Abstract
Biomarkers, pharmacokinetics, pharmacody-namics and immunogenicity data are some of the important information obtained from clinical trials. Obtaining such data, however, can be very cash-, time- and resource-consuming. Ensuring that appropriate mea-sures are taken from the sample collection stage through to the completion of laboratory testing procedures is therefore vital. In this article Haiko Pillu, Director Operations CPU at SGS, explains how to obtain reliable and meaningful results from clinical studies by optimizing the bioanalytical procedures employed.
When clinical trials were first developed to test the effectiveness of drugs in treating diseases, it was usual to carry out tests using multiple drug samples, including for pharmacokinetics (PK) and other assessments such as pharmacodynamics (PD). Samples were taken individually, and the time required for processing each was limited. Typical preparation techniques included centrifugation, aliquoting and freezing. Other preparative techniques would be employed when larger sample preparation flows were necessary.
Sample testing requirements and procedures have now evolved to the stage where today’s clinical studies are routinely larger and more complex. Such studies include PK assessments, more specific sample treatments for PD studies, and the performance of several sequential sample preparation and testing procedures during these specific treatments and processes. The additional complexity means it may now be necessary to allocate up to five hours’ laboratory time to process one, or a batch of samples. These sample or aliquot batches may now be quite large and it is possible that up to ten different assays may be performed on a sample at any one time. In addition, each assay requires its specific test conditions, making treatment schedules more time-consuming. To cope with this added process complexity, and the need to implement new techniques, pilot studies are often undertaken to ensure staff are properly trained and qualified before the full set of analyses required are performed.
Laboratory manuals are now more specific and demanding as a result of these new sample preparation and test requirements. As well as specifying typical sample prep-aration and test conditions, they include more requests, for example, the specification of the timelines taking a sample up to the centrifugation stage, as well as specification of timelines after centrifugation, the storage conditions required, the matrix used, and any other handling steps or special procedures required. This results in more challenging requirements for maintaining repeatable and consistently good sample quality, and for planning of the allocation of human, material and monetary resources.
Some of the questions that need to be answered are: why do we need to set all these parameters? How do we implement these more complex sample treatments? What is the rationale behind this approach? How can we be sure it is effective, and how will the information obtained be delivered from the sample testing laboratory to the clinical site?
Setting the Right Parameters
In an example study from the field of veterinary medicine, a case study entitled ‘The effects of anticoagulant choice and sample processing time on hematologic values of juvenile whooping cranes’ from 20101 describes the procedures used to collect blood from these birds, and the dependence of the subsequent blood test results on a number of factors. The study employed two anti-coagulants, K3 EDTA and lithium heparin (LiHep), with slides being made either immediately or after a period of four to six hours. Afterwards, the test results were analyzed to detect any correlation between the anti-coagulant used and the sample processing time.
The total granulocyte concentration (hetero-phils and eosinophils; H/E concentration) of each sample is shown in Table 1.
Table 1. Statistics (mean ± SD and range) for hematologic values of paired whooping crane blood samples processed immediately by anticoagulant (n = 15).
Relative (%) leukocyte counts of each sample are shown in Table 2.
Table 2. Statistics (mean ± SD and range) for hematological values of paired EDTA- and LiHE-treated blood samples with blood smear preparation completed.
In this specific study, the anti-coagulant used had no effect on the results when sample processing was carried out immediately. However, any time delay in processing the samples gave different results for the two anti-coagulants used.
The Importance of Validation Procedures
As illustrated by the above case study, laboratory testing techniques have to be validated and specific parameters checked regularly if meaningful test results are to be obtained. This multistep process is key in ensuring that final results are fully ‘appropriate and correct’ both in terms of the demands of the analysis and the reliability and reproducibility of the results.
As an example, in the validation of a method for testing peripheral blood mononuclear cells (PBMCs), the first step is to implement and qualify the method, initially by setting up and testing the protocol, which includes testing PBMC stimulation and labeling procedures. Lysis and fixation processes are determined and a permeabilization buffer selected. Preliminary testing of the reproducibility of the method can be carried out when acquisition and analysis templates have been set up and a gating strategy decided.
The second part of the method validation procedure is the in-vitro feasibility testing of the effect of the product on key T cell functions, including CD4 Th1/Th2 and T Regulatory response. Culture conditions and stimulation parameters such as the reagents, stimulus, duration of culture and duration of stimulation are specified, and the dose- and time-effect of the product determined (n= up to six subjects). The in-vitro effect on both CD4 Th1/Th2 and T Regulatory responses is then tested. A duplicate sample for fluorescence-activated cell sorting (FACS) testing is prepared using identical culture conditions and the final step in the study is the processing of the data and statistical analysis.
The third part of method validation is the confirmation of the reliability of FACS testing for analysis of the CD4 Th1/Th2 subset (CD3, CD4, IFN-Gamma, IL-4) and T Regulatory (CD3, CD4, CD25, FoxP3, CD127) response. It includes the determination of the sensitivity of the method and its reproducibility between replicates, runs, analysts, donors and FACS systems. Stability testing, for example the storage of PBMC and monitoring any effects of cryopreservation upon samples, needs to be performed; the inherent stability of reagents determined; and the FACS method adopted shown to be robust.
The fourth part of the method validation procedure is the transfer of the method to the nominated clinical site. As well as providing the procedure itself, this includes the training of clinical site staff in handling procedures, test conditions, and the “gos” and “don’t gos” associated with the protocols. This requires the performance of a pilot study using replicate samples, several analysts taking part, in order to obtain data enabling staff evaluation, and method and training validation. The clinical trial can begin once the results are analyzed and reviewed.
These four steps would each take typically up to one month to perform, the first three being carried out in the bioanalytical laboratory, and the fourth performed at the clinical site undertaking the trial. The fifth and final part of method validation requires much more work to be done in the bioanalytical laboratory and may possibly take up to several months to complete. PBMC stimulation studies, CD4 Th1/Th2 (CD3, CD4, IFN-Gamma, IL-4) FACS analysis, and T Regulatory (CD3, CD4, CD25, FoxP3, CD127) FACS analysis are usually undertaken on more than 200 samples.
Additional Test Procedures
Depending on the particular requirements of specific individual studies, more testing and validation procedures may have to be carried out in order to ensure the stability, reproducibility and robustness of bioanalytical samples. For example, for urine collections, a surfactant such as Tween, may have to be added to the sample to avoid interference in tubes; and it may be necessary to use a non-standard blood collection tube in the determination of cytokines. For tests that require sputum induction, the effect on end parameters of using for example Sputolysin during handling, needs to be evaluated, as it may alter the constituents and viscosity of mucus; and the effect of the presence of binding factors on specific compounds and parameters needs to be assessed when choosing matrices. Any effects of snap freezing at -20°C or -70°C on sample quality have to be determined when cold storage of samples is required.
Summary
In the performance of bioanalytical tests, it is important to carefully consider all of the variables of the process, including the selection of sampling tubes, anti-coagulants, test conditions, sample processing times before testing, and the sample processing and test methods used. Validation of techniques is absolutely vital in ensuring the reliability of test results. Implementing these procedures may appear to be time-consuming, but the work is invaluable. This is especially the case when it is supported by good communication between staff and when they are provided with appropriate training at both the bioanalytical laboratory and the clinical site. This ensures that good final results will be obtained and that staff understand the potential pitfalls there may be in attempting to achieve a successful bioanalytical outcome.
The corollary is that the choice of clinical trial partner is key to the overall success of the trial, and their experience and capabilities should be fully evaluated. To design a good technique that produces excellent data from a clinical trial, specialists developing its bioanalytical features have to determine the experimental set-up and the validation techniques, as well as work with the specialists at the clinical research site responsible for carrying out test procedures. The performance of pilot studies between the clinical site and the bioanalytical lab is the best way to determine the appropriate sample handling procedures, identify optimal conditions for running tests, and ensure analysts have the right experience for performing these tests. Establishing effective and efficient communication between the lab and the clinical site creates synergies between the two that can be exploited to ensure the optimal outcome for clients.
References
- Maurer, Joan; Reichenberg, Betsy; Kelley, Cristin; and Hartup, Barry K., "The Effects of Anticoagulant Choice and Sample Processing Time on Hematologic Values of Juvenile Whooping Cranes" (2010). North American Crane Workshop Proceedings. Paper 125.