Evaluation of basophil activation in mastocytosis with Hymenoptera venom anaphylaxis

Authors

  • D. González-de-Olano,

    1. Allergy Unit, Hospital de Fuenlabrada, Madrid, Spain
    2. Red Española de Mastocitosis (REMA), Spain
    Search for more papers by this author
    • The first two authors contributed equally to the work and should be considered as first authors.

  • I. Álvarez-Twose,

    1. Red Española de Mastocitosis (REMA), Spain
    2. Instituto de Estudios de Mastocitosis de Castilla La Mancha, Hospital Virgen del Valle, Toledo, Spain
    Search for more papers by this author
    • The first two authors contributed equally to the work and should be considered as first authors.

  • J. M. Morgado,

    1. Red Española de Mastocitosis (REMA), Spain
    2. Instituto de Estudios de Mastocitosis de Castilla La Mancha, Hospital Virgen del Valle, Toledo, Spain
    Search for more papers by this author
  • M. I. Esteban López,

    1. Allergy Department, Hospital General de Segovia, Segovia, Spain
    Search for more papers by this author
  • A. Vega Castro,

    1. Allergy Department, Hospital de Guadalajara, Guadalajara, Spain
    Search for more papers by this author
  • M. D. Alonso Díaz de Durana,

    1. Allergy Unit, Fundación Hospital Alcorcón, Madrid, Spain
    Search for more papers by this author
  • L. Sánchez-Muñoz,

    1. Red Española de Mastocitosis (REMA), Spain
    2. Instituto de Estudios de Mastocitosis de Castilla La Mancha, Hospital Virgen del Valle, Toledo, Spain
    Search for more papers by this author
  • A. Matito,

    1. Red Española de Mastocitosis (REMA), Spain
    2. Instituto de Estudios de Mastocitosis de Castilla La Mancha, Hospital Virgen del Valle, Toledo, Spain
    Search for more papers by this author
  • B. de la Hoz Caballer,

    1. Red Española de Mastocitosis (REMA), Spain
    2. Allergy Department, Hospital Ramón y Cajal, Madrid, Spain
    Search for more papers by this author
  • M. L. Sanz,

    1. Allergy and Immunology Department, Clínica Universitaria de Navarra, Pamplona, Spain
    Search for more papers by this author
  • A. Orfao,

    1. Red Española de Mastocitosis (REMA), Spain
    2. Centro de Investigación del Cáncer/IBMCC, Departamento de Medicina and Servicio General de Citometría, University of Salamanca, Salamanca, Spain
    Search for more papers by this author
    • The last two authors contributed equally to the work and should be considered as last authors.

  • L. Escribano

    Corresponding author
    1. Red Española de Mastocitosis (REMA), Spain
    2. Instituto de Estudios de Mastocitosis de Castilla La Mancha, Hospital Virgen del Valle, Toledo, Spain
    • Instituto de Estudios de Mastocitosis de Castilla La Mancha, Hospital Virgen del Valle, Toledo, Spain
    Search for more papers by this author
    • The last two authors contributed equally to the work and should be considered as last authors.


  • How to cite this article: González-de-Olano D, Álvarez-Twose I, Morgado JM, Esteban López MI, Castro AV, Díaz de Durana MDA, Sánchez-Muñoz L, Matito A, de la Hoz Caballer B, Sanz ML, Orfao A, Escribano L. Evaluation of basophil activation in mastocytosis with hymenoptera venom anaphylaxis. Cytometry Part B 2011; 80B: 167–175.

Abstract

Background:

Basophil activation tests (BATs) have been demonstrated to be useful in detecting IgE-mediated sensitization by measuring basophil activation surface markers (CD63 and CD203c). Hymenoptera venom is one of the best known mediators-release trigger in patients with systemic mastocytosis (SM). The aim of this study was to investigate the use of BATs as an additional diagnostic tool in patients with mastocytosis suffering from hymenoptera venom anaphylaxis (HVA).

Methods:

A total of 22 patients with history of HVA and SM, together with a group of 11 patients with HVA in whom SM was ruled out after a complete bone marrow study, were analyzed.

Results:

Among 11 SM patients who had specific serum IgE (sIgE) against hymenoptera venom and an evaluable BAT, a positive BAT was found in nine. Additionally, a positive BAT was detected in three of seven patients who had no sIgE. These three patients had low levels of total IgE compared with control population (mean of 20 vs. 78 IU/mL); one had discontinued immunotherapy after 5 years, when sIgE levels had turned negative, and, in the other two patients, BAT identified the culprit insect.

Conclusions:

BAT is a useful complementary diagnostic tool to sIgE in mastocytosis patients with HVA, and it may contribute to predict or confirm these nearly fatal reactions, especially before discontinuing venom immunotherapy in patients who are negative for skin tests or sIgE or display low total IgE levels; in such cases, it also provides evidence on the culprit insect prompting HVA. © 2010 International Clinical Cytometry Society

Basophils, together with mast cells (MC), are the most relevant effector cells in Type I hypersensitivity reactions. A wide range of stimuli can induce the release of inflammatory mediators in basophils and MC. Cross-linkage of high-affinity Fc receptors for IgE (FcεRI) is the most widely described mechanism of basophil activation (1); however, compared with MC, basophils are more reactive to a greater number of substances that induce release of intracellular mediators independently of IgE/FcεRI cross-linking (e.g., bacterial-derived peptides (2, 3), stromal cell-derived factor (4), cytokines (5), activated complement mediators system (3), and secretory IgA previously primed with interleukin (IL)-3, IL-5, or granulocyte-macrophage colony stimulating factor (6)). Moreover, basophils have been also reported to play a pivotal role in IgG-mediated anaphylaxis, through release of platelet-activating factor (7). Recent studies support that basophils could also express HLA Class II molecules, function as antigen-presenting cells, produce IL-4, and trigger TH2 differentiation both in vivo and in vitro, either directly or through blockage of dendritic cell-induced TH1 responses (8–11).

At present, different basophil activation assays are available, which have proven useful for the identification of IgE-mediated sensitization (12–14). Measurement of basophil activation surface markers (e.g., CD63 and CD203c) correlates with both specific serum IgE (sIgE) and skin tests (15), and it seems to be more useful and sensitive than intradermal skin tests (14). Similarly, the evaluation of the expression of activation markers on basophils by flow cytometry in patients with anaphylaxis after hymenoptera sting has also proven to be of diagnostic value in patients with anaphylaxis (16), as a complementary diagnostic test (17).

Hymenoptera venom is one of the best known triggers for basophil and MC mediators-release in patients with mastocytosis. The prevalence of IgE-mediated allergy to hymenoptera sting in patients with mastocytosis rises up to 5% (18); in these patients, both IgE-mediated and non-IgE-mediated mechanisms have been described as potentially involved in the development of hymenoptera venom anaphylaxis (HVA) (19, 20). Unfortunately, except for a case report of a patient with mastocytosis and basophil D816V c-kit mutation in the context of multilineage hematopoietic cell involvement (21), little is known about the role of basophils in mastocytosis.

In this study, we investigated the potential use of basophil activation tests (BATs) as an additional diagnostic tool to sIgE in a series of 22 adult with mastocytosis suffering from HVA who were referred to the Spanish Network on Mastocytosis (Red Española de Mastocitosis, REMA).

MATERIALS AND METHODS

Patients

A total of 22 patients with systemic mastocytosis (SM) in the absence of skin lesions and a history of HVA were included in this study. Additionally, a group of 11 patients with HVA in whom SM was ruled out after a complete bone marrow (BM) study were analyzed in parallel as control group. The study was approved by the Ethical Committee of the Hospital Virgen del Valle (Toledo, Spain), and all patients gave their informed consent before entering the study.

Serum Determinations

Total IgE, venom-specific IgE antibodies, and baseline serum tryptase levels were measured as part of the routine biochemical and immunomonitoring work-up at the time of the initial allergy study and on further periodical visits. They were all determined by the CAP system using a commercially available kit (Pharmacia Diagnostics AB, Uppsala, Sweden); values of sIgE ≥ 0.35 kU/L were considered to be positive.

Basophil Activation Tests

Basophil activation was measured in acid citrate dextrose-anticoagulated peripheral blood (PB) according to Sanz et al. (22). After blood cell separation, 50 μL of the patient's cell suspension was incubated with 50 μL of Vespula, Polistes dominulus, and Apis extracts (ALK Allergologisk Laboratorium A/S, Horsholm, Denmark) at concentrations of 0.125 μg/mL and 0.03125 μg/mL. To evaluate background basal values (e.g., negative control, without stimulation), 50 μL of stimulation buffer containing IL-3 (2 ng/mL) and heparin (10 μL) (5,000 IU/mL; ROVI, Madrid, Spain) was added to the cell suspension in a separate well. As positive control, a monoclonal anti-IgE receptor antibody (Bühlmann, Allschwil, Switzerland) was used (final concentration of 1 μg/mL). After incubation and 5 min centrifugation at 540g, the basophils were double-labeled with the anti-CD63 phycoerythrin- and anti-IgE-fluorescein isothiocyanate antibody reagents. Flow cytometric analysis of surface markers was performed in a FACScan flow cytometer equipped with a 488-nm argon ion laser (Becton Dickinson Biosciences, BDB, San José, LA) and with the CellQuest software program (BDB). Cut-off levels to establish anti-IgE-fluorescein isothiocyanate and anti-CD63 phycoerythrin expression were defined separately for each patient, based on the negative control. A BAT was considered to be positive when the percentage of activated basophils of the positive control was >10%, and this value was also at least double of the percentage of activated basophils of the negative control. When the percentage of activated basophils of the positive control was <10%, or the negative control was >10%, the BAT was considered as “nonevaluable.”

Bone Marrow Study

Diagnosis of mastocytosis was based on a complete clinical and physical work-up, routine PB counts and differential, routine biochemistry, and serum tryptase levels. Skin biopsy was performed in all cases with suspected cutaneous involvement. A BM study was performed following previously established methods and criteria for morphology (23), histology and immunohistochemistry (24, 25), flow cytometry (26, 27), and KIT mutational status (28, 29). Classification of mastocytosis followed the World Health Organization guidelines (30–32).

Study Design

At the moment of entering the study, serum baseline tryptase, total IgE, and venom-specific IgE antibody levels were investigated in parallel to the BAT. Twenty patients with mastocytosis and HVA and nine controls with HVA had previously attended the REMA. Serum baseline tryptase, total IgE, and sIgE levels were performed at least once a year in these cases as a part of the allergic routine follow-up before they entered the study. In three SM patients and in one control, venom-sIgE values had returned negative (<0.35 kU/L) by the time these patients were included in the study.

At the time of inclusion, sIgE antibodies against hymenoptera venom were detected in 15 SM patients: sIgE to Apis (n = 6), Vespula (n = 9), and Polistes (n = 9). In turn, sIgE antibodies to Apis (n = 5), Vespula (n = 6), and/or Polistes (n = 6) were detected among all 11 control cases.

Venom immunotherapy (VIT) was started in all except one patient who had demonstrated IgE-mediated anaphylaxis (18 SM and 10 controls) at a median time from the anaphylactic episode of 16 months (range, 1–257). Among the 18 SM patients, 4 (22%) underwent Apis IT, 7 (39%) Vespula IT, 5 (28%) Polistes IT, and 2 (11%) went through Vespula and Polistes IT separately. VIT was discontinued according to the allergist's criteria in one patient after 61 months on therapy and 51 months before BAT; two patients started VIT 1 and 9 months after BAT had been performed. In the remaining 15 patients, BAT was performed after a median time on VIT of 48 months (range, 4–235 months); among these patients, four were on VIT for <1 year at the time of the BAT, four between 1 and 5 years, and seven started VIT >5 years ago. Among the 11 controls, VIT was started in 10 at a median time from the anaphylactic episode of 7 months (range, 2–195 months); one patient underwent Apis IT, seven Vespula IT, one Polistes IT, and one went through Vespula and Polistes IT separately. In three patients, VIT was discontinued after 37, 50, and 59 months, and BAT was performed 70, 1, and 85 months later, respectively. In six patients, BAT was performed after a median time of 44 months (range, 2–120 months) on VIT. The remaining patient initiated VIT 15 months after BAT was performed.

Activation-induced expression of surface CD63 by PB basophils was calculated among IgE HVA patients (SM and control) vs. non-IgE HVA. Expression of surface CD63 among IgE HVA vs. non-IgE HVA was also specifically calculated for SM patients. Results were expressed as box-plots. Additional research was performed to find a correlation among activation-induced expression of surface CD63 in SM, the levels of sIgE to the culprit insect, and to the time undergoing VIT, using lineal regression lines.

Statistical Methods

For all continuous variables, median and range were calculated, whereas for categorical variables, frequencies were reported. Categorical and continuous variables were compared through the nonparametric χ2 test and either the Mann–Whitney U or the Wilcoxon test, respectively. The Spearman's correlation coefficient was used to measure the degree of correlation between two variables. A value close to ±1 indicates equivalence (SPSS v15.0; SPSS, Chicago, IL). P values < 0.05 were considered to be associated with statistical significance.

In this study, a true-positive (TP) result was defined as a patient with HVA and a positive BAT and sIgE; a false-positive (FP) result was defined as a patient with HVA, positive BAT, and negative sIgE; a true-negative (TN) result was defined as a patient with HVA and negative BAT and sIgE; a false-negative (FN) result was defined as a patient with HVA, negative BAT, and positive sIgE. Sensitivity was calculated as TP/(TP + FN) and specificity as TN/(TN + FP).

RESULTS

Patients Characteristics

The most relevant features of HVA patients with and without mastocytosis are shown in Table 1. Patients with HVA associated with mastocytosis showed a clear male predominance (82% vs. 54%, P > 0.05), with a similar distribution of the triggers of anaphylactic episodes: wasp sting in 14 of 22 (64%) vs. 9 of 11 (82%) and bee sting in 4 of 22 (18%) vs. 2 of 11 (18%) cases. In the remaining four SM patients, the culprit insect remained unidentified.

Table 1. General Characteristics of the Patients
 Mastocytosis (n = 22)Controls (n = 11)P value
  • Results are expressed as number of cases and percentages in parentheses or as median (range).

  • a

    Mean (range) time undergoing VIT until the basophil activation test was performed, expressed in months.

  • b

    n = 16. Two patients started VIT after the basophil activation test was performed.

  • c

    n = 9. One patient started VIT after the basophil activation test was performed.

  • NS, no statistically significant differences were found.

Age (years)51 (30–78)46 (28–72)NS
Sex   
 Male18 (82%)6 (54%)NS
 Female4 (18%)5 (46%) 
Baseline tryptase (ng/mL)24.6 (6.78–149)14.9 (6–26.5)0.02
Total IgE (IU/mL)64.3 (9.13–455)78 (2–254.2)NS
Immune mechanism   
 IgE15 (68.2%)8 (72.8%)NS
 Non-IgE7 (31.8%)3 (27.2%)NS
Insect   
 Bee4 (18%)2 (18%)NS
 Wasp14 (64%)9 (82%)NS
 Unidentified4 (18%)0NS
Undergoing VIT18 (82%)10 (91%)NS
Time VIT–BATa54 (4–235)b48 (2–120)cNS
Type of IT   
 Apis4 (22%)1 (10%)NS
 Vespula7 (39%)7 (70%)NS
 Polistes5 (28%)1 (10%)NS
 Vespula + Polistes2 (11%)1 (10%)NS

Serum Tryptase, Total IgE, and Specific IgE Levels

Median baseline serum tryptase levels were significantly higher (P = 0.02) among SM vs. controls: median of 24.6 μg/L (range, 6.8–149 μg/L) vs. 14.9 μg/L (range, 6.0–26 μg/L). By contrast, both patient groups showed similar (P > 0.05) total IgE levels (Table 1).

In those 20 SM patients and nine controls followed-up at the REMA before inclusion in the study, median time between serum measurements at admission and those obtained at the time BAT was performed were of 41 months in SM patients (range, 1–367 months) vs. 78 months (range, 5–210 months) in the control group (P > 0.05). Median total IgE levels in the SM group were significantly higher at first determination than at time of BAT: 85 IU/mL (range, 20–5,000 IU/mL) vs. 64.3 IU/mL (range, 9.13–455 IU/mL), respectively (P = 0.02). Conversely, in the control group, total IgE levels at referral were similar to those obtained at BAT: median of 77 IU/mL (range, 2–770 IU/mL) vs. 78 IU/mL (range, 2–254 IU/mL), respectively (P > 0.05). At the time of initial allergic work-up, sIgE antibodies were demonstrated in 15 SM patients (four Apis, two Vespula, two Polistes, five Vespula + Polistes, and two Apis + Vespula + Polistes). In three of the remaining seven cases, sIgE antibodies against Polistes had been detected during the initial allergic work-up but were negative at the time of BAT. VIT had been started in these later cases 48, 63, and 112 months before BAT; two of them were still on VIT at the time of BAT after 48 and 63 months of therapy, whereas the other patient had discontinued VIT 51 months before BAT. Monitoring of total and sIgE levels throughout the years in SM patients undergoing VIT showed a significant decrease in the levels of sIgE antibodies, with no significant changes in total IgE levels (Table 2).

Table 2. Monitoring of IgE Levels in SM Patients Throughout VIT
ITaTotal IgE (IU/mL)P valueSpecific IgE (kU/L)P valueTime from VIT to BAT (months)
FirstLastFirstLast
  • Results are expressed as medians and range in parentheses.

  • a

    Two additional patients underwent IT (Vespula + Polistes), but they were not included in this table because VIT was started after the basophil activation test was performed.

  • IT, immunotherapy; NS, no statistically significant differences were found; VIT, venom immunotherapy; BAT, basophil activation test.

Apis (n = 4)88 (20–1,649)38 (11–100)NS8 (2–23)2 (1.1–4.7)0.0759 (5–235)
Vespula (n = 7)107 (78–5,000)134 (95–455)NS11 (1–100)2 (1–58)0.0324 (6–101)
Polistes (n = 5)56 (21–714)46 (9–81)NS2 (1–10)0.35 (0.3–2)0.0463 (4–112)

In Vitro Basophil Activation

Among those 15 SM patients in whom specific IgE antibodies against venom hymenoptera were detected, BAT was positive in nine, negative in two, and not evaluable in the remaining four cases (Table 3). As shown in Table 4, a high degree of agreement (sensitivity of 82%) existed among patients with an evaluable test and positive sIgE to hymenoptera and basophil activation as measured by the BAT. Those two patients with negative BAT and positive sIgE (SM Cases 5 and 14, Table 4) were on VIT for periods of time (101 and 112 months, respectively) significantly longer (P = 0.04) than those nine SM patients who were positive for both sIgE and BAT, with a mean time of 17 months (range, 5–80 months); additionally, sIgE values decreased in these two patients from 17.5 kU/L and 23.1 kU/L to 1.77 kU/L and 1.14 kU/L, respectively. Among those seven patients who showed no sIgE antibodies, three had a positive BAT and four a negative result. The former three cases (SM Cases 8, 15, and 16, Table 4) had a mean total IgE of 20.6 IU/mL (range, 9.1–23.8 IU/mL). In two of these patients (Cases 15 and 16), the BAT identified the culprit insect. Results of in vitro basophil activation in control patients are shown in Table 3.

Table 3. Basophil Activation Test (BAT) Results in SM Vs. Control Patients
inline image
Table 4. Correlation of Hymenoptera-Specific Serum IgE Levels and the Results of the Basophil Activation Test (BAT)
Type of patientCase IDHymenoptera speciesIn vitro determinations
ApisVespulaPolistes
sIgEBATsIgEBATsIgEBAT
  1. (+), Values ≥ 0.35 kU/L for specific IgE determinations and positive BAT according to the criteria proposed for positivity (22).

  2. (−), Values < 0.35 kU/L for specific IgE determinations and negative BAT according to the criteria proposed for positivity (22).

Mastocytosis1Wasp++++
 2Wasp+++++
 3Wasp++++
 4Wasp++
 5Wasp++
 6Wasp
 7Wasp++++
 8Wasp++
 9Wasp+++
 10Wasp++
 11Wasp+++++
 12Wasp
 13Bee++
 14Bee+
 15Unidentified+
 16Unidentified+
 17Unidentified
 18Unidentified
Control1Wasp
 2Wasp
 3Wasp+++++
 4Wasp++++
 5Wasp++++
 6Wasp+++++
 7Bee
 8Bee+++

Activation-Induced Expression of Surface CD63 on PB Basophils

Among those patients with an evaluable BAT (n = 26), cases with non-IgE HVA (n = 9) showed percentages of CD63+ basophils of 2.1% (range, 0–8.9%) and 2.8% (range, 1–8.7%) similar to those of IgE HVA patients (n = 17), 2.1% (range, 0.3–93.1%) and 2.3% (range, 0.3–84.7%), once PB basophils were stimulated with 0.125 and 0.03125 μg/mL of Apis venom extract, respectively (P > 0.05). Identical doses of Vespula venom were associated with a significantly lower percentage of CD63+ basophils, 2.3% (range, 0.7–57.9%) and 1.9% (range, 0.4–29.8%), among non-IgE HVA patients vs. IgE HVA cases, 53.7% (range, 0–88.5%) and 28.7% (range, 0.4–82.5%) (P = 0.006 and 0.008 for cases of 0.125 and 0.03125 μg/mL, respectively). With Polistes venom, these percentages were of 5% (range, 0.9–79.2%) and 6% (range, 0.9–62.3%) for non-IgE HVA patients and 19.5% (range, 0–91%) and 6.7% (range, 0.5–92.6%) for IgE HVA cases, respectively (Fig. 1A).

Figure 1.

Changes in the expression of surface CD63 after basophil activation. A: Percentage of CD63+ basophils in patients with non-IgE HVA vs. patients with IgE HVA. B: Percentage of CD63+ basophils in SM patients in non-IgE vs. IgE HVA patients with SM. The above-mentioned results express the percent of CD63+ after basophil activation test in each patient with Apis, Vespula, and Polistes venom. No statistically significant differences were found when analyzing the BAT results only to the known culprit insect, either in SM or control patients (data not shown). The bottom and top of the boxes represent the 25th and 75th percentiles, respectively, and the horizontal line inside the box displays the 50th percentile (median). Vertical lines represent the upper and lower boundaries of the 95% confidence intervals.

When considering only SM patients, those cases with non-IgE-mediated HVA (n = 7) showed a percentage of CD63+ basophils after stimulation with 0.125 and 0.03125 μg/mL of Apis venom of 1.7% (range, 0–8.9%) and 2.8% (range, 1–8.7%) vs. 2.4% (range, 0.3–72.2%) and 2.1% (range, 0.5–59.2%) for IgE HVA patients (P > 0.05). Stimulation with Vespula venom showed median values of 2.3% (range, 0.7–57.9%) and 1.9% (range, 0.4–29.8%) for non-IgE HVA SM patients vs. 70.5% (range, 1.5–88.5%) and 54.5% (range, 1.1–82.5%) for IgE SM patients, for the same concentrations of venom (P = 0.001). Finally, Polistes venom median values of CD63+ basophils were of 5% (range, 0.9–79.2%) and 6% (range, 0.9–62.3%) for non-IgE HVA SM cases vs. 16.7% (range, 0–91%) and 7.1% (range, 0.5–62.6%) for IgE HVA SM patients (P > 0.05) (Fig. 1B).

On exploring the relationship between activation-induced expression of surface CD63 on PB basophils and the levels of sIgE to the culprit insect, a lineal correlation was found in controls (r = 0.93 for the 0.125 μg/mL concentration and r = 0.94 for the 0.03125 μg/mL, with P-values equal to 0.001 and <0.001, respectively; Fig. 2A) and in SM (r = 0.5 for 0.125 μg/mL and r = 0.61 for 0.03125 μg/mL, with P-values equal to 0.04 and 0.009, respectively; Fig. 2B). When SM patients with sIgE values >30 kU/L were excluded from the analysis (to use a group with similar values to those of controls), no significant correlation was observed between both parameters (r = [minus]0.11 for 0.125 μg/mL and r = [minus]0.3 for 0.03125 μg/mL, with P-values >0.05 in both cases; Fig. 2C). No correlation was found regarding activation-induced expression of surface CD63 and the type of venom for which the patients were sensitized to (data not showed). Similarly, no significant correlation was found between activation-induced expression of surface CD63 and time under VIT, although values were close to statistical significance among controls (P = 0.05) but not SM patients.

Figure 2.

Correlation between activation-induced expression of surface CD63 on peripheral blood basophils and the levels of specific IgE to the culprit insect. A: Control patients. B: Mastocytosis patients. C: Mastocytosis patients with specific IgE values <30 kU/L.

DISCUSSION

Despite some limitations (15), BAT is a diagnostic technique recommended for the study of IgE-mediated diseases, including hymenoptera allergy, based on evaluation of activation-induced CD63 and/or CD203c expression on PB basophils (17). Previous studies showed that BAT is a diagnostic technique associated with a high sensitivity; because of this, the measurement of basophil activation could be used as a surrogate marker for the presence of allergen-sIgE on basophils and, thus, confirm the existence of sensitization in the absence of a positive sIgE (17), specially in cases where allergen sIgE levels are low because of a rather low total IgE value (<10 kU/L) (15) (two cases in our study). BAT has also been shown to confirm sIgE sensitization in patients with a clinical history of HVA in the absence of positive skin tests or positive allergen-sIgE (three patients in our study); it may also provide information about the culprit insect, leading to a better definition of the type of immunotherapy required (33) (in two of these later three cases).

Regardless of all information available about BAT in IgE-mediated HVA, no previous studies have investigated the potential use of this test in SM patients presenting with HVA. Similarly to what has been described for other allergic patients, our results prove that BAT is a useful tool in the diagnostic work-up of HVA in SM patients. Overall, BAT showed a high sensitivity (82%). The only two false-negative cases corresponded to patients who had started VIT for >8 years before BAT was performed, and both patients showed a significantly lower sIgE levels at the moment of BAT compared with those detected at diagnosis. Interestingly, all positive cases showed a shorter time lapse from initiation of VIT. Recent studies suggest that basophil responsiveness decreases with time on VIT (34) and the need for a higher threshold of antigen to activate the release of mediators after years of VIT could explain, at least to a certain extent, the false-negative results obtained. Interestingly, a positive BAT was detected in three patients in the absence of sIgE antibodies and identified the hymenoptera responsible for the clinical reaction in two of them; all three patients showed low levels of total IgE, in line with previous reports in SM patients (18). Altogether, these results support the high sensitivity of BAT vs. sIgE in this group of patients. Conversely, a significantly lower specificity was observed when comparing SM patients with HVA vs. the control HVA population (57%). Despite all this, in 18% of our cases, anti-IgE antibodies were unable to induce basophil activation and, thus, BAT was not evaluable; such percentage is identical to that described among the normal population (35).

In line with previous reports (36), a positive correlation was found in controls between activation-induced expression levels of surface CD63 and the amount of sIgE; nevertheless, this did not hold true for SM patients where a few patients carrying sIgE antibodies did not show a clear BAT response. These findings deserve further investigations.

VIT in patients with mastocytosis has proven to be an effective treatment in IgE-mediated allergy to hymenoptera (37). A decrease in basophil responsiveness has been demonstrated in repeated BAT in patients without SM undergoing wasp VIT, suggesting that BAT might contribute to a more adequate monitoring of wasp VIT (34). In contrast, it has also been reported that BAT is not helpful for predicting the outcome after a sting challenge, suggesting that other mechanism apart from sIgE may be involved in mediator-release in venom allergy (36). In this study, only one determination of BAT was performed per patient; thus there was no possibility to identify a potential decrease in basophil responsiveness, which might predict the symptoms of a new sting. In our study, one of the three patients with negative sIgE and positive BAT underwent VIT and discontinued it after 5 years, when sIgE levels had become undetectable. These preliminary results support the idea of extending VIT in mastocytosis patients over standard duration (37) and that BAT might be a very useful tool before discontinuing VIT in patients who are negative for skin testing or sIgE (14) and/or those who have low total IgE levels (<10 kU/L) (15).

Because our series is not homogenous and the absolute number of SM patients presenting with HVA are not very high because of the rarity of this condition, our results should be considered as preliminary. Despite this, they support the hypothesis that BAT could be a useful additional tool in the study of HVA in SM. Nevertheless, on the basis of the multiple pathways involved in basophil activation, including non-IgE mechanisms (2, 3–6), other activation-related antigens apart from CD63 and CD203c should be also evaluated (38), including activation markers such as CD107, CD164, and CD13, which may reflect distinct activation pathways (39). Besides, other cells involved in hypersensitivity reactions should also be studied, like the result of BAT in highly sensitized atopic donors' leucocytes (40). Thus, further studies about the function of basophils and other immune cells in mastocytosis (21) and HVA are necessary to confirm our results and establish the precise indications for BAT in the diagnosis of HVA and the follow-up of the response to VIT in both conditions.

Ancillary