Sampling the winners circle
Californias Equine Analytical Chemistry Lab uses state-of-the-art LIMS in testing for drug abuse.
And theyre off! Legendary Churchill Downs in Louisville, KY, was the site of the November 4, 2000, Breeders Cupthe most significant horse racing event in the United States and the gold standard for equine chemistry testing. Ten races showcased horses from juvenile fillies to seasoned stallions with a total purse of nearly $15 million. The most important race is the last, the Breeders Cup Classic, a 1.25-mile contest among the worlds best horses for a $4 million purse. The winners share is $2.4 million. If chemical testing on the winning horse reveals a violation (referred to as an analytical finding), the purse must be returned and protests would be lodged at every possible level.
During any particular week, states Scott Stanley, equine analytical chemist for the Kenneth L. Maddy Equine Analytical Chemistry Laboratory (EACL) at the University of California in Davis (UC Davis), we could be testing samples from a race with a $1 million purse. Only about one-tenth of one percent of samples fail test finding, but when that occurs, a chain of events is set in motion. (See box, Run Swiftly and Carry a Big Purse.)
If there is an analytical finding, the lab has to protect itself in court and the documentation has to hold up to legal scrutiny, Stanley emphasizes. Chain of custody in particular is critical.
Run Swiftly and Carry a Big Purse
|In every state that has authorized horse racing, a state racing commission closely regulates it. In 1999, more than $3 billion was paid to holders of winning tickets in California alone, or 80% of the pari-mutuel pool, with the rest of the monies given to track operators, owners, and other entities. The temptation to manipulate the results of a race is significant. Equine testing labs play a critical role in preventing fraud.
Typically, the purse money goes to the top five horses in pari-mutuel wagering. But any horse in the race can be selected for testing by the race stewards after the race. The top three horses are always tested, but the favorite might be drugged to lose and so a horse that does not run as expected would also be tested.
Many steps are taken to ensure the integrity not only of race results but also of the samples taken from the horses after each race.
The Critical Role of the Lab
EACL is part of the California Animal Health & Food Safety (CAHFS) laboratory system at UC Davis and supports the California Horse Racing Board (CHRB) by providing analytical chemistry testing to the performance horse industries in California.
EACL tests urine samples for drugs of abuse and monitors blood samples for the presence of unauthorized medication. In addition, the lab performs research on therapeutic medications to determine whether their use can alter the outcome of races or interfere with drug screening procedures. For instance, the lab previously studied methocarbamol, a mild muscle relaxant, and two anti-ulcer medications, ranitidine and cimetidine, to determine their effect on athletic performance. Testing protocols were developed and residue levels determined. When one of those substances is detected below the level established by the CHRB, that substance is considered an authorized medication. When results are above the level, its not.
A new state-of-the-art EACL testing and research facility opened in December 1999 that includes eight laboratories equipped with GC-MS, LC-MS, automated sample testing equipment, and automated sample processors. The increase in productivity has been dramatic. It takes the automated robotic immunoassay sample processor about two and a half hours to run seven tests, reports Stanley. Its like having three analysts in-house to do the work. We didnt replace anyone, but we are now able to implement tests that we couldnt do before.
Extensive integration between instruments, rigorous chain of custody documentation, and a high level of security were key requirements in designing the laboratory information management system (LIMS). To support the new facility, EACL selected a 10-user LabSystems Nautilus LIMS that went live on July 5, 2000.
EACL is unique among equine testing labs in the United States in its use of LC-MS and GC-MS. The de facto industry standards have been thin-layer chromatography (TLC) and enzyme-linked immunosorbent assays (ELISAs). EACL also performs ELISA tests but finds GC-MS and LC-MS faster and more efficient than TLC.
Before the launch of the new lab, all samples were sent out of state. EACLs goal is to process at least 20,000 samples yearly. All positive samples with evidence of drugs require a supervisor to review and authorize the results. Before LIMS, Stanley spent 50% of his time reviewing these results. I found that we had to re-aliquot samples mainly because the process failed rather than the sample. Now, we can easily provide preliminary reports by facsimile within 24 hours of when the sample is received.
From Sample to Result
On some racetracks, as many as 70 horses run each day and must be evaluated by the racetracks veterinary. Previously unsound horses must prove to the veterinary that they are sound and qualified to run. Clock times from morning workouts are critical for establishing soundness and setting the stage for afternoon betting.
Immediately following a race, the racetracks stewards select the horses to be tested, and veterinarians take samples from those horses. The samples are barcoded at the track and typically are received by EACL within a few hours of the race. Once the sample is received by EACLs secured sample processing lab, it never leaves that lab. Only the aliquots travel to the testing labs.
EACL uses both the racetracks external tracking number and an internal tracking number for sequential tracking. Samples are logged in by the client and by a barcode. Random quality control (QC) samples that are indistinguishable from other samples are inserted throughout the testing process. Many QC samples include drugs of abuse both as a system test and to ensure that all facets of the system are operating accurately. All samples are tracked in batches and individually. The location of each sample is tracked through worksheets. The LIMS allows different levels of access to the worksheets. Thus, the technician in the sample processing lab can log samples but cannot review results.
Solid-phase extraction and immunoassays are performed. Once processed, samples are kept in a large freezer located in the sample processing lab for 30 days after testing. Samples that test positive are kept indefinitely.
In addition, to ensure sample purity in the processing lab, no drugs or reference materials are allowed into this area, and it is locked down from the rest of the site.
The anonymous nature of the barcode process plays a key part in the sample-handling process. Whether a stake or claim race, the lab has no way of knowing the trainer, race, or even the horse associated with a sample.
Much like a forensics lab, EACL must ensure the integrity of its results with absolute certainty. Hence split samples are always used: Half of the sample taken at the racetrack is sent to EACL, and half is sent to CHRB headquarters in Sacramento, CA, for storage and backup in case the EACL results are disputed.
After leaving the sample processing lab, the aliquots travel to the extraction lab. During sample extraction, acid-neutral and base fractions are extracted from the aliquot. EACLs process uses an in-house nitrogen system to reduce sample oxidation. Three internal standards are added at the same time as the reagents.
Next, the sample is forwarded to instrument screening, where the results of the sample analysis are available in the LIMS. Samples are tested for about 70 compounds, including steroids and caffeine. Internal standards verify that the sample was extracted and injected correctly.
GC-MS is used on the base fraction to test for depressants and tranquilizers, while LC-MS is used to test the acid-neutral fraction. Several hundred compounds are searched for simultaneously, such as codeine, morphine, and procaine. Reports are generated with the results.
One of the reasons for equipping our lab with GC-MS and LC-MS, Stanley points out, is that adding new tests for new drugs to the process is easy. Its not easy to do with TLC.
If the base fraction report highlights a positive sample, and it is not a QC sample, the aliquot is sent to the confirmation lab for further testing. Suspect samples are tested by a different analyst, and 26 individual analytical runs are performed.
When a positive sample is confirmed, the report is provided to the CHRB. When this happens, not only is the purse taken away, but the trainer is fined and his or her license may be suspended, depending on the severity of the violation. Suspension may last anywhere from 15 days to 1 year. Needless to say, with a purse and a reputation at stake, Stanley is subpoenaed to testify in court in almost every case.
In court, the sample-handling process is rarely a deciding factor. Instead, the trainer generally seeks to mitigate the circumstances by attempting to prove that the drug did not affect the performance of the horse. There is some justification for this position, explains Stanley. Lower levels of a drug may have been in the horses system for some time, or a level that might affect the performance of one horse would not have mattered with another. Horses are as individual as people in how they react to a substance.
Implementing a LIMS that provides electronic verification of sample information and processing while tightening the chain of custody now enables EACL to not only better support equine testing but also prepare for future requirements.
LIMS Enables Expansion
EACL started looking at LIMS in 1997 and spent a year and a half in the evaluation process. We were looking at LIMS when the Internet came to prominence and a lot of things changed in the LIMS environment, recalls Stanley. One of our greatest fears was that the LIMS we chose would be obsolete before we could implement it!
EACL narrowed its search to three LIMS vendors whose solutions scored high in marketplace surveys. They chose LabSystems Nautilus solution, which met all of their criteria. The most important requirement for us was that the system be aliquot-driven from start to finish.
LabSystems close collaboration with HFL, Ltd., during the development phase of Nautilus was also a critical deciding factor. We interviewed HFL and visited their labs in the U.K., states Stanley. These discussions helped drive a lot of considerationsboth pro and conthat taught us a lot. We incorporated steps into our own process based upon HFLs processes. And, we created steps that they did not do but which were appropriate for our workflows.
We knew it was a big project, but we didnt know how much work we would have to do ourselves. You still need to define a lot of steps and information for the vendor regardless of their experience with your type of lab, he stresses.
According to Stanley, one of the most important lessons learned is not to underestimate implementation. Planning to implement in 30 to 60 days just wont happen. Six months is more feasible. We have eight labs, took six months to implement the LIMS, and were still learning to use it more efficiently.
After we got the software, we basically didnt do anything for four months. That wasnt the plan, of course. Work was started, but by one person who was working part time. We soon realized the scope of the project and put two people on the effort full time, he continues.
In essence, however, the effort took place from January to July 2000. Now that we have it implemented, it doesnt need a lot of resources to keep the system runningjust a full-time LIMS administrator.
Some 20 laboratories in the United States perform drug testing on racing horses and dogs. With the exception of EACL, TLC is the test labs methodology of choice. TLC has broad spectrum capabilities; however, it is labor-intensive. LC is much more efficient, but the transition to LC-MS is a major step and an expensive one.
And things are changing dramatically right now, Stanley points out. Chip technology will probably become reality in the next five years. This has the potential to take technology down to the 50 to 100 nanoliter rangeso we could test just two to three drops per sample and still be ensured of accurate results. However, lots of R&D will need to be done in this fashion before we can take advantage of it.
With the new lab, we took existing technology and modified it to fit our needs, he explains. We were able to go from 5 to 1 microliter concentration using a 40-microliter sample. Thats five times the reduction, still with conclusive results.
What is interesting is that were the first lab in the world to integrate the LIMS in the way we have, Stanley adds. Were ahead of where I thought wed be at this point.
By planning for an uncertain future, EACL has set a standard for getting ahead and staying ahead in maintaining a rigorous sampling process that will ensure its presence in the winners circle.
Helen Gillespie is an industry analyst, editor/publisher of the LIMS/Letter, and webmaster of LIMSource.com. Send your comments or questions regarding this article to firstname.lastname@example.org or the Editorial Office 1155 16th St N.W., Washington, DC 20036.