| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorial |
1 Department of Clinical Biochemistry St. Bartholomew’s and the Royal London School of Medicine and Dentistry Turner Street London E1 2AD, United Kingdom Fax 44-20-7377-1544 E-mail c.p.price{at}mds.qmw.ac.uk
Multiplexing is a relatively new addition to the Oxford English Dictionary and in the encyclopedic version is defined as "the simultaneous transmission of several messages along a single channel of communication". It is also a relatively new addition to the vocabulary of the clinical chemist, having been imported from the lexicon of the genomics, proteomics, and drug discovery scientists, but it is not a new concept for the clinical chemist who has been exposed to the use of so-called multichannel analyzers for many years, first continuous flow and then discrete analyzers. Some historians of automation might look back to the centrifugal analyzer as the first challenge to multichannel analysis, but there is now at least one example of this technology that provides a profile of analytes on a single sample (1). If these are indeed early examples of multiplexing analyzers, then they distinguish themselves—by today’s standards—by using large amounts of sample and reagents, by a choice of analytes that are not always clinically justified, and by an inability (with a few notable exceptions) to embrace immunoassay and other ligand-binding assay techniques for low-concentration analytes. However, the greatest legacy of the first era of multiplexing analysis is probably the perception that there is too much laboratory testing as well as numerous apocryphal stories of patients being investigated after the generation of an abnormal result, rather than for the presenting symptoms!
So with the advent of microarray platforms, are we seeing the reincarnation of multiplexing analytical strategies and the evolutionary cycle turning full circle? The concept of the multianalyte array based on immunologic capture and interrogation technology is not new, with an editorial on the subject in this Journal almost a decade ago (2). So what has been achieved in the intervening period? Certainly the interrogation of captured molecules using antibodies with different labels was always going to be limited by the number of labels that could be detected simultaneously (3). We are still debating the relative merits of capturing the analytes of interest on a single surface (4), on multiple labeled particles (5), and on chromatographic devices (6). What has definitely changed is the printing technology to produce discrete capture zones, miniaturization of reactant volumes, and signal interrogation technology. Some of these developments are illustrated in the work described by Wiese et al. (7) with the performance of a microarray system for prostate-specific antigen (PSA) and its complexed isoform, together with interleukin-6.
The technology described by Wiese et al. (7) uses a 250-pL spot 250 µm in diameter, which compares with configurations reported by Silzel et al. (8) and Ekins (9). The latter author in fact described performance on spots 10–60 µm in diameter. There are several potential benefits from this "size" of reaction area, including low requirements for reagent and sample, the latter being particularly attractive to the pediatric specialist. This degree of miniaturization also has the potential to enable a reduction in reaction times as a consequence of the reduction in diffusion distances, as well as minimizing the constraints associated with the use of an immobilized capture molecule. Reduction in the size of the reaction area will also improve the signal-to-noise ratio (9). Some of these points are not evident in the system described by Wiese et al. (7), but must be achieved at some point to attain the real application goals for this technology, namely the simultaneous assay of several relevant analytes, point-of-care testing on small samples, or both, with the production of results in 2 min or less.
The ability to print and interrogate multiple discrete spots in a small area must be one of the other major advances in this field in the last decade. The advent of highly efficient fluorophores and of the charge-coupled device (CCD) imaging camera with the ability to attain sensitivity of detection and spatial discrimination has enabled several groups to achieve limits of detection comparable to "macro volume" laboratory methods. Thus Wiese et al. (7) were able to detect 0.31 µg/L for total PSA, and Silzel et al. (8) were able to detect IgG3 down to 15 µg/L. Scorilas et al. (10) claimed a detection limit of 0.001 µg/L for total PSA.
Wiese et al. (7) were able to discriminate between reaction zones at 300-µm centers, enabling a large number of zones to be printed on a small area. This has produced two advantages, the first being that multiple zones can be used for the same analyte to improve precision through signal averaging while also reducing the failure rate that might result from using a single zone. The group also used a capture antibody titration design to extend the analytical range of the assay.
Where might this technology go and how will it be applied in clinical practice? Certainly there are advantages to using capture and interrogation antibody-based systems, the combination of detecting two epitopes improving the specificity of the assay. However, antibody specificity may have its limitations, the clinical situation sometimes demanding both nonspecific and highly specific analyte detection. It has yet to be shown whether antibody specificity will meet some of the testing demands when researchers are looking for isoforms or variants of some proteins. Thus, the potential to link the microarray capture with alternative interrogation systems, such as mass spectrometry, has excited a great deal of interest. This analytical strategy has been described for the detection of prealbumin variants in trying to determine their relevance to the deposition of amyloid (11) as well as in probing the variant forms of PSA (12).
It is also possible to consider alternative capture strategies, the ultimate being to etch the paratope or capture zone directly into the microarray surface. Shi et al. (13) have described the design of a capture pocket constructed in a surface. In this case the analytes of interest included albumin, IgG, and fibrinogen, and the interrogation technology was atomic force microscopy. This type of development may enable the etching of capture zones to create a unique profile of tests for an individual patient. Is this then discretionary microarray analysis?
So how will the microarray system be applied in laboratory medicine? Clearly the miniaturization has considerable appeal in the field of point-of-care testing and in pediatric medicine. There are already some established screening and diagnostic profiles, including allergy tests (14) and drugs of abuse (6). Wiese et al. (7) illustrate the application to the early detection of prostate cancer. The limitations of the total PSA measurement are currently being explored with the introduction of isoform measurement, but we do not yet know whether the best strategy will be an initial measurement of total PSA followed by a cascade of tests, including isoforms, or the use of a profile of tests. Although the diagnostic application of interleukin-6 is less well understood, other candidate markers for prostate malignancy are being proposed. There certainly are additional clinical questions that warrant expansion of the battery of tests available on such an array; examples include the cytokines, metalloproteinases, and their inhibitors as prognostic indicators for metastatic infiltration and bone alkaline phosphatase as a specific marker of bone infiltration.
The isoform profile of human chorionic gonadotropin (hCG) and the association of different isoform patterns with certain diseases demands the ability to probe the profile. There is already evidence in the literature of the limitations of many assays for the various forms of this protein, with the potential to distort our perspective on the clinical application of these analytes (15). It is also important that such an array can provide an assessment of all of the isoforms—a "total" hCG assay. This may not be achievable with an immunometric style of assay because of the diversity of the isoforms and the inability to find two common epitopes in all of the isoforms; an alternative form of interrogation might be more appropriate.
Finally, screening of the newborn is a particularly attractive application for a microarray, even at the point of care, because a sample could be collected for confirmatory tests in the event of a positive screening test. This application presents technologic challenges at the moment because some of the current markers are low-molecular weight analytes that are not amenable to capture and detection in the format described by Wiese et al. (7).
Thus, within the realms of diagnostic laboratory medicine, there clearly are some exciting potential applications for microarray technology. This will be further enhanced by the information deriving from the Human Genome Project and the recognition of more pathophysiologic mechanisms that will generate new markers, identification of variant forms of markers, and associations between markers. The technology itself has the potential to deliver many operational benefits with smaller sample requirements and reduced assay times. Nonetheless, many questions remain unanswered after the work of Wiese et al. (7), in particular the possibility of "carryover" between capture zones (the effect of juxtaposition of high- and low-concentration samples), the analytical imprecision, and the reproducibility of array manufacture. These questions should not detract from the exciting possibilities that this technology offers. However, the appropriate choice of analytes, based on careful clinical and economic analyses, will be the most important determinant of the success of this technology.
References
-hydroxyprogesterone, immunoreactive trypsin, and creatine kinase MM isoenzyme in dried blood spots. Clin Chem 1992;38:2038-2043.[Abstract]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. Binder Autoantibody Detection Using Multiplex Technologies Lupus, July 1, 2006; 15(7): 412 - 421. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
