According to the Nomenclature Review Committee of the World Allergy Organization, drug allergies refer to drug hypersensitivity reactions where a definite immunological mechanism [1] can be demonstrated. The prevalence of self-reported penicillin allergy is high; indeed, a crosssectional survey in Portugal found a 4.5% adult population self-reported penicillin allergy [2]. The prevalence rate of true penicillin allergy is, however, much lower [3, 4], and therefore a firm diagnosis is needed [5]. The clinical picture of drug allergy is very heterogeneous, encompassing distinct diseases such as morbilliform or bullous exanthema, urticaria and angiooedema, pustular exanthemas, fixed drug eruptions, anaphylaxis, blood cell dyscrasia, fever, interstitial lung disease, hepatitis, nephritis, vasculitis, DRESS (drug reaction with eosinophilia and systemic symptoms) and various forms of autoimmune diseases. However, it is important to distinguish between immediate and non-immediate reactions. The former occur within the first hour after the last drug administration and are manifested clinically by urticaria, angiooedema, rhinitis, bronchospasm or anaphylactic shock. The diagnosis of drug allergy often relies on clinical histories, skin tests and a few validated in vitro tests, such as serum-specific IgE assays, which are available only for a few drugs. The sensitivity of these tests is not 100% and, in selected cases, provocation tests are therefore necessary. However, new diagnostic tools, such as the basophil activation test (BAT), have been developed and are still under validation. Their routine use could increase the sensitivity of diagnostic work-ups, thus reducing the need for drug provocation tests and filling the gap between the high number of non-allergic drug allergy suspicions and the lack of specialized centres in all countries, which would enable confirmation or elimination of the diagnosis. There is evidence now that the BAT can contribute to the diagnosis of anaphylactic reactions due to several drugs, particularly muscle relaxants and b-lactams [6, 7]. At present, the most commonly used antigens in BATs are CD63 and CD203c. CD63, the gp53 or lysosyme-associated membrane protein (LAMP)-3, is one of the most reported markers for flow cytometric determination of basophil activation. CD63 is deeply anchored into the basophilic intracytoplasmic granules and is generally weakly expressed on the surface of basophils, monocytes, macrophages and platelets. In cases of allergic inflammation, as a result of fusion between the granule and the plasma membrane, CD63 is expressed with a high density on activated basophils and mirrors histamine release [8]. This up-regulated expression of basophilic CD63 can be detected by multi-colour flow cytometry using specific monoclonal antibodies [9]. As far as b-lactams are concerned, the CD63-based BAT was first evaluated in two studies [7, 10] including, respectively, 58 and 70 patients with immediate reactions to these antibiotics. BAT sensitivity was about 50% and specificity was over 90%. However, unlike the study by Sanz et al. [7], the one by Torres et al. [10] assessed not only patients with positive histories and positive skin tests and/or CAP-FEIA but also patients negative to both these tests and positive to challenges. It is interesting to note that one of the seven patients of the latter group was BAT positive. This group of patients with true allergy and negative skin tests is of crucial importance. They account for up to 33% of b-lactam-allergic patients and are only diagnosed by potentially harmful and cumbersome oral provocation tests [11]. Another more basophil-specific marker, CD203c (neural cell surface differentiation antigen E-NPP3, ecto-nucleotide pyrophosphatase/phosphodiesterase 3), is a multifunctional ectoenzyme exclusively and constitutively expressed on the surface of basophils [12]. Five years Correspondence: Prof. Pascal Demoly, Allergy Department, Hopital Arnaud de Villeneuve, University Hospital of Montpellier and Inserm U657, 34295 Montpellier, Cedex 5, France. E-mail: pascal.demoly@inserm.fr doi: 10.1111/j.1365-2222.2008.02981.x Clinical and Experimental Allergy, 38, 869–871