CDISC has broad support for its fledgling standard whose evolution is critical to the acceleration of technology adoption in clinical trials.


If you missed the CDISC Clinical Data Connectathon at the DIA annual meeting in Denver, you may have missed the event of the seasonand an almost impenetrable crowd. The evening event on 10 July surprised everyone by attracting an estimated 1,000 people. It included active participation by a number of biopharmaceutical organizations.Most of the key clinical trial technology companies attended. 

What is a Connectathon you ask? Well, that takes some explaining.


The Clinical Data Interchange Standards Consortium (CDISC) organized the event, which was jointly sponsored by 23 of its corporate sponsors and members. CDISC is a nonprofit organization founded with the vision of creating data interchange standards for the pharmaceutical industry to streamline the clinical trials process. CDISC hosted the Connectathon as a proof-of-concept demonstration of its ODM XML standard.

The ODM (Operational Data Modeling) group is charged with creating a standard format for representing study metadata, study data, and administrative data associated with a clinical trial. The model uses XML (eXtensible Markup Language) to create a system-neutral exchange format for transferring data between systems and for archiving data. In essence, the ODM XML standard enables the sharing of clinical data and its associated metadata between entirely different information systems.

Transferring data between different systems and archiving data are two of the thorniest issues complicating electronic clinical trials. Creating a standard that achieves consensus among all the major clinical trial software vendors, CROs, central labs, and sponsor companies is a monumental taska task that no one has attempted previously. But without standards in place, clinical data integration typically requires a significant software development effort. The acceptance of industry standards yields data integration efficiencies and process innovations, and speeds technological advancements. 


CDISC organized the Connectathon as a demonstration to show the utility of the ODM XML standard for the exchange of structured information by clinical data producers, reviewers, and analysts. Participants were issued version 1.0 of the ODM standard, an XML dataset, and two tests they could perform to show compliance with the standard. (CDISC also provided food and drinks.)

The Connectathon brought together many vendors that demonstrated the ability to work with a common ODM XML dataset. Systems connect virtually by demonstrating the ability to read and write the same dataset. Thus, if a system can generate an ODM XML dataset, and can load that same dataset, any two systems supporting the standard can theoretically share data. Of course, the gap between the theory and reality is significant, but that shouldnt undermine the value of the Connectathon exercise.

Twenty participating companies collaborated (see box) to demonstrate compliance with the standard. Twelve of those companies submitted test evaluations to the CDISC Testing and Applications Team, which had prepared a test kit. Companies could either demonstrate that they are able to export data in a file compliant with the ODM XML standard or that they can import data into their own system from an ODM XMLcompliant document. Optionally, participants could combine the two tests by loading an ODM XML dataset into their system, modifying the data, and writing the updated data back out into a new ODM XML dataset. Of the 12 companies that submitted test evaluations, 9 completed the combined test.

The goal of the Connectathon was less about showing off a functioning standard than about displaying broad support and commitment to the standard. Real commitment means that participants actually build support for the standard into their products. The Connectathon provided a straightforward test for vendors and other companies, including biopharmaceutical companies and CROs, to show compliance with the evolving standard.Impressively, the Connectathon included a significant number of the clinical trial system vendors standing together in support of a common standard. When was the last time you saw 15 clinical trial system vendors agree to anything? Several participants went over and above the test requirements, displaying additional capabilities or showing live demonstrations of systems interacting via the standard.


Most important is practical application of the CDISC ODM standard. For the Connectathon, CDISC used data transfer between a CRO and a sponsor as the use case, or real-world example, underpinning the technical test. Other use cases include transferring data from an EDC (Electronic Data Capture) system into a CDMS (Clinical Data Management System), loading subject diary data into an EDC system, loading laboratory data into a CDMS, and archiving audit trail information.

In the Connectathon use case, a sponsor company has hired a CRO to collect, clean, and compile the data for a randomized clinical trial. In this scenario, data transfers are typically conducted at study milestones such as when some predetermined quantity of study data is entered, or when the database is locked. The sponsor company uses a commercial data management system and the CRO uses a proprietary data management system. These two systems are completely different, using distinct platforms, databases, and processing conventions. The two companies have agreed, however, to a common set of variable names and code lists and will use the CDISC XML ODM as an intermediary for data interchange.

The two companies agree on a traditional approach to implementing the ODM, such that a Form in the model represents a page in the CRF (Case Report Form), and a StudyEvent in the model represents a visit. The CRF was designed to group related data on a form organized by domainssuch as demographics or concomitant medications. 

Thus, the XML ODM standard enables the sponsor company and CRO to exchange information regardless of the fact that they maintain the data in entirely different CDM systems. Outsourcing and partnerships become simpler when clinical data can be exchanged using a standard format. By agreeing on a data standard, the two organizations can deploy the technologies that best fit their businesses, yet still share the information without the need for expensive and error-prone data integration work. 


The CDISC ODM XML standard enables software innovation. Beyond data interchange, data standards accelerate the acceptance of new technologies. Currently, commercial products are built using conflicting data standards, impeding the ability of organizations to combine best-of-breed technologies toward an optimal solution. For example, EDC adoption has been hampered by the inability to seamlessly integrate with legacy CDM systems. With the acceptance of a data interchange standard, the systems will be able to eliminate data interchange as a significant barrier to entry.

Just as data interchange standards accelerate the acceptance of new technologies, they will significantly reduce the cost to switch from one technology to another. Ideally, companies will build data management processes around the data standard, and as long as they license and develop products to that standard, they are not locked in to any one solution. That should have the positive effect of driving vendors to continually improve their products.

The application service provider (ASP) model for providing software as a service would also benefit from data interchange standards. ASP vendors would use the ODM standard in much the same way that the CRO exchanged data with the sponsor company in the Connectathon use case. For example, a pharmaceutical company could outsource data collection services to an EDC provider via the ASP model, and data exchange could take place via the ODM XML standard. This would have the added advantage of allowing the pharmaceutical company to use different EDC service providers for different trialsor even to change vendors mid-study.

The CDISC ODM group has been hard at work improving the usefulness of the ODM XML data interchange standard. Understanding the limitations of version 1.0, the group has aimed to add significant new features to a 1.1 version scheduled to be released for comment at the end of the summer. The 1.1 version will contain new capabilities such as lab data, an improved audit trail, and a more dynamic study-event model.


The Connectathon was largely a success, but serious challenges remain for the ODM standard. Primarily, these involve creating a standard complete enough for broad-scale use. Although version 1.0 of the ODM standard has significant limitations, those limitations say less about the future of the standard than about achieving an early milestone with known shortcomings. The Connectathon tests were executed with this early version of the standard and have sparked healthy feedback from participants and observers alike. Future versions will address much of what was learned from the Connectathon experience.

Achieving broad support for the standard is another challenge. If the Connectathon is any indication, support from most of the major software vendors, CROs, and pharmaceutical companies can reasonably be expected. 

And, as it turned out, getting food at the Connectathon was more challenging than one might imagine. With many more attendees than expected, the food allotment was consumed in short order. Fortunately, drinks were in plentiful supplydiffusing any potential rioting that might have been fueled by ill-fed data managers.


If you missed the Connectathon, you may have another opportunity. There have been many requests and offers to sponsor another CDISC Clinical Data Connectathon in Europe in conjunction with the Drug Information Association (DIA) Clinical Data Management meeting in Paris at the end of October. Because of the relatively tight timeline, that Connectathon will likely also use version 1.0 of the ODM specification. And given the fact that the last Connectathon drew triple the expected number of attendees, CDISC will surely schedule a larger room. Keep an eye on the CDISC Web site (cdisc.org) for more information.

 Overall, the first-ever CDISC Clinical Data Connectathon was an immense success that provided the ODM group with significant momentum. More than a technical tour de force, the Connectathon showed broad support for the fledgling standard whose evolution is critical to the acceleration of technology adoption in clinical trials. Continued industry support is essential to the ODMs ultimate success, especially given its daunting mission to create a universal clinical data interchange standard.            

Articles

Inside IT

Connecting at the Connectathon

As the public clamors for early access to promising new drugs, sponsors worry about the consequences of experimental therapy conducted outside of controlled trials.


Every researcher and study sponsor dreams of announcing promising results from early clinical studies of a new treatment, but the excitement can turn into a nightmare if the company is unable to deal with requests from dying patients for early access to a wonder drug. Instead of garnering kudos for investing in high-risk research with the hope of identifying a life-saving treatment, a company may find itself the subject of a media expos or a committee hearing on Capitol Hill.

The Food and Drug Administration (FDA) allows certain patients to obtain therapies before approval under compassionate use and treatment IND (Investigational New Drug) programs. But sponsors face the dilemma that widespread use of an experimental therapy outside of controlled trials may generate adverse events and negative results that can make it difficult to gain market approval for the new product. Manufacturers also often have very small amounts of clinical supplies and cannot provide many individuals with a new product that is still in clinical trials.

Patient empowerment is prompting regulators, health care providers, and sponsors to address these issues. FDA is examining the impact of early access programs on studiesand sees a need to clarify its policies. Pharmaceutical companies are exploring options for providing needy patients with access to experimental products as equitably as possible. A related goal is to encourage more patients to enroll in clinical studies, which may be encouraged by expanded information on trial options and improvement in coverage of health care costs related to clinical trial participation. 


Spurred by complaints about access from patients, Representative Dan Burton (Republican, Illinois), chairman of the House Committee on Government Reform, held a hearing 20 June 2001 to explore whether FDA policies or industry practices are denying patients access to experimental therapies that might save lives. Reacting to a 60 Minutes report (6 May 2001), Burton wanted to know why a small biotech company, ImClone, was able to provide a new cancer therapy to some patients but not others, and whether government red tape was the culprit.

Robert Temple, associate director for medical policy at FDAs Center for Drug Evaluation and Research (CDER), explained that FDA cannot compel a manufacturer to provide an experimental therapy to patients and that limited supply is a prime reason that companies often fail to do so. Manufacturers make small amounts of a test therapy for early drug studies, and scaling up to larger batches is expensive and not considered reasonable until there begins to be evidence that the drug is of value, Temple pointed out. Moreover, the process of packing and shipping drugs for single-patient use on an emergent basis can be very disruptive for companies conducting clinical trials. 

 Another concern is that broad early access programs may hurt clinical trial enrollment and study results. If a test drug is readily available, patients may be reluctant to participate in a formal study where they may receive an older therapy or none at all, Temple explained. Use of investigational drugs in less controlled settings and in patients with very advanced disease states may increase adverse reactions that raise questions about the therapys safety and efficacy. Temple said that FDA expects very low response rates in patients who have previously received multiple therapies and that this does not damage the drugs chance for approval, but sponsors are wary of generating a large volume of negative results. 

 FDAs Oncologic Drugs Advisory Committee (ODAC) addressed these issues at meetings in December 2000 and June 2001 as part of an effort to expand information to the public and to physicians about access to investigational cancer drugs. Agency officials asked the panel for advice on when it is appropriate for FDA to allow investigational drugs to be used for the treatment of individual cancer patientswhat is often referred to as compassionate use or a single-patient IND study. FDA permits an individual investigator to treat a patient by filing a single-patient IND that refers to an existing commercial IND. 

Interestingly enough, several patient groups opposed widespread individual access, emphasizing that it is important to develop new therapies rationally and that individual treatment should be deferred until there is adequate information to support a specific use. FDA plans to hold a larger meeting with the National Cancer Institute (NCI), academics, and patient groups to develop consensus on early-access issues. One topic is whether third parties such as NCI could administer expanded access programs for manufacturers. At the House hearing, Burton praised AstraZenecas expanded access protocol for its lung cancer treatment IRESSA, which is administered by the National Organization for Rare Disorders. 

Temple also said that FDA is working on a new rule to update and clarify what avenues are available for sponsors to offer patients access to experimental therapies outside clinical trials. FDAs Treatment IND rule was finalized in 1987 to make unapproved but promising drugslargely AIDS therapiesmore broadly available to patients needing them once clinical studies provide some evidence of efficacy. The revised rule would implement provisions in the FDA Modernization Act of 1997 (FDAMA), which codified existing FDA regulations and policies to expand patient access to experimental drugs and devices. In addition to treatment INDs, those regulations cover emergency situations and individual access. FDA also has been working to expand access to investigational cancer therapies that have been approved in other countries. All these cases, Temple emphasized, depend on obtaining evidence of safety and effectiveness to support a proposed use. 


Clearer FDA access policies, however, may not help sponsors that cannot produce enough of a test therapy to be able to distribute it to patients outside clinical trials. At the Burton hearing, ImClone president Samuel Waksal described how his company has tried, but failed, to provide its promising cancer therapyIMC-C225 (cetuximab)to patients. During five years of clinical research, the drug was tested in more than 700 individuals, including 30 compassionate use patients. But after publishing positive results from clinical trials in May 2000, the company received requests for the drug from 8500 patients. The company tried to handle the volume by putting the names on a first-come, first-served list, but had to close down its compassionate use program when the list had more people than it could possibly serve. 

The main problem, Waksal explained, is that his small company cannot produce enough of the product to meet patient demand while continuing to conduct clinical trials. As a protein derived from a living cell line, IMC-C225 must be produced under highly controlled conditions, with each step thoroughly documented to meet strict FDA guidelines. To obtain initial clinical supplies, ImClone built its own 55,000-square-foot pilot manufacturing facility. As trials and outside demand grew, the company contracted with Lonza Biologics to produce additional product at its New Hampshire contract manufacturing facility. 

In the end, the company decided that its best strategy was to concentrate on getting the drug approved by FDA and to increase product supply as expeditiously as possible. ImClone filed a rolling Biologics License Application (BLA) with FDAs Center for Biologics Evaluation and Research (CBER) in order to gain conditional FDA approval for a limited indication for the therapy in a year. The company also is investing $55 million to construct an 80,000-square-foot commercial manufacturing facility next to its New Jersey pilot plant, in anticipation of market approval. When the new plant is fully operational next year, ImClone hopes to revive its compassionate use or a broader expanded access program for patients, Waksal told the Burton committee. In the meantime, the company has had to turn down emotional pleas from patients and their families. To avoid such problems and ensure that the company can meet future commercial demand for C225, ImClone is weighing construction of another manufacturing facility with a 90,000-liter fermentation capacity, which will cost about $200 million. 


As ImClone foundand FDA officials and many patient advocates agreethe best way for patients to receive promising but unproven therapies is by participating in a controlled clinical trial, particularly at early stages in product testing and development. Clinical trials provide appropriate subject protections and maximize the collection of useful information about the product, which can benefit the entire patient population, Temple explained. FDAs Office of Special Health Issues (OSHI) helps provide information to cancer patients and other individuals on research and access opportunities. 

 OSHI efforts to steer patients to suitable clinical trials could be enhanced by expansion of the FDA-NIH data bank on clinical trials for therapies to treat serious or life-threatening diseases (www.clinicaltrials.gov), which FDAMA required the agency to establish. Most of some 4500 actively recruiting trials now on the Web site come from NIH and academic institutions, with only 157 sponsored by industry. Manufacturers say they have been waiting for FDA to issue a final guidance on what data should be included on the Web site and how to submit it. FDA took a step toward meeting that request by publishing a draft guidance describing procedures for submitting protocol information to the clinical trials data bank.

 A draft guidance issued 29 March 2000 outlined the types of trials and information that sponsors should list.

 FDA hopes to issue a final guidance by year-end. In the meantime, the agency is urging sponsors to list studies voluntarily and has been disappointed that so few choose to do so. 


Another development likely to increase patient participation in clinical studies is expansion of both public and private coverage for medical costs associated with experimental treatment. Most insurers and managed care plans now provide beneficiaries with coverage for costs associated with clinical trials, a trend encouraged by Congress and patient advocates. The patients bill of rights legislation approved by the Senate in July also specifies that health plans should cover the traditional costs of critically ill patients in clinical trials and offers protection for insurers from liability if there are problems with the outcome of a trial. 

Last year, the Clinton administration issued a policy supporting Medicare reimbursement of patient care costs associated with participation in certain clinical trials. To implement the policy, the Health Care Financing Administration (now called the Centers for Medicare and Medicaid ServicesCMS) published a national coverage statement in September 2000 spelling out the criteria a study has to meet to qualify for Medicare coverage. A main requirement is that the study evaluate a product or service covered by Medicare. That could exclude outpatient prescription drugs and many medical productsunless Congress establishes a Medicare prescription drug benefit. The policy also says, however, that Medicare will cover costs associated with trials funded by NIH and other government agencies or conducted under an IND approved by FDA. 

A government panel is supposed to decide what other clinical trials warrant Medicare coverage, which should provide some needed clarification on the program. In the past, Medicare carriers often paid for claims for routine care associated with clinical trials. That was on the decline, though, because federal agencies have been scrutinizing reimbursement and claims more closely. The administrations more explicit coverage policy is needed if more elderly patients are to enroll in clinical studies. 


1.	Food and Drug Administration, draft guidance, Information Program on Clinical Trials for Serious or Life-Threatening Diseases: Implementation Plan, Federal Register 66 (131), 3579835799 (9 July 2001); also available at www.access.gpo.gov/nara/cfr.

2. Food and Drug Administration, draft guidance, Draft Guidance for Industry on Information Program on Clinical Trials for Serious or Life-Threatening Diseases: Establishment of a Data Bank; Availability, Federal Register 65 (61), 1662016623 (29 March 2000); also available at www.access.gpo.gov/nara/cfr.            

View from Washington

Patient Demands vs. Product Development

Patients, research companies, and regulatory authorities may all benefit from proposed changes to current pharmaceutical legislation, but the proposals will provoke much debate.


The European Commission is determined to improve the possibilities for innovative medicines to win authorization in Europe. Proposals it presented in mid-July envisage some significant reforms of current pharmaceutical legislation in the European Union. These reforms are heavily geared towards speeding better medicines to patients and boosting the European pharmaceutical industrys ability to compete at world level in drug design and development.

The proposalsknown as Review 2001will shake up many aspects of European Union (EU) drug regulation, from marketing authorizations to intellectual property rights, from advertising to generic competition. And they aim to give something to everyonepatients, research companies, generic product manufacturers, and regulatory authorities responsible for human health care. According to the Commission, patients will benefit from increasing access to new and innovative medicines, from more and better information for selected prescription medicines, and from reinforced monitoring of products after launch. The research-based pharmaceutical industry should benefit from clearer procedures, faster processing for innovative products, and some improvement in intellectual property protection. The generic pharmaceuticals industry should benefit from the right to perform earlier tests on the products they are copying. And the public in general should benefit from better transparency.


The proposals take the form of three pieces of draft legislation: 

	regulation on marketing authorizations and the functioning of the EMEA

	directive on medicinal products for human use 

	directive on veterinary medicinal products. Text of the drafts can be found on the Web at dg3.eudra.org/F2/review/index.htm. Following are the main themes of the reforms.

 The London-based EMEA (European Agency for the Evaluation of Medicinal Products), in operation since 1995, is to be retained and reinforced, along with its centralized European procedure for the authorization of new medicines. All new medicines based on new chemical entities will be required to use this procedureuntil now it was optional for them, and mandatory only for biotechnology-derived products. The centralized procedure will also be open, on an optional basis, to any other product for which the applicant shows that the product constitutes a significant innovation or that there is an EU interest for patients.

 The evaluation process will be enhanced, with strengthened scientific committees and new working groups and access to experts. The Commission wants EMEA to acquire an increased and flexible scientific expertise, which could be developed in-house or by establishing communication channels outside the agency. The EMEA will use this expertise to provide early scientific advice to companies developing new products, in particular to small and medium companies developing biotechnological or innovative products. And to ensure better coherence, the EMEA will incorporate the Committee on Orphan Medicinal Products and a new Committee on Herbal Medicinal Products, a new advisory board of national authorities with a consultative function on authorization procedures, and a modified management board with representation from patients and industry.

 To speed assessment and decision procedures so that slow development does not neutralize the benefits of innovative products, the Commission proposes a U.S.-style fast-track registration procedure for products of significant therapeutic interest. When a company undertakes to perform additional monitoring and clinical studies for subsequent review, the Commission proposes a conditional marketing authorization. Valid for one year, it would apply to products with an important expected health benefitsuch as for life-threatening conditions where there is no effective therapy. It also proposes a Europe-wide system for compassionate-use availability of medicines before authorization.

 The proposal removes the five-yearly renewal procedure for marketing authorizations, replacing it with reinforced pharmacovigilance and information sharing provisions. Periodic safety reports, in the context of pharmacovigilance, will have to be prepared and reviewed more frequently than in the current system. The removal of the renewal procedure is largely compensated by a strengthening in pharmacovigilance requirements, explained the European Commissioner responsible for the proposals, Erkii Liikanen. This means that there will be an obligation for companies to monitor and analyze all adverse drug reactions.


Under these new proposals, any product may be granted marketing authorization through the EUs centralized procedure, provided that the applicant shows that the medicinal product is a significant therapeutic, scientific, or technical innovation or that an authorization is of interest at EU level to patients or to animal health. An authorization may also be granted subject to certain specific obligations (such as the provision of additional clinical trials data), to be reviewed annually by the EMEA. A Commission regulation will determine the circumstances and arrangements for granting authorizations under this rule. In exceptional circumstances, authorization may also be granted under specific conditions (such as delivery by hospital consultants only, or for highly specific conditions), which the EMEA must reassess annually. Continuation of the initial authorization may be linked to the reassessment of these conditions. 

When an application is submitted for products of major interest from the point of view of public health and in particular from the point of view of therapeutic innovation, the applicant may request an accelerated assessment procedure. If the committee accepts the application, the normal 210-day time limit is reduced to 150 days.

 Products not yet authorized for human use, but potentially of major interest from the point of view of public health, may be made available to certain patients for compassionate reasons. Where compassionate use is envisaged, the Committee for Proprietary Medicinal Products, after consulting the manufacturer or the applicant, may adopt recommendations on the conditions for use, the conditions for distribution, and the patients targeted. Member states will have to take appropriate measures to ensure that the recommendations can be implemented at national level. The EMEA will have to keep an up-to-date list of the products covered by this provision. Civil or criminal liability of the manufacturer or the applicant for marketing authorization will remain unaffected. And no product administered for compassionate reasons may be the subject of a paid transaction, except in special cases determined beforehand in national legislation. 

The EU rules on good clinical practice in the conduct of clinical trials will remain valid for such use. Once actual marketing of a product previously administered for compassionate reasons takes placefollowing the granting of a marketing authorizationthe restrictions on charging will no longer apply.


The EMEA is to set up structures and procedures for giving advice to companies, particularly regarding the development of new therapies. The advice will cover the conduct of the tests and trials necessary to demonstrate quality, safety, and efficacy. Each EMEA committee will have to establish a standing working party with the sole remit of providing scientific advice to companies. In the case of limited markets, the EMEA is to provide help to pharmaceutical companies at the time of submission of their applications. Such help is to include, in particular, taking over responsibility for some translations.

The EMEA will formally acquire the new Committees on Orphan Medicinal Products, and on Herbal Medicinal Products. Its principal scientific body, the Committee for Proprietary Medicinal Products, may be accompanied by experts in specific scientific or technical fields. In addition, it may co-opt a maximum of five additional members chosen on the basis of their specific scientific competence. Its committees will have new powers to establish working parties and expert groups, and to include in their rules of procedure precise procedures for delegating certain tasks to these working parties. And EMEA and Commission staff will be entitled to attend all the meetings of the committees and working parties. 

The EMEA will have duties to ensure early identification of potential sources of conflict between its scientific opinions and those of other EU bodies carrying out a similar task, and to work towards resolving them. The agency will have a new advisory board to give nonbinding advice on any point concerning its activities regarding the procedures for authorizing medicinal products. This board will consist of one representative from each national authority, the EMEA executive director, and representatives of the Commission. 


Companies will have to provide the competent authorities with any information relevant to the evaluation of the risks and benefits of a medicinal product, including appropriate information on post-authorization safety studies. The qualified person for pharmacovigilance will have to reside in the EU. And marketing authorization holders will be required to record and report all suspected serious adverse reactions of which they can reasonably be expected to have knowledge. They must report immediately to the member state on whose territory the incident occurred and to the EMEA, no later than fifteen days following receipt of the information.

After the first four years of marketing, periodic safety updates will have to be submitted every three years (instead of the current five years). The EMEAs role is to be expanded to explicitly cover supervision and pharmacovigilance. And the EMEA is to make available information on adverse reactions, through a database permanently accessible by all member states.

Except in exceptional circumstances, adverse reactions are to be communicated electronically. To facilitate the exchange of information on pharmacovigilance within the EU, the Commission will draw up guidelines (after consulting the EMEA, the member states, and interested parties) on the collection, verification, and presentation of adverse reaction reports. The guidelines will include technical requirements for electronic exchange of pharmacovigilance information in accordance with internationally agreed formats. The agency will also have to publish a reference to an internationally agreed medical terminology, which marketing authorization holders will be obliged to use for reporting of adverse reactions.

And in the context of pharmacovigilance particularly, the authorization holder will have to provide all information on the sales or prescription volumes at EU level and by member state, at the request of the EMEA.

Proposing sweeping changes as they do, the draft directives in Review 2001 will provoke a wide range of responses. And even when the Commission formally makes its final proposals in the autumn, after linguistic and other amendments to the current draft, the debates in the European Parliament and the Council of Ministers are expected to take several years. Even optimists agree it could be 2006 before the first changes reach the statute book.


Meanwhile, on one of the pharmacovigilance questions which will be reinforced by the new proposals, the European Commission is already locked in a battle with the EUs biggest member state. The Commission has rejected a request from Germany to maintain more stringent rules on drug safety monitoring than are required by EU law. Germanys request, says the Commission, is admissible, but unfoundedand therefore must be rejected.

The conflict has arisen from Germanys wish to retain its own legal requirement on drug manufacturers to inform the German authorities of any adverse drug reaction. EU law, modified in 2000 (by Commission Directive 2000/38), requires drug companies to report all adverse drug reactions in the EU to the authorities in the member state concerned, and to report all unexpected drug reactions in non-EU countries to the EMEA and to the authorities in the member state where the product is authorized.

Germany has defended its request on the grounds that its more extensive requirements are necessary for the protection of human life and health. But the Commission says there is no such risk in the EU systemneither the current system nor the new, more electronically-based network of drug safety information that will come fully into effect from 2002 under the terms of Directive 2000/38. Nevertheless, Germany has also brought a case against the Commission Directive in the European Court of Justice, seeking its annulment. That case is still pending.            

View from Brussels

EU Proposes Authorization Changes


Most clinical research professionals have copies of the International Conference on Harmonisation (ICH) guidelines and the 

 sections pertaining to clinical research (21 CFR 50, 21 CFR 56, 21 CFR 312), but the guidelines and regulations can seem abstract. Reading and studying the regulations in Chapter 21 of the 

Case histories gleaned from the findings of FDAs biomedical research monitoring, however, can reveal what FDA considers important in the conduct of clinical trials and the violations commonly cited by FDAs biomedical research inspectors.

GCP auditors and FDA compliance personnel often give engaging and educational presentations at meetings of regulatory affairs associations and local chapters of ACRP (Association of Clinical Research Professionals) and SoCRA (Society of Clinical Research Associates). Other excellent sources for learning more about regulatory affairs are workshops offered by other professional organizations, such as the Food and Drug Law Institute (www.fdli.org), the Drug Information Association (www.diahome.org), and the Regulatory Affairs Professionals Society (www.raps.org).

Meetings may be infrequent, however, and you may be unable to attend all the training workshops that interest you. Although its no substitute for more rigorous study, browsing the FDA (Food and Drug Administration) Warning Letters can give you a sense of the common inspection findings.

FDA issues Warning Letters to companies and individuals when an agency inspection reveals serious violations of specific regulations. Warning Letters describe the inspection and list the personnel involved from FDA and from the investigational site or company. The letter summarizes the findings, cites the specific regulations that were violated, and specifies the action to be taken. A typical warning letter ends with a request for a response within 15 days that outlines what steps will be taken to correct the violations.

 Proprietary information and information revealing subject identity are deleted before the Warning Letters are posted on the FDA Web site (www.fda.gov/foi/warning.htm). The agency does not post responses from companies or individuals.

 To read letters sent to clinical investigators, go to the FDA Web sites Warning Letter Search Form (http://63.75.126.221/scripts/wlcfm/searchwl.cfm), and type clinical investigator in the subject blank on the search form. If youre a monitor, the list of violations will probably look familiarbut one hopes it would be rare to have so many violations at a single site! One useful educational feature of the Warning Letters is that for each violation cited, FDA specifies the applicable regulation from the 

. If you dont have a copy of the CFR, you can read the regulations online at www.access.gpo.gov/nara/cfr/index.html. At the GPO Web site, choose Retrieve CFR sections by citation and enter the title, part, and section of the CFR that you want to review. For 21 CFR 312.60, for example, enter 21 as the title, 312 as the part, and 60 as the section. That search should retrieve the paragraph describing General responsibilities of investigators.

The list of Warning Letters will specify the issuing office, which might be a regional office or one of the centers, such as Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research, or Center for Devices and Radiological Health. Use the search features to find Warning Letters on the particular topics that interest you. If you want to see Warning Letters concerning gene therapy studies, specify gene therapy as the subject. You could also specify Center for Biologics Evaluation and Research as the issuing office, since gene therapy is regulated by that center. If you want to retrieve warning letters sent to CROs or pharmaceutical/biotechnology companies, search using the phrase contract research organization or sponsor.

Just for fun, try a search using the word misbranded. You may be surprised to find warning letters describing Swiss cheese with 31.1% weight-to-weight milkfat. That cheese is misbranded as Swiss cheese because Swiss cheese must have a minimum of 43% milkfat, according to 21 CFR 133.195 (a). (Ill bet you never knew that.) Thats a reminder that FDA also regulates the food industry. Other search words that produce interesting results are false, misleading, and advertising. Spending an afternoon reading Warning Letters can give you a better appreciation of what the FDA does and the scope of Title 21 of the 

, PhD, is president of Corvin Resources, Inc., P.O. Box 280, Apex, NC 27502-0280, (919) 367-8211, fax (919) 303-3728, e-mail: BarbaraR@corvinresources.com.

Notes from the Field

FDA Warning Letters

Michael R. Hamrell, PhD, Ed., (Drugs and Pharmaceutical Sciences, volume 104, Marcel Dekker, New York, 1999), 224 pages, hardcover, ISBN 0-8247-0309-X, $135.

In recent years, clinical audits have become a subject of vital importance to all those involved in pharmaceutical development. As few books have so far been written on this topic, the editor requested a group of experts to address the current issues and trends in the validation and auditing of clinical trials. Divided into nine chapters, this volume provides a comprehensive, state-of-the-art review of both fundamental and methodological aspects of the clinical audit and is a good reference for all those interested in this area. This book will be of special interest to actual and future auditors, but it will also be useful to those responsible for clinical development.

The first chapter, written by the editor, introduces the clinical development process, to put audit into perspective. The second, by Tammala Woodrum, Diana B. Lee, and Jon R. Wallace, gives a nice historical overview of drug regulations, demonstrating that controlling medical research is an evolving process. In fact, regulations for medical research are changing so rapidly that some of the books references are already obsolete. Still, this chapters careful comparison of FDA and ICH GCP guidelines is useful.

The third chapter is dedicated to source documentation. Vernette J. Molloy and Douglas R. Mackintosh ably explain the reasons for adequate source documents. The examples of GCP-compliant source documents and common deficiencies are very helpful.

In chapter 4, Richard Kiernan gives a personal view of the role and organization of a quality assurance (QA) department. This chapter covers the planning and conduct of routine and for-cause audits, and promotes the idea that QA can offer a supplemental role in education and training. 

Chapter 5, by Marijke Korteweg, describes how regulatory requirements and quality go hand-in-hand in international trials and need to be standardized for global use. This chapter will be a helpful reference for those who want to know the historical background of the regulatory framework. It also suggests how theory in this area can be put into practice.

Chapter 6 gives a comprehensive overview of all existing legislation on computer validation. However, the author, Tammala Woodrum, provides no practical guidelines on implementation.

Chapter 7, by Arthur M. Horowitz, explains the different types of FDA inspections and the most common findings. Unfortunately, he does not address inspections by European Health Authorities. 

In chapter 8, Wendy Bohaychuk and Graham Ball analyze the audit findings of 378 investigator site audits. They review 38 items of GCP compliance, assessing and discussing differences between countries. Comparing countries in this way may lead to erroneous conclusions, irrelevant to the real issues in the audits. The authors themselves recognize that differences between countries are related to regulatory requirements, the product, and the standards set by the sponsor or CRO.

The last chapter, by Frank Wells, is dedicated to fraud and discusses how to prevent, detect, and investigate fraud.

Although different authors wrote the chapters, the overlap is limited. The only jarring note is that not all authors respected the International Conference on Harmonisation (ICH) GCP terminology. In spite of the shortcomings mentioned, this book may prove useful to both novice and veteran auditors. 

 is clinical auditor, worldwide regulatory compliance, Bristol-Myers Squibb Pharmaceutical Research Institute, Waterloo, Belgium.

Media Review

The Clinical Audit in Pharmaceutical Development


Although the public historically has had little awareness of clinical research, our field has recently come to the publics attention in the worst possible way. The media have focused on the tragedies associated with very serious, and unexpected, adverse events that have occurred during recent studies. The most prominent of these are the deaths of an 18-year-old subject in a gene transfer study at the University of Pennsylvania and of a 24-year-old healthy volunteer in an asthma challenge study at The Johns Hopkins University.

Public discussions of these tragedies in the lay media have resulted in congressional hearings and other investigations into the conduct of clinical research and a perceived lack of adequate oversight to protect the rights and welfare of human research subjects. These inquiries have led to enhanced scrutiny of all facets of the clinical research communitynot only drugs, but also devices and biologics.

As might be predicted, the clinical research community has reacted in a way that borders on paranoiaparanoia at the prospect of an FDA audit resulting in the issuance of the dreaded Form 483, an action that some liken to the kiss of death for the recipient of such a document. We would like to suggest a different way to look at 483s to foster cooperation between researchers and regulators instead of fear.

 After inspecting a food, drug, medical device, or biologic establishment, FDA prepares a written report of its inspection findings, following a debriefing. This report is intended primarily for internal FDA use and is not provided to the inspected institution at the conclusion of the on-site visit. To provide the facility with its own written list of discrepancies noted during the inspection, FDA developed form FDA 483, Notice of Inspectional Observations, issued by the field investigator.Form 483 should contain only those observations that can be directly linked to a violation of regulationsnot suggestions, guidance, or other comments.   Although the 483 does not contain references to the regulations, each observation should be directly traceable to a section of the applicable regulations.

Items placed on a form 483 are the opinion of the FDA investigator on the scene, and may be subject to review by others within FDA. Current FDA policy requires that the agency write a follow-up letter to clinical investigators and IRBs. When the items on the 483 are confirmed in that letter, they may be taken as an official FDA position. Conversely, there is no guarantee that the follow-up letter will be limited to the items listed on the 483, as additional review by FDA headquarters may identify significant regulatory violations that were not listed on the 483, which will be included in the follow-up letter.

Nevertheless, this does not mean that a 483 should be ignored. A recipient of a 483 should promptly respond to FDA. An effective response addresses each item, indicating agreement and either providing a timeline for correction or requesting clarification of what FDA requires.

 Our colleagues at the Office of Human Research Protections (OHRP) in the Department of Health and Human Services (HHS), in concert with their colleagues at FDAs Office for Human Research Trials (OHRT), remind us to keep our focus on a culture of conscience as much as a culture of compliance. They also encourage us to envision a fully systemic Human Research Protections Program that assures each sponsor, site, and other constituent of the clinical research enterprise that the others are executing their responsibilities. To make such a focus a reality, we believe that it is necessary to reassert the original intent of the 483to be an instrument of prompt notification to allow the inspected entity an opportunity for correction and improvement, not an instrument of punishment.

Over the past six years, in particular, we in clinical research have been subjected to much debate about whether the system designed to protect research subjects is in jeopardy. Instead of debating whether the system is broken or irreparable, perhaps it would be more helpful to ascertain how is it incompleteand what can we do to complete it. To us, this is one of the core messages that Drs. Koski and LePay have been tirelessly trying to convey to anyone who will listen. A good understanding of the way FDA uses the 483, and what it means to receive one, can assist in that effort.

is president of Chesapeake Research Review, Inc., The Chesapeake Building, 9017 Red Branch Road, Suite 100, Columbia, MD 21045, (410) 884-2900, e-mail: felix@irbinfo.com, www.chesapeakeirb.com.
*To whom correspondence should be addressed

What a Form 483 Really Means

Feature Article

Combining the principles of adult learning with the advantages of Web technology can engage participants in important investigator meetings.

Effective Invvestigator Meetings

New systems of ethical oversight for randomized controlled trials could ease the problems caused by increasingly cumbersome restrictions.

Randomized Controlled Trials

EN ISO/IEC 17025 is more evidence that international standards are evolving to keep pace with technology and improve the performance of accredited central laboratories.

A New Challange for Central Laboratories

Meet the EAB

As well as being a committed European, Domenico Criscuolo is a powerful supporter of the Italian clinical research community.

Domenico Criscuolo

A creative mobile monitoring model could help sponsors tap the potential of southern Europe (and Mexico) as areas for clinical research.

Applied Clinical Trials-09-01-2001