A Simple Blood Test for Diagnosing Invasive Aspergillosis?
A Simple Blood Test for Diagnosing Invasive Aspergillosis?
Abstract & commentary
Synopsis: An ELISA test for Aspergillus galactomannan may prove to be a sensitive, noninvasive tool to assist in the early diagnosis of invasive aspergillosis.
Source: Maertens J, et al. Screening for circulating galactomannan as a noninvasive diagnostic tool for invasive aspergillosis in prolonged neutropenic patients and stem cell transplantation recipients: A prospective validation. Blood. 2001;97:1604-1610.
A sandwich elisa method is commercially a-vailable for detecting galactomannan (GM), which is a component of the cell wall of the Aspergillus species found in serum plasma and other sterile body fluids. The test involves using the rat monoclonal antibody EB-A2, which recognizes the 1->5-ß-D-galactofuranoside side chains of the GM molecule to both capture and detect as little as 1 ng/mL GM.
This study set out to validate this GM-ELISA test prospectively by surveying 362 treatment episodes in 191 neutropenic patients and recipients of a hematopoietic stem cell transplant (HSCT) for the presence of sinus and pulmonary signs and subjecting them to a standard diagnostic work-up for invasive fungal infective disease. Serum was collected at least twice weekly from admission to discharge or death or, in the case of HSCT recipients, once weekly as outpatients until immunosuppressive therapy was stopped altogether. The results were not used to diagnose disease prospectively or to decide therapy. Instead, aspergillosis was defined according to the criteria proposed by the EORTC/MSG in which aspergillosis is considered as being proven only when there was histological evidence of tissue invasion and recovery of Aspergillus species from culture. This is probable when there is both clinical and mycological evidence (see Table) or possible if there is either clinical or mycological evidence.
Table-Evidence Required for Defining Cases of Invasive Aspergillosis | ||
Site | Clinical Evidence |
Mycological Evidence |
Pulmonary | a) Any 1 of the following: |
Any 1 of the following: |
CT scan |
1) Microscopic demonstration of septate, |
|
1) halo sign |
acutely branching hyaline hyphae |
|
2) air-crescent sign |
2) culture of Aspergillus species from sputum |
|
3) cavity within an area |
or bronchoalveolar lavage |
|
of consolidation |
||
or |
||
b) Any 2 of the following: |
||
1) symptoms of lower respiratory tract |
||
infection (cough, pleuritic chest pain, |
||
dyspnoea, or haemoptysis) |
||
2) pleural rub |
||
3) a new infiltrate other than those considered |
||
as evidence of aspergillosis but for which |
||
there is no alternative diagnosis |
||
Sinuses | a) Radiographic evidence of invasion |
None |
or |
||
b) Any 2 of the following: |
||
1) upper respiratory symptoms |
||
2) nose ulceration, eschar, epistaxis |
||
3) periorbital swelling |
||
4) maxillary tenderness |
||
5) necrotic lesions or perforation of hard palate |
Use of these criteria allowed cases to be classified on the basis of premortem data. Autopsy was also vigorously pursued and performed unless explicitly refused by the patient or his family. This allowed a more accurate estimate of the incidence of proven cases than would otherwise have been possible since invasive procedures were avoided in thrombocytopenic patients. The GM-ELISA was conducted according to the manufacturers instructions, and GM was considered present when detected in 2 consecutive serum samples.
The diagnostic use of the GM-ELISA test was assessed using both premortem data and after incorporating all the data available. True positives were defined by the presence of proven aspergillosis and GM and true negatives by the absence of both.
None of the 30 proven cases was missed and GM was detected in at least 2 consecutive serum samples. However, GM was also detected in 5 (55.5%) of the 9 probable cases, 4 (7.4%) of the 54 possible cases, as well as in 5 (2%) of the 264 cases without aspergillosis. Hence, while the sensitivity and negative-predictive value with regard to proven aspergillosis were both 100%, the specificity was 96% and the positive-predictive value only 68%.
Comment by J. Peter Donnelly, Phd
If these results are correct, the diagnosis of invasive aspergillosis will have made a quantum leap into the 21st century. Diagnosing this disease in neutropenic patients and HSCT recipients has been notoriously contrary and frustrating, driving clinicians to start antifungal therapy empirically to shift the odds in favor of survival. In this study, empirical antifungal therapy was used in 43% of episodes with all that that entails whereas reliance on the GM-ELISA test would have reduced the use of empirical therapy to a more moderate 12%. True, serum needs to be tested regularly (Maertens and associates screened their patients at least twice weekly and collected on average 12 samples per patient), since a single test is not sufficiently informative, but the additional costs would be partially offset by savings on the drug budget.
The GM-ELISA test cannot be used in isolation as it is complementary to other techniques, particularly CT imaging, used in the work-up of persistently unexplained fever in neutropenic patients and the exploration of invasive fungal infective disease among all hematology patients at high risk. Just as important, the test has to form an integral part of a strategy that also includes the adoption of explicit commonly agreed on criteria for classification. Maertens et al opted to use the criteria proposed by the EORTC/MSG for case definition, which has yet to be published in full. Briefly, these criteria rely on 3 elements, namely: 1) the presence of a host risk factor (eg, neutropenia or HSCT recipient receiving immunosuppressive therapy); 2) clinical evidence (radiographic or clinical signs and symptoms); and 3) mycological evidence (demonstration by culture or microscopy and, importantly, by an indirect test such as the GM-ELISA test). Besides the real possibility of making a diagnosis, the clinician also has the opportunity to explore other avenues for managing invasive aspergillosis including pre-emptive treatment. Furthermore, having decided to begin treatment empirically anyway, a repeatedly negative GM test should encourage a search for an alternative diagnosis as well as stopping antifungal therapy. As Maertens et al state, "The real value of any non-invasive test will largely depend on its potential to discriminate "unproven" IA from alternative aetiologies with similar clinical presentations ie, cases normally classified as probable or possible."
This is not the first study to demonstrate the use of GM-ELISA, but it is the largest so far to attempt to formally evaluate its use in a large prospective series of patients at risk. The current study is not perfect because a credible alternative explanation was not found for the cases that resembled invasive aspergillosis but that nonetheless did not fit the case definition. But Maertens et al should be congratulated for their efforts, and their approach deserves to be adopted as a temporary or tentative standard of best-practice until a better one comes along.
There are some quirks associated with the GM-ELISA test. GM is frequently detected for the first and only time during mucositis, and certain foods and drugs have also tested positive.1 The test has also been around for some 6 years in Europe, but it has yet to gain formal approval for use in North America. This paper should at least help dispel lingering doubts about the potential diagnostic use of GM-ELISA. The body of evidence may not yet be sufficient to meet the demanding standards of the FDA, but its weight should certainly tilt the balance and shift the odds in favor of expediting the process so that the GM-ELISA becomes available to all.
References
1. Ansorg R, et al. Detection of Aspergillus galactomannan antigen in foods and antibiotics. Mycoses. 1997; 40:353-357.
2. Deresinski SC. Conference summaries: ICAAC 2000, IDSA 2000, and ASTMH 2000: Part III. Infectious Disease Alert. 2000;20:41-44.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.