School of Economics and Management University of Växjö
How does Good Laboratory Practice improve quality?
Bachelor’s degree thesis in Business Administration Management Control, FEN 330, Spring 2006 Authors: Malin Jansson & Mirja Wynn-Williams Supervisor: Lecturer Stig Malm Examiner: Professor Rolf G Larsson
Abstract Bachelor’s degree thesis in Business Administration School of Economics and Management, University of Växjö, FEN 330, Spring 2006 Authors: Malin Jansson and Mirja Wynn-Williams Supervisor: Stig Malm
How does Good Laboratory Practice improve quality? Background: The quality systems that steer manufacturing of pharmaceutical products from the testing phase to commercial manufacturing are the national and international regulatory frameworks and legislation. Good Laboratory Practice (GLP) is a quality system concerned with the organizational process and the conditions under which non-clinical health and environmental safety studies are planned, performed, monitored, recorded, archived and reported. In Sweden the Medical Products Agency monitors and regulates compliance with the principles of Good Laboratory Practice. Aim: The aim of our thesis is to explore the reasons why companies/laboratories adopt Good Laboratory Practice. We shall do this by identifying the advantages and disadvantages of adopting GLP principles for companies/laboratories, and how quality is improved by adopting GLP principles. We have summarized our aim in one principal question: How does Good Laboratory Practice function as a tool for quality improvement? Limitations: We will focus on GLP, and this thesis will not consider other Good Practice procedures such as Good Manufacturing Practice or Good Clinical Practice. Only laboratories in Sweden monitored by the Medical Products Agency are targeted in the empirical part of this work. Method: Literature research and interviews with GLP contact persons listed by MPA. Results and conclusions: Though costumers’ needs and legislation seem to be the motivations for companies deciding to comply with GLP, quality improvement seems to be the biggest advantage of validation. Proposal for further research: Do laboratory assistants and technical employees differ from management in their experience and views on GLP?
ABBREVIATIONS CRO
Contract Research Organization
FDA
Food and Drug Administration
GCP
Good Clinical Practice
GLP
Good Laboratory Practice
GMP
Good Manufacturing Practice
GXP
GCP, GLP and/or GMP
MPA
Medical Products Agency
OECD
Organization for Economic Co-operation and Development
PDCA
Plan Do Check Act
QA
Quality Assurance
SOP
Standard Operating Procedures
CONTENTS 1
BACKGROUND AND INTRODUCTION TO GLP ......................................................... 1 1.1 1.2 1.3 1.4 1.5 1.5.1 1.6 1.7
2
THE PHARMACEUTICAL INDUSTRY.................................................................................. 1 GOOD LABORATORY PRACTICE ...................................................................................... 2 NEED FOR RESEARCH ...................................................................................................... 3 WHY DID WE CHOOSE THIS SUBJECT? ............................................................................. 3 AIM ................................................................................................................................ 4 Research question ..................................................................................................... 4 HYPOTHESIS UNDERLYING OUR RESEARCH ..................................................................... 5 RESEARCH DESIGN.......................................................................................................... 5
METHODOLOGIES AND METHODS ............................................................................. 7 2.1 RESEARCH METHODS ...................................................................................................... 7 2.1.1 Qualitative research methodology ............................................................................ 7 2.1.2 Quantitative research methodology .......................................................................... 7 2.1.3 Case study ................................................................................................................. 7 2.1.4 Choosing the right research methods ....................................................................... 8 2.2 HERMENEUTIC RESEARCH PARADIGM ............................................................................. 8 2.3 DESCRIPTIVE, EXPLANATORY, EXPLORATORY, DIAGNOSTIC AND PROBLEM SOLVING RESEARCH ..................................................................................................................................... 9 2.4 DATA COLLECTION METHOD......................................................................................... 10 2.4.1 Literature studies .................................................................................................... 10 2.4.2 Interviews................................................................................................................ 11 2.4.3 Study plan ............................................................................................................... 11 2.4.4 Cumulativeness ....................................................................................................... 12 2.4.5 Representativeness .................................................................................................. 12 2.5 TRUSTWORTHINESS, AUTHENTICITY AND INTEGRITY IN PRODUCING DATA .................. 13 2.5.1 Qualitative research data........................................................................................ 13 2.6 VALIDITY AND RELIABILITY ......................................................................................... 13 2.6.1 Validity.................................................................................................................... 14 2.6.2 Reliability................................................................................................................ 15 2.7 ANALYSIS AND INTERPRETATION STRATEGIES.............................................................. 15 2.7.1 Cluster analysis....................................................................................................... 16 2.7.2 Typologies............................................................................................................... 16
3
THEORY ............................................................................................................................. 18 3.1 HISTORY OF GOOD LABORATORY PRACTICE ................................................................ 18 3.2 GOOD LABORATORY PRACTICE .................................................................................... 19 3.2.1 Scope of GLP principles ......................................................................................... 20 3.2.2 Test facility organization and personnel................................................................. 20 3.2.3 Quality assurance programme................................................................................ 21 3.2.4 Standard Operating Procedures (SOPs)................................................................. 22 3.2.5 Facilities ................................................................................................................. 23 3.2.6 Equipment, material and reagents .......................................................................... 25 3.2.7 Test systems............................................................................................................. 25 3.2.8 Test and reference items ......................................................................................... 26 3.2.9 Performance of the study ........................................................................................ 27 3.2.10 Reporting of study results................................................................................... 29 3.2.11 Storage and Keeping of Records and Materials................................................. 31 3.3 QUALITY SYSTEMS ....................................................................................................... 33 3.3.1 Quality systems and the pharmaceutical industry................................................... 34 3.3.2 Quality systems as a managerial tool ..................................................................... 35 3.4 GLP QUALITY SYSTEM ................................................................................................. 35
4
RESULTS............................................................................................................................. 37 4.1 4.2
COMPANIES INTERVIEWED............................................................................................ 37 REPLIES FROM RESPONDENTS ....................................................................................... 38
4.2.1 4.2.2 4.2.3 4.2.4 4.2.5 4.2.6 4.2.7 4.2.8 4.2.9 4.2.10 5
ANALYSIS AND CONCLUSIONS................................................................................... 50 5.1 5.2 5.2.1 5.2.2 5.2.3 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10
6
DRIVERS FOR SEEKING GLP STATUS ............................................................................ 50 QUALITY ...................................................................................................................... 51 Two paradigms........................................................................................................ 51 How does GLP improve quality?”.......................................................................... 51 PDCA cycle and continuous improvement.............................................................. 52 VALIDATION PROCESS .................................................................................................. 53 TO COMPLY OR NOT TO COMPLY ................................................................................... 54 ARE OUR RESULTS REPRESENTATIVE? .......................................................................... 55 SANCTIONS ................................................................................................................... 55 NO SUCH THING AS A PERFECT SYSTEM ........................................................................ 56 ANALOGY OF THE GLP CONCEPT ................................................................................. 57 FUTURE OF GLP ........................................................................................................... 58 INDIRECT BENEFITS OF GLP ......................................................................................... 60
REFERENCES .................................................................................................................... 61 6.1 6.2 6.2.1 6.3 6.4 6.5
7
Interview questions ................................................................................................. 38 Adoption of GLP principles .................................................................................... 39 Why did laboratories seek validation?.................................................................... 41 Implementation and changes needed ...................................................................... 42 Quality systems ....................................................................................................... 42 What advantages did the laboratories gain from GLP? ......................................... 43 Has GLP improved quality, and if so, how? ........................................................... 44 Quality and continuous improvement ..................................................................... 45 Have laboratories experienced any disadvantages from GLP registration? .......... 46 What are the sanctions for non-compliance? ..................................................... 48
BOOKS .......................................................................................................................... 61 DOWNLOADABLE FROM AUTHORITIES’ AND ORGANIZATIONS’ WEBSITES .................... 61 Läkemedelsverkets författningssamling .................................................................. 61 MANUALS FROM ORGANIZATIONS ................................................................................ 62 ARTICLES ..................................................................................................................... 62 NEWSPAPERS ................................................................................................................ 62
APPENDIX 1 ....................................................................................................................... 63
INTERVIEW QUESTIONS ........................................................................................................ 63 ADOPTION OF GLP PRINCIPLES ................................................................................................... 63 HOW THE GLP FUNCTION TODAY? ............................................................................................. 63 OVERALL QUALITY IMPROVEMENT ............................................................................................. 63 SANCTIONS ................................................................................................................................. 63
1 BACKGROUND AND INTRODUCTION TO GLP 1.1
The pharmaceutical industry
Years of research and testing take place before a new medicine reaches the market, and hundreds of thousands of dollars of costs are incurred before it can be approved and be licensed. In 2003 over 21 billion euros were spent in the European Union and 29 billion euros in the USA on pharmaceutical research (Lääketietokeskus, 2005). By 1938 companies in the USA already had to submit data and convince the FDA that their products were safe before they would be approved as ready for sale. In Sweden the regulatory authority is Läkemedelsverket (or Medical Products Agency in English). In the European Union a licence is required to sell a new drug and it can be applied from either the European Medicines Agency (EMEA) or in Sweden from MPA (GLP Handbook). In February 2006, one pharmaceutical company withdrew its application to market a new drug (EMEA, press releases) because it caused liver damage as a side effect. The company not only withdrew the product, it also announced that it had stopped all further research concerning this medicine, presumably at a large financial loss to the company. This company’s non-clinical laboratory has GLP status. Were the principles of GLP properly adopted? In the 1960s, Thalidomide was a medicine that had been approved by many national medical agencies (it was marketed under various names in different countries, including Neurosedyne in Sweden). It had been tested on mice. It was recommended for morning sickness, but tragically when taken by pregnant women, it resulted in the birth of thousands of deformed babies (Immel, 2000). In the 1960s GLP principles had not yet been developed. Could this disaster been prevented by GLP principles? Regulatory frameworks (covers stages 2-4 in Figure 1) aim to avoid such disasters, as do GLP principles (covers stage 2 in Figure 1) that raise the quality of pharmaceutical procedures. -1-
Stage 1 DISCOVERY ↓ Stage 2 NON-CLINICAL ↓ Stage 3 CLINICAL ↓ Stage 4 POST-APPROVAL ↓ MANUFACTURING Figure 1: Drug development process
The estimated time from stage 1 to manufacturing is approximately 10 years (GLP Handbook).
1.2 Good Laboratory Practice There
have
been
some
disasters
with
new
medicines,
such
as
Neurosedyne/Thalidomide (Immel, 2000). The quality systems that steer and regulate manufacturing of pharmaceutical products from the testing phase to commercial manufacture are the national and international regulation frameworks and legislation. They include GXP principles, which are quality and validation systems for the pharmaceutical industry. Good Manufacturing Practice principles provide regulations for production of pharmaceuticals and Good Clinical Practice analogous regulations for clinical studies. Good Laboratory Practice is a quality system concerned with the organizational process and the conditions under which non-clinical health and environmental safety studies are planned, performed, monitored, recorded, archived and reported. The aims of these Principles of Good Laboratory Practice are to promote public health and environmental safety and to minimize technical trade barriers. Each country has a regulatory authority responsible for monitoring and inspecting its companies’ compliance with these principles.
-2-
1.3 Need for research We have not found any research in Sweden or Finland that focused on principles of Good Laboratory Practice. However, there is a lot of research in related subjects such as safety and information technology in reporting clinical laboratory results and other testing in clinical laboratories, and also research on quality in health care services in general.
1.4
Why did we choose this subject?
Quality has become increasingly important to companies as stakeholder groups, interest groups and members of the public require that their concerns about certain aspects of companies’ operations are taken into account by managers of the companies. As highlighted above, unforeseen consequences raise people’s and authorities’ awareness. GLP principles set out standards for conducting nonclinical research. GLP principles are stipulated in law, but are not yet well known. There is a lot of scope for improvement in education as regards GLP principles. Many students that graduate with degrees that qualify them to work in the pharmaceutical industry and on research have not even heard of GLP. Literature in this field is very limited, often very expensive and not easy to access. We would like promote knowledge of GLP through our thesis. The authors have very different backgrounds. Malin Jansson is a Master of Science in Engineering Biology from the Technical Institute in Linköping, Sweden. She is currently working on microbiological non-pharmaceutical research at the Institute of Molecular Botany at the Ruhr University in Bochum, Germany. She is knowledgeable in research laboratory terminology, laboratory research methods and standard operating procedures. Mirja Wynn-Williams worked on internal control, risk management and quality improvement in a Finnish bank in Sweden in the 1990s. This post was created in the aftermath of the property crash, when banks in Sweden lost billions of Swedish krona through lax lending, lack of proper risk management, poor
-3-
management control and lack of quality improvement work. In those days company reports did not include sections called “Corporate Governance” or “Corporate Responsibility”. We are both especially interested in GLP principles, though for different reasons. We approach the complex of problems in this research from different sides. Malin would like improve her working methods and apply the relevant principles of GLP in her research work. Mirja is interested in quality improvement. Her current specialist sector, business advisory services, lacks any national standards, and quality varies from poor to excellent.
1.5
Aim
The aim of our thesis is to explore the reasons why companies/laboratories adopt Good Laboratory Practice. This will be done by identifying the advantages and disadvantages of adopting principles GLP for companies/laboratories, and how quality is improved by adopting GLP principles.
1.5.1 Research question When we were formulating the question we set out to answer, “How does Good Laboratory Practice function as tool for quality improvement?” we thought that there are two equally important aspects to this question. The first part of our question is “How does Good Laboratory Practice function?” GLP is a quality system that applies to non-clinical research and testing of a medicine. Why is quality so important in non-clinical pharmaceutical research and industry? The last part of our question “a tool for quality improvement” highlights the importance of one aspect of quality work: it has to be managed. Quality improvement is a continuous process and it has to be monitored. Quality improvement helps a company to implement its strategy successfully.
-4-
In adopting any quality system, it is important that all personnel are committed to quality improvement. According to Deming’s principle, top management has a key role in promoting quality. Sustainable quality improvement can be achieved only if the top management are wholeheartedly involved in quality improvement issues (Bergman et al., 1998).
1.6
Hypothesis underlying our research
Why do companies work according to GLP? Is the only reason the demands from authorities (legislation), and is the major benefit that it makes it possible to sell and market pharmaceuticals, or does the quality aspect of the system play an important part? The hypothesis underlying our research is depicted in Figure 2: Other drivers
Patent issues Quality
Legislation
GLP validation
Marketing/selling of medicines
Alternative validations
Safety Other benefits
Figure 2. We assume the main driver for companies to seek GLP validation is that it is stipulated by law. Companies are obliged to comply with the legal and regulatory requirements set for pharmaceutical industries.
1.7 Research design We studied the regulatory framework governing GLP Principles. We needed to study books and publications about GLP, general theories and research methodology theories to be able to conduct the research. We studied MPA’s web pages and its electronic publications. We collected data and information on the laboratories and companies from reports on operations and information on companies’ web pages.
-5-
Before carrying out our interviews, we selected the laboratories and contacted them to ask whether they were willing to participate in our research. We then prepared interview questions to send prior to the interviews so that respondents had time to prepare their answers. We had sets of open-ended and closed-end questions.
-6-
2
METHODOLOGIES AND METHODS
2.1
Research methods
There are two primary ways of carrying out research in social sciences: quantitative and qualitative research methods (Andersen, 1998:31,). A third way is a case study. This is better for showing how and why a company or an organization operates as it does (Yin, 2003).
2.1.1 Qualitative research methodology Qualitative research is carried out to understand meaning, or to describe and understand a specific subject, experience, idea or problem. The data collected for this research includes very little statistical or mathematical data or formulas. The purpose of this research method is to promote understanding and not to explain the causality of a phenomenon. Ideally, the data for qualitative research are gathered through interviews and observations (Andersen, 1998).
2.1.2 Quantitative research methodology Quantitative research methods are used in natural science research. Data collected to underpin this type of research can be measured or verified through numerical data. This type of research is based on statistics and mathematical formulas. The main purpose of quantitative research is to verify the cause of an existing phenomenon or to prove a hypothesis concerning the subject being studied. Surveys and questionnaires are suitable methods of collecting this type of data (Andersen, 1998:31).
2.1.3 Case study According to Yin (2003), the case study approach is the preferred strategy when the questions “how” and “why” should be answered. Other circumstances that favour a case study approach are when the investigator has little control over the events to be studied and if the study is focusing on events happening today.
-7-
2.1.4 Choosing the right research methods According to Andersen (1998:33), the key factors in choosing the right research methods are: •
Research subject
•
Approach to the subject
•
Knowledge sought through research We have chosen to use qualitative research methods, which means seeking through our research wider knowledge and understanding of our subject, which is, “How does GLP improve quality?” We have assumed in our research that laboratories adopt GLP principles mainly because of legal requirements. We chose qualitative research because the interviewees will enlighten us with their experience of how they have carried out safety studies in compliance with GLP principles in their laboratories, and thereby increase our understanding of this subject. The case study approach could also have been chosen. This would be a contemporary study and we do not have any influence on GLP practice. The questions we would like to answer in the thesis are of the “how” and “why” type. However, the case study approach, although theoretically useful, would not be possible in practice. Professional secrecy is one of the main obstacles that would make it impossible to get in-depth information about the GLP systems of the companies surveyed. One other problem is the fact that this kind of approach would require more time from the laboratories’ employees. Moreover, a good case study would require personal meetings and visits to sites for observation, which would also be impossible.
2.2 Hermeneutic research paradigm The hermeneutic research paradigm aims at increasing understanding of the subject being researched. It does not focus on validity, but on trustworthiness and authenticity. The hermeneutic research method focuses on interpreting and understanding (Rosengren et al, 2002:323). It claims that the best understanding of a person’s actions comes from trying to put oneself into that person’s shoes and trying to understand his or her actions from the inside. The hermeneutic approach
-8-
often uses qualitative data. This approach accepts the fact that the researcher’s personal experience affects the study and analysis. Absolute objectivity is not required in the hermeneutic approach (Andersen, 1998). Our research approach is hermeneutic in character. We gather people’s experiences of GLP through the interviews that we undertake to gain a wider understanding of how GLP functions.
2.3 Descriptive, explanatory, exploratory, diagnostic and problem solving research According to Rosengren and Arvidsson (2002), the purpose of results sought from research is the starting point for defining the research task. There are three common research methods: descriptive, explanatory and exploratory research. Descriptive research is often used on subjects where a lot of research has already been carried out and the phenomenon under research is already known. According to Rosengren and Andersson (2002), this approach answers questions that start with where, who, how long and how often. Surveys and questionnaires are often used for this type of research. Explanatory research aims to find out why something is the way it is. According to Rosengren and Arvidsson (2002), the question this research approach often answers is why. This research approach often uses an experiment. Exploratory research is ideal for studying subjects that are unknown, that have not been explored before. The purpose with this approach is to create a hypothesis or a theory that can be tested through research (Andersen, 1998). A lot of data needs to be collected through a field study, for instance. According to Rosengren and Arvidsson (2002), this approach answers questions that start with what. Andersen (1998) presents two more research approaches: diagnostic and problem solving. The aim of diagnostic research is to propose which factors produce positive or negative results based on identified symptoms. The problem solving -9-
approach goes one step further and also aims to suggest solutions to the problems or how to eliminate identified, diagnosed problems. Our research approach will be explanatory and promote wider knowledge and understanding. We consider this approach the most useful for our research. The most common question under this approach is why, but our research question is, “How does GLP function as a tool for quality improvement?”
2.4
Data collection method
The data collected can be secondary data, in other words data compiled by other people, researchers or institutions, or primary data that is collected for the thesis in question (Andersen, 1998). Primary data that we shall collect comprise interviews with the persons responsible for quality work in the companies that have GLP status in Sweden.
2.4.1 Literature studies We have listed all types of publications and material in our list of references. These are secondary data for our research. To gain understanding of the problem and knowledge of the subject we have chosen for our thesis, we started by searching for suitable literature. We found the main regulatory framework in the Swedish Medical Product Agency’s website, which also had links to the OECD regulations and EU legislation. We searched the Internet and found out that Karolinska Institutet organizes five credit courses in GLP. We studied the literature for this course and also online publications for it. We contacted the inspectors at the medical products agencies in Finland and Sweden, and they provided us with further information and publications. We have also studied the two main books for the Management Control System course of the School of Economics and Management of Växjö University. Through colleagues in other research laboratories, Malin accessed a GLP Handbook and articles on GLP. We also obtained warning letters published by the FDA in the USA following inspections of GLP laboratories in the USA.
- 10 -
2.4.2 Interviews According to Andersen (1998), there are many characteristics of an interview conducted for research. An interview can be structured so that the questions are presented from beginning to end. Questions can be standard questions depending on a situation. Standard questions in a standard interview situation help to reduce the time needed for the interview. Both open-ended questions and closed questions can be included in a standard set of questions. Closed questions do not serve our research purpose. We would like to find out how good laboratory practice functions for quality improvement and what are the drivers for seeking GLP status. Some of our questions for the interviewees are standard questions. According to Andersen (1998), a semi-structured interview is suitable for when one has knowledge of some theory and empirical data, but will gain new information from the interviewees that will serve the purpose of the research. We interviewed persons responsible for quality work and also some managers of the companies that are validated in Sweden through semi-structured interviews. Keeping our interviews short, avoiding sensitive information and compiling the results in a short time were all factors that increased the motivation for companies to participate in research (Andersen, 1998).
2.4.3 Study plan First, we studied the literature and publications in our reference list. Our presentation of how GLP principles function and what the companies’ compliance with the principles means is based on these studies. Our studies were limited by confidentiality and business secrets. We also had limited time and resources for this thesis. We could not recompense our interviewees for the working time the interview took. We intended to share the findings of our project with the companies, so we hoped that people would be willing to invest the time we required for the interviews. - 11 -
Laboratories were not selected for inclusion in or exclusion from our research. There are only 12 laboratories listed on the Swedish Medical Products Agency’s GLP inspection programme with GLP status. We invited all these laboratories to participate in our study. We considered it unlikely that all the laboratories would participate in our study, so if 10 laboratories participated, that would mean 83% of GLP laboratory operations subject to MPA’s inspection in Sweden were included in our research. As we had to limit the time required for the interviews, we had only a limited number of questions as the basis for the interviews. The questions were emailed in advance. The interview took place by telephone. Time and resources permitting, we hoped to visit and meet some of these laboratories.
2.4.4 Cumulativeness This concept means continuing from where science has already reached. Scientific theory should incorporate existing theory (Rosengren et al, 2002). There is plenty of literature on the reasons why society has an interest in GLP validation and the history of GLP validation, but the amount of theory on the reasons for each company to comply with GLP is negligible. If such information exists, it may be confidential. We have not found any other work concerning the reasons for companies to operate according to GLP.
2.4.5 Representativeness This means that the results are concluded to be representative of the group under investigation. It is important is that the dropout rate is limited and that the sampling is random and appropriate (Rosengren et al., 2002). However, there is no sampling in this thesis. There are only six people listed in the Medical Products Agency as laboratory contact persons (for the 12 companies) for Good Laboratory Practice Inspections. We tried to keep the dropout rate down through deliberately drawing up our method to take as little time as possible for the personnel. We also formulated our questions to exclude any information that we thought would be hard to answer because of confidentiality. The time available was very limited. All of these factors should, according to Andersen (1998), make companies less reluctant to participate.
- 12 -
2.5 Trustworthiness, authenticity and integrity in producing data 2.5.1 Qualitative research data The data produced in qualitative research focuses on trustworthiness and the authenticity and integrity, rather than validity, of data (Rolf Larsson’s Lecture F4. Kvalitativ metod del 3, spring 2006). The data collected through this research method were interpreted, and this required integrity and trustworthiness from the person doing the research. All data collected from interviews or observations must be recorded and written down carefully with every detail. We recorded and wrote notes during our interviews. We were also in written communication with our interviewees through email so we could clarify anything that remained unclear after the interview. We deliberately gave some time for the interviewees to consider the questions before answering. We wanted them to have time to think about their answers thoroughly. This minimized the effect of the interviewee’s personal mood on the day and the interviewer on the answers. Although we are adopting qualitative research methods, we could expect some basic answers from the persons responsible for GLP at the companies would be the same if we asked them the same questions a month later. However, even within a month’s time, new equipment or personnel, or SOPs, or requirements from new management or government regulations could change the situation, and hence their answers.
2.6 Validity and reliability Validity and reliability are commonly referred to in quantitative research where quantitative data can be measured in an objective way (Rolf Larsson’s Lecture F4. Kvalitativ metod del 3, spring 2006).
- 13 -
2.6.1 Validity To achieve a high degree of validity, the conformity of the theoretical conception and the empirical variables has to be high. Validity consists of two concepts: validity and relevance (Andersen, 1998). Validity says something about the conformity of theory and reality. Relevance says something about how relevant the empirical variables chosen are to the problem. There is no objective way of measuring the validity of qualitative data; a position must be justified (Andersen, 1998). As for the overall relevance of this thesis, the theoretical material we present in this thesis is common to all of the GLP validated laboratories that we surveyed. Compliance with GLP requirements at individual level would be very interesting, but we could not discuss that because of confidentiality. Our goal is to find out the reasons and consequence of GLP validation and increase understanding how compliance with GLP principles functions. Individual answers are therefore irrelevant. A common and simple way of defining validity is that a valid measurement measures what it is supposed to measure. Validity is also absence of systematic errors (Rosengren et al, 2002). Validity can also be external validity, which means that the findings of a study can be generalized beyond the immediate case study (Yin, 2003). This generalization will be applicable to the results of this thesis. The system is global and although the respondents of our survey will be national, the results will be applicable to all companies following the GLP guidelines. Systematic errors are avoided through open questions to which the respondents were totally free to answer in any preferred way. Questions were posed concerning the general advantages and disadvantages of GLP, as well as directly about the advantages for quality. This makes it possible for us to analyse all quality aspects.
- 14 -
2.6.2 Reliability The objective is to be sure that a later investigator following the same procedures as described by an earlier investigator would arrive at the same findings and conclusions (Yin, 2003). For comments regarding this thesis, see 2.6.1.
2.7
Analysis and interpretation strategies
We identified from the data from our interviews the reasons why the laboratories/companies sought GLP validation and how compliance with GLP principles has affected them. We analysed the reasons on the basis of the theories we have studied. The analysis of empirical material should reveal the patterns and relations to be found in the data. All analysis is a sort of simplification. If the aim of the project is not purely describing, it is the task of the researcher to arrive at an understandable explanation of the patterns found (Andersen, 1998). However, the analysis should not include all the information gained in the project, otherwise clarity will suffer (Andersen, 1998). Our research approach is explanatory and promotes wider knowledge and understanding, so the thesis should try to explain the empirical facts collected. Analysis of unstructured interviews is much more demanding on the interviewer in terms of time and creativity than numerical analysis of surveys (Andersen, 1998). Analysis of qualitative data requires that concept definition, categorization, operationalization, analysis and interpretation are woven into each other and take place in parallel and to a greater or lesser extent during the whole research process (Andersen, 1998). We conducted semi-structured interviews, which we hoped would enable categorization and operationalization, but not eliminate the opportunity away to gain new information from our interviewees. Interpretation and discussions of alternative explanations should increase the reliability of the conclusions from the empirical material. After interpreting a phenomenon, alternative possible interpretations should be considered to arrive at the most probable one (Andersen, 1998). - 15 -
Our hypothesis is that legislation is the main reason why organizations comply with GLP principles. That makes it possible for them to produce and sell manufactured goods. We define the reasons why companies choose to comply with GLP and determine the effects of the validation on the companies. In doing so, we look out for other variables affecting the companies than those expected. The following factors should influence the analysis: •
The characteristics of the data
•
The kind of knowledge we wish to produce (Andersen, 1998).
The kind of analysis used depends on the aim of the survey, the researcher’s prior knowledge of the area of research and its nature (Andersen, 1998). The less prior knowledge, the more explorative the work must be (Andersen, 1998:186). Electronic data processing requires coding of the data (Andersen, 1998). As our data cannot be coded, there will be no electronic data processing in our thesis.
2.7.1 Cluster analysis Cluster analysis aims to define groups of survey units from empirical data that share similarities and are differentiated from different groups by different values for defined variables. The critical part of this analysis is to define those variables (Andersen, 1998). One important task in analysis and interpretation is to formulate relevant, comprehensive and precise categories from the collected data (Andersen, 1998).
2.7.2 Typologies One important product of the analysis and interpretation process must be the construction of a typology, which requires searching for attributes that appear frequently in the material (Andersen, 1998). The basis is the single survey unit and the goal is to find subgroups of units that are similar. Though the number of companies is quite limited, we categorize them into groups according to how they
- 16 -
have answered certain questions. This helps us make the material more understandable to the reader. According to Andersen (1998), the question “What does the data say?” is crucial in qualitative research. The data must be selected. During the work, we tried to separate actual information from its interpretation by the interviewee. We analysed the empirical data several times and tried to look beyond the literal answers to see what the interviewees really meant.
- 17 -
3 THEORY 3.1 History of Good Laboratory Practice The policies behind Good laboratory practice principles were created and developed in the USA in the 1970s by the Food and Drug Administration (FDA). This public health agency has protected American consumers by enforcing public health legislation concerning foods, drugs and cosmetics since 1906. In the 1970s the FDA inspected some non-clinical laboratories and revealed that some pharmaceutical studies submitted in support of the safety of regulated pharmaceutical products had not been conducted in accord with good practice. The data were inaccurate or insufficient, and therefore inadequate to ensure the safety of the product. The FDA bases its decisions concerning safety on data from companies’ studies, so the data must be correct and accurate. The FDA started drafting a policy for non-clinical laboratory practices that resulted in Good Laboratory Regulations (Good Laboratory Regulations, 21 CFR Part 58, December 22, 1978 (43 FR 59986)). The regulations, which came into effect in June 1979, establish standards for the conduct and reporting of non-clinical laboratory studies (ISO Guide 25, Agilent Technologies). The OECD countries followed the FDA’s example and compiled Principles of Good Laboratory Practice to promote public health and environmental safety in all OECD countries. These common principles are also intended to minimize technical trade barriers and make test data comparable between countries so that duplication of work can be avoided. Each OECD country has a regulatory authority responsible for monitoring and inspecting companies’ compliance with these principles. Once validated in one OECD country, this makes it possible to operate in the whole OECD community. GLP principles can also be adopted in countries outside the OECD. GLP status in a non-OECD country can be approved by authorities in an OECD country and will then be accepted in all other countries (GLP Handbook).
- 18 -
3.2
Good Laboratory Practice
The OECD’s principles of Good Laboratory Practice spell out a clear principle: laboratories should plan, prepare, perform, record and report non-clinical safety studies and testing carefully and accurately. The validity of the data must not be compromised at any stage of a study. The test materials, all activities and methods, and the results should be documented and stored in such way that the study can be reconstructed. An internal, independent, quality assurance unit is required in a laboratory or company to ensure the quality of the data, the whole process and procedures. In this thesis we refer to the OECD’s principles of Good Laboratory Practice utilizing it for definitions of terms. They can be downloaded from www.olis.oecd.org. According to the OECD’s Principles of Good Laboratory Practice revised in 1995 and 1996, “The purpose of these Principles of Good Laboratory Practice is to promote the development of quality test data. Comparable quality of test data forms the basis for the mutual acceptance of data among countries. If individual countries can confidently rely on test data developed in other countries, duplicative testing can be avoided, thereby saving time and resources. The application of these Principles should help to avoid the creation of technical barriers to trade, and further improve the protection of human health and the environment” (OECD’s GLP principles, Section I, Introduction). The GLP principles address the following, and so set requirements on them: •
Recruiting, appointing and training of staff and allocating of responsibilities
•
Handling of test materials
•
Choice of test methods and standard operating procedures
•
Quality assurance and quality measurement
•
Maintenance and calibration of equipment and machines
•
Samples and test data
•
Documentation, keeping and storing records and materials
- 19 -
It is ultimately the responsibility of the management of a company or organization to ensure that it complies with the GLP principles, so it is in the best interest of the company or organization to raise awareness of GLP.
3.2.1 Scope of GLP principles According to OECD principles section I, item 1, “These Principles of Good Laboratory Practice should be applied to the non-clinical safety testing of test items contained in pharmaceutical products, pesticide products, cosmetic products, veterinary drugs as well as food additives, feed additives, and industrial chemicals. Non-clinical health and environmental safety studies covered by the Principles of Good Laboratory Practice include work conducted in the laboratory, in greenhouses, and in the field. “Unless specifically exempted by national legislation, these Principles of Good Laboratory Practice apply to all non-clinical health and environmental safety studies required by regulations for the purpose of registering or licensing pharmaceuticals, pesticides, food and feed additives, cosmetic products, veterinary drug products and similar products, and for the regulation of industrial chemicals.”
3.2.2 Test facility organization and personnel The management of a laboratory organization or company is ultimately responsible for the quality and integrity of the testing and studies conducted in its laboratories (P. A. Carlsson et al, 1998). Section II of GLP principles sets this requirement out in the first item: “Each test facility management should ensure that these Principles of Good Laboratory Practice are complied with, in its test facility.” What are the duties of management required by the GLP principles? The management should select qualified individuals to serve as the Study Director and Deputy Study Director, and also provide a sufficient number of qualified laboratory staff. The job descriptions, qualifications, training and experience of all
- 20 -
the personnel should be recorded and updated regularly. Each study conducted at the laboratory requires the appointment of a responsible Study Director and Deputy Study Director. The Study Director drafts a new study plan, which must be approved by the management before it is started. The management should ensure that there are Standard Operating Procedures that have to be approved by the management before implementation. They should also ensure that there is a Quality Assurance Programme and that the Quality Assurance unit carries out its duties in accordance with GLP principles. The Study Director shall assume full responsibility for the GLP compliance of all activities and procedures of a study, and must approve and sign the final report of the results. The Study Director is also responsible for ensuring that no data in the study or testing are compromised at any stage. If something is not conducted in accordance with GLP principles, the Study Director must take corrective action (GLP Handbook).
3.2.3 Quality assurance programme GLP principles require that a laboratory/company has a quality assurance unit and QA inspector (head of QA unit) that monitor and inspect the facilities and the operations carried out in the laboratory. The company/laboratory has to have a quality
assurance
statement.
The
regular
inspections
by
the
laboratory’s/company’s own Quality Assurance Unit are reported and the reports kept. The inspections should cover organizational issues (such as staff levels, qualifications, planning of work, understanding duties), procedures for QA, facilities (such as health, safety, hygiene requirements), animal facilities (such as housing of different animal species separately, separation of clean and dirty areas), test substances, laboratory operations (such as different laboratory operations for different activities, monitoring of environmental conditions, facilities for the retention of samples and records) and equipment (such as condition, tolerance). These inspections also cover manual data recording and computer generated data, as well as standard operating procedures, preparations of reports and archiving of them (P. A. Carlsson et al, 1998). Detailed raw data audits should be carried out as part of these inspections. Errors such as wrong - 21 -
species, wrong factors or data used in calculations, inappropriate statistics applied and misspellings or missing or wrong references to test substances must be listed in audit reports. These records are available to government inspections and sponsors. These internal inspections are an excellent opportunity to promote best practices and prevent errors, rather than detect them. We refer to the details of this GLP principle requirement and the internal inspections in the enclosed OECD’s GLP principles.
3.2.4 Standard Operating Procedures (SOPs) The OECD principles state that GLP laboratories should have written SOPs and they should cover the following categories of laboratory activities: test and reference substance, apparatus and reagents, record-keeping, reporting, storage and retrieval, test system, quality assurance procedures, and health and safety precautions. SOPs need not to be limited to these categories. OECD principles list the key issues in each required category (see OECD’s principles). SOPs depend on the organization and activities of each laboratory. GLP principles require that SOPs should be approved by management. What company CEO would want to authorize sixty different instructions in non-clinical laboratory activities? What responsibility would such an approving signature entail? The purpose of authorization is to decide on an appropriate method for a given procedure, standardize the laboratories’ performance of the procedure, and communicate these decisions to everyone involved (P. A. Carlsson et al, 1998). A rational approach to effective authorization is to include this in a company policy in which this authority to approve SOPs is delegated to someone with management status and special insight and knowledge. Those persons empowered with this authority must be high-ranking managers in the organization, as they take full responsibility for authorizing the procedures. This is a requirement of government officials in some countries.
- 22 -
The SOPs must communicate their message effectively in the language currently used in the laboratory. Instructions should be written in the right order using clear short words and terms, and short sentences where possible. Everyone involved should have immediate access to SOPs. Serious deviations from SOPs should be immediately reported to the Study Director and fully documented in the study data. Such deviations may delay the whole study, but rescheduling and redoing the test in full compliance with SOP is better than having the data of the whole study declared invalid in the final report. SOPs should be reviewed regularly. SOPs should also include a procedure for updating a SOP. SOPs bring other advantages to the company/laboratory in addition to production of quality data: they clarify working routines and responsibilities. Good SOPs signal the management’s commitment to quality documentation and GLP (P. A. Carlsson et al, 1998).
3.2.5 Facilities The OECD principles state that GLP studies should be carried out in testing facilities of suitable size, construction and location to meet the requirements of the study and minimize disturbances that would interfere with the validity of the study. The building materials used in construction of GLP laboratories should be such that they are easy to clean but do not allow test materials to accumulate and cross-contaminate others. Any internal or external potential threat to the study should be minimized. Testing facilities should be designed so as to provide an adequate degree of separation of the various aspects of the study (OECD principles). The objectives of the study and how to achieve them should be carefully considered. The functions or activities must be separated so that they would not compromise the study. Walls, filters and isolators can be used for physical separation. Entries to certain areas should be restricted. If clean and dirty materials are moved around the facility and they cannot be handled in separate areas, they should be handled at different times of day. - 23 -
Areas within a laboratory should be defined for carrying out certain activities. Special attention to areas concerned with test material control and mixing with vehicles is required. Areas should be designated so that they allow laboratory staff to carry on their work without risk of getting in each other’s way or mixing up different materials. Separation can also be achieved by letting the staff carry out different function at different times, allowing time for cleaning between working shifts. The GLP Handbook lists these as important areas to be separated: • Storage of test items under different conditions • Storage of control items • Handling of volatile materials • Weighing • Mixing of different dose formulations, e.g. in the diet or as solutions or suspensions • Storage of prepared doses • Cleaning equipment • Offices and refreshment rooms • Changing rooms According to the GLP Handbook, the following areas should be kept separated in animal facilities to meet the requirements set by GLP principles: • Species • Studies • Quarantine • Changing rooms • Receipt of materials • Storage – bedding and diet – test doses – cages • Cleaning equipment
- 24 -
• Necropsy • Laboratory procedures • Utilities • Waste disposal
3.2.6 Equipment, material and reagents All appropriate equipment including computer systems with adequate capacity should be available for proper conduct of the study, and they should be periodically inspected, cleaned, maintained and calibrated to ensure accurate performance. Dates and details of maintenance, repairs, calibration and any nonroutine work should be recorded and retained (GLP Handbook). The OECD principles state that materials used in a study should not adversely interfere with the test systems. Requirements concerning labelling and providing details of all chemicals, reagents and solutions used in the study are specified in the OECD principles.
3.2.7 Test systems The definition of a test system according to GLP is very varied. Test systems are often animals, but they can also be plants, bacteria, organs, cells or analytical equipment (GLP Handbook). Analytical apparatus used to generate physical or chemical data should be suitably located and of appropriate design and adequate capacity. The integrity of the physical or chemical test systems should be ensured (OECD principles). If the test system is biological, proper conditions should be established and maintained for the storage, housing, handling and care of biological test systems to ensure the quality of the data. Newly received animal and plant test systems should be isolated until their health status has been evaluated. At the experimental starting date of a study, test systems should be free of any disease or condition that might interfere with the study. Any diagnosis or treatment of any disease
- 25 -
before or during a study must be recorded. Records of source, date of arrival, and arrival condition of test systems should be maintained. Biological test systems should be acclimatized to the test environment for an adequate period before the first application of the test or reference item. All information needed to identify the test systems properly should appear on their housing or containers (OECD principles). Housing or containers for test systems should be cleaned at appropriate intervals during use. Any material coming into contact with the test system should be free of contaminants at levels that would interfere with the study. Bedding for animals should be changed as required by sound husbandry practice. Use of pest control agents should be documented. Test systems used in field studies should be located so as to avoid interference in the study from spray drift and from past usage of pesticides (OECD principles).
3.2.8 Test and reference items Reference items or substances are important because they are used to calibrate instruments. The accuracy of the analytical method is therefore determined by the accuracy of the reference item (ISO Guide 25, Agilent Technologies). Records should be kept including test item and reference item characterization, date of reception, expiry date, and quantities received and used in studies (OECD principles). Methods of synthesis, fabrication or derivation of test and control articles should be documented and included with the study (ISO Guide 25, Agilent Technologies). Handling, sampling and storage procedures should be controlled so that stability is ensured as far as possible and contamination or mixup are prevented. Storage containers for the test and reference items should carry identification information, expiry date and specific storage instructions (OECD principles). Storage containers should be assigned to a particular test article for the length of the study (ISO Guide 25, Agilent Technologies). Each test and reference item should be appropriately identified. It must be possible to define each different batch of the test. In cases where the test item is - 26 -
supplied by the sponsor, there should be a mechanism to verify the identity of the test item subject to the study. The stability of test and reference items under storage and test conditions should be known for all studies (OECD principles). The stability of each test or reference item should be determined. This can be done either before starting the study or according to written SOPs that provide for periodic reanalysis of each batch (ISO Guide 25, Agilent Technologies). A sample for analytical purposes from each batch of test item should be retained for all studies except short-term studies (OECD principles).
3.2.9 Performance of the study Study Plan
A written plan for each study should exist before the start of the study. The study plan should be approved by the dated signature of the Study Director and verified for GLP compliance by Quality Assurance personnel. Changes to the study plan must be scientifically justified and approved by the dated signature of the Study Director and kept with the study plan (ISO Guide 25, Agilent Technologies). Deviations from the study plan should be described, explained, acknowledged and dated in a timely fashion by the Study Director and maintained with the study raw data (OECD principles). Changes must not obscure the original (ISO Guide 25, Agilent Technologies). Content of the Study Plan
The study plan should contain, but not be limited to the following information:
Identification of the Study, Test Item and Reference Item
A descriptive title and a statement that reveals the nature and purpose of the study and the test items and reference item to be used must be identified by a code or name Information Concerning the Sponsor and the Test Facility
Name and address of the sponsor, the test facilities, test sites and Study Director.
- 27 -
Dates
The date of approval of the study plan by the signature of the Study Director. The date of approval of the study plan by the signature of the test facility management and sponsor if required by national regulation or legislation in the country where the study is being performed. Other important dates are the proposed experimental starting and completion dates. Test Methods
Reference to the OECD Test Guideline or other test guideline or method to be used. Issues (where applicable)
•
Reason for selection and characterization of the test system.
•
Method of administration and reason for selecting it.
•
Detailed information on the experimental design.
Records
A list of records must be retained.
Conduct of the Study According to the OECD principles:
1.
A unique identification should be given to each study. All items concerning this study should carry this identification. Specimens from the study should be identified to confirm their origin. Such identification should enable traceability of the specimen and study.
2.
The study should be conducted in accordance with the study plan.
- 28 -
3.
All data generated during the conduct of the study should be recorded directly, promptly, accurately and legibly by the individual entering the data. These entries should be signed or initialled and dated.
4.
Any change in the raw data should be made so as not to obscure the previous entry, should indicate the reason for change and should be dated and signed or initialled by the individual making the change.
5.
Data generated as a direct computer input should be identified at the time of data input by the individual(s) responsible for direct data entries. Computerized system design should always provide for the retention of full audit trails to show all changes to the data without obscuring the original data. It should be possible to associate all changes to data with the persons having made those changes, for example, by use of timed and dated (electronic) signatures. Reason for changes should be given.
3.2.10
Reporting of study results
GLP principles provide detailed information and requirements for how a final report for each long-term study should be prepared. For short-term studies, a standardized final report may be prepared. These reports should be dated and signed by the Study Director, principal investigators and/or scientists who were involved in the study. The signing of the data by the Study Director indicates compliance with GLP principles, and his acceptance of responsibility for the validity of the data. The Study Director is responsible for the scientific interpretation of the results of the study (GLP Handbook). Reasons for amendments (corrections and additions) to a final report should be clearly specified and approved, then signed and dated by the Study Director. Reformatting of the final report to comply with the submission requirements of a regulatory authority does not constitute an amendment to the final report (OECD’s GLP principles). - 29 -
The detailed contents of the final reports according to OECD’s GLP principles shall be: “1. Identification of the Study, the Test Item and Reference Item
2.
a)
A descriptive title;
b)
Identification of the test item by code or name (IUPAC, CAS number, biological parameters, etc.);
c)
Identification of the reference item by name;
d)
Characterization of the test item including purity, stability and homogeneity.
Information Concerning the Sponsor and the Test Facility a)
Name and address of the sponsor;
b)
Name and address of any test facilities and test sites involved;
c)
Name and address of the Study Director;
d)
Name and address of the Principal Investigator(s) and the phase(s) of the study delegated, if applicable;
e)
Name and address of scientists having contributed reports to the final report.
3. Dates Experimental starting and completion dates. 4.
Statement A Quality Assurance Programme statement listing the types of inspections made and their dates, including the phase(s) inspected, and the dates any inspection results were reported to management and to the Study Director and Principal Investigator(s), if applicable. This statement would also serve to confirm that the final report reflects the raw data.
5.
Description of Materials and Test Methods a)
Description of methods and materials used;
- 30 -
b) 6.
7.
Reference to OECD Test Guideline or other test guideline or method.
Results a)
A summary of results;
b)
All information and data required by the study plan;
c)
A presentation of the results, including calculations and determinations of statistical significance;
d)
An evaluation and discussion of the results and, where appropriate, conclusions.
Storage The location(s) where the study plan, samples of test and reference items, specimens, raw data and the final report are to be stored.”
We have presented this section in detail, as we consider that compliance with this section is crucial to ensure success in high quality testing and research. The fundamental principle of this section is that laboratories should conduct nonclinical testing and safety studies carefully and document all the activities so that the study can be reconstructed in future at any time.
3.2.11
Storage and Keeping of Records and Materials
For regulatory purposes and for the company’s own research purposes, the records and materials of a study conducted according to GLP principles should be stored and the records safeguarded. If any data are lost or details are not recorded properly, the validity of the data is compromised. In other words, what was done, how, by whom and when must be accurately included in the records. The archives should be maintained in a systematic, organized way, and storing and record keeping should be conducted simultaneously so that nothing is lost or forgotten. GLP principles set out the requirements for reconstruction of a study, so it is important to store and keep records and materials in a safe place for a period of time. How long they must be stored is specified by the regulatory authorities.
- 31 -
The archives should be indexed and recorded in a way that makes accessing them easy. Only authorized people should have access to the archives, and the moving of records in and out should be monitored and recorded. The following should be stored and kept according to OECD’s GLP principles: “a)
The study plan, raw data, samples of test and reference items, specimens, and the final report of each study;
b)
Records of all inspections performed by the Quality Assurance Programme, as well as master schedules;
c)
Records of qualifications, training, experience and job descriptions of personnel;
d)
Records and reports of the maintenance and calibration of apparatus;
e)
Validation documentation for computerized systems;
f)
The historical file of all Standard Operating Procedures;
g)
Environmental monitoring records.
“In the absence of a required retention period, the final disposition of any study materials should be documented. When samples of test and reference items and specimens are disposed of before the expiry of the required retention period for any reason, this should be justified and documented. Samples of test and reference items and specimens should be retained only as long as the quality of the preparation permits evaluation.
- 32 -
“If a laboratory or test facility or an archive contracting facility goes out of business and has no legal successor, the archive should be transferred to the archives of the sponsor(s) of the study(s).” As more and more data are recorded and stored electronically, the security and safety of hardware and software increases in importance. Laboratories and companies should have proper authorization systems and restricted access to appropriate files. There should be adequate user ID management, regular changes of passwords and firewalls installed. When storing information in electronic form, it is also important to store the data in a form that will be retrievable in the future. As hardware and software are all the time being developed and upgraded, it is important to retain hardware or software that can retrieve the stored files, or alternatively a procedure to convert the files to an upgraded version should be defined and undertaken accordingly. In a large company or laboratory, the computer function may be centralized. The requirements spelled out in GLP principles should be clearly communicated to this function, which should provide back-up systems for computer failures and means of retrieving any data lost in breakdowns. Normally, a company has an emergency or crisis management plan for unexpected events beyond its control, such as war or a state of emergency declared by the government, natural disasters, fire or attack by terrorists or animal rights activists or persons opposed to genetically modified crops. The appropriate requirements of GLP principles should be fully incorporated into this plan to ensure no loss of crucial data.
3.3
Quality systems
The purpose of quality systems is to improve quality within the company. The roots of today’s quality systems originate from the USA. In the 1920s W. A. Shewhart developed methods for product control, and in the 1950s Joseph Juran and W. E. Deming developed methods for process control (Bodin et al, 2003). The most significant quality improvement took place in Japan, where kaizen
- 33 -
(continuous quality improvement) and kanban (just in time) were developed by Japanese companies to gain a competitive edge over competitors. The Total Quality Management (TQM) philosophy was developed in the 1980s, and the Malcolm Baldrige National Quality Award based on TQM philosophy was established as a standard in 1987. This was the model for the Swedish standards established by the Swedish Standards Institute, SIS (Sweden), which was established by a government decree in 1990 (Bodin et al, 2003). The International Organization for Standardization is responsible for the international quality management system ISO 9000 series of standards. Each company can choose the standards suitable for its operations. ISO 9000 is a certification granted to companies that manufacture goods. It specifies that they have properly recorded all of their processes, which meet certain specifications. The standards deal with factors such as quality management system, management responsibilities, resource management (material and human resources), and the processes a company uses and how it measures data to monitor its quality control and then use the results to improve quality requirements. EN ISO 9001-2000 is a quality assessment scheme for laboratory medicine. Guidance for document management, for instance, is in ISO 9000 and ISO 14000.
3.3.1 Quality systems and the pharmaceutical industry Quality systems, and national and international standards are created to improve the quality of products and services. Quality has many dimensions and works on many levels. According to definitions of quality, customer satisfaction is the ultimate benchmark of quality (Bergman et al, 1998:298). Quality according to many definitions is essentially satisfying customers’ needs and expectations, preferably over-satisfying these needs (Bergman et al, 1998). For pharmaceutical products, the safety of a product is its key quality factor after its therapeutic purpose: a medicine must have no significant adverse side effects. In general terms, one could say that for a medicine, good quality means that it is good for its therapeutic purpose and has no significant adverse effect, so the pharmaceutical industry should manufacture such medicines to satisfy their customers.
- 34 -
3.3.2 Quality systems as a managerial tool Elements of management control systems include strategic planning, budgeting, resource allocation, performance measurement, evaluation, reward, responsibility centre allocation and transfer pricing (R. Anthony et al 2003). The overall purpose of a management system is to help a company to implement its strategies. When a company decides to implement a quality system as a part of its management system, this can enhance its performance in many ways. Improved quality can affect profitability in many ways: •
More satisfied customers, who buy again
•
Lower staff turnover and days of sick leave
•
Improved competitive position in the market
•
Shorter standstill times, delays
•
Facilitates effective use of capital resources
•
Less defects, unusable products and rework
•
Improved productivity
(Bergman et al, 1998, including its Figure 2.1) Deming describes this in the following figure (Deming, 1986): Improve quality }
→ Costs decrease because of less rework } → Productivity improves } → fewer mistakes,
fewer delays, fewer snags, better use of machine time and materials
} → Capture market } → Stay in business
} → Provide jobs and more jobs } → with better quality and lower prices
3.4 GLP quality system A good understanding of Good Laboratory Principles and their adoption helps laboratories in their quality improvement. GLP is a managerial tool covering the organizational process and the conditions under which non-clinical laboratory studies are designed, conducted, monitored, recorded and reported. A good documentation practice ensures high ethical standard in any study a laboratory undertakes. The intention of GLP is to promote the quality and validity of the data forming part of the studies. The requirement spelled out in GLP principles on
- 35 -
internal quality assurance within a company must be seen as one part of an overall quality system of the company’s total quality management system. The duties of management personnel as described by GLP principles entail great responsibility. According to Deming’s principle, it is the top management that have a key role in promoting quality. Only through top management being wholeheartedly involved in quality improvement issues can a sustainable quality improvement be achieved (Bergman et al, 1998). Compliance with GLP principles increases the company’s profile as a provider of safe and high quality products and enhances its competitive position in the market. The scientific work that companies produce can also be commercially utilized as it is produced to international standards (P.A. Carlsson et al., 1998).
- 36 -
4
RESULTS
4.1
Companies interviewed
We invited all the twelve GLP laboratories listed in MPA’s GLP programme to participate in our research. These twelve GLP laboratories were represented by six contact persons on MPA’s list because one person represented seven laboratories owned by the same pharmaceutical company. All these six people were interviewed and contributed valuable information to our study, which covers the operations of all the GLP laboratories monitored by MPA in Sweden. The companies have a wide range of activities and different sizes. Here follows a short description of the companies in alphabetical order.
Active Biotech The present company, situated in Lund, originated from part of Pharmacia that was bought by Active Biotech in 1997. Active Biotech focuses on developing pharmaceuticals for medical fields in which the immune system plays a central role.
AstraZeneca AstraZeneca was formed in 1999 through the merger of Astra AB of Sweden and Zeneca Group PLC of the UK. The company has research, production and marketing facilities in Sweden. Astra AB, which was founded in 1913, is headquartered in Södertälje. Astra was an international pharmaceutical group engaged in the research, development, manufacture and marketing of pharmaceutical products, primarily for four main product groups: gastrointestinal, cardiovascular, respiratory and pain control. Some research effort was also put into the central nervous system.
BioVet The company was founded in 1983 in Stockholm. BioVet is a veterinary medicine laboratory that concentrates on pets and racehorses. It specializes in histopathology, cytology and immunology tests. - 37 -
Quintiles Transnational Quintiles is a global contract research company established in North Carolina, USA in 1982. It provides the full spectrum of product development and commercialization activities, from early compound development, laboratory services and regulatory submission to sales and marketing. Quintiles employs about 16,000 people in 50 countries, including 400 working on Quality Assurance and 1,000 on information technology and communications.
National Veterinary Institute (SVA) The National Veterinary Institute is a Swedish national authority that strives for good animal and human health, a good environment and sustainable food production. GLP studies are only a small part of its operations: only two of SVA’s four hundred employees work on GLP studies carried out by SVA as a Contract Research Organization. SVA provides diagnostic activities and preparedness by developing diagnostic methods and laboratory products. Bacterial diagnostics, substrate products and cell culture media are mainly prepared for SVA’s own diagnostics. Blood products are produced from horses, cattle, sheep, goats, rabbits and poultry from SVA’s own Swedish herds and flocks. SVA also supplies the vaccines required in preventing disease. It has its own research, production and marketing of autogene vaccines.
Visionar Biomedical Visionar, which was established in 2000 in Uppsala, Sweden, has five employees. It is a contracting research company providing services for pharmaceutical and biotechnology companies. Visionar helps its customers even with the evaluation of their preclinical project strategy, or gives a “second opinion” about the potential of a product.
4.2
Replies from respondents
4.2.1 Interview questions The questions asked can be divided into three groups. First, we wanted to know about the validations process: why did the company want to implement GLP,
- 38 -
when did it start the process of compliance, how long did that take and what are its experiences of the process? Secondly, we wanted to know what changes GLP had brought to the company, especially positive and negative effects on quality. Thirdly and finally, we asked what effect losing GLP quality status would have on the company (for a complete list of questions, see appendix 1). A summary of the answers that does not link the answers to a specific company will be given here.
4.2.2 Adoption of GLP principles The time that GLP has been in use in the various laboratories differs a lot. One of the businesses employed the first workers engaged in GLP on 20th June 1979 as GLP became statutory through legislation in the USA, whereas one has been validated only since 2001 (see Table 1). Table 1. Dates of introduction of GLP and time needed for implementation. GLP since
Duration of validation process
1979
-
About 1990
-
1996
1-2 years
2000
1 year
2000
1-2 years
2001
6 Months
Of the two respondents not specifying a duration, one said that the company had been working according to GLP for such a long time before validation was conducted in Sweden that it was impossible to say how long the validation process took, and the other said that the company was already validated when it was bought by the current firm. For the laboratories that sought validation immediately after the regulatory framework became into force in Sweden in the 1990s, it took between one and two years to gain GLP status, whereas for the laboratories that sought GLP - 39 -
validation after 2000, it took only one year, and for the laboratory that gained GLP status most recently, it took only six months to be approved by MPA as a GLP laboratory. There has been compliance with GLP principles for over ten years in Sweden, but our research shows these principles are still not well known. Bioanalyst training, for instance, does not include any training in GLP principles. The laboratories provided internal and also external training to keep staff updated on GLP principles. All but two laboratories stated that the staff they recruited had no knowledge of GLP: one respondent said that if there were two applicants otherwise equally qualified, but one had experience of GLP, then the one that had experience of GLP would be recruited; another respondent did not answer the question. Regulatory frameworks including GMP and GLP principles aim to raise the quality of pharmaceutical procedures. However, although GMP and GLP principles are stipulated in law, they are not yet well known. This is a problem for pharmaceutical companies needing to employ graduates that are qualified to work in the pharmaceutical industry, as hardly any of them have heard of the principles of GLP. Three of the six respondents stated that new recruits to GLP laboratories had no knowledge of GLP at the time of recruitment. They had to be trained in GLP principles. The interviewees’ overall experiences of adopting and complying with GLP principles were good. All but one respondent stated several ways through which quality had been improved by adopting and adhering to GLP principles. The only laboratory that had no experience of quality improvement stated that the question whether GLP had improved quality of its operations did not apply because it had been set up with a business strategy of doing only GLP studies, in other words this laboratory did not start its operations until MPA had approved its status as a GLP laboratory, so it had achieved consistent GLP high quality from the start.
- 40 -
4.2.3 Why did laboratories seek validation? Two laboratories stated that they sought GLP validation because it was a requirement from authorities for doing the type of studies (GLP studies) the laboratories wanted to conduct. It was necessary for the firm to be able to keep on working as before. Working according to GLP became obligatory in the USA in 1979. Pharmaceutical companies that wanted to keep on registering drugs there had to introduce GLP in the daily work. Two of the firms are or were part of large pharmaceutical businesses and were very early on implementing GLP. One of them indicates that it would have been too expensive to go outside of the firm for all GLP studies that had to be made. The other four organizations are more or less contract research laboratories and it was either demands from their customers or the opportunity to be able to perform more studies that made them comply with GLP. The reasons given for complying with GLP included: “To be able to perform GLP compliant studies” (otherwise it would have been an additional cost to make another laboratory do the tests for which GLP is mandatory). “It is necessary to be able to register drugs in the USA.” “There is more work for a GLP validated laboratory.” “We’re working as a contract laboratory and it was demanded by our customers.” “It was demanded by the customers. They wanted to have a study performed that required GLP status.” “It was a way to make money.” Of the six interviewees, one respondent for a GLP laboratory said it had chosen as its business strategy to set up a GLP laboratory to make money, in other words its motivation was purely business related. As the respondent said, “We sought GLP status for purely business reasons: we wanted to make money.” A reason given by three respondents was that it was a requirement from their customers, in other words the customers for which they were supplying services
- 41 -
demanded that the non-clinical studies were conducted in compliance with GLP principles. These laboratories wanted to satisfy their customers’ needs.
4.2.4 Implementation and changes needed According to four of the companies, the key changes in the process of seeking GLP status were, writing SOPs (Standard Operating Procedures) and agreeing what to write in them. SOPs must be established for the personnel working according to GLP. On the other hand, once this tedious and time-consuming task had been completed, it proved a great help in training new recruits. The SOPs can be used as a reference book above all for new employees, but also for others. All of the activities in the laboratory will be documented in a way that makes all routines easy to access. The SOPs also make the work more reproducible, since every one is working according to the same manual. It is also necessary to start up a partly new organization for the GLP. The QA and its independent audit function is part of that, as some of the contact persons mentioned. Since a GLP study has to follow GLP in every link of the chain, one company had to start to buy the living material that it uses in the studies from a different firm than before, one working according to GLP. Some were working in a very GLP-like manner before and did not have to go through such a hard process of change to implement GLP. “The adoption wasn’t that extensive, but there were many details. Others had to start to work and document in a completely new way. It took much more time than we had expected. It was a much bigger work.”
4.2.5 Quality systems Table 2. Quality systems employed before and parallel with GLP Quality system before GLP
Quality System parallel with GLP
Written routines
-
GMP, GCP
GMP, GCP, six sigma, lean, ISO 9000
Not specified
-
EN 45001
ISO 17025
- 42 -
Quality system before GLP
Quality System parallel with GLP
EN 17005
EN 17005 (Swedac)
-
-
Half of the companies were working according to other named quality systems before they implemented GLP. All of these also have other systems parallel with GLP today. ISO 9000 is a certification granted to companies that manufacture goods. This is really parallel to GMP (Good Manufacturing Practice) and not really to GLP. EN ISO 9001-2000 is a quality assessment scheme for laboratory medicine.
4.2.6 What advantages did the laboratories gain from GLP? According to our research findings, adherence to GLP principles brings the following advantages: 1. Increased business through GLP compliance: •
requirement of customers
•
GLP status attracts more new customers
•
GLP studies lead to other development studies
•
GLP status an excellent reference
2. Operations become more efficient, systematic and better organized; 3. Easier induction of recruits; 4. Human capital retained as knowledge within the company when an employee leaves.
Respondents’ answers One respondent said, “It’s a quality label that proves the non-clinical studies were conducted in adherence to GLP principles as agreed with the customer, so the customer can then apply for the Stage 2 clinical trials required to get the medicine approved by MPA.”
- 43 -
Another said, “GLP is beneficial because it forces a company to work in a structured and organized way. As our company already worked according to such principles, there were no major changes for us.” Another respondent noted, “Having GLP status enhances our chances of gaining more GLP and other studies — more business. It is like an advertisement.” Another respondent pointed out, “Adherence to GLP principles helps preserve knowledge. If a researcher leaves the company without documenting everything he had worked on, then a lot of knowledge would disappear with him. More comprehensive and improved documentation leads to knowledge being retained in the company. We cannot afford to lose knowledge! Other advantages of following GLP are that every SOP must be written down and stored. For the recruits, it is especially advantageous that all processes are clearly and straightforwardly set down in writing, like an encyclopaedia that can be picked up and read through any time necessary. There is then no need always to ask somebody.” Another respondent said, “Operations are run more systematically and we can find things because we have better documentation. We also review what we file and store in the archives after the studies have been completed. That is part of documentation, so it improves quality.” Another respondent summarized the benefits from adhering to GLP principles as “GLP is mandatory for preclinical safety documentation, and thus necessary to adhere to.”
4.2.7 Has GLP improved quality, and if so, how? GLP does indeed improve quality. Our research revealed several ways: •
providing a competitive edge as an indicator of superior quality that attracts more customers;
•
routinely adopting GLP principles in other studies, even when not required, increases overall quality not just in GLP studies;
- 44 -
•
improved, more detailed documentation and record keeping reduces the risk of mistakes and faults;
•
defining tasks and responsibilities, and ways of doing things, enhances performance;
•
regular inspections by authorities, QA and sponsors prevents any cheating or falsification;
•
increased safety of the studies due to stringent compliance with SOPs and study master schedules;
•
increased communication (everyone must follow SOPs) and dialogue between Study Director, employees, sponsors and inspectors;
•
improved quality through continuous improvement of the quality system.
4.2.8 Quality and continuous improvement Continuous quality and improvement work is an important part of the GLP system according to several of the respondents: “In our company GLP isn’t so much an instrument of control, but rather a proactive method.” “At first you have to set the quality base for the work according to the principles of GLP and then continuously maintain that quality and work to improve quality.” “You think about what you do and how you do certain things. It is an ongoing process.” One respondent said that in the end, the whole reason for the GLP system and also for the work of the company is to deliver good quality to the customer. Traceability has improved for many of the organizations. “When we had to follow studies backwards in time, after the validation, it was possible to see exactly what had happened to the sample.”
- 45 -
One firm has become more systematic in the way the instrumentation is calibrated and maintained. This must also be done in a controlled and regular manner. One other respondent said, “GLP forces the company to work in a structured and organized way.” The documentation has become better, more extensive and more uniform. This has made it possible to save knowledge more efficiently. “The company can’t afford to lose knowledge.” Answers from respondents also included the following comments: “We all have the same understanding of the processes and do things in almost the same way, so in this way operations have become more efficient. It is easier for us to gain new assignments as we advertise that we have GLP status, and we get other assignments, development assignments. It’s an advert.” Written regulations, guidelines and SOPs are of little use unless they are followed. Every employee needs to take responsibility for adhering to the principles, and needs to follow SOPs and all guidelines in the smallest detail. One respondent summarized this in her reply:
“We try to COMMUNICATE very often . . . and EVERYONE needs to be INVOLVED in the LOOP.”
4.2.9 Have laboratories experienced any disadvantages from GLP registration? Four of the respondents talked about bureaucracy as a problem with GLP. One of the respondents meant that it is absolutely essential to find a balance. The SOPs and the increasing documentation will increase bureaucracy, but it must not take too much time and the system must not become inert. With a good balance, it is possible to get most of the benefits of GLP. Another meant that there is no way of getting over the bureaucracy problem. It is a part of the system. Another important aspect is connected to the very soul of science. “Science is to document. If you
- 46 -
don’t save your results, the knowledge won’t stay and there won’t be any result of the studies performed.” Sometimes the inspectors can be very meticulous: “How do you know that the freezer stays at -14º C?” “Because the thermometer says so.” “How do you know that the thermometer displays the correct temperature?” and so on. GLP makes studies more expensive, since there has to be a QA. The composition of the QA, states one of the respondents, is very important for how the GLP work functions. One other thing that is important for how smoothly GLP implementation works and how it is continued is the attitude of the management. The specific points were raised by respondents can be summarized as follows: •
GLP increases costs for the customer;
•
keeping SOPs updated and providing staff training to keep them up to date increases costs;
•
sometimes customers expect the laboratory to have even more documentation;
•
GLP-related problems can be created by inexperience or personality, or management’s attitude (whether they see GLP as a necessary evil or an opportunity to improve quality, and QA staff are crucial in this respect);
•
increased bureaucracy;
•
the specific “triangle problem” of CROs.
The “triangle problem” refers to a situation where a Study Director at a CRO who is responsible for compliance with the Master Schedule of the GLP study is “pressured” by the sponsor providing the money for the GLP study as to how the study should be carried out. One respondent said, “So there are difficulties sometimes in matching GLP with reality, but in our case it is the customer that has the money who is the person that decides in the end, and not the Study Director. This we have discussed with the Medical Products Agency’s inspector, but it is GLP that says that the Study Director has the overall responsibility. In some way we try to cope with the system. It is a little bit difficult sometimes.” - 47 -
The Study Director also has to meet the MPA inspector’s statutory requirements for approval, so in some respects a sponsor or an inspector could have the last word even though according to GLP Principles it is the Study Director who has the responsibility for a GLP study.
4.2.10
What are the sanctions for non-compliance?
When we asked respondents what the consequences for their laboratory would be if they lost GLP status, loss of business (financial consequences or preventing them from getting approval for their potential medicines) was the most common answer. According to Senior Expert Dr Ragnar Hede of the Medical Products Agency, there are no legal sanctions for non-compliance, in other words failing to comply with GLP does not entail prosecution or sanctions such as fines or imprisonment. Even in a case of cheating, falsification or plagiarism, which is very unlikely with GLP studies but unfortunately takes place in some studies at some universities, the criminal definition of fraud in Swedish law does not cover this type of action. Non-compliance can lead to loss of GLP status. An inspector can through an inspection of a GLP laboratory find three types of deficiencies in compliance: minor, major and critical. The GLP laboratory replies to MPA in writing stating how any minor deficiencies have been eliminated. In the major and critical deficiency cases, suspension and loss of GLP status may follow. The question on the consequences for the company for non-compliance was answered quite differently. New drugs could be stopped, customers could be lost, and increased project costs could be a consequence because compliant CROs would have to be hired for the GLP studies. One respondent answered that it would also save money, since it is an extra cost to stay validated. For all (100%) the CROs, loss of GLP status would lead to loss of business and customers, with serious financial consequences.
- 48 -
For the in-house pharmaceutical GLP laboratories, loss of GLP status would prevent them from getting approval for their potential medicines, in other words there would be financial consequences.
Extracts from some respondents answers: “We would lose a lot of business and there would be financial consequences.” “I think we would lose some of our customers, because it is a requirement for them that they use laboratories that have GLP.” “A consequence of non-compliance would be that potential medicines would be blocked.”
- 49 -
5
ANALYSIS AND CONCLUSIONS
We set out to find out how Good Laboratory Practice functions for quality improvement and what drivers there are for seeking GLP status.
5.1
Drivers for seeking GLP status
We have assumed in our research that laboratories adopt GLP principles mainly because of legal requirements. Our expectation of why laboratories/companies seek GLP compliance turned out to be incorrect. We expected legislation to be the major reason why laboratories/companies sought GLP validation, but the main reason was that laboratories wanted to meet customers’ needs and requirements, as well as authorities’ requirements.
Legislation
Customer demands
Marketing/ selling of medicines
GLP validation Alternative quality systems
Quality Financial benefits
Authorities
Figure 3. We have in open interviews with GLP laboratories learned that the main driving force for GLP validation is demands from the customers. Figure 2 had to be corrected correspondingly, as above. Legislation seems to be directly influencing the need for GLP validation only when the company wants to register drugs in the USA. In this case the benefit is also the one we had expected: marketing and selling of medicines. In companies working as CROs, customers’ demands, often due to legislation and authorities’ demands, are the major driver for GLP validation. Compliance with GLP for these companies is beneficial financially, since they can accomplish more assays for more companies or do not have to hire GLP competence. The major benefit according to respondents, though, is increased quality. Documentation (SOPs and - 50 -
records) becomes more comprehensive, which improves reproducibility and safety. Training and communication must be part of the system, and continuous quality improvement is also embedded in the system.
5.2
Quality
The first part of our goal was to find out how good laboratory practice functions for quality improvement.
5.2.1 Two paradigms According to definitions of quality, customer satisfaction is the ultimate benchmark of quality (Bergman et al, 1998:298). According to Deming’s principle, top management has a key role in promoting quality. Both paradigms have also been stated by some of our respondents. They are aware that customers are their key stakeholders and aware of the importance of the management thinking.
5.2.2 How does GLP improve quality?” According to the OECD’s Principles of Good Laboratory Practice revised in 1995 and 1996, the purpose of these Principles of Good Laboratory Practice is to promote the development of quality test data. Adoption of GLP principles not only improves the safety of non-clinical studies, it can also improve the overall quality of operations. Meeting customers’ needs (they require studies conducted in accordance with GLP) can gain more satisfied customers, and with more satisfied customers, it is easier to gain more business. The overall quality of operations can be improved by adopting GLP in other than GLP studies. Management of the organization becomes more efficient as GLP addresses key areas including personnel, processes, documentation and reporting. SOPs and record keeping in detail help everyone involved in the studies and operations to access information required. By complying with GLP principles,
- 51 -
managers have assurance through the company’s internal quality assurance unit that the studies conducted in their laboratories are performed to high standard. As GLP principles specify standard operating procedures, very few procedures need to be repeated.
As mentioned above: When a company decides to implement a quality system as a part of its management system, this can enhance its performance in many ways. Improved quality can affect profitability in many ways: •
More satisfied customers, who buy again
•
Lower staff turnover and days of sick leave
•
Improved competitive position in the market
•
Shorter standstill times, delays
•
Facilitates effective use of capital resources
•
Less defects, unusable products and rework
•
Improved productivity
Our respondent did not mention any positive effects of the points in grey originating from the GLP system, but all other factors were said to be advantages of the GLP system.
5.2.3 PDCA cycle and continuous improvement Quality improvement work is an ongoing process. The more systematically quality improvement is embedded in operations, the greater the improvement in operations. According to the answers we got, this is the case for GLP, which makes it a little like kaizen, which is continuous quality improvement. Continuous dialogue and communications with staff, QA, inspectors and sponsors keeps everyone updated and in the loop of improvement, and any weaknesses or uncertainties in the whole organization can be minimized or eliminated. This extract from the mission statement of one of our respondents summarizes GLP’s function in quality improvement very well:
- 52 -
“We thrive on continuous IMPROVEMENT and QUALITY enhancements through attention to detail, consistent feedback, state-of-the-art systems, TRAINING and TEAMWORK.”
4. Conclusions for creating - best practices - benchmarking Is there a need to do something better?
1. Identify a need for improvement by reviewing, analysing and defining current status of activities
A P Act Plan C D Check Do
3. Monitor effects of action taken - results - staff/customer satisfaction - performance
2. Implement improvement in for example - staff (training) - processes,safety - doc./reporting - IT, communic.
Figure 4. The PDCA cycle shows the loop for continuous improvement of a quality system.
5.3
Validation process
An interesting point about the process by which the laboratories obtained GLP validation was that for the laboratories that sought validation immediately after the regulatory framework became into force in Sweden in the 1990s, it took between one and two years to gain GLP status, whereas for the laboratories that sought GLP validation after 2000, it took only one year, and for the laboratory
- 53 -
that gained GLP status most recently, it took only six months to be approved by MPA as a GLP laboratory. This could be because the guidelines or policies for GLP are so vague. What to do is outlined, but not at all how. It might be possible to validate faster when there are concrete examples to use as models, as then one does not have to build a totally new method from very theoretical documents. It also might be because the process of validation was compulsory for a company in the 1990s and optional before. This might influence the resources invested by the company.
5.4
To comply or not to comply
Despite the fact that GLP principles have existed for thirty years, opinions about what GLP principles mean are somewhat divided. One company thought that it was difficult to keep GLP and other studies separate, but stated that they had to be kept separate because there were different requirements for non-GLP and GLP studies. Another company stated that GLP had to be applied to all studies conducted at the test site. That was a requirement of inspectors. According to our other sources, this respondent was right. According to the GLP Handbook, compliance with all aspects of GLP principles is required to gain the status of a GLP laboratory; there is no possibility of complying with only a selection of requirements and still claiming GLP compliance. A GLP laboratory has to comply with all requirements at all times. If a GLP laboratory closes down, a special procedure concerning the archives, test data and disposal of microorganic test material must be followed (GLP Handbook). The only company that seems to work according to GLP principles only when doing a GLP study for a customer (in other cases according to the company’s own principles) seems to be more dissatisfied about the GLP system than the others. It might be necessary that the whole organization totally complies with GLP to be able to benefit from the system properly.
- 54 -
5.5
Are our results representative?
There is a risk that we have received too positive responses. Almost all our respondents were involved in the process of validation and the decision to comply, or are now working very actively with GLP high up in the hierarchy. It is possible that we would have received less positive responses if we had asked laboratory employees. However, we believe the favourable view of GLP of senior management would be more representative because they can more easily see the overall picture and the benefits of GLP to efficient operation of the whole organization.
5.6
Sanctions
In Sweden the threat of losing GLP status is enough sanction to make GLP laboratories adhere to GLP principles. As stated in the results, faults in nonclinical testing and studies can have serious financial consequences. Noncompliance and sanctions in the USA have even more severe consequences. The FDA in the USA provides information on two cases of severe defrauding of nonclinical laboratory testing (in the USA falsification of drug and food studies is a crime).
The FDA found out through its inspections that Industrial BioTest laboratories falsified its reporting of GLP studies and its laboratory work, dead animals were replaced with fresh animals without the replacement being recorded, and documentation on substances were fabricated. Three officials of IBT were found guilty of falsifying research and jailed for many years. Chemical companies that had used IBT’s products had to spend millions of dollars repeating the tests and studies to keep their products on the market. IBT no longer exists (P.A. Carlsson et al, 1998). Biometric Testing’s two Vice Presidents pleaded guilty to conspiring to falsify reports of animal tests on certain drug products to prove them harmless although the tests had not been carried out. The company went bankrupt (P.A. Carlsson et al, 1998). In Sweden such sanctions may be considered in the future, but not yet.
- 55 -
5.7
No such thing as a perfect system
Even with rigorous GLP principles for non-clinical studies and monitoring of their compliance, testing in humans can go wrong. Six male volunteers had to be rushed into intensive care in London in March 2006 immediately after being injected with a trial monoclonal antibody intended to treat chronic inflammatory conditions and leukaemia by affecting the immune system. An interim report found no evidence of human error, contamination or failure to follow GLP protocols by the German company TeGenero that had conducted the non-clinical studies and manufactured the drug. The intense reaction to the drug was unprecedented and could not have been predicted (The Times, 4 April 2006). We raised some problems in the Introduction that have risen in real life concerning the safety of pharmaceutical products. In the case in February of withdrawal of a medicine that caused liver damage as side effect, the company had fully complied with GLP principles, as had TeGenero in the other case we presented. Drug development is a complex and very long process. Statutory requirements cannot alone guarantee total safety of non-clinical studies and products developed based on the studies. Human input is crucial in quality improvement. Even with state-of-the-art equipment and calibration, human error or negligence may threaten a GLP study. All statutory requirements, regulations, quality systems, guidelines and policies require that everyone involved in the organization takes responsibility for quality and follows regulations in detail, otherwise quality cannot be improved. It is important to note that adoption of the regulatory framework is ineffective unless all personnel contribute and comply with the regulatory framework. Motivation of laboratory personnel is crucial to the success of adopting and complying with GLP principles. According to Bergman et al (1998), it is vital that the whole organization is involved in sustainable quality improvement work. However, even following the regulations into the smallest detail cannot always guarantee success with the GLP studies. All scientific studies include an element of interpretation. Choosing the right test samples and species of animals, and interpretation of data are crucial in developing science. Open and creative
- 56 -
dialogue and communication should therefore be included in the quality improvement work, in addition to compliance with all regulations. Regulations restrict scientific creativeness in conducting studies, but when drafting the master schedule of a GLP study and also in analysing the results, creativeness should be allowed to ensure that every health and safety aspect has been covered and researched. Unknown territory in, for example, monoclonal antibodies should be researched even more profoundly than known territory in drug development. Openness should be encouraged. New thoughts and ideas are created in situations when one’s mind is open to new ideas.
5.8
Analogy of the GLP concept
A key characteristic of GLP is that it is a total concept. It is not enough that a laboratory meets the required standards. All its suppliers must too. This is like how air safety affects an airline company. The company’s pilot must exactly follow instructions from air traffic control when flying his plane. He must be qualified to international standards to fly the plane, and the airline company must meet the strictest international standards to be allowed to use a European airport. The airline company’s suppliers must also meet the strictest standards. The manufacturer of its aircraft must meet very demanding standards for its aircraft to be allowed to fly. Nowadays, these are not just legal requirements, they are commercial necessities. Airline companies are linking in alliances (Finnair flights to Britain also have a British Airways flight number and British Airways flights to Finland also have a Finnair flight number). Airlines will not link with a partner airline unless they are sure the partner airline has the same highest standards of safety, otherwise its own reputation will be damaged. Airlines that cannot find partners will lose business.
- 57 -
Both also have full documentation of every single thing that happened, allowing the whole sequence of events to be analysed: the GLP laboratory in its detailed records of all procedures, the aircraft in its black box. Thankfully, they are very rarely needed because airlines and GLP laboratories maintain such high standards of quality.
5.9
Future of GLP
We conclude our analysis with a prediction. This highlights our key finding that the main reason for a company to comply with standards can change: the company begins by complying with standards to meet legal requirements, but then finds that strict compliance with standards improves overall efficiency, attracts customers and increases business. Testing laboratories and modem manufacturers have the option of complying with certain standards (GLP for the laboratories, Windows for the modem manufacturers). Nowadays, all modem manufacturers have realized the advantages of 100% Windows compatibility so compliance is universal, but still only some laboratories comply with GLP. We believe the similarities between the cases suggests in time a much larger percentage of laboratories will seek GLP status. The situation of a laboratory as regards GLP is like the situation of a manufacturer of a personal computer modem and its software. We had expected the motivation for laboratories to seek GLP status was to satisfy the legal requirement that the national testing authority had created to ensure the safety of the public who would be using medicines tested by the laboratory. Similarly, years ago the only reason for a modem to comply with certain standards was to be approved by the national testing authority (the national telephone company) and so be permitted to connect to the telephone network. Again, the national testing authority (the national telephone company) had set its standards to protect the safety of the public, because a faulty modem could be electrically to dangerous users of the telephone network.
- 58 -
However, there are now also other strong motivations for a modem manufacturer to follow very strict design guidelines and manufacturing standards, and to document them thoroughly. Customers need total confidence that a modem will automatically work with their computer. Unless they can see proof of this on the package, they will buy another brand, and the company’s business will collapse. Nowadays, this means confirmation by Microsoft or similar that the modem and software have been tested to be Windows compatible. As with GLP laboratories, there must be verification of compliance at both government authority and lower inspection levels. Also, compliance must be total. Complying with 99% of GLP regulations is useless as the laboratory will then lose GLP status. Similarly, a modem that is 99% Windows compatible will not work properly and the computer may crash. In both cases maintaining a high reputation for quality and reliability is essential. A laboratory losing GLP status would go out of business; so would a modem manufacturer if computer magazines reported problems with its latest modems. Finally, comprehensive documentation is not only essential in both cases, it is a real business advantage. Documenting compliance with GLP practices or Windows standards is essential in both cases for approval. For GLP laboratories it is also needed to be able to repeat the testing exactly; for a modem manufacturer it is essential for smoothly upgrading the modem and software to the next version of Windows. But in both cases there is also another key benefit: detailed documentation makes it much easier to preserve the knowledge of staff who have left and to bring new employees up to speed quickly. In both cases companies have found that strict compliance with standards improves efficiency so much that this more than compensates for the extra burden of complying with the standards. This is a key motivator to strive to maintain the standards. However, there is one crucial difference between these two cases. Nowadays, every modem manufacturer must be able to display proof on the box that the
- 59 -
modem is 100% Windows compatible. Nobody will buy a modem that cannot claim that. But GLP status is still optional. Some laboratories want GLP status, some do not think they need it. In view of the great similarities between these two cases, we think this will be only a difference of timing. For commercial reasons, all modem manufacturers already comply with Windows standards, but only some laboratories comply with GLP. We think as laboratories realize the advantages of GLP, more and more of them will adopt GLP, till GLP status becomes essential for commercial reasons in some cases.
5.10 Indirect benefits of GLP In our thesis we have provided direct answers to our question “How does Good Laboratory Practice improve quality” by explaining the direct benefits to quality of GLP. However, it is also clear that GLP has very important indirect benefits to quality, too. As companies and laboratories see that the strict following and detailed recording of procedures required by GLP has economic benefits through improving overall efficiency, they become even more motivated to enforce the highest standards of quality. This is positive feedback that further improves quality, benefiting the laboratory, its customers and ultimately the public, who will be using the pharmaceutical products being tested. GLP improves quality, and the improved quality benefits the laboratory’s owners, employees, customers and the public.
- 60 -
6 REFERENCES 6.1
Books
Andersen, Ib, Den uppenbara verkligheten - Val av samhällsvetenskaplig metod, 1998 Anthony, Robert, Govindarajan, Vijay, Management Control Systems, McgrawHill Education, 2003 Bergman, Bo och Klevsjö, Bengt Kvalitet från behov till användning. Studentlitteratur, 1998 Bodin, Sven, Sahlen, Tom, Sjögren, Carina, Dokumenthantering i företag och organisationer - en kvalitetsfråga, Folkrörelsernas Arkivförbund, Stockholm, 2000 Carlsson, P.A., Dent, N. J. Good Laboratory and Clinical Practices, Butterworth – Heineman Ltd, Oxford, 1998 Rosengren , Karl Erik, Arvidsson, Peter, Sociologisk metodik, 2002 Yin, Robert K., Case study research- Designs and Methods, 3rd edition, 2003
6.2 Downloadable from authorities’ and organizations’ websites Food and Drug Administration FDA, 21 CFR58.35 Eudralex, Volume 4, Medicinal Products for Human and Veterinary Use. Good Manufacturing Practices EMEA: European Medicines Agency (Europeiska läkemedelsverket) Lääketietokeskus electronic and printed publications, Helsingfors, Finland, 2005
6.2.1 Läkemedelsverkets författningssamling GMP LVFS 2001:1
- 61 -
LVFS 2001:5 LVFS 2001:8 LVFS 2003:11 LVFS 2004:6 LVFS 2004:11 GLP LVFS 1999:8
6.3
Manuals from organizations
ISO Guide 25, ISO 9000 GLP for analytical laboratories, Agilent Technologies Handbook- Good Laboratory Practice, UNDP/World Bank/WHO (not dated)
6.4
Articles
Immel, Barbara K, A Brief History of the GMPs for Pharmaceuticals, Pharmaceutic Technology, 2001
6.5
Newspapers
The Times, 4 April 2006
- 62 -
7
APPENDIX 1
INTERVIEW QUESTIONS
Adoption of GLP principles -
How long has GLP been adopted in your laboratory? Why did your laboratory apply for GLP registration? How long did the validation process take from the decision to adopt GLP principles and seek GLP status to the approval by Medical Products Agency? What quality processes or systems did your laboratory have before seeking validation according to GLP Principles? In your view, what were the key changes required in the laboratory operations in order to gain approval by Medical Products Agency?
How the GLP function today? -
What is your overall experience of conforming to Good Laboratory Practice Principles? How has your laboratory directly or indirectly benefited from GLP status? Has GLP improved quality? How? Was it easy or difficult to implement overall? Could GLP principles be utilized more effectively to benefit the company? How? Have you experienced any disadvantages from GLP registration? How could these disadvantages be overcome?
Overall quality improvement -
Any other quality systems such as ISO or GMP adopted in parallel?
Sanctions -
What would the consequences for the company be of non-compliance and a loss of GLP status?
- 63 -
