Gnathostoma spinigerum advanced third-stage larvae RNA (920 ng) was used for cDNA synthesis. The cDNA library was constructed and colonies showing positive immunoscreening with serum samples from gnathostomiasis patients were selected for cloning procedure. A clone (named Gslic18) with a strong positive reaction to IgG antibodies in serum samples from gnathostomiasis patients, but not reactive to serum samples from patients with other parasitic infections or those in the negative control group, was selected through immunoscreening. The cDNA insert of Gslic18 was subcloned into a pQE-31 expression vector and transformed into Escherichia coli XL-1 Blue (Qiagen, Hilden, Germany) as the expression system. The recombinant Gslic18 (rGslic18) protein was expressed as insoluble protein and solubilised using urea solution (8 M urea, 0.1 M NaH₂PO₄, 0.01 M Tris-HCl, pH 8.0). Then, the rGslic18 carrying 6-Histidine (6-His)-tagged residues was purified using Ni-NTA His Bind Resin (Novagen, Darmstadt, Germany) according to the manufacturer’s protocol. The purified protein concentration was determined [3]. All human serum samples used for diagnostic values were supplied by the frozen bank (-70 °C). The samples were divided into three groups: (i) the negative control group (n = 20) comprised samples from healthy adult volunteers who underwent stool examination [4] and were found to be free from any intestinal parasitic infection at the time of blood collection (a pooled serum consisting of samples from the healthy individuals was also used as a negative control for each assay); (ii) the gnathostomiasis group (n = 32), which included samples from parasitologically confirmed gnathostomiasis patients (n = 9) and from patients showing clinical symptoms of suspected cutaneous and visceral gnathostomiasis, and neurognathostomiasis (n = 23) [1] with a history of eating food possibly contaminated with Gnathostoma larvae and were positive 24/21 kDa G. spinigerum antigen by immunoblotting [2]; and (iii) the third group (n = 114), which consisted of serum samples from patients with parasitic infections other than gnathostomiasis. Their infections were confirmed by parasitological methods except in cases of cysticercosis which were diagnosed by a computerised tomography scan and found positive by the immunological method [5] (Table 1).

Type of serum samples	Number of positive/Total number
	ICT kit	Immunoblotting
Healthy control	0/20	0/20
Confirmed gnathostomiasis	9/9	9/9
Suspected gnathostomiasis	21/23	23/23
Cysticercosis	1/4	0/4
Taeniasis	0/10	0/10
Opisthorchiasis viverrini	0/15	0/15
Fascioliasis	1/5	0/5
Paragonimiasis	1/10	0/10
Angiostrongyliasis	1/10	0/10
Strongyloidiasis	0/10	0/10
Hookworm infection	0/10	0/10
Capillariasis	0/10	0/10
Ascariasis	0/10	0/10
Trichinellosis	0/3	0/3
Sparganosis	0/2	0/2
Malaria	0/10	0/10
Filariasis	0/5	0/5

The rapid ICT kit (named the “KAN gnathostomiasis kit, K, Khon Kaen University; A, Adtec Inc. Oita; N, National Institute of Infectious Diseases, Tokyo”) using the rGslic18 antigen was optimised based on the ELISA result using the same antigen at Adtec Inc., Oita, Japan. One mg/ml of anti-human IgG (H + L) (Medical and Biological Laboratories Co., Ltd., Nagoya, Japan) was absorbed at control line (C), 0.5 mg/ml of the rGslic18 antigen was absorbed at test line (T), and OD 1.0 of colloidal gold conjugated with anti-human IgG (Adtec Inc.) was sprayed on a piece of glass fiber (conjugate pad) (Fig. 1). The kit consists of an immunochromatographic device, sample buffer for diluting serum sample and buffer for chromatography. The diagnostic procedure is as follows: dilute serum samples with sample buffer in 1:50 and spot the aliquot (5 μl) onto inscription “sample”, and apply buffer (60 μl) onto inscription “buffer” (Fig. 1a). A criterion of the diagnostic result is whether a red band appears at the test line within 15 min or not. The intensity of the bands was estimated visually (unaided) according to the reference board (Fig. 1d, with level 2 as the cutoff level). Positive results were judged by the appearance of red band at the test (T) line and the control (C) line (Fig. 1b). Negative results were judged by the disappearance of a red band at test (T) line and appearance at control (C) line (Fig. 1c). The diagnostic parameters of sensitivity and specificity were computed as previously [6]. The diagnostic values of the ICT kit were evaluated with those of immunoblotting [2] and percentages of total concordance were computed (percent concordance (%) = (the number of concordances between ICT and the immunoblotting reference × 100)/the number of tested samples). Informed consent was obtained from all human adult participants and from parents or legal guardians of minors. The study protocol was accepted by the Khon Kaen University Ethics Committee for Human Research (HE571438).