Characteristics of phenotypic antibiotic resistance of Helicobacter pylori and its correlation with genotypic antibiotic resistance: A retrospective study in Ningxia

Geographic differences exist in the antibiotic resistance patterns of Helicobacter pylori. Personalized treatment regimens based on local or individual resistance data are essential. We evaluated the current status of H. pylori resistance in Ningxia, analyzed resistance‐related factors, and assessed the concordance of phenotypic and genotypic resistance.


| INTRODUC TI ON
Helicobacter pylori infection continues to be a major health problem worldwide. It is estimated that more than half of the world's population is infected with H. pylori. 1 Furthermore, approximately 92% of gastric cancer is caused by H. pylori infection, and patients infected with H. pylori are four to six times more likely to develop gastric cancer than uninfected patients. 2,3 The virulence factor CagA plays a crucial role in the pathogenesis of H. pylori, which increases the risk of gastric atrophy and intestinal saprophy. 4 Eradication of H. pylori can delay or block the occurrence of atrophy and intestinalization of the gastric mucosa and even reverse gastric mucosal atrophy and other gastric precancerous lesions, significantly reducing the incidence and mortality of gastric cancer. [5][6][7] However, eradication of H. pylori is not an easy task. Historically, a single drug is not effective in eradicating H. pylori. 8 Only a few antibiotics, such as metronidazole, amoxicillin, and clarithromycin, are effective in eradicating H. pylori. A typical treatment regimen usually consists of two or three of these antibiotics, in combination with antacids and/or bismuth. 9 The availability of limited effective treatment options and widespread use of some antibiotics in the population (such as quinolones for infectious diarrhea), coupled with the adaptability of the species, is causing an increase in H. pylori resistance. 10 According to the WHO, the resistance of H. pylori in the Western Pacific region is increasing every year, with metronidazole, clarithromycin, and levofloxacin resistance rates rising from 52%, 32%, and 12% in 2006-2008 to 57%, 35%, and 31% in 2012-2016, respectively. 11 As of 2015, H. pylori in China is dominated by metronidazole resistance, and the resistance rates of clarithromycin and levofloxacin are around 20% and increasing year by year, with higher rates in the north than in the south. The resistance rates of amoxicillin and rifampicin are low and have no increasing trend. 12 Antibiotic resistance patterns vary between geographic locations and susceptible populations and may change over time. Therefore, antibiotics resistance patterns of specific regions and populations are the basis of empirical regimens for H. pylori eradication. 13 Traditional antibiotic susceptibility testing has several challenges and limitations. The growth of H. pylori is slow and requires a special three-gas incubator to culture. Antibiotic resistance phenotypes are determined by using the Kirby-Bauer (K-B) disk diffusion, agar dilution, or E-test methods. The whole process may take several weeks, 14 the success rate of culture is approximately 50%-70%, and these tests are not provided by the laboratory departments of most domestic hospitals. Studies have reported the resistance genes of H. pylori and methods, such as real-time PCR, DNA hybridization, and targeted sequencing, have verified the mutations related to clarithromycin and levofloxacin susceptibility. 15,16 Therefore, genetic testing may be a reliable way to obtain antibiotic susceptibility results for H. pylori and overcome the limitation of the small sample in an unsuccessful culture.
Located in northwest China, Ningxia is a region with a large population of the Hui ethnic group and has a high incidence of chronic gastric diseases, including gastric cancer. 17 The prevalence of H. pylori infection in Ningxia is above 60%, 18 and the current status of H. pylori resistance and its relationship with mutations in these resistance genes are unclear. Personalized therapy based on antibiotic resistance genes has gradually attracted attention. 19 It is important to standardize regional eradication programs, develop optimal eradication strategies, improve H. pylori eradication rates, and reduce local morbidity and mortality related to chronic gastric diseases and gastric cancer. Therefore, we reviewed the phenotypic and genotypic antibiotic resistance of H. pylori infection in Ningxia and analyzed the resistance-related factors. This was the largest and most extensive H. pylori antibiotic resistance study conducted in Ningxia and included patients from 13 different hospitals.

| H. pylori isolation and identification
After the gastroscopic biopsy operation, the specimens were im-

| Phenotypic antibiotic susceptibility testing
The sensitivity to the antibiotic metronidazole, clarithromycin, levofloxacin, amoxicillin, tetracycline, and furazolidone was tested.
Sensitivity was determined using Kirby-Bauer (K-B) disk diffusion method as described in our previous study. 20 Specifically, a standard inoculum with a turbidity of 0. Antibiotic paper sheets were pressed onto the agar surface and incubated for 48 h, and the diameter of the inhibition circle was measured. The following antibiotic paper sheets were used: clarithromycin (15 μg), levofloxacin (5 μg), metronidazole (5 μg), amoxicillin (10 μg), tetracycline (30 μg), and furazolidone (100 μg) (Oxoid, UK). The susceptibility and resistance of the strains were determined according to Table S1. H. pylori strain NCTC-11637 was used as the quality control standard strain. Among them, strains resistant to three or more antibiotics were defined as multidrug-resistant (MDR).  Table S2. The PCR amplification products were sequenced using Sanger sequencing. Point mutations and amino acid substitutions previously reported to be associated with antibiotic resistance were captured using Chroma 2.6.5 software (Technelysium Pty Ltd, QLD).

| Statistical analysis
Data were analyzed using SPSS (version 25.0). Differences in proportions were calculated using the chi-squared test or Fisher's exact test. p < 0.05 was considered a statistically significant difference.
Evaluation of agreement between antibiotic-resistant phenotypes and genotypes was conducted using Kappa coefficients. The figures were prepared using Origin® 2019 software (OriginLab Corp.).

| RE SULTS
We isolated 1955 H. pylori strains from 2719 participants who underwent antibiotic susceptibility testing for inclusion in this study.
The overall isolation and culture success rate was 72%. A flow chart of the study design is shown in Figure 2. Table 1 Table S3 shows the resistance of H. pylori to six antibiotics in various regions of Ningxia. Most of the strains showed high resistance to metronidazole, levofloxacin, and clarithromycin. The overall resistance rate to clarithromycin was 40.4%. The resistance rate to clarithromycin was lower in Guyuan (27.7%). The resistance rate to levofloxacin (42.4%) was higher than that to clarithromycin, and similar rates were observed in all regions. The resistance rate of metronidazole was alarmingly high (94.2%), reaching 100% in some areas.
However, the resistance to furazolidone, amoxicillin, and tetracycline was low, with the number of resistant strains being 17, 5, and 1, respectively. Of these strains, 82.6% (19/23) were secondary resis- however, this was observed in less than 1% of the strains. The rate of MDR also varied in different geographical regions ( Figure S2). The MDR rate of H. pylori in Wuzhong was less than 20%, whereas the MDR rates in Yinchuan, Zhongwei, and Shizuishan were more than 25%.
Factors affecting the resistance to clarithromycin, levofloxacin, and metronidazole are shown in Figure S3. The rate of resistance to clarithromycin (p < 0.001) and levofloxacin (p < 0.001) was significantly higher in women than in men. Strains isolated from patients without ulcers (gastritis, bile reflux, and polyps) showed higher resistance rates to clarithromycin (p = 0.001) and levofloxacin (p = 0.018) than those isolated from patients with the ulcerative disease. Sex However, the concordance between phenotypic and genetic resistance for amoxicillin, furazolidone, tetracycline, and metronidazole was not significant.

| DISCUSS ION
The incidence of gastric cancer in Ningxia is the highest in China. 21 H. pylori infection is a clear and controllable factor in the development of gastric cancer and thus, its eradication is one of the most effective methods of preventing the occurrence of gastric cancer.
However, the eradication rate of H. pylori infection has been declining over the past 20 years due to antibiotic resistance. 22 In this study, we reviewed the resistance of 1955 H. pylori strains to six commonly used antibiotics in five districts of Ningxia, a region with a high incidence of gastric cancer.
Over time, resistance to clarithromycin, levofloxacin, and metronidazole has increased significantly worldwide. 11,12,23,24 Additionally, national multicenter data monitored by the China Center for Helicobacter pylori Molecular Medicine for 2021 showed that the primary and secondary resistance rates of metronidazole, clarithromycin, and levofloxacin were 87.87% and 93.48%, 37.00% and 76.93%, 34.21%, and 61.58%, respectively. 25 In our study, the primary clarithromycin resistance rate was 26.6%, and the secondary resistance rate was 70.7% in Ningxia. The clarithromycin resistance rate was much higher than 15%, 10 suggesting that empirical triple therapy containing clarithromycin is no longer applicable in Ningxia.
Recently, the concept of H. pylori diagnosis and personalized treatment has become gradually accepted and recognized by clinicians F I G U R E 2 Flowchart depicting the study design. PPI, proton pump inhibitor.
worldwide. 10,26,27 The World Gastroenterology Organization's updated H. pylori guidelines for 2021 state that the main determinant of eradication success with a PPI-amoxicillin-clarithromycin regimen is the presence of clarithromycin resistance before treatment. 28 PCR assays can detect H. pylori resistance with high sensitivity and specificity. The use of such assays in areas with high resistance may greatly assist in the selection of treatment for individual patients. The primary resistance rate of clarithromycin in this study was 26.6%, which means that 73.4% of primary infected patients in Ningxia were sensitive to clarithromycin, indicating the majority of patients can continue with the clarithromycin regimen and reduce the blind use of antibiotics, such as tetracycline and furazolidone, to slow down the development of antibiotic resistance. Additionally, the Chinese expert consensus recommends levofloxacin as a secondline empirical antibiotic. 29 Other guidelines recommend levofloxacin as a salvage treatment option based on antibiotic susceptibility testing. 10,30,31 The observed resistance rate for levofloxacin (42.4%) exceeds the threshold (25%) for this antibiotic as rescue treatment or empirical regimens, 32 indicating that it is a poor choice for patients with H. pylori infections in Ningxia. Considering the high resistance rate to clarithromycin and levofloxacin in Ningxia, regimens containing clarithromycin and levofloxacin should not be used without prior antibiotic susceptibility testing. The primary resistance rate of metronidazole in this area was more than 90%, and the secondary resistance rate was as high as 100%, which is alarming. This means that any regimen containing metronidazole will give poor eradication treatment results.
Fortunately, the resistance rate of H. pylori to amoxicillin, tetracycline, and furazolidone was very low in Ningxia. This was in contrast to the high resistance rates in Wuwei (12.42%, 19.25%, and 20.5%, respectively), 25 which is also a region in northwest China.
A previous prospective study by our team showed that both a PPI + bismuth + amoxicillin + furazolidone quadruple 14-day regimen and a PPI + amoxicillin high-dose dual-therapy 14-day regimen had eradication rates greater than 90% in Ningxia. 33 Moreover, some studies have shown that resistance to amoxicillin, furazolidone, and tetracycline is not expected to develop despite the failure of previous eradication treatments with these three antibiotics. 34,35 Therefore, resistance to amoxicillin, tetracycline, and furazolidone is not a concern when choosing eradication therapy. However, the adverse reactions induced by allergies to amoxicillin, tetracycline, and furazolidone should be considered before establishing the treatment regimen.
Only 3.3% of the strains were sensitive to all six antibiotics, which was significantly lower than the 64.8% reported in Ningbo, China. 36 The rate of MDR (22.9%) was significantly higher than the previously reported global rate (9.6%) and the rate reported in Spain (14.2%). 37,38 However, it is consistent with the results reported in South Korea in 2019 and the national reports in China in 2021. 25,39 Additionally, our study revealed geographic differences in the antibiotic resistance and MDR of H. pylori in Ningxia. For example, the resistance rate to clarithromycin was approximately 40% in Yinchuan, Wuzhong, Zhongwei, and Shizuishan, but only 27.7% in Guyuan.
Amoxicillin-, furazolidone-, and tetracycline-resistant strains were mostly identified in Yinchuan. The investigation of antibiotic resistance patterns and MDR strains in different geographic regions is critical in developing locally-appropriate eradication treatment regimens.
There is considerable controversy regarding the relationship between sex, age, endoscopic presentation, pathological findings, and antibiotic resistance. The findings of Vilaichone et al. and Miftahussurur et al. showed no statistically significant relationship between antibiotic resistance and age, sex, and endoscopic findings. 40,41 Contrastingly, a Spanish study from a dyspeptic group showed that age was the only factor associated with metronidazole TA B L E 1 Basic information for patients with Helicobacter pylori infections. resistance. 38 Additionally, a study from Israel showed a correlation between sex and antibiotic resistance only in the Arab population; however, no difference was reported in the Jewish population. 42 In the present study, we found that antibiotic resistance of H. pylori was not associated with ethnicity or pathological in Ningxia. However, we found that infected women were more resistant to levofloxacin than infected men. This result was similar to previous studies 36, [42][43][44] and may be due to differences in the use of antibiotics between the sexes. For example, female patients are more likely to suffer from urinary tract infections and pelvic inflammatory disease due to their anatomy and therefore are more likely to develop resistance to quinolones and nitroimidazole antibiotics. 36 Furthermore, we found higher rates of resistance to levofloxacin in the ≥60 age group. This result is consistent with that of a multicenter study in China 43 and may be due to increased exposure to antibiotics with age. This reminds us to avoid levofloxacin for H. pylori eradication in women aged ≥60 years in our region Additionally, the ulcerative disease was a protective factor for clarithromycin and levofloxacin resistance. This is consistent with the results of two previous studies from China (Beijing), and Italy. 45,46 However, another study showed that the eradication rate of H. pylori was higher in patients with ulcerative disease than in those without ulcerative disease, 47 indicating that patients with the ulcerative disease may have a lower rate of antibiotic resistance. On one hand, this might be due to the virulence of H. pylori, as studies have shown that type I H. pylori infection is associated with ulcerative disease and has a higher eradication rate. [48][49][50] On the other hand, it may also be due to geographical differences.

Concordance analysis of phenotypic and genotypic resistance
revealed that the major mutation associated with clarithromycin resistance was the A2143G mutation in 23 S rRNA, and the mutation associated with levofloxacin resistance was a substitution at amino acids 87 and 91 of gyrA. in this region. However, this study had some limitations. First, this was a retrospective cross-sectional study with an inherent selection bias to assess trends in H. pylori resistance over time. Second, we used the K-B disk diffusion method to assess antibiotic susceptibility, which may not be sufficient to accurately reflect the actual level of antibiotic resistance. In addition, we only obtained endoscopic data for 61.1% (1058/2719) of subjects, which may be another limitation of this study. However, we have established several subcenters in the province of Ningxia for H. pylori infection, treatment, and antibiotic resistance detection systems, which will allow us to follow up with dynamic studies.
In conclusion, the increasingly prominent problem of antibiotic resistance urgently requires the identification of new and effective prevention and treatment options. Eradication therapy remains the most important option in the absence of an effective vaccine. The antibiotic resistance profile identified by this study will provide a basis for empirical H. pylori eradication protocols in Ningxia. We also demonstrated rapid and accurate molecular detection of clarithromycin and levofloxacin resistance. In the future, precise treatment based on the detection of molecular resistance to clarithromycin and levofloxacin will be the focus of our research.

ACK N OWLED G M ENTS
We would like to thank Editage (www.edita ge.com) for English language editing. We are also thankful for the support provided by the

CO N FLI C T O F I NTER E S T S TATEM ENT
None.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.