Science1 publisher2 min readPublished
Malaga's nanoparticle assay found 49 of the 50 confirmed amoxicillin allergies in its test set
The 98% sensitivity behind that count came from a single case-control comparison at the University of Malaga, in which ImmunoCAP, the most widely used commercial assay, detected about 17% of the same patients.
The Scientist · Science desk
What happened
- A University of Malaga team has built an in vitro platform of magnetic nanoparticles for diagnosing allergy to beta-lactam antibiotics, including amoxicillin and other penicillins.
- Blood came from 94 people in all, 44 of them able to take the antibiotics without reacting.
- The platform was measured against two existing tests: ImmunoCAP, the most widely used commercial assay, and conventional RAST.
- The authors say the method avoids drug provocation testing with the medication itself, and that the laboratory run needs no lengthy centrifugation.
- The authors say larger multicenter studies and clinical validation are needed before the technology goes into hospitals for routine use.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint Both arms of this study were sorted before the assay saw them: allergy confirmed on one side, tolerance demonstrated on the other. That cannot tell you how the test behaves in the patient a clinic actually sends for provocation, whose history is vague and whose skin test came back negative.
- decision Standing down a provocation test turns on false positives as much as false negatives, and the published account does not include a specificity figure. An allergy service cannot price the decision without one.
- exposure The population this would reach is the mislabelled majority, people held off first-line penicillins by a diagnosis nobody ever tested, and a bench assay reaches them without booking a monitored chair.
- capability A validated version would move de-labelling work out of the day unit and into the laboratory, which removes the reaction risk the current confirmation route carries.
Sensitivity counts only the people already known to have the condition, and here that group is the 50 patients with confirmed amoxicillin allergy [1]. So 98% is 49 found and one missed [2]. ImmunoCAP's roughly 17% on the same 50 works out at about eight [3], a difference of some 40 patients between the two assays [4]. A negative result is what a clinic would have to act on to skip a provocation test, and the one miss in this cohort is the patient who would then take the drug [2].
The particles are 30 nm across, an iron oxide core inside a silica shell [2]. Ezequiel Perez-Inestrosa, professor of organic chemistry at the University of Malaga and one of the study's authors, described them as acting "like a tiny magnet covered with thousands of molecular hooks" [3]. "If the patient's blood contains the IgE antibodies responsible for the allergy, they are captured much more readily than with conventional methods," he said [4].
Maria Jose Torres, professor of medicine and another of the researchers leading the work, said that "In the tests performed, the system was able to detect virtually all true cases of allergy and reduced both false negatives and false positives" [7].
Torres said penicillin allergy is overdiagnosed [8], and Perez-Inestrosa put a spread on it: "Although between 8% and 25% of the population report being allergic to these antibiotics, only between 1% and 10% actually are. This incorrect label leads to the use of alternative drugs that are generally less effective, more toxic and contribute to the emergence of bacterial resistance" [9]. If every truly allergic person sits inside the group that reports an allergy, the widest reading of those two ranges leaves 24 of every 25 labels wrong [5].
The paper, by Jose A. Cespedes and colleagues, is in Materials Today Bio [15], and the work involved the IBIMA BIONAD Platform, the Inflammatory Diseases Network and the Regional University Hospital of Malaga alongside the university [5]. Yolanda Vida and Maria I. Montanez, of the university's organic chemistry department, are the lead authors [6].
What to watch
- Whether a multicenter cohort of patients with unclear histories and negative skin tests reproduces the amoxicillin sensitivity, and what specificity comes with it.
- Whether the group publishes specificity and negative predictive value in a consecutively recruited clinic population rather than a pre-sorted one.
- Whether any allergy service writes a protocol that lets a negative in vitro result stand in for a provocation test, and at what safety threshold.