Analysis of on-treatment cancer safety events with daprodustat versus conventional erythropoiesis-stimulating agents-post hoc analyses of the ASCEND-ND and ASCEND-D trials.

BACKGROUND: The prespecified on-treatment analysis of ASCEND-ND (NCT02876835) raised concerns about a higher relative risk of cancer-related adverse events (AEs) with daprodustat vs darbepoetin in patients with anaemia of CKD. This concern was not observed in dialysis patients in ASCEND-D (NCT02879305). METHODS: ASCEND-ND randomized 3872 patients to daprodustat or darbepoetin. ASCEND-D randomized 2964 patients to daprodustat or conventional erythropoiesis-stimulating agents (ESAs). In both studies ESA comparators used different dosing intervals (3/week, 1/week, every 2 or every 4 weeks). The prespecified on-treatment approach examined relative risks for cancer AEs up to the last dose date + 1 day. In these analyses, owing to different dosing intervals between arms, Cox models were used to estimate the daprodustat effect by various follow-up periods (censoring at last dose date, last dose date + dosing intervals, or end of study). RESULTS: In ASCEND-ND, the safety of daprodustat vs darbepoetin on cancer-related AEs depended on the duration of follow-up after last dose date: hazard ratio (HR) 1.04 [95% confidence interval (CI) 0.77, 1.40] at end of study [HR 1.12 (95% CI 0.81, 1.56) for last dose date + dosing interval; HR 1.50 (95% CI 1.04, 2.15) for last dose date + 1 day]. In ASCEND-D, no excess risk of cancer-related AEs was observed with any model examined. CONCLUSIONS: Prespecified on-treatment analyses for cancer-related AEs appeared to result in biased risk estimates in ASCEND-ND by preferentially under-counting events from patients assigned to darbepoetin. Analyses accounting for longer darbepoetin dosing intervals, or extending follow-up, resulted in attenuation of effect estimates towards neutrality, similar to ASCEND-D, where ESA comparator dosing intervals are closer to daprodustat. TRIAL REGISTRATION: The ASCEND-ND trial is registered with ClinicalTrials.gov (NCT02876835); the ASCEND-D trial is registered with ClinicalTrials.gov (NCT02879305).

Treatment effect measures under nonproportional hazards.

In a clinical trial with a time-to-event endpoint the treatment effect can be measured in various ways. Under proportional hazards all reasonable measures (such as the hazard ratio and the difference in restricted mean survival time) are consistent in the following sense: Take any control group survival distribution such that the hazard rate remains above zero; if there is no benefit by any measure there is no benefit by all measures, and as the magnitude of treatment benefit increases by any measure it increases by all measures. Under nonproportional hazards, however, survival curves can cross, and the direction of the effect for any pair of measures can be inconsistent. In this paper we critically evaluate a variety of treatment effect measures in common use and identify flaws with them. In particular, we demonstrate that a treatment's benefit has two distinct and independent dimensions which can be measured by the difference in the survival rate at the end of follow-up and the difference in restricted mean survival time, and that commonly used measures do not adequately capture both dimensions. We demonstrate that a generalized hazard difference, which can be estimated by the difference in exposure-adjusted subject incidence rates, captures both dimensions, and that its inverse, the number of patient-years of follow-up that results in one fewer event (the NYNT), is an easily interpretable measure of the magnitude of clinical benefit.

A shrinkage estimator for subgroup analysis without the exchangeability assumption.

Shrinkage estimators for exploratory subgroup analyses are intuitively appealing and can greatly improve estimation over standard analysis approaches; however, adoption of these estimators has been limited by reliance on the exchangeability assumption. This paper describes a new shrinkage estimator that does not rely on this assumption. Rather than assuming that treatment effect sizes within subgroups are randomly distributed around an overall mean, this new estimator assumes that the difference between the effect sizes in any given pair of subgroups is randomly distributed around zero. The estimator is illustrated using data from a clinical trial in which the treatment effect size in one region was substantially different from the sizes in other regions. Simulation results show that the estimator has properties that are comparable to or superior to a standard shrinkage estimator when exchangeability is assumed, while allowing the flexibility to handle situations where exchangeability cannot be assumed.

Content Validity of HIT-6 as a Measure of Headache Impact in People With Migraine: A Narrative Review.

BACKGROUND: The short-form Headache Impact Test (HIT-6) is a widely used patient-reported outcome measure that assesses the negative effects of headaches on normal activity. It was developed using the general headache population and prior to the establishment of the now well-accepted FDA patient-reported guidance. OBJECTIVE: The objective of this narrative review was to examine existing qualitative research in patients with migraine and headache, providing insight into the relevance and meaningfulness of HIT-6 items to the lives of migraine patients. METHODS: Articles were identified through database searches (National Library of Medicine and Google Scholar) and review of reference lists of candidate articles. RESULTS: A total of 3227 articles were identified through database and hand searching. Of these, 12 contained patient- or expert-generated qualitative information regarding headache patients' experience (8 specific to migraine [episodic and chronic] patients and 4 citing general headache patients). The combined publications described a total of 283 patient interviews. Overarching themes and specific information were identified that provide support of the relevance of content for each HIT-6 item to migraine patients' lives. Identified effects of headaches on patients with migraine included limitations in daily activities, needing to lie down during headaches, feeling tired, being irritated by headaches, difficulty concentrating, and the experience of pain. Further, previous research specific to the HIT-6 indicated that patients understood the instructions, items, and response scales as intended by the instrument authors. CONCLUSIONS: This narrative literature review demonstrates qualitative research support for the relevance of the items of the HIT-6 in migraine patients, supporting its ongoing use in clinical migraine research and practice.

Some remaining challenges regarding multiple endpoints in clinical trials.

Despite recent advance in methods for handling multiple endpoints in clinical trials, some challenges remain. This paper discusses some of these challenges, including confusion surrounding the terminology used to describe the multiple endpoints, the justification for simultaneously testing for non-inferiority and superiority in a non-inferiority trial, lack of agreement on the situations under which multiple objectives do or do not lead to the need for a multiplicity correction, and choice of the most appropriate multiple comparisons procedure. In addition, this paper will discuss the position of the recent FDA draft guidance, Multiple Endpoints in Clinical Trials, on these issues. Copyright © 2017 John Wiley & Sons, Ltd.

Improving Collection and Analysis of Overall Survival Data.

Advances in anticancer therapies have provided crucial benefits for millions of patients who are living long and fulfilling lives. Although these successes should be celebrated, there is certainly room to continue improving cancer care. Increased long-term survival presents additional challenges for determining whether new therapies further extend patients' lives through clinical trials, commonly known as the gold standard endpoint of overall survival (OS). As a result, an increasing reliance is observed on earlier efficacy endpoints, which may or may not correlate with OS, to continue the timely pace of translating innovation into novel therapies available for patients. Even when not powered as an efficacy endpoint, OS remains a critical indication of safety for regulatory decisions and is a key aspect of the FDA's Project Endpoint. Unfortunately, in the pursuit of earlier endpoints, many registrational clinical trials lack adequate planning, collection, and analysis of OS data, which complicates interpretation of a net clinical benefit or harm. This article shares best practices, proposes novel statistical methodologies, and provides detailed recommendations to improve the rigor of using OS data to inform benefit-risk assessments, including incorporating the following in clinical trials intending to demonstrate the safety and effectiveness of cancer therapy: prospective collection of OS data, establishment of fit-for-purpose definitions of OS detriment, and prespecification of analysis plans for using OS data to evaluate for potential harm. These improvements hold promise to help regulators, patients, and providers better understand the benefits and risks of novel therapies.

Pure and hybrid causal effects on variables associated with an incident event.

This article is concerned with the assessment of the causal effect of a treatment on a variable that is defined only in the subset of patients who experience a specific event. In this case, the treatment can affect the variable directly, as well as indirectly through its effect on the occurrence and severity of the event. In this article we describe the pure (direct) and hybrid (direct and indirect) causal effects and methods typically used to assess them. When the treatment has a strong effect on the occurrence of the event, we found no method with adequate properties to address the pure causal effect; this remains an intractable statistical problem with no clear solution. Among the valid methods for assessment of the hybrid causal effect, power depends greatly on whether the treatment effect is primarily on the occurrence of the event or on the variable of interest.

Predicting analysis time in event-driven clinical trials with event-reporting lag.

For a clinical trial with a time-to-event primary endpoint, the rate of accrual of the event of interest determines the timing of the analysis, upon which significant resources and strategic planning depend. It is important to be able to predict the analysis time early and accurately. Currently available methods use either parametric or nonparametric models to predict the analysis time based on accumulating information about enrollment, event, and study withdrawal rates and implicitly assume that the available data are completely reported at the time of performing the prediction. This assumption, however, may not be true when it takes a certain amount of time (i.e., event-reporting lag) for an event to be reported, in which case, the data are incomplete for prediction. Ignoring the event-reporting lag could substantially impact the accuracy of the prediction. In this paper, we describe a general parametric model to incorporate event-reporting lag into analysis time prediction. We develop a prediction procedure using a Bayesian method and provide detailed implementations for exponential distributions. Some simulations were performed to evaluate the performance of the proposed method. An application to an on-going clinical trial is also described.

The issue of multiplicity in noninferiority studies.

BACKGROUND: In a clinical study to evaluate noninferiority (NI) of an experimental drug relative to an established therapy, it is common to further test superiority of the experimental drug after NI is established. It has been shown that no multiplicity issue exists between NI and superiority tests of the primary endpoint. However, when there is an additional, or secondary, endpoint that will be tested for superiority in a hierarchical fashion to the NI testing of the primary endpoint, it is not clear whether the overall type I error rate is strictly controlled among all the tests. PURPOSE: In this article, our goal is to evaluate if the family-wise type I error rate is strictly controlled in this setting. We also evaluate a multiplicity adjustment procedure based on the Hochberg procedure. METHODS: We use the closed testing principle to evaluate the family-wise type I error rate. Some simulations are performed to appreciate the magnitude of the potential inflation of the type I error rate. RESULTS: It is demonstrated that the family-wise type I error rate is not controlled and an appropriate multiplicity adjustment procedure must take into account the NI and superiority tests of the two endpoints. When the test statistic used for superiority testing of the primary endpoint is the same as that for the NI testing, a multiplicity adjustment method using the popular Hochberg procedure is shown to be potentially conservative. LIMITATION: The assessment is based on a simplified set-up where there is only one secondary endpoint tested for superiority in addition to the primary endpoint. CONCLUSION: It is necessary to evaluate the issue of multiplicity in non-inferiority studies to assure strict control of the family-wise type I error rate.

On the clinical meaningfulness of a treatment's effect on a time-to-event variable.

The standard analysis of a time-to-event variable often involves the calculation of a hazard ratio based on a survival model such as Cox regression; however, many people consider such relative measures of effect to be poor expressions of clinical meaningfulness. Two absolute measures of effect are often used to assess clinical meaningfulness: (1) many disease areas frequently use the absolute difference in event rates (or its inverse, the number-needed-to-treat) and (2) oncology frequently uses the difference between the median survival times in the two groups. While both of these measures appear reasonable, they directly contradict each other. This paper describes the basic mathematics leading to the two measures and shows examples. The contradiction described here raises questions about the concept of clinical meaningfulness.

Statins for secondary prevention of cardiovascular disease: the right dose.

Since the publication of the 4S trial in 1994, there has emerged a consensus that statins save lives and decrease myocardial infarctions and strokes in coronary artery disease (CAD) patients irrespective of baseline serum cholesterol. However, there is controversy over the correct dose and the utility of the treatment-to-goal (cholesterol, low-density lipoprotein) approach. To answer remaining questions about the optimal statin dose in CAD patients, we have performed simple and meta-analyses of 3 large long-term (approx. 5 years) dose-clinical response studies (TNT, IDEAL, and SEARCH) and compared the results with older data including long-term safety data. The results show that raising the dose of simvastatin or atorvastatin to 80 mg confers no mortality advantage, an increase in adverse reactions and only a slight decrease in myocardial infarctions and stroke versus a lower dose. These results suggest a cost-effective approach of a single safe dose (40 mg of inexpensive generic simvastatin or atorvastatin) for almost all CAD patients and makes treatment-to-goal and cholesterol monitoring (except to check for medication compliance) unnecessary; moreover, it is likely to improve the weakness in statin use - medication compliance.

Indirect comparisons in the comparative efficacy and non-inferiority settings.

International Conference on Harmonization E10 concerns non-inferiority trials and the assessment of comparative efficacy, both of which often involve indirect comparisons. In the non-inferiority setting, there are clinical trial results directly comparing an experimental treatment with an active control, and clinical trial results directly comparing the active control with placebo, and there is an interest in the indirect comparison of the experimental treatment with placebo. In the comparative efficacy setting, there may be separate clinical trial results comparing each of two treatments with placebo, and there is interest in an indirect comparison of the treatments. First, we show that the sample size required for a trial intended to demonstrate superiority through an indirect comparison is always greater than the sample size required for a direct comparison. In addition, by introducing the concept of preservation of effect, we show that the hypothesis addressed in the two settings is identical. Our main result concerns the logical inconsistency between a reasonable criterion for preference of an experimental treatment to a standard treatment and existing regulatory guidance for approval of the experimental treatment on the basis of an indirect comparison. Specifically, the preferred treatment will not always meet the criterion for regulatory approval. This is due to the fact that the experimental treatment bears the burden of overcoming the uncertainty in the effect of the standard treatment. We consider an alternative approval criterion that avoids this logical inconsistency.

Predictors of cardiovascular events in patients with hypertension and left ventricular hypertrophy: the Losartan Intervention for Endpoint reduction in hypertension study.

OBJECTIVE: We assessed readily available patient characteristics, including albuminuria (not included in traditional cardiovascular risk scores), as predictors of cardiovascular events in hypertension with left ventricular hypertrophy (LVH) and developed risk algorithms/scores for outcomes. METHODS: The Losartan Intervention For Endpoint reduction in hypertension (LIFE) study compared effects of losartan-based versus atenolol-based therapy on cardiovascular events in 9193 patients with hypertension and LVH. Univariate and multivariate analyses identified baseline variables with significant impact on development of the primary composite endpoint (cardiovascular death, stroke and myocardial infarction) and its components. Multivariate analysis used a Cox regression model with stepwise selection process. Risk scores were developed from coefficients of risk factors from the multivariate analysis, validated internally using naïve and jack-knife procedures, checked for discrimination and calibration, and compared with Framingham coronary heart disease and other risk scores. RESULTS: LIFE risk scores showed increasing endpoint rates with increasing quintile (first to fifth quintile, composite endpoint 2.8-26.7%, cardiovascular death 0.5-14.4%, stroke 1.2-11.3%, myocardial infarction 1.4-8.1%) and were confirmed with a jack-knife approach that adjusts for potentially optimistic bias. The Framingham coronary heart disease and other risk scores overestimated risk in lower risk patients and underestimated risk in higher risk patients, except for myocardial infarction. CONCLUSION: A number of patient characteristics predicted cardiovascular events in patients with hypertension and LVH. Risk scores developed from these patient characteristics, including albuminuria, strongly predicted outcomes and may improve risk assessment of patients with hypertension and LVH and planning of clinical trials.

Impact of diabetes mellitus on regression of electrocardiographic left ventricular hypertrophy and the prediction of outcome during antihypertensive therapy: the Losartan Intervention For Endpoint (LIFE) Reduction in Hypertension Study.

BACKGROUND: Diabetes mellitus is associated with increased cardiovascular (CV) morbidity and mortality and with greater ECG left ventricular hypertrophy (LVH); however, it is unclear whether diabetes attenuates regression of hypertensive LVH and whether regression of ECG LVH has similar prognostic value in diabetic and nondiabetic hypertensive individuals. METHODS AND RESULTS: A total of 9193 hypertensive patients (1195 with diabetes) in the Losartan Intervention For Endpoint (LIFE) Reduction in Hypertension Study were treated with losartan- or atenolol-based regimens and followed up with serial ECG and blood pressure determinations at baseline and 6 months and then yearly until death or study end. ECG LVH was defined with gender-adjusted Cornell voltage-duration product (CP) criteria >2440 mm . ms. After a mean follow-up of 4.8+/-0.9 years, patients with diabetes had less regression of CP LVH (-138+/-866 versus -204+/-854 mm . ms, P<0.001), remained more likely to have LVH by CP (56.0% versus 48.1%, P<0.001), and had higher rates of CV death, myocardial infarction, stroke, and all-cause mortality and of the LIFE composite end point of CV death, myocardial infarction, or stroke. In multivariable Cox proportional hazards models, in-treatment regression or absence of ECG LVH by CP was associated with between 17% and 35% reductions in event rates in patients without diabetes but did not significantly predict outcome in patients with diabetes. CONCLUSIONS: Hypertensive patients with diabetes have less regression of CP LVH in response to antihypertensive therapy than patients without diabetes, and regression of ECG LVH is less useful as a surrogate marker of outcomes in hypertensive patients with diabetes. These findings may in part explain the higher CV morbidity and mortality in hypertensive patients with diabetes, and the absence of a demonstrable improvement in prognosis in diabetic patients in response to regression of ECG LVH suggests a more complex interrelation between underlying LV structural and functional abnormalities and outcome in these patients.

Renal function and risk for cardiovascular events in type 2 diabetic patients with hypertension: the RENAAL and LIFE studies.

OBJECTIVE: To investigate whether a threshold exists for cardiovascular risk in type 2 diabetic patients with hypertension, the association between renal function and cardiovascular risk was examined across the entire physiological range of serum creatinine. DESIGN AND METHODS: The RENAAL and LIFE studies enrolled 1513 and 1195 patients with type 2 diabetes and hypertension, respectively. The relationship between baseline serum creatinine and the risk for a composite outcome of myocardial infarction, stroke or cardiovascular death was examined using Cox regression models. To adjust for heterogeneity between studies and treatment groups, these factors were included as strata when applicable. The analyses were conducted with adjustment for age, gender, smoking, alcohol use, blood pressure, heart rate, total and high-density lipoprotein (HDL) cholesterol, hemoglobin, albuminuria and prior cardiovascular disease. RESULTS: The hazard ratios across the baseline serum creatinine categories < 0.9 mg/dl, 0.9-1.2 mg/dl, 1.2-1.6 mg/dl, 1.6-2.8 mg/dl and >or= 2.8 mg/dl were 0.51 (95% confidence interval 0.34, 0.74), 0.74 (0.55, 1.00), 1.00 (reference), 1.24 (0.96, 1.59) and 1.67 (1.17, 2.91), respectively. Baseline serum creatinine (per mg/dl) strongly predicted the composite cardiovascular endpoint in LIFE [2.82(1.74,4.56), P < 0.001], RENAAL [1.41(1.12,1.79), P < 0.001], as well as the combined studies [1.51(1.21,1.87), P < 0.001]. CONCLUSION: A progressively higher risk for the composite cardiovascular endpoint was observed with incremental baseline serum creatinine in type 2 diabetic patients with hypertension, even within the normal range. Thus, there appears to be no serum creatinine threshold level for an increased cardiovascular risk. Baseline serum creatinine was a major independent risk factor for cardiovascular disease (www.ClinicalTrials.gov number NCT00308347).

Does albuminuria predict cardiovascular outcomes on treatment with losartan versus atenolol in patients with diabetes, hypertension, and left ventricular hypertrophy? The LIFE study.

OBJECTIVE: Our current aims were to investigate whether 1) baseline urinary albumin-to-creatinine ratio (UACR) predicted cardiovascular outcomes, 2) changes in UACR differed between treatments, 3) benefits of losartan were related to its influence on UACR, and 4) reduction in albuminuria reduced cardiovascular events. RESEARCH DESIGN AND METHODS: In 1,063 patients with diabetes, hypertension, and left ventricular hypertrophy, UACR was measured for a mean of 4.7 years. The primary composite end point included cardiovascular death, myocardial infarction, and stroke. Cox models were run including and excluding baseline and time-varying UACR. RESULTS: Increasing baseline albuminuria related to increased risk for cardiovascular events. Reductions in UACR at years 1 and 2 were approximately 33% for losartan vs. 15% for atenolol (P < 0.001). Benefits of losartan seem to be most prominent in patients with the highest level of baseline UACR, although treatment by albuminuria interaction was only significant for total mortality. Approximately one-fifth of the superiority of losartan was explained by the greater reduction of albuminuria. Risk of the primary end point was related to the in-treatment UACR. CONCLUSIONS: Lowering of albuminuria in patients with hypertension and diabetes appears to be beneficial and should be the subject of additional study in future clinical trials.

The effect of losartan versus atenolol on cardiovascular morbidity and mortality in patients with hypertension taking aspirin: the Losartan Intervention for Endpoint Reduction in hypertension (LIFE) study.

OBJECTIVES: We conducted a subgroup analysis in the Losartan Intervention For Endpoint reduction in hypertension (LIFE) study to determine whether aspirin interacted with the properties of losartan, an angiotensin-II receptor antagonist. BACKGROUND: Negative interactions between angiotensin-converting enzyme inhibitors and aspirin have been reported. There are no data reported from clinical trials about possible interactions between angiotensin-II receptor antagonists and aspirin. METHODS: The LIFE study assigned 9,193 patients with hypertension and left ventricular hypertrophy (LVH) to losartan- or atenolol-based therapy for a mean of 4.7 years, with 1,970 (21.4%) taking aspirin at baseline. The primary composite end point (CEP) included cardiovascular death, stroke, and myocardial infarction (MI). The present cohort was stratified by aspirin use at baseline. RESULTS: Blood pressures were reduced similarly in the losartan with aspirin (n = 1,004) and atenolol with aspirin (n = 966) groups. The CEP was reduced by 32% (95% confidence interval 0.55 to 0.86, p = 0.001) with losartan with aspirin compared to atenolol with aspirin, adjusted for Framingham risk score and LVH. The test for treatment versus aspirin interaction, excluding other covariates, was significant for the CEP (p = 0.016) and MI (p = 0.037). CONCLUSIONS: There was a statistical interaction between treatment and aspirin in the LIFE study, with significantly greater reductions for the CEP and MI with losartan in patients using aspirin than in patients not using aspirin at baseline. Further studies are needed to clarify whether this represents a pharmacologic interaction or a selection by aspirin use of patients more likely to respond to losartan treatment.

The impact of morbid events on survival following hospitalization for complicated myocardial infarction.

BACKGROUND: Little is known about the importance of morbid events with respect to longer term survival following MI hospital discharge. AIMS: Establish the risk of death associated with morbid events following initial discharge from MI hospitalization. METHODS: We examined the rates of morbid events (reinfarction, stroke/TIA, revascularization, heart failure (HF) hospitalization, cardiovascular hospitalization and all-cause hospitalization) and the relationships of these events to subsequent death in patients who survived the initial hospitalization for MI (n = 5301) in the OPTIMAAL trial. Events were classified as Early (< or = 30 days post discharge) and Late (> 30 days post discharge) for an average of 2.7 years follow-up. RESULTS: Death rates were higher in the Early period (0.20 deaths/patient year) than in the Late period (0.05 deaths/patient year). Once a morbid event, excluding revascularization, occurred, the acute hazard ratios (HR, determined by Cox regression) for death on the day of event were higher than at time periods following the event and were highest for reinfarction and stroke/TIA. The acute HRs for death for all 6 morbid events were especially high for events occurring during the Late period. The highest chronic HR for death was associated with HF and all-cause hospitalizations. By contrast, the chronic HR for death from revascularization in both the Early (HR = 0.3) and Late (HR = 0.4) period indicated reduced risk. CONCLUSIONS: The results document event rates following hospitalization for MI, provide quantification of the associated risk for death, and may be useful in designing clinical trials. The serious morbid events examined may serve as potential surrogate endpoints in long-term studies and identify patients that should be targeted for aggressive management.