GLP Protocols and Study Conduct-It Just Takes A Little Planning
GLP Protocols and Study Conduct-It Just Takes A Little Planning
EXECUTIVE SUMMARY
Good laboratory practices (GLPs) have been universally recognized
as rules that govern the conduct of non-clinical safety studies. The
GLP goal is to ensure the quality, integrity, and reliability of the data
developed from such studies, which in turn allows the appropriate
assessment of the safety and effectiveness of the drug or device that is
the object of the study. The US Food and Drug Administration relies on
documented adherence to GLP requirements by nonclinical laborato-
ries in judging the acceptability of safety data submitted in support of
research or marketing permits (1). Laboratory management is respon-
sible for communicating and supporting a commitment to the reporting
of laboratory results and delivery of services in an efficient and timely
manner. The “Good” in GLP cannot happen without well thought-out
protocols and proper conduct of the study itself. This article focuses on
these two critical elements to provide a roadmap for a journey to com-
plete and compliant GLP study.
INTRODUCTION
All nonclinical laboratory studies that support, or are intended to sup-
port, applications for products regulated by FDA must adhere to 21
CFR Part 58 (2). Section § 58.120, which addresses protocol require-
ments, and § 58.130, which defines the conduct of a nonclinical labora-
tory study, are especially important sections.
Any person who submits an application for a re- and handling, analytical methodologies, data analysis
search or marketing permit that includes data from and documentation of the study), then the template ap-
a study in connection with the application must proach may be acceptable.
include a true and correct statement that the study However, this approach is not without pitfalls,
was conducted in accordance with GLP require- because there may be a tendency for developers
ments. Admission of false information may result in using a template to actually skip over the contents,
the cancellation, suspension, or modification of an missing some key elements. Likewise, this same
approved research or marketing permit, or denial scenario may apply to those who review and ap-
or disapproval of an application for such permit. prove such documents. When this occurs, it is quite
Therefore, it is essential that the testing facility likely that the protocol will require amending after
strictly adhere to the requirements of GLP and the its initial approval, often because the individuals
approved study protocol. who must execute the protocol have encountered a
Although it is not required by regulatory require- problem in the documentation.
ments, studies not intended for submission should Developers of any protocol should strive to mini-
be conducted in a manner comparable to GLP. mize, if not eliminate, the need for amendments. Keys
Maintaining a standard of compliance assures that to strategically pre-planning a well-developed protocol
regulatory studies, when conducted, are not nega- are found in Table I.
tively impacted. Protocols developed for a nonclinical laboratory
study should comply with a standard format and con-
PRE-PLANNING AND tent to ensure that all elements are addressed. Failure
DEVELOPING A SUSTAINABLE PROTOCOL to have such requirements is another reason for having
Much has been written about protocol development to amend the study protocol. The study director, who is
with regard to format and content. Developing proto- responsible for designing the protocol format and con-
cols is a time consuming and costly endeavor to the tents, can then assign the task of protocol preparation
extent that facilities often use “cookie cutter” templates to appropriate personnel and monitor its development.
for their development. If the GLP study is similar to The protocol for a nonclinical laboratory study must
that of a previously performed protocol (with regard provide specific information required to properly con-
to the study goal, risks, populations, sample collection duct the study. The protocol must clearly state the
PROTOCOL DEVIATION REPORTING after the occurrence or after becoming aware of the
Referred to here are unplanned deviations. Most will incident). Maintenance of a protocol deviation report
agree that a protocol that necessitates a so-called log is often the responsibility of quality assurance or
“planned deviation” is poorly conceived and is lacking under the administration of the study director, who
in planning—meaning it should never have left the is ultimately responsible for review and approval of
launch pad. protocol deviation reports. A sample protocol devia-
Deviations from protocols are, unfortunately, not tion report is found in Figure 1.
uncommon events. Human error is generally the In addition, the study director must also review
culprit lurking behind a deviation. While it is best to and approve summary and trend analysis reports,
avoid deviating from the prescribed activities defined recurring deviations, and effectiveness of corrective
in the protocol, strategies must be made in advance for actions as well as ensure timely completion of correc-
efficient and expeditious management of these events. tive actions identified in protocol deviation reports.
It is essential to provide a mechanism for docu- Summary and trend analysis reports are important in
menting exceptions from approved methods de- discovering recurring deviations and effectiveness of
scribed in a protocol for a nonclinical laboratory corrective actions.
study. All unplanned deviations must be recorded The following are two basic reasons for most
and specifically assigned an identifier by which it deviations:
can be used in creation of a deviation report and for • The source of the deviation is internal, where the
tracking and trending. The supervisor of the as- people responsible for conducting the study have
signed area must immediately be notified when the created (unwittingly) the set-up for deviations to
deviations occur (preferably no later than 24 hours occur
Figure 1:
ExampleFigure: Example
protocol deviation report. protocol deviation report.
Future
Action:
By:____________________________________________
Date:
___________________________
Final
Approval:
Quality
Assurance:
_______________________________
Date:
___________________________
60 Journal of GXP Compliance
Page
1
of
1
Jeff Morgan
• The source of the deviation is external, where the data can and have led to serious errors in an oth-
participants in the study fail to follow instructions. erwise good study. This is where logs and training
records become invaluable. Call logs, training logs,
Internal Sources of Deviations written communication and responses should be
Internal sources of deviations further fall into the developed so that communication is clear and con-
following categories: cise, and can be traceable, not only to avoid devia-
• Poor planning. Examples include incorrect or tions, but as a tool in investigating the root cause
insufficient quantities of study supplies, poor of deviations. Additional root causes for deviations
quality study supplies, unprepared study sites, from an approved protocol are found in Table IV.
study sites incorrect (i.e., cannot handle the vol-
ume, have insufficient storage or clinical facilities), External Sources of Deviations
and facilities issues (i.e., insufficient cold storage External sources of deviations are similar in nature to
for samples, problems with validated systems, those mentioned with internal issues, but an additional
including system failures that may impact product layer of problems may occur if written instructions to
quality or may require increased environmental patients have not been “road tested” by non-clinical
monitoring). personnel for clarity and understanding to determine
• Insufficient training. Examples include the lack how easy the instructions are to follow. This is a big
of specialized individuals to process and evalu- challenge, because well-written subject instructions
ate specimens taken from patients during a study can make or break a study.
(i.e., ensuring serum samples are not hemolyzed For example, during one study, the subject was
or being aware of how to judge whether lipemia or responsible for collecting a 24-hour urine sample and
specimens with high bilirubin may interfere with was given instructions written by a medical profession-
test results). Training can also relate to counsel- al. During the evaluation of data, the patient appeared
ing a study subject on a variety of issues such as to be an outlier because the analyte in the urine was
fasting before blood draws, timing of taking study extremely elevated. However, as the data were further
medications, proper collection of in-home samples scrutinized, it became apparent that the patient under-
or spotting a problem that could lead to disquali- stood that the sample was to be taken once, and again
fying a subject, such as excessive weight gain. 24 hours later instead of collecting all urine output over
• Lack of experience. While experience must be 24 hours. The error was discovered because the sample
thoroughly assessed during the selection of study volume was consistent with a patient with kidney
sites, it is possible that through employee turnover failure, but the blood work showed just the opposite.
the individuals in place during the site assessment The takeaway from this is if the patient is given written
may have moved on at some point in the study. instructions, they need to be written in “patient-ese”
It is important for the study director to be kept and not with confusing, conflicting medical terminol-
appraised of any study site personnel changes and ogy or jargon.
evaluate the experience of the replacement per- One-off deviations can and will occur. However,
sonnel to ensure optimal quality throughout the when the same category, cause, or class of deviation
study’s lifecycle. recurs, then a larger issue is introduced. Not only is the
• Poor communication. Most problems of this reason for the present problem important to discover,
nature arise from assumption. Nonclinical labora- the reason for the recurrence must also be pinpointed.
tory studies are expensive, time consuming, and Repeat problems or deviations must first be completely
must be exact in the execution. Assumptions by dissected for the root cause, reviewed for previous
the study director or sponsor regarding the avail- attempts at corrective and preventive action, analyzed
ability of supplies, proper study subject prepara- to discover why the corrective and preventive action
tion, managing the subjects and their samples and was inadequate, and re-evaluated to permanently fix
TABLE IV: Common root causes for deviations from an approved protocol.
Material Use of non-conforming material or reference standards where the testing and specification have been
validated; includes the use of expired material.
Method Use of an inadequately validated method.
Document Use of a SOP (i.e., redlined) that is not approved for use.
Operator Errors made by study personnel or instances of not following approved procedures (without an approved
protocol deviation report).
Equipment Equipment or facility breakdowns, failures, or instruments found to be out of calibration.
Environment Air, surface, or water sampling results out of specification.
Schedule Departure from established schedules such as environmental monitoring, preventive maintenance, and
calibration programs.
Validation/qualification Use of equipment or method inadequately validated or qualified, or specifications established without
adequate supporting data.
Vendor Changes in vendor supplied product (i.e., documentation or certification of purchased parts, changes in
product specifications, changes in packaging), inability to deliver pre-shipment samples, and contract
test lab errors.
the situation. Retraining the retrained is not a viable apparent, indicate the need for further investiga-
option. tion to determine corrective action or indicate that
Well-trained area supervisors should possess the the QA unit or the study director will determine
proper skills for determining the cause of the devia- the corrective action.
tion within their area and recommending appropri- • Review the report with QA to determine the
ate corrective action. While deviation investigation impact of the deviation on the study. Obtain and
techniques are beyond the scope of this article, it is record approval by QA for the report. Confirm
sufficient to say that using tried and true root cause the cause of the deviation and determine the next
analysis should be in an area supervisor’s toolkit. course of action.
• Do not accept any planned deviation that has a
IF YOU HAVE TO DEVIATE, DO IT CORRECTLY significant effect on the study. The study direc-
The following are recommended when deviations are tor must evaluate these deviations prior to their
necessary: potential implementation.
• Evaluate the deviation to determine whether the • Review each protocol deviation report to deter-
deviation could affect the data collected for the test mine whether the deviation described should be
article or otherwise corrupt the study data. If this permanently incorporated into the protocol. If it is
is the case, send the report to the study director for determined that the deviation should be incorpo-
review, discussion, and approval. If the deviation rated, amend the protocol according to established
report is approved, return the report to the quality procedures. Significant repeated unauthorized
assurance (QA) department for inclusion in the protocol deviations by a testing facility will war-
final study report. rant further investigation by the sponsor and may
• Recommend and assign a corrective action in the result in termination or suspension of the testing
report to prevent the future occurrence of devia- facility’s participation in the study. The takeaway
tions due to the same or similar reason. here is that any deviation is a serious event and
• Describe the corrective action specifically to allow requires the utmost attention to detail.
verification. If the corrective action is not yet
PROTOCOL AMENDMENTS
TABLE V: Suggested content of a protocol amendment.
A protocol amendment is prepared when any perma-
nent modification to the objective, methodology, or Protocol number
analytical methods of the approved protocol is planned Protocol amendment number
and initiated. Date of protocol amendment
The Code of Federal Regulations succinctly directs Study title
what must be done in the event of a change to a
Nature of the revisions with reference to the applicable sections
protocol: of the protocol
“All changes in or revisions of an approved protocol
Reason for protocol amendment
and the reasons therefore shall be documented, signed
by the study director, dated, and maintained with the Name of individual initiating the amendment
protocol” (3). Signatures indicating approval of the amendment
When an amendment is necessary, inform the QAU Dated signature of individual responsible for filing the
of the proposed change prior to initiating and imple- completed document
menting the protocol amendment. Make amendments to
approved protocols according to established procedures
and request QA to assign a number for the amendment,
generally by assigning the number of the corresponding with safety and effectiveness to the patient. Thorough
protocol followed by a numeric suffix. For example, assessment, communication, pre-planning, and an
DEV-1-01 WOULD INDICATE THE FIRST AMEND- adherence to basic elements of protocol development
MENT FOR THE PROTOCOL NUMBERED DEV-1. and execution practices may seem on the surface to be
Subject the protocol amendment to the same review overkill, but will save time and expenses in avoid-
and approval requirements as those of the original ing protocol false-starts, numerous amendments, and
protocol. Include the same individuals in the review of potential deviations that can derail an otherwise good
the amendment that were responsible for review and study. Taking the time to get it right the first time is
approval of the original protocol. a guaranteed advantage over poorly or inadequately
Typical information to be included in a protocol prepared protocols or inadequate GLP resources, inef-
amendment is shown in Table V. ficient laboratories, or confused study participants. It’s
After the amended protocol is returned, incorporate just the right thing to do.
recommendations and circulate the revised protocol
for review and approval. Obtain the dated signature of REFERENCES
the study director to indicate approval of the amended 1. FDA, Food and Drug Administration Compliance Program
protocol. If necessary, obtain the dated signature of the Guidance Manual, Chapter 48–Bioresearch Monitoring, Good
sponsor. Laboratory Practice, February 21, 2001.
Once approved, submit original protocol amend- 2. FDA, 21 CFR Part 58-Good Laboratory Practice for Nonclini-
ments to the study director and a copy to the spon- cal Laboratory Studies.
sor. Maintain and control the original approved study 3. FDA, 21 CFR 58.120(b).
protocol and all approved protocol amendments in an
organized manner within the QAU files according to GENERAL REFERENCE
applicable regulatory requirements. WHO, Handbook: Good Laboratory Practice (GLP): Quality Practic-
es For Regulated Non-Clinical Research And Development-2nd
CONCLUSIONS edition, UNDP/World Bank/WHO Special Programme for
GLP study protocols and their conduct are critical Research and Training in Tropical Diseases (TDR), World
in the accurate and reliable assessment of whether a Health Organization, Geneva, 2009. GXP
given drug or device will meet its goal of intended use