Clostridium difficile has been casting an ever-growing specter over North America and around the world, particularly with the emergence of a virulent new strain known as BI/NAP1/027. Tens of thousands of hospitalized patients have been stricken with C. difficile-associated diarrhea and hundreds have died. The current guidelines on the diagnosis and treatment of C. difficile infection (CDI) were produced in 1995 by the Society for Healthcare Epidemiology of America (SHEA) and are badly out of date. Hence, a new set, currently in production, can’t be made available soon enough for clinicians struggling to contain the deadly organism.
Nevertheless, the threat of CDI may soon begin to diminish, thanks to the recent fast-paced and landmark leaps in knowledge about the organism. For example, it was recently discovered that toxin B—rather than toxin A, as was long believed—is the key to the ability of C. difficile to sicken and kill its victims. Furthermore, progress has been made in the accurate diagnosis of CDI and methods used to determine which patients are likely to have recurrences of the infection; a Phase II trial of a potential vaccine against C. difficile is under way; and therapies to potentially cure CDI and prevent recurrences are moving through Phase II and III testing.
“We’re on the cusp of change,” affirmed Thomas Louie, MD, professor of medicine, University of Calgary, in Canada, and an investigator in trials of the anti-CDI drug candidate OPT-80. “We are on the threshold of actually realizing better diagnosis of CDI by molecular methods, and we’re also marching toward better therapy.”
A Widespread Problem
The astounding statistics relating to C. difficile now are familiar to most physicians. But until the late 1990s, CDI was not a major player on the nosocomial-infection field, affecting only about 30 to 40 cases out of every 100,000 individuals. However, with the appearance of the BI/NAP1/027 strain, CDI started marching into the record books. Subsequently, its prevalence rose to 50 cases per 100,000 individuals, and C. difficile went on a killing spree, felling hundreds of people in hospital outbreaks in the province of Quebec, Canada, as well as in outbreaks in the states of Georgia, Illinois, Maine, New Jersey, Oregon and Pennsylvania.
The virulent strain has now spread to at least 40 states in the United States, all 10 provinces in Canada and several European countries. The rates of infection and mortality also have continued to increase, according to a paper co-authored by Dale Gerding, MD, a leading expert in C. difficile detection and strain typing (O’Connor JR et al. Gastroenterology 2009;136:1913-1924). Dr. Gerding and his co-authors also point out that C. difficile favors not only the sick but also the elderly: Its rate per 1,000 hospital admissions rises from 3.5 among patients younger than age 40 years, to 38.3 in the 71- to 80-year-old age bracket, to a whopping 74.4 in people over 90 years of age. The 30-day death rate associated with CDI rises in sync, at 2.6%, 6.2% and 14% in these age groups, respectively.
A Cunning Culprit
Hospitalized patients are most at risk for developing the potentially fatal diarrhea associated with C. difficile. Infection with the bacteria is precipitated by treatment with antibiotics or other mechanisms that disrupt the normal gastrointestinal flora. C. difficile also often evades measures to stop its spread because alcohol-based hand gels—today’s hand-hygiene infection control product of choice—do not kill the disease-causing spores of C. difficile. Instead, hand washing and thorough housekeeping measures are more effective at limiting the microorganism’s spread.
C. difficile carries a very large punch in a very small package. Just five genes control its pathogenicity, including the genes that encode toxins A and B. The two toxins have a very efficient and irreversible way of causing cell death: They slip inside the cell membrane and cause the collapse of the actin cytoskeleton—sort of like the wolf blowing down the little pig’s straw house.
It was believed for many years that toxin A was the main culprit in these bodily crimes, and that toxin B was merely an accomplice. Thus, most approaches for detecting the invader were developed around the presence of toxin A. But the whodunit has been turned on its head by Dr. Gerding and a team of 11 other researchers: In a letter published in Nature this spring, the investigators discussed their examination of hamsters infected with genetically manipulated strains of C. difficile bacteria, some of which did not produce toxin A and others that did not produce toxin B (Lyras D et al. Nature 2009;458:1176-1179). The researchers found that 16 of 17 hamsters infected with C. difficile containing an inactivated toxin A gene died. In contrast, only four of 19 hamsters infected with C. difficile containing an inactivated toxin B gene died. And all 10 hamsters infected with wild-type, or nonmanipulated, C. difficile died.
Accurate Detection of CDI— A Work in Progress
The trick, then, is to figure out how to detect and stop this tiny killer before it causes too much harm. The first step in that process is confirming whether an individual is infected with C. difficile. The current gold standard for diagnosis is toxin immunoassays, including those that detect both toxins A and B. However, these tests are relatively inaccurate, pointed out Dr. Gerding, associate chief of staff for research, Edward Hines Jr. Veterans Affairs Hospital, Hines, Va., and professor, Loyola University Chicago, Stritch School of Medicine.
“Current EIA [enzyme immunoassay] tests are insensitive,” said Dr. Gerding, who is also a member of the team preparing the next set of SHEA CDI guidelines. “PCR [polymerase chain reaction]-based tests will be much more sensitive. And testing for toxin B is essential.”
British researchers have written a review paper in which they conclude that the positive predictive values of immunoassays for toxins A and/or B are indeed “unacceptably low”—under 50% in some cases (Planche T et al. Lancet Infect Dis 2008;8:777-784). Hence, they suggest first testing with the highly sensitive rapid screening test for either toxins A and B or glutamate dehydrogenase (GDH)—an easily detected enzyme made by C. difficile—and then testing the stool of positive patients for cytotoxicity in cell culture.
“The two-stage ‘GDH triage’ approach also has been advocated by a group at Johns Hopkins,” said Ciaran Kelly, MD, associate professor of medicine, Harvard Medical School, Boston, and another leading expert in C. difficile. However, he added, “It is reasonable but not ideal. Any two-stage testing approach will lead to increased costs and delays in diagnosis. The attainable goal is a single, sensitive, specific and rapid test either using molecular techniques—for example, PCR—or improvements on the time-honored tissue culture cytotoxicity assay.”
Dr. Gerding agreed, and noted that some of the PCR assays in development amplify the genetic markers of the BI/NAP1/027 strain. However, he added that determining which strain a patient has is not as important as simply finding out whether CDI is present and if so, its severity and hence how best to treat it. He has co-authored a paper strongly urging the development of a standardized and validated system for stratifying CDI severity (Belmares J et al. Expert Rev Anti Infect Ther 2008;6:897-908).
Dr. Kelly and eight collaborators from Harvard and University College Dublin in Ireland have made a move along these lines. They devised a rule that can be used at the bedside to predict which patients with CDI are most at risk for recurrent infections (Hu MY et al. Gastroenterology 2009;136:1206-1214). The investigators derived the rule from multivariate regression analyses of data they collected from 63 patients with CDI who were hospitalized at Beth Israel Deaconess Medical Center in Boston. They found three factors that when combined correctly classified 71.9% of patients: age over 65 years, the presence of severe or fulminant underlying illness as determined by the Horn index and use of additional antibiotics after discontinuation of metronidazole or vancomycin for the initial CDI. Dr. Kelly told Gastroenterology & Endoscopy News that he and his colleagues at Harvard have been using this tool in clinical practice for several months and that more clinicians are beginning to adopt the protocol since the publication of the paper.
“The main limitation [in the clinic], however, is the lack of proven methods to avoid/reduce recurrent disease,” he said. “However, those are being developed, such as OPT-80 from Optimer Pharmaceuticals, Inc., and an anti-toxin A/anti-toxin B monoclonal antibody combination from Mass Biologics and Medarex.”
Effective Treatments Emerging
Indeed, some hope is on the horizon for clinicians struggling to help patients with initial and recurrent episodes of CDI.
Sanofi Pasteur announced in February 2009 that it is carrying out a Phase IIb study in the United Kingdom of a vaccine against C. difficile. A news release from the company said the candidate vaccine uses a toxoid-based approach, which it has used successfully in licensed vaccines against conditions such as tetanus and diphtheria. Approximately 600 patients will be enrolled at about 30 centers. Each subject will be randomized to one of four groups, with three groups receiving vaccine and the fourth a placebo. All subjects will also receive standard-of-care antibiotics.
Recurrence is by far the biggest problem associated with CDI: 15% to 20% of patients have a second episode of CDI within 60 days of their first one, and 65% of those who have had a third occurrence are likely to experience even more. Vancomycin—alone or in combination with metronidazole—is recommended by many experts as the treatment of choice for severe CDI, but in fact it does not appear to reduce recurrence.
One approach to CDI that has proven highly effective—and unfortunately unpalatable to many patients and physicians—is the introduction of feces from a healthy friend or relative into the patient’s bowel using a nasogastric tube or colonoscope. The principle behind the fecal transplant is that it restores the normal bioflora to the patient’s gut, making it an unfriendly breeding ground for C. difficile. The procedure recently received mainstream attention when it was showcased on an episode of “Grey’s Anatomy.”
Lawrence Brandt, MD, and Seth Persky, MD, of Montefiore Medical Center/Albert Einstein College of Medicine, in New York City, published the first academic paper on the colonoscopic fecal transplant for CDI, describing the treatment of a 60-year-old woman (Am J Gastroenterol 2000;95:3283-3285).
“Every time she came off her vancomycin or her metronidazole, she got debilitating diarrhea from the C. difficile,” recalled Dr. Brandt, chief, Division of Gastroenterology, Montefiore Medical Center. “So I gave her this fecal transplant from her husband. And at 11 o’clock that night, she called me at home and said, ‘I don’t know what you did in there, but whatever you did I haven’t felt this good in a year.’ She’s been healthy ever since—and that was close to 10 years ago.”
Dr. Brandt has now completed 12 fecal transplants, all via colonoscopy. A few other clinicians also are using this technique, including physicians at St. Mary’s/Duluth Clinical Health System, Duluth, Minn., who also have reported a very high success rate (Aas J et al. Clin Infect Dis 2003;36:580-585) and a team led by Dr. Louie that has completed approximately 50 cases with a 95% success rate.
The “yuck” factor will ensure that this approach, unfortunately, will not spread like wildfire, said Dr. Louie. Instead, he is bullish about a new medication called fidaxomicin, or OPT-80, that is in Phase III clinical trials.
Dr. Louie, who is medical director for infection prevention and control for Calgary’s hospitals, presented results from the first Phase III trial of OPT-80 at the 2009 European Congress of Clinical Microbiology and Infectious Diseases in Helsinki, Finland, in May. The same results were presented by his colleague Mark Miller, MD, at the 2009 Digestive Disease Week (DDW) meeting in Chicago in June. Dr. Louie does not have any financial ties to Optimer, the makers of OPT-80.
OPT-80, recently renamed fidaxomicin, is the first of a new class of antibiotics called macrocyclics. It inhibits the activity of RNA polymerase in C. difficile, causing its death. The Phase III trial led by Dr. Louie was a double-blind, head-to-head, randomized trial comparing fidaxomicin—given in twice-daily oral doses of 200 mg for 10 days—and vancomycin—given as four-times-daily oral doses of 125 mg for 10 days. The study included 629 patients who were enrolled at more than 100 sites in Canada and the United States.
The characteristics of the patients in each treatment group were similar. Six of the patients did not take any medication in the course of the study, and hence only the remaining 623 (300 on fidaxomicin and 323 on vancomycin) were included in the safety analysis; 596 patients (287 on fidaxomicin and 309 on vancomycin) were included in the modified intent-to-treat (mITT) analysis; and 548 (265 on fidaxomicin and 283 on vancomycin) formed the per-protocol cohort. The primary end point of the study was a clinical cure, defined as requiring no further medication two days after completion of the treatment.
The researchers did not observe a significant difference in the rate of clinical cure between patients taking fidaxomicin and those taking vancomycin—92.1% and 89.8%, respectively, among the per-protocol subjects, and 88.2% and 85.8%, respectively, among the mITT cohort. Dr. Louie and his collaborators performed subgroup analyses that indicated cure rates that were similar in both treatment groups among inpatients, outpatients, those harboring the epidemic strain BI/NAP1/027 and those under or over 65 years of age.
There was, however, a significant difference between the two antibiotics in recurrence rate: The rate was 13.3% among patients on fidaxomicin and 24% among those on vancomycin in the per-protocol analysis (P=0.004), and 15.4% and 25.3%, respectively, in the mITT analysis (P=0.005). The rate of global cure—defined as a cure not followed by a recurrence—also was significantly higher with fidaxomicin: It was 77.7% for patients taking fidaxomicin compared with 67.1% for those taking vancomycin in the per-protocol analysis, and 74.6% and 64.1%, respectively, in the mITT analysis (P=0.006 for both). Similar types and rates of adverse events were associated with both treatments, as well as similar all-cause mortality rates (5.3% for fidaxomicin, 6.5% for vancomycin).
“In some ways, vancomycin and OPT-80 treatment of CDI are different biological probes for the role of the normal microflora in recurrence of disease,” said Dr. Louie. “Both are equally effective in killing C. difficile. The difference, which we showed in earlier studies, is that OPT-80 allows components of the normal bacterial flora to remain present, whereas vancomycin acts more as a broad-spectrum agent. It is the more-intact, normal microflora that defends against recurrent disease.”
A second, confirmatory, Phase III trial of OPT-80 versus vancomycin is expected to be completed by the end of this year.
Data on another agent to cause a stir among the gastroenterology community also was presented at the 2009 DDW meeting. There, Israel Lowy, MD, PhD, senior director of clinical science and infectious diseases at Medarex, Princeton, N.J., presented a Phase II efficacy study of two fully human monoclonal antibodies targeting C. difficile toxins A and B, known as CDA1 and CDB1, respectively. The antibodies and clinical studies were developed in a collaboration between Mass Biologics of the University of Massachusetts Medical School in Boston and Medarex.
In this study, individuals who had persistent diarrhea despite treatment with either metronidazole or vancomycin were randomized to receive either a single, 10 mg/kg IV infusion of CDA1 and CDB1, or normal saline. The study enrolled 200 patients at 30 different sites; 101 were randomized to receive active treatment and 99 received placebo. The two groups of patients had similar rates of infection with the BI/NAP1/027 strain (32.5%, CDA1 + CDB1 group; 25.7%, placebo group), prior CDI (approximately 30% for both) and inpatient status at enrollment (approximately 50% for both). The investigators followed the subjects for 84 days post-infusion to assess safety and efficacy of the treatment.
The researchers reported a 70% reduction in CDI recurrence rate in the ITT analysis (6.9% for CDA1 + CDB1 and 25.3% for placebo; P=0.0004). The significant difference in recurrence rate among the groups was observed even for those over age 65 years, at just under 7% for those who received the antibodies and just under 40% for those who received placebo (6.9% vs. 37.5%, respectively; P=0.006), and also for those with a prior history of CDI. There was a trend toward significance in recurrence rate among patients with the epidemic strain (8% vs. 31.6%, respectively; P=0.06) and also in severe diarrhea during the initial episode of CDI (29.7% vs. 43.4%, respectively; P=0.056). The length of initial hospitalization was similar in both groups (9.54 vs. 9.37 days, respectively); however, in a separate analysis, the investigators observed a significant difference in the percentage of patients who required a subsequent hospitalization (8.91% vs. 20.2%, respectively; P=0.028).
“The data suggest that the antibodies, in addition to clearly protecting patients from recurrence—including the patients in subgroups at higher risk for recurrence—also had a moderating effect on the extent of diarrhea during the initial episode, as well as a positive impact on the likelihood of requiring a new hospitalization during the three-month follow-up period,” Dr. Lowy told Gastroenterology & Endoscopy News.
Dr. Lowy also reported significantly lower rates of dehydration (0% for CDA1 + CDB1 vs. 5% for placebo; P=0.03) as well as hypotension (0% for CDA1 + CDB1 vs. 7% for placebo; P=0.01) for patients who received the antibodies.
Further testing of this antibody combination is being carried out by Merck & Co., Inc., which licensed the antibodies from Medarex and Mass Biologics in April 2009.
The world is holding its breath as all of these developments continue apace. It is anybody’s guess as to when, if ever, a definitive set of diagnostic tests and treatments for CDI will emerge. What is certain, however, is that the next few months and years will bring continued progress toward reaching this goal, that more patients will be spared a painful and distressing course of diarrhea and that a greater number of those who are affected will survive the ordeal.