Most individuals sensitized to cats react to Fel d 1, and standardized extracts are based on particular concentrations of this allergen. In contrast, a distinctive major dog allergen recognized by the specific IgE of dog-sensitized individuals has not been identified. Several dog allergens, including Can f 1, Can f 2, and Can f 3, have been traditionally considered relevant to particular individuals sensitized to dogs, with Can f 2 (lipocalin) and Can f 3 (albumin) being responsible for cross-reactivity among mammals. More recently, other dog allergens (Can f 4, Can f 5, and Can f 6) have been identified and proven clinically relevant to particular individuals and also responsible for cross-reactivity among species [1-3].
Mammalian allergens can occur in a variety of sources, and different raw materials are used for the production of allergenic extracts. These materials include dander, hair, epithelium, pelt, skin scrapings, and hides. Dog raw materials differ in allergenic composition, with dander containing greater Can f 1 concentrations than epithelia. Dander is the most common material used for the preparation of dog allergenic extracts.
The fact that no major clinically relevant dog allergens that affect most sensitized individuals exist, combined with the heterogenic composition of dog raw materials, is responsible for the presence of a number of products in the market with differential diagnostic value. Standardized allergenic extracts for immunotherapy are lacking.
In the May issue of Allergy, Polovic et al.  described the interesting finding that dog saliva contains allergens, not previously identified in dander, and suggested that saliva could have an increased diagnosis potential, compared with other materials. The authors of this article collected saliva samples from 14 individual dogs belonging to 11 different breeds and compared their allergenic content with that measured in a dog dander sample commercially obtained.
As expected, a large variation of different allergen concentrations was detected between individual dogs, and the authors propose that particular dog breeds could be less allergenic than others, based on selective allergenic content in saliva, and therefore, particular breeds could be recommended to dog-sensitized individuals. While this theory has been previously proposed, based on Can f 1 content measured in dander , due to the multiple factors responsible for dog allergen production and accumulation, the allergenic content in saliva may not correlate with that present in dander, and findings should not be extrapolated.
Studies that investigate the environmental components responsible for allergen accumulation are very difficult to perform due to a number of confounding factors involved. In fact, an attempt to determine the validity of the proposition that dog-sensitized individuals can tolerate some dog breeds better than others concluded that scientific evidence is lacking to provide recommendations in that regard .
The findings described in the article by Polovic et al.  should be verified using a larger sample size, and the nature of the differential allergens detected in saliva should be identified, particularly those common to different breeds. From a manufacturing and regulatory perspective, the qualitative and quantitative allergen compositions of raw materials are the critical parameters responsible for the quality and consistency of the associated allergenic extracts.
While the finding of large allergen diversity in dog saliva is scientifically interesting, this material may not be suitable for the production of allergenic extracts due to the potentially associated challenges of obtaining consistent formulations comprising a number of allergens. The intrinsic nature of sensitization to dog allergens is unique, and consistent diagnostic tools are needed. The production of individual dog allergens is an emerging technology with applications for diagnosis and treatment to target specific patient populations.