Quantitative assessment of adverse events in clinical trials: Comparison of methods at an interim and the final analysis

In clinical study reports (CSRs), adverse events (AEs) are commonly summarized using the incidence proportion (IP). IPs can be calculated for all types of AEs and are often interpreted as the probability that a treated patient experiences specific AEs. Exposure time can be taken into account with time-to-event methods. Using one minus Kaplan–Meier (1-KM) is known to overestimate the AE probability in the presence of competing events (CEs). The use of a nonparametric estimator of the cumulative incidence function (CIF) has therefore been advocated as more appropriate. In this paper, we compare different methods to estimate the probability of one selected AE. In particular, we investigate whether the proposed methods provide a reasonable estimate of the AE probability at an interim analysis (IA). The characteristics of the methods in the presence of a CE are illustrated using data from a breast cancer study and we quantify the potential bias in a simulation study. At the final analysis performed for the CSR, 1-KM systematically overestimates and in most cases IP slightly underestimates the given AE probability. CIF has the lowest bias in most simulation scenarios. All methods might lead to biased estimates at the IA except for AEs with early onset. The magnitude of the bias varies with the time-to-AE and/or CE occurrence, the selection of event-specific hazards and the amount of censoring. In general, reporting AE probabilities for prespecified fixed time points is recommended.     

Joint Models of Longitudinal Data and Recurrent Events with Informative Terminal Event

This article presents semiparametric joint models to analyze longitudinal data with recurrent events (e.g. multiple tumors, repeated hospital admissions) and a terminal event such as death. A broad class of transformation models for the cumulative intensity of the recurrent events and the cumulative hazard of the terminal event is considered, which includes the proportional hazards model and the proportional odds model as special cases. We propose to estimate all the parameters using the nonparametric maximum likelihood estimators (NPMLE). We provide the simple and efficient EM algorithms to implement the proposed inference procedure. Asymptotic properties of the estimators are shown to be asymptotically normal and semiparametrically efficient. Finally, we evaluate the performance of the method through extensive simulation studies and a real-data application.     

Ultrasonic acoustic emissions from the sapwood of cedar and hemlock : an examination of three hypotheses regarding cavitations

Measurements are reported of ultrasonic acoustic emissions (AEs) measured from sapwood samples of Thuja occidentalis L. and Tsuga canadensis (L.) Carr. during air dehydration. The measurements were undertaken to test the following three hypotheses: (a) Each cavitation event produces one ultrasonic AE. (b) Large tracheids are more likely to cavitate than small tracheids. (c) When stem water potentials are >−0.4 MPa, a significant fraction of the water content of sapwood is held by `capillary forces.' The last two hypotheses were recently discussed at length by M. H. Zimmermann. Experimental evidence consistent with all three hypotheses was obtained. The evidence for each hypothesis respectively is: (a) the cumulative number of AEs nearly equals the number of tracheids in small samples; (b) more water is lost per AE event at the beginning of the dehydration process than at the end, and (c) sapwood samples dehydrated from an initial water potential of 0 MPa lost significantly more water before AEs started than lost by samples dehydrated from an initial water potential of about −0.4 MPa. The extra water held by fully hydrated sapwood samples may have been capillary water as defined by Zimmerman.

We also report an improved method for the measurement of the `intensity' of ultrasonic AEs. Intensity is defined here as the area under the positive spikes of the AE signal (plotted as voltage versus time). This method was applied to produce a frequency histogram of the number of AEs versus intensity. A large fraction of the total number of AEs were of low intensity even in small samples (4 mm diameter by 10 mm length). This suggests that the effective `listening distance' for most AEs was less than 5 to 10 mm.

     

Ontology-Based Combinatorial Comparative Analysis of Adverse Events Associated with Killed and Live Influenza Vaccines

Vaccine adverse events (VAEs) are adverse bodily changes occurring after vaccination. Understanding the adverse event (AE) profiles is a crucial step to identify serious AEs. Two different types of seasonal influenza vaccines have been used on the market: trivalent (killed) inactivated influenza vaccine (TIV) and trivalent live attenuated influenza vaccine (LAIV). Different adverse event profiles induced by these two groups of seasonal influenza vaccines were studied based on the data drawn from the CDC Vaccine Adverse Event Report System (VAERS). Extracted from VAERS were 37,621 AE reports for four TIVs (Afluria, Fluarix, Fluvirin, and Fluzone) and 3,707 AE reports for the only LAIV (FluMist). The AE report data were analyzed by a novel combinatorial, ontology-based detection of AE method (CODAE). CODAE detects AEs using Proportional Reporting Ratio (PRR), Chi-square significance test, and base level filtration, and groups identified AEs by ontology-based hierarchical classification. In total, 48 TIV-enriched and 68 LAIV-enriched AEs were identified (PRR>2, Chi-square score >4, and the number of cases >0.2% of total reports). These AE terms were classified using the Ontology of Adverse Events (OAE), MedDRA, and SNOMED-CT. The OAE method provided better classification results than the two other methods. Thirteen out of 48 TIV-enriched AEs were related to neurological and muscular processing such as paralysis, movement disorders, and muscular weakness. In contrast, 15 out of 68 LAIV-enriched AEs were associated with inflammatory response and respiratory system disorders. There were evidences of two severe adverse events (Guillain-Barre Syndrome and paralysis) present in TIV. Although these severe adverse events were at low incidence rate, they were found to be more significantly enriched in TIV-vaccinated patients than LAIV-vaccinated patients. Therefore, our novel combinatorial bioinformatics analysis discovered that LAIV had lower chance of inducing these two severe adverse events than TIV. In addition, our meta-analysis found that all previously reported positive correlation between GBS and influenza vaccine immunization were based on trivalent influenza vaccines instead of monovalent influenza vaccines.     

Discrete-time nonparametric estimation for semi-Markov models of chain-of-events data subject to interval censoring and truncation

Chain-of-events data are longitudinal observations on a succession of events that can only occur in a prescribed order. One goal in an analysis of this type of data is to determine the distribution of times between the successive events. This is difficult when individuals are observed periodically rather than continuously because the event times are then interval censored. Chain-of-events data may also be subject to truncation when individuals can only be observed if a certain event in the chain (e.g., the final event) has occurred. We provide a nonparametric approach to estimate the distributions of times between successive events in discrete time for data such as these under the semi-Markov assumption that the times between events are independent. This method uses a self-consistency algorithm that extends Turnbull's algorithm (1976, Journal of the Royal Statistical Society, Series B 38, 290-295). The quantities required to carry out the algorithm can be calculated recursively for improved computational efficiency. Two examples using data from studies involving HIV disease are used to illustrate our methods.     

Ultrasonic Acoustic Emissions from the Sapwood of Thuja occidentalis Measured inside a Pressure Bomb

An improved method of counting acoustic emission (AE) events from water-stressed stems of cedar (Thuja occidentalis L.) is presented. Amplified AEs are analyzed on a real time basis by a microcomputer. The instrumentation counts AE events in a fashion nearly analogous to scintillation counting of radioactive materials.

The technique was applied to measuring ultrasonic AEs from the stems of cedar inside a pressure bomb. The shoots were originally fully hydrated. When the shoots are dehydrated in the bomb by application of an overpressure very few AEs were detected. When the bomb pressure is reduced after dehydration of the shoot, AE events could be detected. We conclude that ultrasonic AEs are caused by cavitation events (= structural breakdown of water columns in the tracheids of cedar) and not by the breaking of cellulose fibers in the wood.

     

Nonparametric analysis of recurrent events and death

This article is concerned with the analysis of recurrent events in the presence of a terminal event such as death. We consider the mean frequency function, defined as the marginal mean of the cumulative number of recurrent events over time. A simple nonparametric estimator for this quantity is presented. It is shown that the estimator, properly normalized, converges weakly to a zero-mean Gaussian process with an easily estimable covariance function. Nonparametric statistics for comparing two mean frequency functions and for combining data on recurrent events and death are also developed. The asymptotic null distributions of these statistics, together with consistent variance estimators, are derived. The small-sample properties of the proposed estimators and test statistics are examined through simulation studies. An application to a cancer clinical trial is provided.     

Semiparametric models for cumulative incidence functions

In analyses of time-to-failure data with competing risks, cumulative incidence functions may be used to estimate the time-dependent cumulative probability of failure due to specific causes. These functions are commonly estimated using nonparametric methods, but in cases where events due to the cause of primary interest are infrequent relative to other modes of failure, nonparametric methods may result in rather imprecise estimates for the corresponding subdistribution. In such cases, it may be possible to model the cause-specific hazard of primary interest parametrically, while accounting for the other modes of failure using nonparametric estimators. The cumulative incidence estimators so obtained are simple to compute and are considerably more efficient than the usual nonparametric estimator, particularly with regard to interpolation of cumulative incidence at early or intermediate time points within the range of data used to fit the function. More surprisingly, they are often nearly as efficient as fully parametric estimators. We illustrate the utility of this approach in the analysis of patients treated for early stage breast cancer.     

Long-term safety and tolerability of donepezil 23 mg in patients with moderate to severe Alzheimer's disease

ABSTRACT: BACKGROUND: Donepezil (23 mg/day) is approved by the US Food and Drug Administration for the treatment of patients with moderate to severe Alzheimer's disease (AD). Approval was based on results from a 24-week, randomized, double-blind study of patients who were stable on donepezil 10 mg/day and randomized 2:1 to either increase their donepezil dose to 23 mg/day or continue taking 10 mg/day. The objective of this study was to assess the long-term safety and tolerability of donepezil 23 mg/day in patients with moderate to severe AD. METHODS: Patients who completed the double-blind study and were eligible could enroll into a 12-month extension study of open-label donepezil 23 mg/day. Clinic visits took place at open-label baseline and at months 3, 6, 9, and 12. Safety analyses comprised examination of the incidence, severity, and timing of treatment-emergent adverse events (AEs); changes in weight, electrocardiogram, vital signs, and laboratory parameters; and discontinuation due to AEs. RESULTS: 915 double-blind study completers were enrolled in the open-label extension study and 902 comprised the safety population. Mean treatment duration in this study was 10.3 +/- 3.5 months. In total, 674 patients (74.7%) reported at least one AE; in 320 of these patients (47.5%) at least one AE was considered to be possibly or probably study drug related. The majority of patients reporting AEs (81.9%) had AEs of mild or moderate severity. There were 268 patients (29.7%) who discontinued early, of which 123 (13.6%) were due to AEs. Patients increasing donepezil dose from 10 mg/day in the double-blind study to 23 mg/day in the extension study had slightly higher rates of AEs and SAEs than patients who were already receiving 23 mg (78.0% and 16.9% vs 72.8% and 14.0%, respectively). The incidence of new AEs declined rapidly after the first 2 weeks and remained low throughout the duration of the study. CONCLUSION: This study shows that long-term treatment with donepezil 23 mg/day is associated with no new safety signals. The elevated incidence of AEs in patients increasing the dose of donepezil from 10 mg/day to 23 mg/day was limited to the initial weeks of the study.     

A multistate model for bivariate interval-censored failure time data

SUMMARY: Interval-censored life-history data arise when the events of interest are only detectable at periodic assessments. When interest lies in the occurrence of two such events, bivariate-interval censored event time data are obtained. We describe how to fit a four-state Markov model useful for characterizing the association between two interval-censored event times when the assessment times for the two events may be generated by different inspection processes. The approach treats the two events symmetrically and enables one to fit multiplicative intensity models that give estimates of covariate effects as well as relative risks characterizing the association between the two events. An expectation-maximization (EM) algorithm is described for estimation in which the maximization step can be carried out with standard software. The method is illustrated by application to data from a trial of HIV patients where the events are the onset of viral shedding in the blood and urine among individuals infected with cytomegalovirus.     

Accounting for multiplicities in assessing drug safety: a three-level hierarchical mixture model

Multiple comparisons and other multiplicities are among the most difficult of problems that face statisticians, frequentists, and Bayesians alike. An example is the analysis of the many types of adverse events (AEs) that are recorded in drug clinical trials. We propose a three-level hierarchical mixed model. The most basic level is type of AE. The second level is body system, each of which contains a number of types of possibly related AEs. The highest level is the collection of all body systems. Our analysis allows for borrowing across body systems, but there is greater potential-depending on the actual data-for borrowing within each body system. The probability that a drug has caused a type of AE is greater if its rate is elevated for several types of AEs within the same body system than if the AEs with elevated rates were in different body systems. We give examples to illustrate our method and we describe its application to other types of problems.     

Clinical tolerability of artesunate-amodiaquine versus comparator treatments for uncomplicated falciparum malaria: an individual-patient analysis of eight randomized controlled trials in sub-Saharan Africa

ABSTRACT: BACKGROUND: The widespread use of artesunate-amodiaquine (ASAQ) for treating uncomplicated malaria makes it important to gather and analyse information on its tolerability. METHODS: An individual-patient tolerability analysis was conducted using data from eight randomized controlled clinical trials conducted at 17 sites in nine sub-Saharan countries comparing ASAQ to other anti-malarial treatments. All patients who received at least one dose of the study drug were included in the analysis. Differences in adverse event (AE) and treatment emergent adverse event (TEAE) were analysed by Day 28. RESULTS: Of the 6,179 patients enrolled (74 % <5 years of age), 50 % (n = 3,113) received ASAQ, 20 % (n = 1,217) another ACT, and 30 % (n = 1,849) a non-ACT (combination or single-agent) treatment. Overall, 8,542 AEs were recorded. The proportion of patients experiencing at least one gastro-intestinal AE on ASAQ was 43 % (and higher than that with artemetherlumefantrine and dihydroartemisinin-piperaquine at two sites), and was 23 % for any other AEs (not different from other treatments). Specifically, the risk of diarrhoea, vomiting, cough and weakness was lower with artemether-lumefantrine; artemether-lumefantrine and dihydroartemisinin-piperaquine carried a higher risk of pruritus, chloroquine-SP had a higher risk of nausea. Parasitological recurrence increased the risk of occurrence of any AE. No other difference was detected. Comparing AE to TEAE in patients who had pre-treatment occurrence and grades of intensity recorded, AEs were significantly more related to the pretreatment prevalence of the symptom (p = 0.001, Fisher test); AEs overestimated TEAEs by a factor ranging from none to five-fold. The overall incidence of serious AEs (SAEs) with ASAQ was nine per 1,000 (29/3,113) and mortality was one per 1,000 (three deaths, none drug-related); both were similar to other treatments. CONCLUSION: ASAQ was comparatively well-tolerated. Safety information is important, and must be collected and analysed in a standardized way. TEAEs are a more objective measure of treatment-induced toxicity.     

Nonparametric estimation in a Markov "illness-death" process from interval censored observations with missing intermediate transition status

Summary .  In many clinical trials patients are intermittently assessed for the transition to an intermediate state, such as occurrence of a disease-related nonfatal event, and death. Estimation of the distribution of nonfatal event free survival time, that is, the time to the first occurrence of the nonfatal event or death, is the primary focus of the data analysis. The difficulty with this estimation is that the intermittent assessment of patients results in two forms of incompleteness: the times of occurrence of nonfatal events are interval censored and, when a nonfatal event does not occur by the time of the last assessment, a patient's nonfatal event status is not known from the time of the last assessment until the end of follow-up for death. We consider both forms of incompleteness within the framework of an "illness–death" model. We develop nonparametric maximum likelihood (ML) estimation in an "illness–death" model from interval-censored observations with missing status of intermediate transition. We show that the ML estimators are self-consistent and propose an algorithm for obtaining them. This work thus provides new methodology for the analysis of incomplete data that arise from clinical trials. We apply this methodology to the data from a recently reported cancer clinical trial ( Bonner et al., 2006 , New England Journal of Medicine 354, 567–578) and compare our estimation results with those obtained using a Food and Drug Administration recommended convention.     

Bayesian nonparametric analysis of restricted mean survival time

The restricted mean survival time (RMST) evaluates the expectation of survival time truncated by a prespecified time point, because the mean survival time in the presence of censoring is typically not estimable. The frequentist inference procedure for RMST has been widely advocated for comparison of two survival curves, while research from the Bayesian perspective is rather limited. For the RMST of both right- and interval-censored data, we propose Bayesian nonparametric estimation and inference procedures. By assigning a mixture of Dirichlet processes (MDP) prior to the distribution function, we can estimate the posterior distribution of RMST. We also explore another Bayesian nonparametric approach using the Dirichlet process mixture model and make comparisons with the frequentist nonparametric method. Simulation studies demonstrate that the Bayesian nonparametric RMST under diffuse MDP priors leads to robust estimation and under informative priors it can incorporate prior knowledge into the nonparametric estimator. Analysis of real trial examples demonstrates the flexibility and interpretability of the Bayesian nonparametric RMST for both right- and interval-censored data.     

Semiparametric transformation models with random effects for joint analysis of recurrent and terminal events

Summary .  We propose a broad class of semiparametric transformation models with random effects for the joint analysis of recurrent events and a terminal event. The transformation models include proportional hazards/intensity and proportional odds models. We estimate the model parameters by the nonparametric maximum likelihood approach. The estimators are shown to be consistent, asymptotically normal, and asymptotically efficient. Simple and stable numerical algorithms are provided to calculate the parameter estimators and to estimate their variances. Extensive simulation studies demonstrate that the proposed inference procedures perform well in realistic settings. Applications to two HIV/AIDS studies are presented.     

Flexible Estimation of Differences in Treatment-Specific Recurrent Event Means in the Presence of a Terminating Event

Summary .  In this article, we consider the setting where the event of interest can occur repeatedly for the same subject (i.e., a recurrent event; e.g., hospitalization) and may be stopped permanently by a terminating event (e.g., death). Among the different ways to model recurrent/terminal event data, the marginal mean (i.e., averaging over the survival distribution) is of primary interest from a public health or health economics perspective. Often, the difference between treatment-specific recurrent event means will not be constant over time, particularly when treatment-specific differences in survival exist. In such cases, it makes more sense to quantify treatment effect based on the cumulative difference in the recurrent event means, as opposed to the instantaneous difference in the rates. We propose a method that compares treatments by separately estimating the survival probabilities and recurrent event rates given survival, then integrating to get the mean number of events. The proposed method combines an additive model for the conditional recurrent event rate and a proportional hazards model for the terminating event hazard. The treatment effects on survival and on recurrent event rate among survivors are estimated in constructing our measure and explain the mechanism generating the difference under study. The example that motivates this research is the repeated occurrence of hospitalization among kidney transplant recipients, where the effect of expanded criteria donor (ECD) compared to non-ECD kidney transplantation on the mean number of hospitalizations is of interest.     

Neural encoding schemes of tactile information in afferent activity of the vibrissal system

When rats acquire sensory information by actively moving their vibrissae, a neural code is manifested at different levels of the sensory system. Behavioral studies in tactile discrimination agree that rats can distinguish different roughness surfaces by whisking their vibrissae. The present study explores the existence of neural encoding in the afferent activity of one vibrissal nerve. Two neural encoding schemes based on “events” were proposed (cumulative event count and median inter-event time). The events were detected by using an event detection algorithm based on multiscale decomposition of the signal (Continuous Wavelet Transform). The encoding schemes were quantitatively evaluated through the maximum amount of information which was obtained by the Shannon’s mutual information formula. Moreover, the effect of difference distances between rat snout and swept surfaces on the information values was also studied. We found that roughness information was encoded by events of 0.8 ms duration in the cumulative event count and event of 1.0 to 1.6 ms duration in the median inter-event count. It was also observed that an extreme decrease of the distance between rat snout and swept surfaces significantly reduces the information values and the capacity to discriminate among the sweep situations.     

Nonparametric inference of general while-alive estimands for recurrent events

Measuring the treatment effect on recurrent events like hospitalization in the presence of death has long challenged statisticians and clinicians alike. Traditional inference on the cumulative frequency unjustly penalizes survivorship as longer survivors also tend to experience more adverse events. Expanding a recently suggested idea of the “while-alive” event rate, we consider a general class of such estimands that adjust for the length of survival without losing causal interpretation. Given a user-specified loss function that allows for arbitrary weighting, we define as estimand the average loss experienced per unit time alive within a target period and use the ratio of this loss rate to measure the effect size. Scaling the loss rate by the width of the corresponding time window gives us an alternative, and sometimes more photogenic, way of showing the data. To make inferences, we construct a nonparametric estimator for the loss rate through the cumulative loss and the restricted mean survival time and derive its influence function in closed form for variance estimation and testing. As simulations and analysis of real data from a heart failure trial both show, the while-alive approach corrects for the false attenuation of treatment effect due to patients living longer under treatment, with increased statistical power as a result. The proposed methods are implemented in the R-package WA , which is publicly available from the Comprehensive R Archive Network (CRAN).     

N-estimation from retrospectively ascertained events with applications to AIDS.

It is a common sampling scheme in retrospective studies that the data set includes only individuals who satisfy a certain sampling criterion. In this paper we consider the situation when the sampling criterion is a specified event, and assume that an earlier event can be retrospectively identified given the occurrence of the specified event. A semiparametric method, which is a compromise between nonparametric and parametric methods, is employed for the estimation of the expected number of the specified events (namely, the N-estimation) occurring in arbitrarily given intervals. A number of statistical properties of the estimates are developed. Due to the limitation of semiparametric models, our estimates should be regarded as conservative estimates since in general they underestimate the actual number of the specified events. This type of limitation, however, cannot be avoided with nonparametric or semiparametric models. Applications to acquired immunodeficiency syndrome (AIDS) cases are considered. The blood transfusion AIDS cases reported to the Centers for Disease Control are analyzed in detail.     

A Survey of the FDA's AERS Database Regarding Muscle and Tendon Adverse Events Linked to the Statin Drug Class

Background

Cholesterol management drugs known as statins are widely used and often well tolerated; however, a variety of muscle-related side effects can arise. These adverse events (AEs) can have serious impact, and form a significant barrier to therapy adherence. Surveillance of post-marketing AEs is of vital importance to understand real-world AEs and reporting differences between individual statin drugs. We conducted a review of post-approval muscle and tendon AE reports in association with statin use, to assess differences within the drug class.

Methods

We analyzed all case reports from the FDA AE Reporting System (AERS) database linking muscle-related AEs to statin use (07/01/2005–03/31/2011). Drugs examined were: atorvastatin, simvastatin, lovastatin, pravastatin, rosuvastatin, and fluvastatin.

Results

Relative risk rates for rosuvastatin were consistently higher than other statins. Atorvastatin and simvastatin showed intermediate risks, while pravastatin and lovastatin appeared to have the lowest risk rates. Relative risk of muscle-related AEs, therefore, approximately tracked with per milligram LDL-lowering potency, with fluvastatin an apparent exception. Incorporating all muscle categories, rates for atorvastatin, simvastatin, pravastatin, and lovastatin were, respectively, 55%, 26%, 17%, and 7.5% as high, as rosuvastatin, approximately tracking per milligram potency (Rosuvastatin>Atorvastatin>Simvastatin>Pravastatin≈Lovastatin) and comporting with findings of other studies. Relative potency, therefore, appears to be a fundamental predictor of muscle-related AE risk, with fluvastatin, the least potent statin, an apparent exception (risk 74% vs rosuvastatin).

Interpretation

AE reporting rates differed strikingly for drugs within the statin class, with relative reporting aligning substantially with potency. The data presented in this report offer important reference points for the selection of statins for cholesterol management in general and, especially, for the rechallenge of patients who have experienced muscle-related AEs (for whom agents of lower expected potency should be preferred).