Sergio Venturini

Gamma shape mixtures for heavy-tailed distributions

An important question in health services research is the estimation of the proportion of medical expenditures that exceed a given threshold. Typically, medical expenditures present highly skewed, heavy tailed distributions, for which (a) simple variable transformations are insufficient to achieve a tractable low-dimensional parametric form and (b) nonparametric methods are not efficient in estimating exceedance probabilities for large thresholds. Motivated by this context, in this paper we propose a general Bayesian approach for the estimation of tail probabilities of heavy-tailed distributions, based on a mixture of gamma distributions in which the mixing occurs over the shape parameter. This family provides a flexible and novel approach for modeling heavy-tailed distributions, it is computationally efficient, and it only requires to specify a prior distribution for a single parameter. By carrying out simulation studies, we compare our approach with commonly used methods, such as the log-normal model and nonparametric alternatives. We found that the mixture-gamma model significantly improves predictive performance in estimating tail probabilities, compared to these alternatives. We also applied our method to the Medical Current Beneficiary Survey (MCBS), for which we estimate the probability of exceeding a given hospitalization cost for smoking attributable diseases. We have implemented the method in the open source GSM package, available from the Comprehensive R Archive Network.

Testing Hardy–Weinberg equilibrium: An objective Bayesian analysis

We analyze the general (multiallelic) Hardy-Weinberg equilibrium problem from an objective Bayesian testing standpoint. We argue that for small or moderate sample sizes the answer is rather sensitive to the prior chosen, and this suggests to carry out a sensitivity analysis with respect to the prior. This goal is achieved through the identification of a class of priors specifically designed for this testing problem. In this paper, we consider the class of intrinsic priors under the full model, indexed by a tuning quantity, the training sample size. These priors are objective, satisfy Savages continuity condition and have proved to behave extremely well for many statistical testing problems. We compute the posterior probability of the Hardy-Weinberg equilibrium model for the class of intrinsic priors, assess robustness over the range of plausible answers, as well as stability of the decision in favor of either hypothesis.

Combination of letrozole, metronomic cyclophosphamide and sorafenib is well-tolerated and shows activity in patients with primary breast cancer.

Purpose:To assess whether the combination of letrozole, metronomic cyclophosphamide and sorafenib (LCS) is well tolerated and shows activity in primary breast cancer (BC).Methods:Thirteen oestrogen receptor-positive, postmenopausal, T2-4, N0-1 BC patients received the LCS combination for 6 months. In these patients we examined the pharmacokinetics of sorafenib and cyclophosphamide, toxicity of the regimen, the clinical response to therapy and changes in the levels of biologically relevant biomarkers.Results:Adequate plasma concentrations of sorafenib were achieved in patients when it was dosed in combination with L+C. The mean plasma concentrations of C were consistently lower following administration of LCS, compared with administration of L+C only. The most common drug-related grade 3/4 adverse events were skin rash (69.3%), hand-foot skin reaction (69.3%) and diarrhoea (46.1%). According to RECIST Criteria, a clinical complete response was observed in 6 of 13 patients. A significant reduction in tumour size, evaluated with MRI, was also observed between baseline and 14 days of treatment in all 13 patients (P=0.005). A significant reduction in SUV uptake, measured by 18FDG-PET/CT, was observed in all patients between baseline and 30 days of treatment (P=0.015) and between baseline and definitive surgery (P=0.0002). Using modified CT Criteria, a response was demonstrated in 8 out of 10 evaluable patients at 30 days and in 11 out of 13 evaluable patients at the definitive surgery. A significant reduction in Ki67 expression was observed in all patients at day 14 compared with baseline (P<0.00001) and in 9 out of 13 patients at the definitive surgery compared with baseline (P<0.03). There was also a significant suppression of CD31 and VEGF-A expression in response to treatment (P=0.01 and P=0.007, respectively).Conclusions:The LCS combination is feasible and tolerable. The tumour response and target biomarker modulation indicate that the combination is clinically and biologically active.

Understanding Personal Mobile Technologies: Decomposing and De-Averaging the Value of a Smartphone

ABSTRACT: The study focuses on the multifaceted motives for adopting personal technologies. Specifically, it uses earlier models of technology adoption to develop a model of smartphone acceptance. The model is unique in that it decomposes attitudinal beliefs into three components: functional value, hedonic value, and symbolic value. Latent class analysis facilitates the identification of three user types. The analysis shows that value drivers, control beliefs, and normative beliefs play different roles for determining smartphone acceptance, depending on three different individual characteristics (i.e., playfulness, public self-consciousness, and innovativeness). The paper makes a contribution to the information systems literature by providing an analysis of the drivers of overall value perceptions for multipurpose information appliances and of the role of individual differences among potential users in forming these attitudes.

New Omics Information for Clinical Trial Utility in the Primary Setting

Cancer is a complex cellular disease caused by multiple factors via genetic mutations (hereditary or somatic) or environmental factors. The emerging omics technologies, including genomics, epigenomics, transcriptomics, proteomics, metabolomics, and interactomics, are increasingly being used for cancer research and personalized medicine; they have provided new opportunities in the molecular analysis of human cancer with unprecedented speed and detail. The omic approach has brought powerful ability to screen cancer cells at different levels from gene to metabolite and to search for novel drug targets, expounding the drug mechanism of action, identifying adverse effects in unexpected interaction, validating current drug targets, exploring potential applications for novel drugs, and enabling the translation from bench to bedside. As a clinical research tool, the neoadjuvant approach in breast cancer is the perfect setting for individualization of treatment based on clinical, pathological, image-guided, or molecular assessment, based on the omics techniques of tumors during treatment; neoadjuvant treatment offers the ability to discern treatment effect in vivo and may allow smaller trials targeting specific breast cancer subtypes.

Role of targeted agents in neuroendocrine tumors: Results from a meta-analysis

ABSTRACT Background: Several randomized phase III trials in neuroendocrine tumors (NETs) showed the clinical role of new targeted agents and their impact on tumor response and outcome of whose patients affected by advanced NET. In this study, we summarize the available clinical data related to clinical efficacy of targeted therapies in the treatment of advanced NETs. Methods: A meta-analysis of randomized studies in accordance with the PRISMA guidelines was performed after searching the databases of PubMed, the Cochrane Library, and the ASCO University Meeting for relevant publications. Results: One thousand 9 hundred and 8 cases were included in the meta-analysis; among these, 1012 were in the experimental arm and 896 were in the control arm. The pooled analysis of the use of target agents in NETs revealed significantly increased of progression free survival compared to control group (hazard ratio = 0.59, 95% CI:0.42-0.84; P = 0.003). Subgroup analysis of patients according to tumor site showed a difference in favor of pancreatic neuroendocrine tumors. Moreover, targeted therapies improved the overall survival (hazard ratio = 0.79, 95%CI: 0.63-0.98; P = 0.03), and response rate (hazard ratio = 3.33, 95% CI 2.02-5.49; P < 0.00001) in all types of NETs. Conclusion: Our analysis supports the routine use of targeted agents for treatment of neuroendocrine tumors with particular regards to the pancreatic neuroendocrine tumors.

Generalized Quantile Treatment Effect: A Flexible Bayesian Approach Using Quantile Ratio Smoothing

We propose a new general approach for estimating the effect of a binary treatment on a continuous and potentially highly skewed response variable, the generalized quantile treatment effect (GQTE). The GQTE is defined as the difference between a function of the quantiles under the two treatment conditions. As such, it represents a generalization over the standard approaches typically used for estimating a treatment effect (i.e., the average treatment effect and the quantile treatment effect) because it allows the comparison of any arbitrary characteristic of the outcomes distribution under the two treatments. Following Dominici et al. (2005), we assume that a pre-specified transformation of the two quantiles is modeled as a smooth function of the percentiles. This assumption allows us to link the two quantile functions and thus to borrow information from one distribution to the other. The main theoretical contribution we provide is the analytical derivation of a closed form expression for the likelihood of the model. Exploiting this result we propose a novel Bayesian inferential methodology for the GQTE. We show some finite sample properties of our approach through a simulation study which confirms that in some cases it performs better than other nonparametric methods. As an illustration we finally apply our methodology to the 1987 National Medicare Expenditure Survey data to estimate the difference in the single hospitalization medical cost distributions between cases (i.e., subjects affected by smoking attributable diseases) and controls.

Hypoxia-related biological markers as predictors of epirubicin-based treatment responsiveness and resistance in locally advanced breast cancer

Purpose To identify hypoxia-related biomarkers indicative of response and resistance to epirubicin treatment in patients with locally advanced breast cancer. Patients and Methods One hundred seventy-six women with T2-4 N0-1 breast tumours were randomly assigned to receive epirubicin 120 mg/m2/1-21 (EPI ARM), epirubicin 120 mg/m2/1-21 + erythropoietin 10.000 IU sc three times weekly (EPI-EPO ARM) and epirubicin 40 mg/m2/w-q21 (EPI-W ARM). Sixteen tumour proteins involved in cell survival, hypoxia, angiogenesis and growth factor, were assessed by immunohistochemistry in pre-treatment samples. A multivariate generalized linear regression approach was applied using a penalized least-square minimization to perform variable selection and regularization. Results VEGF and GLUT-1 expression were significantly positively associated with complete response (CR) to treatment in all leave-one-out iterations. Bcl-2 expression was inversely correlated with pCR, whilst EPO expression was positively correlated with pathological complete response (pCR). Haemaglobin and HIF-1a nuclear expression were inversely correlated with pCR. HB and HIF-1a expression were associated with a higher risk of relapse and overall survival. Conclusion Hypoxic biomarkers determines the epirubicin resistance in breast cancer. Assessment of such biomarkers, may be useful for predicting chemosensitivity and also anthracycline-based treatment outcome.