Postgraduate Medicine: Volume 121: No.4

Results of 2 Randomized, Placebo-Controlled, Double-Blind Studies

Pamela R. Kushner, MA, MD; Andrew M. Snoddy, PhD; Larry Gilderman, DO; and David A. Peura, MD

Abstract: Objective To investigate the efficacy and safety of a 14-day treatment period with lansoprazole 15 mg for frequent heartburn in patients who are likely to select a nonprescription medication before consulting a prescriber. Methods Adults with untreated frequent heartburn ≥ 2 days a week over the past month were recruited for 2 identical multicenter, double-blind studies conducted with a 1-week screening and heartburn medication washout, a 1-week placebo run-in, a 2-week placebo-controlled treatment, and a 1-week placebo follow-up. After the washout and placebo run-in, subjects were randomly assigned to receive lansoprazole 15 mg or placebo once daily for 14 days in a double-blind fashion. Antacid tablets were permitted as rescue medication. Endpoints included percentage of 24-hour days without heartburn (primary), percentage of night-times without heartburn, and percentage of subjects without heartburn during day 1 of treatment (secondary endpoints). Data were collected daily via an interactive voice response system. Results In studies 1 and 2, 282 and 288 subjects, respectively, were randomly assigned to lansoprazole, and 282 in each study received placebo. The mean percentage of days without heartburn was greater among lansoprazole recipients compared with placebo recipients (P < 0.0001). Significantly more subjects treated with lansoprazole also reported no night-time heartburn and no heartburn during day 1 of the 14-day treatment. Adverse events were infrequent and were similar for lansoprazole and placebo groups. Conclusion During the 14-day treatment period in a population with frequent heartburn who were likely to select a medication without consulting a prescriber, lansoprazole 15 mg once daily showed rapid and sustained effectiveness throughout a 24-hour period and was well tolerated.

Keywords: lansoprazole; frequent heartburn; self-selection; night-time heartburn; nonprescription; placebo controlled

Heartburn is defined as an upward-moving, uncomfortable burning or painful retrosternal sensation.¹ Frequent heartburn (ie, mild symptoms occurring ≥ 2 days a week) or moderate-to-severe symptoms occurring > 1 day a week is considered troublesome by most patients.¹ An estimated 22% of adults in the United States reported heartburn occurring at least once per month during a 3-month period in a telephone survey involving more than 21 000 adults.² Overeating, stress, smoking, pregnancy, and consumption of certain foods and beverages, such as spicy or greasy foods, coffee, alcohol, chocolate, peppermint, tomato products, fatty foods, citrus, and carbonated drinks are commonly associated with heartburn.^3,4 Frequent heartburn has a significant negative impact on quality of life, particularly if it affects quality of sleep. In a nationwide telephone survey conducted by the American Gastroenterological Association, > 60% of respondents with heartburn reported sleep disturbance, which in turn impaired their ability to function the following day.⁵

The majority of patients with frequent heartburn take nonprescription medications to control symptoms, but often find these medications less than satisfactory in terms of meeting their expectations of efficacy.⁵ Treatments for heartburn include antacids, histamine₂-receptor antagonists (H₂RAs), antacid and H₂RA combinations, and proton pump inhibitors (PPIs). The efficacy and safety of the PPI lansoprazole (Prevacid^®) has been well documented for the symptomatic treatment of both daytime and night-time heartburn,^6,7 and it would appear to be an appropriate self-care therapy for frequent heartburn. However, these studies were conducted in patients with confirmed gastroesophageal reflux disease (GERD). When medication is first considered by subjects with frequent heartburn, they are unlikely to have consulted with a physician and would not have a confirmed diagnosis. Furthermore, no studies of lansoprazole in patients who are interested in selecting nonprescription medications without consulting a prescriber have been reported in the literature to date. Therefore, the objective of the 2 randomized studies reported was to investigate the efficacy and tolerability of lansoprazole 15 mg in a 14-day regimen for the treatment of frequent heartburn in patients who are likely to choose nonprescription medications.

The planned sample size for each trial was 576 subjects, with 288 in each group. The study population comprised male and female subjects ≥ 18 years old with frequent heartburn occurring ≥ 2 days per week over the past month. Subjects were recruited for each study from clinical research organizations and site databases containing subjects with known heartburn. If medication had been used, the heartburn was to have been responsive to antacids, nonprescription H₂RAs, or a short-term nonprescription or prescription PPI. Patients had to discontinue use of nonprescription H₂RA and PPI treatment before the start of the run-in phase and had to experience at least two 24-hour days with heartburn during the run-in. Only nonpregnant and nonlactating women were included, and women of child-bearing potential were required to use an acceptable method of contraception. Patients were excluded if they had a history of endoscopically proven erosive esophagitis or physician-diagnosed GERD confirmed by endoscopy; a history of pathologic intraesophageal pH monitoring; an underlying medical condition or concomitant medication that might interfere with the evaluation of heartburn treatment or might constitute a safety concern; clinically significant and/or unstable renal or hepatic disease; a requirement for continuous treatment with H₂RAs or PPIs, or with gastric prokinetic drugs; required antacids for any other indications throughout the study; and/or had a requirement for continuous treatment with a prescription antifungal or warfarin. Prior use of antisecretory agents was permitted if they were discontinued after provision of informed consent.

The 2 identical multicenter, randomized, double-blind, placebo-controlled trials (clinical trials.gov: study 1 = NCT00389948; study 2 = NCT00390390) included a 1-week screening and heartburn medication washout period, a 1-week placebo run-in to confirm the frequency of heartburn, a 2-week placebo-controlled treatment phase, and a 1-week placebo follow-up. Subjects in each study were enrolled at 1 of 34 locations in the United States. The washout period was ≥ 1 day for antacids or ≥ 7 days for an H₂RA or a PPI. After the washout and placebo run-in phases, subjects were randomly assigned to receive lansoprazole 15 mg or matching placebo once daily for 14 days. This phase was followed by administration of placebo for 7 days. Lansoprazole delayed-release 15 mg capsules or identically appearing placebo capsules were packaged in blister strips. The capsule was to be swallowed whole with a glass of water once a day immediately after waking and before breakfast. Commercially available Gelusil^® tablets were permitted as rescue medication (≤ 12 a day), but subjects were encouraged to minimize use to allow the effects of the study medication to be fully assessed. Use of the study and rescue medication was recorded in daily diaries and assessed by the investigator at each clinic visit. Treatment compliance was assessed by returned capsule count at each visit. The study design was approved by Investigational Review Boards for the New York University School of Medicine (study 1) and by Quorum Review, Inc. (study 1 and study 2), following the principles of the Declaration of Helsinki, and included provision of written informed consent by the subjects.

The primary efficacy endpoint was percentage of heartburn-free 24-hour days during 14 days of double-blind treatment. Secondary endpoints were percentage of night-times (time between going to bed and the next dose of study medication, which was taken before eating) without heartburn during 14 days of treatment and percentage of subjects without heartburn on day 1. The 2 treatment groups were also compared on 5 subject global assessment questions to assess treatment satisfaction. The questions were: 1) overall, how satisfied have you been with the study medication you received for your heartburn during the last 2 weeks? (6 possible answers, from extremely satisfied to not at all satisfied); 2) how much have you benefited from the study medication you received for your heartburn during the past 2 weeks? (4 possible answers, from not at all to a lot); 3) would you recommend the study medication you received for your heartburn to someone else with your condition? (Yes, no, unsure); 4) given your experience with other medications for your night-time heartburn, would you say your study medication provided relief? (5 levels from much better to much less); 5) given your experience with other medications for your daytime heartburn, would you say your study medication provided relief? (5 answers, from much better to much less) Most measures of efficacy were assessed from data recorded by the subjects in an interactive voice response system diary after daily self-assessment. The diary was designed to collect information on the occurrence of daytime and night-time heartburn episodes, maximum heartburn severity, and study and rescue medication consumption. Heartburn was defined for the subjects as an upward-moving, uncomfortable sensation behind the breastbone, frequently accompanied by a burning feeling. Heartburn severity was rated on a scale of 0 to 3 (0 = no heartburn; 1 = mild: heartburn present but easily tolerated; 2 = moderate: heartburn sufficient to cause interference with normal daily activities or sleep; and 3 = severe: incapacitating heartburn, with the subject unable to perform normal daily activities or sleep).

Safety assessments included monitoring and recording of all adverse events (AEs), including serious events. Screening assessments included hematology, blood chemistry, vital signs, electrocardiogram, and physical condition, and a urine pregnancy test for women of child-bearing potential. All safety analyses were conducted using the safety analysis population.

The intent-to-treat (ITT) population included all randomized subjects who took ≥ 1 dose of the study medication and had ≥ 1 post-baseline efficacy assessment. The per-protocol set included a subset of subjects in the ITT population without any major protocol violation. The safety analysis population included all subjects who took ≥ 1 dose of double-blind study medication, whether or not the subject was randomly assigned. Primary and secondary variables were analyzed sequentially.

Statistical tests used to compare the 2 treatment groups were 2-sided, with a probability of type I error of 0.05. For the primary efficacy variable, the percentage of 24-hour days between day 1 and day 14 with no heartburn (95% confidence limits on the mean, and the median, minimum, and maximum values for each treatment) were calculated using the ITT population. The 2 treatment groups were compared by analysis of covariance of the percentage of days with no heartburn, including the treatment and center as factors and percentage of days in which heartburn was experienced during the run-in phase as a covariate. The percentage of night-times with no heartburn between day 1 and day 14 was analyzed similarly. The number and percentage of subjects with no heartburn on day 1 were calculated for each treatment group and analyzed using the Cochran-Mantel-Haenszel test, stratified by each center.

The 2 treatment groups were also compared for individual global assessment questions by using a Cochran-Mantel-Haenszel test with table scores stratified by each center. Safety analysis was conducted using the safety population, based on the frequency, duration, severity, and outcome of treatment-emergent AEs. Adverse events were summarized by the number and percentage of subjects, individual AEs, and severity.

After completion of the run-in phase, 1134 subjects were randomly assigned to receive lansoprazole (n = 570) or placebo (n = 564) and were included in the ITT data set for statistical analysis. The disposition of the subjects in the 2 studies is shown (Table 1), and the baseline demographic and heartburn characteristics are summarized (Table 2). There were no statistically significant differences in demographic parameters or heartburn characteristics between treatment groups in the 2 studies.

At baseline, heartburn moderately to greatly affected the ability to eat or drink what was wanted (about 75% of subjects in each study) and affected sleep (> 60% and 58% of subjects in study 1 and study 2, respectively); half of the subjects in each study elevated their heads for sleeping. Thirty-five percent (study 1) and 29% (study 2) of subjects could not sleep through the night, approximately 33% and 31%, respectively, took naps when possible, and 22% and 23% slept in a chair or in a sitting position. In study 2, subjects in the placebo group were more likely to have heartburn triggered by physical activity (23%) than those in the lansoprazole group (17%).

Compliance remained high during the 14-day double-blind phase of both studies. In study 1, 88% and 90% of lansoprazole and placebo recipients, respectively, took ≥ 13 of the 14 tablets dispensed, and in study 2, 90% and 89% of subjects, respectively, were similarly compliant. Use of rescue medication was considerably lower in the lansoprazole group than in the placebo group in each study (Table 3).

In the double-blind treatment phase of the 2 studies, subjects who received lansoprazole were heartburn-free on 60% of 24-hour days compared with approximately 45% of 24-hour days for subjects who received placebo. The difference between lansoprazole and placebo in mean percentage of 24-hour days with no heartburn was 14.2% in study 1 and 19.7% in study 2, respectively (both P < 0.0001) (Figure 1).

Lansoprazole was effective in increasing the mean percentage of night-times with no heartburn compared with placebo. In the lansoprazole group, 79.5% of nights in study 1 and 81.6% of nights in study 2 were heartburn-free. In the placebo group, the values were 76.3% and 77.0%, respectively. The treatment differences were statistically significant (P = 0.003 in study 1; P = 0.0001 in study 2) (Figure 2).

Lansoprazole rapidly reduced the frequency of heartburn in both studies. Half of the subjects in the lansoprazole group had no heartburn during day 1 of double-blind treatment compared with approximately one-third of the subjects in the placebo group. The treatment difference between lansoprazole and placebo was 17% (P < 0.0001) in the first study and 13% in the second (P = 0.0011) (Figure 3).

The majority of patients reported they had no heartburn or mild heartburn at the end of the treatment phase. In study 1, 84.1% of subjects who received lansoprazole reported no heartburn or mild heartburn compared with 74.4% of subjects receiving placebo. In study 2, no heartburn or mild heartburn was reported in 85.2% of subjects who received lansoprazole compared with 73.8% of subjects receiving placebo.

In studies 1 and 2, > 60% of subjects treated with lansoprazole 15 mg were very or extremely satisfied with the treatment, and > 50% reported significant benefit from the study medication (P < 0.0001 compared with placebo). Additionally, in the 2 studies, a significantly greater percentage of recipients of lansoprazole 15 mg compared with the recipients of placebo reported somewhat or much better symptom relief with study medication relative to their experience with other medications for both daytime (70% vs 57%) and night-time heartburn (66% vs 53%; P < 0.0005).

During the first 7 days after the double-blind treatment phase in study 1, 46.5% of the 24-hour days were heartburn-free in subjects treated with lansoprazole compared with 46.7% of placebo recipients. In study 2, the values were 44.0% versus 46.5%. The differences were not statistically significant. On the first day of the follow-up period, 62% of subjects in study 1 and 57% in study 2 who had taken lansoprazole were heartburn-free compared with 43% and 42% of placebo recipients.

Post hoc analysis of pooled data from both studies showed that there were no significant differences between the younger and older age groups in the response to lansoprazole (Table 3). A second post hoc pooled analysis also showed that the difference in the number of heartburn-free 24-hour days during lansoprazole treatment was not statistically significant between men and women (Table 3).

The number of subjects who were non-white was small, and subjects who were black, Asian, and of other races were combined for post hoc efficacy analysis. Overall, 72% of the randomized subjects in the pooled data set were white, 13% were Hispanic, and 15% were black, Asian, or of other races. Treatment differences were 20.7% (P < 0.0001), 9.3% (P = 0.16), and 5.2% (P = 0.03) for white, Hispanic, and black, Asian, and other races, respectively.

Overall, lansoprazole 15 mg was well tolerated in the 2 randomized studies. Most AEs were mild and were similar across the 2 treatment groups in both studies. In study 1, 27 (10%) subjects in the lansoprazole group and 26 (9%) in the placebo group experienced 1 or more AEs during the 14-day treatment. In study 2, 38 (13%) subjects in the lansoprazole group and 23 (8%) in the placebo group experienced 1 or more AEs during the 14-day treatment. The most common AEs reported in the 2 studies are presented (Table 4). A total of 5 subjects treated with lansoprazole and 1 who received placebo across both studies experienced AEs suspected to be drug-related.

In study 1, 1 subject discontinued lansoprazole because of a “swollen throat” that was suspected to be related to the study drug, and 1 discontinued placebo because of moderate diarrhea and a moderate hot flush. In study 2, 4 subjects in the lansoprazole group discontinued because of AEs. One discontinued because of severe diarrhea, and 1 because of moderate stomach pain, dizziness, and headache; these events were all suspected to be related to the study drug. One subject discontinued because of mild headache, vomiting, and diarrhea, and 1 discontinued because of severe bacterial pneumonia; these events were not suspected to be related to the study drug.

In the 2 identical, double-blind, placebo-controlled studies reported, a 14-day treatment regimen with lansoprazole 15 mg showed rapid and sustained effectiveness and was well tolerated for the treatment of frequent heartburn in patients likely to self-select nonprescription medication without consulting a prescriber. No rebound of symptoms was observed after discontinuation of lansoprazole at the end of the study. Post hoc analysis showed that there were no differences in efficacy between patients aged < 65 or ≥ 65 years or between men and women.

The use of PPIs in patients who choose nonprescription medication has been previously documented in a 14-day study including 3557 subjects with frequent heartburn.⁸ An actual use study showed that patients were able to accurately self-select a PPI for personal use, comply with the dosing regimen, and appropriately seek advice from a primary care provider for longer-term management of frequent heartburn,⁹ which supports the use of nonprescription PPIs prior to consultation with a prescriber. However, a European pharmacy survey suggests that educating patients about the importance of consulting a physician may be justified.¹⁰

Lansoprazole has been available for prescription use in the United States since 1995 and is widely prescribed for a number of indications.^6,7,11-14 Prescription lansoprazole has been found to be well tolerated, with a similar AE profile¹⁴ to that seen in our 2 studies. Although PPIs such as lansoprazole provide rapid and sustained symptom relief that can last throughout an entire treatment period, a substantial proportion of patients with frequent heartburn still rely on other medications such as antacids and H₂RAs that do not provide adequate or sustained suppression of symptoms.⁸ Antacids treat heartburn symptoms rapidly but have a short duration of action.⁸ Histamine₂-receptor antagonists have a longer duration of action than antacids, but repeated dosing, which is essential for the treatment of frequent heartburn, has been reported to lead to tolerance or tachyphylaxis as early as the second or third day of treatment.^15,16

We observed a placebo response for all our primary and secondary endpoints, but this was comparable with that seen in similar studies in this setting.⁸ A detailed discussion of the mechanisms of placebo response is beyond the scope of this article, but possible factors include variations in the level of endogenous opioid system activation, “conditioning” of the subject, and expectancy.^17–19 Importantly, the results of our studies showed significant and clinically meaningful differences between placebo and lansoprazole responses. This is particularly true for night-time heartburn.

In the 2 studies reported, lansoprazole 15 mg once daily was superior to placebo in the treatment of frequent heartburn. The mean percentage of 24-hour days with no heartburn was significantly higher among subjects treated with lansoprazole 15 mg than among those receiving placebo (P < 0.0001). Lansoprazole 15 mg was also more effective than placebo in increasing the percentage of night-times with no heartburn during 14 days of treatment. A significantly larger percentage of patients reported no heartburn during day 1 of treatment with lansoprazole 15 mg compared with placebo. Furthermore, the subjects in both studies were very or extremely satisfied with lansoprazole treatment, particularly when compared with previous treatments. Both lansoprazole and placebo treatments were safe and well tolerated.

A recently completed study compared lansoprazole doses of 15 mg and 30 mg with placebo in patients with frequent night-time heartburn.²⁰ In this study, both doses had similar effects on the percentage of 24-hour days with no heartburn. Taken together with the results of our study, it is suggested that for patients with frequent heartburn, it may be possible to use a lower dose than that traditionally used for healing gastric ulcers¹³ and erosive esophagitis.²¹

A limitation of these studies is that the duration of treatment may have been too short to fully assess the effects of lansoprazole 15 mg for the treatment of frequent heartburn. However, the studies were designed to support the short-term efficacy and tolerability of lansoprazole among subjects self-selecting the treatment in an over-the-counter setting rather than long-term efficacy; this objective was achieved. The post hoc analyses by age and gender show that lansoprazole appears to be effective for the treatment of heartburn in these patient subgroups, but analyses of larger patient populations are required to confirm these preliminary findings. Further studies are also required to determine whether treatment with lansoprazole 15 mg once daily for 14 days can have an impact on quality of life, but the results of the global assessments suggest that most subjects in our studies were satisfied with the treatment. To our knowledge, however, use of these global assessment questions in patients with frequent heartburn has not been validated, somewhat limiting the conclusions that can be drawn. However, a beneficial effect of lansoprazole 15 mg on quality of life after 4 weeks has been reported previously.²²

The results of the studies reported here are clinically important because they demonstrate the efficacy and tolerability of another option for the treatment of frequent heartburn. The availability of low-cost, fast-acting PPIs without a prescription offer the possibility of changing the way in which frequent heartburn is managed by consumers and physicians.²³ The availability of lansoprazole without a prescription may reduce demands on health care insurance by patients who have coverage and allow patients who are not insured to obtain relief from their symptoms without the cost of consulting a physician. The use of a nonprescription PPI for treatment of frequent heartburn will allow the patient to provide more information to the physician at the consultation, aiding in the diagnosis of the patient’s symptoms.

In conclusion, the results of 2 double-blind, placebo-controlled studies of a 14-day duration demonstrate that lansoprazole 15 mg once daily is an effective, fast-acting, and well-tolerated treatment for frequent heartburn and should be considered an option for short-term use by patients with frequent heartburn who wish to select a nonprescription medication without consulting a prescriber.

The study was sponsored by Novartis Consumer Health, Inc. Editorial support for this article was provided by the editorial staff at Embryon.

Back to the table of contents for the July 2009 issue