42

The Safety Department’s Role in Clinical Research, Marketing and Sales, Labeling, Regulatory, Due Diligence, and Legal Issues

The clinical research (or medical research) department is the section of a company (or CRO) that does clinical trials and studies. There are many ways to structure such departments. Some companies have their research department do only phase II and III (developmental) trials, with a separate group handling phase I (clinical pharmacology and early safety studies mainly) and another group handling phase IV (postmarketing) studies and commitments. Alternatively, there may be separate groups handling different types of products: biologics, drugs, devices, or over-the-counter products. Other structures are by country or geographic region: U.S. studies, European Union studies, and “rest of the world” studies. Multiple hybrids of these models exist with and without outsourcing of some of the functions.

Whatever the structure of the clinical research division may be, the drug safety department must play a continuous role in these groups’ day-to-day activities because clinical studies almost always produce adverse events (AEs). The drug safety group must ensure that it receives all appropriate AEs from all trials. “Appropriate” usually means all serious AEs (SAEs), some or all non-SAEs, and all pregnancy cases (not AEs, of course, but in practice handled like AEs from a logistic point of view; see Chapter 15). In most countries of the world, there is a requirement to report certain clinical trial SAEs to the health authorities in either 7 or 15 days (expedited reports) or periodically (e.g., yearly). Unless the clinical research divisions are able to receive and process such AEs themselves, these cases must be sent to the drug safety department.

The clinical research and drug safety groups must establish a process to ensure that all SAEs are reported to the company within 24 to 48 hours after occurrence at a clinical site. This allows the company to process them and report the cases to the appropriate health authorities, particularly for the 7- or 15-calendar-day SAEs.

The investigators must be trained and sensitized (on a continual basis) to report SAEs. Usually, this means reporting all SAEs on an expedited basis, and reporting the nonserious AEs weekly, monthly, or only at the end of the study. Note that the obligation of the investigator to report SAEs to the sponsor is clearly stated in the FDA regulations: “An investigator must immediately report to the sponsor any SAE, whether or not considered drug related, including those listed in the protocol or investigator brochure and must include an assessment of whether there is a reasonable possibility that the drug caused the event” (21CFR312.64(b)). Similar obligations apply in other countries.

The process involves the capture of the AE on the case report form or in the electronic data capture (EDC) system, with transmission of the information to the company or CRO. Transmission to the company or CRO is done either electronically in the EDC system or by the older means of written case report forms or dedicated SAE collection forms that are faxed or e-mailed. This information must contain all the data necessary for the SAE case to be entered into the safety database and for preparation of an E2B transmission or a MedWatch or CIOMS I form for submission to the concerned health authorities, the clinical investigators, and IRBs/Ethics Committees when required. This process often involves queries to the investigator (done either by drug safety directly or via the clinical research department, or CRO). Care must be taken to avoid having multiple company departments sending repetitive questions to the investigator; usually a single point of contact is best from the company. This process can become exceedingly complex if SAE volume is very high, if multiple worldwide sites are involved, if a clinical research organization or other intermediaries are involved, if a cooperative group is involved (e.g., in oncology studies), or if a governmental agency (e.g., National Institutes of Health) is involved.

Some studies generate very large numbers of SAEs, such as oncology studies in very ill patients receiving toxic study drugs. If studies run long enough, just about every patient (whether on study drug, comparator, or placebo) will have an AE (though not necessarily an ADR) at some point during the trial. Normally, all SAEs are transmitted to the company within 1 or 2 days. However, for some studies, various protocol customizations may be done to ensure that the critical and important SAEs (those that are unexpected or related to the study drug) reach the company within 1–2 days and the health agencies within 7–15 days, though all SAEs should be sent to the sponsor immediately. Nonserious AEs may be reported to the company in a less urgent fashion (e.g., monthly or even at the end of the study) and to the agencies in the annual or special periodic reports.

In some cases, arrangements may be made with the health agencies overseeing a study to report certain expected SAEs periodically rather than as expedited reports. Most health agencies (including the FDA and European Union agencies) are willing to negotiate such arrangements. For large multinational trials, this may require negotiations with multiple national health authorities to reach a workable consensus that meets all needs. For example, a protocol might state that, because drug X in previous trials has been shown to produce mild gastrointestinal, urinary, and pulmonary hemorrhage in this patient group, these cases will not be reported until the end of the trial unless the patient dies or the patient requires a transfusion (for example). Hemorrhage from other body sites would be SAEs and reported to the company in 1–2 days.

Many companies, especially the larger ones, maintain two “safety databases.” One very common model has a safety database maintained by the drug safety department and a clinical trial database maintained by the clinical research/statistics groups. The drug safety database holds two sets of data: (1) the SAEs from clinical trials and (2) all spontaneous serious and nonserious reports; that is, nonserious clinical trial AEs are not kept in this safety database. This database is used for the regulatory reporting of spontaneous and clinical trial SAE expedited reports, Periodic Safety Update Reports (PSURs), and annual clinical trial safety reports (e.g., IND annual reports). The clinical research department maintains the other database that holds all safety and efficacy data from clinical trials (but no spontaneous reports) and is used in preparing the final clinical study report for each study as well as safety sections for Marketing Authorization dossiers and integrated safety summaries for New Drug Application (NDA) submissions.

The data in the two databases for any given clinical trial patient may (unfortunately) be different. One reason is due to different mechanisms of data collection. In paper-based (non-EDC) trials, the drug safety database usually receives data from the investigator using a dedicated SAE reporting form rather than multiple pages from the case report form. The clinical trial database receives the data on the pages in the CRF that are used for SAE/AE data collection. The data may differ if the SAE collection form is not filled in at the same time or by the same person as the CRF. Thus, the same SAE information is collected in two different places for each patient and may not be reconciled at the investigator site. Also, follow-up information may be entered into the CRF and not sent on the separate SAE collection form or vice versa. In addition, the drug safety group receives a clinical trial SAE within 24–48 hours after the investigator first notes it. The clinical trial database may not receive the CRF data for several weeks if the CRFs are “harvested” only every 6 weeks or so by the medical monitor. The use of EDC usually alleviates this problem if all the safety data are collected in the EDC and transmitted to the drug safety group. If the EDC system only collects safety data and notifies the company by an e-mail alert that a new SAE has been collected, the investigator may still end up sending a separate SAE sheet to the company, which could differ from the EDC data. So unless the safety data are collected and transmitted in only one place and manner, there is the risk of differing and incorrect or out-of-date information being present in the two company safety databases.

Other differences between the databases may occur if different people (one person in the drug safety department and one in the clinical research department) code each case differently. Follow-up information or corrected data may reach one database and not the other. More subtle differences may arise in the data that are collected. For example, for the clinical research department, the drug compliance (what percentage of the study drug was actually taken by the patient and which doses were missed) is critical. To the safety department, it is less important that the patient took 75% of the study drug versus 85% or 95%. To drug safety, it is a much more binary (yes/no) question of whether or not the drug was ingested at all (though, of course, dose-related effects may occur). Drug safety usually spends less time clarifying the dosing schedule followed by the patients than the clinical research department does.

It is thus necessary to reconcile the cases within the two databases to ensure that the data, or at least the key data, are identical. This can be done at the end of the trial or during the trial as each patient or groups of patients complete the study. This can be a very time-consuming procedure requiring detailed case review by the drug safety and clinical research groups. Differences must be ironed out, and sometimes new queries to the investigative sites are generated from the reconciliation. Another vexing problem that occurs occasionally for 15-day alert cases concerns important follow-up information that arrives in clinical research and is not sent to the drug safety group for submission to the health authorities as a follow-up alert report. This produces a late alert report because the data were in-house for weeks or months before the 15-day alert was sent to the health agencies.

The drug safety group may be involved in the creation by clinical research of the final study report for each individual study as well as the final integrated safety summary for the submission dossier (e.g., MA or NDA). This includes data reconciliation as well as supplying the “narrative” (or “capsule summary”) or other data sets prepared for each case to clinical research to aid them in preparing their final study reports. If there are postmarketing data held by drug safety, these will usually need to be supplied to the dossier as well.

The clinical research and drug safety groups usually collaborate when a signal is being worked up or an ad hoc health authority question is received. For example, if a health agency asks about a particular AE or groups of AEs (e.g., acute pancreatitis or all pancreatic AEs), it is usually necessary to pull the non-SAEs from the clinical research database and the SAEs from the drug safety database to capture all the AEs in question. Because of the way data are entered into the clinical research database (either at the end of the study or when batches of data arrive at the data entry site), the drug safety database is usually more up to date for SAEs at any one time because drug safety usually enters data immediately upon receipt, and clinical research may enter data periodically, when a group of CRFs arrives, or even at the end of the study. If EDC is used and the data are efficiently collected and distributed to both databases, this problem may be largely alleviated.

The use of electronic data capture, where each investigator enters data into a database in real time, making them available immediately to clinical research, statistics, drug safety, and others involved in the study, has changed to some degree the way data are handled and analyzed by companies and health authorities. The concept of a single (or distributed) central database (or data warehouse) holding all information that can be viewed by any party that needs to see it can now be created fairly easily. Thus, an investigator could enter a patient’s study data into the study database by simply logging on to a personalized, secure URL. The efficacy and safety data would go (“be pushed”) to a central server (perhaps somewhere in the “cloud”) and would be immediately accessible to the clinical research department, the drug safety group, operations, senior management, CROs, partners, accounting (to pay investigators after a certain number of patients are enrolled), and any other groups or individuals who need access. Various limits on the amount of data that is viewable can be used, allowing access only to the data needed. Alternatively, all the data can go into one central site, and each department can “pull” the data it needs when it needs them. From the drug safety point of view, anything that prevents data entry into two databases at different times, by different people using different forms or data entry screens, is a significant virtue. Setting up such systems requires much thought and work up-front before the study starts. It is hoped that drug safety is involved in creating the system, processes, and data rules from the beginning.

The clinical research department and the safety department also interact at development, project planning, and risk management meetings as well as on signaling, training sessions, preparation of annual reports, investigator brochure updates, and so on. The drug safety group is also frequently charged with sending blinded or unblinded safety information to drug safety monitoring committees or data monitoring boards. These are outside groups that evaluate the safety profile and status of a study on an ongoing basis during the study to ensure patient safety and data integrity. This may become tricky if the drug safety group breaks individual patients’ blinding codes to report the case unblinded to the health agencies and data committees but, at the same time, endeavors to maintain the blind for the clinical research and biostatistics departments (see Chapter 37 on E2A). Drug safety may also assist the clinical research department in preparing data for meetings with health authorities, health agency advisory committee meetings, responses to ad hoc queries, and so forth.

Sometimes the dealings between the two departments may become strained. The clinical research physicians and staff dealing with the drug in question often feel it is “their baby” and become protective and defensive about it. This may be because they “believe” in the drug and want to give it the benefit of the doubt as well as for potential financial reasons: bonuses and rewards are better for drugs that succeed in clinical trials and get approved for marketing. They also may feel (sometimes correctly) that they are much more familiar with the drug, its activity, safety, and behavior, than the drug safety group, which may handle a whole array of drugs and must, of necessity, devote less time to the drug in question. In addition, the clinical research group deals with both safety and efficacy and may feel they have a better appreciation of the balance between risk and benefit compared with the drug safety folks, who tend to see the risk side and rarely are involved in the benefit side. However, the drug safety personnel usually have better understanding of seriousness, labeledness, and (perhaps) causality, because they have more experience across many drug classes. From a practical point of view, a system of adjudication and resolution of disagreements (e.g., serious vs. nonserious) must exist and must function rapidly. In most, but not all situations, drug safety tends to have the final say on reportability of cases to the health authorities. The company does not want to be in a position where clinical research does not want a case reported as a 7- or 15-day report and drug safety does. A paper trail showing that the company did not take the most conservative position (i.e., reporting the case when there is some doubt) may be embarrassing at best and harmful at worst.

CROs

If the sponsor is using one or more CROs for various clinical research (or safety or regulatory functions), the safety department should be in close and continued touch with the external CRO(s). If the CRO is handling some or all of the safety functions for the company, one or more internal groups should track, both in real time and with periodic audits and quality reviews, how the safety work is being handled. Although outsourced, the pharmaceutical company still has the legal (and moral) responsibility for the safety of its drug.

Stay updated, free articles. Join our Telegram channel

Join