Increasing role for modified bismuth-containing quadruple therapies for Helicobacter pylori eradication

Eradication rates for Helicobacter pylori (H. pylori) infection achieved with standard clarithromycin-based triple regimen have generally fallen to unacceptable levels world-wide.^{1, 2} As with other infectious diseases, the most appropriate treatment should be identified on the basis of susceptibility obtained by testing the infecting organism or empirically based on local antibiotic usage, documented prevalence of antibiotic resistance, and local treatment outcome data.³ The best, most rational, and successful option after failure of empiric therapy is tailored therapy, that is, to test susceptibility before prescribing drugs.⁴ If the strain is susceptible to the original regimen, a reason other than resistance must be explored, especially adherence. If susceptibility testing is unavailable, the best locally proven successful alternative should be used. Fluoroquinolones (levofloxacin), macrolides (clarithromycin) and metronidazole exhibit cross resistance to other drugs within the same class. With fluoroquinolones and macrolides resistance is all-or-none meaning it cannot be overcome by increasing the dose, or duration of therapy. In most regions macrolide and fluoroquinolone resistance has increased to the point where they are no longer considered acceptable choices for empiric triple therapy. Neither should be used empirically for patients who have received a similar drug previously (e.g., patients with chronic infectious bronchopneumopathy who received fluoroquinolones, or with diarrhea who received ciprofloxacin).⁵

A recent open-label, single-arm study, conducted in Croatia by Marušic et al. evaluated the efficacy of a modified bismuth-containing quadruple therapy consisting of colloidal bismuth subcitrate 240 mg b.i.d., pantoprazole 40 mg b.i.d, metronidazole 500 mg b.i.d. and moxifloxacin 400 mg once daily for 14 days in patients who failed eradication with standard triple therapy.⁶ Moxifloxacin is a fourth-generation synthetic fluoroquinolone with similar efficacy as levofloxacin for H. pylori eradication.⁷ H. pylori eradication rates were 80.6% intention-to-treat (ITT) and 88% per-protocol (PP).⁶ Their experience suggested that this modified bismuth-containing quadruple regimen might prove useful and is as safe as non-bismuth-containing therapies.

Bismuth-containing quadruple regimens, consisting of a proton pump inhibitor (PPI), a bismuth, metronidazole, and tetracycline, are now recommended as first line and second line therapies in regions where clarithromycin resistance has resulted in low-cure rates. The Maastricht V and the Toronto Consensus conferences recommend 14 days’ full dose of bismuth-containing quadruple therapy.⁸, ⁹ This regimen is becoming available in Europe as a proprietary combination therapy but is packaged for 10 days (Pylera^®). The optimum formulation remains unknown and in many areas of the world tetracycline HCl is still unavailable. A number of substitutions have been published such as substitution of amoxicillin for tetracycline or for metronidazole, substitution of a fluoroquinolone for tetracycline (as in Marušic et al.⁶), and reducing the frequency of bismuth and or tetracycline administration to b.i.d. ^10–13 High-cure rates have been reported despite high rates of resistance in part because metronidazole resistance can be overcome by increasing the dose to 1500 or 1600 mg/day and extending the duration to 14 days.^10–13 Fluoroquinolone resistance cannot be overcome and in this therapy fluoroquinolone resistance results in a triple therapy consisting of a PPI, 1000 mg of metronidazole, plus bismuth. The regimen is most similar to a study from China, in which the regimen included a PPI, amoxicillin, bismuth and levofloxacin, achieving 97% cure rates with fluoroquinolone-susceptible strains and approximately 70% with fluoroqui-nolone-resistant strains.¹⁴ If similar outcomes were present in the Marušic et al. study ⁶ a cure rate PP of 88% would be present with a fluoroquinolone resistance rate of slightly more thn 35% which is within the range the authors suspected (Figure 1).¹⁵ However, the results with this metronidazole-bismuth-PPI combination could well be markedly different than with the amoxicillin-bismuth combination. Figure 1 also shows one of innumerable alternatives. The lack of susceptibility data allows only speculation about resistance rates, cure rates in the presence or absence of fluoroquinolone or metronidazole resistance, and outcome in any other population or even in another sample from the same population.¹⁵

Figure 1.—

Hp-normogram showing the cure rate with susceptible infections on the left vertical axis and those with resistant infections on the right vertical axis.¹⁵ The proportion with resistance is shown on the horizontal axis. The cure rates are connected with a line allowing one to visualize the population cure rates for any prevalence of resistance. The cure rate (solid line) was taken from a somewhat similar Chinese study as describe in the text.¹⁴ The cure rates with the current regimen (dotted line) are based on arbitrarily chosen cure rates with susceptible and resistant organisms. The dashed line shows the prevalence of fluoroquinolone resistance that would achieve an 88% cure rate in each population.

It should be evident that it is critical for clinical trials to report treatment-specific susceptibility results. Only such data allows one to understand the outcome of the therapy in relation to antimicrobial resistance and thus be generalizable. It is time for all studies to at a minimum collect and store biopsies for susceptibility testing which if unavailable locally can be done elsewhere. The results of the Marušic et al, while interesting and suggestive, are population-specific and are not generalizable to other populations. We predict that studies without susceptibility data will soon no longer be publishable.