Registries and Directories of Clinical Trials, plus Disclosure of Their Results and Archiving Their Data
In science the credit goes to the man who convinces the world, not the man to whom the idea first occurs.
–Sir Francis Darwin (1848-1925). From Eugenics Review (April 1914).
One definition of a registry in Webster’s Dictionary is “an official record book,” which is the way that registries are used in this chapter in relation to clinical trials. The International Conference on Harmonisation E2E Guideline (Pharmacovigilance Planning) defines a registry as “a list of patients presenting with the same characteristics.” Directories are defined as “an alphabetical or classified list.” While these definitions overlap, a distinction will be made between these terms in this chapter to facilitate the discussion of various types of information on patients, diseases and clinical trials that is increasingly being required by professional organizations, medical journals, US regulations, and the public.
In this chapter, an overall frame of reference for viewing the several types of listings, directories and registries of patients, and clinical trials and their results is presented. A schematic diagram (Fig. 83.1) illustrates how a company’s recruitment strategy leads to dissemination of information about a trial and how this information reaches a patient. Figure 83.1 also shows that after the trial is completed, the data are disseminated in several different ways, including via a potentially new public registry. One of the ultimate goals of encouraging greater public awareness of clinical trials and their results is to help minimize risks to patients taking drugs. Chapter 57 discusses various aspects of that specific topic.
A registry is not a clinical trial. Several types of registries keep track of patients who are taking a specific drug (e.g., thalidomide) or obtain data on specific adverse events. In many cases, their establishment is mandated by a regulatory agency as a condition of a drug’s approval. They can help a sponsor discern whether adverse events associated with its drug are really signals or not,
particularly for a drug that has limited distribution and use (such as an orphan drug for a rare disease). That issue is often complex because a larger number of patients with heterogeneous characteristics take a new drug after it is marketed than during its investigational period. Therefore, postmarketing data in the form of a registry provide important information to better understand the drug’s benefit-to-risk relationship.
particularly for a drug that has limited distribution and use (such as an orphan drug for a rare disease). That issue is often complex because a larger number of patients with heterogeneous characteristics take a new drug after it is marketed than during its investigational period. Therefore, postmarketing data in the form of a registry provide important information to better understand the drug’s benefit-to-risk relationship.
 Figure 83.1 Schematic illustration of how a company’s recruitment strategy leads to dissemination of information about a trial and reaches patients, via directories and other methods. The results of a clinical trial are disseminated via a variety of methods and approaches, including a proposed registry available to the public. Inclusion of the data in regulatory submissions is not shown. |
TYPES OF REGISTRIES
Patient Registries
Patient registries are private and confidential to the group(s) that establish them, whether that group is a pharmaceutical company, a central pharmacy that dispenses a drug, a physician who prescribes drugs, investigators who treat or diagnose patients, or relevant regulatory agencies. An exposure registry lists patients who are using a certain drug (e.g., thalidomide) or have a specific treatment and information about them (e.g., laboratory values showing that they are continuing to be monitored for potential problems such as liver impairment, pregnancy status). Patients are generally followed over time to collect adverse event data. These data can be used as part of a cohort study to identify rare adverse events.
Patient privacy must be respected and there are laws, such as the Health Insurance Portability and Accountability Act, that are carefully considered when a patient registry is established. Many registries will not include patient names but will indicate how information can be provided to their physician (e.g., about new risks to the patient) who will be able to inform the patient and/or make appropriate medical decisions.
These registries include data on patients who were previously treated with a drug, are currently receiving a marketed drug, or are receiving an investigational drug in a clinical trial. It may include all patients who received a specific drug, a subgroup of patients such as pregnant women who became pregnant while on the study or shortly prior to the study, or another specific subgroup that the sponsor wants (or needs) to follow and observe over time (e.g., the registry might be for children under the age of 18 exposed to the drug in order to track their growth or other characteristics, or of babies born to women who were pregnant while in the study).
One example is the prospective and retrospective registry established to assess angiographic and clinical outcomes of percutaneous coronary intervention in patients with stent thrombosis (Burzotta et al. 2007). All patients with angiographically confirmed stent thrombosis undergoing percutaneous coronary intervention in the enrolling centers during a two-year period
will be entered in the registry, and their clinical outcome during hospitalization and at one and six months after discharge will be assessed. Many additional patient registries can be readily found through the National Institutes of Health PubMed database (www.pubmed.gov) as well as many others.
will be entered in the registry, and their clinical outcome during hospitalization and at one and six months after discharge will be assessed. Many additional patient registries can be readily found through the National Institutes of Health PubMed database (www.pubmed.gov) as well as many others.
Disease Registries
These include the National Cancer Institute’s Surveillance, Epidemiology, and End Results Program and Aramis. Both are examples of longitudinal registries. Disease registries also contain data on patients and can be set up prior to launching a product to understand the disease’s natural history, including how the disease has changed over time (possibly over 100 years or longer) or they may be aimed at tracking the natural history of the disease in specific patients. These registries can be extremely helpful when little information is known about the natural history of a disease, which is particularly common for rare diseases. Such information can assist a sponsor to learn which clinical endpoints can be measured in pivotal trials. These registries may also be used as part of a case-control study where drug exposure in the group of exposed cases is compared with a control group from others inside the registry or from a group that is external to the registry. Many disease registries also can be rapidly located through the use of PubMed.
Pregnancy Registries
While the reasons for creating a pregnancy registry are similar to those previously mentioned, the primary aim of this type of registry is to keep track of those women who are pregnant and who are known (or believed) to have taken the drug or treatment while pregnant in order to determine the health of the baby. The ingestion of the drug may have been intentional (i.e., a study conducted in pregnant women) or unintentional. In either case, the child will be assessed after birth and for a period of time determined by the various groups that approved (and sponsored) the trial.
The potential problem in the case of an abnormality in a newborn being a false positive event is a fear of most drug developers. This occurred with the drug Bendectin, which was used to treat morning sickness more than 25 years ago. There are many articles, reviews, and book chapters on this subject (Huber 1993). The methodology of some pregnancy registries has been seriously flawed, particularly when they focused solely on one issue (e.g., proximity to power lines) without considering many additional and confounding factors.
Pregnant women are excluded from almost all Phase 1, 2, and 3 clinical trials, but participants sometimes become pregnant while on treatment, despite pregnancy warnings and contraceptive measures. Therefore, all data on babies born to women who were taking an investigational drug are of great importance for assessing the safety of the drug in pregnant women. Even after some drugs are marketed, it is deemed essential to follow pregnancies extremely closely, especially if the drug is known to be teratogenic in animals. Despite strict programs initiated to minimize the chance of pregnancy occurring (e.g., isotretinoin), some pregnancies do occur, and the Food and Drug Administration (FDA) and Roche introduced a series of progressively stringent restrictions over a period of several years with isotretinoin. Examples of patient registries taking the drug where prevention of pregnancy is paramount are those for thalidomide, isotretinoin, and alosetron.
Clinical Trial Registries
The term clinical trial registry has been commonly used in the newspapers and literature to denote a listing of clinical trials, usually giving basic details of the study design and entry criteria (sometimes referred to as “elements”). However, this type of registry is usually a listing of trials, whether classified by disease, geographical location, other criteria, or unsorted. Thus, a clinical trial registry is really a directory and not a registry in the sense of the previously described types of registries. This type of registry is further described below in the section on “Clinical Trial Directories.”
Registries in the Postmarketing Period
An all-patient and all-prescribing physician registry was recently established for Tysabri when the product was reintroduced, after being withdrawn for safety reasons. The objectives were to establish risk minimization and risk assessment goals for clinical practice and to also determine the incidence of some adverse events. All patients had to read and sign a form before they could receive the drug. Physicians had to read material about the drug’s risks and agree to comply with the company’s program, which involved completing a form on patients receiving the drug that had to be submitted to the company every six months. It is easy to imagine the many practical issues and challenges involved in making this type of program a success. Approvals with limited distributions (e.g., drug is only available from one or a few pharmacies, or through a special program such as the methadone program) have been applied to a number of drugs in the United States [e.g., gamma hydroxy butyrate (Xyrem), thalidomide, methadone] under Subpart H approval.
COMPANY WAREHOUSES OF CLINICAL TRIAL DATA
Many companies have created their own internal “data warehouses,” using outside vendor systems and/or in-house staff to assemble, combine, and sometimes merge data from different databases within the organization. These data warehouses are limited to internal company data on their drugs, placebo responses, and other related data. These types of data warehouses can be established for both investigational and marketed drugs, and allow a company to readily prepare meta-analyses and conduct subgroup analyses to generate hypotheses that may be tested. The data may also be used to evaluate the magnitude and time course of placebo responses in different patient populations or to conduct other analyses.
CLINICAL TRIAL DIRECTORIES FOR VARIOUS STAKEHOLDERS
As described earlier, directories differ from registries in being simple listings of studies that contain a minimal amount of information, usually entrance criteria. Both the www.clinicaltrials.gov website and the CenterWatch website tend to be incomplete, primarily because listing of most trials on these websites is not mandatory, as their primary function is to inform patients and their relatives and friends about clinical trials that are available for appropriate patients to enter. The US government’s directory at www.clinicaltrials.gov is mandatory for drug trials to treat life-threatening diseases
and is encouraged (but is not mandatory) for other trials. Listing the trial is free. CenterWatch charges fees for listing trials on their website (www.centerwatch.com). Individual companies also list on their own websites many of the trials they are sponsoring.
and is encouraged (but is not mandatory) for other trials. Listing the trial is free. CenterWatch charges fees for listing trials on their website (www.centerwatch.com). Individual companies also list on their own websites many of the trials they are sponsoring.
| ||||||||||||||||||||||||||||||||||||||||||
Many additional registries of clinical trials exist and a few that were readily found on PubMed are current clinical trials in the treatment of colorectal cancer (Mooney and Schoenfeldt 2004b), rhabdomyosarcoma (Anderson and Schoenfeldt 2005), pancreatic cancer (Mooney and Schoenfeldt 2004a), and earlystage breast cancer (Abrams, Gravell, and Cheson 2001).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
