GLPs and the Importance of Standard Operating Procedures

Good laboratory practice (GLP) is a standard by which laboratory studies are designed, implemented, and reported to assure the public that the results are correct and the experiment can be reproduced exactly, at any time in the future. In less technical terms, GLP is the cornerstone of all laboratory-based activities in any organization that prides itself on the quality of the work it performs. And, despite its immediate association with the pharmaceutical sector (Figure 1), GLPs can (and should) be applied to virtually all industries in which laboratory work is performed, including companies involved in drug manufacturing, food and drink production, and engineering. In addition, commercial testing laboratories (for toxicology, metabolism, materials and safety, for example), research establishments, and universities - in fact, all laboratories engaged in product or safety testing or research and development - should adopt and apply the doctrines of GLP.

GLP is not a luxury. It is a necessity for any professional laboratory wishing to gain and retain the respect of its employees, clients, regulators and perhaps most importantly, its competitors. If a company is seen to be applying and adhering to the highest standards of laboratory practice, it will gain significant competitive advantage and will compete successfully for business and recognition within its operational environment. Conversely, without rigidly enforced GLPs - and good manufacturing practices (GMPs) - a scientific organization will not achieve the commercial success and respect that its products and personnel deserve.

A Brief History of GLP

Up until the early 1970s, many private and public laboratories applied GLP-type principles in one form or another. Repeat-dose safety studies involving a variety of techniques concerned with animal handling, dosing, and observation were regularly conducted. In 1972, New Zealand formally introduced GLP as the Testing Laboratory Registration Act, which covered staff records, procedures, equipment, and facilities. The act prompted the establishment of a related council "to promote the development and maintenance of GLP in testing." In the same year, Denmark also introduced a law to promote GLP.

Figure 1. The ubiquitous nature of GLPs in the pharmaceutical industry (adapted from Edwards and Murji, 2001, see reading list).

Significant events took place in 1975 when Senator Edward Kennedy (Dem-MA) and members of FDA made allegations against research laboratories in the USA (Searle and Hazelton) related to preclinical research studies. Both sites were subsequently investigated, which revealed serious problems with the conduct of safety studies submitted to the agency. Violations included poor record keeping and data storage, inadequate personnel training, poor test facility management, and even fraud.

By January 1986, scientists at Searle had developed a document, Good Laboratory Practice, which was designed to be used as guidance to evaluate research activities, and submitted it to both FDA and the Pharmaceutical Research and Manufacturers Association of America (PhRMA). In August of the same year, FDA released a draft GLP document based on the Searle paper and published GLP regulations in the Federal Register. At the same time, the agency was creating 606 new positions to monitor biological research, and it also began a pilot inspection program to establish baseline levels of competence. Items listed as "major failings" included failure to have a quality assurance (QA) department, failure to test every batch of manufactured product, and failure to maintain standard operating procedures (SOPs).

Our Laboratories, Inc.

In December 1978, FDA published final GLP regulations and made compliance with them the law in the United States in June 1979. These regulations were collected in Title 21: "Food and Drugs" of the Code of Federal Regulations (CFR) as Part 58: "Good Laboratory Practice for Nonclinical Laboratory Studies," and they applied to all nonclinical safety studies intended to support research permits or marketing authorizations of products regulated by FDA. Subsequently, FDA's Office of Regulatory Affairs (ORA) released two Guidance for Industry documents to ensure the proper and consistent interpretation of the directives by industry and by FDA's field investigators. Further changes to the GLP rules were proposed in 1984, and in September 1987, FDA published its "Final Rule" - Compliance Program Bioresearch Monitoring: Good Laboratory Practices, which was expanded to incorporate the following:

• the requirement for a QA department

• the requirement for protocol preparation (study plan)

• the characterization of test and control materials

• the requirement to retain specimens and samples.

Since then, the requirement for laboratories to apply and comply with GLP principles has extended from pharmaceutical companies to many other types of research and testing establishments throughout the developed world. In Europe, adherence to the principles of GLP is governed by European Union (EU) law and, in compliance with EU Directives, an inspection program confirms that "toxicological studies for the regulatory assessment of industrial chemicals, medicines, veterinary medicines, food and animal feed additives, cosmetics, and pesticides must be conducted in accordance with GLP."

Within 14 years, therefore, GLP moved from an ad hoc concept to legally enforceable code, designed to control and regulate the quality of laboratory-based operations. Since that time, the principles have remained fairly consistent and have not changed much; they have simply expanded to include other scientific activities and devices (such as computers).

The Standard Operating Procedure

SOPs are documents that describe how to perform various routine operations in a GMP-based manufacturing or research facility. They provide a general framework enabling the efficient implementation and performance of all the functions and activities, and contain step-by-step instructions that technicians and production personnel consult on a daily basis to complete their tasks in a reliable and consistent manner. SOPs are, in essence, written commitments to the regulatory bodies that describe the performance of routine tasks; they are required to ensure the successful conduct of a study but can also serve as a valuable training tool.

General factors that must be covered by an SOP are as follows:

• the receipt, handling, and storage of test and control substances

• test systems and methods

• the use, care, cleaning, and calibration of equipment

• facility use (such as cleanrooms)

• documentation and preparation of reports, data collection, handling, storage, and retrieval

• the completion of documentation, such as pro forma result and data collection forms

• any other routine procedure.

SOPs must be comprehensive, covering all the necessary details to enable staff to complete the procedure. Information that should be listed includes the preferred suppliers of chemicals, reagents, and equipment; the catalogue numbers of reagents and the model numbers of equipment; the storage conditions and stability of chemicals and test substances; and acceptance criteria for valid procedures. In some instances, it may also be necessary to specify such factors as centrifugation speeds listed as X g rather than rpm (unless the rotor radius is stated), incubation conditions with tolerances, volumes with tolerances (if relevant), and any other information that enables the accurate reproducibility of procedures.

SOPs should be written by personnel who are familiar with the procedure and then approved by the management team. They should be written in language that will be understood by the person using the SOP and checked for accuracy and quality. An SOP should be signed and dated by authorized personnel and, ideally, include the printed name of the signatory. All SOPs should be reviewed regularly, and the review date must appear on the documentation. They should be accessible to all and refer, when relevant, to product quality specifications. Above all, an SOP must be clearly presented and followed.

SOP format. There are many ways to format an SOP. When developing a format, consider the essential and informational categories such as the title and purpose (why it is written), the scope (where does it apply), the responsible party (who must apply it), references and other documents, safety considerations (such as the use of protective clothing or disposal procedures), the principles and preliminary operations, procedures that must be followed (including calculations), and documentation requirements (see Figure 2).

Within the scope of GLP, several factors may not specifically be referred to in an SOP - those that should be addressed and considered under the general dogma of regulatory compliance. Examples of these items include equipment, experimental procedures, routine procedures, facilities, and laboratory reagents and test substances, which are described in further detail below.

Equipment. Types of equipment and apparatus that might be included are pipettes; balances; refrigerators and freezers; HVAC systems and laminar flow hoods; computers; analytical machinery (ELISA, spectrophotometers, densitometers, and image analyzers); high performance liquid chromatography (HPLC), gas chromatography (GC), and atomic absorption equipment; fermentors; and centrifuges. The equipment should be for the intended purpose; kept in good, clean, working order; and used by trained personnel only - that is, according to SOPs describing use, care, calibration, and cleaning routines. Regular calibration and servicing is essential; full records of maintenance and equipment checks must be kept in an easily retrievable form. If maintenance is done by external contractors, such work is acceptable without an SOP. If equipment is serviced and calibrated by internal staff, however, SOPs must be in place to cover the work.

Experimental procedures must be fully documented and contain all pertinent details. For example, lot numbers, catalog numbers, supplier details, and dates for when chemicals and reagents arrive and when they are opened must be listed, as well as temperatures and times, specific identification of equipment used and, perhaps most importantly, any deviations from the SOP. All this information must be recorded in lab notebooks with individual page numbers. Experimental details and results should be easily located; a log page at the front of a notebook can help track the recordings and observations. Any reference to computer files containing data should also be catalogued in the notebook. Data files should always be backed-up in case of computer failure, corruption, or deletion.

Routine procedures. Daily or scheduled tasks must be fully described in written SOPs. Just as for other SOPs, SOPs on routine procedures must not only describe technical details but also a system for reporting results and the methods for cleaning and calibrating any equipment. Even routine procedures must be validated as appropriate for their intended use. Assays, for example, must be accurate in the conditions in which they will be used (for instance, cesium affects the detection of hepatitis B). The validation of an assay should include factors such as

• specificity and selectivity - evidence that cross-reactivity by factors such as antibodies is detectable

• accuracy - as shown by comparison with other assays

• precision - with limited variability

• recovery (spike recovery) - showing the assay detects known amounts of added analyte, for instance.

Validation protocols and reports must be written and available for every procedure, and the most important criterion for any set of instructions is that the SOP is followed.

Facilities must be fit and suitable for the purpose of the work; that is, size, construction, and location should be appropriate, and the building should allow for the separation of activities. Animal handling facilities, in particular, should be large enough, allow for the isolation of incoming or diseased animals, allow for the separation of both species and studies, have storage space for feed and bedding, and allow for the adequate disposal of waste and refuse to minimize vermin, odor, disease, infestation, and environmental contamination.

The Pros and Cons of GLP Compliance

SOPs should ensure that facilities are designed to permit separate areas for similar but unrelated work and sample preparation. In addition, buildings must be validated for the required functionality (air handling and sterilization, for example) and monitored (for air pressure differences, flow patterns, or sterilization), with easily accessible results. The property must be regularly maintained and serviced, with full maintenance and modification records kept. All facilities, laboratory areas, benches, floors, corridors, and cupboards need to be neat, both during and after experimentation - which really only amounts to good housekeeping. Finally, the facility must be of sufficient size to accommodate data storage and archiving.

Reagents and test substances. Several GLP rules apply to all laboratory reagents and test substances. For instance, preparation of these ingredients should use a standardized and fully documented methodology that lists supplier information, lot numbers of component chemicals and reagents, dates of preparation, and the names of the staff involved in the preparation. Substances should be tested against reagents and standards of known reactivity before being released for use, and the results of these tests must be recorded, preferably using a Reagent Preparation Record. All prepared solutions should have unique lot numbers (a laboratory-based system is acceptable as long as the provided numbers are actually unique). This information log prevents duplication and should comprise lot numbers, dates, descriptions, and signatures. Containers are often overlooked. For both test and control substances, the container should be inert to the stability of the substance and clearly labeled with the following details:

• contents or identity

• potency (titer, concentration, and activity, for example)

• storage temperature

• preparation date

• unique lot number

• container number (in case there are multiple containers of the same lot)

• date first opened

• expiration date (determined by experimentation or reference to manufacturer's recommendation)

• signature of the person who prepared the contents.

Much of this information can be recorded separately with a unique identification number log. Test substances should also be retained for reference until a report is written, if tests last longer than four weeks.

Documentation. Although not actually referred to in many GLP guidelines, documentation quality is crucial for laboratory and manufacturing compliance. Some general documentation requirements are included in the UK Medicines Control Agency's Orange Book, ensuring all pro forma paperwork, SOPs, results reports, product quality specification (PQS) sheets, and manufacturing instructions be

• typed, and clearly and unambiguously written

• written legibly with indelible ink

• corrected, if necessary, with a single cross-out line, then signed and dated with a reason for the alteration, if relevant

• officially authorized and signed

• dated and also dated when reviewed

• reviewed on time and retrieved when out of date

• designed for ease of use and completion by appropriate staff

• coded with document versions to ensure use of correct editions of similar forms

• carefully controlled to ensure only correct documents are circulated.

Computer-generated documents should be closely controlled with password protection, and records should be kept of previous versions.

Recording experimental details. Experimental details must include lot numbers (internal and supplier's) of reagents used and the incubation temperatures, times, and conditions. These records should be stored in an unalterable, easily retrievable format, using laboratory notebooks that contain logs or indices and consecutively numbered pages. Cross-references to other data (charts, printouts, and computer files, for example) should be straightforward and transparent. Above all, these records must be accurate and up-to-date.

Retaining data. All results must be recorded and retained. Raw data (such as chromatograms and printouts) and processed results must be easily retrievable. Back-ups of computer-generated information should be maintained.

Reporting results. All reports, whether final study reports or test request forms, must be checked for accuracy, signed, dated, and countersigned by authorized personnel. Ideally, the names of signatories should be printed because handwritten signatures can be difficult to identify. Reported results should be complete (containing all valid results) with a foolproof presentation - to avoid any ambiguity - according to the relevant SOP. Full study reports should also contain a QA statement that confirms work was performed according to GLPs, includes audit dates, and lists aspects of the study that did not conform to GLPs, if any.

Web Resources for Writing SOPs

Archiving and retrieving. An effective archiving system must keep all related data (experimental, calibration details, slides, specimens, and reference material) together, under suitable storage conditions. Specimens and materials should be stored only as long as they are stable. The data should be indexed - with cross-references to any data or information stored at a different location - so they are easy to locate and retrieve. Data require official authorization before disposal.

SOPs in Review

The compliance of all laboratories - and analytical ones in particular - is defined by their SOPs. FDA laboratory inspections focus on the adequacy of SOPs and on how those SOPs are followed. Therefore, properly written SOPs are critical. Poorly written SOPs - or well-written SOPs that are not properly followed - are a major source of preapproval inspection 483 observations by FDA. Yet SOPs should not necessarily be written for FDA; they should be written for those technicians who will use them. Although compiling SOPs can be somewhat daunting to the beginner, they are easily mastered and, with practice, can become almost second nature.

As described, two paramount rules must be followed. An SOP should be detailed enough to adequately define the task that it purports to describe, and an SOP should be general enough to prevent user confusion or limit his or her ability to work efficiently or impede decision-making processes. It is particularly important not to get restricted by semantics. An SOP for calibrating the temperature of a water bath could, for example, state that calibration frequency should be daily. This sounds harmless enough, but it would necessitate calibrating the water bath every day, regardless of whether the instrument was used or not. Most laboratory water baths are used every day, but if the SOP states the frequency as "daily when in use," the bath would only need to be calibrated on days when it was actually used. The difference in the text is minor, but it could save a lot of unnecessary work (and time).

Language clarity and accuracy is another crucial factor. Decision-making processes or scientific judgment can be limited by the syntax used in an SOP. If a protocol required, for example, a 0.1-M solution of trisodium citrate, the SOP might specify that 29.4 g of trisodium citrate dihydrate (MW5294.1) should be weighed into a tared 1-L volumetric flask, dissolved in ultrapure water, and diluted to volume, even if only 300 mL of the solution is required for the experiment. If, on a given occasion, there was only enough trisodium citrate to make 0.5 L of the reagent (14.7 g in 500 mL of water), this could, according to FDA, constitute a breach of the SOP.

Suggested Reading

Common sense would suggest that the reagent had been prepared just as accurately, producing enough to complete the experiment, but the SOP had not, technically, been followed. A better option would be to write "prepare a sufficient quantity of the reagent at a concentration of 0.1 M," which allows the user some discretion concerning the choice of sample weight and final volume. Maintaining flexibility is an important factor in making an SOP useful.

Carol DeSain, in her book Documentation Basics That Support Good Manufacturing Practices and Quality System Regulations, suggests using active words for procedural directives: "Add this," "Pour that," "Observe color." Always name the equipment or instrumentation used, along with its part numbers. All SOPs should include on the first page the name of the author or authors and the signatures of reviewers and approvers, along with title, date, number of pages, and revision number. Each subsequent page should contain the title, page number, and revision number.

SOP length. Some people in the industry claim SOPs should never exceed three pages. For others, length is simply a function of the procedure being described and cannot be restricted to a certain number of pages or words. Shorter SOPs tend to be adhered to more closely, especially in the pharmaceutical industry, where a given company can have hundreds of documented protocols on record. The matter of the level of detail has been discussed above. Concise instructions, however, are useful in training, are not necessarily restricted to specific applications, and often contribute to the smooth running of a laboratory. Whatever the length of the SOP, each page must contain the (abbreviated) title, the SOP number and edition, and a pagination reference (1 of 10, for example). There is no specific FDA-approved system for layout and style.

For SOP review and approval, every laboratory should have designated personnel who are authorized to have a copy of a manual containing the current version of documents, including analytical methods or SOPs. Only current versions of each document should be in use. Previous revisions of documents should be archived for reference purposes and for historical tracking of changes with time. Once a new document is approved, it should be issued by a central "study director" and then included in the authorized manual. The previous version should be returned to the original author, and a receipt for its return obtained. This ensures that only current documents are used, and out-of-date versions are removed from common access. QA is responsible for auditing laboratories for compliance and for informing the study director of any shortcomings. Everyone working in a laboratory is responsible for applying GLP principles and informing the study director (or the head of the department, the direct line manager, or the supervisor) if GLP is not being adhered to. Management is responsible for providing conditions that facilitate and enhance GLP adherence.

Consequences of Noncompliance

Irrespective of what an SOP is for, how it is written, or who uses it, the cardinal rule is that the instructions are followed. It is much better to get an SOP right the first time than to have to go through several iterations or rounds of approval. FDA is less interested in the layout and style of the SOP but in whether it is being followed and a good level of compliance is being maintained. Noncompliance with GLPs is extremely serious and compromises the validity of results obtained in a study and, in extreme circumstances, could even render a study void. The consequences of noncompliance include

• rejection of study findings by regulatory authorities

• requirement that a study be repeated (with attendant cost implications)

• inability to complete the study

• failure to achieve product registration and license

• loss of client business and company credibility (with severe business consequences)

• withdrawal of product license.

All of these consequences are avoidable. Compliance with GLP is imperative, and the cornerstone of that compliance is a written, approved, and followed SOP. Perhaps as a final thought, a documentation writer should heed a quotation attributed to English poet, critic, and lexicographer Samuel Johnson: "Language is the only instrument of science, and words are but the signs of ideas: I wish, however, that the instrument might be less apt to decay, and that signs might be more permanent, like the things which they denote."BPI