The outsourcing of genetic analysis

Dr Stephen Little, ceo of DxS, discusses the use of genetic analysis in clinical trials. By using a specialised pharmacogenomics service provider, CROs can offer full genotyping support to their clients

Most clinical trials involve laboratory analysis of the subjects to ensure safety and/or to assess the behaviour of the drug.

As most sponsors do not have the facilities to carry out the work in-house, the analysis is normally outsourced to a specialist laboratory. Recently there has been considerable interest in including a genetic component as part of the laboratory analysis.

This article briefly reviews why genetic analysis is emerging as a valuable addition to the range of testing techniques and considers the issues involved in finding an appropriate laboratory to perform the work.

Clinical trials are expensive to initiate, perform and analyse so it is hardly surprising that methods that promise to increase the value or reduce the cost of a trial are increasing in popularity.

more information

Inherited variation within the full complement of human genes is responsible for all of our individuality, including the likelihood of developing a disease and response to therapy, and this knowledge can be used during the clinical trial. The main benefit of including genetic analysis during clinical development is that it gives an informative view of the individuals who make up the clinical trial population. This information can be used either prospectively to actively include/exclude individuals or retrospectively to correlate an individual's drug response to some aspect of his/her genetic make-up.

In some cases the size of the clinical trial can be reduced if a sub-group of patients likely to show improved efficacy can be selected. In other cases it is possible to correlate variation of drug levels with variation in drug metabolism enzymes

The majority of pharmacogenetic studies analyse inherited genetic markers. There are two main benefits associated with inherited variation. The first is that genetic information is predictive. The second is that the full complement of genetic information is carried by almost every cell in the body. This means that a simple blood spot or cheek scrape sample can reveal information about genes which are expressed only in, for example, the brain or the liver.

A good example of the retrospective use of genetic analysis is the issue of Long QT prolongation in clinical early stage clinical trials. Certain drugs have the ability to delay cardiac repolarisation, an effect that is manifested on the surface electrocardiogram as prolongation of the QT interval.

The combination of the ability of a drug to cause a drug's effect to prolong QT/QTc interval has resulted in a substantial number of regulatory actions, including withdrawal from the market (terfenidine, cisapride, astemizole, grepafloxacin), relegation to second-line status (bepridil, thioridazine), and denial of marketing authorisation (lidoflazine)

If long QT prolongation is observed in a trial subject then it is clearly in the trial sponsor's interests to determine if this may be due to hereditary factors in the patient's genetic makeup rather than being an inherent property of the drug.

Indeed, this analysis is recommended by the FDA in a draft discussion document.1 A second example showing the benefit of prospective genetic profiling is given on-line at www.manufacturing-chemist.info.

genetic benefits

Often genetic analysis will be requested by the sponsor as part of the study protocol and the drug company or clinical research organisation (CRO) will already have identified a suitable laboratory to perform the analysis. Occasionally there will be an opportunity to use genetic analysis to improve the value of the trial which the sponsor has overlooked. In these cases the CRO can propose a genetic component and thereby increase their chances of winning the study.

Many of the issues associated with genetic analysis outsourcing are common to any laboratory analysis and include topics such as quality standards, certification and turnaround time. Other questions are specific to genetic testing.

Normally suitability would be addressed by the sponsor during an audit visit prior to the first contract. As with all laboratory outsourcing, the critical issue is to ensure that the sponsor has complete confidence in the quality of the analysis provided.

In common with any type of laboratory analysis it is essential that the data that is provided is correct. At present there are very few genetic analysis products on the market so most if not all of the tests will have been developed in house by the laboratory. This means that the sponsor should expect to be able to see not only performance data for the assay but also a file of development data showing how the assay was designed and shown to be suitable for its purpose.

discrete variables

Much of the testing performed by a genetic service provider is SNP genotyping - i.e. determining the state of a single nucleotide at a single polymorphic site. For any given SNP with two alleles there are only three possible states. Both the maternal and paternal chromosome can carry the first allele (AA), the maternal and paternal chromosomes can carry different alleles (AB) or they can both carry the second allele (BB). The SNP genotyping is a discrete variable so the issue of reference ranges associated with biochemistry are not usually relevant.

Not all genotyping laboratories are able to offer a full range of analyses. Ideally a laboratory would offer nucleic acid extraction and banking, SNP genotyping, micro-satellite genotyping, RNA profiling and DNA sequencing.

The genetic analysis laboratory may have specialised techniques and technology but many of the laboratory procedures are common to any central laboratory. The client needs to know:

• How samples are received and labelled and what checks are made to ensure the correct sample is being analysed;

• What LIMS system is in use and how does this control and monitor the laboratory processes;

• What control procedures are in place to deal with any discrepancies which may arise;

• How is the report generated from the raw data;

• What happens to samples following analysis;

No national or international agreement exists yet regarding the certification of clinical laboratories that carry out testing on patients in studies - this is in contrast to pre-clinical work which is regulated by GLP. In the absence of firm guidelines there are two considerations

First, the service provider must operate quality systems compliant with the requirements of the ICH guidelines on Good Clinical Practice. The ISO9001:2000 quality standard is internationally recognised and shows the laboratory meets these standards.

Second, the pharmacogenomic analysis may well be used to make inclusion/exclusion decisions regarding the trial. In this case the test is being used as a diagnostic and additional regulations apply. For example, in Europe (from November 2003) diagnostic services should be covered by the CE mark and in the US diagnostics of this type are covered by standards known as CLIA'88. The service provider should meet these standards

spot testing

Time is of the essence when the data will be used as exclusion/inclusion criteria for a study. For post trial analysis the time pressure is much less. Some laboratories are able to offer 24-hour turnaround if required, but to this must be added the time to ship the samples to the laboratory. For a multinational study this creates a problem - one solution is the DxS Genetwork - this is an international alliance of reference laboratories, all of which are able to offer the same genetic analyses using a common set of reagents. Tests can be shipped to the nearest Genetwork laboratory and the use of common reagents and systems guarantees a consistent and reliable result.

International clinical trials can take place at trial sites dispersed among many countries or continents. The collection and transport of samples can be greatly simplified by selection of a service provider who offers testing on dried blood spots. Using sensitive genetic analysis systems a single dried blood spot can provide sufficient genetic material for several hundred separate SNP tests.

Blood spots collected and dried on the specialised solid support contained within the card are sterile, safe and stable at room temperature for many years, making it the ideal material for pharmacogenomic analysis

This approach is particularly beneficial in paediatric trials where it may be very difficult to obtain sufficient blood for large scale DNA preparation.

Often the inclusion of a pharmacogenomic element within a clinical trial will involve a number of decisions and options that are relatively unfamiliar to the sponsor. It can be valuable to select a service provider with expert knowledge to help in the design, ethics committee approval, implementation and interpretation of the trial.