36

CIOMS

This chapter summarizes the functions of the Council for International Organizations of Medical Sciences (CIOMS) and the reports issued by working groups created by CIOMS. These reports have been crucial for the International Conference on Harmonization (ICH) and the development of safety regulations in North America, Europe, Japan, and elsewhere. They are worth reviewing. Keep in mind that not all proposals from the CIOMS reports were adopted, and those that were adopted were not necessarily adopted directly and without change by ICH and national regulatory authorities.

From the CIOMS website (Web Resource 36-1): “CIOMS is an international, non-governmental, nonprofit organization established jointly by WHO (World Health Organization) and United Nations Educational, Scientific and Cultural Organization (UNESCO) [Web Resource 36-2] in 1949.” The membership of CIOMS includes 60 international member organizations, representing many of the biomedical disciplines, national academies of sciences, and medical research councils. The main objectives of CIOMS are

• To facilitate and promote international activities in biomedical sciences, especially when the participation of several international associations and national institutions is deemed necessary
  
• To maintain collaborative relations with the United Nations and its specialized agencies, in particular with WHO and UNESCO
  
• To serve the scientific interests of the international biomedical community in general

CIOMS has several long-term programs, including one on drug development and use. Starting in the early 1980s, working groups composed of experts from industry and governments have been examining key issues in drug safety. They have issued many reports, several of which have served as seminal documents for procedures and regulations that ICH, the U.S. Food and Drug Administration (FDA), the European Union, Japan, and other drug safety authorities have issued. The key documents are summarized below.

The goal of this working group was “to develop an internationally acceptable reporting method whereby manufacturers could report post-marketing adverse drug reactions rapidly, efficiently and effectively to regulators.” It noted the fact that postmarketing surveillance is necessary because premarketing studies in animals and humans have “inherent limitations.” It noted the need for standardization internationally.

The report established several conventions that have largely been adopted, including the following:

• The concept and format of a report (“a CIOMS I report”) from the manufacturer receiving the event to the regulators.
  
• “Reactions” are different from “events.” “Reactions” are reports of clinical occurrences that have been judged by a physician or healthcare worker as having a “reasonable possibility” that the report has been caused by a drug. “Events” have not had a causality evaluation made, and thus may or may not be related to or associated with the drug.
  
• Causality is discussed. No particular method of assessing causality is recommended. The report recommends that manufacturers not separate out those spontaneous reports that they receive into those that seem to be drug-related and those not seemingly drug-related. The physician, by making the report to the manufacturer, indicates that there is some level of causality possible in the report. This is a “suspected reaction.” This has become a fundamental concept in most spontaneous reporting systems around the world, wherein all spontaneous reports from physicians (now extended to all healthcare providers, and in some countries, such as the United States and Canada, to consumers) are to be considered possibly related to the drug; that is, they are “reactions,” not “events.”
  
• Because labels for marketed drugs differ from country to country, it is recommended that all reactions be collected at one point and then submitted to local authorities on a country-by-country basis based on whether the reactions are labeled locally.
  
• The report discusses the four minimum requirements for a valid report: (1) an identifiable source (reporter), (2) a patient (even if not precisely identified by name), (3) a suspect drug, and (4) a suspect reaction.
  
• The report recommends that all reports be sent in as soon as received and no later than 15 working days after receipt, to create a common worldwide deadline. This concept has been adopted, but the 15 working days has been changed to 15 calendar days because of differences in the designation of “working days” and nonworking days (holidays) around the world. The reporting clock starts the date the report is first received by anyone anywhere in the company.
  
• The CIOMS I form was created. It is essentially the same form still used now. This form is to be used for reporting to regulatory authorities.
  
• Reactions are to be reported in English.

This working group proposed a standard for Periodic Safety Update Reports (PSURs) of reactions received by manufacturers on marketed drugs. This standard, with modifications from the ICH and other organizations, has been widely adopted. The document defined several key terms:

• CIOMS Reportable Cases or Reports: “serious, medically substantiated, unlabeled ADRs with the 4 elements (reporter, patient, reaction, suspect drug).”
  
• Core Data Sheet (CDS): A document prepared by the manufacturer containing all relevant safety information, including adverse drug reactions (ADRs). This is the reference for “labeled” and “unlabeled.” This concept, which has been widely accepted, has since gotten more complex, and one must distinguish labeling from listing (e.g., unlabeled and unlisted).
  
• International Birth Date (IBD): The date that the first regulatory authority anywhere in the world has approved a drug for marketing.
  
• Data Lock-Point (Cut-Off Date): The closing date for information to be included in a particular safety update.
  
• Serious: Fatal, life-threatening, involves or prolongs inpatient hospitalization.

• Reports should be semiannual and not cumulative (unless cumulative information is needed to put a safety issue into context).
  
• The same report goes to all regulatory authorities on the same date irrespective of the local (national) approval date of the drug.
  
• Reactions reported should be from studies (published and unpublished), spontaneous reports, published case reports, cases received from regulatory authorities, and other manufacturers. Duplicate reports should be eliminated.
  
• The manufacturer should do a “concise critical analysis and opinion in English by a person responsible for monitoring and assessing drug safety.”

A sample simulated PSUR is included based on a fake drug, “Qweasytrol.”

The CIOMS III guideline is now out of print but established and extended several fundamental concepts now in use in much of the world. The idea of the CDS introduced in CIOMS II was extended to the Core Safety Information (CSI). The CDS contains all of the key core data (not just safety data) on a drug. The CSI contains (only) core safety information and is a subset of the CDS. Several fundamental concepts were introduced:

• The CSI is the core safety information that should appear in all countries’ labeling for that drug. Additional information could be added at the national level, but the core information should be included in all countries’ labels. The CSI (and national labels) are guides for healthcare professionals and contain the most relevant information needed for the drug’s use.
  
• Marketing considerations should not play a role in preparing the CSI.
  
• The CSI was proposed primarily as a medical document and not as a legal or regulatory document.
  
• Every drug should have a CSI prepared and updated by the manufacturer.
  
• Adverse events (AEs) due to excipients should be included.
  
• AEs that have no well-established relationship to therapy should not be included.
  
• The CSI should include important information that physicians are not generally expected to know.
  
• As soon as relevant safety information becomes sufficiently well established, it should be included. The specific time when it is included occurs when the safety information crosses the “threshold for inclusion,” which is defined as the time when “it is judged that it will influence physicians’ decisions on therapy.”
  
• Thirty-nine factors were proposed that can be ranked and weighed for an AE for a particular drug to see whether the information has crossed the threshold. An extensive discussion on the threshold is given:

• Ten general principles were proposed:

• Where possible, frequencies should be provided, although it is admitted that this is very difficult with spontaneous safety data. A proposed classification is:
  
  
  
  • Very common: ≥1/10 (≥10%)
    
  • Common (frequent): ≥1/100 and <1/10 (≥1% and <10%)
    
  • Uncommon (infrequent): ≥1/1000 and <1/100 (≥0.1% and <1%)
    
  • Rare: ≥1/10,000 and <1/1000 (≥0.01% and <0.1%)
    
  • Very rare: <1/10,000 (<0.01%)

Many of these recommendations have been adopted in one form or another around the world, though not in their totality. The revised edition (1998/1999) of this document appeared as CIOMS V (see below).

From the preface of the report: “CIOMS IV is to some extent an extension of CIOMS II and III. It examines the theoretical and practical aspects of how to determine whether a potentially major, new safety signal signifies a shift, calling for significant action in the established relationship between benefits and risks; it also provides guidance for deciding what options for action should be considered and on the process of decision-making should such action be required.”

The report looks at the general concepts of benefit–risk analysis and discusses the factors influencing assessment, including stakeholders and constituencies, the nature of the problem (risk), the indication for drug use and the population under treatment, constraints of time, data and resources, and economic issues. It recommends a standard format and content for a benefit–risk report:

• Introduction
  
  
  
  • Brief specification/description of the drug and where marketed
    
  • Indications for use, by country, if there are differences
    
  • Identification of one or more alternative therapies or modalities, including surgery
    
  • A very brief description of the suspected or established major safety problem
  
  
• Benefit evaluation
  
  
  
  • Epidemiology and natural history of the target disease(s)
    
  • Purpose of treatment (cure, prophylaxis, etc.)
    
  • Summary of efficacy and general toleration data compared with
    
    
    
    • Other medical treatments
      
    • Surgical treatment or other interventions
      
    • No treatment
  
  
• Risk evaluation
  
  
  
  • Background.
    
  • Weight of evidence for the suspected risk (incidence, etc.).
    
  • Detailed presentations and analyses of data on the new suspected risk.
    
  • Probable and possible explanations.
    
  • Preventability, predictability, and reversibility of the new risk.
    
  • The issue as it relates to alternative therapies and no therapy.
    
  • Review of the complete safety of the drug, using diagrammatic representations when possible (risk profiles); when appropriate, focus on selected subsets of serious AEs (e.g., the three most common and three most medically serious adverse reactions).
    
  • Provide similar profiles for alternate drugs.
    
  • When possible, estimate the excess incidence of any adverse reactions known to be common to the alternatives.
    
  • When there are significant adverse reactions that are not common to the drugs compared, highlight important differences between the drugs.
  
  
• Benefit–risk evaluation
  
  
  
  • Summarize the benefits as related to the seriousness of the target disease and the purpose and effectiveness of treatment.
    
  • Summarize the dominant risks (seriousness/severity, duration, incidence).
    
  • Summarize the benefit–risk relationship, quantitatively and diagrammatically if possible, taking into account the alternative therapies or no treatment.
    
  • Provide a summary assessment and conclusion.
  
  
• Options analysis
  
  
  
  • List all appropriate options for action.
    
  • Describe the pros and cons and likely consequences (impact analysis) of each option under consideration, taking alternative therapies into account.
    
  • If relevant, outline plans or suggestions for a study that could provide timely and important additional information.
    
  • If feasible, indicate the quality and quantity of any future evidence that would signal the need for a reevaluation of the benefit–risk relationship.
    
  • Suggest how the consequences of the recommended action should be monitored and assessed.

Several examples of benefit–risk analyses are given (quinine and allergic hematologic events, felbamate and blood dyscrasias, dipyrone and agranulocytosis, temafloxacin and renal impairment and hypoglycemia, remoxipride and blood dyscrasias, clozapine and agranulocytosis, sparfloxacin and phototoxicity).

No example of a real benefit–risk report is given using this format. This type of report seems eminently possible in situations where the risk is small and there is no urgent or immediate action needed to protect the public health. However, in situations in which immediate action is needed, usually in multiple markets around the world, the preparation of such a report is probably not feasible.

Since this CIOMS IV report, several other guidelines and documents on benefit–risk analysis have been published by the FDA, European Medicines Evaluation Agency, ICH, and others (see Chapters 30–31). Most of these documents use similar conceptual frameworks for benefit–risk analyses but do not follow or propose the rigid CIOMS IV format. Clearly, however, this document served as a stimulus to a much closer and intense examination of benefit–risk analyses around the world. The document is worth reading, in particular for the specific case studies noted above.

The CIOMS V report is a 380-page document that covers a wide variety of current issues in drug safety. A summary of some of the proposals follows. Not all these recommendations are universally accepted or required.

The sources of individual case reports are recommended as follows:

Traditionally, the primary source of safety information on marketed drugs was spontaneous reports, with occasional literature reports also appearing. New types of reports are now appearing, including internet reports, solicited reports from patient support programs, surveys, epidemiologic studies, disease registries, regulatory and other databases, and licensor and licensee interactions. Consumer reports were often not analyzed unless medical validation was obtained.

The CIOMS V report makes various recommendations, some of which are noted below:

• Consumer reports
  
  
  
  • Consumer reports should be scrutinized and should receive appropriate attention.
    
  • The quality of a report is more important than its source.
    
  • Spontaneous reports are always considered to have an implied causal relationship to the drug.
    
  • Respect privacy and the laws and regulations governing it.
    
  • If a report is received from a third party, that party should be asked to encourage the consumer to report the information to his or her physician or to authorize the sponsor/authority to contact the physician directly.
    
  • All efforts should be made to obtain medical confirmation of serious unexpected consumer reports. The regulators may be in a better position to get this information if companies have been unsuccessful.
    
  • If an event is considered not to be drug related, it should be retained in the company database but not reported.
    
  • Even in the absence of medical confirmation, any ADR with significant implications for the medicine’s benefit–risk relationship should be submitted on an expedited or periodic basis.
    
  • Consumer reports should be included in PSURs in an appendix or as a statement indicating they have been reviewed and do or do not suggest new findings.
  
  
• Literature
  
  
  
  • Cases may appear in letters to the editor.
    
  • There may be a long lag time between the first detection of a signal by a researcher and his or her publication of it.
    
  • Publications may be a source of false information and signals.
    
  • Companies should search at least two internationally recognized literature databases using the International Normalized Nomenclature name at least monthly.
    
  • Broadcast and lay media should not ordinarily be monitored. If such information is made available to the company, it should be followed up.
    
  • Judgment should be used in regard to follow-up, with the strongest efforts made for serious unexpected ADRs.
    
  • If the product source or brand is not specified, a company should assume it was its product. The company should indicate in any report that the specific brand was not identified if this is the case.
    
  • If there is a contractual agreement between two or more companies (e.g., for comarketing), the contract should specify the responsibility for literature searches and reporting.
    
  • English should be the standard language for literature report translations.
    
  • Regulators should accept translation of an abstract or pertinent sections of a publication.
    
  • References cited in a publication on apparently unexpected/unlisted and serious reactions should be checked against the company’s existing database of literature reports. Articles not previously reported should be retrieved and reviewed as usual. Routine tracking down of all such sources is unrealistic unless faced with a major safety issue.
    
  • The clock starts when a case is recognized to be a valid case (reporter, patient, drug, event).
  
  
• The internet
  
  
  
  • Protection of privacy is particularly important regarding internet cases.
    
  • A blank ADR form should be provided on a website to facilitate reporting.
    
  • A procedure should be in place to ensure daily screening of a company’s or regulator’s website(s) to identify potential case reports.
    
  • Companies and regulators do not need to routinely surf the net beyond their own sites other than to actively monitor relevant special home pages (e.g., disease groups) if there is a significant safety issue.
    
  • The message should be consistent around the world because the internet does not respect geographic (or linguistic) boundaries.
  
  
• Solicited reports
  
  
  
  • Solicited ADR reports arising in the course of interaction with patients should be regarded as distinct from spontaneous unsolicited reports.
    
  • They should be processed separately and so identified in expedited and periodic reporting.
    
  • To satisfy postmarketing regulations, solicited reports should be handled in the same way as study reports: causality assessments are needed. Serious unexpected ADRs should be reported on an expedited basis.
    
  • Serious expected and nonserious solicited reports should be kept in the safety database and reported to regulators on request.
    
  • Signals may arise from solicited reports, so they should be reviewed on an ongoing basis.
  
  
• Aspects of clinical trial reports
  
  
  
  • In general, safety information reported expeditiously to regulatory authorities should be reported to all phase I, II, and III investigators who are conducting research with any form of the product and for any indication.
    
  • It is less important to notify phase IV investigators; they will ordinarily use the available up-to-date local official data sheet as part of the investigator’s brochure.
    
  • Quality of life studies should be handled like clinical trial data.
  
  
• Epidemiology: observational studies and use of secondary databases
  
  
  
  • Structured epidemiologic studies should have the same reporting rules for suspected ADR cases as clinical trials.
    
  • For epidemiologic studies, unless there is specific attribution in an individual case, its expedited reporting is generally not appropriate.
    
  • If relevant, studies should be summarized in PSURs.
    
  • Promptly notify regulators (within 15 days) if a study result shows an important safety issue (e.g., a greater risk of a known serious ADR for one drug versus another).
    
  • For manufacturers, expedited reports from comparator drug data should be forwarded to the relevant manufacturer(s) for their regulatory reporting as appropriate.
  
  
• Disease-specific registries and regulatory ADR databases
  
  
  
  • A registry is not a study. Cases should be treated as solicited reports (causality assessment required).
    
  • Although there are numerous ADR databases created by regulatory authorities, it is unnecessary to attempt to routinely collect them for regular review. If a company possesses data from a regulatory database, it should review those data promptly for any required expedited reporting. Careful screening should be done to avoid duplicates.
    
  • It is advisable to mention in the PSUR that the databases have been examined even if no relevant cases have been found.
  
  
• Licenser–licensee interactions
  
  
  
  • When companies codevelop, comarket, or copromote products, it is critical that explicit contractual agreements specify processes for exchange of safety information, including timelines and regulatory reporting responsibilities.
    
  • The time frame for expedited regulatory reporting should normally be no longer than 15 calendar days from the first receipt of a valid case by any of the partners.
    
  • The original recipient of a suspected ADR should ideally conduct any necessary follow-up; any subsequent follow-up information sent to the regulators should be submitted by the same company that reported the case originally.
  
  
• Clinical case evaluation
  
  
  
  • The company or regulatory authority staff can propose alternate clinical terms and interpretations of the case from those of the reporter, but unless the original reporter alters his or her original description in writing, the original terms must also be reported.
    
  • When a case is reported by a consumer, his or her clinical description should be retained even if confirmatory or additional information from a healthcare professional is obtained.
    
  • There is an important distinction between a suspected ADR and an “incidental” event. An incidental event occurs in reasonable clinical temporal association with the use of the drug product but is not the intended subject of the spontaneous report (it did not prompt the contact with the company or regulator). There is also no implicit or explicit expression of possible drug causality by the reporter or the company’s safety review staff. They should be included as part of the medical history and not be the subject of expedited reporting. Incidental events should be captured in the company database.
  
  
• Assessing patient and reporter identities
  
  
  
  • When cases do not meet the minimum criteria (patient, reporter, event, drug) even after follow-up, the case should be kept in the database as an “incomplete case.”
    
  • The regulatory reporting clock starts in the European Union at the first contact with a healthcare professional, but in the United States and Canada, it starts when the case is initially reported to the company, even by a consumer.
    
  • One or more of the following pieces of information automatically qualify a patient as identifiable: age, age category (e.g., teenager), sex, initials, date of birth, name, or patient number.
    
  • Even in the absence of such qualifying descriptors, a report referring to a definite number of patients should be regarded as a case as long as the other criteria for validity are met. For example, “Two patients experienced…” but not “A few patients experienced…. ”
    
  • For serious, unexpected, suspected reactions, the threshold for reporting in the absence of confirmatory identity should be lowered.
  
  
• Criteria for seriousness
  
  
  
  • Hospitalization refers to admission as an inpatient and not to an examination or treatment as an outpatient.
    
  • All congenital anomalies and birth defects, without regard to their nature or severity, should be considered serious.
    
  • There is a lack of objective standards for “life threatening” and “medical judgment” as seriousness criteria; both require individual professional evaluation that invariably introduces a lack of reproducibility.
    
  • Within a company, the tools, lists, and decision-making processes should be harmonized globally.
  
  
• Criteria for expectedness
  
  
  
  • The terminology associated with expectedness depends on which reference safety document is being used and for what purpose:
    
    
    
    • “Listed” or “unlisted” refers to the ADRs contained in the CSI for a marketed product or within the development CSI (DCSI) in the investigator’s brochure.
      
    • “Labeled” or “unlabeled” refers to the ADRs contained in official product safety information for marketed products (e.g., summary of product characteristics in the European Union or the package insert in the United States).
    
    
  • Determining whether a reported reaction is expected is a two-step process: first, is the reaction term already included in the CSI? Second, is the ADR different regarding its nature, severity, specificity, or outcome?
    
  • Expectedness should be strictly based on inclusion of a drug-associated experience in the ADR section of the CSI. Special types of reactions, such as those occurring under conditions of overdose, drug interaction, or pregnancy, should also be included in this section.
    
  • Disorders mentioned in “contraindications” or “precautions” as reasons for not treating with the drug are not expected ADRs unless they also appear in the ADR section.
    
  • If an ADR has been reported only in association with an overdose, it should be considered unexpected if it occurs at a normal dose.
    
  • For a marketed drug CSI, events cited in data from clinical trials are not considered expected unless they are included in the ADR section.
    
  • For expedited reporting on marketed drugs, local approved product information is the reference document for expectedness (labeledness).
    
  • For periodic reporting (PSUR), the CSI is the reference document for expectedness (listedness).
    
  • Disclaimer statements for causality (e.g., “X has been reported but the relationship with the drug has not been established”) are discouraged; however, even if used, the reaction X is still unexpected.
    
  • Class labeling does not count as “expected” unless the event in question is included in the ADR section.
    
  • Lack of expected efficacy is not relevant to whether an AE is expected.
    
  • If the treatment exacerbates the target indication, it would be unexpected unless already detailed in the CSI.
    
  • Unless the CSI specifies a fatal outcome for an ADR, the case is unexpected as long as there was an association between the reaction and the fatality.
  
  
• Case follow-up approaches
  
  
  
  • Highest priority for follow-up are cases that are serious and unexpected; followed by serious, expected; and nonserious, unexpected.
    
  • Cases “of special interest” (e.g., ADRs under active surveillance at the request of the regulators) also deserve high priority, as do any cases that might lead to a labeling change.
    
  • For any cases with legal implications, the company’s legal department should be involved.
    
  • When the case is serious and if the ADR has not resolved at the time of the initial report, it is important to continue follow-up until the outcome has been established or the condition stabilized. How long to follow up such cases requires judgment.
    
  • It is recommended that collaboration with other companies be done if more than one company’s drug is suspected as a causal agent in a case.
    
  • Follow-up for unexpected deaths and life-threatening cases should be done within 24 hours.
    
  • If a reporter fails to respond to the first follow-up attempt, reminder letters should be sent as follows:
    
    
    
    • A single follow-up letter for any nonserious expected case.
      
    • For all other cases, a second follow-up letter should be sent no later than 4 weeks after the first letter.
      
    • In general, when the reporter fails to respond or is incompletely cooperative, the two follow-up letters should reflect sufficient due diligence.
  
  
• Role of narratives
  
  
  
  • A company case narrative is different from the reporter’s clinical description of a case, though the reporter’s comments should be an integral part of the company narrative. The reporter’s verbatim words should be included for the adverse reactions.
    
  • Alternate causes to that given by the reporter should be described and identified as a company opinion.
    
  • The same evaluation should be supplied to all regulators.
    
  • Narratives should be prepared for all serious (expected and unexpected) and nonserious unexpected cases but not for nonserious expected cases.
    
  • Narratives should be written in the third person past tense. All relevant information should be in a logical time sequence.
    
  • In general, abbreviations (except laboratory parameters and units) and acronyms should not be used.
    
  • Time to onset of an event from the start of treatment should be given in the most appropriate time units (e.g., hours), but actual dates can be used if helpful to the reader.
    
  • If detailed supplementary records are important to a case (e.g., autopsy report), their availability should be mentioned in the narrative.
    
  • Information may be supplied by more than one person (e.g., initial reporter and supplementary information from a specialist); all sources should be specified.
    
  • When there is conflicting information provided from different sources, this should be mentioned and the sources identified.
    
  • If it is suspected that an ADR resulted from misprescribing (e.g., wrong drug or wrong dose) or other medication error, judgmental comments should not be included in the narrative because of legal implications. Only the facts should be stated (e.g., “four times the normal dose was administered,” “the prescription was misread and a contraindicated drug for this patient was given”).
    
  • The narrative should have eight sections that serve as a comprehensive stand-alone “medical story”:
    
    
    
    • Source of the report and patient demography.
      
    • Medical and drug history.
      
    • Suspect drug(s), timing and conditions surrounding the onset of the reaction(s).
      
    • The progression of the event(s) and their outcome in the patient.
      
    • If the outcome is fatal, provide relevant details.
      
    • Rechallenge information, if applicable.
      
    • The narrative preparer’s medical evaluation and comment.
  
  
• PSURs: content modification
  
  
  
  • For reports covering long time periods (e.g., 5 years), it is more practical to use the CSI current at the time of PSUR preparation.
    
  • Clinical trial data should be supplied only if they suggest a signal or are relevant to a possible change in the benefit–risk relationship.
    
  • If there are more than 200 individual case reports, submit only summary tabulations and not line listings (which may be supplied on request by the regulator).
    
  • For 5-year reports, follow-up information on cases described in the previous report should be provided only for cases associated with new or ongoing safety issues.
    
  • Inclusion of literature reports should be selective and cover publications relevant to safety findings, independent of listedness.
    
  • For PSURs with large numbers of cases, discussion and analysis for the overall safety evaluation should be by system organ class rather than by listedness or seriousness.
    
  • An abbreviated PSUR saves time and resources if little or no new safety information is generated during the time period covered. Criteria for an abbreviated report:
    
    
    
    • No serious unlisted cases
      
    • Few (e.g., ≤10) serious listed cases
      
    • No significant regulatory actions for safety
      
    • No major changes to the CSI
      
    • No findings that lead to a new action
  
  
• PSURs: a bridging report
  
  
  
  • A summary bridging report is a concise document that provides no new information and integrates two or more previously prepared PSURs to cover a specified period.
    
  • Its format follows that of a regular PSUR, but the content should consist of summary highlights of the reports being summarized.
  
  
• PSURs: an addendum report
  
  
  
  • This report is prepared on special request of the regulators to satisfy regulators who require reports covering a period outside the routine PSUR reporting cycle (e.g., if the reports are based on the local approval date in that country rather than on the IBD).
    
  • It updates the most recently completed PSUR.
    
  • It follows the usual PSUR format.
  
  
• PSURs: miscellaneous proposals
  
  
  
  • A brief (e.g., one-page) stand-alone overview (executive summary) should be provided.
    
  • Manufacturers should be allowed to select the IBDs for their old products to facilitate synchronization of PSURs.
    
  • If there is no CSI for an old product, the most suitable local labeling should be considered for use.
    
  • The evaluation of cases in a PSUR should focus on unlisted ADRs, with analyses organized primarily by system organ class (body system).
    
  • Discussion of serious unlisted cases should include cumulative data.
    
  • Complicated PSURs and those with extensive new data may require more than 60 days to prepare adequately and the regulators should be flexible.
    
  • The possibility of “resetting” the PSUR clock (from annual to semiannual reports as the result of a new indication or dosage form) should be allowed by the regulators.
  
  
• PSURs: population data
  
  
  
  • Detailed calculations on exposure (the denominator) are ordinarily unnecessary, especially given the unreliability of the numerator; rough estimates usually suffice, but the method and units used should be explained clearly.
    
  • Drug exposure data are approximate and usually represent an overestimate.
    
  • For special situations, such as dealing with an important safety signal, attempts should be made to obtain exposure information covering the relevant covariates (e.g., age, gender, race, indication, dosing details).

Stay updated, free articles. Join our Telegram channel

Join