Latex Agglutination Turbidimetric Immunoassay Versus Enzyme-Linked Immunosorbent Assay for Helicobacter pylori: An Observational Study.

Background: Helicobacter pylori antibody levels in the blood are currently measured using an enzyme-linked immunosorbent assay (ELISA). In April 2016, FUJIFILM Wako Pure Chemical Corporation launched the “L-type Wako Helicobacter pylori antibody J” test, which is based on the latex agglutination turbidimetric immunoassay. In this study, we investigated the usefulness of the Wako test. Methods: We measured H. pylori antibody levels using both ELISA and Wako tests in 180 patients who underwent esophagogastroduodenoscopy at our hospital between September 2017 and February 2019. Ninety patients had H. pylori infections. We calculated the diagnostic accuracy, sensitivity, and specicity of each test and the concordance rate between the two tests. If the lower limits of 90% condence intervals (CI) for each diagnostic validity exceeded the 85% threshold, the usefulness of the diagnostic test was conrmed. Results: Diagnostic accuracy, sensitivity, and specicity were 94.4% (90% CI; 90.8–97.0%), 94.4% (90% CI; 88.7–97.8%), and 94.4% (90% CI; 88.7–97.8%), respectively, using the Wako test, and 94.4% (90% CI; 90.8– 97.0%), 88.9% (90% CI; 81.9–93.8%), and 100% (90% CI; 96.0–100%), respectively, using ELISA. The concordance rate between the two tests was high (κ = 0.8444). Conclusions: We conrmed the usefulness of the Wako test, especially when screening for H. pylori infection, due to its high sensitivity. Trial registration: We retrospectively registered data of this study.


Background
Helicobacter pylori, a gram-negative bacillus, was discovered in 1983 by Warren and Marshall and is known to persistently infect the human gastric mucosa and induce histological gastritis accompanied by neutrophilic in ltration [1]. It has been demonstrated that H. pylori infection is associated with the development of gastric ulcer/duodenal ulcers and gastric cancer [2]. In 1994, the International Agency for Research on Cancer, an external organisation of the World Health Organisation, reported that H. pylori is a cause of gastric cancer [3].
People in Japan were noti ed that eradication therapy for H. pylori-infected gastritis would be covered by health insurance starting February 2013; thus, the diagnosis of H. pylori infection became more prevalent.
Diagnostic methods for H. pylori infection with endoscopy include the culture test, microscopic test, and rapid urease test. The tests without endoscopy include the urea breath test (UBT), blood H. pylori antibody test, and faeces H. pylori antigen test [4,5]. According to the Japanese Society for Helicobacter Research guidelines [6], UBT is the most accurate among the tests, and its sensitivity and speci city are 98% [7].
The blood H. pylori antibody test is convenient and widely used to diagnose H. pylori infection in current clinical practice. The prevailing methods for this antibody test are the enzyme-linked immunosorbent assay (ELISA, Eiken Chemical, Tokyo, Japan) and the chemiluminescent method, both of which have limitations: analysis is time-consuming and requires special measuring devices.
In April 2016, FUJIFILM Wako Pure Chemical Corporation launched the "L-type Wako H. pylori antibody J" (L-HP•J) test (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan), which is based on the latex agglutination turbidimetric immunoassay.
Since this kit can be used with a general-purpose automatic analyser, it is less complicated than ELISA.
Many samples can be analysed simultaneously, rapidly, and more economically using this kit. Another advantage of this kit is that the presence of H. pylori infection can be diagnosed, and drugs can be prescribed to eradicate H. pylori on the same day as the blood test. Consequently, this kit might contribute to reduced medical costs; however, its accuracy has not been con rmed. In this study, we evaluated the diagnostic validity of the L-HP•J test, which has not yet been demonstrated, by calculating and comparing its diagnostic accuracy, sensitivity, speci city, and concordance rate with those of the standard ELISA method.

Patients
This observational study was conducted in clinical settings and included consecutive patients who underwent gastrointestinal endoscopy between September 2017 and February 2019. The following patients were excluded: those who had received eradication therapy for H. pylori or had undergone gastrectomy; those who had taken proton pump inhibitors within 2 weeks before undergoing the UBT; those who had taken antibiotics within 4 weeks before the endoscopy; those who had undergone head and neck surgery and were unable to undergo the UBT; those in whom advanced gastric cancer was detected endoscopically; and those who were suspected of having type A gastritis endoscopically (cases with severe atrophic gastritis of the corpus and mild or no atrophic gastritis of the antrum). Our medical staff interviewed the patients to determine whether they had received eradication therapy for H. pylori or had undergone gastrectomy in the past. Major reasons for undergoing gastrointestinal endoscopy included screening, testing positive for H. pylori antibodies, having abdominal symptoms such as upper abdominal pain, and workup for early gastric cancer. Ultimately, 180 patients were enrolled in the study. and automated (by auto) steps (Fig. 1). First, the specimen and buffer are manually dispensed into the wells of a 96-well microplate. This step may require a few minutes per sample, as processing speed is dependent on the technician's skill level. An automated washing step is then performed to remove any unbound antibodies. A number of additional incubation and washing steps are performed before the automated analysis of the microplate. The entire process takes approximately 87 minutes to analyse all 96 specimens. In contrast, the latex agglutination turbidimetric immunoassay does not require the incubation and washing steps that the ELISA does; hence, one specimen can be analysed in 10 minutes, and the results of 100 specimens can be obtained in 16 min. Reagents were provided by the FUJIFILM Wako Pure Chemical Corporation. Fresh plasma was used for analysis via ELISA, and cryopreserved sera (− 20 °C; harvested within 4 weeks) or fresh plasma were thawed and used for analysis via the L-HP•J test.
Endoscopic ndings Esophagogastroduodenoscopy (EGD) was performed with the Evis Lucera Elite system (Olympus Corporation, Tokyo, Japan) and a GIF-H290Z scope (Olympus Corporation). EGD was performed by endoscopists who have 5000 cases or more of experience in EGD. The grading of gastric mucosal atrophy by endoscopic ndings was based on the Kimura-Takemoto classi cation [8]: no atrophy (C-0); closed-type (C-1 to C-3), in which an endoscopic atrophic border is in the lesser curvature of the gastric body and does not extend beyond the cardia; and open-type (O-1 to O-3), in which an endoscopic atrophic border extends beyond the cardia and progresses into the greater curvature.
Findings that suggest active gastritis, such as spotty or diffuse redness from the gastric body to the fundus, accompanied by disappearance of the regular arrangement of collecting venules, enlarged/tortuous folds, mucosal swelling, and sticky mucus according to the Kyoto classi cation of gastritis [9], were considered as presence of H. pylori infection (Fig. 2).
Criteria for determining H. pylori infection status and methods to evaluate the kits H. pylori infection was diagnosed by integrating the endoscopic ndings and the UBT results as follows: Subjects who had no gastric mucosal atrophy or any ndings suggestive of active gastritis on EGD as described above, and whose UBT results were negative (≥ 2.5‰ was de ned as positive) (UBIT, Otsuka Pharmaceutical), were considered uninfected, while those who had ndings of either gastric mucosal atrophy or active gastritis detected by EGD with a positive UBT result were considered as infected. Patients whose endoscopic ndings and UBT results had a discrepancy, such as active gastritis detected endoscopically, indicating a current H. pylori infection, but negative UBT results, and conversely, patients who were not infected according to their endoscopic ndings but whose UBT results were positive, were excluded from the study.
Subjects whose endoscopic ndings and UBT results concurred in terms of the presence/absence of H. pylori infection were diagnosed as being positive or negative for H. pylori, and the sensitivity, speci city, and diagnostic accuracy of the L-HP•J test were determined. ELISA was performed in a similar manner, and the concordance rate with the L-HP•J test was calculated. In cases misdiagnosed by the L-HP•J test, the test results correlated with the clinical ndings.

Sample size calculation
According to a previous report from Japan, the sensitivity and speci city of the L-HP•J test for H. pylori positive cases were 96.4% (currently infected; 56 cases) and 97.7% (uninfected; 86 cases), respectively. Therefore, we set the expected value as 95%, and threshold value as 85% for both sensitivity and speci city. Based on the binomial test, a sample size of 85 cases was deemed to be required for both uninfected and currently infected cases using the following criteria: one-sided alpha of 5% and power of 90%. Ultimately, a total of 180 patients (90 uninfected and 90 currently infected cases) were enrolled to compensate for any dropouts.

Statistical analyses
All continuous variables are expressed as the median and range. For diagnostic performance, diagnostic accuracy, sensitivity, and speci city are presented as percentages with a con dence interval (CI) of 90%. To evaluate concordance between the test methods for positivity and negativity, the κ statistic was used.
McNemar's test, which evaluates marginal homogeneity between two test methods, was used to compare the L-HP•J test and ELISA. A P-value < 0.05 was considered statistically signi cant. Statistical analysis was performed using Statistical Analysis System (SAS) software 9.4 (TS1M5, SAS Institute Inc., Cary, NC, USA).

Ethics
This study was approved by the Institutional Review Board of Cancer Institute Hospital (No. 2016 − 1208), and performed in compliance with the principles of the Declaration of Helsinki and its later amendments. While recording the data for this study, all personal identifying information was removed. Informed consent for the use of pathological specimens and imaging data for research purposes was obtained from each patient.

Breakdown of H. pylori infection status
Among the 180 patients, 90 patients (50%) were uninfected and 90 patients (50%) had a H. pylori infection. Regarding the patient characteristics of uninfected and currently infected cases, the median age and percentage of male patients was 57 years (range; 37-76), and 65 years (range; 28-87), 45.6% and 57.8%, respectively. The uninfected group was younger in age. Among the currently infected subjects, the degree of gastric mucosal atrophy (the Kimura-Takemoto classi cation) was C-1 to C-3 in 51.1% (46/90) and O-1 to O-3 in 48.9% (44/90). Although the major comorbidity in the currently infected subjects was early gastric cancer, this was seen only in one patient among the uninfected subjects (Table 1).

Comparison of concordance rates with ELISA (conventional test)
The number of subjects who were positive for anti-H. pylori antibody using the ELISA and L-HP•J test were 80 and 90, respectively. The concordance rates between the L-HP•J test and ELISA were high (κ = 0.8444) (Table 3). Subsequently, using McNemar's test, we analysed the tendency for positive or negative results for H. pylori between the L-HP•J test and ELISA. The results revealed that ELISA had a signi cant tendency to evaluate the test cases as negative (P = 0.0129) ( Table 3). ELISA, Enzyme-Linked Immunosorbent Assay; L-HP•J, L-type Wako Helicobacter pylori antibody J; HP, Helicobacter pylori.

Investigation of cases misdiagnosed by the L-HP•J test
Cases misdiagnosed by the L-HP•J test included ve false negative cases (5.6%) and ve false positive cases (5.6%) ( Table 4). Among the false positive cases, one subject had a high titre, while the other four subjects had titres slightly above the cut-off value. All false negative subjects had open-type mucosal atrophy. In this study, we investigated a newly launched kit that can be used with a general-purpose automatic analyser: The L-HP•J test kit. The L-HP•J test can be performed using commercially available automated analysers, such as Hitachi 7180 (Hitachi High-Tech), JCA-BM6050 (JEOL Ltd., Tokyo, Japan), ARCHITECT c16000 (Canon Medical Systems, Tokyo, Japan), and DxC700AU (Beckman Coulter K.K., Tokyo, Japan). The L-HP•J test was launched in April 2016 by FUJIFILM Wako Pure Chemical Corporation and is based on the latex agglutination turbidimetric immunoassay. To our knowledge, this is the rst report to evaluate the diagnostic validity of this newly launched kit. We demonstrated that the diagnostic accuracy, sensitivity, and speci city of the L-HP•J test was approximately 95%, and the concordance rate between the L-HP•J test and conventional ELISA was high, with a κ statistic of 0.8444. Although ELISA showed high speci city, the 90% CI lower limits for sensitivity of ELISA did not exceed the threshold value of 85%. In addition to this, ELISA had a signi cant tendency to give false negative results (P = 0.0129). From these ndings, it can be considered that ELISA tends to underestimate the antibody titre, and that the L-HP•J test is a relatively more accurate kit.
The latex agglutination turbidimetric immunoassay is superior to ELISA in terms of throughput and reducing labour and is expected to greatly contribute to rapid testing. Kita et al. demonstrated that the diagnosis of H. pylori infection is improved when two strains from genotypes commonly seen in the Japanese population (#3 and #5) [10] are used. The L-HP•J test we evaluated in this study used the same two antigens as the chemiluminescent test. Our ndings show that the accuracy of the L-HP•J test is not inferior to ELISA and that it has a greater sensitivity. Regarding the false positive and false negative cases detected using the L-HP•J test, all but one false positive case had low-positive titres (between 4 U/mL and 8.7 U/mL). It has been reported that approximately 10% of individuals with low-positive titres detected using the L-HP•J test are uninfected [11]. Thus, we believe that this problem arises due to the established cut-off value. Regarding the false negative cases, there were many subjects with open-type mucosal atrophy. This is because, in patients with severe atrophic gastritis, H. pylori colonisation is decreased or has spontaneously disappeared, resulting in the decrease of serum antibody [12]. It is important to use additional tests, such as the UBT, to diagnose H. pylori infection correctly if the L-HP•J test is not consistent with endoscopic ndings, such as in the following two scenarios: (1) when an individual has an open-type mucosal atrophy and endoscopic ndings suggest current infection, but the L-HP•J test result is negative; or (2) when endoscopic ndings suggest no infection, but the L-HP•J test result has low-positive titres (between 4 U/mL and 8.7 U/mL).
In the present study, we considered both the results of the UBT, which has a high diagnostic accuracy [13], and endoscopic ndings based on the Kyoto classi cation of gastritis. We then diagnosed the presence or absence of H. pylori infection in those subjects whose UBT results and endoscopic ndings were consistent. Therefore, subjects who were previously infected with H. pylori and those who had type A gastritis could be excluded from the study. Moreover, this was a prospective study and antibody measurements using both kits were performed in-house, which increases the reliability of the study.
There are some limitations of the study. First, it represented a single-centre observational study. To con rm the generality of this kit, a multi-centre study is needed in the future. Second, in some subjects, fresh plasma was used for detection using ELISA and part of the L-HP•J test, while stored serum (harvested within 4 weeks) was also used for some of the L-HP•J tests; therefore, there were differences between subjects in terms of the conditions of the specimens used. However, supplemental data from FUJIFILM Wako Pure Chemical Corporation showed no difference in the titres between fresh plasma and stored serum (harvested within 4 weeks) under conditions of -20 °C. Third, we used the UBT as the gold standard for diagnosing H pylori infection; however, its accuracy is not 100% [13]. Fourth, we investigated only noninfected and currently infected cases, not previously infected cases. Therefore, a further study should be conducted to analyse the association between previously infected cases and the antibody titre of the L-HP•J test.

Conclusions
The L-HP•J test has a high diagnostic performance and is not inferior to ELISA. It is useful, especially in screening for Helicobacter-pylori infection, due to its high sensitivity. It can be used with various generalpurpose automatic analysers because the kit is versatile, easy to use, and convenient. It can, therefore, contribute to the streamlining of clinical laboratory procedures.

Declarations
Ethics approval and consent to participate All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and later versions. Informed consent Informed consent to be included in the study, or the equivalent, was obtained from all patients.

Consent for publication
Consent for publication has been obtained from patients.
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests