Safety outcomes in clinical trials

Approaches to mitigating the safety risk of efficacy are also discussed. This dichotomized viewpoint leads to the thought process wherein patient safety in clinical trials, including those for community-acquired pneumonia CAP , is considered solely in terms of prevention, reporting, and analysis of adverse events. Efficacy, under this approach, has nothing to do with safety.

This separation is, of course, happily temporary, as efficacy and safety ultimately are married in the form of a risk-benefit analysis for the new molecule. Indeed, their artificial separation at earlier stages of the drug development pathway is extremely advantageous from the perspective of ensuring process, diligence, and clarity in the drug-development world. Although this definition leaves the door wide open for a broad consideration of what should be considered an adverse event, the conventional approach has been to consider only toxicities e.

Although this approach is suitable for analytic purposes, the risk is that it perpetuates our inherent tendency to separate patient safety and efficacy considerations when we design and conduct clinical trials of antimicrobials. As responsible physicians and drug developers, we must actively guard against the unintended consequences of any artificial mental dichotomy between safety and efficacy.

With such a current and appropriate focus on safety in its traditional sense, efficacy can receive unintentional underweighting. This mental diversion might occur because of an overly narrow perspective on what constitutes patient safety in clinical trials the core message of this article or otherwise because of the press of time lines, heavy workload, or other clinical-development considerations.

In addition, during the conduct of a clinical trial, multiple constraints on the assessment of efficacy may be imposed. As discussed below, acceptable options, such as the use of a data-monitoring committee also known as a data safety—monitoring board , are time-consuming and expensive and may present a challenge in terms of ensuring the timely evaluation of data in a short-term clinical trial.

Therefore, perhaps the best time to consider efficacy as a patient safety issue is during the design of a clinical trial.

What are the likely reasons for a poor efficacy outcome in a clinical trial? In my opinion, the top 3 reasons with which we must concern ourselves are 1 incorrect dose selection, 2 incorrect dose selection, and 3 incorrect dose selection! Other areas in which we might err include 1 the choice of active comparator, 2 the selection of adjunctive antimicrobial therapy i.

Given the state of the art in dose selection for antibacterials, as championed by Drs. William Craig, George Drusano, Paul Ambrose, Johan Mouton, Helmut Dehrendorf, and many others, sponsors have the pharmacokinetic-pharmacodynamic PK-PD and other tools necessary to make extremely informed decisions about dose selection, before any patient with infection is enrolled in a clinical trial.

However, at times, even the best dose-selection processes will fail because of unanticipated and unknowable confounding factors. Using CAP as a springboard, let us explore some examples. Vancomycin, a glycopeptide, had been used successfully in therapy for pneumonia not primarily CAP for decades. Since that experience became understood, the pharmaceutical sponsors of new antimicrobials under study for pneumonia, especially hospital-acquired pneumonia, have performed additional trials to exclude some of these problems; examples include the conduct of pulmonary PK trials involving healthy volunteers tigecycline [ 4 ], telavancin [ 5 ], and iclaprim [ 6 ] and analysis of a possible surfactant interaction telavancin [ 5 ].

Robust exploration of activity in animal models of infection, with organisms relevant to the intended use, has also been performed for other earlier-development-stage molecules [ 7 , 8 ]. Considerations in a dose-selection rationale for community-acquired pneumonia CAP. A relevant example is the emergence of community-associated methicillin-resistant S. Fortunately, this organism remains a rare cause of CAP, especially in the clinical trial setting [ 9 ].

Even so, it is appropriate to apply relevant exclusion criteria i. A related potential issue is that if the MIC of the target pathogens in a CAP trial were to be higher than that expected, the best a priori PK-PD target-attainment analyses would be worthless. Unanticipated drug-drug interactions resulting in lower exposure to the study drug could also be imagined [ 10 ].

Choice of the active comparator regimen. In addition, the comparator must have an appropriate spectrum of activity; one concern, for example, is the glimmer of emerging resistance of S. Because ceftriaxone is a staple active control therapy for current CAP trials, this trend must be watched closely; the same applies for the potential emergence of community-associated MRSA, as alluded to above. Additional requirements to ensure appropriate efficacy include the use of an appropriate dose and dose frequency which may have changed since the drug's initial regulatory approval and anticipated good tolerability for the patient population under study.

The ICH has issued very useful guidance on the choice of control group [ 12 ]. Choice of adjunctive antimicrobial therapy. Given the current state of the art of rapid diagnostic testing, this is not a sure thing. Optimal adjunctive therapy should be used to ensure the best overall outcome for both treatment groups again, in noninferiority trials, a higher response rate in the control group has design benefits.

A future example may be the study of a new cephalosporin that does not have activity against community-associated MRSA. One current conundrum in CAP also relates to cephalosporins—because the spectrum of these study drugs does not include atypical pathogens, how do we provide optimal therapy for patients, without overlapping coverage that confounds efficacy?

This issue is especially problematic because, in the United States, Canada, and, increasingly, Europe, clinical practice guidelines and clinician expectations incorporate adjunctive macrolide therapy [ 13 ]. Inclusion of appropriate nonantimicrobial adjunctive therapies. Outcome also can be compromised by inadequate nonantimicrobial adjunctive therapy. Possible examples for CAP include poor pulmonary toilet methods, inadequate respiratory therapy support, inadequate mobilization, delayed hospital admission, and premature hospital discharge.

Mitigating the impact of prior antimicrobial therapy. Prior effective antimicrobial therapy may spuriously improve efficacy in CAP clinical trials, as demonstrated by the experience with daptomycin [ 3 ]. Unless this effect is recognized during trial conduct and analysis, the safety issue becomes apparent only later—in postmarketing use, when it may be difficult to detect. The solution—to avoid any prior antimicrobial use—poses major logistical consequences; for example, in the United States, treatment and reimbursement guidelines encourage prompt administration of the intial antimicrobial dose for patients with CAP, which may be difficult to achieve in a clinical trial setting because of the delays incurred during patient identification and the stepwise assessment needed for enrollment.

Another, more pragmatic solution is to limit therapy to a single dose of a short-acting antimicrobial. Better approaches to this issue are needed. Selection of the appropriate patient population. A final consideration relating to efficacy as a patient safety parameter is ensuring the inclusion of the appropriate patient population in the trial and excluding those who cannot benefit. Of course, inclusion and exclusion criteria of clinical trials attempt to do this, but the focus may be more on exclusion of patients at risk of adverse events or drug-drug interactions, as well as the special populations e.

Know the risks and potential benefits of clinical studies and talk to your health care provider before participating. Read our disclaimer for details. Last Update Posted : December 6, See Contacts and Locations. Study Description. The PROVECT study is a retrospective, multicentre study assessing the effectiveness and the safety outcomes in patients with non-valvular atrial fibrillation NVAF treated by apixaban between and 5 years follow up.

The study will use harmonized and federated hospital electronic health records EHRs from 10 Belgian hospitals. Outcomes of interest are major bleeding events leading to hospitalization safety , stroke and systemic thromboembolic events effectiveness , and all-cause mortality as an exploratory endpoint, and after confirming the data availability because the death events are not always recorded into hospital EHRs.

Detailed Description:. Although such data can be generalized to different populations, healthcare professionals in Belgium and healthcare authorities have been frequently requesting local scientific data to support the effectiveness and safety outcomes in patients treated by apixaban in local Belgium settings.

This scientific data gap is particularly evident for routine care data allowing to understand the apixaban use patterns in the Belgian poly-medicated, frail and new patients starting an apixaban treatment. MedlinePlus related topics: Atrial Fibrillation. Drug Information available for: Apixaban. FDA Resources. Outcome Measures. Time to major bleeding will be defined as the number of days from the index date to the occurrence of the first major bleeding event requiring hospitalization.

Event rate was defined as number of events divided by participant-years for first occurrence of GI bleeding events after index date was reported. Event rate was defined as number of events divided by participant-years for first occurrence of intracranial hemorrhage events after index date was reported. Event rate was defined as number of events divided by participant-years for first occurrence of other bleeding events after index date was reported. Patients who died during a hospitalization during the follow-up period will be labelled by binary indicators.

Time to discontinuation will be defined as the number of days from the date of index apixaban prescription to the date of discontinuation. Length of stay will be defined as the number of days a patient received care during an inpatient stay. Clinical outcomes can be measured by activity data such as hospital re-admission rates, or by agreed scales and other forms of measurement 1. A clinical endpoint generally refers to one of the target outcomes of the trial, but may also refer to any disease or sign that strongly motivates the withdrawal of the patient from the trial, then often termed humane clinical endpoint.

Target outcomes and primary endpoint Like all scientific investigations, clinical trials seek to confirm or reject a hypothesis generally the hypothesis that a new drug is efficient and safe at the selected doses in fighting a given disease. In order to conduct such investigation, we must select the most representative outcome, agree on the target value endpoint and calculate the size of the sample allowing a reasonable estimation of the true effect power of the trial.

It is good scientific practice to only ask one question at a time, therefore a single primary endpoint should be selected whenever possible. Primary and secondary endpoints All drugs have safety risks. Therefore, the only reason that a patient would want to take a drug would be if the drug i improved survival, ii resulted in a benefit that was detectable by the patient improvement in symptoms, improvement in functional capacity , or iii decreased the chances of developing a condition or disease complication that is itself apparent to the patient and is undesirable e.

Therefore, a primary endpoint should be a direct measure of one of these.