For a long time, my grandfather suffered from indigestion. In his day, gastritis (i.e., inflammation of the stomach lining) was thought to be due to excessive acid secretion.1,2 And damage from stomach acid could cause peptic ulcers.3-5
Doctors noted that businessmen who had stressful jobs and were heavy smokers suffered gastritis and developed ulcers.6 But women’s symptoms were usually considered psychosomatic.1,7
The treatment was aimed at reducing or neutralizing excess acid. Men were told to lower their stress level, eat a bland diet (dairy products, no citrus or spicy foods and no caffeine) and take over-the-counter antacids such as Pepto-Bismol or TUMS.1,3,6,8 Women were often referred to a psychiatrist.1,7
My grandfather was a classic case. A lifelong hardscrabble farmer, his livelihood depended on the weather, over which he had no control. He smoked cigarettes, regularly ate fried food, drank strong coffee and like most men then and now, avoided going to the doctor for any reason. I don’t know how long he suffered because he never admitted it, but he was often cranky.
Things came to a head when he began coughing up blood and was rushed to the hospital. The doctors immediately recognized the symptoms and diagnosed a bleeding peptic ulcer—a common medical condition in those days. He was given repeated transfusions to replace his serious blood loss.
When the bleeding could not be controlled, the only remaining option was surgery. Unfortunately, 10–25% of postsurgical patients became “gastric cripples.”2,3,6 The excised stomach tissue left patients unable to enjoy food. They had difficulty maintaining a normal body weight and never regained complete health.2,6
In any case, surgery was not an option for my grandfather because of his frail physical condition and the limited facilities at the rural community hospital. He died in 1973 at the age of 74.
Suppressing Acid
Three years later, the Food and Drug Administration (FDA) approved cimetidine. It was the first H2 receptor antagonist.8,9
In the 1960s, James Black, a pharmacologist at SKF Research Institute in England, had discovered propranolol, the first beta-adrenergic blocker. It was subsequently marketed to treat angina, hypertension, myocardial infarction, migraine and glaucoma. Black reasoned that since there were at least two types of adrenergic receptors, there might also be more than one type of histamine receptor.8
Black and his colleagues aimed to control gastric acid secretion, which is stimulated by histamine. They began screening a variety of compounds, looking for an antihistamine that selectively prevented acid secretion in anesthetized rats. After four years and the failure of more than 200 analogs, Smith Kline & French executives in Philadelphia asked Black to abandon this seemingly futile project.8 But Black defiantly continued the research and just stopped sending progress reports to the U.S. headquarters.
In 1976, after 12 years of research, SKF launched cimetidine (Tagamet®), the first antihistamine that selectively inhibited the histamine H2 receptor.8 It was soon followed by ranitidine (Zantac®).
H2 receptor antagonists seemed to confirm the excessive acid theory. The drugs often healed ulcers outright.1,2 In the 1980s, 2–4% of Americans carried Tagamet tablets in their pockets.6
But cimetidine was a treatment, not a cure. When patients stopped taking it, their ulcers typically returned. By 1983, it was clear that for most ulcer patients, lifelong treatment was going to be necessary.1,2
Given that 10% of the world’s population suffered from peptic ulcer disease at some point in their lives, the H2 antagonists became bestsellers.1,2
In the early 1990s, FDA approved the first small molecule proton pump inhibitors (PPIs). PPIs like omeprazole (Prilosec®) and lansoprazole (Prevacid®) are the most effective suppressors of gastric acid secretion.8 Like H2 antagonists, PPIs promoted ulcer healing, and they revolutionized treatment.8,10
Despite these advances, gastritis, peptic ulcer disease, and stomach cancer remained a global health concern. In Japan, for example, gastritis was so common it was considered “normal,” and the Japanese also suffered the world’s highest rate of stomach cancer.2,11
Stomach Bugs
The next great advance in gastritis research came from, of all places, a basement pathology laboratory in Perth, Australia.
J. Robin Warren graduated from medical school at the University of Adelaide in 1961 and wanted to specialize in psychiatry. Unfortunately, the residency program did not accept him, so he chose clinical pathology instead. In those days, he said, “Clinical Pathology meant mainly laboratory hematology, which I thoroughly enjoyed.”9 In addition to blood smears and bone marrow specimens, Warren had the opportunity to examine a variety of other tissues (including the stomach) for parasites, fungus, and bacteria.9
When he joined Royal Perth Hospital as a pathologist in 1968, gastroenterologists were using a rigid gastroscope, and gastric biopsies were difficult to obtain.9,12 Then, with the introduction of the flexible endoscope in the 1970s, they could directly view the stomach lining, easily perform biopsies, and more accurately diagnose gastritis and peptic ulcers.8,12
In the biopsy specimens of peptic ulcers and stomach tumors, Warren sometimes saw yeast, fungi, or bacteria amongst the necrotic debris. The mixed variety of bacteria were mostly dead organisms just passing through or perhaps contamination.12
By 1979, Warren had developed a particular interest in stomach pathology, especially gastritis. He also played around with various microbiological stains and found a silver stain that differentiated Gram-negative bacilli from the complex mass of gastric tissue structures.12
In June 1979, Warren examined a routine biopsy taken from a patient with active chronic gastritis. The inflamed stomach lining, not surprisingly, contained white blood cells. But there were also numerous bacteria lining the mucosal layer. Electron microscope images clearly showed small curved and spiral bacilli attached to the microvilli of the stomach’s epithelium.12
For more than 100 years, physicians were taught that bacteria cannot survive in the stomach’s strong acid environment.2,12 “When I said they were there, no one believed it.”9 Warren’s colleagues challenged him to find more cases. And he did. “Eventually I was finding them in about a third of the gastric biopsies.”12
As a pathologist, Warren did not have direct contact with patients. He relied on gastroenterologists to send him specimens. They took biopsies from visible lesions, such as ulcers, and asked Warren to assist with their diagnoses.12
Warren wanted to also examine the intact stomach mucosa, which would allow him to observe the bacteria without the complication of other lesions. Most gastroenterologists did not consider gastritis important, and they thought taking biopsies of normal gastric tissues was “ludicrous.”12
But with a kind of “scholarly stubbornness,” Warren stuck to the facts and was not influenced by other people’s opinions. “Win [Warren’s wife] was the only person to accept my work and encourage me.”9 After two years, Warren had found the spiral bacteria in many clinical cases and was ready to publish his findings.
Then, Barry Marshall arrived.
Most gastroenterologists did not consider gastritis important, and they thought taking biopsies of normal gastric tissues was “ludicrous.”
A Partner, at Last
Growing up in Australia, Barry Marshall and his brothers had access to the technical books and tools of their father, a steam engine and train mechanic. They frequently got into trouble building and dismantling things. Mrs. Marshall had trained as a nurse, and Barry read her medical and nursing books, too.7
In college, Marshall was interested in science, but electrical engineering required math beyond his abilities. Instead, he chose medical school, “which was equally interesting and did not require daily exposure to calculus.”7 He enjoyed the biological sciences, particularly biochemistry, and aimed for an academic career that would combine research with clinical medicine.
Marshall was a “lateral thinker” and took a broad approach to problem-solving. “In school my grades always suffered because I was continually mucking about with irrelevant side issues which I often found to be more interesting.”7
After internship and residencies in internal medicine, Marshall began specialist training. On each specialist rotation, he was encouraged to conduct a clinical research project. In 1981, his gastroenterology supervisor referred him to Warren, who had requested some research assistance.7
During their first meeting in August 1981, Warren showed Marshall his microscope slides and explained the histopathology of the gastric mucosa.7 Warren had found the spiral bacteria in stomach biopsy specimens from 27 patients.2
Perhaps, because of his broad training as an internist, Marshall was more receptive to the notion of infectious disease than gastroenterologists. He was not totally convinced, but he thought that studying this previously unidentified gastric bacteria would at least be an interesting project.7
Warren handed him the list of patients and asked him to review their medical records and document their diseases. Among them was a 50-year-old woman Marshall had seen on the gastroenterology ward.2,7 She suffered from severe stomach pain. The usual battery of tests revealed nothing, and she was referred to a psychiatrist, who put her on an antidepressant, “for want of a better treatment.”7 The only abnormal findings had been some redness in her stomach and Warren’s bacteria in the stomach biopsy.2,7
Marshall saw no obvious connection between the bacteria and the 27 patients’ medical ailments. But he agreed to provide Warren with more stomach biopsies. Finally, Warren had a clinical collaborator.12
Initial Steps
For the rest of 1981, Marshall and his colleagues collected biopsy specimens from patients who were scheduled for endoscopy. In addition to sampling the ulcer borders, which were often disturbed and always inflamed, Warren asked them to take samples a few centimeters away from the ulcer.2
The frequent appearance of Gram-negative bacteria in inflamed gastric mucosa prompted Marshall and Warren to do a pilot study on 20 patients. The curved bacilli were present in more than half of those biopsy specimens, and the number of bacteria closely correlated with the severity of gastritis.13
Marshall’s next rotation (in the first half of 1982) took him to the hematology department, where he cared for bone marrow transplant patients. His research project was studying heatstroke in marathon runners, but he managed to continue his work with Warren as well.2
Marshall wrote a protocol for a formal clinical trial in patients who were scheduled for elective endoscopy at Royal Perth Hospital.2 The study was designed to confirm the association between gastritis and the bacteria, to culture and identify the bacteria, and to identify factors that made patients susceptible to infection.13
They collected specimens from 100 endoscopy patients. In the first 34 patients, they could clearly see masses of the Gram-negative bacteria under the microscope, but their attempts to culture it were unsuccessful.2,12,13
Patient 35 was a 70-year-old man with a history of duodenal and gastric ulcers.2 On Thursday, April 8, 1982, the man’s biopsy was sent to the hospital’s microbiology lab. Normally, the culture would have been read two days later. But because of the long Easter weekend, the culture plate remained in the incubator until Tuesday, April 13. What emerged on the plate was “a rather pure culture.”2 Realizing that the bacterium was actually slow growing, they lengthened the incubation time.2,12,13
The curved or S-shaped bacteria had a smooth coat and usually 4 sheathed flagella at one end. On the culture plates, it was sensitive to a variety of antibiotics, including tetracycline, erythromycin, kanamycin, gentamicin, and penicillin.13
The bacteria closely resembled Campylobacter, a pathogenic bacteria found in the intestinal tract. Warren tentatively named this new organism Campylobacter pyloridis but admitted the name might not be appropriate because of differences in morphology.13
Other Campylobacter species have a single unsheathed flagella and lack many of the proteins and lipids found in Warren’s new bacterium.14 Because of these differences, a new genus, Helicobacter, was eventually created. Campylobacter pyloridis became Helicobacter pylori.
The final data from Patient 100 were collected in May 1982. Marshall compiled and completed the data analysis with the assistance of a statistician at the hospital.2
They found a close correlation between peptic ulcers and the presence of bacteria, and most of the patients with peptic ulcers also had gastritis. Even in the gastritis patients who did not have an ulcer, H pylori was present. On the other hand, it was not found in most patients with a normal stomach mucosa.13
Marshall had collected extensive medical histories on each patient, but he could not find any environmental, lifestyle, dietary, or medical factors that predisposed patients to infection. They concluded, “Although cause-and-effect cannot be proved in a study of this kind, we believe that pyloric campylobacter [H pylori] is etiologically related to chronic antral gastritis and, probably, to peptic ulceration also.”13
They found a close correlation between peptic ulcers and the presence of bacteria, and most of the patients with peptic ulcers also had gastritis.
Doubting Doctors
For his remaining rotation (in the second half of 1982), Marshall was assigned to Port Hedland Hospital, 1250 miles north of Perth. He had become very excited about Warren’s spiral bacteria, but he was still a physician in training. He managed to squeeze his H pylori research between his remaining training and patient commitments.7
In October 1982, Marshall presented the trial results to the Royal Australian College of Physicians meeting in Perth. It received a mixed response.2 Gastroenterologists thought the notion that a germ caused ulcers was just “too weird.”6 In late 1982, Marshall submitted an abstract to the Gastroenterology Society of Australia, but it was rejected.5
Marshall and Warren then submitted back-to-back letters to the editor of The Lancet. Warren’s one-page letter described his initial observations of H pylori and its association with gastritis. He speculated that the bacteria were able to survive in the stomach by growing in close contact with the stomach epithelium where it was protected from stomach acid by the overlying mucus layer.15
In his two-page letter, Marshall briefly reviewed literature references to spiral gastric bacteria. For more than 100 years, researchers had seen it but considered it of no particular importance—either a contaminant or part of the normal gut flora. Marshall also described the culture methods they had devised, which allowed them, for the first time, to extensively characterize this slow-growing bacteria.16 He said that if H pylori was associated with gastritis, as Warren proposed, it might also be a factor in other “poorly understood” gastric diseases, such as peptic ulcers.16
Research Commitment
By the time Marshall completed his training at the end of 1982, he had decided to specialize in gastroenterology or microbiology, so that he could continue his research on H pylori. Unfortunately, there were no available jobs at Royal Perth Hospital. Instead, he accepted an endoscopy training post at Fremantle Hospital, the smallest teaching hospital in Perth.2,7
In January 1983, Marshall began isolating H pylori from patients at Fremantle Hospital, but he also stayed in touch with Warren at Royal Perth Hospital.7
In September 1983, Marshall presented the Royal Perth trial results at the 2nd International Workshop on Campylobacter Infections in Brussels, Belgium.7,12 The European microbiologists at the conference were impressed and followed up on the research in their labs.5–7 But gastroenterologists remained skeptical.
Marshall and Warren then submitted a full report of their clinical trial results to The Lancet.2 The editors wanted to publish the manuscript, but they could not find reviewers who would recommend acceptance.12
So, Marshall contacted Martin Skirrow, who had chaired the Brussels conference. Skirrow repeated the work in his lab and confirmed the Australians’ results. “[Skirrow] informed The Lancet and shortly afterwards they published our paper, unaltered.”12
In the paper, which appeared on June 16, 1984, Warren and Marshall proposed that gastritis was caused by H pylori and that subsequent damage to the mucus layer coating the stomach predisposed patients to peptic ulcers.13
Defining the Disease
At Fremantle Hospital, Marshall was taking biopsies from “every patient who came through the door.”6 Between January 1983 and September 1984, he had collected data from 350 patients.14 The results confirmed and extended the observations Marshall and Warren had observed in their Royal Perth Hospital study.
Marshall and his Fremantle colleagues found that H pylori was rarely present in biopsy specimens that showed no evidence of inflammation. On the other hand, H pylori was associated with gastritis, and stomach ulcers occurred in patients with gastritis. It appeared that H pylori could also damage the duodenal epithelium, leading to duodenal inflammation and ulcers.14
Interestingly, most patients with either stomach or duodenal ulcers had normal levels of acid secretion, and some people who had high acid levels never developed ulcers.14 Although stomach acid is necessary for ulcers to form, it was not sufficient to explain their occurrence.3 Rather, Marshall found that H pylori infection diminished mucus production, which allowed even normal levels of acid to penetrate the mucosa and cause ulcer formation.14
In addition to antibiotics, Marshall found that bismuth salts suppressed H pylori on the culture plates.2,5 Bismuth had been prescribed for more than 100 years and was the only antacid remedy known to heal ulcers. It also decreased the chances of ulcer recurrence.13,17 Marshall’s data indicated that bismuth’s efficacy was more likely due to its antimicrobial action than its antacid properties.
Marshall treated some gastritis patients with bismuth and found that it eradicated their H pylori infections. This lent further support to the notion that the symptoms of gastritis and the cause of ulcers were closely related to H pylori.17
Warren and Marshall proposed that gastritis was caused by H pylori and that subsequent damage to the mucus layer coating the stomach predisposed patients to peptic ulcers.
If Marshall and Warren were correct, antibacterial agents might provide a permanent cure.14,17 Unfortunately, gastroenterologists remained extremely skeptical.2 There was still no clear evidence that H pylori was the primary cause of gastritis; it might just be a harmless organism in the mucosa.18 For example, H pylori appeared in autopsy specimens of many individuals with no history of ulcers.4
“No amount of logical reasoning could budge what people knew in their hearts to be true… A bacteria cause was preposterous.”2 They disregarded Marshall, a young physician with thin research credentials in a remote part of Australia.7
In addition, several things did not fit Marshall and Warren’s hypothesis. For example, most patients infected with H pylori have no symptoms.12 About half of the world’s population is infected with H pylori, but fewer than 20% of them have peptic ulcers.4,19–21
Marshall had also seen patients with peptic ulcers, despite no H pylori infection.14 In reviewing those patients’ medical history, he concluded that the most likely cause of their ulcers was ingestion of NSAIDs like aspirin and ibuprofen.2,14 In addition to inhibiting prostaglandin synthesis, NSAIDs diminish the secretion of protective mucus in the stomach.8
Providing Proof
To satisfy the skeptics, Marshall realized he would need to present evidence that met Koch’s criteria. In 1890, Robert Koch published four postulates that, he said, would establish that a specific microbe caused a corresponding disease.5
Marshall and Warren had already fulfilled the first two criteria. In patients with gastritis, they had almost always found H pylori adhering to the gastric mucosa (Koch’s first postulate). They had also cultured and characterized H pylori from more than 150 patients with gastritis (Koch’s second postulate).
Koch’s third postulate required that inoculation with the microbe, taken from a purified culture, would produce the specified disease in a susceptible individual.
In January 1984, Marshall began inoculating piglets, mice, and rats with H pylori.2 Unfortunately, biopsies showed the animals’ stomachs remained normal. They did not develop gastritis.2,7 It turns out that H pylori infect only primates.6
Marshall had been arguing with the skeptics for two years, and he became more frustrated when his animal experiments failed.7,19 He concluded that the only way to prove the point was to infect himself with H pylori. Several factors convinced him that this was a “reasonable” thing to do.2
First, he had developed a crude H pylori antibody blood test. He tested blood from healthy blood donors at Fremantle and discovered that 43% of them were infected with H pylori but symptom-free. This meant that the infection “was not necessarily fatal.”2
Second, he treated four of his patients with bismuth and metronidazole, and they were all cured.2
Finally, because he had personally collected all of the research data, he reasoned that he was the only person in the world who could make an informed decision about the risks of an H pylori infection.2
“Maybe it was safe for me to try swallowing the bacteria to see what really happened.”2 Warren didn’t like the idea and said so.9
Knowing that the local ethics committee might deny his request and that he risked losing his medical license, Marshall decided it was better to ask forgiveness than permission.2,7 He told no one, not even his wife. But he did gain the “unofficial support” of a few senior colleagues at Fremantle, because he needed them to do the endoscopies and microbiology analysis.2
Self-Experimentation
In July 1984, the pre-study endoscopy confirmed that Marshall had a healthy stomach, with no signs of inflammation, histopathology, or H pylori.2,19
Three weeks later, on Day 1 of the experiment, Marshall took a dose of cimetidine. The H2 blocker would lower his level of stomach acid, which might make it easier for H pylori to colonize.2,18
The H pylori test material came from a 66-year-old man who suffered from indigestion but did not have an ulcer. On a culture plate, this specimen was sensitive to ampicillin, erythromycin, tinidazole, doxycycline, and rifampicin.18
At 11:00 a.m., 32-year-old Barry Marshall drank about 10 ml of the reconstituted H pylori suspended in peptone water (a microbial growth medium).2,18 He later said, “If only I knew that people would be so interested, I would have taken a photograph!”19
“Maybe it was safe for me to try swallowing the bacteria to see what really happened.” – Barry Marshall
Marshall expected to develop an asymptomatic infection, and for the next few days, he remained symptom-free.19 He noticed only some “audible gurgling.”18 But after Day 5, he was sleeping poorly and waking early. At 6:00 a.m. on Day 8, he ran to the bathroom to vomit.2,18 He flushed the clear, slimy vomit, not thinking to save a sample for later testing. But he did note that it lacked the usual acid taste.2
Marshall was surprised at the severity of the infection. “I didn’t expect to become as ill as I did.”7 His wife complained about his “putrid breath,” which his Fremantle colleagues also noticed.2 Clearly, Marshall had developed the same symptoms as his H pylori-positive patients, and this satisfied Koch’s third postulate.18
On Day 10, an endoscopy showed severe gastritis and epithelial damage. Biopsy specimens were taken and showed spiral bacilli adhering to the surface of his stomach, as well as white blood cells in the mucosal layer. Marshall’s colleagues identified the cultured bacteria as H pylori. And it was sensitive to the same antibiotics as in the specimen from the 66-year-old donor. These results satisfied Koch’s fourth postulate.18
Marshall had proven that H pylori was the pathogen: It colonized in his completely normal stomach and caused acute inflammation, mucus depletion and epithelial cell damage.18
Fortunately, by Day 14, Marshall’s symptoms subsided. At that point, he decided to terminate the experiment, and on his wife’s orders, began treatment with tinidazole.2,18,19 A biopsy (taken before he began tinidazole treatment) showed no bacteria, and the stomach epithelium was returning to normal.18 Also, his blood was negative for H pylori antibodies.2
Marshall’s body had effectively eradicated the H pylori infection. In hindsight, he believed tinidazole, as a single agent, would not have worked, anyway.2 Even at this early stage, he had learned that multi-drug combinations were required to eradicate H pylori in patients.
Repeat Performance
When Arthur Morris saw Marshall’s report, which was published in the Medical Journal of Australia, the 29-year-old New Zealand microbiologist decided to repeat the experiment. Like Marshall, Morris developed gastritis and an H pylori infection. Also, his measurements confirmed Marshall’s anecdotal observation: the gastric pH was, indeed, near neutral.2
But unlike Marshall, who had eradicated H pylori without any specific therapy, Morris developed a chronic infection.22 It took several years of bismuth and antibiotic treatment for Morris to finally clear the infection and cure his residual gastritis.4,5,12,22
Marshall suspected that a chronic H pylori infection and persistent inflammation of the stomach mucosa (such as Morris experienced), when coupled with acid secretion, set the conditions for developing peptic ulcers.18
But neither Marshall nor Morris developed an ulcer. The connection between H pylori and ulcers was eventually deduced from epidemiological studies. That data indicated that patients who develop chronic gastritis 10–20 years after an initial H pylori infection have a 3- to 12-times greater likelihood of developing a peptic ulcer.3,4,23,24
Public Pressure
The editors of the Medical Journal of Australia required extensive revisions, but they eventually agreed to publish Marshall’s self-experiment study in 1985. The editors of The Lancet then cited the Australian paper in an editorial, “giving it far more notoriety than it might otherwise have had.”2
Journalists and science reporters contacted Marshall and Warren for interviews.6,7,12 Detailed articles describing their research appeared in The New York Times, Readers Digest, The New Yorker, and various tabloids.2,7
Although gastroenterologists still refused to prescribe antibiotics, desperate patients from around the world saw the articles and inundated Marshall with requests.7 He spent many hours each week mailing treatment advice.2
In 1990, Abbott Laboratories, which funded some of Marshall’s research, immortalized his self-administration experiment in a comic, to advertise its new antibiotic, clarithromycin.2
Shifting to Treatment
At Fremantle Hospital, Marshall was routinely screening his endoscopy patients using his crude antibody blood test. By the end of 1984, he had identified a dramatically increased number of H pylori-infected patients. He was able to eradicate H pylori in most of them in just two weeks with a combination of bismuth and metronidazole. If that failed, he combined bismuth with amoxicillin.2
Local primary care physicians noticed Marshall’s remarkable results and sent him more patients. He cured most of them. And, patients with so-called “psychosomatic gastric symptoms” were pleased to have a real diagnosis: H pylori gastritis.2
In 1985, Marshall moved back to Royal Perth Hospital, where he expanded his collaboration with Warren. From April 1985 to August 1987, they recruited 100 patients for a new trial to compare the effects of antibiotic treatment to the current standard-of-care remedies.17
Marshall purposely selected patients with the most severe H pylori infections: cigarette smokers who also had confirmed ulcers.2 (Smokers were known to have rapid ulcer recurrence.)
The first and second groups in the double-blind trial were given standard treatments: cimetidine (an H2 antagonist) and bismuth, respectively. Because antibiotic treatment was considered experimental, the third and fourth groups received the antibiotic (tinidazole) in combination with either cimetidine or bismuth. The treatment period for all four groups lasted eight weeks.17
H pylori was eradicated in three-fourths of those who received the tinidazole-bismuth combo. Almost all of those patients experienced healed ulcers, and the ulcers recurred in only four of them during the 12-month follow-up period. On the other hand, H pylori persisted in all of those who received cimetidine, only 40% of them experienced ulcer healing, and the ulcers recurred in all of them during the 12-month follow-up.17
In short, once H pylori was eradicated, the patients did well, even those who smoked or had a previous history of chronic relapsing ulcers. Marshall said, “Detection of H pylori should be part of the routine management of patients with acid peptic disease, and eradication of the bacterium [should be] a major therapeutic goal.”17
Many other investigators, using variations of this protocol, reported the same or better results.2
Diagnosing H pylori
The media buzz caught the attention of Proctor and Gamble, the makers of the bismuth-containing antacid, Pepto-Bismol. P&G realized the economic potential of Marshall’s work and funded a fellowship for him to continue his research in the U.S.7
In 1986, Marshall moved with his family to the University of Virginia where he set up a gastroenterology lab focused on treatment and diagnostics.7
Dutch researchers had reported that H pylori contain massive quantities of urease, the enzyme that converts urea to ammonia and carbon dioxide. Ammonia production accounted for the observed near-neutral gastric pH in infected individuals and enabled H pylori to survive in the stomach.2 Later studies showed that ammonia, along with an H pylori-generated cytotoxin, also irritates stomach epithelial cells and triggers the inflammatory reaction.4
After a little tinkering, Marshall developed a simple diagnostic test to detect urease in gastric biopsies. The test, which he called CLOtest (for Campylobacter Like Organism test), could diagnose H pylori infections in a few minutes without elaborate lab facilities.2
Marshall also devised a simple test to detect the urease-generated carbon dioxide expired in an H pylori-infected patient’s breath.2 This office-based diagnostic test delivers results in 10–20 minutes.10,20
Proctor & Gamble set up a business relationship with Marshall and assisted him with his diagnostics patents. Marshall also met medical device executives who funded studies for the regulatory approval and US launch of CLOtest and the Urea Breath Test.7 Both diagnostic tests are still marketed by Kimberly Clark.
Official Acceptance
In the early 1990s, many investigators reported finding H pylori in their patients who had gastritis and ulcers, and antimicrobial therapy resulted in ulcer cures.4,6 The weight of the evidence was compelling. “By 1992, I could go to meetings and receive as much praise as criticism.”7
Local primary care physicians noticed Marshall’s remarkable results and sent him more patients. He cured most of them.
In February 1994, the National Institutes of Health convened a conference in Washington, D.C.2 Gastroenterologists, epidemiologists, infectious disease experts, and other specialists, including Marshall, issued a consensus statement, announcing that the key to treating peptic ulcers was detection and eradication of H pylori.7,25
After 10 years of persistent research, Marshall had finally achieved “official acceptance.” In 1996, he and his family moved back to Perth. He became a clinical professor of medicine at the University of Western Australia and continued research related to H pylori.7
Caveats
The studies by Marshall and others clearly showed that permanent H pylori eradication required multiple drugs. The initial triple-drug combination consisted of a bismuth compound plus two antibiotics (chosen from amoxicillin, clarithromycin, metronidazole, or tetracycline).8,20
But by the mid-2010s, the efficacy of triple therapy had fallen below 80%, due to increasing antibiotic resistance.20,26,27 Rifabutin, a derivative of rifamycin and mainly used to prevent or treat tuberculosis, is highly effective against H pylori. It has been used off-label to treat refractory cases.28
The current first-line therapy is a quadruple regimen consisting of two antibiotics, a bismuth compound, and a PPI for 14 days.20,26 PPIs have increased in importance, because suppressing acid secretion enhances ulcer cure rates. Also, because of successful H pylori eradication, NSAIDs now account for more than half of peptic ulcer cases, and those patients are treated with PPIs.10,19,29
Stomach Cancer
At the start of the 20th century, stomach cancer was the leading cause of cancer death in the U.S.3 Many factors contribute to the disease, including diet, socioeconomic status and genetics.4,11,19
Roughly 1–3% of H pylori-infected people develop stomach cancer.4,6,21,23 But it was difficult to establish a causal relationship, because the time between initial H pylori infection and the development of cancer is so long—usually decades.4 The accumulated evidence from preclinical and clinical studies now unequivocally supports the view that H pylori is, indeed, a factor.21
In 1994, WHO’s International Agency for Research on Cancer classified H pylori as a Class 1 carcinogen. WHO also recommended H pylori eradication to prevent stomach cancer.4,24
Now that antibiotic treatment is the standard-of-care for gastritis and peptic ulcers, stomach cancer has almost disappeared from the Western world.6 In Japan, where more than 99% of stomach cancer cases are associated with H pylori-gastritis, antibiotic treatment has significantly reduced both the infection rate and the incidence of stomach cancer.11,24
In 2005, Barry Marshall and Robin Warren received the Nobel Prize in Physiology or Medicine for their discovery of H pylori and its role in gastritis and peptic ulcer disease. After a long and distinguished career at Royal Perth Hospital, Warren died in July 2024. Marshall continues research and co-directs the Marshall Center for Infectious Diseases and Training at the University of Western Australia.
