Paul D. Feigin

A double-blind, delayed-start trial of rasagiline in Parkinson's disease.

BACKGROUND A therapy that slows disease progression is the major unmet need in Parkinsons disease. METHODS In this double-blind trial, we examined the possibility that rasagiline has disease-modifying effects in Parkinsons disease. A total of 1176 subjects with untreated Parkinsons disease were randomly assigned to receive rasagiline (at a dose of either 1 mg or 2 mg per day) for 72 weeks (the early-start group) or placebo for 36 weeks followed by rasagiline (at a dose of either 1 mg or 2 mg per day) for 36 weeks (the delayed-start group). To determine a positive result with either dose, the early-start treatment group had to meet each of three hierarchical end points of the primary analysis based on the Unified Parkinsons Disease Rating Scale (UPDRS, a 176-point scale, with higher numbers indicating more severe disease): superiority to placebo in the rate of change in the UPDRS score between weeks 12 and 36, superiority to delayed-start treatment in the change in the score between baseline and week 72, and noninferiority to delayed-start treatment in the rate of change in the score between weeks 48 and 72. RESULTS Early-start treatment with rasagiline at a dose of 1 mg per day met all end points in the primary analysis: a smaller mean (+/-SE) increase (rate of worsening) in the UPDRS score between weeks 12 and 36 (0.09+/-0.02 points per week in the early-start group vs. 0.14+/-0.01 points per week in the placebo group, P=0.01), less worsening in the score between baseline and week 72 (2.82+/-0.53 points in the early-start group vs. 4.52+/-0.56 points in the delayed-start group, P=0.02), and noninferiority between the two groups with respect to the rate of change in the UPDRS score between weeks 48 and 72 (0.085+/-0.02 points per week in the early-start group vs. 0.085+/-0.02 points per week in the delayed-start group, P<0.001). All three end points were not met with rasagiline at a dose of 2 mg per day, since the change in the UPDRS score between baseline and week 72 was not significantly different in the two groups (3.47+/-0.50 points in the early-start group and 3.11+/-0.50 points in the delayed-start group, P=0.60). CONCLUSIONS Early treatment with rasagiline at a dose of 1 mg per day provided benefits that were consistent with a possible disease-modifying effect, but early treatment with rasagiline at a dose of 2 mg per day did not. Because the two doses were associated with different outcomes, the study results must be interpreted with caution. (ClinicalTrials.gov number, NCT00256204.)

Maximum likelihood estimation for continuous-time stochastic processes

This paper is mainly concerned with the asymptotic theory of maximum likelihood estimation for continuous-time stochastic processes. The role of martingale limit theory in this theory is developed. Some analogues of classical statistical concepts and quantities are also suggested. Various examples that illustrate parts of the theory are worked through, producing new results in some cases. The role of diffusion approximations in estimation is also explored. MAXIMUM LIKELIHOOD ESTIMATION; CONTINUOUS-TIME STOCHASTIC PROCESSES; ASYMPTOTIC THEORY; MARTINGALE LIMIT THEORY; DIFFUSION APPROXIMATIONS

A novel host-proteome signature for distinguishing between acute bacterial and viral infections.

Bacterial and viral infections are often clinically indistinguishable, leading to inappropriate patient management and antibiotic misuse. Bacterial-induced host proteins such as procalcitonin, C-reactive protein (CRP), and Interleukin-6, are routinely used to support diagnosis of infection. However, their performance is negatively affected by inter-patient variability, including time from symptom onset, clinical syndrome, and pathogens. Our aim was to identify novel viral-induced host proteins that can complement bacterial-induced proteins to increase diagnostic accuracy. Initially, we conducted a bioinformatic screen to identify putative circulating host immune response proteins. The resulting 600 candidates were then quantitatively screened for diagnostic potential using blood samples from 1002 prospectively recruited patients with suspected acute infectious disease and controls with no apparent infection. For each patient, three independent physicians assigned a diagnosis based on comprehensive clinical and laboratory investigation including PCR for 21 pathogens yielding 319 bacterial, 334 viral, 112 control and 98 indeterminate diagnoses; 139 patients were excluded based on predetermined criteria. The best performing host-protein was TNF-related apoptosis-inducing ligand (TRAIL) (area under the curve [AUC] of 0.89; 95% confidence interval [CI], 0.86 to 0.91), which was consistently up-regulated in viral infected patients. We further developed a multi-protein signature using logistic-regression on half of the patients and validated it on the remaining half. The signature with the highest precision included both viral- and bacterial-induced proteins: TRAIL, Interferon gamma-induced protein-10, and CRP (AUC of 0.94; 95% CI, 0.92 to 0.96). The signature was superior to any of the individual proteins (P<0.001), as well as routinely used clinical parameters and their combinations (P<0.001). It remained robust across different physiological systems, times from symptom onset, and pathogens (AUCs 0.87-1.0). The accurate differential diagnosis provided by this novel combination of viral- and bacterial-induced proteins has the potential to improve management of patients with acute infections and reduce antibiotic misuse.

Stable convergence of semimartingales

Under a nesting condition on the sequence of histories, stable weak convergence of semimartingales to processes with conditionally independent increments is considered. Apart from ensuring the stability property, the nesting condition is more natural in some applications than an alternative measurability condition which appears in the literature.

Limit distributions for linear programming time series estimators

We consider stationary autoregressive processes of order p which have positive innovations. We propose consistent parameter estimators based on linear programming. Under conditions, including regular variation of either the left or right tail of the innovations distribution, we prove that the estimators have a limit distribution. The rate of convergence of our estimator is favorable compared with the Yule--Walker estimator under comparable circumstances.

Estimation for Autoregressive Processes with Positive Innovations

We consider stationary autoregressive processes of order p which have positive parameters and positive innovations. The main results concern the rate of consistency of parameter estimators for the case p = 2. These estimators are defined in terms of estimating equations. Relevant asymptotic theory is developed in the wider context of vector autoregressive processes with positive innovations having a distribution with regularly varying left or right tails. These weak convergence results may be of independent interest.

Linear functionals and Markov chains associated with Dirichlet processes

By investigating a Markov chain whose limiting distribution corresponds to that of the Dirichlet process we are able directly to ascertain conditions for the existence of linear functionals of that process. Together with earlier analyses we are able to characterize those functionals which are a.s. finite in terms of the parameter measure of the process. We also show that the appropriate Markov chain in the space of measures is only weakly convergent and not Harris ergodic.

Designing a call center with an IVR (Interactive Voice Response)

A call center is a service operation that caters to customer needs via the telephone. Call centers typically consist of agents that serve customers, telephone lines, an Interactive Voice Response (IVR) unit, and a switch that routes calls to agents. In this paper we study a Markovian model for a call center with an IVR. We calculate operational performance measures, such as the probability for a busy signal and the average wait time for an agent. Exact calculations of these measures are cumbersome and they lack insight. We thus approximate the measures in an asymptotic regime known as QED (Quality and Efficiency Driven) or the Halfin–Whitt regime, which accommodates moderate to large call centers. The approximations are both insightful and easy to apply (for up to 1000’s of agents). They yield, as special cases, known and novel approximations for the M/M/N/N (Erlang-B), M/M/S (Erlang-C) and M/M/S/N queue.

Evaluating the accuracy of different sperm counting chambers by performing strict counts of photographed beads

PurposeTo suggest a method for evaluating the accuracy of different kinds of sperm counting chambers by eliminating errors concerned with human skills or semen properties. In this method, various concentrations were prepared from stocks of commercially available latex beads (Accu-beads, Hamilton-Thorn Research, Beverly, MA). Samples from identical preparations were loaded into different types of chambers, namely, hemocytometer (Neubauer, improved double, Superior Ltd, Germany), Makler (Sefi Medical Instruments, Haifa, Israel) and Horwell (Horwell Ltd, London, UK). Beads were counted by both direct microscopic observation and by strict scanning of their photographed images.ResultsIn all cases, counts by direct observation were about 5% higher than strict counts of the same photographed beads. Counting photographed beads showed high reproducibility (average CV of 5.1%) between samples in the two wells of the hemocytometer. Counts of photographed beads, sampled from identical stocks, were on average slightly lower in the Makler chamber (20.7 × 106/ml) and much higher in the Horwell chamber (47.4 106/ml) than counts in the hemocytometer (21.5 × 106/ml). Samples from three different batches of Accubeads revealed slight variation in counts between the batches and an average concentration of 11% above the number indicated on the commercial product.ConclusionsA technique that combined loading latex beads from identical stock into various chambers, proper covering of the tested samples and strict counting of photographed beads provided precise and reproducible results. By eliminating most errors related to human skills and semen properties, this method is suitable for evaluating the accuracy of counting chambers.

The association of DNA sequence variation at the MAOA genetic locus with quantitative behavioural traits in normal males

Monoamine oxidase A (MAOA) catalyses the oxidative deamination of biogenic amines including neurotransmitters, mainly norepinephrine and serotonin in the brain and peripheral tissues. A nonsense mutation in the gene was shown to be involved in a rare X-linked behavioural syndrome, which includes impaired impulse control, aggression and borderline mental retardation (Brunner syndrome). Several recent studies have shown the association of genetic variation of a VNTR in the gene promoter with various pathological behavioural traits. In the present study the association of MAOA genetic variation with a large set of quantitative behavioural traits in normal individuals has been examined. DNA samples from 421 unrelated males were genotyped for 14 SNPs and for the promoter VNTR at the MAOA locus. An additional 16 SNPs were genotyped at apparently neutral loci across the X chromosome to serve as a genomic control for possible false positive associations due to population structure. Behavioural traits were measured using the NEO psychometric questionnaire, which is based on a 5-axis model of personality, and consists of 30 different quantitative traits. There was a robust association of the A2 (“straightforwardness”) facet with common allelic variants at the promoter VNTR. Most of the tested traits were not associated with the VNTR despite reasonable power, thus demonstrating that the VNTR influence on quantitative behavioural traits in normal males may be very specific. In contrast, several traits of the C (“conscientiousness”) axis were associated with less common SNP-defined haplotypes. Hence, it appears that common genetic variation at the VNTR contributes to the behavioural attribute of “straightforwardness”, while rare haplotypes defined by SNPs downstream of the transcription start site may contribute to “conscientiousness”. This study is used to address the validation, interpretation and limitation of genetic association studies of quantitative behavioural traits.