1p12, 12q13.13, 13q12.3, 22q11, and Xp21, and gains at 2p11.1, 3q13.12, 10p11.21, 10q23.1, 11p15, 14q13.2-q13.3, and 17q21.33. In addition, seven CNIs previously implicated as prognostic markers were selected: losses at 8p22 and 16p11.2 and gains at 10p13, 11q13.5, 12p13, 20q13, and Xq28. For all breast cancers combined, the final full model including 19 CNIs, clinical covariates, and tumor marker-approximated subtypes (estrogen receptor [ER], progesterone receptor, ERBB2 amplification, and Ki67) significantly outperformed a model containing only clinical covariates and tumor subtypes (C-Index _{full model}, train[test]  =  0.72[0.71] ± 0.02 vs. C-Index _{clinical + subtype model}, train[test]  =  0.62[0.62] ± 0.02; p<10⁻⁶). In addition, the full model containing 19 CNIs significantly improved prognostication separately for ER–, HER2+, luminal B, and triple negative tumors over clinical variables alone. In summary, we show that a set of 19 CNIs discriminates risk of recurrence among early-stage breast tumors, independent of ER status. Further, our data suggest the presence of specific CNIs that promote and, in some cases, limit tumor spread.     

Bayesian methods for expression-based integration of various types of genomics data

We propose methods to integrate data across several genomic platforms using a hierarchical Bayesian analysis framework that incorporates the biological relationships among the platforms to identify genes whose expression is related to clinical outcomes in cancer. This integrated approach combines information across all platforms, leading to increased statistical power in finding these predictive genes, and further provides mechanistic information about the manner in which the gene affects the outcome. We demonstrate the advantages of the shrinkage estimation used by this approach through a simulation, and finally, we apply our method to a Glioblastoma Multiforme dataset and identify several genes potentially associated with the patients’ survival. We find 12 positive prognostic markers associated with nine genes and 13 negative prognostic markers associated with nine genes.     

Bayesian hierarchical spatially correlated functional data analysis with application to colon carcinogenesis

Summary .   In this article, we present new methods to analyze data from an experiment using rodent models to investigate the role of p27, an important cell-cycle mediator, in early colon carcinogenesis. The responses modeled here are essentially functions nested within a two-stage hierarchy. Standard functional data analysis literature focuses on a single stage of hierarchy and conditionally independent functions with near white noise. However, in our experiment, there is substantial biological motivation for the existence of spatial correlation among the functions, which arise from the locations of biological structures called colonic crypts: this possible functional correlation is a phenomenon we term crypt signaling . Thus, as a point of general methodology, we require an analysis that allows for functions to be correlated at the deepest level of the hierarchy. Our approach is fully Bayesian and uses Markov chain Monte Carlo methods for inference and estimation. Analysis of this data set gives new insights into the structure of p27 expression in early colon carcinogenesis and suggests the existence of significant crypt signaling. Our methodology uses regression splines, and because of the hierarchical nature of the data, dimension reduction of the covariance matrix of the spline coefficients is important: we suggest simple methods for overcoming this problem.     

<Emphasis Type="Italic">P</Emphasis>rediction-<Emphasis Type="Italic">O</Emphasis>riented <Emphasis Type="Italic">M</Emphasis>arker <Emphasis Type="Italic">S</Emphasis>election (PROMISE): With Application to High-Dimensional Regression

In personalized medicine, biomarkers are used to select therapies with the highest likelihood of success based on an individual patient’s biomarker/genomic profile. Two goals are to choose important biomarkers that accurately predict treatment outcomes and to cull unimportant biomarkers to reduce the cost of biological and clinical verifications. These goals are challenging due to the high dimensionality of genomic data. Variable selection methods based on penalized regression (e.g., the lasso and elastic net) have yielded promising results. However, selecting the right amount of penalization is critical to simultaneously achieving these two goals. Standard approaches based on cross-validation (CV) typically provide high prediction accuracy with high true positive rates (TPRs) but at the cost of too many false positives. Alternatively, stability selection (SS) controls the number of false positives, but at the cost of yielding too few true positives. To circumvent these issues, we propose prediction-oriented marker selection (PROMISE), which combines SS with CV to conflate the advantages of both methods. Our application of PROMISE with the lasso and elastic net in data analysis shows that, compared to CV, PROMISE produces sparse solutions, few false positives, and small type I + type II error, and maintains good prediction accuracy, with a marginal decrease in the TPRs. Compared to SS, PROMISE offers better prediction accuracy and TPRs. In summary, PROMISE can be applied in many fields to select regularization parameters when the goals are to minimize false positives and maximize prediction accuracy.