Factors associated with early catheterization in patients randomized to the conservative strategy in the ISCHEMIA Trial

Background

In the ISCHEMIA trial, individuals randomized to the conservative strategy (CON) could undergo coronary catheterization (cath) for suspicion of an endpoint event, persistent symptoms despite optimal medical therapy, or through protocol non-adherence. Understanding the reasons for cath in CON participants can aid in ISCHEMIA results interpretation.

Purpose

To describe the frequency of and factors associated with early cath in ISCHEMIA CON participants.

Methods

A prespecified, post-hoc analysis of the 2591 CON participants was performed with multivariable analyses to identify independent factors associated with cath within 6 months of randomization (“early cath”).

Results

Overall 8.7% (225/2591) of CON participants underwent an early cath: with 4.6% (119/2591) for a suspected endpoint, 1.6% (41/2591) for medical treatment failure, and 2.6% (67/2591) for protocol non-adherence; 67% of all these caths (151/225) occurred within the first 3 months from randomization. Independent factors associated with cath among CON participants included daily (HR=5.84, CI: 2.73–12.47, p<0.01) and weekly (HR=2.64, CI: 1.52–4.58, p<0.01) baseline angina vs no angina, severe (HR=2.02, CI: 1.03–3.95, p=0.04) and moderate baseline quality of life impairment vs no impairment (HR=2.03, CI: 1.24–3.33, p=0.01), randomization in Europe vs Asia (HR=1.83, CI: 1.15–2.9, p=0.01), with the proviso that all these characteristics were associated with cath occurring within the first 3 months of follow-up (very early cath), but not those between 3 and 6 months (proportional hazard assumption violation). Other factors independently associated with early cath were new or increasing angina pattern over 3 months pre-randomization (HR=1.79, CI: 1.33–2.39, p<0.0001) and increases in anti-anginal medication use during follow-up (HR=1.45, CI: 1.06–1.98, p=0.02). Baseline LDL-C <70mg/dL (HR=0.65 CI: 0.46–0.91, p=0.01) and a subsiding angina pattern during follow-up (HR=0.65, CI: 0.6–0.71, p<0.01) were independently associated with a reduced hazard of early cath. Neither ischemia severity nor extent of atherosclerosis on coronary imaging showed association with cath in CON participants at 6 months.

Conclusions

The rate of early cath in the ISCHEMIA CON strategy was low and driven mainly by a suspected endpoint event. Severe/moderate baseline angina and quality of life impairment were independently associated with very early cath. Chances of early cath were greater with worsening pre-randomization angina and need for additional antianginal medication, and less with well controlled LDL-C and decreasing angina pattern. The baseline severity of ischemia or extent of disease on coronary imaging were not related to early cath. These results give important insight into the coronary disease treatment trajectory in the conservative strategy of the ISCHEMIA trial, further inform real-life decision making and point to the efficacy of optimal medical therapy in reducing the need for cath.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): National Institutes of Health, National Heart Lunch and Blood Institute

Statistical controversies in clinical research: data access and sharing-can we be more transparent about clinical research? Let's do what's right for patients

Calls for greater transparency and 'open data access' in clinical research are widespread, from sources including the Executive Office of the President, which in 2013 called for increased access to the results of federally funded research. In 2015, The Institute of Medicine issued a report advocating for a multi-stakeholder effort to foster responsible data sharing, and there are many others. Open science is good for researchers, good for innovation, and good for patients. The question at the center of the open-science efforts for clinical trials should not be whether data should be shared, but rather how we can usher in responsible methods for doing so. Unfortunately, there remain numerous perceived barriers to complete transparency around clinical trial data. This paper reviews the current status of data disclosure, the barriers to achieving it and a suggestion for the future.

Strategic use of statistical thinking in drug development

The role of the statistician and statistical thinking in the pharmaceutical industry has evolved greatly in the last four or five decades. Regulatory developments and the changing face of the science of drug development have driven this evolution. The increasing regulatory requirement for statistical input in critical areas has facilitated a wider range of applications. The pace of change of science in general has brought statisticians into contact with a wider range of potential customers. More importantly, it has allowed the statistician to become increasingly involved in strategic issues with the possibility of influencing the direction of the business. However, it is not clear that the statistical profession in industry is adequately prepared for these opportunities either in attitude or training. Changing the statisticians' approach to their role and acquiring the correct training and experience are critical for the profession to optimize their contribution to the drug discovery and development processes.

Guidelines for quality assurance in multicenter trials: a position paper

In the wake of reports of falsified data in one of the trials of the National Surgical Adjuvant Project for Breast and Bowel Cancer supported by the National Cancer Institute, clinical trials came under close scrutiny by the public, the press, and Congress. Questions were asked about the quality and integrity of the collected data and the analyses and conclusions of trials. In 1995, the leaders of the Society for Clinical Trials (the Chair of the Policy Committee, Dr. David DeMets, and the President of the Society, Dr. Sylvan Green) asked two members of the Society (Dr. Genell Knatterud and Dr. Frank Rockhold) to act as co-chairs of a newly formed subcommittee to discuss the issues of data integrity and auditing. In consultation with Drs. DeMets and Green, the co-chairs selected other members (Ms. Franca Barton, Dr. C.E. Davis, Dr. Bill Fairweather, Dr. Stephen George, Mr. Tom Honohan, Dr. Richard Mowery, and Dr. Robert O'Neill) to serve on the subcommittee. The subcommittee considered "how clean clinical trial data should be, to what extent auditing procedures are required, and who should conduct audits and how often." During the initial discussions, the subcommittee concluded that data auditing was insufficient to achieve data integrity. Accordingly, the subcommittee prepared this set of guidelines for standards of quality assurance for multicenter clinical trials. We include recommendations for appropriate action if problems are detected.

An approach to the assessment of therapeutic drug interactions with fixed combination drug products

Several complex clinical and statistical issues are involved in the development of a combination drug product. The medical rationale for the combination, the intended clinical use, drug-drug interactions, and dose response are just a few of the considerations. The factorial design is a useful experimental model for evaluating combination therapies. This paper outlines some of the considerations involved in the use of this design in the study of combination drugs. Questions related to design and methods of analysis are discussed in general and through use of an example. Such a design was applied to study the combination of a vasodilator with a diuretic in the treatment of hypertension. The sample size derivation, the evaluation of "interaction," the use of ANOVA versus regression techniques, and methods of data display are reviewed. It is recognized that although these are complex issues, substantially more information per patient can be obtained than with standard parallel-design approaches. The design could be an especially powerful tool early in the development of a drug.

Data monitoring and interim analyses in the pharmaceutical industry: ethical and logistical considerations

The characteristics of data monitoring and the need for the use of data monitoring committees in clinical trials sponsored by the pharmaceutical industry differ from those of trials sponsored by government. Data monitoring is a continuous process in industry trials due to the regulatory requirements and the need to more thoroughly evaluate safety of new compounds. As part of this process, interim analyses are employed to make decisions about treatment effects. In some cases, such analyses may require the use of an external data monitoring committee to assist in the data review, analysis and decision making. A number of examples of interim analyses, with and without data monitoring committees, are discussed. Issues surrounding the need for external data monitoring committees and recommendations are presented. In particular the issues of sponsor participation in the data monitoring committee and controls of the decision making process are considered.

Continuous intravenous cimetidine decreases stress-related upper gastrointestinal hemorrhage without promoting pneumonia

OBJECTIVES

To determine whether a continuous i.v. infusion of cimetidine, a histamine-2 (H2) receptor antagonist, is needed to prevent upper gastrointestinal (GI) hemorrhage when compared with placebo and if that usage is associated with an increased risk of nosocomial pneumonia. Due to the importance of this latter issue, data were collected to examine the occurrence rate of nosocomial pneumonia under the conditions of this study.

DESIGN

A multicenter, double-blind, placebo-controlled study.

INTERVENTIONS

Patients were randomized to receive cimetidine (n = 65) as an iv infusion of 50 to 100 mg/hr or placebo (n = 66).

SETTING

Intensive care units in 20 institutions.

PATIENTS

Critically ill patients (n = 131), all of whom had at least one acute stress condition that previously had been associated with the development of upper GI hemorrhage.

MEASUREMENTS AND MAIN RESULTS

Samples of gastric fluid from nasogastric aspirates were collected every 2 hrs for measurement of pH and were examined for the presence of blood. Upper GI hemorrhage was defined as bright red blood or persistent (continuing for > 8 hrs) "coffee ground material" in the nasogastric aspirate. Baseline chest radiographs were performed and sputum specimens were collected from all patients, and those patients without clear signs of pneumonia (positive chest radiograph, positive cough, fever) at baseline were followed prospectively for the development of pneumonia while receiving the study medication. Cimetidine-infused patients experienced significantly (p = .009) less upper GI hemorrhage than placebo-infused patients: nine (14%) of 65 cimetidine vs. 22 (33%) of 66 placebo patients. Cimetidine patients demonstrated significantly (p = .0001) higher mean intragastric pH (5.7 vs. 3.9), and had intragastric pH values at > 4.0 for a significantly (p = .0001) higher mean percentage of time (82% vs. 41%) than placebo patients. Differences in pH variables were not found between patients who had upper GI hemorrhage and those patients who did not, although there was no patient in the cimetidine group who bled with a pH < 3.5 compared with 11 such patients in the placebo group. Also, the upper GI hemorrhage rate in patients with one risk factor (23%) was similar to that rate in patients with two or more risk factors (25%). Of the 56 cimetidine-infused patients and 61 placebo-infused patients who did not have pneumonia at baseline, no cimetidine-infused patient developed pneumonia while four (7%) placebo-infused patients developed pneumonia.

CONCLUSIONS

The continuous i.v. infusion of cimetidine was highly effective in controlling intragastric pH and in preventing stress-related upper GI hemorrhage in critically ill patients without increasing their risk of developing nosocomial pneumonia. While the number of risk factors and intragastric pH may have pathogenic importance in the development of upper GI hemorrhage, neither the risk factors nor the intragastric pH was predictive. Therefore, short-term administration of continuously infused cimetidine offers benefits in patients who have sustained major surgery, trauma, burns, hypotension, sepsis, or single organ failure.

Cimetidine 800 mg twice daily for healing erosions and ulcers in gastroesophageal reflux disease

Although H2-receptor antagonists have been the mainstay of therapy for gastroesophageal reflux disease (GERD), none of these agents has been approved by the FDA as effective in healing lesions. Since proton pump inhibitors may be associated with long-term disadvantages, a healing regimen with cimetidine would be useful clinically. This multicenter, randomized, double-blind study was conducted to evaluate the efficacy of cimetidine 800 mg b.i.d. in healing lesions and in providing symptomatic relief in patients with ulcerative or erosive esophagitis. Patients with greater than or equal to 8 heartburn episodes during a 1-week screening period, reflux confirmed by esophageal pH monitoring, and esophageal ulcers or erosions confirmed by endoscopy were randomized to treatment with placebo or cimetidine for 12 weeks. Cimetidine provided significantly greater (p less than 0.01) improvement (74% vs. 51%) and complete healing (67% vs. 36%) of esophageal lesions than did placebo. In these patients with erosive or ulcerative esophagitis, the median time to achieve 24 h without heartburn was 13 days with cimetidine and 30 days with placebo (p = 0.01). The mean heartburn severity score in the cimetidine group decreased rapidly during the first week and was consistently lower than in the placebo group. Cimetidine, 800 mg twice daily, is effective in promoting healing of esophageal ulcers and erosions and in providing heartburn relief in patients with symptomatic erosive/ulcerative GERD.

Beneficial effects of pinacidil on blood lipids: comparisons with prazosin and placebo in patients with hypertension. Pinacidil-Prazosin and Pinacidil-Placebo Research Groups, Lilly Research Laboratories

In two randomized, double-blind clinical trials comparing pinacidil with prazosin and with placebo in patients with hypertension, a number of statistically significant and potentially beneficial effects on blood lipids were detected in the patients taking pinacidil. Patients treated with pinacidil exhibited significant average decrements from baseline in concentrations of total and low-density lipoprotein cholesterol and triglycerides and a significant average increment in high-density lipoprotein cholesterol. The mean effects seen in the pinacidil group were significantly greater than those in the placebo group for both total cholesterol (-9.8 vs +4.2 mg/dl, p less than 0.001) and triglycerides (-21.6 vs +8.6 mg/dl, p less than 0.001). The effects seen in patients given pinacidil were also significantly greater than those seen in the patients treated with prazosin for both high-density lipoprotein cholesterol (+3.6 vs -1.0 mg/dl, p = 0.002) and triglycerides (-14.8 vs +30.3 mg/dl, p less than 0.001). Negative effects of hydrochlorothiazide and propranolol on blood lipids were not apparent in patients given pinacidil. Thus, pinacidil treatment of hypertension is associated with a beneficial effect on blood lipids, which may be of clinical significance.

Dose-effect and concentration-effect relationships of pinacidil and hydrochlorothiazide in hypertension

To determine dose-effect and concentration-effect relationships in hypertension for pinacidil and hydrochlorothiazide when given alone and together, we conducted a randomized, double-blind, 4 X 3 factorial, modified fixed-dose multicenter trial. Three hundred and eighty-four patients with essential hypertension (supine diastolic blood pressure, 95 to 110 mm Hg) were assigned to one of 12 groups that received all combinations of four doses of pinacidil (0, 12.5, 25, and 37.5 mg, b.i.d.) with three doses of hydrochlorothiazide (0, 12.5, and 25 mg, b.i.d.). Significant dose- and concentration-effect relationships were seen for pinacidil and hydrochlorothiazide on diastolic blood pressure. For pinacidil, dose- and concentration-effect relationships were steeper after the dose than before the dose. A significant interaction with hydrochlorothiazide was noted such that, when combined with 12.5 mg hydrochlorothiazide, 12.5 mg pinacidil had near-maximal effects on blood pressure at both peak and trough. Edema occurred in 47% of those who received 37.5 mg pinacidil monotherapy (19% discontinued). The administration of 12.5 mg pinacidil with 12.5 mg hydrochlorothiazide appears to be optimal for efficacy and safety.

Efficacy of once-daily cimetidine in preventing recurrence of duodenal ulcer

In a prospective multicenter trial, 88 patients with acute duodenal ulcers that were healed with ranitidine were randomly assigned to receive maintenance treatment with either cimetidine 400 mg (n = 45) or placebo (n = 43) at bedtime for six months. Ten percent of the patients experienced moderate or severe pain both during the day and at night while on placebo during the maintenance phase. The average proportion of cimetidine patients experiencing moderate or severe pain during the day or night was 50% and 80% lower than placebo, respectively. Ulcer-like symptoms prompted endoscopy in 44% (19 of 43) of the placebo patients compared with 18% (eight of 45) of patients receiving cimetidine (P = 0.009). At the completion of the maintenance study, cumulative symptomatic ulcer recurrence rates were 28% (12 of 43) for those on placebo compared with 13% (six of 45) for cimetidine patients. The adverse drug effects noted were similar between treatment groups, with no unexpected reactions reported. A low dose of cimetidine (400 mg) at bedtime effectively reduced the incidence of gastrointestinal symptoms that were severe enough to prompt endoscopy as well as the actual recurrence of ulcers in those patients who had responded to initial therapy with ranitidine, but who continued to be at increased risk of reulceration.

The efficacy of cimetidine in the treatment of "resistant" duodenal ulcers

In a prospective multicenter trial, 43 patients with acute duodenal ulcers unhealed after four weeks of treatment with an H2-receptor antagonist, ranitidine, were switched to treatment with another H2-receptor antagonist, cimetidine. Sixty-eight percent of the unhealed patients were successfully healed; of these patients, 81% were free of daytime pain and 89% were free of nighttime pain. Of those with residual pain, 71% and 50% showed improvement in daytime and nighttime pain severity, respectively. There were no unexpected adverse reactions reported or clinically significant changes in laboratory values measured during the study. It is concluded that cimetidine is highly effective both in healing duodenal ulcers unresponsive to ranitidine therapy and in providing continued relief of daytime and nighttime pain.

Monitoring versus interim analysis of clinical trials: a perspective from the pharmaceutical industry

The definitions of "interim analysis" and "monitoring" of clinical trials are often ambiguous in the current literature. The resulting confusion can lead to erroneous conclusions and misguided decisions, especially when activities that are operational or observational are evaluated in a probabilistic sense as inferential. The authors seek to define "interim analysis" and "monitoring" in a mutually exclusive fashion. These definitions will then provide the opportunity to review and categorize existing clinical trial practices and procedures. This will clarify such issues as "when to look" and "when to pay a price" (e.g., test size and power) and characterize such issues in the context of pharmaceutical industry drug development.

Clinical pharmacokinetics of pinacidil, a potassium channel opener, in hypertension

Pinacidil is a potassium channel opener that decreases blood pressure by reducing peripheral arterial resistance. In two multicenter trials, we studied the concentrations and apparent clearance of pinacidil (406 patients) and concentrations of its pyridyl-N-oxide metabolite (147 patients). Responding patients had plasma samples collected hourly for 12 hours on 2 occasions after weeks to months of treatment. Pinacidil dose was titrated from 12.5 to 75 mg b.i.d. The peak concentration of pinacidil and N-oxide and the area under the concentration-time curve (AUC) were proportional to the dose of pinacidil, with an average pinacidil concentration of 268 micrograms/L (1.02 microM) and N-oxide concentration of 172 micrograms/L (0.65 microM) for every 1 mg/kg pinacidil administered. Clearance of pinacidil (Clp = Dose/AUC) was 31 L/hr in patients younger than 45 years and 27 L/hr in those older than 60. Clp was significantly smaller in white patients compared with other races (Clp = 28 vs. 34 L/hr). Clp was significantly less in patients taking hydrochlorothiazide (27 vs. 31 L/hr) and greater in smokers (33 vs. 29 L/hr). Concomitant propranolol use did not influence Clp.

Vasodilator monotherapy in the treatment of hypertension: comparative efficacy and safety of pinacidil, a potassium channel opener, and prazosin

We compared antihypertensive effects of monotherapy with pinacidil (N = 197) or prazosin (N = 204) in a randomized, parallel, double-blind dose-titration study in which hydrochlorothiazide or propranolol could be added for adverse events or lack of efficacy. Pinacidil (12.5 to 75 mg b.i.d.) was a more potent vasodilator, producing a mean decrease in supine diastolic blood pressure (baseline = 102 to 103 +/- 9 mm Hg) of 18.8 +/- 10.0 (SD) mm Hg compared with 15.5 +/- 9.2 mm Hg with prazosin (1 to 10 mg b.i.d.; p less than 0.001). Patients responding to each drug had similar average blood pressure levels during 12-hour monitoring (137/85 mm Hg). More patients taking pinacidil required hydrochlorothiazide for edema (p = 0.008) and more taking prazosin required hydrochlorothiazide and propranolol for lack of efficacy (p less than 0.001). Tachycardia (15% to 20%) and palpitation (13% to 15%) were frequent events with both drugs. Edema (38.2% vs 22.3%) was more frequent with pinacidil (p less than 0.001) and postural hypotension (4.7% vs 1.0%) and asthenia (20.2% vs 13.2%) were more frequent with prazosin (p = 0.025; 0.062). No significant laboratory toxicity was noted. In conclusion, both pinacidil and prazosin are effective as monotherapy for hypertension. Monotherapy with pinacidil is limited by adverse events related to vasodilatation and monotherapy with prazosin is limited by lack of efficacy.

Cellular electrophysiology of clofilium, a new antifibrillatory agent, in normal and ischemic canine Purkinje fibers

Intracellular electrophysiological studies were performed on isolated canine cardiac tissues to investigate further the reported ability of clofilium (3 X 10(-8)--10(-6) M) to selectively increase action potential duration (APD) and refractoriness. In Purkinje fibers from normal dogs, clofilium did not influence (1) the rate of rise of the action potential (Vmax) elicited from normal or depolarized (10 mM potassium) resting potentials, (2) the Vmax of premature potentials elicited during the repolarization phase of a previous action potential or (3) the rate of diastolic depolarization of spontaneously firing Purkinje fibers. The diastolic interval was altered by inserting a single premature impulse during diastole or by varying the basic cycle length. Clofilium (3 X 10(-7) M) slightly reduced the time constant for the relation between diastolic interval and APD in concentrations that caused a maximal increase in APD of nonpremature impulses. In dogs subjected to occlusion of the left anterior descending coronary artery 48 hr before study, the APD of surviving Purkinje fibers was longer in the infarcted zone than in the normal zone. Clofilium (3 X 10(-8) M) increased APD in both zones but more so in the normal areas, thus reducing the disparity of APD between zones. Similarly, clofilium (3 X 10(-8) and 3 X 10(-7) M) increased the effective refractory period in both zones but more so in the normal area. The increase of APD and refractoriness in normal as well as depolarized or ischemic tissues in the absence of marked changes in Vmax and conduction may decrease the likelihood of reentrant arrhythmias and underlie the antifibrillatory effects in anesthetized dogs.