Frequentist rules for regulatory approval of subgroups in phase III trials: A fresh look at an old problem.

BACKGROUND: The number of Phase III trials that include a biomarker in design and analysis has increased due to interest in personalised medicine. For genetic mutations and other predictive biomarkers, the trial sample comprises two subgroups, one of which, say B+ is known or suspected to achieve a larger treatment effect than the other B-. Despite treatment effect heterogeneity, trials often draw patients from both subgroups, since the lower responding B- subgroup may also gain benefit from the intervention. In this case, regulators/commissioners must decide what constitutes sufficient evidence to approve the drug in the B- population. METHODS AND RESULTS: Assuming trial analysis can be completed using generalised linear models, we define and evaluate three frequentist decision rules for approval. For rule one, the significance of the average treatment effect in B- should exceed a pre-defined minimum value, say ZB->L. For rule two, the data from the low-responding group B- should increase statistical significance. For rule three, the subgroup-treatment interaction should be non-significant, using type I error chosen to ensure that estimated difference between the two subgroup effects is acceptable. Rules are evaluated based on conditional power, given that there is an overall significant treatment effect. We show how different rules perform according to the distribution of patients across the two subgroups and when analyses include additional (stratification) covariates in the analysis, thereby conferring correlation between subgroup effects. CONCLUSIONS: When additional conditions are required for approval of a new treatment in a lower response subgroup, easily applied rules based on minimum effect sizes and relaxed interaction tests are available. Choice of rule is influenced by the proportion of patients sampled from the two subgroups but less so by the correlation between subgroup effects.

A recycling framework for the construction of Bonferroni-based multiple tests.

In this paper we describe Bonferroni-based multiple testing procedures (MTPs) as strategies to split and recycle test mass. Here, 'test mass' refers to (parts of) the nominal level alpha at which the family-wise error rate is controlled. Briefly, test mass is split between different null hypotheses, and whenever a null hypothesis is rejected, the part of alpha allocated to it may be recycled to the testing of other hypotheses. These recycling MTPs are closed testing procedures based on raw p-values associated with testing the individual null hypotheses, and the class of such MTPs includes, for example, serial and parallel gatekeeping, fallback and Holm procedures. Graphical displays and a concise algebraic notation are provided for such MTPs. This recycling approach has pedagogical advantages and may facilitate the tailoring of MTPs for different purposes.

Impact of Helicobacter pylori eradication on heartburn in patients with gastric or duodenal ulcer disease -- results from a randomized trial programme.

BACKGROUND: Helicobacter pylori infection has been proposed as a protective factor against the development of gastro-oesophageal reflux disease. AIM: To study heartburn and endoscopic findings before and after H. pylori eradication therapy in patients with peptic ulcer disease. METHODS: In a multicentre trial programme, patients (n = 1497) were randomized to the omeprazole triple therapy group or to the control group, and were followed for 1-6 months after treatment. Patients in whom the infection was eradicated were compared with those in whom infection persisted. The severity of heartburn was measured at baseline and at each return visit. Endoscopy was performed 6 months after therapy in two of the five studies. RESULTS: In patients with duodenal ulcer, there was a significantly lower prevalence of heartburn after successful eradication of H. pylori relative to that after failed eradication (estimated odds ratio, 0.48). The reduction in the prevalence of heartburn in patients with gastric ulcer was independent of the post-treatment H. pylori status. In studies in which ulcer relapse was included in the model, this factor emerged as a significant factor for heartburn. The observed incidence of oesophagitis at the last visit was not influenced by H. pylori status. CONCLUSIONS: Eradication of H. pylori in patients with peptic ulcer disease was associated with a reduced prevalence of heartburn. Prevention of ulcer relapse could be the true cause of this reduction.

The HOMER Study: the effect of increasing the dose of metronidazole when given with omeprazole and amoxicillin to cure Helicobacter pylori infection.

BACKGROUND: Helicobacter pylori eradication with omeprazole, amoxycillin, and metronidazole is both effective and inexpensive. However, eradication rates with different dosages and dosing vary, and data on the impact of resistance are sparse. In this study, three different dosages of omeprazole, amoxycillin, and metronidazole were compared, and the influence of metronidazole resistance on eradication was assessed. METHODS: Patients (n = 394) with a positive H. pylori screening test result and endoscopy-proven duodenal ulcer in the past were enrolled into a multicenter study performed in four European countries and Canada. After baseline endoscopy, patients were randomly assigned to treatment for 1 week with either omeprazole, 20 mg twice daily, plus amoxycillin, 1,000 mg twice daily, plus metronidazole, 400 mg twice daily (low M); or omeprazole, 40 mg once daily, plus amoxycillin, 500 mg three times daily, plus metronidazole, 400 mg three times daily (medium M); or omeprazole, 20 mg twice daily, plus amoxycillin, 1,000 mg twice daily, plus metronidazole, 800 mg twice daily (high M). H. pylori status at entry was assessed by a 13C urea breath test and a culture. Eradication was defined as two negative 13C-urea breath test results 4 and 8 weeks after therapy. Susceptibility testing using the agar dilution method was performed at entry and in patients with persistent infection after therapy. RESULTS: The eradication rates, in terms of intention to treat (ITT) (population n = 379) (and 95% confidence interval [CI]) were as follows: low M 76% (68%, 84%), medium M 76% (68%, 84%), and high M 83% (75%, 89%). By per-protocol analysis (population n = 348), the corresponding eradication rates were: low M 81%, medium M 80%, and high M 85%. No H. pylori strains were found to be resistant to amoxycillin. Prestudy resistance of H. pylori strains to metronidazole was found in 72 of 348 (21%) of the cultures at entry (range, 10%-39% in the five countries). The overall eradication rate in prestudy metronidazole-susceptible strains was 232 of 266 (87%) and, for resistant strains, it was 41 of 70 (57%; p <.001). Within each group, the results were as follows (susceptible/resistant): low M, 85%/54%; medium M, 86%/50%; and high M, 90%/75%. There were no statistically significant differences among the treatment groups. 23 strains susceptible to metronidazole before treatment were recultured after therapy failed; 20 of these had now developed resistance. CONCLUSIONS: H. pylori eradication rates were similar (approximately 80%) with all three regimens. Metronidazole resistance reduced efficacy; increasing the dose of metronidazole appeared not to overcome the problem or significantly improve the outcome. Treatment failure was generally associated with either prestudy or acquired metronidazole resistance. These findings are of importance when attempting H. pylori eradication in communities with high levels of metronidazole resistance.

Efficacy of two different dosage regimens of omeprazole, amoxycillin and metronidazole for the cure of Helicobacter pylori infection.

BACKGROUND: While addition of metronidazole to the omeprazole-amoxycillin combination has been shown to be advantageous, the optimal dosage and drug distribution of the antimicrobials has not been sufficiently evaluated. AIM: To investigate the efficacy of two different regimens of omeprazole, amoxycillin and metronidazole for the cure of Helicobacter pylori infection. METHODS: Two hundred and fifty-five patients with H. pylori associated duodenal ulcers were randomly treated with either a 1-week regimen of omeprazole 20 mg b.d., amoxycillin 1000 mg b.d. and metronidazole 800 mg b.d. (OAM b.d.) or a combination of omeprazole 40 mg o.d., amoxycillin 500 mg t.d.s. and metronidazole 400 mg t.d.s. (OAM t.d.s.). All patients subsequently received omeprazole 20 mg o.d. for an additional 3 weeks. H. pylori status was assessed by histology and 13C-UBT prior to treatment and 8 weeks after randomization. Additional biopsies were obtained for H. pylori culture to determine primary and secondary resistance to metronidazole by agar dilution. RESULTS: Two hundred and thirty-seven patients were included in the intention-to-treat analysis and 198 patients in the per protocol analysis. With intention-to-treat analysis, the cure rate was 77% after treatment with OAM b.d. (95% CI, 69%-85%) and 76% after OAM t. d.s. therapy (95% CI, 67%-83%). Ulcer healing (intention-to-treat analysis) was documented in 95% of patients in the OAM b.d. group (n=122) and in 97% of patients in the OAM t.d.s. group (n=115). Adverse events were reported in 26 (20%) and in 18 (14%) patients in the OAM b.d. and OAM t.d.s. groups, respectively. None resulted in discontinuation of treatment. Overall primary resistance of H. pylori against metronidazole was found in 22 of 116 strains (19%). CONCLUSIONS: The combination of omeprazole, amoxycillin and metronidazole achieves about an 80% cure rate of H. pylori infection even in active ulcers. The total daily dose, and the choice of twice or three times daily dosing does not seem critical with this regimen.

Antimicrobial susceptibility testing of Helicobacter pylori in a large multicenter trial: the MACH 2 study.

Culture and susceptibility testing of Helicobacter pylori strains was performed in a large multinational, multicenter randomized clinical trial. Culture was carried out on gastric biopsy samples obtained from 516 patients at entry and had a sensitivity of 99% when the [(13)C]urea breath test was used as a reference. Susceptibility testing was performed for clarithromycin and metronidazole on 485 strains by an agar dilution method and the epsilometer test (Etest) and for amoxicillin by an agar dilution method only. Resistance to clarithromycin (>1 microgram/ml) was found in 3% of the H. pylori strains, with a perfect correlation between Etest and agar dilution methods. Resistance to metronidazole (>8 microliter/ml) was found in 27% of the strains by agar dilution, but there were important discrepancies between it and the Etest method. No resistance to amoxicillin was found. The logarithms of the MICs of the three antibiotics against susceptible strains had a distribution close to normal. The impact of resistance was tested in the four arms of the trial. There were not enough clarithromycin-resistant strains to evaluate the impact of resistance on the cure rate of clarithromycin-based regimens. For metronidazole-resistant strains, the impact noted in the clarithromycin-metronidazole arm was partially overcome when omeprazole was added (76% eradication for resistant strains versus 95% for susceptible strains). Secondary resistance to clarithromycin occurred in strains from 12 of 105 patients (11.4%) after the failure of a clarithromycin-based regimen to effect eradication. The detection of point mutations in clarithromycin-resistant strains was performed by a combination of PCR and restriction fragment length polymorphism. Mutations (A2142G and 2143G) were found in all strains tested except one. This study stresses the importance of performing susceptibility tests in clinical trials in order to explain the results of different treatments.