Abstract
A high rate of resistance (49.5 to 72.7%) to amoxicillin (AMX) was observed in Helicobacter pylori after two or three unsuccessful eradication attempts. Unsuccessful eradication regimens significantly increase resistance to not only clarithromycin (CLR) and metronidazole (MNZ) but also AMX.
TEXT
Currently available eradication regimens for Helicobacter pylori are triple-drug combination regimens comprising a proton pump inhibitor (PPI) and two antibiotic drugs, and clarithromycin (CLR), metronidazole (MNZ), and amoxicillin (AMX) are commonly used antibiotics (12). Although H. pylori bacteria easily become resistant to CLR and MNZ, H. pylori has been thought to seldom become resistant to AMX (6). In the present study, the resistance rates after unsuccessful eradication attempts were examined.
A total of 343 patients (189 males and 154 females; mean age, 55.8 years) with H. pylori infection were enrolled between September 2004 and October 2010. H. pylori infection was defined by H. pylori culture positivity. Of the total, 22 patients had no history of antibacterial therapy for eradication, 211 patients had one treatment failure, 99 patients had two treatment failures, and 11 patients had three treatment failures (first-line treatment, triple therapy with CLR [800 mg/day], AMX [1,500 mg/day], and PPI for 7 days; second-line treatment, triple therapy with MNZ [500 mg/day], AMX [1,500 mg/day], and PPI for 7 days; third-line treatment, triple therapy with fluoroquinolone [levofloxacin, 400 mg/day; gatifloxacin, 400 mg/day; or sitafloxacin, 400 mg/day], AMX [2,000 mg/day], and PPI for 7 days) (8, 13). All patients underwent esophagogastroduodenoscopy and gastric biopsy for bacterial culture 6 to 12 months after the eradication failure at Keio University Hospital and National Tokyo Medical Center.
Susceptibilities of H. pylori isolates to AMX, CLR, and MNZ were determined by the agar dilution method according to the guidelines established by the Clinical and Laboratory Standards Institute (CLSI) (1, 7). Isolates were considered resistant to MNZ if the MIC of the drug was ≥8 μg/ml and to CLR if the MIC was ≥1 μg/ml (9). For AMX, the interpretive standard (susceptible, ≤0.03 μg/ml) established by the Japanese Society of Chemotherapy was used (3). Isolates were defined as high-level resistant and resistant to AMX if the MIC was ≥0.5 μg/ml and ≥0.06 μg/ml, respectively (11), in this study. Differences between groups were compared by Fisher's exact test or the chi-squared test.
The rates of resistance to AMX in the groups with no history of eradication treatment, one treatment failure, two treatment failures, and three treatment failures were 13.6%, 26.5%, 49.5%, and 72.7%, respectively. The high-level rates of resistance to AMX in the group with no history of eradication treatment, one treatment failure, two treatment failures, and three treatment failures were 0%, 0.9%, 6.1%, and 18.2%, respectively (Table 1). The rates of resistance to AMX in the group with two treatment failures and that with three treatment failures were significantly higher than that in the group with no history of eradication treatment and that with one treatment failure. To the best of our knowledge, the present study is the first to report the increase in rates of resistance to AMX after unsuccessful H. pylori eradication.
Table 1.
Eradication failures and resistance rates
| Agent | Prior treatment | % resistance (no. of resistant strains/no. tested)a |
MIC of agent |
||||
|---|---|---|---|---|---|---|---|
| With AMX MIC (μg/ml) of: |
Other resistance | ||||||
| ≥0.06 | ≥0.5 | 50% | 90% | Range | |||
| AMX | None | 13.6 (3/22) | 0 (0/22) | <0.015 | 0.06 | <0.015–0.12 | |
| One failure | 26.5 (56/211) | 0.9 (2/211) | <0.015 | 0.12 | <0.015–0.5 | ||
| Two failures | 49.5 (49/99) ++ ### | 6.1 (6/99) # | 0.03 | 0.25 | <0.015–4 | ||
| Three failures | 72.7 (8/11) +++ ## | 18.2 (2/11) ## | 0.12 | 0.5 | <0.015–4 | ||
| CLR | No treatment | 9.1 (2/22) | 0.03 | 0.25 | <0.015–8 | ||
| One failure | 89.6 (189/211) +++ | 16 | 32 | <0.015–64 | |||
| Two failures | 88.8 (88/99) +++ | 16 | 32 | <0.015–64 | |||
| Three failures | 72.7 (8/11) +++ | 16 | 64 | <0.015–64 | |||
| MNZ | None | 13.6 (3/22) | 2 | 8 | 0.25–32 | ||
| One failure | 4.7 (10/211) | 1 | 2 | 0.5–32 | |||
| Two failures | 72.7 (72/99) +++ ### | 16 | 64 | 1–64 | |||
| Three failures | 72.7 (8/11) ++ ### | 16 | 32 | 4–32 | |||
AMX resistance, MIC ≥ 0.06 μg/ml; AMX high-level resistance, MIC ≥ 0.5 μg/ml; CLR resistance, MIC ≥ 1μg/ml; MNZ resistance, MIC ≥ 8 μg/ml. ++, P < 0.01 versus results for nontreatment group; +++, P < 0.001 versus results for nontreatment group; #, P < 0.05 versus results for one-failure group; ##, P < 0.01 versus results for one-failure group; ###, P < 0.001 versus results for one-failure group.
The MIC90 of AMX showed 2-fold increases with every eradication failure. The MIC90 of CLR showed a 128-fold increase after triple therapy with CLR, AMX, and PPI, and the MIC90 of MNZ showed a 32-fold increase after triple therapy with MNZ, AMX, and PPI (Table 1). While the 23S rRNA point mutation is a main cause of CLR resistance (4) and the single mutation of rdxA or frxA is one of the main causes of MNZ resistance (5), multiple mutations in penicillin binding protein 1 (PBP1) would contribute to a greater increase in the level of AMX resistance (11) and then could result in a gradual increase in AMX resistance.
We amplified the bacterial DNA by PCR and sequenced the PBP1 genes in 30 strains between September 2008 and April 2010 (forward, 5′-CACRAGCACCGGTAAGATTT-3′; reverse, 5′-GCGACAATAAGAGTGGCA-3′). The sequences obtained were compared with the published sequences of H. pylori PBP1 (L26695; GenBank accession number AE000511). Table 2 shows the substitutions detected in AMX-resistant strains. Strains with high-level resistance to AMX had 1 to 3 substitutions, and low-level-resistant strains (MICs of 0.06 to 0.25 μg/ml) had 0 to 2 substitutions. The accumulation of PBP1 mutations could result in a gradual increase in AMX resistance. The Asn535 → Asp substitution was also detected in not only AMX-resistant strains but also 3 of 15 (20%) AMX-susceptible strains.
Table 2.
Substitutions in penicillin binding protein 1 of H. pylori strains
| Strain or group (na) | AMX MIC (μg/ml) | Substitution at penicillin binding protein 1 position: |
||||||
|---|---|---|---|---|---|---|---|---|
| 374 | 406 | 414 | 535 | 593 | 599 | 601 | ||
| KS0461 | 4 | Asp | Gly | Gly | ||||
| KS0478 | 0.5 | Asp | ||||||
| KS0487 | 0.5 | Leu | Arg | |||||
| KS0439 | 0.25 | Ala | Ala | |||||
| KS0476 | 0.25 | Asp | Gly | |||||
| KS0470 | 0.25 | Asp | ||||||
| KS0444 | 0.12 | Ala | Ala | |||||
| KS0464, KS0479 | 0.12 | Asp | ||||||
| KS0493 | 0.12 | Pro | ||||||
| KS0434 | 0.06 | Ala | ||||||
| KS0466, KS0491 | 0.06 | Asp | ||||||
| KS0503 | 0.06 | Arg | ||||||
| KS0502 | 0.06 | |||||||
| AMX-susceptible strains (15) | ≤0.03 | Val | Glu | Ser | Asn/Aspb | Thr | Ala | Val |
n, no. of strains.
535Asn → Asp was detected in the amoxicillin-susceptible strains KS0447, KS0452, and KS0467.
The AMX resistance rates were 13.6% (6/45) in the strains susceptible to both CLR and MNZ, 32.2% (66/205) in the strains resistant to CLR but susceptible to MNZ, 45.5% (5/11) in the strains resistant to MNZ but susceptible to CLR, and 48.8% (40/82) in the strains resistant to both CLR and MNZ. The AMX resistance rate in the strains resistant to CLR or MNZ was significantly higher than that in the strains susceptible to both CLR and MNZ. The rate of resistance to AMX in the strains resistant to both CLR and MNZ was significantly higher than that in the strains susceptible to MNZ (Table 3). Efflux pump systems in bacteria, which can eject drugs and toxic compounds, including antibiotics, have a critical role in the development of multidrug resistance. We recently reported that the expression of the TolC efflux pump (hefA) was significantly increased under MNZ exposure (14). The efflux pump of H. pylori is also associated with the development of resistance to CLR, in addition to 23S rRNA point mutations (2). In addition to the known mutations in the gene coding for PBP, activated efflux systems may also play a role in H. pylori resistance to AMX.
Table 3.
Amoxicillin resistance rate and susceptibility to clarithromycin and metronidazole
| Resistance | % amoxicillin resistance (no. of resistant strains/no. tested)a | MIC |
|
|---|---|---|---|
| 50% | 90% | ||
| CLR susceptible, MNZ susceptible | 13.6 (6/45) | <0.015 | 0.06 |
| CLR resistant, MNZ susceptible | 32.2 (66/205) + | 0.03 | 0.12 |
| CLR susceptible, MNZ resistant | 45.5 (5/11) + | 0.03 | 0.5 |
| CLR resistant, MNZ resistant | 48.8 (40/82) # +++ | 0.03 | 0.25 |
Amoxicillin resistance: MIC ≥ 0.06 μg/ml. +, P < 0.05 versus results for CLR-susceptible, MNZ-susceptible group; +++, P < 0.001 versus results for CLR-susceptible, MNZ-susceptible group; #, P < 0.05 versus results for CLR-resistant, MNZ-susceptible group.
In conclusion, contrary to our expectations, resistance to AMX in H. pylori was gradually induced after unsuccessful eradication attempts. The data are clearly consistent with the association of resistance rates and eradication failures. If AMX-resistant H. pylori strains were to spread further, serious problems would arise, resulting in increasing eradication failures (10). Our results suggest that clinicians should be aware of AMX resistance together with resistance to other antibiotics in the future.
Footnotes
Published ahead of print on 12 April 2011.
REFERENCES
- 1. Clinical and Laboratory Standards Institute 2010. Performance standards for antimicrobial susceptibility testing: twentieth informational supplement, M100-S20. CLSI, Wayne, PA [Google Scholar]
- 2. Hirata K., et al. 2010. Contribution of efflux pumps to clarithromycin resistance in Helicobacter pylori. J. Gastroenterol. Hepatol. 25(Suppl. 1):S75–S79 [DOI] [PubMed] [Google Scholar]
- 3. Japan Society for Chemotherapy 2000. Antimicrobial Susceptibility Subcommittee on Helicobacter pylori MIC breakpoints for clarithromycin and amoxicillin. Jpn. Soc. Chemother. 48:561–567 [Google Scholar]
- 4. Masaoka T., Suzuki H., Kurabayashi K., Kamiya A. G., Ishii H. 2004. Second-line treatment of Helicobacter pylori infection after dilution agar methods and PCR-RFLP analysis. Aliment. Pharmacol. Ther. 20(Suppl. 1):68–73 [DOI] [PubMed] [Google Scholar]
- 5. Masaoka T., et al. 2006. Could frameshift mutations in the frxA and rdxA genes of Helicobacter pylori be a marker for metronidazole resistance? Aliment. Pharmacol. Ther. 24(Suppl. 4):81–8716803605 [Google Scholar]
- 6. Nishizawa T., Suzuki H., Hibi T. 2009. Quinolone-based third-line therapy for Helicobacter pylori eradication. J. Clin. Biochem. Nutr. 44:119–124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Nishizawa T., et al. 2007. Rapid detection of point mutations conferring resistance to fluoroquinolone in gyrA of Helicobacter pylori by allele-specific PCR. J. Clin. Microbiol. 45:303–305 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Nishizawa T., et al. 2008. Gatifloxacin-based triple therapy as third-line regimen for Helicobacter pylori eradication. J. Gastroenterol. Hepatol. 23(suppl. 2):167–170 [DOI] [PubMed] [Google Scholar]
- 9. Nishizawa T., et al. 2006. Gatifloxacin resistance and mutations in gyrA after unsuccessful Helicobacter pylori eradication in Japan. Antimicrob. Agents Chemother. 50:1538–1540 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Nishizawa T., Suzuki H., Masaoka T., Iwasaki E., Hibi T. 2007. A new eradication resistance index as a predictor of metronidazole-containing second-line treatment of Helicobacter pylori. Digestion 76:215–220 [DOI] [PubMed] [Google Scholar]
- 11. Rimbara E., Noguchi N., Kawai T., Sasatsu M. 2008. Mutations in penicillin-binding proteins 1, 2 and 3 are responsible for amoxicillin resistance in Helicobacter pylori. J. Antimicrob. Chemother. 61:995–998 [DOI] [PubMed] [Google Scholar]
- 12. Suzuki H., Nishizawa T., Hibi T. 2010. Helicobacter pylori eradication therapy. Future Microbiol. 5:639–648 [DOI] [PubMed] [Google Scholar]
- 13. Suzuki H., Nishizawa T., Muraoka H., Hibi T. 2009. Sitafloxacin and garenoxacin may overcome the antibiotic resistance of Helicobacter pylori with gyrA mutation. Antimicrob. Agents Chemother. 53:1720–1721 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Tsugawa H., et al. 2011. Enhanced bacterial efflux system is the first step to the development of metronidazole resistance in Helicobacter pylori. Biochem. Biophys. Res. Commun. 404:656–660 [DOI] [PubMed] [Google Scholar]