OR WAIT null SECS
With globalization of trials comes the difficulties of sample logistics. Enter the central laboratory model.
In the early 1990s, the concept of the central laboratory was developed and implemented by laboratories delivering services to major pharmaceutical companies. The goal was to consolidate the test results and data originating in different clinical sites, which was previously analyzed in local labs. Bringing the samples to one single laboratory would avoid consolidation of biased test results among different laboratories, all of which could be using different analytical platforms, kits, and reference values.
This concept was first applied to clinical studies conducted in the United States. Soon after, the courier industry started offering solutions for biologic sample transportation, which allowed the central lab concept to be applied globally.
Later, the concept of the affiliated laboratory was created. The affiliated laboratory covered geographic regions that had difficulties exporting biologic samples. As a consequence, the central laboratory became more global and started to build different types of associations with analytical laboratories in different parts of the world.
Nowadays, clinical trials are globalized and clinical sites are not just in the backyard anymore. Places that may sound exotic such as Coquimbo, Chile, and Belém do Pará, Brazil, are actually well prepared for Phase II and III studies and perform according to Good Clinical Practice requirements.
In parallel, logistics costs and constraints are increasing, especially after September 11th. The increased project logistic costs may, in some instances, be higher than the laboratory costs. Also, a considerable number of samples are not able to reach the analytical facility under adequate conditions to be properly analyzed. The patient recall for sample drawing may take up to one week, sometimes more, and represents a delay in the study schedule. Additionally, this recall may negatively affect patient adherence.
Query solution and clinical site support processes usually are best dealt with by local teams and staff. Differences in language and time zones between central labs and clinical sites may be problematic if dealt with from a distance.
Additionally, clinical site staff may be responsible for logistical tasks, although that is not exactly in their scope of expertise. These tasks could include sample and invoice preparation, courier processes or dry ice providers. There also are difficulties in shipping lab materials to remote sites, which could increase study costs and cause logistic constraints. These logistic constraints usually relate to moving samples and lab kits to central labs in reference countries, generally the United States or Europe, which is forcing trial sponsors to consider new approaches regarding central laboratory services.
Considering that the analytical equipment and reagents most frequently used are available in several countries in the five continents, a new concept is becoming more popular among the most innovative trial sponsors and global central labs: partner laboratory harmonization.
Harmonization is a process led by the global central lab and its partners for specific projects. It aims to integrate comparable results from different laboratories, avoiding possible bias generated by technical differences among them.
This process allows laboratories to exchange results instead of exchanging samples, thus overcoming obstacles previously mentioned and representing fewer costs and delays for the sponsor. Furthermore, it is also fair to say that access to new research environments, with new sets of investigators and patient populations, will also be more easily achievable.
Depending on the level of harmonization that a group of labs decides to implement, they may reach a very close technical comparability and can even be considered as a single entity by the trial sponsor, delivering the same service and results all over the world. Of course, there are different levels of harmonization, which could involve internal and external parameters.
Another possibility that has to be considered is the harmonization process involving local hospital laboratories because of some laboratory set-up particularities such as turn-around time, responsiveness, and special test menus that are performed in emergency situations.
Analytical platform harmonization. In terms of analytical platform harmonization, the first step is to compare the analytical equipment, methodologies, kits, and reagents used in the laboratory test either in general or for the specific trial. If differences are found, they have to be addressed through correlation tests. This will show that the results coming from different equipment can be considered homogeneous and, therefore, can be consolidated in the study databank.
Differences are found with some frequency, particularly for basic hematology and hormone tests. Additionally, the equipment preferences are different in different places, such as the United States and Europe. Also, keep in mind that the global central lab leading the harmonization process usually has more than one platform for the same test, whereas the local lab is less flexible in terms of analytical equipment.
Several aspects are part of an analytical equipment setup decision. One aspect to consider is technical support and kit availability in the country, which is usually considered by the local laboratory during its clinical laboratory setup phase. Nowadays, the analytical equipment and kit providers frequently have global distribution and are represented in the countries where the clinical trials are conducted.
A correlation test program can be necessary depending on the tests involved, regardless of differences between the analytical setups.1 This correlation means the exchange of samples between labs and comparisons of the results to check if the harmonization is consistent and can be demonstrated statistically during the study. The exchange procedure frequency will have to be agreed on by the involved laboratories.
There is the virtual central laboratory concept. In short, there is a calibrator that runs together with the trial samples. The calibrator and the sample results are compared, and the result is adjusted based on the calibrator value that is used by all the laboratories participating in the trial. The virtual central laboratory concept has been used by Covance (Zeist, Holland)2 for several years. However, a potential pitfall is that this concept does not consider monitoring of the local labs.
Last, but not least, the IT platform plays a major role in the process because the final product is the data, which will have to be generated, transmitted, and stored in a way that could be consolidated within the study database accordingly. This will require a lot of effort and investment in validation documents and processes, together with safe and flexible connectivity with different clients and databases.3,4
Reference value harmonization. This aspect is important for the data analysis and data management process. It can be a difficult task depending on the population and tests involved. Safety test reference values are easily harmonized because most of them follow international standards. Safety tests are the main analysis done at local labs in order to speed up a project, especially during the screening phase.
Hematological parameters can be harmonized, provided that the equipment/methodologies are previously discussed among the partner laboratories (because of the different sensitivities among the existing analyzers). Often, a correlation study is necessary. Also, there can be differences regarding population aspects such as genetic and nutritional characteristics, among others.
The way of reporting urinalysis needs to be discussed previously because there might be some differences in reporting qualitative aspects. The reporting conventions are not always the same. For virology tests, despite the fact that reference values are usually qualitative (negative/positive), the methods linearity and sensitivity demand an exchange of information among the partner laboratories. And molecular biology test harmonization depends on the protocols used for the result analysis and reporting, especially for genotyping tests.
In my experience, harmonization is very difficult for coagulation and immunophenotyping tests. Those parameters are highly dependent on population characteristics and reagents used by the laboratories. However, samples for immunophenotyping tests are not stable enough to be transported from one country to another, which means that the protocol and database have to carefully harmonize these parameters.
Certification, accreditation, and external QC programs. The third step requires that all laboratories involved in a project be able to show their national and international certification and accreditation. The most frequently requested are the ISO 9001 certification and CAP accreditation.5
CAP accreditation could be very costly to many local labs and has to be taken into account. Participation on the CAP proficiency testing surveys is more common and affordable. The certifications for environmental protection and sustainability are becoming very popular among the clinical trial sponsor companies. Therefore, ISO 14000 accreditation is a positive certificate for clients and sponsors.
There are some quality control programs specific to certain areas of the clinical lab (such as the NGSP program for glicated hemoglobin), and these must also be a part of the discussion.
The requirements from different regulatory agencies also have to be considered, such as 21 CFR part 11 from FDA, which refers to electronic system validation.2
Laboratory routines and reports. The fourth step requires that routines, not just the equipment, go through a harmonization process. The stability data in different temperatures need to be checked. Calibration frequencies, preventive equipment maintenance, and repeat thresholds also have to be discussed among participating laboratories.
Reporting and data transfer format also are easily harmonized, despite the fact that there may be some difficulties with a few tests, such as urinalysis results.
Today, there is a multidisciplinary group of international specialists that are enforcing the adoption of the CDISC protocol as a kind of standard regarding data transfer format. This tends to facilitate the workflow for IT structure, since there are several formats and platforms for transferring the data requested by different sponsors.
To ensure laboratory routine consistency throughout different partner laboratories, some SOPs will also have to be discussed and incorporated into the lab's SOP list.
Sponsors want to receive a reliable clean database from a single source and they usually request different types of connectivity. Therefore, the local lab needs to be prepared to cope with different communication strategies. The common point among them is that the local lab will be requested to send the data to the global central lab or to the global data manager. However, how to do this can be a difficult task.
The easiest way is to allow the local lab's Laboratory Information Management System (LIMS) to transfer the data to the central lab electronically. Other central labs insist on an interface construction between their system and the local lab's. This is not so easy or simple because it represents two different LIMS running at the same lab. Both labs will have to be fed with the same data for production and sample tracking control. In addition, we should not forget that every interface among different LIMS is considered critical for the 21 CFR part 11 validation processes.6
The global lab will conduct the harmonization process and define what level of harmonization is required. The global lab is the one that will have access to the sponsor's requirement.
There are some analyses that usually are centralized and will remain under the global central lab's responsibility, such as pharmacokinetics, study endpoints, and global data management, which can be done also by a third-party company.
The local lab will conduct the trial in its region. In addition to the sample analysis, it will support the clinical sites and take care of the inbound and outbound samples and kit logistics. The data will be transferred to the global data manager according to the global central lab's specifications.
The local lab performance has an important impact on the relationship between the investigator and the study sponsor. Sometimes, remote project management interferes negatively in this regard because it is not a concern for the global central lab that is managing the clinical trial project with a given principal investigator at his/her clinical site. This is an additional advantage to having a local laboratory supporting the sites.
At the end of the responsibility chain is the sponsor, who is responsible for the data submitted to the regulatory agencies. Therefore, the local labs have to be prepared to host periodic audits with the sponsor or global central lab's audits. The scope of these audits is usually related to a particular project or can focus on the laboratory and QA processes.
Some sponsors ask for a qualification visit before starting the harmonization process. This can be a useful procedure that strengthens the level of integration between local and global laboratory service providers.
Last but not least, the regulatory agencies are also prepared to make inspections at the local labs in order to validate or not validate the data coming from different laboratories.
As for sample drawing kits, it is increasingly difficult to put sample drawing kits shipped directly from the global central labs in place at different clinical sites. Not only do costs contribute to these difficulties, but so do customs requirements. The major sample drawing material providers are represented globally, allowing kits to be assembled and dispatched locally. This is a meticulous and critical process, which requires enough resources and competencies in the local labs.
For some global central labs, the kits are linked to the sample tracking system, which obliges the global central lab to dispatch kits already bar coded. This will cause the local labs to have additional work whenever the sample drawing tubes do not fit the analytical equipment. This issue has to be addressed during the harmonization process discussion.
Clinical trial sponsors are facing a dilemma. They have to choose between the previous central laboratory concept and face the logistics problem, which could increase costs and cause time schedule delays, or learn to work with decentralized laboratory services and spend a little more on data management and international audit expenses. (There are a good number of medium-and large-sized global central labs looking for solutions in different regions, such as Brazil, India, and Russia.)
The harmonization concept between global central labs and local central labs is an interesting one to be implemented and a viable alternative to be explored. Nevertheless, the costs of this harmonization process for local labs, including preventive equipment maintenance, should be taken into account to evaluate the feasibility of the whole concept. However, we are fully convinced that the implementation of a proper Quality Control program saves time, resources, and money, even for small laboratories.
I want to recognize the efforts of Mr. Didier Nicolas, retired from MDS Pharma Services, for developing and supporting the affiliated laboratory concept.
Dr. Francisco Leão, Jr, is with the division of clinical investigation at Fleury SA, São Paulo, Brazil, email: firstname.lastname@example.org
1. D. Burnett and C. Blair, "Standards for the Medical Laboratory—Harmonization and Subsidiarity," Clinica Chimica Acta, 309 (2) 137–45 (July 20, 2001).
3. K.L. Keatley, "A Review of US EPA and FDA Requirements for Electronic Records, Electronic Signatures, and Electronic Submissions," Quality Assurance, 7 (2) 77–89 (April–June 1999).
4. G.C. Giudi, G. Lippi, G.P. Solero, G. Poli, M. Plebani, "Managing Transferability of Laboratory Data," Clinica Chimica Acta, 374 (1–2) 57–62 (December 2006).
5. B. White, "The Impact of ISO 15189 and ISO 9001 Quality Management Systems on Reducing Errors," Vox Sanguinis, 83, Suppl 1, 17–20 (August 2002).
6. M.S. Bobka, "The 21 CFR (Code of Federal Regulations) Online Database: Food and Drug Administration Regulations Full-Text," Medical Reference Services Quarterly, 12 (1) 7–15 (Spring 1993).