Aki Vehtari

One vs Three Years of Adjuvant Imatinib for Operable Gastrointestinal Stromal Tumor A Randomized Trial

CONTEXT Adjuvant imatinib administered for 12 months after surgery has improved recurrence-free survival (RFS) of patients with operable gastrointestinal stromal tumor (GIST) compared with placebo. OBJECTIVE To investigate the role of imatinib administration duration as adjuvant treatment of patients who have a high estimated risk for GIST recurrence after surgery. DESIGN, SETTING, AND PATIENTS Patients with KIT-positive GIST removed at surgery were entered between February 2004 and September 2008 to this randomized, open-label phase 3 study conducted in 24 hospitals in Finland, Germany, Norway, and Sweden. The risk of GIST recurrence was estimated using the modified National Institutes of Health Consensus Criteria. INTERVENTION Imatinib, 400 mg per day, orally for either 12 months or 36 months, started within 12 weeks of surgery. MAIN OUTCOME MEASURES The primary end point was RFS; the secondary end points included overall survival and treatment safety. RESULTS Two hundred patients were allocated to each group. The median follow-up time after randomization was 54 months in December 2010. Diagnosis of GIST was confirmed in 382 of 397 patients (96%) in the intention-to-treat population at a central pathology review. KIT or PDGFRA mutation was detected in 333 of 366 tumors (91%) available for testing. Patients assigned for 36 months of imatinib had longer RFS compared with those assigned for 12 months (hazard ratio [HR], 0.46; 95% CI, 0.32-0.65; P < .001; 5-year RFS, 65.6% vs 47.9%, respectively) and longer overall survival (HR, 0.45; 95% CI, 0.22-0.89; P = .02; 5-year survival, 92.0% vs 81.7%). Imatinib was generally well tolerated, but 12.6% and 25.8% of patients assigned to the 12- and 36-month groups, respectively, discontinued imatinib for a reason other than GIST recurrence. CONCLUSION Compared with 12 months of adjuvant imatinib, 36 months of imatinib improved RFS and overall survival of GIST patients with a high risk of GIST recurrence. TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT00116935.

Understanding predictive information criteria for Bayesian models

We review the Akaike, deviance, and Watanabe-Akaike information criteria from a Bayesian perspective, where the goal is to estimate expected out-of-sample-prediction error using a bias-corrected adjustment of within-sample error. We focus on the choices involved in setting up these measures, and we compare them in three simple examples, one theoretical and two applied. The contribution of this paper is to put all these information criteria into a Bayesian predictive context and to better understand, through small examples, how these methods can apply in practice.

Risk of recurrence of gastrointestinal stromal tumour after surgery: an analysis of pooled population-based cohorts.

BACKGROUND The risk of recurrence of gastrointestinal stromal tumour (GIST) after surgery needs to be estimated when considering adjuvant systemic therapy. We assessed prognostic factors of patients with operable GIST, to compare widely used risk-stratification schemes and to develop a new method for risk estimation. METHODS Population-based cohorts of patients diagnosed with operable GIST, who were not given adjuvant therapy, were identified from the literature. Data from ten series and 2560 patients were pooled. Risk of tumour recurrence was stratified using the National Institute of Health (NIH) consensus criteria, the modified consensus criteria, and the Armed Forces Institute of Pathology (AFIP) criteria. Prognostic factors were examined using proportional hazards and non-linear models. The results were validated in an independent centre-based cohort consisting of 920 patients with GIST. FINDINGS Estimated 15-year recurrence-free survival (RFS) after surgery was 59·9% (95% CI 56·2-63·6); few recurrences occurred after the first 10 years of follow-up. Large tumour size, high mitosis count, non-gastric location, presence of rupture, and male sex were independent adverse prognostic factors. In receiver operating characteristics curve analysis of 10-year RFS, the NIH consensus criteria, modified consensus criteria, and AFIP criteria resulted in an area under the curve (AUC) of 0·79 (95% CI 0·76-0·81), 0·78 (0·75-0·80), and 0·82 (0·80-0·85), respectively. The modified consensus criteria identified a single high-risk group. Since tumour size and mitosis count had a non-linear association with the risk of GIST recurrence, novel prognostic contour maps were generated using non-linear modelling of tumour size and mitosis count, and taking into account tumour site and rupture. The non-linear model accurately predicted the risk of recurrence (AUC 0·88, 0·86-0·90). INTERPRETATION The risk-stratification schemes assessed identify patients who are likely to be cured by surgery alone. Although the modified NIH classification is the best criteria to identify a single high-risk group for consideration of adjuvant therapy, the prognostic contour maps resulting from non-linear modelling are appropriate for estimation of individualised outcomes. FUNDING Academy of Finland, Cancer Society of Finland, Sigrid Juselius Foundation and Helsinki University Research Funds.

Rao-Blackwellized particle filter for multiple target tracking

In this article we propose a new Rao-Blackwellized particle filtering based algorithm for tracking an unknown number of targets. The algorithm is based on formulating probabilistic stochastic process models for target states, data associations, and birth and death processes. The tracking of these stochastic processes is implemented using sequential Monte Carlo sampling or particle filtering, and the efficiency of the Monte Carlo sampling is improved by using Rao-Blackwellization.

Bayesian approach for neural networks—review and case studies

We give a short review on the Bayesian approach for neural network learning and demonstrate the advantages of the approach in three real applications. We discuss the Bayesian approach with emphasis on the role of prior knowledge in Bayesian models and in classical error minimization approaches. The generalization capability of a statistical model, classical or Bayesian, is ultimately based on the prior assumptions. The Bayesian approach permits propagation of uncertainty in quantities which are unknown to other assumptions in the model, which may be more generally valid or easier to guess in the problem. The case problem studied in this paper include a regression, a classification, and an inverse problem. In the most thoroughly analyzed regression problem, the best models were those with less restrictive priors. This emphasizes the major advantage of the Bayesian approach, that we are not forced to guess attributes that are unknown, such as the number of degrees of freedom in the model, non-linearity of the model with respect to each input variable, or the exact form for the distribution of the model residuals.

Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC

Leave-one-out cross-validation (LOO) and the widely applicable information criterion (WAIC) are methods for estimating pointwise out-of-sample prediction accuracy from a fitted Bayesian model using the log-likelihood evaluated at the posterior simulations of the parameter values. LOO and WAIC have various advantages over simpler estimates of predictive error such as AIC and DIC but are less used in practice because they involve additional computational steps. Here we lay out fast and stable computations for LOO and WAIC that can be performed using existing simulation draws. We introduce an efficient computation of LOO using Pareto-smoothed importance sampling (PSIS), a new procedure for regularizing importance weights. Although WAIC is asymptotically equal to LOO, we demonstrate that PSIS-LOO is more robust in the finite case with weak priors or influential observations. As a byproduct of our calculations, we also obtain approximate standard errors for estimated predictive errors and for comparison of predictive errors between two models. We implement the computations in an R package called loo and demonstrate using models fit with the Bayesian inference package Stan.

Bayesian model assessment and comparison using cross-validation predictive densities

In this work, we discuss practical methods for the assessment, comparison, and selection of complex hierarchical Bayesian models. A natural way to assess the goodness of the model is to estimate its future predictive capability by estimating expected utilities. Instead of just making a point estimate, it is important to obtain the distribution of the expected utility estimate because it describes the uncertainty in the estimate. The distributions of the expected utility estimates can also be used to compare models, for example, by computing the probability of one model having a better expected utility than some other model. We propose an approach using cross-validation predictive densities to obtain expected utility estimates and Bayesian bootstrap to obtain samples from their distributions. We also discuss the probabilistic assumptions made and properties of two practical cross-validation methods, importance sampling and k-fold cross-validation. As illustrative examples, we use multilayer perceptron neural networks and gaussian processes with Markov chain Monte Carlo sampling in one toy problem and two challenging real-world problems.

A survey of Bayesian predictive methods for model assessment, selection and comparison

To date, several methods exist in the statistical literature for model assessment, which purport themselves specifically as Bayesian predic- tive methods. The decision theoretic assumptions on which these methods are based are not always clearly stated in the original articles, however. The aim of this survey is to provide a unified review of Bayesian predictive model assessment and selection methods, and of methods closely related to them. We review the various assumptions that are made in this context and discuss the connections between different approaches, with an emphasis on how each method approximates the expected utility of using a Bayesian model for the purpose of predicting future data.

A multi-metabolite analysis of serum by 1H NMR spectroscopy: early systemic signs of Alzheimer's disease.

A three-molecular-window approach for (1)H NMR spectroscopy of serum is presented to obtain specific molecular data on lipoproteins, various low-molecular-weight metabolites, and individual lipid molecules together with their degree of (poly)(un)saturation. The multiple data were analysed with self-organising maps, illustrating the strength of the approach as a holistic metabonomics framework in solely data-driven metabolic phenotyping. We studied 180 serum samples of which 30% were related to mild cognitive impairment (MCI), a neuropsychological diagnosis with severely increased risk for Alzheimers disease (AD). The results underline the association between MCI and the metabolic syndrome (MetS). Additionally, the low relativeamount of omega-3 fatty acids appears more indicative of MCI than low serum omega-3 or polyunsaturated fatty acid concentration as such. The analyses also feature the role of elevated glycoproteins in the risk for AD, supporting the view that coexistence of inflammation and the MetS forms a high risk condition for cognitive decline.

Bayesian analysis of the neuromagnetic inverse problem with ℓp-norm priors

Magnetoencephalography (MEG) allows millisecond-scale non-invasive measurement of magnetic fields generated by neural currents in the brain. However, localization of the underlying current sources is ambiguous due to the so-called inverse problem. The most widely used source localization methods (i.e., minimum-norm and minimum-current estimates (MNE and MCE) and equivalent current dipole (ECD) fitting) require ad hoc determination of the cortical current distribution (l(2)-, l(1)-norm priors and point-sized dipolar, respectively). In this article, we perform a Bayesian analysis of the MEG inverse problem with l(p)-norm priors for the current sources. This way, we circumvent the arbitrary choice between l(1)- and l(2)-norm prior, which is instead rendered automatically based on the data. By obtaining numerical samples from the joint posterior probability distribution of the source current parameters and model hyperparameters (such as the l(p)-norm order p) using Markov chain Monte Carlo (MCMC) methods, we calculated the spatial inverse estimates as expectation values of the source current parameters integrated over the hyperparameters. Real MEG data and simulated (known) source currents with realistic MRI-based cortical geometry and 306-channel MEG sensor array were used. While the proposed model is sensitive to source space discretization size and computationally rather heavy, it is mathematically straightforward, thus allowing incorporation of, for instance, a priori functional magnetic resonance imaging (fMRI) information.