Proteomics and severe acute respiratory syndrome (SARS): emerging technology meets emerging pathogen.

Early recognition and rapid initiation of infection control precautions were the most important strategies for controlling severe acute respiratory syndrome (SARS) during the global epidemic that occurred from March to June of 2003. This approach was hampered, however, by the fact that clinical and laboratory features did not distinguish patients with SARS from those with other respiratory illnesses and that there was no reliable rapid diagnostic test (1)(2)(3).

Currently, the “gold standard” for the laboratory diagnosis of SARS coronavirus (CoV) infection is antibody detection by indirect immunofluorescence assay or ELISA. The median time to seroconversion in SARS patients, however, is 17–20 days after the onset of symptoms (4); therefore, rapid diagnosis by antibody detection is not possible. ELISA-based antigen detection tests are well known to offer high specificity and reproducibility, but they lack sensitivity (5). As for the direct detection of viruses, culturing of the virus from clinical specimens is dangerous and insensitive; the detection of viral RNA by reverse transcription-PCR is expensive and labor-intensive and relies on the availability of expertise, and false-positive results may result from contamination. In addition, although SARS-CoV can be detected in nasopharyngeal specimens, sputum, and stool, peak viral shedding from these sites, and therefore the most optimal time for testing, occurs around 10 days after illness onset and then decreases (4)(6).

In this issue of Clinical Chemistry, reports by Kang et al. (7) and Yip et al.(8) present evidence that a new proteomics technology, surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF), may provide a solution to these problems. SELDI-TOF is a proteomic technology involved in quantitative analysis of protein mixtures. This technique uses trapping surfaces that allow differential capture of proteins based on the intrinsic properties of the proteins themselves. It can yield comprehensive profiles of peptides and proteins in biological fluids without the need to first carry out protein separations. Microliter quantities of samples are applied to the surface of a protein-binding plate with physical properties that bind a class of proteins. The bound proteins are laser desorbed and ionized for mass spectral analysis. Masses of proteins ranging from small peptides of <1000 Da up to proteins of >300 kDa are calculated based on time of flight. As mixtures of proteins are analyzed within different samples, a unique sample fingerprint or signature will result for each sample tested. In the case of SARS and other infections, these fingerprints reflect proteins specific to the infectious agent itself, as well as proteins related to the host’s response to that infectious agent.

Kang et al. (7) were not only able to show that SELDI-TOF was sensitive and specific (97.3% and 99.4%, respectively) as a diagnostic test for SARS, but they were able to do it with a small quantity (20 μL) of serum, without the need for high-level biosafety facilities, and could provide results within 3 h of initiating testing. Additional important findings were that the tests were positive early in the course of disease. All seven SARS patients tested within 24 h of onset of fever were positive, and patients could be grouped according to the severity of their disease. The ability to rapidly diagnose patients could allow for the early implementation of appropriate infection control measures and treatment. Being able to predict the clinical course of illness could also allow optimal tailoring of therapy, with more potent but toxic agents being reserved for those with more severe disease.

Yip et al. (8) also showed that patients infected with SARS have a unique protein profile. They went a step further and identified one of the SARS protein biomarkers as serum amyloid A and showed that increased concentrations correlated with increased severity of disease. Characterizing the proteins that make up SELDI-TOF profiles in this manner not only provides insight into the pathophysiology of the disease, but may also lead to the development of more direct diagnostic tools and novel therapeutic targets relevant to SARS-CoV.

As promising as SELDI-TOF appears to be based on these two studies, there are limitations in both studies that are worth commenting on. For example, Kang et al. (7) determined the sensitivity and specificity of their assay by comparing its use in SARS patients with that in control patients. The controls included healthy persons as well as persons with other infections. Although it was important to use healthy patients as controls during the initial validation of these assays, from a diagnostic perspective including these types of controls in the calculations of sensitivity and specificity of an assay may artificially enhance the apparent sensitivity and specificity of the assay. From an infectious diseases perspective, one is not trying to separate healthy patients from infected ill patients, but rather one is trying to identify the cause of symptoms in patients presenting with similar symptoms to aid in the selection of the most appropriate therapy for a specific organism.

In addition, in both studies (7)(8), the patients with SARS were well characterized in terms of timing of specimen collection and severity of disease. This was not the case for the control patients who had infections other than SARS. This raises the question of whether the pattern obtained in SARS patients is unique to SARS or whether a similar pattern would have been obtained with the other infectious diseases had the patients been matched for severity of disease and timing of specimen collection. As reported by Yip et al. (8), serum amyloid A concentrations remained increased in one patient because of a superimposed bacterial infection, suggesting that a severe bacterial infection could give the same increase in amyloid A as SARS.

Although Kang et al. (7) and Yip et al.(8) chose to apply SELDI-TOF to the diagnosis of SARS, there are many other, more common infectious diseases, such as influenza, that would also benefit from more rapid and reliable diagnostic and prognostic tests. This need has been made more acute by the expanding availability of therapeutic agents and infection control implications for many of these diseases. It would be of interest to determine whether SELDI-TOF reveals profiles specific or prognostic for these other, more common infectious diseases. Such profiles may be able to be used to develop a clinically useful diagnostic algorithm for patients presenting with cough, fever, and a pulmonary infiltrate.

Before SELDI-TOF can be implemented into a clinical laboratory, there are factors, including those outlined above, that need to be addressed. In particular, the expected variations in profiles need to be determined for healthy persons without infection, for persons with inflammatory conditions without infections, and for persons with a variety of different infections at different time periods in their course of illness. In addition, instrumental and analytical costs, availability, and expertise need to be considered.

Notwithstanding these limitations, SELDI-TOF has numerous potential applications that make it an attractive emerging technology. Theranostics is a new field linking rapid diagnostic tests to appropriate treatment choices and response to therapy as a means to optimize patient care. SELDI-TOF and the field of proteomics clearly meet the expectations of this newly emerged concept (9). Not only can SELDI-TOF aid in the diagnosis of infections caused by known agents, it also has the potential to aid in the diagnosis of newly emerging infectious diseases before the etiologic agent is known. In addition, it can offer insight into prognosis, provide rapid real-time data regarding response to treatment, and help identify targets useful for the development of specific diagnostic tests and specific therapeutics. Clearly, SELDI-TOF and the field of proteomics are emerging technologies that have the potential to positively impact the field of infectious diseases. We encourage more studies to be completed in this area.