A murine model was developed that recapitulates key features of clinical hypersensitivity to Escherichia coli asparaginase. Sensitized mice developed high levels of anti-asparaginase IgG antibodies and had immediate hypersensitivity reactions to asparaginase upon challenge. Sensitized mice had complete inhibition of plasma asparaginase activity (P = 4.2 × 10−13) and elevated levels of mouse mast cell protease 1 (P = 6.1 × 10−3) compared with nonsensitized mice. We investigated the influence of pretreatment with triprolidine, cimetidine, the platelet activating factor (PAF) receptor antagonist CV-6209 [2-(2-acetyl-6-methoxy-3,9-dioxo-4,8-dioxa-2,10-diazaoctacos-1-yl)-1-ethyl-pyridinium chloride], or dexamethasone on the severity of asparaginase-induced allergies. Combining triprolidine and CV-6209 was best for mitigating asparaginase-induced hypersensitivity compared with nonpretreated, sensitized mice (P = 1.2 × 10−5). However, pretreatment with oral dexamethasone was the only agent capable of mitigating the severity of the hypersensitivity (P = 0.03) and partially restoring asparaginase activity (P = 8.3 × 10−4). To rescue asparaginase activity in sensitized mice without requiring dexamethasone, a 5-fold greater dose of asparaginase was needed to restore enzyme activity to a similar concentration as in nonsensitized mice. Our results suggest a role of histamine and PAF in asparaginase-induced allergies and indicate that mast cell–derived proteases released during asparaginase allergy may be a useful marker of clinical hypersensitivity.
Asparaginase is one of the integral components of combination chemotherapy regimens used in the treatment of acute lymphoblastic leukemia (ALL) and lymphoma. The mechanism of action of asparaginase is not completely understood, but the active enzyme depletes asparagine and possibly glutamine systemically (Wu et al., 1978; Asselin et al., 1989; Chan et al., 2014). Common adverse events to asparaginase include allergic reactions, often accompanied by the development of anti-asparaginase IgG antibodies (Pieters et al., 2011). Serum anti-asparaginase IgG levels have been found to increase before the onset of allergic reactions (Liu et al., 2012), suggesting that high antibody titers are required to induce clinical hypersensitivity. Furthermore, lower serum asparaginase activity was found in patients with anti-asparaginase antibodies compared with patients without detectable antibodies, and the enzyme activities were inversely correlated with the antibody levels (Liu et al., 2012). Anti-asparaginase IgG antibodies have been proposed to directly neutralize asparaginase activity (Albertsen et al., 2002; Pieters et al., 2011), and lower systemic exposure to asparaginase is associated with lower exposure to concomitantly administered dexamethasone and an increased risk of central nervous system relapse (Yang et al., 2008; Kawedia et al., 2012), although some studies showed no associations between anti-asparaginase IgG antibodies and ALL outcomes (Cheung et al., 1986; Larson et al., 1998; Asselin, 1999; Woo et al., 2000; Panosyan et al., 2004).
Classic allergies involve cell-associated antigen-specific IgE antibodies and require low doses of antigen, low titers of circulating antibody, and are mediated by the release of histamine (Finkelman et al., 2005). An alternative pathway for allergy, which appears to play a role in asparaginase-induced reactions (Liu et al., 2012), requires repeated exposure to the antigen, high antigen-specific IgG antibody levels, a large antigen dose, and the release of platelet activating factor (PAF) (Finkelman et al., 2005; Finkelman, 2007). Immunologic studies have shown that antihistamines or PAF receptor antagonist can block the symptoms of an allergic reaction depending on the mechanism of allergy induced by the antigen (Strait et al., 2002). Understanding the pathway of asparaginase allergy will inform strategies for ameliorating the severity of hypersensitivity reactions and can help identify possible markers for detecting sensitized patients before receiving the offending drug.
To investigate therapeutic strategies for mitigating allergies and maintaining plasma concentrations of asparaginase, we created a murine model of asparaginase allergy. The model recapitulates several features of clinical hypersensitivity reactions developed to Escherichia coli asparaginase. Our results indicate the involvement of both histamine and PAF in asparaginase-induced allergies and support the importance of monitoring asparaginase activity if pretreatment of patients with antihistamines or glucocorticoids is used to prevent allergy.
Materials and Methods
Asparaginase Sensitization Protocol.
Eight-week-old female BALB/c mice received 10 μg i.p. doses of E. coli asparaginase (BioVendor Laboratory Medicine Inc., Candler, NC; >96.0% purity as determined by reverse phase high-performance liquid chromatography and SDS-PAGE) formulated with aluminum hydroxide adjuvant (Imject Alum; Thermo Scientific, Rockford, IL) on days 0 and 14 of treatment (Fig. 1A) to be sensitized to asparaginase. Control (nonsensitized) mice received intraperitoneal doses of adjuvant with vehicle alone (normal saline). Asparaginase allergies were induced in sensitized mice by challenging with a 100 μg i.v. dose of E. coli asparaginase on day 24 of treatment. The onset of hypersensitivity was detected by monitoring decreases in rectal temperature using a digital thermometer (model BAT-12; Physitemp Instruments, Clifton, NJ) for 2 hours after the asparaginase challenge. Prechallenge plasma samples for determining anti-asparaginase antibody levels were collected on day 23 of treatment by retro-orbital puncture (Fig. 1A), and postchallenge samples were collected by cardiac puncture at the end of the experiment for measuring antibody levels, asparaginase activity, and mouse mast cell protease 1 (mMCP-1) levels. The area under the temperature versus time curve was calculated using the trapezoidal rule. Lower area under the curve (AUC) values indicate more severe reaction (drop in rectal temperature), and differences in the severity of asparaginase-induced allergies between treatment groups was determined by comparing the AUC values of different groups. Mice were housed in an American Association of Laboratory Animal Care–accredited facility and treated using Institutional Animal Care and Use Committee–approved protocols in accordance with National Institutes of Health guidelines
Four different agents were investigated for their ability to mitigate asparaginase-induced allergic reactions: the antihistamines cimetidine (H2 receptor antagonist; Sigma-Aldrich, St. Louis, MO) and triprolidine (H1 receptor antagonist; Sigma-Aldrich), the PAF receptor antagonist CV-6209 [2-(2-acetyl-6-methoxy-3,9-dioxo-4,8-dioxa-2,10-diazaoctacos-1-yl)-1-ethyl-pyridinium chloride; Santa Cruz Biotechnology, Santa Cruz, CA], and dexamethasone (American Pharmaceutical Partners, Inc., Schaumburg, IL). As described in Fig. 1B, antihistamine (cimetidine and triprolidine, 200 μg i.p.) and CV-6209 (66 μg i.v.) were given 30 and 5 minutes before the asparaginase challenge on day 24, respectively. Five different regimens of dexamethasone were investigated as described in Fig. 1, B and C. Single-dose dexamethasone was given 2 hours before the asparaginase challenge on day 24 (2 mg/kg i.p.), and multiple-dose dexamethasone was given 2 hours before each asparaginase dose on days 0, 14, and 24 (2 mg/kg i.p.). Pretreatment with oral continuous dexamethasone (4 mg/l in drinking water with a consumption rate of 5 ml/day yields an approximate dose of 0.8–1.0 mg/kg per day for the mice) starting 7 days before, same day (day 0), or 7 days after the initial asparaginase immunization dose was also evaluated for its influence on the severity of asparaginase-induced allergies. This regimen of dexamethasone achieves similar plasma dexamethasone concentrations as those achieved clinically (10–200 nM) in pediatric ALL patients given doses of 8 mg/m2 per day (Yang et al., 2008, 2009; Kawedia et al., 2011).
Asparaginase Activity Determination.
The asparaginase activity in plasma samples was determined by monitoring the enzymatically coupled oxidation of NADH to NAD+ in a 96-well format, as described previously (Fernandez et al., 2013). Briefly, 10 μl of plasma from each sample was added to a 96-well microplate (Corning, Lowell, MA), and an enzyme reaction mixture containing asparagine, α-ketoglutaric acid, glutamic oxalacetic transaminase, β-NADH, and malic dehydrogenase was added to each sample. A BioTek EL x 808 IU microplate reader (Winooski, VT) was used to determine the asparaginase activity. All asparaginase standards used for calibration purposes were prepared using mouse plasma, and samples with values over the standard curve were diluted with mouse plasma to get their enzyme activities within the linear range of the assay.
Determining the Ex Vivo Asparaginase Activity Inhibition of Samples from Sensitized Mice.
The neutralization of asparaginase activity by postimmunization plasma was determined by adding 0.2 IU/ml of asparaginase to plasma samples collected after the asparaginase challenge (day 24). Plasma from sensitized mice, nonsensitized mice, or nonsensitized mice after addition of 5 μg/μl of rabbit polyclonal anti-asparaginase IgG antibodies (Abcam, Cambridge, MA) were tested for residual asparaginase activity as described above.
Detection of Anti-Asparaginase IgG Antibodies.
An enzyme-linked immunosorbent assay–based method was used to detect anti–E.coli asparaginase IgG antibodies, as described previously (Wang et al., 2000; Liu et al., 2012). Plasma samples from nonsensitized mice were used as negative controls, and rabbit polyclonal anti-asparaginase IgG antibodies were used as positive controls. Anti-mouse IgG (Sigma-Aldrich), anti-mouse IgE (Thermo Scientific), and anti-rabbit IgG secondary antibodies (Sigma-Aldrich) were used for detection of antibodies in samples or controls.
Determining mMCP-1 Concentrations.
Concentrations of mMCP-1 were determined by enzyme-linked immunosorbent assay using the Ready-SET-Go commercial kit from eBioscience (San Diego, CA).
Differences between rectal temperature AUC, asparaginase activity, anti-asparaginase IgG levels, and mMCP-1 concentrations were determined using a pairwise Wilcoxon rank sum test. All statistical analysis was performed with the R statistical software (version 2.13.2).
Mice were sensitized to E. coli asparaginase using the adjuvant aluminum hydroxide to create a murine immune response similar to clinical hypersensitivity reactions to asparaginase (Figs. 1A and 2, A–E). Sensitized mice challenged with 100 μg of asparaginase (22.5 IU or ∼1125 IU/kg per mouse) had immediate hypersensitivity reactions characterized by a sudden drop in rectal temperature (Fig. 2A) and a decrease in physical activity, whereas nonsensitized mice did not experience changes in rectal temperature (Fig. 2B; P = 3.3 × 10−10) or activity upon challenge. Compared with nonsensitized controls, sensitized mice experienced a peak drop in rectal temperature >6°C within 30 minutes of the asparaginase challenge and had negligible asparaginase activity in samples collected 2 hours after the challenge (Fig. 2C; P = 4.2 × 10−13). The loss of enzyme activity is likely due to high levels of anti-asparaginase IgG antibodies detected before the challenge on day 23 relative to nonsensitized mice (Fig. 2D, red open circles; P = 1.1 × 10−7). The antibodies bind to asparaginase upon challenge and result in a drop in the measurable antibody titer (Fig. 2D; P = 6.7 × 10−5, comparing black to red open circles of sensitized mice), presumably due to the formation of immune complexes that lead to a decrease in asparaginase activity. Using an ex vivo activity inhibition assay, we found that plasma samples from sensitized mice neutralize >30% of the added asparaginase activity (0.2 IU/ml), with the degree of activity inhibition correlated with anti-asparaginase IgG levels (Supplemental Fig. 1, A and B; R2 = 0.161, P = 0.04). Consistent with the reports of elevated levels of mMCP-1 in various pathways of anaphylaxis (Strait et al., 2002), the mMCP-1 concentrations measured in sensitized mice were elevated nearly 35-fold compared with nonsensitized mice (Fig. 2E; P = 6.1 × 10−3).
Having established a model of asparaginase-induced allergy, we next tried to develop pretreatment strategies to mitigate the severity of the reactions. The initial pretreatment agents we selected were based on receptor antagonists of known chemical meditators of allergy (Fig. 1, B and C). We found that cimetidine, triprolidine, CV-6209, or the three agents used in combination all mitigated the severity of the reaction compared with no pretreatment (Fig. 3; P < 1.3 × 10−3). Antihistamines and CV-6209 mitigated the drop in rectal temperature, but pretreatment with the combination of antihistamines and CV-6209 reduced the severity of the allergic reaction compared with no pretreatment (Fig. 3; P = 1.2 × 10−5), antihistamine alone (Fig. 3; P = 1.6 × 10−4), and single agent CV-6209 (Fig. 3; P = 3.2 × 10−4). It was unclear whether triprolidine and/or cimetidine were responsible for reducing the severity of asparaginase-induced allergy; therefore, we tested each antihistamine as a single agent in sensitized mice. Cimetidine showed little effect compared with nonpretreated sensitized mice, and this group had more severe reactions compared with mice pretreated with a combination of cimetidine and triprolidine (Fig. 3; P = 1.1 × 10−3). In contrast, mice pretreated with single agent triprolidine showed a decrease in the severity of allergies compared with nonpretreated, sensitized mice (Fig. 3; P = 0.02), and their rectal temperature AUC was similar to that of mice pretreated with cimetidine and triprolidine in combination (Fig. 3; P > 0.05). These results suggest that triprolidine but not cimetidine is required for mitigating asparaginase-induced allergies in our murine model.
PAF receptor antagonists, such as CV-6209, are not approved for clinical use in most countries. Therefore, we wanted to test whether dexamethasone could achieve similar effects as CV-6209 on asparaginase-induced allergies. Five dexamethasone pretreatment strategies were investigated: single-dose dexamethasone (Fig. 1B; 2 mg/kg i.p.) given 2 hours before the asparaginase challenge, multiple doses of dexamethasone (Fig. 1C; 2 mg/kg i.p.) given 2 hours before each immunization dose (Fig. 1C; days 0 and 14) and before the challenge (day 24), and oral dexamethasone (Fig. 1C; 4 mg/l in drinking water) starting at 7 or 0 days before or 7 days after the initial asparaginase immunization dose. Pretreatment with single or multiple doses of dexamethasone was also evaluated with concomitant triprolidine. We found that mice pretreated with any of the dexamethasone regimens showed a decrease in the severity of the reaction compared with nonpretreated, sensitized mice (Fig. 3; P ≤ 0.03), except for mice receiving oral dexamethasone 7 days after the initial asparaginase immunization. We found no difference in rectal temperature AUC between single-dose dexamethasone pretreatment (with or without concomitant triprolidine), multiple-dose dexamethasone pretreatment (with or without concomitant triprolidine), and mice receiving oral dexamethasone (Fig. 3; P > 0.05). Furthermore, all mice pretreated with any regimen containing dexamethasone had more severe reactions compared with mice pretreated with antihistamines and CV-6209 in combination (Fig. 3; P ≤ 0.06).
Asparaginase activity was determined in samples collected after the asparaginase-induced hypersensitivity to determine if pretreatment was able to restore activity (Fig. 4). Only oral dexamethasone given 0 or 7 days before the initial asparaginase was able to partially rescue activity compared with nonpretreated, sensitized mice (Fig. 4; P = 4.8 × 10−2 and 8.3 × 10−4, respectively). All other pretreatment strategies resulted in similar activity levels as nonpretreated, sensitized mice (Fig. 4; P > 0.05). Nevertheless, mice receiving oral dexamethasone 0 or 7 days before the initial asparaginase had lower asparaginase activity than nonsensitized mice (Fig. 4; P = 3.8 × 10−5 and 1.9 × 10−6, respectively), suggesting that other methods of lowering anti-asparaginase IgG antibody levels may be required to completely rescue asparaginase activity. Similar pretreatment experiments with antihistamine, CV-6209, and dexamethasone were also tested in mice that were not sensitized to asparaginase to confirm that there was no direct influence of premedication on asparaginase activity (Supplemental Fig. 2). These results indicate that H1 and PAF receptor antagonists mitigate allergic reactions but do not influence the asparaginase activity.
The inability of pretreatment strategies to fully rescue asparaginase activity is likely due to high anti-asparaginase IgG levels that are unaffected by the pretreatment strategies we used. Mice immunized with asparaginase all developed high levels of anti-asparaginase IgG antibodies in prechallenge samples (day 23) compared with nonsensitized mice (Fig. 5A; P < 0.04), including mice that received multiple doses of dexamethasone. However, mice pretreated with oral dexamethasone on the day of or 7 days before the initial asparaginase immunization showed diminished antibody levels compared with sensitized, nonpretreated mice (Fig. 5B; P = 1.0 × 10−3 and 4.1 × 10−4, respectively) in samples available on day 17 of treatment (Fig. 1A). These results indicate that oral dexamethasone was able to partially restore asparaginase activity due to its effect on anti-asparaginase IgG antibody levels.
We hypothesized that many of the pretreatment strategies were masking the severity of the hypersensitivity reactions rather than blocking the release of the chemical mediators of allergy. To confirm this, we measured concentrations of mMCP-1 in samples that were collected after the asparaginase-induced hypersensitivity. All sensitized mice showed elevated levels of mMCP-1 compared with nonsensitized mice (Fig. 6; P ≤ 0.02), except for mice receiving a single dose of dexamethasone (with or without concomitant triprolidine, P > 0.05). Interestingly, this observation was not observed with any other dexamethasone pretreatment regimen. It is possible that the pretreatment with a single dose of dexamethasone mitigated the reactions by inhibiting the release of histamine during the allergic reaction, whereas other dexamethasone regimens masked the hypersensitivity through other mechanisms that likely affect antibody production. Similar data were previously reported, suggesting that dexamethasone can inhibit the release of histamine from mast cells and basophils (Schleimer et al., 1982, 1987; Berenstein et al., 1987).
We next investigated if increasing the dose of asparaginase used during the challenge could overcome the high antibody titers and restore asparaginase activity without requiring dexamethasone. In the absence of pretreatment, the severity of asparaginase-induced reactions increased in a dose-dependent manner, and all groups challenged with higher doses of asparaginase had more severe reactions compared with sensitized mice challenged with the standard 100-μg dose (Fig. 7A; P ≤ 2.1 × 10−4). Furthermore, dose increases ≥5-fold (5× and 10×) were able to restore enzyme activity to concentrations similar to or higher than that of nonsensitized mice (Fig. 7B). Although increasing the dose of asparaginase used during the challenge from 1- to 10-fold showed a trend toward lower antibody titers in samples collected after the challenge (data not shown), the trend failed to reach statistical significance. mMCP-1 levels were elevated in a dose-dependent manner and correlated with the severity of the reactions (Fig. 8; R2 = 0.577, P = 3.0 × 10−16).
Our pretreatment studies showed that triprolidine and CV-6209 were the best for mitigating the severity of asparaginase-induced allergies, and our dose adjustment experiments indicated that a 500-μg dose of asparaginase per mouse was required to rescue asparaginase activity in sensitized mice. Therefore, we combined both strategies to determine if we could restore activity while simultaneously mitigating the severity of the allergy. We found that the pretreatment, even at the higher dose of asparaginase, was capable of blocking the allergic reaction compared with sensitized, nonpretreated mice (Fig. 7A; P = 9.8 × 10−4), and the dose adjustment restored asparaginase activity to similar concentrations as nonsensitized mice (Fig. 7B; P > 0.05). The anti-asparaginase antibody levels and mMCP-1 concentrations measured for pretreated and nonpretreated mice given a 5-fold increased dose were similar (P > 0.05, data not shown), suggesting that mice in both groups had similar potential for developing hypersensitivity reactions of comparable severities.
Clinical immune responses to asparaginase during the treatment of leukemia almost exclusively occur upon re-exposure to asparaginase after an initial exposure (Albertsen et al., 2002; Pieters et al., 2011; Fernandez et al., 2014a), implying that these patients were sensitized to asparaginase administered initially. Reactions in sensitized patients are usually characterized by an immediate allergic response after administration of asparaginase, by the development of high levels of anti-asparaginase IgG antibodies, and by low serum asparaginase activity concentrations (Peterson et al., 1971; Spiegel et al., 1980; Muller et al., 2001; Panosyan et al., 2004; Pieters et al., 2011; Liu et al., 2012). The sensitization protocol we used in mice (Fig. 1, A and B) resulted in the development of an asparaginase-mediated immune response with a similar phenotype as clinical hypersensitivity to E. coli asparaginase.
Our results have several implications for improving use of asparaginase. Our data support that an H1 histamine receptor antagonist (triprolidine) used in combination with a PAF receptor antagonist (CV-6209) is best for mitigating asparaginase-induced reactions (Fig. 3). Although we did not measure IgE antibodies, these response data are consistent with a role for both anti-asparaginase IgG (through FcγR-dependent release of PAF) and IgE antibodies (through FcεR1-dependent release of histamine) being involved in eliciting asparaginase hypersensitivity reactions. However, pretreatment with these agents did not rescue plasma asparaginase activity (Fig. 4). These experiments clearly demonstrate that the symptoms and signs of asparaginase allergies can be completely masked by pretreatment and yet result in subtherapeutic drug concentrations, thus emphasizing the importance of monitoring asparaginase activity after pretreatment with antihistamines, glucocorticoids, or any other medication used for mitigating the severity of allergic reactions. In addition, the low asparaginase activity measured in sensitized mice is likely due to an accelerated clearance and antibody neutralization, because the maximum ex vivo inhibition we were able to induce using rabbit anti-asparaginase IgG antibodies (5 μg/μl) was only 50% (Supplemental Fig. 1). Similar neutralization results were previously reported (Peterson et al., 1971; Baechtel and Prager, 1973), suggesting that anti-asparaginase IgG antibodies cannot completely neutralize asparaginase activity and that other mechanisms are involved in the rapid in vivo disappearance of enzyme activity. This is further supported by the >50% loss of asparaginase activity (Fig. 4) seen in mice pretreated with dexamethasone 7 days before the initial asparaginase immunization dose, although they had negligible antibody levels (Fig. 5B) compared with other pretreatment groups. This is similar to the accelerated asparaginase clearance seen in pediatric patients with relapsed ALL receiving E. coli asparaginase and negative for anti-asparaginase IgG antibodies (Panetta et al., 2009).
We found that dexamethasone could be used to mitigate the reaction, although at the dosage schedules we studied it was not as effective in masking the allergic reaction as pretreatment with triprolidine and CV-6209 in combination (Fig. 3; P ≤ 0.06). However, pretreating mice with oral dexamethasone before exposure to asparaginase lowered anti-asparaginase IgG levels (Fig. 5B) and partially maintained asparaginase activity (Fig. 4). These results are similar to recent clinical findings that showed a lower incidence of asparaginase hypersensitivity in patients that received dexamethasone 4 days before asparaginase administration during remission induction and support the use of glucocorticoids before beginning asparaginase (Pui, 2013; Vora et al., 2013).
Our dose adjustment studies indicate that strategies to lower anti-asparaginase IgG antibody titers (by overcoming their inhibitory effects by presenting a higher concentration of antigen) can restore enzyme activity. However, the effect of large asparaginase doses on treatment-related toxicities is unknown. The dose-adjustment experiments also show that the severity of asparaginase-induced reactions increases in a dose-dependent manner (Fig. 7A). Thus, when challenging patients after a suspected reaction, a lower dose of asparaginase might be expected to result in a less serious reaction but may not overcome the antibody levels and result in lower serum asparaginase activity levels.
Histamine and PAF are considered primary chemical mediators of allergic shock by acting on smooth muscle endothelial cells (Finkelman, 2007). However, histamine or PAF are unlikely to be useful biomarkers of asparaginase-induced allergies due to their short serum half-life. Histamine concentrations reach peak levels within 5 minutes of the onset of symptoms and have a serum half-life of a few minutes (Ferrer et al., 2010), whereas PAF has a serum half-life <15 minutes (Vadas et al., 2008). Several studies have shown that elevated levels of tryptase can be used as a biomarker of clinical hypersensitivity (Ordoqui et al., 1997; He et al., 2004; Payne and Kam, 2004; Fernandez et al., 2014b), and in accord with those studies, we found that mMCP-1 was elevated in sensitized mice from samples collected 2 hours after the asparaginase challenge (mMCP-1 is a mucosal mast cell granule-specific β-chymase that is released during basophil or mast cell degranulation in mice similar to tryptase in humans; Caughey, 2007). Furthermore, mMCP-1 concentrations were correlated with the severity of allergic reactions (Fig. 8), and previous studies showed that increases in plasma mMCP-1 levels may be induced by either IgG or IgE-mediated allergies (Strait et al., 2002). Unlike histamine and PAF, tryptase reaches peak concentrations ∼30–60 minutes after allergy onset, has a serum half-life of 90 minutes, and exhibits low serum concentrations in normal, nonallergic individuals (Laroche et al., 1991). Although other studies have suggested that histamine levels may be a more useful immediate (e.g., 10 minutes) indicator of the onset of allergies compared with the release of mast cell proteases in mice (Nabe et al., 2013), a biomarker of allergy that is elevated and stable for several hours after an allergic reaction may be more useful as an eventual clinical biomarker of asparaginase-induced allergies. Taken together, our results suggest that tryptase may be a possible clinical biomarker for confirming episodes of hypersensitivity to asparaginase.
Aluminum hydroxide adjuvant was required for mice to develop immune responses to asparaginase, as mice immunized with asparaginase without adjuvant did not develop allergic reactions when challenged (data not shown). Selection of an adjuvant can play a large role in determining the characteristics of the immune response generated (e.g., antibody isotypes, subclass, affinity, and titer) (Hadjipetrou-Kourounakis and Moller, 1984; Kenney et al., 1989; Muller and Boos, 1998). Anti-asparaginase antibodies in ALL patients have been reported to be predominantly IgG1 and IgG4 (Cheung et al., 1986; Korholz et al., 1987), which is consistent with a TH2 response (Visciano et al., 2012). Immunization with aluminum hydroxide generally elicits a TH2 response with the development of antigen-specific IgG1 and IgE antibodies (Kenney et al., 1989). In accordance with these reports, asparaginase-sensitized mice developed anti-asparaginase IgG1 antibodies, but no IgE antibodies were detected (data not shown). Therefore, clinical hypersensitivity reactions to asparaginase that are associated with the development of anti-asparaginase IgG antibodies and the asparaginase-induced allergic reaction in our murine model both appear to be TH2 driven. Nevertheless, there are several similarities and differences between the murine and human immune system that may influence the translation of our results to human allergies. Both species produce IgE antibodies that bind with high affinity to the FcεRI receptor, and cross-linking the receptor results in the release of chemical mediators of anaphylaxis, including histamine and PAF (Finkelman, 2007). Both species also produce IgG antibodies that induce the release of PAF by binding to the FcγRIII receptor, and both can mediate allergic reactions by the activation of complement through the production of anaphylatoxins. In contrast, the human and murine immune systems differ in the ability to activate complement through their respective IgG isotypes (Snapper and Finkelman, 1999), in the cells that express FcεR1 (Bieber et al., 1992; Maurer et al., 1994, 1996; Kinet, 1999), and in the quantity of granules within basophils, which influences the role of chemical mediators of allergy during hypersensitivity reactions (Lee and McGarry, 2007; Tsujimura et al., 2008). It is also possible that other chemical mediators of allergies in addition to PAF and histamine, such as cysteinyl leukotrienes and prostaglandin D2, play a role during human allergic reactions to asparaginase (Ono et al., 2009).
In conclusion, we created the first murine model of immediate hypersensitivity reaction to asparaginase. Several previous studies have sensitized mice to asparaginase and demonstrated the development of anti-asparaginase antibodies (Roberts et al., 1966; Baechtel and Prager, 1973; Goldberg et al., 1973; Uren and Ragin, 1979; Kamisaki et al., 1981; Yagura et al., 1981; Fernandes and Gregoriadis, 2001; Cantor et al., 2011) and demonstrated delayed hypersensitivity reactions to asparaginase (Kawamura et al., 1985; Gaspar et al., 1996). Our model is the first to show the development of acute symptomatic allergies in sensitized mice and to identify methods of reducing the severity of the symptoms and restoring asparaginase activity. We demonstrate that pretreatment with H1 receptor antagonists, PAF receptor antagonist, or dexamethasone can all partially mitigate the severity of asparaginase-induced allergic reactions to a similar extent; however, only dexamethasone was able to partially maintain plasma asparaginase activity. This may be particularly important for selecting a pretreatment agent in patients that are at a high risk of developing glucocorticoid-induced osteonecrosis (Vora, 2011). Our results also support the need for a clinically available PAF receptor antagonist for mitigating immune responses (Roberts et al., 1988; Hozawa et al., 1995; Kingsnorth et al., 1995) and emphasize that it may be important to monitor asparaginase activity if pretreatment medications are used. Furthermore, our results indicate that strategies that can lower anti-asparaginase antibodies might be useful to restore enzyme activity in a similar manner as our dose escalation approach.
Participated in research design: Fernandez, Smith, Finkelman, Relling.
Conducted experiments: Fernandez.
Performed data analysis: Fernandez, Smith.
Wrote or contributed to the writing of the manuscript: Fernandez, Smith, Karol, Ramsey, Liu, Pui, Jeha, Evans, Finkelman, Relling.
- Received October 16, 2014.
- Accepted January 7, 2014.
The study was supported by the National Institutes of Health National Cancer Institute [Grants P30-CA21765, R01-CA142665, R37-CA36401]; and by the American Lebanese Syrian Associated Charities.
- acute lymphoblastic leukemia
- area under the curve
- 2-(2-acetyl-6-methoxy-3,9-dioxo-4,8-dioxa-2,10-diazaoctacos-1-yl)-1-ethyl-pyridinium chloride
- mouse mast cell protease 1
- platelet activating factor
- U.S. Government work not protected by U.S. copyright