Carola Lemne Handbook for Clinical Investigators 2008

8/2/2019 Carola Lemne Handbook for Clinical Investigators 2008

1/110

Carola Lemne

Handbook forClinicalInvestigators


2/110

All rights reserved. No part of this publication may be reproduced ortransmitted in any form or by any means, electronic or mechanical,including photocopying, recording, or any information storage andretrieval system, without permission in writing from the publisher.

Art.nr 7223

Carola Lemne and Studentlitteratur 2002Translated by: Carola LemneSecond editionCover art: Pernilla Eriksson

Printed in SwedenStudentlitteratur, LundWeb-address: www.studentlitteratur.se

Printing/year 1 2 3 4 5 6 7 8 9 10 2006 05 04 03 02

eISBN10 91-44-03181-5

eISBN13 978-91-44-03181-1
http://www.studentlitteratur.se/http://www.studentlitteratur.se/


3/110

Studentlitteratur 3

Contents

Introduction 5

1 General 7

Some background facts 7Clinical trials some definitions 8Clinical trials why? 10Is there room for clinical trials within ordinary health

care? 12Good Clinical Practice 13

2 Sponsors 17The investigators responsibility 17The sponsors responsibility 19SOP Standard Operating Procedure 21The Investigators Brochure 22The clinical trial monitor 23The agreement with the sponsor 25IND studies 28

3 The carrying through 29

Before 29Preparations visits, planning 29The Study Protocol 30Ethical considerations 45The ethics committee 48Patient information and informed consent 50Approval from the regulatory authority 53Approval from the radiation protection committee or

similar 55Protecting the patients confidentiality 55


4/110

4 Studentlitteratur

Information for the staff 57The contract with the Principal 58Financial disclosure 59Insurance 60Multi-center trials 60Study documentation 63

During and after the study 65Finding suitable patients 65Delegating tasks and responsibility 66Monitoring 66Adverse events Side-effects and other events 68Audits, inspections and quality assurance 70

Carelessness and fraud in clinical trials 72Documentation and archiving 73

4 Processing results 76Data and computer processing 76Analysis of results 77Publication 78

5 Conclusion 80Overview and time perspectives 80Conclusion 83

References 85Guidelines 85Reference literature 85

Appendix I Suggestions for drawing up a protocol 86

Appendix II Glossary of common terms used in clinical research 92

Appendix III Declaration of Helsinki 101

Useful addresses: 107

Index 109


5/110

Introduction

Studentlitteratur 5

Introduction

The term clinical trials is used for all the prospective studies of drugs or medical devices that are intended for investigating anddemonstrating the efcacy and/or safety of drug therapies and

other methods of medical treatment. The carrying out of clinicaltrials has become increasingly formalized. Governmental authori-ties have passed legislation, ordinances and guidelines; pharmaceu-tical companies have introduced formal Standing Operating Pro-cedures; and most recently international indeed, global guidelines for Good Clinical Practice have been established.This has occurred not simply due to demands from governmentalauthorities, but mainly because all the parties involved have cometo realize that conclusive and high-quality study results can only beachieved through careful and consistent planning and execution.The discipline necessary for clinical drug trials is just as necessaryand useful for all other types of clinical research projects.This handbook is intended as an aid for investigators (that is, cer-tied physicians and dentists) and others, prior to participating inany type of clinical trial, although it emphasizes drug trials.


6/110

Introduction

6 Studentlitteratur


7/110

General

Studentlitteratur 7

1 General

Some background factsThroughout the world thousands of drug trials are performed everyyear. Even if a large number of these concern new strengths or indi-cations for existing drugs, or even generic preparations, a consider-able number of studies concern completely new substances. Clini-cal trials have been performed to varying degrees for each new drug,preparation and indication. For drugs intended for long-term use incommonly occurring diseases, such as hypertension, it is notuncommon today for several thousands of patients, many of whomhave been treated for a year or more, to be included in the databasesubmitted for registration.

This means that the number of clinical trials has also steadilyincreased, and this has been observed all over Europe, the US andalso increasingly in other parts of the world. As an example therewere slightly over 1 000 applications to start clinical trials submit-ted to the Swedish Medical Products Agency in 1982. In 1990 thecorresponding number was nearly 1 770 and in the year 2000 it hasstabilized at around 2 200. Of these applications several could con-cern the same study; that is, more than one physician was involvedin the same trial. This type of study, a multi-center trial, has becomeincreasingly common; applications in Sweden (the year 2000)involved around 500 different studies, of which about 200 wereinternational. There are no exact statistics on how many patientswere included in all these trials, but an approximate estimate indi-cates that in Sweden alone more than 50 000 patients (and healthyvolunteers) participated in clinical studies in 1998!

All medical disciplines are represented in clinical trials. The larg-est number of trials (just over 25 %) involve internal medicine. Thisis followed by outpatient care, where an increasing number of trials


8/110

Chapter 1

8 Studentlitteratur

are performed, and next comes general surgery. Trials are being runin private practice and in company employee clinics. Since thebeginning of 1995, the same standards have begun to be placed onmedical devices as those already existing for drugs. This means thatmedical/technical products, such as infusion pumps, must be testedand documented in a similar fashion to drugs; that is, by perform-ing clinical trials. In short, each physician and nurse can thereforecount on encountering this phenomenon in his or her workplacesooner or later.

Clinical trials some denitionsThe development of new drugs follows a general pattern, albeitwith some variation. First comes the pre-clinical stage, beginning inthe laboratory when possible new drugs are developed, these daysoften based on advanced knowledge about the bodys cellular func-tions combined with advanced chemical techniques for tailoringnew substances. Promising substances from in vitro experimentsthen enter an extensive animal-studies phase. This is partly anattempt to investigate the substances possible effects, and partly tostudy the safety prole. Should the desired effects occur, therequired toxicity studies are begun in order eventually to test thesubstance in humans. There are now harmonized guidelines forwhat to do and in certain cases even in what sequence.

Once the pre-clinical program has advanced so far that there is anunderstanding about the substances properties both positive andnegative applications are submitted to regulatory authorities andto ethical committees for carrying out the rst tests on humans;

that is, to begin clinical trials.Clinical trials are dened as a systematic administration of drugs

(this includes common drugs, radioactive drugs, natural and relatedremedies, and some preparations for external application), or use of a medical device for the purpose of discovering or conrming theirefcacy, patterns of adverse effects, pharmacokinetics, etc.

Clinical drug trials are usually classied into four different mainphases, which have different purposes and characteristics.

Phase I comprises the rst trials on humans, as well as subsequenttraditional clinical pharmacology studies. These early studies are


9/110

General

Studentlitteratur 9

mostly performed on healthy volunteers and are mainly intended toestablish the tolerable dosage interval for humans. The investiga-tions also study if and how the drug is absorbed, distributed, meta-bolized and excreted from the body. If possible, an attempt is alsomade to measure any effect, but since healthy volunteers are themost common participants in Phase I studies, this isnt always pos-sible. One exception from using healthy volunteers in Phase I maybe with trials involving cytotoxic drugs. Naturally, it is not advisableto administer these drugs to healthy volunteers instead an oppositeapproach is used, i.e. the drug is administered to gravely ill patientswho have been previously treated but without success. During thecourse of a drug development program several other phase I studies

are performed to investigate the drugs effects in special patientcategories, like for instance patients with kidney or liver disease.If the Phase I studies show that the substance may be given to

humans in reasonable doses and without overly troubling adversereactions, Phase II generally follows. These are the rst studies onpatients having the disease intended for treatment and are meantto discover the minimum and maximum effective dose. The pur-pose is to extensively investigate how the drug behaves in thepatients bodies, which effects can be measured, how these effectsrelate to a given dose, and so on. As mentioned above, these studiesare performed on patients having the disease of interest, and aneffort is generally made to enroll as homogenous patients as poss-ible. Because there is uncertainty about the drugs possible effects, itis desirable to hold the number of possible confounding factors to aminimum. Consequently, it is highly desirable to have as healthypatients as possible (apart from the disease of interest) and whoresemble each other as much as possible. In this way it is easier to

discern the possible benecial effect of the drug itself. Once Phase IIis completed, there should be a good understanding about thedrugs effects, suitable doses, and a preliminary understandingabout the adverse reaction prole.

Phase III follows. Phase III studies are designed to document thedrugs effects and adverse reaction pattern in normal patients.These studies include patients of various ages, patients with concur-rent diseases, patients who are taking concurrent medication etc.Phase III is often divided into Phase IIIa and Phase IIIb. Phase IIIaare those studies that are to be included in the registration docu-


10/110

Chapter 1

10 Studentlitteratur

mentation, while Phase IIIb are studies that are performed prior toregistration but are not included in the registration application. Forexample, these can be studies performed in order to conform to thetreatment practices of different countries or to investigate specialgroups of patients. Once the Phase III program is complete, it is pos-sible to begin talking about the drugs properties and common side-effects in a relatively representative cross-section of patients withsome certainty; and the necessary documentation for getting thedrug registered should now be available.

However, the drugs development does not stop once the drug isregistered and permission to market is obtained. There often fol-lows several years of research in Phase IV studies. These are studies

in which the registered drug is used for the registered indicationsusing a registered dose and formulation. For example these studiescan be large studies designed to discover rare adverse reactions(called Post-Marketing Surveillance Studies), they can be studiesthat are designed to discover more mechanistic properties, or theycan be long term mortality studies, such as those large cardiovascu-lar studies that investigate the effect of anti-hypertensive drugs inpreventing myocardial infarction.

It may be worth pointing out that if a registered drug is beinginvestigated for another indication than the registered one, orusing a new dosage or formulation, it is actually a Phase III, or evena Phase II study.

Another important point to remember is that the different phasesdescribed above seldom follow each other in strict chronologicalorder. After the rst Phase I studies to establish maximum tolerateddose it is not unusual to go directly into limited Phase II studies on

a smaller number of patients. The intention is to establish proof of concept as early as possible in the development of a drug. If suchproof is obtained the development continues with additional PhaseI and Phase II studies in parallel.

Clinical trials why?Clinical trials are required in order to register a drug, and also thesedays to obtain permission to sell various medical devices. They have


11/110

General

Studentlitteratur 11

the purpose of documenting the efcacy and safety of drugs andtreatment methods that may some day be administered to thou-sands of patients. Physicians have a responsibility for their patientshealth and included in this is the responsibility that the treatmentmethods used have a documented efcacy and safety. Participationin evaluating these new therapies and treatment methods oughtthus to be an integrated part of all medical practice.

However participating in clinical trials is sometimes boring,often puts ones patience to the test, and always involves more workthan initially imagined. What makes a doctor become an investi-gator?

Being involved in a clinical trial often gives pure professional sat-

isfaction. The physician gains more in-depth knowledge in the cur-rent topic and an opportunity to establish contact with other col-leagues interested in the eld. He/she keeps better up-to-date withdevelopments in the eld. The publication of the results can lead tovaluable merits and the opportunity to present the results at con-gresses that he/she might possibly not have been otherwise able toattend. The trial the study can often be a stimulating interrup-tion in the daily routine and often creates the opportunity to havea more in-depth contact with the patients involved. In additionthere are happy side-effects such as well-organized and produc-tive project meetings during the study and naturally the eco-nomic compensation for the extra work put in.

Regardless of the reasons put forth for participating in a clinicaltrial, the most important question remains Do I personally feelthat this study could be of use to the patients or for understandingthe medical problem under consideration?

If the answer is no it is perhaps wisest to refrain. Clinical trials

entail so much effort in themselves that exclusively economicmotives are often insufcient for keeping alive the personal motiva-tion that is needed for obtaining qualitatively good results. Clinicaltrials require thoroughness to all details throughout, and the temp-tation to take a shortcut can be substantial if there is no genuineinterest in the results.

There is often a sponsor behind a clinical trial. A sponsor may bea pharmaceutical company, a research fund or some other unit thathas an interest in having a specic trial performed, and which sup-ports the trial in various ways, e.g. economically. There are various


12/110

Chapter 1


reasons why a sponsor wants to conduct a clinical trial. It is alwaysa question of evaluating some form of treatment, but the studiesmay have different formats depending on the sponsor.

Clinical trials sponsored by pharmaceutical companies are oftendesigned to produce the necessary documentation for demonstrat-ing the drugs efcacy and safety, and are required in order to getthe drug registered. In addition, they often aim to expand theknowledge of an already registered drug such as new indications,for example.

Clinical trials sponsored by authorities, funds or academic insti-tutions can focus on evaluating treatment that lack commercialpromise, for instance because the indication is too small (orphan

drug), or methods that have been long used, but have never beenthoroughly investigated before.Regardless of the type of clinical trial in question, it is the doctors

commitment that matters. It is the physician who determineswhether the study may be practical to perform at the clinic. Thephysician decides which patients are suitable for the study, and thephysician is a guarantor for the patients welfare and integrity inthe clinical research. Finally, it is also the physician who, with his/her own name, carries the responsibility for the reliability of theresults.

Is there room for clinical trials within ordinaryhealth care?

There is often a heavy workload in todays health care, and it mayfeel impossible to nd the time for clinical research and clinical tri-als. While it is often ofcially said that continuous evaluation andverication of new treatments are a part of a physicians normalduties, the reality is that there is rarely time set aside for this. Parti-cipation in clinical trials requires careful advance preparation aswell as consultation with superiors, colleagues and other coworkerswho may become involved.

Talking with patients in connection to clinical trials usually takesmore time than required when doing regular rounds or visits, espe-


13/110

General


cially in the beginning of the study when the project is to beexplained and walked through. The time invested initially in care-fully explaining the entire project is later paid back by having moreenthusiastic and careful patients who more rarely withdraw fromthe study. Thus, this is time well spent, and it is important thereforeto plan for this time right from the start.

In connection with their examinations, the patient often hasmany more questions than usual for the nurses and staff, and thusadditional time may need to be reserved to cover this. Speciallytrained and hired nurses with research experience are invaluable inthis respect!

Participating patients may experience suddenly appearing com-

plaints, or get other diseases that may make it necessary to discussthe treatment with the physician responsible for the study. Thus, itis necessary to be prepared for the fact that additional work mayoccur outside of normal working hours.

In addition, the additional time required for updating patientcharts, meeting sponsors and any other investigators, complement-ing missing data and a vast number of small but time-consumingdetails very often can only be attended to during leisure time.

However, careful review and planning can make it possible tohandle these demands during the physicians allotted workinghours and thereby allowing not only the study to be performed,but also assuring high quality results.

Good Clinical Practice

Good Clinical Practice (GCP) is a term coined for labeling a collec-tion of recommendations, rules and guidelines about how goodclinical research ought to be performed. Thus, it is really a matter of Good Clinical Research (or Trial) Practice. The phenomenon isexceptionally international. The FDA (Food and Drug Administra-tion USA) was the rst to issue these kind of rules and guidelinesback in the 1960s. These guidelines are incorporated in US federallaw, and violation of these can lead to prosecution and althoughit rarely occurs, it remains fully possible. In England the pharma-ceutical industry (ABPMI) formulated its own guidelines, which


14/110

Chapter 1


were voluntary. The French health care ministry issued guidelinesin 1987 that later were incorporated into law and encompassedboth Good Clinical Practice and a system for verifying that compli-ance to the law actually occurred. The European Union (EU) estab-lished a commission that worked out guidelines applicable to theentire EU and which were rst issued in May, 1990. These guide-lines were issued as directives in April 2000; and thus became man-datory. Many nations inside and outside of EU such as Swedenand Hungary have however already chosen to incorporate GCPinto law. Japan has produced similar guidelines and WHO has alsoreleased guidelines intended for use outside the USA, Europe and

Japan.

Work began several years ago to further harmonize the US, Euro-pean and Japanese guidelines. This work is being carried out withinthe framework of a large international cooperative forum under thecollective name of International Conference of Harmonization(ICH). The aim is to arrive at an agreement about what documenta-tion is needed for drugs (and to a certain degree also medicaldevices), both during the development phase in order to get thedrug registered, and for ongoing monitoring of drugs on the mar-ket. In this way research need not be unnecessarily duplicated, anda universal and high standard will be obtained for all nations. Aconsensus has already been reached for some aspects, howeverthere remains much work before drug research and developmentcan be carried out and documented on a truly global basis. How-ever, one of the aspects that has come far along in its developmentis specically GCP. The ICH guidelines for GCP, ICH-6, has com-pleted the entire process, were made ofcial in May 1996, and areapplicable to all clinical trials begun after January 1997 within

ICHs jurisdiction, i.e. the US, EU and Japan. The new EU directiveis built upon these guidelines, and work has begun in Japan torevise Japanese law in order to conform to ICH-6. The US authori-ties have previously made several changes, and at the time of thiswriting no additional changes have been notied. Given this abun-dance of guidelines, it is easy to wonder which ones actually apply.In the future it is intended that the ICH GCP guidelines are to bethe basis for fundamental legislation such as EU directives whichin turn will be mirrored in each countrys own legislation. Minorlocal differences can sometimes appear, as they do already. Should


15/110

General


differences between the two remain, it is the national legislationthat applies, with the ICH GCP guidelines a solid second.

The same principles apply to the clinical testing of medicaldevices as for clinical drug trials. This is regulated within the EU bytwo directives (93/42/EEC, MDD; 90/385/EEC, AIMD) Thesedescribe concepts so similar to the GCP used in clinical drug trialsthat the concept of GCP can be applied more or less equally to suchactivities as drug trials and evaluations of treatment methods ormedical devices of various kinds. However, there are some differ-ences concerning documentation, time allocated for ofcialprocessing, etc. These differences will be specically discussed inthe appropriate chapters below if nothing specic is stated, the

described procedures are applicable for both drugs and medicaldevices.GCP has several purposes. The two principal ones are 1) protec-

tion of the patients own self-interest, i.e. safety and integrity, andto 2) establish that clinical research is carried out correctly usinghigh standards of quality and in such manner that it produces reli-able data that may be veried afterwards. The patients own interestis primary, and their safety and integrity are (among other actions)protected by emphasizing the ethics committees role and makingthem strictly obligatory. The content and quality of the informa-tion given to the patients is of very central signicance. In additionthere are detailed instructions about how adverse reactions are to becollected and reported. Both quality and verication functions areensured via instructions on how the study project is to be set up,how data is to be gathered, veried regularly, and then stored forany later inspection.

Since the ICH collaboration, the various rules are steadily

approaching harmonization. The US and EU requirements havepreviously differed on a number of points, but these are now farfewer in number and usually have no signicant practical implica-tions. However, in some cases, special requirements can appear thatnecessitate compliance with all the FDA directions. For example,this may occur when the drug being tested is manufactured andexported directly from the USA. According to US federal legislation,no drug may be exported for investigative use if US-federal regula-tions are not adhered to when the drug is later used. Considerablediscussion may arise in these cases, since local legislation and direc-


16/110

Chapter 1


tives are not always in complete agreement (see IND studies).How this is to be resolved should be discussed with the sponsor ona case-to-case basis.


17/110

Sponsors


2 Sponsors

The investigators responsibilityAn investigator is the medically responsible person (doctor or den-tist) who takes a personal responsibility for carrying out the studyaccording to the applicable directives and who ensures that the cor-rect therapy is administered to all patients enrolled in the project.

Just what is it that an investigator takes responsibility for in con-nection with a study or clinical drug trial?

First and foremost it is the responsibility for both the patientswell-being and that the treatment being offered is the most appro-priate in any given case. In order to ensure this, there are a greatmany things that are the investigators lot throughout a study:

To begin with, the investigator must be certied and well-meritedwithin the eld being studied (this applies to the person bearingthe ultimate responsibility at each participating clinic; othercoworkers at the clinic are not required to have the same degreeof specialist knowledge). The investigator must have goodresearch skills and be familiar with all that is generally requiredin clinical trials (like GCP) and especially with the drug beinginvestigated. The investigator is also responsible for ensuring

that the resources needed actually exist. Investigators are oftenrequested to prove their suitability via attested curriculum vitae. The investigator must be knowledgeable of the drug or medical

device under investigation (see Investigators Brochure) andhave good knowledge about the contents of the trial protocol. Itis formally the investigators responsibility to ensure that the trialprotocol holds all the necessary information for designing andcarrying out the study (see The Study Protocol).

The investigator must ensure that the personnel and equipmentare adequate for taking care of the patients, and that the person-


18/110

Chapter 2


nel are well informed about the trial and procedures to be fol-lowed in any acute situation. The investigator must also informand receive approval from the head of clinic, or the equivalent if applicable, and make sure that no other projects are going onthat can disturb the course of the study.

One of the investigators most important responsibilities is of course the information to patients and that the enrollment (orinclusion) of patients is carried out in an ethically correct man-ner (see Patient information and informed consent). Thepatients personal physician, or the equivalent, must also beinformed that the patient is participating in the study if thepatient is in agreement with this.

It is also the investigators responsibility to ensure that approvalhas been granted by both the ethics committee and the regula-tory authority, if applicable (see The ethics committee andApproval from the regulatory authority).

Once the study has begun, the investigator is responsible forensuring that the protocol is followed, that drugs are distributedand collected correctly, and that the right data is collected. Theinvestigator must also be available for the regular visits that asponsor makes during the trial (see Monitoring).

The investigator is of course always responsible for the patientssafety and that all decisions affecting their medical treatment bemade only by the investigator. One of the most important tasksfor an investigator is thus to be responsible for a thorough andcorrect reporting of adverse events. The investigator is alsoresponsible for keeping informed about developments in theeld that the trial covers, and to stay familiarized with any newinformation concerning any drug being used.

After the studys completion, the investigator is responsible forensuring that the patients will continue to receive proper care.

Itemized this way, the duties may seem insurmountable, but in real-ity many of these responsibilities are already an integrated part of properly functioning health care. Concerning those details thatspecically concerns the study, one can often get advice and assist-ance from a serious sponsor.


19/110

Sponsors


The sponsors responsibilityThe sponsor is an individual, an institution or a company that ini-tiates, organizes and monitors a clinical trial and/or arranges the

nancing. Most commonly, it is a pharmaceutical company or amanufacturer of a medical device who is the sponsor, but otherorganizations or even private individuals may also be the sponsors.The denition of a sponsor is thus rather broad and to be a formalsponsor carries a considerable responsibility for the conduct of thetrial. It is of course possible for companies and institutions todonate general research funds without becoming formal sponsorsof all clinical trials carried out. The text to follow in this chaptermainly describes the responsibilities inherent to companies actingas sponsors of clinical trials or to the investigator if the study has noexternal sponsor.

The sponsor is responsible for that all information about the drugor medical device, its manufacture, packaging and accompanyingdocumentation are correct. The sponsor is also responsible for reg-ularly updating the investigator about new data regarding the drugor medical device being investigated (see Investigators Brochure).It is also the sponsors responsibility to select investigators for the

study and to assure that these are competent and willing to adhereto GCP, etc.

The investigator and sponsor share mutual responsibility forensuring that approval for the study has been granted by both theethics committee and the applicable regulatory authority (see Theethics committee and Approval from the regulatory authority).The sponsor is further obligated for ensuring itself that any permitsneeded to gain direct access to the patients medical records areobtained (see Protecting the patients condentiality).

The sponsor is required to have a written, documented system(see SOP) for monitoring and following up clinical trials. Thiscomprises having specially selected and trained persons clinicalmonitors who are the sponsors representatives and who workclosely with the investigators. These have specic duties before,during and after a study (see The clinical monitor).

The sponsor is responsible for all drugs involved as concerns theirmanufacture, packaging, quality and labeling. Routines arerequired for emergency situations in which the treatment codes for


20/110

Chapter 2


individual patients may need to be broken, and the sponsor isresponsible for having someone available around the clock who cananswer questions in such situations. The sponsor must also savesamples of each drug used in a trial in case there is a later need toperform control analyses. Concerning medical devices the manu-facturer is correspondingly responsible for manufacturing, quality,performance, etc.

Concerning adverse events, the sponsor is responsible for report-ing all such to the ethics committee and/or regulatory authorityaccording to a well-dened system (see Adverse events side-effects and others).

The sponsor is responsible for gathering all data from all trials

and must also have an internal verication system for monitoringthe quality of the trials (see Audits, inspections and quality assur-ance).

Finally, the sponsor, as well as the investigator, is required toarchive all documents and collected data. The ICH GCP guidelinesdescribe a relatively intricate procedure in which the requiredarchiving time is based on when the product was registered in theparticipating ICH countries. This results in the investigator havingto hang on to the information until conrmation arrives from thesponsor that it is no longer needed. Concerning data from drugstudies, as a result of the international product liability laws, it isalso required that the company continues to save much of thestudy documentation as long as the drug is being sold anywhere inthe world, and after that, for an additional number of years. Thedocuments to be stored are those necessary for reconstructing thestudy afterwards; for example the original data. There is a list in theICH GCP guidelines covering which documents must be stored.

It is now relatively common that a company out-sources theactual performance of the trial to independent companies knownas Contract Research Organizations (CROs). However, the sponsorretains responsibility for the study even if it is not their own per-sonnel who carries out each stage. All such duties that are con-tracted out to a CRO must be done so in writing.

As mentioned above, it is important to realize that for any studylacking an outside sponsor, it is the investigator who assumes allresponsibility that would otherwise be the responsibility of thesponsor, such as quality control etc. (see Monitoring).


21/110

Sponsors


SOP Standard Operating ProcedureCompanies that carry out clinical trials are required to have aninternal, established and written procedure for how these are to be

carried out a Standard Operating Procedure (SOP). This SOPshould specify how all parts of a clinical trial are to be performedand monitored from the sponsors point of view. SOPs have the aimto satisfy the high standards of Good Clinical Practice and some-times also to be a practical handbook for the sponsors personnel.

A SOP describes both what the company is required to do andwhat the company requires of the investigator. This should not beviewed as something negative, instead it is a guarantee that thecompany wants to keep a high standard in its clinical trials. TheSOPs are constructed to adhere to GCP and the requirements of various authorities. Because clinical research is an exceptionallyinternational phenomenon, and a SOP is written to be applicablethroughout the world, some of the requirements may sometimesseem strange from a local perspective. The areas most often result-ing in discussions between sponsor and investigator include thepatient and the patients integrity/secrecy rights (these two sectionswill be discussed in greater detail later in this handbook). In an

eagerness to satisfy all conceivable requirements concerning docu-mentation, some companies can go too far in their demands onhow things are to be documented etc. Should any requirementseem unreasonably troublesome, the investigator may be wise toverify personally that the requested item is actually specied in thevarious GCP guidelines sometimes doing so may result in avoid-ing administrative routines that makes the study particularly dif-cult to carry out without contributing any additional quality.

In other words, prior to undertaking a trial it may be wise to askthe sponsor if they have a Standard Operating Procedure which isrequired of them and if there are any specic demands, beyondthat of GCP, that these SOPs poses on the investigator and how thetrial is to be performed. Since the sponsors responsibilities includeassuring itself that the investigator is aware of which requirementsthe GCP and other guidelines place on the performance of thestudy there ought to be many occasions to discuss this with thesponsor, during the preparations for the stydy.


22/110

Chapter 2


The Investigators BrochureIn clinical trials of drugs or medical devices it is the sponsors obli-gation to provide complete information about the drug or device to

the participating physicians. This is usually done using an Investi-gators Brochure. It should not be older than one year, and is tocontain information about the substances composition and prop-erties, results from animal studies performed for pharmacology,pharmacokinetic and toxicology studies and, depending on howlong the drug is supposed to be administered, carcinogenic and ter-atogenic studies. The results of previous studies in humans mustalso be reviewed. It commonly contains a short summary of thecompounds properties, a description of the reasons for developingthis specic kind of drug and an evaluation of safety aspects. If pre-vious studies have led to suspicions that any particular type of adverse reactions might arise, this should be stated. The ICH GCPguidelines contains relatively detailed instructions about whatshould be included in an Investigators Brochure.

Concerning medical devices, the contents take another form,naturally. For example, technical performance must be stated aswell as which standards the device fullls. Also required are sections

on usage instructions, instructions for installation, maintenance,cleaning etc. If the device has direct contact with the body, or withsome other device or drug that itself has direct contact with thebody, the material comprising the device and the process of itsmanufacture must be described in order to be able to determine if there is any risk for unfavorable biological reactions.

An Investigators Brochure is also required for drugs or devicesthat are already on the market, although the brochure is often con-siderably shorter. For example, for a drug it may sufce to simplyreproduce the SmPC (Summary of Product Characteristics), PackageInsert or similar, plus a summary of the experience and theoriesthat led to consideration of initiating a new test.

It is the investigators responsibility to read and become familiarwith the Investigators Brochure or its equivalent.

It is the sponsors duty to keep the investigator informed aboutnewly received results, especially those concerning safety aspects,and it is the investigators duty to keep himself up-to-date with thisinformation. Concerning protracted studies it is therefore impor-


23/110

Sponsors


tant that the Investigators Brochure or its equivalent is updated atleast once yearly. Those Investigators Brochures that were applica-ble during a study must be archived by the sponsor, and the inves-tigators documentation (study le) is to contain the most recentrevision, along with a list of which other revisions were applicableduring the study (see Documentation and archiving).

The information found in the Investigators Brochure is con-dential and may not be disseminated without the sponsors permis-sion. In general the sponsor probably has no objection to spreadingthe clinical data about the drug, however the sponsor may want tokeep technical data and results such as data from animal studiesand pharmacological studies condential, at least until the drug is

registered. There are thus good reasons for the sponsor to desirekeeping the clinical information condential for a given timeperiod, and this is something that must be respected, of course.Before the Investigators Brochure is handed over, the investigator isoften requested to sign a special secrecy agreement, in which isstated how the information is to be handled by the investigator.This is often especially emphasized and detailed concerning medi-cal devices.

The clinical trial monitorThe sponsor employs a person who is responsible for the regularcontacts with the investigator, and is called the clinical trial moni-tor, clinical research associate, manager or executive or one of sev-eral other titles. This person may have a diverse background, oftenmedical or pharmaceutical training, and is required to have special

training for clinical trials, with an education that is suitable to thecurrent studys characteristics. Broadly dened, it is the clinicalmonitors job to see to it that the trial is carried out and reportedaccording to the specications stated in the protocol, while main-taining protection for the patients integrity and safety.

The clinical monitor has a number of duties before, during andafter the study.

Prior to the study , it is the monitors responsibility to ensure that theinvestigator and other personnel have the necessary skills. In addi-


24/110

Chapter 2


tion, the monitor must also make sure that the investigator hasstudied the existing information concerning the drug or equip-ment, is well-informed about the protocols content and under-stands all commitments prior to undertaking the study, such asthose concerning GCP, for example. Also included in the task is anevaluation of the available resources and patients at the clinic.Procedures for the drugs or devices handling needs to be reviewed,as well as when and how the patient code may be broken. Themonitor goes through the protocol and the Case Report Formtogether with the investigator and also ensures that other personnelinvolved are informed. The monitor also needs to make sure thatpermits are applied for and received from ethics committees and

applicable regulatory authorities. During the study the clinical monitor must keep in regular contactwith the investigator and make regular visits. During these visitsthey discuss the progress of the study how many patients havebeen included, withdrawn, etc. The monitor checks to make surethat the patients consent has been given before beginning any of the study procedures. In addition, each Case Report Form isreviewed not only to make sure that the protocol is followed, but

also to discover any mistakes or overlooked information. Thisongoing monitoring considerably improves the possibility of hav-ing a correct and complete database at the end of the study. Theclinical monitor also performs random tests against the originaldata; for example, by controlling a given category of information inthe original patient charts, and then comparing it with the infor-mation in the Case Report Form (also called Source Data Verica-tion, SDV). The clinical monitor also checks whether the drug ormedical device is being properly handled. It is the duty of the clin-ical monitor to inform the investigator about new data concerningthe drug in question. The clinical monitor also regularly checkswhether ongoing and required resources are still available, and thatall staff concerned are properly informed about the study, etc.

The monitor must document all contact with the investigator,whether this takes place via a visit, written correspondence or tele-phone conversation. In addition, there must be a list covering themonitors visits, preferably countersigned by the investigator or

other co-worker. Should any signicant decision be arrived at dur-


25/110

Sponsors


ing or following the visit, these too must be documented throughwritten correspondence.

After the end of the study , the clinical monitor gathers all Case Report

Forms, remaining test drug, code envelopes etc., and veries that allnecessary information is archived in a satisfactory manner by theinvestigator, and ensures that a report covering the study is writtento be sent to the ethics committee and/or applicable authority.

Thus, the investigator and the clinical monitor are going to getwell-acquainted with each other. Should the person in question notappear to be suitably compatible, the sponsor can of course alwaysbe approached and asked if another monitor could be appointedinstead.

The agreement with the sponsorThe agreement between a sponsor and investigator will, of course,vary from case to case depending on the details of each study.Sometimes the study may be entirely initiated by the investigatoralone, and the sponsor only contributes with certain nancial sup-port, usually in the form of an independent research grant. It is nor-mally wise to have a contract concerning these independentgrants, clearly stating who has the GCP responsibility for monitor-ing, reporting adverse events etc. This responsibility normally fallson the investigator when a company has had very minor involve-ment, but the sponsor donating the grant usually expects a reportabout how the research is progressing, most often on a yearly basis,and usually wants a report on any serious adverse events that may

occur. It is important to remember that in this type of independentresearch, it is the investigator who bears the full responsibility formonitoring.

Another situation arises in cases of more conventional clinical tri-als. The sponsor, usually a pharmaceutical company, a companymanufacturing medical devices, or the like, has a vested interest inthe results, and accordingly has itself taken the initiative in con-ducting the study. Even if the investigators prime motivation forparticipating in the study should be an interest in the study itself,the work that the investigator and others contribute to the study


26/110

Chapter 2


should, of course, be compensated. The details of this compensa-tion should be carefully contracted in writing before the studystarts. Even though oral agreements are just as binding as writtenones in most legislation, it may be difcult afterwards to provewhat was really said should conict arise. This is specially empha-sized in the ICH GCP guidelines, via a statement that such writtenagreement between the sponsor and the investigator or institution,must exist prior to study start.

Issues that ought to be discussed are the size of the compensa-tion, the applicable rules for its payment, and when it is to be paid.

The size of the compensation can vary enormously depending onthe nature of the trial. Long and complicated trials are generally

compensated with higher amounts than short and simple studies.For investigators and other personnel working as employees of ahospital or institution, there might need to be an agreement con-cerning how much of the study may be performed during ordinaryworking hours, and how much has to take place outside these hours.In an increasing number of countries a special contract is requiredwith the responsible employer for the portion taking place duringordinary working hours (see The contract with the Principal).

In addition to those details that are conducted during ordinaryworking hours, there is almost always some work that the investiga-tor, and also perhaps even the nurse(s) need to attend to outside of ofce hours. This can include lling in the Case Report Forms, par-ticipating in meetings, etc. Compensation for this work is generallypaid directly to the investigator or nurse(s) in various ways. Accord-ing to local legislation, this may be seen as a spare-time occupation,and each employer may have different provisions concerning this.It may therefore be wise for the prospective investigator to talk with

the employer about any possible duty to report such information tothe employer.

Another duty for reporting this information has become a hotitem of discussion in the last few years; namely whether this com-pensation must be reported to the ethics committee in connectionwith the application for conducting a trial. This is currently not thecase in most countries, nor do the ICH GCP guidelines so require.However, the new EU directive states that the ethics committeeshould consider the nancial compensation to investigators as wellas trial subjects. It is yet early days of the new directive and it is


27/110

Sponsors


uncertain how this will be interpreted and/or implemented in thevarious member states, and if so, how it will effect trial applications.

Once there is an agreement on the amount of compensation,there remain several topics to be discussed.

One should, for example, specify the basis on which the compen-sation is to be calculated. The compensation is normally related tothe number of participating patients, and to whether all thepatients can be evaluated or not. The study protocol or nancialagreement ought to state which patients are evaluable for a pro-spective analysis, and this can accordingly be used as the basis forcalculating the amount of compensation.

Finally, there must be an agreement on when the compensation

is to be paid. For short studies, it may be entirely adequate that theentire amount be paid at the conclusion of the study. For longerstudies, the compensation is usually paid out in increments, andrules for this should be specied. The nal payment is generally notpaid until all patients are completed, all data collected and thereport written.

It has become increasingly common that companies have printedstandard contracts covering these aspects. Also included are provi-sos on what is to take place should the trial not proceed asintended, or if, due to newly available data, the study needs to bechanged or canceled altogether. This sort of contract may some-times feel overly formal, but it provides security to all partiesinvolved should things start falling apart. Even if the contractualwordings may feel unnecessarily bureaucratic at times, this kind of contract is warmly recommended. In many countries it has becomemandatory to show these type of contracts to the investigatorsmain employer.

It is not entirely uncommon for the investigator to agree to par-ticipate in a clinical study without ever having discussed the eco-nomic issues. While this may speak well of the investigators rea-sons for participating in the study, it is not at all suspect todiscuss pecuniary compensation for ones work, and an open dis-cussion before can avoid many conicts later. The relationsbetween the investigator and the company should be one of mutualrespect for each others contributions and roles in developing newdrugs, and a professional discussion of monetary matters is de-nitely a part of this relation.


28/110

Chapter 2


IND studiesIND refers to Investigational New Drug and is a US concept thatcould be increasingly affecting non-US investigators. When a drug

is to be tested in the US, the available pre-clinical and clinical doc-umentation is sent to the FDA as a IND-application. In the appli-cation the company reviews the intended study program for docu-menting the drug, and describes specic main studies that areplanned. The specic trials that the FDA authorizes via its approvalof this application must then be carried out according to US federallaw, irrespective of which country they are performed in. Especiallystrict review occurs for drugs that are tested, manufactured andexported abroad directly from the US. That is, these applicationswill not be approved unless the doctor signing the application com-mits himself to following the US federal regulations!

On such occasions, a FDA 1572 form will have to be lled in. Thisdocument commits the undersigned to follow US federal law onhow clinical trials are to be performed. Some of these laws may con-ict with local national law and guidelines the most commonconicts concern composition of ethics committees, some aspectsof study monitoring, and the wording of the patient information. It

is wise to discuss with the sponsor in advance if this is the case inones own country and if so, how these matters are to be resolved. Ithas lately become accepted by the FDA that non US investigatorssubstitute a letter indicating that they will comply with ICH GCPguidelines, however, this should be agreed with the FDA in advancefor the study in question.


29/110

The carrying through


3 The carrying through

BeforePreparations visits, planningBefore the investigator and the sponsor reach an agreement aboutstarting a clinical trial, the sponsor, often represented by the clini-cal monitor (see The clinical monitor), pays a visit (often several)to the clinic. This is an excellent opportunity to discuss the proto-col in more detail, but many other matters will also be discussed.

The sponsor will want to know more about the participating phy-sicians qualications and experience with clinical trials or otherresearch projects. Other aspects sponsors are often interested in

include: are there resources available in the form of time and staff to carry

out the project? how does the investigator evaluate the proposed methods to be

used in the study and are those methods routinely used? are there enough patients for inclusion? are other projects underway at the center that can hinder the

proposed trial?

will the timetable hold?

The investigator in turn ought to take the opportunity to ask ques-tions like:

How does the sponsor intend to monitor the study once under-way? (more on this topic further into the handbook)

Who will be analyzing the trials data and how will they be pub-lished?


30/110

Chapter 3


If applicable, how will the sponsors economic support be paid and when?

The rst visit(s) are usually spent in wide-ranging discussions aboutgeneral topics. Once both partners are in agreement that a clinicaltrial will be started, the sponsors visits will take on a more practicalnature. Agreements are made about how to inform the staff,whether any specic training is required, how the patients chartsare to be recorded, how often the sponsor is to visit, and so on.

These meetings provide the investigator a good opportunity toget to know the sponsors personnel who will be working on thestudy. It also provides an opportunity to evaluate the sponsors abil-

ity and attitude about running clinical trials a very importantissue.

The Study ProtocolThe protocol the trials overall plan must contain all the infor-mation needed for carrying out the study.

A proposed protocol from the sponsor is often available at anearly phase, but since it is the protocol that determines how thestudy is performed, it is vital that an opportunity exists for discuss-ing the nal draft in detail. Both practical and scientic issuesshould be reviewed. A well-designed protocol is the foundation of awell-performed study, and the time spent in working through theprotocol is time well spent. The protocol discusses not only themain purpose of the study, but also which actions are required toreach this goal. It is not only the purely practical matters that need

to be discussed, but also the proposed research methodology itself.Selected portions of the protocol will be described in detail in a

later section of this handbook, but the overall contents will be pre-sented below. It is worth pointing out that it is the investigatorsduty to become very well acquainted with the protocol.

The title page usually states the studys title, the principal investi-gators name, position and telephone number, the other participat-ing investigators names (with or without address), the sponsorsname, address and telephone number, the sponsors contact per-son, and a date that indicates when the protocol was written. Each


31/110



time the protocol is revised a new date is added for identifying theapplicable protocol. It is also a very good idea to date each page inorder to avoid confusion among the different revisions. While thismay seem obvious, one of the most common problems occurringduring checks made by the sponsor or authorities is in identifyingwhich protocol was actually being used (not to mention which onewas approved by the ethics committee).

Summary and introduction

The protocol often begins with a summary or synopsis that con-cisely describes what type of study is intended, the study design,

the treatment methods to be used, the dosage (for drugs), andbriey how the study will be carried out.Following this, an introduction or description of the background

factors appears that describes both the reasons for carrying out thisspecic study and the treatment methods to be used. In drug trials,the new, unknown compound is usually described in some detail.The purpose is to give a background for why the study is being car-ried out and to state the reasons for why these specic treatmentmethods were chosen. The general idea is to include sufcientdetail to enable a knowledgeable but outside assessor to read thismaterial and then be able to agree that the idea of carrying out thisparticular study seems sound. This may make it necessary that thesection be quite detailed, especially when new drugs or treatmentprinciples are involved. On the other hand this automaticallyensures that a good background description has been written,which is needed in any case once the results are to be reported orpublished, and thus is never wasted effort.

Objectives

The study objectives must be clearly stated. It is generally wise notto address too many research questions at once. Certainly, severalobjectives can exist, but a single objective must be pointed out asbeing primary, if for no other reason than to build the basis for sta-tistical calculation of the necessary number of patients to include.The importance of having a well-dened objective cannot be over-emphasized. The entire studys design, selection of measurement


32/110


33/110


34/110

Chapter 3


characteristics of the selected patients may affect the treatmentmethods efcacy, and thus a detailed description is needed abouthow an even distribution of these characteristics will be achieved inthe different study groups (stratied randomization).

Practical execution

The protocol is also to describe the practical execution of the study,visit by visit. Here should be explained what needs to be done at dif-ferent occasions, the times between visits, the maximum allowabledeviations from these intervals, and so on. At an early stage carefulconsideration should be paid to whether it is truly possible to fol-

low the study plan. For example, is the clinic closed for vacationsduring a period that patients should be receiving a treatment? Arethe treatments planned for a part of the day that could conict withother activities?

There should also be a owchart that briey but clearly lists theactions to be done at each visit. If the protocol is written in English,which is usually the case, it can be a good idea to translate the ow-chart to the national language(s). This allows it to be used by thestaff as an easily comprehended guide. The protocol should alsostate which treatment the patients are to receive following thestudys conclusion. This can be especially sensitive in trials studyingdrugs or devices in an early phase of development and intended foruse in a serious disease where no effective treatment currentlyexists. A very difcult ethical problem can arise when the studyends and thus the question of whether to terminate a possibly func-tioning treatment while on the other hand there still isnt sufcientdocumentation to demonstrate that the drug is safe and efcacious

in the long run. This needs to be thoroughly discussed in advanceand described in the protocol so that the ethics committee has achance to evaluate the problem and the patients can be properlyinformed.

There must be both a specication of and reasons stated for thechoice of treatment; and concerning drugs, justications of the dos-age, dose interval and time of day for taking the drug are also nec-essary. It is important that the drugs in comparable studies are usedat adequate doses; that is, that the study is a comparison of differentdrugs at clinically comparable dosages. Likewise it is important that


35/110



the treatment method being tested against is a currently acceptedone. The ethical aspects of a placebo-controlled study should beespecially considered. Very often placebo treatment for short timeperiods present no problems, especially if no proven treatment forthe studied disease exists (see Ethical considerations); it is whenthe treatments begin to approach longer time periods, or in emer-gency situations, that problems can arise. The permissibility of additional medication during the study should also be stated, aswell as how one is to proceed with patients who require dose adjust-ments for one reason or another. It is also necessary to describe howone is to evaluate patient compliance, if applicable.

Timetable

The protocol should also include a study timetable stating the pre-liminary date that the nal report is expected.

Composition and handling of experimental compounds

A description of the experimental drugs composition is required;for example whether capsules or tablets will be used, or whether thecompounds resemble each other etc. Packaging size, dosing instruc-tions and labeling should also be stated. The labeling must clearlystate that the drugs are intended only for use in clinical trials. Theclinical trial must be identiable, usually using a study number, andboth the investigator and the manufacturer must be identied. Thedrug must be marked (for example, ABC 123/placebo) as well as theamount, dosage and mode of administration. Instructions for stor-

age, including a statement about keeping the compound out of reach of children, are also required. Double-blind studies requirethat each patient have a specially packaged drug labeled with thepatient number or its equivalent. An expiration date for the drugmust also be included. It is important that this labeling be done insuch manner that it is impossible to see the difference between thetwo drugs. The labeling must normally be written in the nationallanguage(s). One possible exception to this is using English labelingfor substances that are never handled by any other person than theinvestigator, such as radio labeled, directly administered sub-


36/110

Chapter 3


stances. Drugs intended for clinical testing may not be adminis-tered to other patients; and this is usually stated in the protocol.

Another question that needs to be discussed concerning the drugitself, is how it is to be handled. A careful accounting is required of which substance has been given to which patient, as well as whenand what the patients return. In principle, it must be possible toprecisely account for where each table has gone, and the various tal-lies must agree. After the end of the study, the drug should then bedestroyed and the destruction documented; this is normally doneby the sponsor, but the pharmacy can sometimes be of assistancewith this.

Use of medical devices

The national language(s) should be used in instructions for usage,maintenance, and cleaning of medical devices, the contents of these instructions must also be described, and a descriptionincluded about how these are to be distributed to the staff andpatients. The protocol must also state whether a cooperative con-tact will be made with the hospitals medical technology section.

The device should be clearly labeled that it is intended only foruse in clinical trials, its identity (designation, manufacturingnumber, batch number, date of manufacture etc.) and the manufac-tures and investigators names and addresses. This labeling shoulduse the national language(s).

The sponsors continuing commitments

Another practical detail that should be stated in the protocol is how

the sponsor will honor its commitments concerning its controlmeasures during the study. These include, for example, regular vis-its (see Monitoring), following up information provided to thestaff, any quality controls (see Audits, inspections and qualityassurance) and so on. In some of the cases, such as when a com-pany merely contributes economic support or free substances with-out having a direct interest in the study itself, the company will notprovide these functions, and this should clearly be stated in theprotocol. In these cases, it is the investigator alone who is the guar-antor that the study is carried out properly, including the responsi-


37/110



bility to monitor the study. Should this arrangement be a multi-center study with many participants, it becomes a heavy responsi-bility to be assumed by the investigator.

Randomization

Randomization procedures are an essential aspect errors madehere can invalidate the results and thus these must be described indetail. Both the time at which the randomization is to occur andwhether it is stratied must be stated. Stratication can becomenecessary in studies where inherent factors present in the popula-tion may be expected to have prognostic signicance for the out-

come (confounding factors). One such example is the knowledgethat smokers and nonsmokers react differently to a given treat-ment. In this case it is necessary to try to achieve an even distribu-tion of smokers and nonsmokers in the various treatment groupsbeing studied; and especially concerning smaller studies, it maytherefore be necessary to stratify the actual randomization in orderto obtain an equal preparation of smokers/nonsmokers in the twotreatment groups. Stratication can signicantly complicate therandomization procedure if more than one or two factors are to beweighed in. However, the techniques in this eld have advancedconsiderably in recent years, and today there are good possibilitiesof conducting a stratied randomization where a large number of factors can be considered. This is done using computer programsthat can be programmed for a large number of factors. However,this assumes that the randomization is done centrally at one loca-tion; it can also cause several practical complications concerninghow the blinding is to be maintained, how to ensure that the

patients receive the correct treatment, etc. Thus, the choice of astratied randomization must be well justied and thoughtthrough in respect to the practical consequences.

Breaking the treatment codes

Another required section is instructions on who is allowed to breakthe treatment codes for a patient, under which circumstances andhow this is actually to be done. Randomized, double-blind studiescommonly use a list of randomization codes for all patients and one


38/110

Chapter 3


envelope per patient with an individualized code. These envelopesare usually retained in duplicate, with one copy for the investigator,and the other(s) for the safety committee (if applicable), and spon-sor. This ensures that there always is a possibility of breaking thecode in case of emergency with any given patient.

Breaking the code (for example, for an interim analyses for anentire study) may be necessary at times for reasons of both efcacyand safety. Unplanned interim analyses should preferably not beattempted while the study is in progress, but at times it is a neces-sary evil brought about by safety reasons (to name but one). Insome cases there might be good reasons to plan for interim analy-ses up front. This could be studies where the disease studied is very

serious or when the desired outcome is to inuence mortality. Insuch cases it is of utmost importance to detect serious adversereactions, and decreased or increased efcacy. There are accepted,techniques for these kinds of interim analysis. Should interimanalyses be planned for, the reasons and techniques must be thor-oughly described. In the revised version of the Helsinki declara-tion it has been stated that physicians should cease any investiga-tion if the risks are found to outweigh the potential benets or if there is conclusive proof of positive and benecial results. To dothis while a study is ongoing necessitates interim analyses, and itis thus not unlikely that the number of planned interim analyseswill increase.

Withdrawals

A separate section should exist that discusses patients who with-

draw from the study for whatever reason. There may be special pro-cedures to be performed in case of withdrawals and there may alsobe circumstances where the patients data may still be usable tothe study in certain aspects. All patients included in the study mustbe reviewed afterwards, and even if withdrawn may still beincluded in certain analyses depending on the reason for withdraw-ing. When the number of required patients for the study is calcu-lated, a certain drop-out rate ought to be expected and planned forin the early planning stages of the study in order to secure a suf-cient number of evaluable patients.


39/110



Evaluation of efcacy

Methods included in the study, such as those used for the evalua-tion of efcacy and safety, must be described. This is especiallyimportant for multi-center trials, i.e. those with more than one par-ticipating center. For the study to be reliable, the efcacy must bemeasured with the same techniques at all centers. At times thiscould require the use of rather complicated standardization meth-ods to guarantee uniformity. It has become increasingly commonthat the efcacy variable is evaluated by a central unit even whenthe data has been gathered elsewhere, for example, that all ECGs beevaluated at one center, or that all blood samples are analyzed by asingle laboratory; however, this is not always possible. Should cen-

tral evaluation not be possible, it is necessary for the various inves-tigators to have the possibility of meeting regularly throughout thestudy in order to maintain coordination at this level. The protocolmust also state how the efcacy variable is to be evaluated and whatrequirements are necessary in order to classify the patientsresponse to treatment, etc. This is especially important if the ef-cacy variable is subject to subjective evaluation, such as an estimateon a descriptive scale. These types of variables may necessitate

repeated trial runs in which the participating investigators practicemaking reproducible and consistent evaluations, and get a chanceto discuss differences in their evaluations.

This is extra important in studies in which, for whatever reason,the treatments are given openly. One common method is to use anindependent end-point committee, which is a group of specialistswho have no other contact with the study, and who can evaluatethe efcacy blindly; that is, without knowing which treatmentany particular patient has received. One type of such techniquethat has come into use in large mortality studies is the PROBEdesign, which stands for Prospective, Randomized, Open, BlindedEvaluation. The patients are openly treated, often using regularlyprescribed drugs, but the evaluation of all end-points, that is theprimary efcacy variables, are made by an independent committeewho receives all its materials blinded, meaning that all informa-tion that identies which drug treatment the patient has receivedhas been omitted. This or similar kinds of studies naturally require

much consideration in order to work properly, but may be the only


40/110

Chapter 3


option for carrying out very large and long studies. Another varia-tion of large mega-studies are the large cardiovascular studies runon an international basis, such as the ISIS studies. These studies arerun using blinded drugs and sometimes placebo, but the actualstudy conduct and data collection have been simplied in theextreme. No true monitoring in the form of site visits has beendone. The ICH GCP guidelines have seized upon this kind of studydesign as being both important and possible, and states that a spe-cic exemption from the monitoring requirement at each study sitemay be acceptable for certain types of studies, in which case thismust be motivated and described in the protocol. In a similar man-ner the new EU directives describe non-commercial clinical trials

as an area where simplied regulations for, for instance, drug labe-ling, can be applied. Non-commercial trials are dened as clinicaltrials where the pharmaceutical industry is not participating assponsors.

Statistics

The statistical aspects require their own, thoroughly described sec-tion that discusses the basis for the sample size (which often entailsan estimate of the expected efcacy and its variability), the desiredstatistical power and an estimate of which differences can be clini-cally valuable to discover. Sometimes it is simply impossible toexactly predict the size of the expected efcacy and a more generalreasoning must sufce.

In this section, a description of how the study results are to beanalyzed must be stated. If possible, this section should cover the

choice of the intended statistical methods and their levels of signif-icance.

Also needed is a denition of which patients will be included inthe analyses, the intention-to-treat (namely, all patients rand-omized into the study) or on-treatment (meaning all patientswho correctly completed the study, also often called the per-proto-col analysis). Subgroup analyses may also occur, which has impli-cations for the calculation of the total number of required patients.The use of subgroup analyses should be planned and clearly statedbefore the trial begins.


41/110



As discussed above, an increasing number of studies may requireinterim analyses usually for safety reasons. Because the perform-ance of such analyses can affect the statistical calculations for thestudy in its entirety, it is wise to seek out professional help in orderto perform these analyses correctly, not only concerning how to doan interim analysis, but also for estimating the effect they will haveon the nal analyses.

Because the various statistical calculations are the foundation fordesigning the entire study and can have far-reaching consequencesfor the number of patients, the potential for carrying out the studyand so forth, the importance of a sound statistical description can-not be pointed out too often. Unfortunately, the statistical founda-

tions in many studies are rather meager, something that many aninvestigator has bitterly regretted afterwards

Safety and ethical aspects

All clinical trials, indeed all human experimental research, must beapproved by an ethics committee, and if it concerns the testing of experimental compounds, often also by the local regulatory author-ities (see Approval from the regulatory agency). Since this is regu-lated by guidelines and local regulations it is not always specied inthe protocol that an application is to be made. However, thereneeds to be a section in the protocol about the ethical aspects of theresearch, and each investigator must reach an individual decisionthat the possible benets outweigh the possible risks, and that riskis minimized. Written information for the patient is obligatory andmust be appended to the protocol. Also required is a description of how information to the participants is to be presented, and how

their consent is to be obtained. In addition, in many countries it isalso required to submit a copy of any advertisement concerningpatient recruitment. Careful instructions concerning how adverseevents are to be reported must be included, including a time framespecifying the nature of the adverse event and its severity (seeAdverse events side-effects and others). The ethical principlesthat governs all clinical research are described in the Helsinki Dec-laration, which is often appended to the protocol, and is includedin this handbook (Appendix III). Ethical aspects are treated in moredetail in the section entitled Ethical considerations.


42/110

Chapter 3


There is another safety aspect than those purely related toadverse reactions. This concerns the carrying out of the study andthe test results. In intervention studies, especially those intendedto reduce mortality for a given disease, the test results obtainedduring the study may make it unethical to continue, a fact thatthe revised version of the Helsinki declaration emphasizes. It mayturn out that the new therapy does not have a sufcient efcacyor which results in the same ethical problem the new therapyleads to a much improved efcacy. In either case it is unethical tocontinue treating the patients with the inferior alternative. Inorder to discover this, it is necessary to build into the study acheckpoint along the way, in which the result up to that point can

be analyzed (interim analyses). This is possible, although rathercomplicated from a statistical point of view; and it ought to becarefully discussed in advance how this is to be done and whodecides when to break off the study. Very often an independentsafety committee is established for this purpose, whose only func-tion is to control the safety aspects, but is otherwise uninvolved inthe study.

This committee is to monitor the study by adhering to a previ-ously drawn up program, and the committees members are theonly ones who have access to the trial code and these results whilethe trial is running.

Information/delegation to the staff

The ICH GCP guidelines also emphasizes the relationship betweenthe investigator and other staff involved and the necessity to decide

who does what, and document this, before the start of the study.This can be done in the protocol, but perhaps more easily in a sep-arate delegation and signature list.

Other points that might need to be specied in the protocol aresections about insurance for and possible compensation to patients, a description of how the results are to be published, andhow the archiving of trial documentation is to occur.

A vast number of other details, depending on the type of study,can also be included in the protocol. Because the protocol is thestudys bible, many practical details may be included. This often


43/110



makes the protocol rather bulky, which is something that one sim-ply has to put up with. Once the study is underway, a simpliedowchart is usually sufcient, and the protocol then serves as a ref-erence source.

The Case Report Form

The Case Report Form (Case Report or Record Form, called theCRF) is the document into which all the data necessary for anal-ysis is to be recorded. The forms design may vary, but it often takesthe shape of one notebook per patient and is often more extensivethan expected. All essential information is recorded in these note-

books, patient by patient and visit by visit. It is advantageous if each section in the notebook is accompanied by instructions con-cerning what should be recorded at every specic occasion. Whilethis information may be found in the study protocol, it considera-bly facilitates practical operations to repeat the instructions in theCRF. Once the study is underway it is the CRF that is in daily use,and it is practical and convenient not to have to refer to the studyprotocol during each visit.

The CRF is often designed to facilitate data analysis. At times thismakes them somewhat clumsy from the clinicians point of view.Unreasonably impractical CRFs should not be accepted, since theyendanger a smooth and consistent data collection. It does not helpmuch if a CRF is very easy to enter into a computer database, if itsconstruction has made it almost impossible to complete in the rstplace.

Naturally, the actual appearance of the form will vary considera-bly from study to study and from sponsor to sponsor. A more com-

pact form with much information to be reported on each page mayseem attractive due to its lesser bulk. However a more sparse formrequiring less data per page may be easier to ll in, despite any ini-tial shock on seeing its bulk for the rst time.

The data in the CRF is to be lled in by the investigator or anyperson formally delegated to carry out the procedure in question.Those receiving authorization to ll in the data in the CRF andpreferably a specication of which data it applies to must be doc-umented on a special signature-list, which is to be updated asneeded during the study and archived afterwards (see Study docu-


44/110

Chapter 3


mentation). It is ultimately the investigator who accepts responsi-bility for the correctness of these data via his or her signature.

If for some reason the data needs to be revised in the CRF, the olddata should be crossed over, not erased, and the revision is to bedated and initialed. In addition a reason for the revision needs to bestated if it isnt immediately obvious in context.

The data that is recorded in the CRF also generally needs to beentered into the patients regular chart, which is the source docu-ment. All pertinent data must be recorded in the patient chart incase it is needed in the future for data verication (see sections:Monitoring, Audits and Documentation and archiving).However, in some cases some of the original data collected for the

study can be of such nature that they are not very suitable forarchiving in the patients regular chart and doing so might indeedbe contrary to local legislation about what belongs in a patientchart. In several countries the patient chart should contain only theinformation necessary for the proper medical treatment of thepatients disease. Quite often a substantial amount of the data col-lected in a trial is not directly related to the treatment of thepatient. For example, this may occur if a large number of extraexaminations are conducted exclusively for reasons related to thestudy, and lack signicance for the patients continued care. Thereis a not insignicant risk that much of such data will later beexcised from the chart. It is then better to record archive them in, oralong with, the CRF. When this procedure is chosen it should bestated in the patients chart from the study start. This is also trueconcerning data documented directly into the CRF, such as repliesto specic questionnaires or forms. The location where the data isrst documented is, of course, original data in this case the CRF

itself. This is entirely acceptable, but requires description in thepatient chart at the start of the study, so that one can always knowwhere to nd the original information. According to the ICH GCPguidelines a statement must be added to the protocol that thismethod (i.e. using the CRF as source data for some variables) hasbeen chosen and which data it applies to.


45/110



Ethical considerationsThe relationship between doctors and their patients during researchwere rst widely disseminated in writing in 1947 as the Nuremberg

Code. This was later further developed as the Helsinki Declarationand was adopted in 1964, and later revised in 1975, 1983, 1989,1996 and 2000 and is under continuous revision. This documentmaintains that a doctors rst responsibility is always to protect thepeoples health. It also ascertains that the development of new,effective treatment methods unavoidably involve experimentationusing humans, experiments that can be potentially risky. Whatmust then be complied with in order that such experiments, suchas clinical trials, can be carried out?

First and foremost, established scientic methodology must beused. The experiment must be substantiated using laboratory dataand animal data as much as possible, and investigators are requiredto be extensively read in the relevant literature.

A carefully constructed protocol must be written and submittedto an independent group for evaluation (an ethics committee).

The person running the test must be scientically qualied andthe risk to the individual patients participating must not exceed the

expected benecial results neither for the patient himself/herself nor for a larger group of patients. It also emphasizes that theresearch subjects integrity is to be protected at all times, that theparticipation must be voluntary and consented to only after beingaccurately informed about every aspect of the experiment.

In addition, when the study subjects are also the doctors ownpatients, it becomes especially important that the patients depend-ence is not exploited, and that it is perfectly clear that a refusal toparticipate will not in any manner affect the patients own treat-ment. A respect for the patients best interest must always be kept inmind and the doctor is always responsible for the patient receivingthe best possible treatment at every given occasion. It is also statedthat the researcher is responsible for the correctness of the ndingsin publication, and that any study not adhering to these ethicalrules will not be accepted for publication.

In the latest version of the Helsinki Declaration the protection of patients have been further strengthened in several areas. One of these concerns the use of placebo. This has hitherto been stated as


46/110

Chapter 3

46

Documents

Carola Lemne Handbook for Clinical Investigators 2008