Antigen Choice for Skin Testing

Edwyn L. Boyd

The patient’s history is the most useful diagnostic tool available for identifying suspected allergenic offenders. Antigen selection for both screening and completion of testing may be in part empiric, but it is also based on portions of the history that suggest specific triggers, both seasonal and perennial. Prior to selecting the type and number of antigens to be tested, it is important to consider the patient’s age, occupation, living accommodations (including prior fire or water damage to the dwelling), pets, hobbies, and frequent travel to different geographic areas. Consequently, even though such a situation would be ideal, neither the screening panel nor a complete testing battery can be developed as a “one size fits all” tool because each patient has a different lifestyle with different allergen exposures.

Proper selection of antigens for allergy testing is critical to achieving a positive clinical outcome in patients receiving immunotherapy. Generally this process begins after the patient fails environmental avoidance and the proper use of pharmacotherapy. Avoidance of allergens, both known and suspected, is the basic tenet of allergy management, representing the treatment of choice. Unless the patient has only a few obvious specific triggers, however, avoidance may be difficult, if not impossible, to accomplish.

When the decision is made to perform allergy testing, the first step is the selection of an appropriate antigen panel for initial screening. Screening batteries are very important and should be routinely employed because they provide the patient and practitioner with an economical, rapid, same-day “yes/no” answer about the presence or absence of immunoglobulin E (IgE)-mediated disease. In 1982 King 1 found a diagnostic sensitivity of 99.2% using an in vitro screen of only six antigens. This initial screen included one representative pollen for trees (mountain cedar), grasses (Bermuda), and weeds (ragweed), plus two molds (Alternaria and Mucor) and one dust mite. A subsequent study by Lehr et al² showed an efficiency of nearly 100% using a “midiscreen” of nine antigens, and even better success with 13 antigens. It is currently accepted that the patient with negative results using a screening battery of up to 14 antigens will probably not benefit from further testing.

There is a significant financial difference between the use of a screen versus initial testing with 30 to 50 antigens. If positive results are obtained using the screening battery, however, then further testing with an expanded panel, including a realistic number of antigens, is performed to complete the testing process. Details of the process may be found in the references cited.

Prior to selecting allergens for use in the testing battery, practitioners must be familiar with relevant antigens in their own geographic region. This implies knowledge of not only prevalence, but also clinical correlation with the likelihood of producing symptoms. Furthermore, in the overall management of the allergic patient, “sins of omission” can be equally as detrimental as “sins of commission.” That is to say, not only may relevant allergens be omitted from the patient’s treatment set, causing a suboptimal outcome, but ignorance of the phenomenon of cross-reactivity, resulting in the inclusion of multiple similar allergens in the prescription, could provide effective overdosing, causing a serious systemic reaction.

Cross-reactivity, the sharing of like epitopes (antigenic combining sites), is likely to be a problem when more than one allergen from the same plant family is selected for testing and treatment, and is especially important among the grasses. The grasses exist within a single family with four subfamilies, two of which have the most clinical significance. Testing (and treatment) should include only one representative from the major subfamilies. For the Pooideae, either timothy or perennial rye grass is generally chosen. Bermuda grass is the usual representative of the Chloridoideae.³

Trees, on the other hand, are distributed among many unrelated families, making the likelihood of cross-reactivity producing overdosage less problematic. However, it is noteworthy that juniper, cedar, and cypress (members of the cypress family) cross-react strongly. In areas where oak is predominant, selection of a single oak should suffice for adequate coverage of the entire family.

Like trees, the weeds are also distributed among unrelated families, but short, giant, false, and western ragweed are similar enough that selection of one of these is adequate for testing and treatment. Short ragweed is the representative most often chosen for this family.

According to Ramanarayanan,4 cross-reactivity between grasses in the same genus is >95%, within the same tribe and subfamily >90%, within related tribes and same subfamily >75%, within distant tribes and same subfamily >50%, and within distant tribes and different subfamilies >20%. Weeds and trees within the same genus have a cross-reactivity of >95%, within the same family >75%, within related families >50%, and within unrelated families >20%.

Dust mites are not as significant as a cause of allergy in arid, higher elevations as they are in other climes. Significant antigenic similarity exists between Dermatophagoides farinae and D. pteronyssinus, but they are different enough that many practitioners choose to test and treat for each individually.

Molds are encountered indoors and outdoors in every geographic area of the United States. They thrive in warm, humid, climates, and thus are much more significant to allergic patients in the Gulf Coast states than to those living in the cooler Rocky Mountain and arid southwestern regions. Furthermore, there is little, if any, snow and ice on the ground in the South during the winter months, thus exposing the inhabitants to year-round, high concentrations of mold. This is in contrast to residents of northern and Rocky Mountain states, where such ground cover during the winter months reduces outdoor mold exposure significantly.

Alternaria is probably the most important outdoor mold. Other molds of clinical significance throughout the United States include, but are not limited to, Hormodendrum, Cephalosporium, Pullularia, and Helminthosporium. There are many other molds that are more clinically significant regionally than nationally.

Pollens are ubiquitous offenders across North America, and there are hundreds of different species of pollen-producing, seed-bearing trees and plants with the potential to induce allergic symptoms in genetically predisposed individuals. Selection of relevant seasonal pollens to include in a screening panel or complete testing battery must be based on the patient’s history and exposure to those allergens. One or two representative pollens from each of the weeds, trees, and grasses should be chosen based on the patient’s geographic region and the knowledge that that allergen is a known offender in that locale. Generally, weed pollen is responsible for symptoms in the fall, grass pollen in the summer, and tree pollen in the spring. Exceptions include regions where mountain cedar is prevalent, causing symptoms from early December through January, and in southern California, where grasses can be considered almost perennial offenders.

To stimulate the immune system and ultimately cause symptoms, pollens from plants must fulfill Thommen’s postulates. That is, their pollen must be small, produced in abundance in close proximity to human populations, be wind-borne, and possess a protein allergen capable of stimulating the immune system to induce a symptom producing reaction. Some plants have more potential for causing symptoms than others. This potential is related to the amount of pollen they produce and its potency, the extent of the area of their distribution, and the dispersion of their pollen based on its weight. In the southeastern part of the United States, for example, pine trees are found in abundance and they produce a tremendous amount of pollen, but they are not highly relevant offenders in the production of allergic symptoms. Another example would be crabgrass. Although it is found throughout North America, it does not produce a great amount of pollen, thereby limiting its concentration in the air and subsequent exposure to humans.

The choice of a screening battery must take into account the patient history, with attention to exposure and symptom production linked to season and circumstance, the prevalence of antigens in the region in question, and the clinician’s experience. A reasonable screening battery for relevant allergens in the southeastern United States would include 11 antigens: short ragweed, Bermuda grass, Timothy grass, hickory, oak, D. farinae, D. pteronyssinus, Alternaria, Hormodendrum,, cat, and dog. Other examples of regional screening test batteries and antigens to be considered for additional testing may be found in Appendix 2 at the end of this book.

Standardized allergenic extracts are commercially available, and should be used for testing and treatment when available. Standardization allows for an increased reproducibility of testing, as the variability in biologic activity of the extract between lot numbers within the same manufacturer and between manufacturers is lessened (but not totally eliminated). Prior to the availability of standardized extracts, broad variability in potency existed between lot numbers within the same manufacturer, as well as among the products of different manufacturers. Some standardized extracts are available in different concentrations. One example is dust mite, which is offered in strengths of 3000, 5000, and 10,000 allergy units (AUs).

Unfortunately, not all extracts are available in standardized form. In fact, of the hundreds of allergens available for purchase, only extracts for cat hair and pelt, D. farinae, D. pteronyssinus, short ragweed, Bermuda grass, Timothy grass, Kentucky bluegrass, perennial rye grass, meadow fescue, orchard grass, red top, sweet vernal, and Hymenoptera venom are available as standardized products.

The concentration of nonstandardized extracts is generally expressed as weight per volume (w/v), representing the weight of antigen in grams extracted in 100 mL of solution. The w/v concentration in most common use is a 1 : 20 strength, although other strengths are also available.

A third class of extracts has recently been introduced. These are “well-characterized extracts,” which have not met the Food and Drug Administration criteria for standardization but about which much more data are available than for the common w/v extracts. This manufacturerdriven initiative for better quality control and reproducibility among extract lots is in its infancy, but it holds great promise for the future.

Because there are literally hundreds of allergens from which to choose, the selection of antigens for testing and treatment can be a daunting task. Commercial manufacturers generally know which allergens are more likely offenders in each geographic region, and they can be immensely helpful in assisting the clinician with choices. The local county extension agent can also be very helpful, as can a local university botany department.

In conclusion, once the decision to perform allergy testing is made, the next task facing the clinician is selection of the allergens that are most appropriate for testing and treatment of a particular patient. Using the principles detailed here, a reasonable number of allergens can be chosen, and the process of screening, subsequent additional testing, and treatment can begin.