Dependent generalized Dirichlet process priors for the analysis of acute lymphoblastic leukemia.

We propose a novel Bayesian nonparametric process prior for modeling a collection of random discrete distributions. This process is defined by including a suitable Beta regression framework within a generalized Dirichlet process to induce dependence among the discrete random distributions. This strategy allows for covariate dependent clustering of the observations. Some advantages of the proposed approach include wide applicability, ease of interpretation, and availability of efficient MCMC algorithms. The motivation for this work is the study of the impact of asparginage metabolism on lipid levels in a group of pediatric patients treated for acute lymphoblastic leukemia.

Recalcitrant chronic bladder pain and recurrent cystitis but negative urinalysis: What should we do?

PURPOSE: Lower urinary tract symptoms (LUTS) may be associated with chronic urinary tract infection (UTI) undetected by routine diagnostic tests. Antimicrobial therapy might confer benefit for these patients. MATERIALS AND METHODS: Over 10 years, we treated patients with chronic LUTS. Pyuria was adopted as the principal biomarker of infection. Urinary leucocyte counts were recorded from microscopy of fresh midstream urine (MSU) samples. Antibiotics were prescribed and the prescription adjusted to achieve a measurable clinical response and a reduction in pyuria. RESULTS: We treated 624 women [mean age = 53.4 years; standard deviation (SD) = 18] with chronic LUTS and pyuria. Mean duration of symptoms prior to presentation was 6.5 years. Only 16% of MSU cultures submitted were positive (≥105 cfu ml-1). Mean treatment length was 383 days [SD = 347; 95% confidence interval (CI) = 337-428]. Treatment was associated with a reduction in total LUTS (F = 98; p = 0.0001), 24-h frequency (F = 75; p = 0.0001), urinary urgency (F = 90; p = 0.0001), lower urinary tract pain (F = 108; p = 0.0001), voiding symptoms (F = 10; p = 0.002), and pyuria (F = 15.4; p = 0.0001). Full-dose first-generation antibiotics for UTI, such as cefalexin, nitrofurantoin, or trimethoprim, were combined with methenamine hippurate. We recorded 475 adverse events (AEs) during 273,762 treatment days. There was only one serious adverse event (SAE). We observed no increase in the proportion of resistant bacterial isolates. CONCLUSION: This large case series demonstrates that patients with chronic LUTS and pyuria experience symptom regression and a reduction in urinary tract inflammation associated with antimicrobial therapy. Disease regression was achieved with a low frequency of AEs. These results provide preliminary data to inform a future randomized controlled trial (RCT).

Lower urinary tract symptoms that predict microscopic pyuria.

INTRODUCTION AND HYPOTHESIS: Urinary dipsticks and culture analyses of a mid-stream urine specimen (MSU) at 105 cfu ml-1 of a known urinary pathogen are considered the gold standard investigations for diagnosing urinary tract infection (UTI). However, the reliability of these tests has been much criticised and they may mislead. It is now widely accepted that pyuria (≥1 WBC µl-1) detected by microscopy of a fresh unspun, unstained specimen of urine is the best biological indicator of UTI available. We aimed to scrutinise the greater potential of symptoms analysis in detecting pyuria and UTI. METHODS: Lower urinary tract symptom (LUTS) descriptions were collected from patients with chronic lower urinary tract symptoms referred to a tertiary referral unit. The symptoms informed a 39-question inventory, grouped into storage, voiding, stress incontinence and pain symptoms. All questions sought a binary yes or no response. A bespoke software package was developed to collect the data. The study was powered to a sample of at least 1,990 patients, with sufficient power to analyse 39 symptoms in a linear model with an effect size of Cohen's f2 = 0.02, type 1 error probability = 0.05; and power (1-ß); 95% where ß is the probability of type 2 error). The inventory was administered to 2,050 female patients between August 2004 and November 2011. The data were collated and the following properties assessed: internal consistency, test-retest reliability, inter-observer reliability, internal responsiveness, external responsiveness, construct validity analysis and a comparison with the International Consultation on Incontinence Modular Questionnaire for female lower urinary tract symptoms (ICIQ-FLUTS). The dependent variable used as a surrogate marker of UTI was microscopic pyuria. An MSU sample was sent for routine culture. RESULTS: The symptoms proved reliable predictors of microscopic pyuria. In particular, voiding symptoms correlated well with microscopic pyuria (χ2 = 88, df = 1, p < 0.001). The symptom inventory has significant psychometric characteristics as below: test-retest reliability: Cronbach's alpha was 0.981; inter-observer reliability, Cronbach's alpha was 0.995, internal responsiveness F = 221, p < 0.001, external responsiveness F = 359, df = 5, p < 0.001. The correlation coefficients for the domains of the ICIQ-FLUTS were around R = 0.5, p < 0.001. CONCLUSION: This symptoms score performed well on the standard, psychometric validation. The score changed in response to treatment and in a direction appropriate to the changes in microscopic pyuria. It correlated with measures of quality of life. It would seem to make a good candidate for monitoring treatment progress in ordinary clinical practice.

Prognostic Value of Late Gadolinium Enhancement Cardiovascular Magnetic Resonance in Cardiac Amyloidosis.

BACKGROUND: The prognosis and treatment of the 2 main types of cardiac amyloidosis, immunoglobulin light chain (AL) and transthyretin (ATTR) amyloidosis, are substantially influenced by cardiac involvement. Cardiovascular magnetic resonance with late gadolinium enhancement (LGE) is a reference standard for the diagnosis of cardiac amyloidosis, but its potential for stratifying risk is unknown. METHODS AND RESULTS: Two hundred fifty prospectively recruited subjects, 122 patients with ATTR amyloid, 9 asymptomatic mutation carriers, and 119 patients with AL amyloidosis, underwent LGE cardiovascular magnetic resonance. Subjects were followed up for a mean of 24±13 months. LGE was performed with phase-sensitive inversion recovery (PSIR) and without (magnitude only). These were compared with extracellular volume measured with T1 mapping. PSIR was superior to magnitude-only inversion recovery LGE because PSIR always nulled the tissue (blood or myocardium) with the longest T1 (least gadolinium). LGE was classified into 3 patterns: none, subendocardial, and transmural, which were associated with increasing amyloid burden as defined by extracellular volume (P<0.0001), with transitions from none to subendocardial LGE at an extracellular volume of 0.40 to 0.43 (AL) and 0.39 to 0.40 (ATTR) and to transmural at 0.48 to 0.55 (AL) and 0.47 to 0.59 (ATTR). Sixty-seven patients (27%) died. Transmural LGE predicted death (hazard ratio, 5.4; 95% confidence interval, 2.1-13.7; P<0.0001) and remained independent after adjustment for N-terminal pro-brain natriuretic peptide, ejection fraction, stroke volume index, E/E', and left ventricular mass index (hazard ratio, 4.1; 95% confidence interval, 1.3-13.1; P<0.05). CONCLUSIONS: There is a continuum of cardiac involvement in systemic AL and ATTR amyloidosis. Transmural LGE is determined reliably by PSIR and represents advanced cardiac amyloidosis. The PSIR technique provides incremental information on outcome even after adjustment for known prognostic factors.

Variable selection in covariate dependent random partition models: an application to urinary tract infection.

Lower urinary tract symptoms can indicate the presence of urinary tract infection (UTI), a condition that if it becomes chronic requires expensive and time consuming care as well as leading to reduced quality of life. Detecting the presence and gravity of an infection from the earliest symptoms is then highly valuable. Typically, white blood cell (WBC) count measured in a sample of urine is used to assess UTI. We consider clinical data from 1341 patients in their first visit in which UTI (i.e. WBC ≥ 1) is diagnosed. In addition, for each patient, a clinical profile of 34 symptoms was recorded. In this paper, we propose a Bayesian nonparametric regression model based on the Dirichlet process prior aimed at providing the clinicians with a meaningful clustering of the patients based on both the WBC (response variable) and possible patterns within the symptoms profiles (covariates). This is achieved by assuming a probability model for the symptoms as well as for the response variable. To identify the symptoms most associated to UTI, we specify a spike and slab base measure for the regression coefficients: this induces dependence of symptoms selection on cluster assignment. Posterior inference is performed through Markov Chain Monte Carlo methods.

Subpopulation Treatment Effect Pattern Plot (STEPP) analysis for continuous, binary, and count outcomes.

BACKGROUND: For the past few decades, randomized clinical trials have provided evidence for effective treatments by comparing several competing therapies. Their successes have led to numerous new therapies to combat many diseases. However, since their conclusions are based on the entire cohort in the trial, the treatment recommendation is for everyone, and may not be the best option for an individual. Medical research is now focusing more on providing personalized care for patients, which requires investigating how patient characteristics, including novel biomarkers, modify the effect of current treatment modalities. This is known as heterogeneity of treatment effects. A better understanding of the interaction between treatment and patient-specific prognostic factors will enable practitioners to expand the availability of tailored therapies, with the ultimate goal of improving patient outcomes. The Subpopulation Treatment Effect Pattern Plot (STEPP) approach was developed to allow researchers to investigate the heterogeneity of treatment effects on survival outcomes across values of a (continuously measured) covariate, such as a biomarker measurement. METHODS: Here, we extend the Subpopulation Treatment Effect Pattern Plot approach to continuous, binary, and count outcomes, which can be easily modeled using generalized linear models. With this extension of Subpopulation Treatment Effect Pattern Plot, these additional types of treatment effects within subpopulations defined with respect to a covariate of interest can be estimated, and the statistical significance of any observed heterogeneity of treatment effect can be assessed using permutation tests. The desirable feature that commonly used models are applied to well-defined patient subgroups to estimate treatment effects is retained in this extension. RESULTS: We describe a simulation study to confirm that the proper Type I error rate is maintained when there is no treatment heterogeneity, and a power study to show that the statistics have power to detect treatment heterogeneity under alternative scenarios. As an illustration, we apply the methods to data from the Aspirin/Folate Polyp Prevention Study, a clinical trial evaluating the effect of oral aspirin, folic acid, or both as a chemoprevention agent against colorectal adenomas. The pre-existing R software package stepp has been extended to handle continuous, binary, and count data using Gaussian, Bernoulli, and Poisson models, and it is available on the Comprehensive R Archive Network. CONCLUSION: The extension of the method and the availability of new software now permit STEPP to be applied to the full range of clinical trial end points.

Cardiac Structural and Functional Consequences of Amyloid Deposition by Cardiac Magnetic Resonance and Echocardiography and Their Prognostic Roles.

OBJECTIVES: This cross-sectional study aimed to describe the functional and structural cardiac abnormalities that occur across a spectrum of cardiac amyloidosis burden and to identify the strongest cardiac functional and structural prognostic predictors in amyloidosis using cardiac magnetic resonance (CMR) and echocardiography. BACKGROUND: Cardiac involvement in light chain and transthyretin amyloidosis is the main driver of prognosis and influences treatment strategies. Numerous measures of cardiac structure and function are assessed by multiple imaging modalities in amyloidosis. METHODS: A total f 322 subjects (311 systemic amyloidosis and 11 transthyretin gene mutation carriers) underwent comprehensive CMR and transthoracic echocardiography. The probabilities of 11 commonly measured structural and functional cardiac parameters being abnormal with increasing cardiac amyloidosis burden were evaluated. Cardiac amyloidosis burden was quantified using CMR-derived extracellular volume. The prognostic capacities of these parameters to predict death in amyloidosis were assessed using Cox proportional hazards models. RESULTS: Left ventricular mass and mitral annular plane systolic excursion by CMR along with strain and E/e' by echocardiography have high probabilities of being abnormal at low cardiac amyloid burden. Reductions in biventricular ejection fractions and elevations in biatrial areas occur at high burdens of infiltration. The probabilities of indexed stroke volume, myocardial contraction fraction, and tricuspid annular plane systolic excursion (TAPSE) being abnormal occur more gradually with increasing extracellular volume. Ninety patients (28%) died during a median follow-up of 22 months (interquartile range: 10 to 38 months). Univariable analysis showed that all imaging markers studied significantly predicted outcome. Multivariable analysis showed that TAPSE (hazard ratio: 1.46; 95% confidence interval: 1.16 to 1.85; p < 0.01) and indexed stroke volume (hazard ratio: 1.24; 95% confidence interval: 1.04 to 1.48; p < 0.05) by CMR were the only independent predictors of mortality. CONCLUSIONS: Specific functional and structural abnormalities characterize different burdens of cardiac amyloid deposition. In a multimodality imaging assessment of a large cohort of amyloidosis patients, CMR-derived TAPSE and indexed stroke volume are the strongest prognostic cardiac functional markers.