David Oakes

Bivariate Survival Models Induced by Frailties

Abstract When two observed survival times depend via a proportional-hazards model on the same unobserved variable, called a frailty, this common dependence induces an association between the observed times. This article considers the class of bivariate survival distributions that can arise in this way, extends it to allow negative association, and shows that the observable bivariate distribution determines the unobserved frailty distribution up to a scale parameter. A cross-ratio function, easily estimated even from censored data, is the key to both the characterization results and inferential procedures and diagnostic plots that are introduced and illustrated by real examples. The models considered are the natural counterpart for survival data of the latent variable models that have long been used in factor analysis for continuous data or in latent structure analysis for binary data. The main distinguishing feature of survival data is the possibility of censoring, which makes the standard methods difficu...

Detection of Huntington’s disease decades before diagnosis: the Predict-HD study

Objective: The objective of the Predict-HD study is to use genetic, neurobiological and refined clinical markers to understand the early progression of Huntington’s disease (HD), prior to the point of traditional diagnosis, in persons with a known gene mutation. Here we estimate the approximate onset and initial course of various measurable aspects of HD relative to the time of eventual diagnosis. Methods: We studied 438 participants who were positive for the HD gene mutation, but did not yet meet the diagnostic criteria for HD and had no functional decline. Predictability of baseline cognitive, motor, psychiatric and imaging measures was modelled non-linearly using estimated time until diagnosis (based on CAG repeat length and current age) as the predictor. Results: Estimated time to diagnosis was related to most clinical and neuroimaging markers. The patterns of association suggested the commencement of detectable changes one to two decades prior to the predicted time of clinical diagnosis. The patterns were highly robust and consistent, despite the varied types of markers and diverse measurement methodologies. Conclusions: These findings from the Predict-HD study suggest the approximate time scale of measurable disease development, and suggest candidate disease markers for use in preventive HD trials.

Treatment of ADHD in children with tics: A randomized controlled trial

BACKGROUND The treatment of children with attention deficit hyperactivity disorder (ADHD) and Tourette syndrome (TS) has been problematic because methylphenidate (MPH)--the most commonly used drug to treat ADHD--has been reported to worsen tics and because clonidine (CLON)--the most commonly prescribed alternative--has unproven efficacy. METHODS The authors conducted a multicenter, randomized, double-blind clinical trial in which 136 children with ADHD and a chronic tic disorder were randomly administered CLON alone, MPH alone, combined CLON + MPH, or placebo (2 x 2 factorial design). Each subject participated for 16 weeks (weeks 1-4 CLON/placebo dose titration, weeks 5-8 added MPH/placebo dose titration, weeks 9-16 maintenance therapy). RESULTS Thirty-seven children were administered MPH alone, 34 were administered CLON alone, 33 were administered CLON + MPH, and 32 were administered placebo. For our primary outcome measure of ADHD (Conners Abbreviated Symptom Questionnaire--Teacher), significant improvement occurred for subjects assigned to CLON (p < 0.002) and those assigned to MPH (p < 0.003). Compared with placebo, the greatest benefit occurred with combined CLON + MPH (p < 0.0001). CLON appeared to be most helpful for impulsivity and hyperactivity; MPH appeared to be most helpful for inattention. The proportion of individual subjects reporting a worsening of tics as an adverse effect was no higher in those treated with MPH (20%) than those being administered CLON alone (26%) or placebo (22%). Compared with placebo, measured tic severity lessened in all active treatment groups in the following order: CLON + MPH, CLON alone, MPH alone. Sedation was common with CLON treatment (28% reported moderate or severe sedation), but otherwise the drugs were tolerated well, including absence of any evident cardiac toxicity. CONCLUSIONS Methylphenidate and clonidine (particularly in combination) are effective for ADHD in children with comorbid tics. Prior recommendations to avoid methylphenidate in these children because of concerns of worsening tics are unsupported by this trial.

Thrombogenic Factors and Recurrent Coronary Events

BACKGROUND Thrombosis is a pivotal event in the pathogenesis of coronary disease. We hypothesized that the presence of blood factors that reflect enhanced thrombogenic activity would be associated with an increased risk of recurrent coronary events during long-term follow-up of patients who have recovered from myocardial infarction. METHODS AND RESULTS We prospectively enrolled 1045 patients 2 months after an index myocardial infarction. Baseline thrombogenic blood tests included 6 hemostatic variables (D-dimer, fibrinogen, factor VII, factor VIIa, von Willebrand factor, and plasminogen activator inhibitor-1), 7 lipid factors [cholesterol, triglycerides, HDL cholesterol, LDL cholesterol, lipoprotein(a), apolipoprotein (apo)A-I, and apoB], and insulin. Patients were followed up for an average of 26 months, with the primary end point being coronary death or nonfatal myocardial infarction, whichever occurred first. The hemostatic, lipid, and insulin parameters were dichotomized into their top and the lower 3 risk quartiles and evaluated for entry into a Cox survivorship model. High levels of D-dimer (hazard ratio, 2.43; 95% CI, 1.49, 3.97) and apoB (hazard ratio, 1.82; 95% CI, 1.10, 3.00) and low levels of apoA-I (hazard ratio, 1.84; 95% CI, 1.10, 3.08) were independently associated with recurrent coronary events in the Cox model after adjustment for 6 relevant clinical covariates. CONCLUSIONS Our findings indicate that a procoagulant state, as reflected in elevated levels of D-dimer, and disordered lipid transport, as indicated by low apoA-1 and high apoB levels, contribute independently to recurrent coronary events in postinfarction patients.

Urate as a Predictor of the Rate of Clinical Decline in Parkinson Disease

BACKGROUND The risk of Parkinson disease (PD) and its rate of progression may decline with increasing concentration of blood urate, a major antioxidant. OBJECTIVE To determine whether serum and cerebrospinal fluid concentrations of urate predict clinical progression in patients with PD. DESIGN, SETTING, AND PARTICIPANTS Eight hundred subjects with early PD enrolled in the Deprenyl and Tocopherol Antioxidative Therapy of Parkinsonism (DATATOP) trial. The pretreatment urate concentration was measured in serum for 774 subjects and in cerebrospinal fluid for 713 subjects. MAIN OUTCOME MEASURES Treatment-, age-, and sex-adjusted hazard ratios (HRs) for clinical disability requiring levodopa therapy, the prespecified primary end point of the original DATATOP trial. RESULTS The HR of progressing to the primary end point decreased with increasing serum urate concentrations (HR for highest vs lowest quintile = 0.64; 95% confidence interval [CI], 0.44-0.94; HR for a 1-SD increase = 0.82; 95% CI, 0.73-0.93). In analyses stratified by alpha-tocopherol treatment (2000 IU/d), a decrease in the HR for the primary end point was seen only among subjects not treated with alpha-tocopherol (HR for a 1-SD increase = 0.75; 95% CI, 0.62-0.89; vs HR for those treated = 0.90; 95% CI, 0.75-1.08). Results were similar for the rate of change in the Unified Parkinsons Disease Rating Scale score. Cerebrospinal fluid urate concentration was also inversely related to both the primary end point (HR for highest vs lowest quintile = 0.65; 95% CI, 0.44-0.96; HR for a 1-SD increase = 0.89; 95% CI, 0.79-1.02) and the rate of change in the Unified Parkinsons Disease Rating Scale score. As with serum urate concentration, these associations were present only among subjects not treated with alpha-tocopherol. CONCLUSIONS Higher serum and cerebrospinal fluid urate concentrations at baseline were associated with slower rates of clinical decline. The findings strengthen the link between urate concentration and PD and the rationale for considering central nervous system urate concentration elevation as a potential strategy to slow PD progression.

Serum Urate as a Predictor of Clinical and Radiographic Progression in Parkinson Disease

OBJECTIVE To determine whether concentration of serum urate, a purine metabolite and potent antioxidant that has been linked to a reduced risk of Parkinson disease (PD), predicts prognosis in PD. DESIGN Prospective study. SETTING The Parkinson Research Examination of CEP-1347 Trial (PRECEPT) study, which investigated the effects of a potential neuroprotectant on rates of PD progression, was conducted between April 2002 and August 2005 (average follow-up time 21.4 months). PARTICIPANTS Eight hundred four subjects with early PD enrolled in the PRECEPT study. MAIN OUTCOME MEASURES The primary study end point was progression to clinical disability sufficient to warrant dopaminergic therapy. Cox proportional hazards models were used to estimate the hazard ratio (HR) of reaching end point according to quintiles of baseline serum urate concentration, adjusting for sex, age, and other potential covariates. Change in striatal uptake of iodine I 123-labeled 2-beta-carbomethoxy-3-beta-(4-iodophenyl)tropane ([(123)I]beta-CIT), a marker for the presynaptic dopamine transporter, was assessed with linear regression for a subset of 399 subjects. RESULTS The adjusted HR of reaching end point declined with increasing baseline concentrations of urate; subjects in the top quintile reached the end point at only half the rate of subjects in the bottom quintile (HR, 0.51; 95% confidence interval [CI], 0.37-0.72; P for trend < .001). This association was markedly stronger in men (HR, 0.39; 95% CI, 0.26-0.60; P for trend < .001) than in women (HR, 0.77; 95% CI, 0.39-1.50; P for trend = .33). The percentage of loss in striatal [(123)I]beta-CIT uptake also improved with increasing serum urate concentrations (overall P for trend = .002; men, P = .001; women, P = .43). CONCLUSIONS These findings identify serum urate as the first molecular factor directly linked to the progression of typical PD and suggest that targeting urate or its determinants could be an effective disease-modifying therapy in PD. Trial Registration clinicaltrials.gov Identifier: NCT00040404.

Impact of sustained deprenyl (selegiline) in levodopa-treated Parkinson's disease: a randomized placebo-controlled extension of the deprenyl and tocopherol antioxidative therapy of parkinsonism trial.

Deprenyl (selegiline) delays the need for levodopa therapy in patients with early Parkinsons disease, but the long‐term benefits of this treatment remain unclear. During 1987 to 1988, 800 patients with early Parkinsons disease were randomized in the Deprenyl and Tocopherol Antioxidative Therapy of Parkinsonism trial to receive deprenyl, tocopherol, combined treatments, or a placebo and were then placed on active deprenyl (10mg/day). A second, independent randomization was carried out in early 1993 for 368 subjects who by that time had required levodopa and who had consented to continuing the deprenyl treatment (D subjects) or changing to a matching placebo (P subjects) under double‐blind conditions. The first development of wearing off, dyskinesias, or on–off motor fluctuations was the prespecified primary outcome measure. During the average 2‐year follow‐up, there were no differences between the treatment groups with respect to the primary outcome measure (hazard ratio, 0.87; 95% confidence interval, 0.63, 1.19; p = 0.38), withdrawal from the study, death, or adverse events. Although 34% of D subjects developed dyskinesias and only 19% of P subjects did (p = 0.006), only 16% of D subjects developed freezing of gait but 29% of P subjects did (p = 0.0003). Decline in motor performance was less in D subjects than P subjects. Levodopa‐treated Parkinsons disease patients who had been treated with deprenyl for up to 7 years, compared with patients who were changed to a placebo after about 5 years, experienced slower motor decline and were more likely to develop dyskinesias but less likely to develop freezing of gait.

Heterozygosity for a mutation in the parkin gene leads to later onset Parkinson disease

Background: The vast majority of the parkin mutations previously identified have been found in individuals with juvenile or early onset PD. Previous screening of later onset PD cohorts has not identified substantial numbers of parkin mutations. Methods: Families with at least two siblings with PD were ascertained to identify genes contributing to PD susceptibility. Screening of the parkin gene, by both quantitative PCR and exon sequencing, was performed in those families with either early onset PD (age onset ≤50 years) or positive lod score with a marker in intron 7 of the parkin gene. Results: A total of 25 different mutations in the parkin gene were identified in 103 individuals from 47 families. Mutations were found in both parkin alleles in 41 of the individuals, whereas a single mutation in only one of the two parkin alleles was observed in 62 individuals. Thirty-five of the subjects (34%) with a parkin mutation had an age at onset of 60 years or above with 30 of these 35 (86%) having a detectable mutation on only one parkin allele. Few significant clinical differences were observed among the individuals with two, one, or no mutated copies of the parkin gene. Conclusion: Mutations in the parkin gene occur among individuals with PD with an older age at onset (≥60 years) who have a positive family history of the disease. In addition, the clinical findings of parkin-positive individuals are remarkably similar to those without mutations.

Head injury and Parkinson's disease risk in twins.

Head injury is an inconsistently reported risk factor for Parkinsons disease (PD). Many related variables might confound this association, such as differences in childhood and adolescent lifestyles or genetically determined risk‐taking behaviors. Twin studies circumvent some of these problems, because twins are genetically and environmentally much more similar than typical cases and control subjects.

Significant Linkage of Parkinson Disease to Chromosome 2q36-37

Parkinson disease (PD) is the second most common neurodegenerative disorder, surpassed in frequency only by Alzheimer disease. Elsewhere we have reported linkage to chromosome 2q in a sample of sibling pairs with PD. We have now expanded our sample to include 150 families meeting our strictest diagnostic definition of verified PD. To further delineate the chromosome 2q linkage, we have performed analyses using only those pedigrees with the strongest family history of PD. Linkage analyses in this subset of 65 pedigrees generated a LOD score of 5.1, which was obtained using an autosomal dominant model of disease transmission. This result strongly suggests that variation in a gene on chromosome 2q36-37 contributes to PD susceptibility.