Fish-collagen allergy: a pitfall of the prick-to-prick test with raw fish

Dear Editor, A 20‐year‐old Japanese woman presented with a history of nausea and weals after eating baked Japanese amberjack. She also had a history of urticaria after eating fish fingers, although the actual species of fish involved was unknown. Four months before admission to our department, she experienced generalized erythema and weals with throat discomfort after eating baked eel. Other past medical history was unremarkable. Laboratory tests revealed positive IgE specific for tuna, salmon [6·37 IU mL−1 (class 3) and 117 IU mL−1 (class 6), respectively; normal < 0·27 IU mL−1, fluorescence enzyme immunoassay (FEIA)], cod and plaice [4·47 UA mL−1 (class 3) and 1·15 UA mL−1 (class 2), respectively; normal < 0·34 UA mL−1, CAP‐RAST], but negative for sardine, Pacific mackerel, horse mackerel, shrimp, squid, octopus and crab. Tests for Anisakis‐specific IgE also showed negative results. We were unable to measure IgE against Japanese amberjack and eel. We then performed the prick‐to‐prick test with raw fish, including Japanese amberjack, tuna, salmon, Japanese sea perch and red sea bream. However, the patient did not respond to any of these fish. The most common fish allergen is known to be parvalbumin.1 It is a thermostable, calcium‐binding sarcoplasmic protein of various species of fish. In addition, fish collagen has also been identified as an important allergen for fish allergy.2 Inconsistent results of specific IgE and skin prick tests, despite apparent allergic episodes to fish, have prompted us to explore causative allergens and fish in detail. To this end, we attempted to identify specific IgE to parvalbumin and fish collagen by enzyme‐linked immunosorbent assay (ELISA), as previously reported.3 Briefly, parvalbumin and fish collagen (type I collagen) were purified from the meat and skin of Pacific mackerel, respectively. Purified allergens (1 μg mL−1) were coated on a microtitre plate, followed by incubation with serum from the patient. An enzyme reaction was produced using an ELISA POD Substrate TMB Kit Popular® (Nacalai Tesque, Inc., Kyoto, Japan) and the colour developed was measured by absorbance at 450 nm. Concomitantly, we performed ELISA with crude extracts from Pacific mackerel and horse mackerel following homogenization of the fish meat with phosphate buffer (10 mmol L−1, pH 7·0) and heat treatment (100 °C, 10 min). As a result, the patient was found to have specific IgE to fish collagen, but not parvalbumin (Fig. 1a). In addition, she also had Pacific mackerel‐ and horse mackerel‐specific IgE (Fig. 1a), which had not been detected using the FEIA method. Results of specific IgE from enzyme‐linked immunosorbent assay and skin prick tests. (a) IgE against fish collagen, Pacific mackerel and horse mackerel was detected. (b) A prick‐to‐prick test was performed with raw and heated fish. The diameter of positive control (1% histamine hydrochloride) was 7 × 7 mm. Collagen is a protein that becomes water soluble with heat treatment. Therefore, collagen is not present as a water‐soluble protein in raw materials.3 In addition, IgE‐binding epitopes of collagens are known to be heat stable.4 Therefore, we examined whether the prick‐to‐prick test with heat‐treated fish would result in positive reactions. Fish samples were heat treated in a microwave oven (500 W × 15 s per 10 g sample) and subjected to a prick‐to‐prick test together with raw fish. As expected, we observed weal reactions in the skin prick test with heated fish, including Japanese amberjack, tuna, salmon, Japanese sea perch, red sea bream and horse mackerel, while no responses were detected with untreated raw fish (Table 1 and Fig. 1b). Purified fish collagen also induced a positive reaction, but parvalbumin did not (Table 1). Results of prick‐to‐prick test −, equivalent to negative control; +, < 50% of positive control; ++, > 50% of positive control. Positive control: 1% histamine dihydrochloride. Negative control: saline. Results of prick‐to‐prick test −, equivalent to negative control; +, < 50% of positive control; ++, > 50% of positive control. Positive control: 1% histamine dihydrochloride. Negative control: saline. Fish collagen is an important allergen for fish allergy. While fish collagens from various species of fish cross‐react with each other, they are not cross‐reactive with collagens from mammals.2 3 Nevertheless, even if eating fish is avoided, patients with fish‐collagen allergy carry a risk of exposure to causative allergens in products such as gelatin, a denatured collagen that is used as a supplement for foods and possibly for cosmetics, although this is not as commonly used as bovine and porcine collagens. The prick‐to‐prick test is a reliable test for immediate‐type food allergy, although an oral challenge test is needed to confirm the causative foods. In our patient, the initial skin prick test with raw fish did not result in positive reactions. We performed the additional prick‐to‐prick test with heated fish to make the collagen soluble, based on the assumption that negative results with raw fish could have been due to its insolubility. Indeed, negative skin reactions with raw fish became positive after heat treatment. It should also be noted that, despite negative results for IgE against Pacific mackerel and horse mackerel as assessed using FEIA methods, we detected positive results for specific IgE against these fish using ELISA. This could be explained by the fact that we used heated fish extracts for ELISA assays. Our results suggest that the prick‐to‐prick test with raw fish may risk overlooking fish‐collagen allergy. Performing the prick‐to‐prick test with heat‐treated fish in addition to raw fish is recommended in patients with suspected fish allergy, although this may present a higher risk of adverse events than an ordinary skin prick test with standardized samples. Funding sources: none. Conflicts of interest: none declared.