Article written by: Hector Gamboa, RQAP-GLP
The regulations for Good Laboratory Practices (GLP) for nonclinical laboratory studies in America came into effect in June 1979. Since then, the GLP regulations (including its revision in 1987) have been recognized nationally and internationally as a robust and sound quality system to ensure the quality and integrity of safety data. It has inspired other nations, for instance, to come up with the OECD Principles of GLP which in turn has been compared with and deemed compatible (and complementary) with international quality standards such as ISO 17025. Throughout the years that the GLPs have been implemented changes in technology and business practices have given rise to new needs in nonclinical testing facilities. The Society of Quality Assurance (SQA), a premiere organization for Quality Professionals in the Regulated Research industry, government and academia, has been in the forefront of calling for tweaking the GLPs and modernizing it to meet the current needs of the changing nonclinical testing facilities. As a response to many years of pleas from industry and quality professionals, on August 24, 2016 the US FDA has released its proposed rule (NPRM) in revising the current GLP regulations. Did the NPRM address the needs of the GLP community? Did it, to borrow President Donald Trump’s words, make GLP great again? The SQA has evaluated the FDA’s NPRM and this article shows the key areas of concern by the SQA as well as a couple of added objections.
The following shows the key areas of concern of the SQA regarding the NPRM.
The NPRM has a comment regarding whether the regulations have “…sufficient flexibility to allow the regulated parties to meet these requirements in a manner that best suits their organizational needs.” The current regulations have adequate flexibility to perform any type of safety study and maintain clear responsibilities of the study director, sponsor, testing facility management, QAU, and study personnel, whether the study is performed by the sponsor, performed by a contract research organization, or performed as a multi-site study. The current regulations can also be applied to the extent practical in support of FDA Animal Rule studies.
The proposed rule lacks clear accountability for who is responsible for various aspects of a study (e.g., it is unclear how a contributing scientist differs from a principal investigator). The proposed regulations attempt to provide a rigid structure dictating how studies are conducted rather than providing a framework that allows testing facility management and the study director to perform the study.
Animal Rule/Animal Care – Welfare
SQA recommends creation of a separate regulation to address animal rule studies, since they have a significant impact on public health, and should therefore be clearly defined and regulated. The reiteration of animal care and use requirements is not necessary and blurs the lines of responsibility between study director, testing facility management, attending veterinarian, and Institutional Animal Care and Use Committee.
Conflict with OECD
The remit of the OECD GLP Principles is environmental chemicals, rather than pharmaceuticals and medical devices. The latter have been included as common practice because the same monitoring agencies tend to monitor both chemicals and pharmaceuticals. The OECD GLP Principles, and specifically Consensus Document 13, are driven by environmental chemical testing at field sites. FDA should consider if this is the appropriate model to follow in the United States for FDA GLP regulated articles.
Multi-site studies have been performed in the United States prior to the promulgation of the current GLPs, and the preamble to the 1976 proposed rule addresses studies or portions of studies run by contractors. The roles and responsibilities are clear in the current GLPs, and individual accountability is maintained for studies run by a sponsor, studies run entirely by a contract research organization, or studies run at multiple facilities. Additional requirements regarding multi-site studies will not add value, will add complexity, and may negatively impact study conduct.
The proposed definition weakens one of the fundamental precepts of GLPs, i.e., the study director serves as the single point of study control. The landmark cases that led to the development of FDA GLPs were replete with examples of multiple points of study control, the absence of study control, or unqualified individuals in positions of study control. Diminishing the role of the study director, as this proposal does, by placing other individuals in equivalent or superior positions of study control will lead to the same problems that existed before the current GLP regulations were issued. Study director responsibilities were addressed in the 1978 Final Rule preamble, comment #59, “The study director is charged with the technical direction of a study including interpretation, analysis, documentation, and reporting of results.”
Responsibility for the availability of procedures, resources, and facilities should reside with the testing facility management of an organization, rather than a scientist that is responsible for the technical conduct of a study. The difference between management responsibility for administrative activities and technical conduct of a study was thoroughly addressed in the 1978 Final Rule preamble, comment #58, “Duties which are more administrative than scientific are the responsibility of management; however, management may delegate appropriate administrative duties to the study director.” Management with executive responsibility should not apply to management of a “test site” unless the intention is to have every test site have a complete quality management system. A consulting cardiologist or ophthalmologist could satisfy the definition of a test site, thereby requiring such individuals to have a complete quality management system; this is improbable and would not be feasible.
SQA questions the need to define a “fully implemented GLP Quality System.” FDA GLPs have defined a robust quality management system for nonclinical laboratory studies for over 30 years. It has been recognized nationally and internationally as a sound and effective quality system regulation and has also served as the basis of commercial quality system standards, e.g., ISO 17025. The assertion that it is not a fully implemented quality system is incongruous with its development and over 30 years of successful implementation.
The ability to view QA audit findings does not take into account how QA was designed to be FDA’s ‘left hand’ on site, with total ability to voice compliance concerns without repercussion (as evidenced by their reporting relationship to be separate of and independent from study activity). If this proposal is accepted, management may request QA not to report significant findings. Personnel may focus on the number of findings/observations rather than the importance of the findings/observations and content of the audits. If this requirement is kept, SQA anticipates a reduction in the number of QA audits performed to meet a minimal standard. If implemented as proposed, the amended GLPs may adulterate the mechanism for QA audits to be used by management to improve quality and compliance. We concur and support the FDA Commissioner’s position as stated in the 1978 Preamble, comment #91, “The Commissioner shares the concerns of the comments that general FDA access to QAU inspection reports would tend to weaken the inspection system. He believes that FDA’s review of quality assurance programs is important, and he recognizes the need to maintain a degree of confidentially if QAU inspections are to be complete and candid.” In the event that FDA determines a specific and significant need to review QA audit reports (i.e., in order to determine if systematic failures exist), there currently exists a formal mechanism by which to do so.
Notification to FDA When Archived Materials are Relocated
Almost every study performed at a contract research organization will have a data transfer from its archive to another archive after the study is finalized. SQA is unaware of any situations where study data could not be found to support an FDA investigator’s review. This requirement would add additional record keeping and reporting requirements to almost all contracted studies without enhancing the expedient retrieval of raw data. The processes used by testing facilities has been successful and a new requirement for reporting data transfer to FDA is not necessary.
Differentiating quality assurance inspections into “facility-based, process-based, and study-based” adds unnecessary complexity. The requirement to “inspect every study at frequencies adequate to assure the integrity of the study” has effectively provided QA with the authority to perform inspections and audits whether study-based, process-based, or facility-based.
Tracking Draft Reports
This requirement will be unenforceable. SQA concurs that it is important to have effective collaboration between sponsors and contract research organizations to complete studies in a timely manner. However, at what point is a final draft of a study report the final draft? Until concurrence is reached by the sponsor and study director, there may be multiple iterations of a draft report. Defining, tracking, and archiving all drafts will be cumbersome and time consuming when the study director should only have responsibility for the signed final report. The significance of signing the final report must be maintained.
Revising the meaning of testing facility to separate it from exposure of the test system to the test article has numerous ramifications to the structure and operation of the quality system and to the integrity of nonclinical laboratory studies. This revision will permit sponsors to declare themselves to be the testing facility when in reality they have little to no direct involvement in the conduct of the study. Contract testing facilities (under the current regulations) will need to align their respective quality systems with multiple sponsors and will no longer be able to operate their organizations in a consistent and reliable manner. The end result may be compromised data integrity and study compliance. Redefining testing facility also has ramifications for how quality assurance will be performed. Should the sponsor declare itself the testing facility and testing facility management with executive authority, they are likely to designate their quality assurance organization as the lead quality assurance organization. Such a construct would put the sponsor in control of when, what, and how frequently quality assurance inspections are performed, thereby further diminishing contract testing facility (under the current regulations) operational control and result in inconsistent and unreliable quality assurance oversight of testing facility operations.
If one were to review the NPRM in detail what seems to stand out is the apparent attempt to assimilate the GLPs with the Quality System Regulations (QSR), which is the current Good Manufacturing Practice requirements for medical devices intended for human use, instead of the modernization being pushed by quality and industry professionals in the GLP world. In fact, a few of the new terms and items in the NPRM came from the QSRs. Adopting these to the GLPs is somewhat like trying to shoehorn a large elephant’s foot inside a person’s size 7 shoe. In a lot of cases the rules applicable to manufacturing are not applicable for early development and are often unnecessary. Take for instance the proposed requirement for a “Management Representative” (MR). Under the QSRs the MR is defined as:
“…a member of management who, irrespective of other responsibilities, shall have established authority over and responsibility for:
(i) Ensuring that quality system requirements are effectively established and effectively maintained in accordance with this part; and
(ii) Reporting on the performance of the quality system to management with executive responsibility for review.”
The definition of MR in the NPRM was worded very similar to the QSRs. The problem with this is that this requirement is akin to the addition of an extra layer of fat – it just isn’t necessary under the GLP setting. In GLP, one of the responsibilities of the Quality Assurance Unit (QAU) is to “assure management that the facilities, equipment, personnel, methods, practices, records, and controls are in conformance with the regulations”. What is the value of the addition of a MR if the MR’s mandate is already under the ambit of the QAU’s role already? If the FDA is merely concerned about proper hierarchy in the organization – that someone senior should report to the top management as oppose to a mere quality assurance staff member then facilities often have Quality Assurance Managers or Directors who can relay the report to top management. In fact this is what happens in a lot of cases, particularly in larger companies. In addition, this requirement for another layer of management level (i.e. the MR) seems to ignore the fact that there are small organizations that would incur increased cost and undue burden with this proposed requirement.
Next, the proposed addition of a GLP Quality System definition in the NPRM seems to imply that the existing GLP regulations do not have a quality system in itself (or a mechanism for its management). In addition, the proposed definition seems to overemphasize harmonization with definitions in the QSR but lacks important components of the GLPs such as planning, performing, monitoring, recording, archiving, and reporting. The OECD Principles of GLP, on the other hand, offers the following definition:
“Good Laboratory Practice (GLP) is a quality system concerned with the organisational process and the conditions under which non-clinical health and environmental safety studies are planned, performed, monitored, recorded, archived and reported.”
The OECD’s definition offers a better description of the GLPs as a quality system as this is more reflective of how regulated industry professionals experienced in the conduct of nonclinical studies understand and implement the GLPs.
While there are some points where the NPRM offers appealing and sensible proposals (e.g. a grace period of for archival after the completion of studies and allowing for process-based inspections for procedures that are repetitive in nature wherein study-based inspections are impractical to undertake), in general the NPRM has shown to be quite disappointing. Instead of tweaking the current GLPs as needed and instead of the modernization that regulated industry professionals have hoped the FDA would deliver, the FDA seems to be merely taking the opportunity to radically revise the GLPs just to meet the GMPs (and ISO). Different worlds, different purposes. By making the GLPs fit the requirements of the cGMPs and ISO, there will be a one size fits all type of standard instead of separate specific rules for specific needs. It won’t be too far-fetched to expect more GMP inspectors inspecting GLP facilities expecting its operations to be conducted like the cGMPs. This definitely would not make GLP great again.