Introduction

Mast cells (MCs, Fig. 1) are immune cells of hematopoietic origin found in all human tissues, especially at the environmental interfaces. They act as both effector and regulatory cells and play a central role in adaptive and innate immunity (Anand et al. 2012; Gri et al. 2012). Their important role in immunological as well as non-immunological processes is reflected by the large number of mediators (>200) including pre-stored ones such as histamine and tryptase as well as numerous mediators synthesized de novo in response to allergic or non-immune triggers such as chemokines and cytokines, by which MCs may influence other cells (Lundequist and Pejler 2011; Ibelgaufts 2016). Their evolved arrays of sensory and response mechanisms engender diverse havoc when MC dysfunction emerges.

Fig. 1
figure 1

May-Grünwald/Giemsa stain of a resting human mast cell and a mast cell following activation-induced degranulation. Note the loss of granule staining. Mast cells obtained from the human bone marrow, magnification 1000×

The umbrella term mast cell activation disease (MCAD; Akin et al. 2010) comprises the full spectrum of primary systemic MC disease, i.e., systemic mastocytosis (SM) which is further divided into several subtypes (Valent et al. 2007; Tables 1 and 2), primary MC activation syndrome (MCAS; Table 3; Molderings et al. 2011a; Hamilton et al. 2011; Valent et al. 2012), and MC leukemia (MCL). Pathogenetically, MCAD denotes a group of polygenic MC disorders (Molderings 2015, 2016) characterized by aberrant release of variable subsets of MC mediators and also an accumulation of either morphologically altered and immunohistochemically identifiable mutated MCs due to MC proliferation (SM and MCL) or morphologically ordinary MCs due to decreased apoptosis (MCAS; Kohno et al. 2005; Aichberger et al. 2009; Karlberg et al. 2010a). According to recent molecular genetic findings (Molderings 2015, 2016; Haenisch et al. 2014; Lasho et al. 2016), the subclasses and clinical subtypes of MCAD do not represent distinct disease entities but should be more accurately regarded as variable presentations of a common generic state of MC dysfunction (Molderings et al. 2007, 2010; Hermine et al. 2008; Akin et al. 2010). Due to both the widespread distribution of MCs and the great heterogeneity of aberrant mediator expression patterns, symptoms can occur in virtually all organs and tissues; hence, the clinical presentation of MCAD is very diverse, sometimes to the even-further-confounding point of presenting opposite abnormalities in different patients (or even in the same patient at different times, or in different sites in the same patient at the same time). While the prevalence of SM in Europeans ranges between 0.3 and 13 per 100,000 (Haenisch et al. 2012; Cohen et al. 2014; van Doormaal et al. 2013), the prevalence of MCAS may be as high as 17 % (in Germany; Molderings et al. 2013a, b).

Table 1 WHO 2008 diagnostic criteria for systemic mastocytosis (Valent et al. 2001)
Table 2 Classification of systemic mastocytosis (modified form Valent et al. 2007)
Table 3 Current provisional criteria to define mast cell activation syndrome (MCAS; modified from Afrin and Molderings 2014)

This review focuses on the current state of drug therapy in SM and MCAS and describes perspectives of promising new approaches for drug treatment. Compounds in various stages of preclinical and clinical development are summarized in tables. We first describe drugs that are currently available and either are used on a regular basis in MCAD therapy or have been used successfully in single MCAD cases. In this context, it should be noted that there is no official guideline for treatment of MCAD.

Treatment options

Due to its genetic roots, MCAD generally is regarded as incurable. Recent mutational studies revealed that each patient has an individual pattern of genetic and epigenetic alterations which may affect the intracellular signal transduction pathways and receptive sites involved in sensory perception. As a consequence, mediator formation and release as well as inhibition of apoptosis and/or increase in proliferation are determined by individual genetic and epigenetic conditions (Fig. 2) and represent potential targets for therapy. Hence, there is need of highly personalized therapy for the disease. Unfortunately (with regard to easy detection), most genetic alterations (with a few exceptions such as certain mutations in tyrosine kinase KIT, e.g., KITD816V) do not alter the morphology and immunohistochemistry of the surface of the affected MCs. Thus, in most cases except for patients with the reliably identifiable D816V mutation, it cannot be decided by simple tests whether MCs found in biopsies are genetically altered MCs or physiological MCs.

Fig. 2
figure 2

Scheme of conditions responsible in MCAD for the development of individual phenotypes

First-line treatment options

Step 1 in managing most situations of inappropriate MC activation is identifying the individual patient’s unique triggers (chemical, physical, or otherwise) as precisely as possible and then desensitizing when possible (in truth, rarely) and otherwise practicing avoidance. With respect to drug treatment, only a few clinical therapeutic trials have been conducted in SM (midostaurin, cladribine, masitinib; Table 4), and there have been no therapeutic trials in MCAS yet. Most information about therapeutic effectiveness in MCAD has been found in small case series (Table 4) and single case reports, perhaps unsurprising given the mutational heterogeneity of the disease and thus the heterogeneity of its patterns of clinical presentation and therapeutic responsiveness. Therefore, in the future, it may be helpful to establish an international patient registry in partnership with existing registries so that issues related to molecular and clinical MCAD phenotypes can be adequately addressed. As the primary feature of MCAD is inappropriate MC activation (Molderings et al. 2011a, b; Pardanani 2013; Cardet et al. 2013), mainstays of first-line management are identification and avoidance of triggers plus therapies to control MC mediator production (both primary as well as secondary/reactive; Table 5) as well as their action (Table 6).

Table 4 Case series and clinical therapeutic trials in systemic mastocytosis and mast cell activation syndrome
Table 5 First-line drugs which can potentially be used in the treatment of mast cell (MC) activation disease and their target location and mechanisms of action
Table 6 Symptomatic treatment (orally as needed) in MCAD (modified from Molderings et al. 2014)

Subordinate therapeutic options

Continuous diphenhydramine infusion

Occasional patients suffer nearly continuous anaphylactoid and/or dysautonomic states poorly controlled by intermittently dosed epinephrine, antihistamines, and steroids. As discussed in more detail below, some such patients are particularly triggered by a wide range of medication excipients, making it challenging for them to tolerate trials of any adulterated (non-pure) medications, and yet some modicum of stability is required to pursue medication trials in such patients. Diphenhydramine is a well-tolerated histamine H1 receptor blocker (that among other non-threatening adverse affects can cause dizziness and an increase in appetite) which can quickly suppress MC activation and is used to treat allergic reactions and anaphylaxis. However, its half-life is as short as 1 h (www.drugbank.ca/drugs/DB01075). Intermittently dosed, though, its initial therapeutic serum level rapidly declines to subtherapeutic levels and the patient seesaws into yet another flare. The safety of continuous diphenhydramine infusion was established in trials of the “BAD” regimen (diphenhydramine [Benadryl], lorazepam [Ativan], and dexamethasone) in refractory chemotherapy-induced emesis in adult and pediatric patients (Dix et al. 1999; Jones et al. 2007). In a small series of ten MCAS patients suffering almost continuous anaphylactoid/dysautonomic flares, continuous diphenhydramine infusion at 10–14.5 mg/h appeared effective in most patients at dramatically reducing flare rates and appeared safely sustainable at stable dosing for at least 21 months (Afrin 2015). Stabilization has enabled successful trials of other helpful medications, but no patient has yet successfully stopped continuous diphenhydramine infusion.

Acute and chronic immunosuppressive therapies

Though typically not first-line, acute and chronic immunosuppressive therapies can be considered (Fig. 3; Table 7) and may be particularly appropriate for patients possibly manifesting an autoimmune component of the disease as might be suggested by the presence, for example, of anti-IgE or anti-IgE-receptor antibodies. Glucocorticoids may exert beneficial effects in MCAD, including a decrease in production of stem cell factor (SCF, and possibly other cytokines) and a decrease in MC activation, by various mechanisms which have been extensively reviewed by Oppong et al. 2013. Glucocorticoids at doses >20 mg prednisone equivalent per day are frequently needed to effectively control otherwise refractory acute (and chronic) symptoms. Their chronic toxicity profile is disadvantageous for long-term use, but such toxicities have to be accepted in some cases. The influence of azathioprine, methotrexate, ciclosporine, hydroxyurea, and tamoxifen on MC activity can vary from no to moderate effect depending on individual disease factors. As in therapy of rheumatoid arthritis, azathioprine and methotrexate can be used in daily doses lower than those used in cancer or immunosuppressive post-transplant therapy. Effective MCAD therapy with ciclosporine requires doses as high as those used in transplantation medicine (M. Raithel, personal communication). Methotrexate has to be administered parenterally to be effective (unpublished observation, G.J. Molderings), and in the risk-benefit analysis, a possible non-immunologic histamine release from MCs (Estévez et al. 1996) has to be considered. Hence, use of the compound should be limited to MCAD with methotrexate-sensitive comorbidities (e.g., rheumatoid arthritis and vasculitis).

Fig. 3
figure 3

Suggested treatment options for mast cell activation disease. All drugs should be tested for tolerance in a low single dose before therapeutic use, if their tolerance in the patient is not known from an earlier application. For further details of indication, see text

Table 7 Second- and third-line drugs which can potentially be used in the treatment of mast cell activation disease and their target location and mechanisms of action

Recently, the humanized anti-IgE murine monoclonal antibody omalizumab has been described in multiple case reports as safe and effective in MCAD (e.g., Molderings et al. 2011b; Kontou-Fili et al. 2010; Bell and Jackson 2012; Kibsgaard et al. 2014), though a definitive trial has yet to be conducted. Since treatment with omalizumab has an acceptable risk-benefit profile, it should be considered in cases of MCAD resistant to at least a few lines of therapy. The drug’s expense likely consigns it to third-line (or later) treatment (Table 7). If elevated prostaglandin levels induce symptoms such as persistent flushing, inhibition of cyclooxygenases by incremental doses of acetylsalicylic acid (ASA; 50–350 mg/day) may be used with extreme caution, since ASA can induce MC degranulation probably due its chemical property as an organic acid. The leukotriene antagonist montelukast (possibly more effective at twice-daily dosing; personal observation, L.B. Afrin) and the 5-lipoxygenase inhibitor zileuton may be useful adjuvants in people with MCAD, particularly in those with refractory gastrointestinal and urinary symptoms (Tolar et al. 2004; Turner et al. 2012; Akhavein et al. 2012).

Studies of kinase inhibitors, both on-market (e.g., imatinib, nilotinib, dasatinib) and experimental (e.g., midostaurin, masitinib), have yielded variable responses in SM ranging from no response to partial or even complete responses (Fig. 3; Table 8). As with all drugs used in therapy of MCAD, their therapeutic success seems to be strongly dependent on the individual patient, again underscoring the observed mutational heterogeneity of the disease. In formal studies in SM patients, although some kinase inhibitors reduced MC burden as reflected by histological normalization in bone marrow and improved laboratory surrogate markers (e.g., tryptase level in blood), at best only partial improvement of mediator-related symptoms was achieved (Droogendijk et al. 2006; Gotlib et al. 2008; Verstovsek et al. 2008; Vega-Ruiz et al. 2009). There has been repeated suggestion that symptoms in MCAD may be due more to mediator release from normal MCs secondarily activated by pathologically overactive, mutated MCs (Galli and Costa 1995; Rosen and Goetzl 2005; Boyce 2007; Kaneko et al. 2009; Fig. 2 in Molderings et al. 2014), helping to explain why intensity and pattern of symptoms do not correlate with degree of MC proliferation and infiltration (Topar et al. 1998; Hermine et al. 2008; Broesby-Olsen et al. 2013; Erben et al. 2014; Quintás-Cardama et al. 2013). Distinction in pathways in the MC which promote MC proliferation vs. mediator production/release may explain why kinase inhibitors reduce MC burdens and MC-driven symptoms to different degrees (Droogendijk et al. 2006; Gotlib et al. 2008; Verstovsek et al. 2008; Vega-Ruiz et al. 2009; Table 8). However, in some case reports, kinase inhibitors have been significantly effective at relieving symptoms. Thus, in spite of potential serious adverse effects of these drugs, a therapeutic trial may be justified in individual cases at an early stage. Partial and complete responses have been reported with some of these agents in MCAS too (e.g., Afrin 2010, 2011, 2012, 2015; Afrin et al. 2015a). Dosing of the kinase inhibitors in the individual often is considerably lower than how such drugs are dosed for other applications (e.g., imatinib, sunitinib; Afrin et al. 2015a). Possibly due to the causative mutations in multiple genes leading to simultaneous activation of multiple intracellular pathways, multitargeted kinase inhibitors such as midostaurin and sunitinib may be more effective than drugs which selectively downregulate only one intracellular pathway.

Table 8 Kinase inhibitors which can potentially be used as fourth-line drugs in the treatment of mast cell activation disease and their target location and mechanisms of action

In the mastocytosis patient with significant MC burden and/or an aggressive clinical course, cytoreductive drugs are prescribed (Lim et al. 2009; Valent et al. 2010). Unfortunately, effective cytoreductive therapies in SM presently are few in number and typically offer only modest response rates, qualities, and durations. Cytoreductive options include interferon-α and 2-chlorodeoxyadenosine (cladribine, 2-CdA; Fig. 3 and Table 9). Interferon-α is frequently combined with prednisone and is commonly used as cytoreductive therapy for aggressive SM. It ameliorates mastocytosis-related organopathy in a proportion of cases but can be associated with considerable adverse effects (e.g., flu-like symptoms, myelosuppression, depression, hypothyroidism), which may limit its use in MCAD (Simon et al. 2004; Butterfield 2005). PEGylated interferon-α has been shown to be as efficacious as and less toxic than the non-PEGylated form in some myeloproliferative neoplasms, but it has not been specifically studied in MCAD. 2-Chlorodeoxyadenosine is generally reserved for last-choice treatment of patients with aggressive SM who are either refractory or intolerant to interferon-α. Potential toxicities of 2-CdA include significant and potentially prolonged myelosuppression and lymphopenia with increased risk for opportunistic infections.

Table 9 Last-choice drugs which can potentially be used in the treatment of mast cell activation disease and their target location and mechanisms of action. R-review article (further references therein)

Last resorts

Polychemotherapy, including intensive induction regimens of the kind used in treating acute myeloid leukemia, as well as high-dose therapy with stem cell rescue, are approaches restricted to rare, selected patients. Allogeneic stem cell transplantation sometimes yields remissions in mastocytosis long thought impermanent (Spyridonidis et al. 2004; Nakamura et al. 2006; Bae et al. 2013; Gromke et al. 2013), though recent data may offer new hope (Ustun et al. 2014).

Investigational drugs

There are several drugs approved for indications other than MCAD which already have been successfully used in isolated cases with MCAD (Table 10). In cases of unsuccessful first- to fourth-line therapy, these compounds may be considered as treatment options.

Table 10 Drugs successfully (or not) used off-label to treat isolated cases of mast cell activation disease

A variety of drugs have been shown to inhibit MC growth, to decrease MC mediator release, and/or to relieve mediator-induced symptoms in in vitro and in vivo animal models (Table 11). Some of these drugs are approved for certain indications (such as ambroxol, statins, mefloquine, and ruxolitinib) and, thus, may be used (if accessible given financial considerations for some agents) if MCAD patients suffer from both the disorder of indication (e.g., hypercholesterolemia—statins, mucous congestion—ambroxol, polycythemia vera—ruxolitinib) and MCAD. An important question is what the role of the other compounds without approved indications should be in clinical practice. There are several challenges that may hamper the clinical introduction of novel targeted therapies in general. Some of these challenges include inherent problems in the translation of preclinical findings to the clinic, the presence of multiple coactive deregulated pathways in the disease, and questions related to the optimal design of clinical trials (e.g., eligibility criteria and endpoints). In particular, the testing of novel targeted treatment in an isolated fashion may be problematic and may in fact underestimate the effectiveness of these novel compounds. It is reasonable to assume that combination therapy will be the key to target parallel critical pathways.

Table 11 Investigational drugs which might have activity against mast cell activation disease since they induce apoptosis of mast cells and/or suppress mast cell mediator release in vitro and/or in vivo

General considerations on drug treatment of MCAD

Although no biomarkers of symptomaticity or therapeutic response are yet validated, the tolerability and efficacy of most therapies tried in MCAD (starting, and escalating in dosage and composition, cautiously) become clinically evident within 1–2 months. Modest experiments with alternative dosages and/or dosing frequencies are not unreasonable. Therapies clearly shown clinically helpful should be continued; therapies not meeting this high bar should be halted to avoid the troublesome polypharmacy that can easily develop in such patients. With no predictors of response yet available, a cost-based approach to sequencing therapeutic trials in a given patient seems reasonable. It is not even clear yet that medications targeted at mediators found elevated in diagnostic testing (e.g., antihistamines in patients with elevated histamine, non-steroidal anti-inflammatory drugs in patients with elevated prostaglandins, leukotriene inhibitors in patients with elevated leukotrienes) are reliably effective, again perhaps unsurprising given the multitude of MC mediators and the complexity of the signaling networks dysregulated by the multiple mutations in MC regulatory elements present in most MCAD patients. Successful regimens appear highly personalized.

Multiple simultaneous (or nearly so) changes in the medication regimen are discouraged since such can confound identification of the specific therapy responsible for a given improvement (or deterioration). Ineffective or harmful agents should be stopped promptly. Prescribers should be aware that although rapid demonstration of intolerance of a new medication (or a new formulation of a previously well-tolerated medication) often suggests excipient reactivity as further discussed below, some active drug molecules themselves (e.g., cromolyn) sometimes cause an initial symptom flare which usually soon abates. Temporary waiver of gluten-, yeast-, and cow milk protein-containing foods during the initial 3–4 weeks of drug therapy can improve the response rate (Biesiekierski et al. 2011; Rodrigo et al. 2013; own unpublished experiences). When MCAD is suspected, therapies that strongly activate the immune system (e.g., vaccinations with live vaccines or autohemotherapy) must be given with caution (especially if similar therapies were previously already poorly tolerated), as such interventions sometimes dramatically worsen MCAD acutely and/or chronically.

Any drug can induce intolerance symptoms in the individual MCAD patient. In some MCAD patients, the disease creates such remarkable states of not only constitutive MC activation but also aberrant MC reactivity that such patients unfortunately experience a great propensity to react adversely to a wide variety of medication triggers. Those MCAD patients begin demonstrating (either acutely or subacutely) odd/unusual/weird/strange/bizarre/unexpected symptoms soon after beginning new medications. It is very important to note that such patients often demonstrate even a greater propensity to react to medication excipients (i.e., fillers, binders, dyes, preservatives) than to the active ingredients. When the patient tries one or more alternative formulations of a medication with the same active ingredient but sharing as few as possible (preferably none) of the excipients in the offending formulation, the patient may discover the medication to be at least tolerable and perhaps even quite effective. Furthermore, such a scenario obviously provides the patient (and physician and pharmacist) a great opportunity to identify one or more of the specific excipients which are triggering abnormal reactivity in the patient’s dysfunctional MCs, and it is those specific excipients—not the medication as a whole—that should be added to the patient’s allergy list and screened against all present medications being taken by the patient and against all future medications proposed for the patient. An MCAD patient’s physician would be wise to not assume, just because an excipient is very widely used in many medication products and appears innocuous and well tolerated in the vast majority of patients, that the same excipient will necessarily be tolerated well in MCAD patients (unpublished observation of the authors). Sometimes the specificity of the reaction is quite extraordinary. For example, patients who react to wood-based microcrystalline cellulose might tolerate cotton-based microcrystalline cellulose without any difficulty at all, or vice versa. In some cases, the pharmacist is unable to identify alternative commercially available formulations sharing few to none of the excipients in the offending formulation, and in those cases, a compounding pharmacist may need to be engaged to identify/develop a custom-compounded formulation the patient can tolerate. (There can be geographic and financial challenges in accessing compounding pharmacies, though.) Occasionally, MCAD patients may be so remarkably reactive to such a wide range of excipients that they can only tolerate a given medication when provided as pure drug salt, reconstituted in water (without preservatives). Intolerance symptoms can be mediated by IgE antibodies, though this scenario appears to be rare since the symptoms are usually not ameliorated by the anti-IgE monoclonal antibody omalizumab (unpublished observation, G.J. Molderings). Alternatively, they may be mediated by IgG antibodies, raising the question of whether gamma globulin (if itself tolerable) might be a helpful adjunct therapy in such patients (perhaps by directly targeting the MC surface’s IgG receptors or via indirect pathways). Recently, a MC-specific receptor termed MRGPRX2 has been identified which appears to be crucially involved in pseudo-allergic drug reactions (McNeil et al. 2015; Seifert 2015).

Drugs which should not be used in MCAD

Several drugs have the ability to trigger MC mediator release. A compilation of drugs known to be associated with a high risk of release of mediators from MCs is given in Table 12. However, there often are therapeutic alternatives to these drugs (Table 12).

Table 12 Compilation of drugs associated with a high risk of release of mediators from mast cells and their therapeutic alternatives (compiled from Mousli et al. 1994; Sido et al. 2014; Afrin et al. 2015b; McNeil et al. 2015)

Conclusions and future perspectives

The therapeutic management of individuals with MCAD is complex and requires reviewing the entire spectrum of symptoms. The paucity of randomized, controlled studies makes treatment of refractory disease challenging and requires patience, persistence, and a methodical approach on the parts of both patient and managing provider(s). Delayed control of the symptoms may increase morbidity. Effective therapy often consists simply of antihistamines and MC-stabilizing compounds supplemented with medications targeted at specific symptoms and complications (Table 13). Current treatment options for refractory disease are based mainly on observational studies and case reports. Until larger randomized, controlled trials become available to give more guidance on therapy for refractory disease, clinicians should use the available data in conjunction with their clinical expertise and the adverse effect profile of the available drugs to make treatment decisions. More research is certainly needed to better understand MCAD pathobiology, in particular to determine which deregulated genes contribute to a specific symptom or symptom cluster. The greatest challenge in translational research for the discovery of new rational therapies requires a highly interactive interdisciplinary approach engaging basic science labs and clinicians. Understanding of the key components might hasten the progress of novel treatment for all these devastating MCAD phenotypes.

Table 13 Schematic summary of selected potential targets of pharmacological interventions in MCAD