Partners Asthma Center Grand Rounds

Jonathan Arm, M.D., M.R.C.P.
Critical Analysis of Immunotherapy in Treatment of Asthma

The controversy that surrounds the role of allergen immunotherapy in the treatment of asthma is highlighted by the following two recently expressed and widely contrasting views.

"There can be few indications for the use of allergen immunotherapy in the routine management of asthma in view of its low efficacy, the risk of adverse effects, the relatively high cost of treatment, and the availability of safer and more effective drug therapy" (Peter Barnes).

"Immunotherapy is the only treatment available today that has the potential to suppress the basic mechanism and to reduce the underlying cause of allergic asthma. Many ... studies have unequivocally shown that properly administered immunotherapy reduces both asthma symptoms and the need for concomitant medications" (Philip Fireman, President of the Academy of Asthma, Allergy and Immunology).

In this discussion I will first review the efficacy of immunotherapy in the management of anaphylaxis, especially due to stinging insects, and of seasonal allergic rhinitis. We will examine some of the scientific basis for immunotherapy, and then look at immunotherapy in asthma, drawing on a number of previous reviews on the subject and a recent meta-analysis. We will then look critically at two recent and important studies on allergen immunotherapy in asthma published in the New England Journal of Medicine, one on ragweed immunotherapy in adults, the other on immunotherapy in children. Finally, we will try to draw some balanced conclusions based on the objective data reviewed.

Immunotherapy in Allergy to Stinging Insects

In the United States there are approximately 60-70 deaths each year from bee sting (hymenoptera) anaphylaxis. Immunologic reactions to hymenoptera can be categorized into several basic types: large local reactions and hives (reactions confined to the skin); angioedema (which may involve the larynx and cause upper airway obstruction); bronchospasm; and anaphylaxis. All but the first two are potentially life-threatening reactions. For those who have suffered an anaphylactic reaction, the risk of suffering another severe anaphylactic reaction on repeat sting within the next few years is 50-70%. On the other hand, immunotherapy with hymenoptera venom reduces the risk of a systemic reaction to 3%, with most of those remaining reactions being mild. Any person who has suffered more than a local cutaneous reaction to bee sting should be referred for immunotherapy; to do otherwise would be negligent.

Immunotherapy in Hay Fever

The benefit of allergen immunotherapy has been well documented for seasonal allergic rhinitis and conjunctivitis ("hay fever"). As has been convincingly shown in many studies, patients maintained on immunotherapy have a marked reduction in symptoms and need for medications.

Immunotherapy is indicated in seasonal allergic rhinitis for those patients in whom severe symptoms interfere with their lifestyle despite use of topical nasal steroids and the newer generation of non-sedating antihistamines.

Scientific Basis for Immunotherapy

More and more information is available about the mechanism by which immunotherapy may modify the allergic response. Allergen challenges in the skin, nose, and airways may elicit not only an immediate reaction (developing within minutes and resolving within one hour) but also a late reaction, occurring usually by about three hours, lasting typically 6-12 hours, and often being more severe than the immediate response. The late-phase reaction is considered particularly important because it is associated with an influx of inflammatory cells into the site of challenge, amplifying and prolonging the allergic response. Allergen immunotherapy is effective in inhibiting both the immediate and the late-phase reactions. In addition, studies of the early and late cutaneous reactions to antigen challenge have been able to demonstrate that following immunotherapy there is a decrease in the release of mediators (e.g., histamine, prostaglandins, and leukotrienes) from mast cells during the immediate response.

During allergen immunotherapy with ragweed, there is a marked rise in circulating concentrations of ragweed-specific immunoglobulin G (IgG) and inhibition of the seasonal rise in serum IgE that one expects in these ragweed allergic patients. The observed increase in IgG has led to the concept that this immunoglobulin acts as a blocking antibody. Circulating IgG may in some way block access of the allergen to IgE or to the mast cell, or it may bind to the mast cell and through recruitment of inhibitory IgG receptors inhibit the mast cell response.

The response of T-lymphocytes in persons receiving allergen immunotherapy has also been examined. Observations indicate that allergen immunotherapy causes decreased lymphocyte production of the cytokines interleukin 4 and interleukin 5 and increased production of interferon gamma. Their findings are significant because they suggest that allergen immunotherapy leads to decreased cytokine production by the T helper lymphocyte type 2 (TH2) cells (involved in the allergic response) and increased production by the T helper lymphocyte type 1 (TH1) cells (involved in immunologic responses to infections).

Immunotherapy in Asthma

There is good reason to think of asthma as an allergic disease in which allergen immunotherapy might be effective. Especially among children, the vast majority of persons with asthma are also atopic. A recent article in the New England Journal of Medicine emphasized the role of allergen sensitivity and allergen exposure in causing severe asthma among children. This study of asthma among children living in inner cities across the United States found a rate of hospitalization that was 3-4 times higher among children with cockroach allergy who were exposed to high levels of cockroach antigen when compared with children who were cockroach allergic but had low levels of cockroach exposure or children who had high levels of cockroach exposure but were not cockroach allergic. Several studies have implicated exposure to high levels of house dust mite in the development of asthma. One study linked asthma deaths to sensitization to, and exposure to, the mold spore Alternaria.

Bronchoalveolar lavage data from patients with asthma support the importance of the allergic response: in the lavage fluid one finds a significantly increased number of T lymphocytes expressing messenger RNA for interleukins 4 and 5 (the TH2 type of cytokines) and no increase in interleukin 2 or interferon gamma (the TH1 type of cytokines). With data such as these, one can make a strong argument that in many individuals asthma is an allergy-driven disease.

Let me begin our consideration of allergen immunotherapy in the treatment of asthma with mention of two studies in which a beneficial outcome was clearly demonstrated. In the first, a placebo-controlled trial, persons with allergy to animal danders (cats or dogs) were treated with cat or dog dander extracts or placebo injections. Airway reactivity to the allergen was measured before and after the intervention. Placebo had no effect, whereas allergen immunotherapy resulted in a 10-fold decrease in the sensitivity of the airways to the allergen. The beneficial effect was significant in those individuals receiving immunotherapy to cat but not to dog dander, probably reflecting the quality of the allergen preparation that was used.

In the second study, early- and late-phase reactions to inhaled house dust mite antigen were compared before and after specific allergen immunotherapy to house dust mites. Allergen immunotherapy reduced the number of persons developing a late asthmatic response to allergen challenge by 50%, and elimination of the late asthmatic response was associated with clinical improvement in these patients' asthma.

In 1989 Dr. John Ohman published a set of useful criteria by which to judge studies done to evaluate allergen immunotherapy in asthma (Table 1). These include the following: 1) double-blinded and placebo-controlled; 2) random assignment into treatment groups with stratification based on the degree of allergic sensitization; 3) adequate clinical criteria for allergic asthma (both skin test positivity and symptoms on exposure to the particular allergen); 4) use of standardized allergen preparations; and 5) adequate indices of response with appropriate statistical analyses.

Return to Text

Table 1: Optimal Design of a Trial of
Allergen Immunotherapy*

  • Selection of subjects with allergic asthma
    evidence for sensitization
    evidence that symptoms are related to allergen exposure
  • Use of standardized extracts
  • Randomization with stratification according to severity of asthma
  • Double-blilnd, placebo-controlled
  • Adequate statistical analyses
* Criteria proposed by Dr. John L. Ohman

Return to Text

Using these criteria to select studies from the medical literature, Dr. Abramson and colleauges performed a meta-analysis regarding the efficacy of allergen immunotherapy in the treatment of asthma. The studies showed a wide variation in the odds ratio for improvement with immunotherapy, with some positive and some negative studies. Overall, however, based on their meta-analysis, these authors found a significant benefit from treatment with allergen immunotherapy, in terms of improved symptoms, reduction in medication use, and decreased airway reactivity to the specific allergen.

Recent Controlled Clinical Trials of Immunotherapy for Asthma

Two recent randomized trials of allergen immunotherapy for asthma bear closer scrutiny. One study was performed in adults (aged 16-70) with ragweed allergy. Subjects not only had positive skin test reactions to ragweed, but had to have much stronger reactions to ragweed than other relevant seasonal antigens. Patients were excluded if they had severe asthma (long-term oral steroids or recent hospitalization), were cigarette smokers, had received prior immunotherapy, or were exposed to perennial allergens (e.g., pets) that they could not or would not eliminate. Individuals meeting the inclusion criteria entered an observation phase from July to October during which they recorded their peak flows, symptom scores, and medication use. Patients were recruited into the study only if they demonstrated worsening of their asthma symptoms, a fall in their peak flows, and an increase in their medication use during ragweed pollen season. The strictness of the entry criteria is demonstrated by the fact that despite screening 1000 persons with asthma, only 127 were eligible to enter the observation phase and only 90 were eligible to enter the actual study. Subjects were randomly assigned to receive immunotherapy with short ragweed or placebo. Seventy-seven persons entered the treatment phase, 64 completed one year, and 53 completed two years. The relatively high drop-out rate, particularly from the control group, confounds interpretation of the results.

The increase in ragweed-specific IgE that was observed in the placebo group was abolished in the ragweed immunotherapy group during the two years of active treatment, giving evidence that sufficient ragweed allergen was administered. In terms of the clinical outcome, during the first year of treatment the allergen immunotherapy group did not have a significant drop in peak flow or increase in medication use, unlike the placebo-treated group. However, in the second year there was little difference between the two groups: both groups had only a small deterioration in their athma during ragweed season. Potential explanations for the decreased efficacy of allergen immunotherapy during the second year in this study include less ragweed exposure for both groups that year or drop-out from the placebo-treated group of those persons with more severe symptoms.

The other recent study, also from investigators at Johns Hopkins Medical School, involved children (aged 5-12 years) with asthma and sensitivity to perennial allergens. All of the children had asthma for more than one year, used asthma medications at least 5 days out of the week, had two or more positive skin tests to relevant perennial allergens, and had serum IgE levels above the 95th percentile of the normal range. Eligible children were entered into an observational stabilization phase during which their asthma care was optimized. During this period, which lasted on average for more than one year, the children (and their parents) received asthma education, instruction in allergen avoidance (including a home visit), optimization of their medication regimen, and follow-up physician visits every three weeks. Three hundred and fifty children were recruited; 120 children actually entered the randomized trial. Subjects were randomized to receive immunotherapy with from 2 to 7 allergen extracts or with placebo for a mean duration of three years. The primary outcome measure was medication usage for control of asthmatic symptoms. The immunologic responses to immunotherapy in these children included a significant rise in allergen-specific IgG and a significant decrease in the size of the immediate allergen skin test response compared to baseline and to placebo.

However, no significant differences in medication use were found between the allergen immunotherapy and placebo groups. Both groups improved during the double-blinded treatment phase of the study; overall, no clinical benefit from allergen immunotherapy was demonstrated. Nevertheless, it may be argued that in routine clinical practice few patients achieve the high medical compliance (>90%) required for entry into the study, and even fewer will employ such vigorous allergen avoidance measures or receive such intensive follow-up. In this sense it might be said that the study simply failed to demonstrate that immunotherapy was more effective than conventional medical therapy combined with intensive allergen avoidance.

In a series of post-hoc subgroup analyses, children who were younger at the time of entry into the study seemed to derive significant benefit from the immunotherapy. The significance of this finding is uncertain, although it is possible, as the authors suggest, that institution of immunotherapy early in life soon after the onset of allergic asthma is necessary in order to modulate the fundamental processes in allergic inflammation.

One can see from close review of these two carefully conducted clinical trials that the value of allergen immunotherapy is less dramatic and more difficult to demonstrate in the treatment of asthma than it is in hay fever or bee sting allergy.

Adverse Reactions to Immunotherapy

What are the risks involved in allergen immunotherapy? Between 1985 and 1989, there were 17 deaths associated with allergen immunotherapy. In 15 of these deaths detailed information was available for analysis. Thirteen of the 15 persons had asthma, and in 12 of these 13 the asthma was considered severe or labile. In the majority of these fatalities, airway compromise was the primary or contributing cause of death. Based on this information, one can estimate the risk of death from allergen immunotherapy to be approximately 1 in every 2 million shots given. However, most of these fatalities occur in those with severe or labile asthma. Of individuals given immunotherapy, it is estimated that asthma is the indication in perhaps only 25%. Therefore, one would deduce that the risk of death from immunotherapy is considerably greater in those with labile or severe asthma.

Given that there are safe and effective medications available to treat persons with mild to moderate asthma and that persons with severe asthma are at greater risk for severe adverse reactions, including death, from immunotherapy, the window of therapeutic utility for allergen immunotherapy in asthma becomes quite narrow.

Future Directions

Proponents of allergen immunotherapy argue that it is the only treatment presently available that alters the immunologic basis of allergic disease. Various alternative forms of immunotherapy are being explored. These include immunotherapy with peptides designed to attenuate the TH2 cell response to allergen yet escape recognition by B cells and avoid the risk of anaphylactic reactions. Several studies are now in phase III clinical trials. Encouraging results have also been obtained in animals with immunotherapy using plasmid DNA.


Let me conclude with my own set of questions to consider when selecting a patient with asthma for possible allergen immunotherapy (Table 2). First, every effort should be made to optimize conventional medical treatment, including rigorous pursuit of allergen avoidance. In the patient who remains symptomatic despite these efforts, consider the following: Is there evidence for allergic sensitization to one or more allergens? Does the history indicate a relation between exposure to these allergens and asthma symptoms or exacerbations? Is allergen exposure a major part of the person's overall asthma? Does the overall severity of asthma warrant this approach? Is an adequate allergen preparation (well characterized and standardized) available for immunotherapy? Is the patient likely to be compliant, including usually weekly office visits for the first 3-6 months? Does the risk of a severe adverse reaction justify the treatment? "Yes" answers to all of these questions would identify a potential candidate for allergen immunotherapy in asthma.

Return to Text

Table 2: Selection of Asthmatic Subjects
for Consideration of Allergen Immunotherapy

  • Demonstration that asthma is allergen-induced symptoms correlate with exposure evidence of sensitization
  • Adequate employment of allergen avoidance
  • Optimization of conventional asthma therapy
  • Availability of a standardized allergen
  • Appropriate risk-benefit ratio

Return to Text

Suggested Readings

Creticos PS, Reed CE, Norman PS, et al. Ragweed immunotherpay in adult asthma. N Engl J Med 1996; 334:501-6.

Adkinson NF Jr, Eggleston PA, Eney D, et al. A controlled trial of immunotherapy for asthma in allergic children. N Engl J Med 1997; 336:324-31.

About the Author:

Jonathan P. Arm, M.D., M.R.C.P., is an allergist and member of the Division of Rheumatology, Allergy, and Immunology at the Brigham and Women's Hospital. He is actively engaged in basic science research on the mechanism of inflammation in asthma.