|
|
Background We created the orphan drug Human Botulism Immune Globulin Intravenous (BIG-IV), which neutralizes botulinum toxin, and evaluated its safety and efficacy in treating infant botulism, the intestinal-toxemia form of human botulism.
Methods We performed a five-year, randomized, double-blind, placebo-controlled trial statewide, in California, of BIG-IV in 122 infants with suspected (and subsequently laboratory-confirmed) infant botulism (75 caused by type A Clostridium botulinum toxin, and 47 by type B toxin); treatment was given within three days after hospital admission. We subsequently performed a 6-year nationwide, open-label study of 382 laboratory-confirmed cases of infant botulism treated within 18 days after hospital admission.
Results As compared with the control group in the randomized trial, infants treated with BIG-IV had a reduction in the mean length of the hospital stay, the primary efficacy outcome measure, from 5.7 weeks to 2.6 weeks (P<0.001). BIG-IV treatment also reduced the mean duration of intensive care by 3.2 weeks (P<0.001), the mean duration of mechanical ventilation by 2.6 weeks (P=0.01), the mean duration of tube or intravenous feeding by 6.4 weeks (P<0.001), and the mean hospital charges per patient by $88,600 (in 2004 U.S. dollars; P<0.001). There were no serious adverse events attributable to BIG-IV. In the open-label study, infants treated with BIG-IV within seven days of admission had a mean length of hospital stay of 2.2 weeks, and early treatment with BIG-IV shortened the mean length of stay significantly more than did later treatment.
Conclusions Prompt treatment of infant botulism type A or type B with BIG-IV was safe and effective in shortening the length and cost of the hospital stay and the severity of illness.
|
The equine botulism antitoxin used for adult patients has not been used to treat patients with infant botulism in the United States, because of its serious side effects when given to adults (including serum sickness and anaphylaxis),13 its short half-life (five to seven days),14 and its potential for lifelong sensitization to equine proteins. For these reasons, we created Botulism Immune Globulin Intravenous (Human) (BIG-IV), a human-derived botulism antitoxin that neutralizes botulinum toxin. We evaluated its safety and efficacy in a five-year, randomized, double-blind, placebo-controlled clinical trial statewide, in California, and a subsequent six-year nationwide, open-label study.
Methods
Patients and Eligibility
The study population for the randomized trial consisted of all infants admitted to hospitals in California because of suspected botulism from February 24, 1992, to March 24, 1997. The open-label study population consisted of patients who had suspected infant botulism, initially in California and later nationwide, between the end of the randomized trial and licensure of BIG-IV in October 2003. Follow-up observation continued until all patients had been discharged from the hospital after admission for the management of infant botulism and its complications.
Infants were eligible for the trial if they had acute flaccid paralysis consistent with infant botulism according to the history, physical examination, and laboratory findings at admission and had been hospitalized for less than three days (72 hours). This enrollment limitation reflected concern that any efficacy of BIG-IV might decrease over time as motor-nerve intoxication proceeded.
A laboratory-confirmed case of infant botulism was defined as an illness consistent with the known paralyzing action of botulinum toxin in which C. botulinum toxin or organisms were eventually identified in the patient's feces or enema specimen. In both the randomized trial and the open-label study, enrollment was based on bedside clinical diagnosis; in almost all cases, laboratory confirmation of the diagnosis was not obtained until after infusion of the study medication.
Potential subjects for the randomized trial were identified when a physician, hospital, or local public health department in California contacted the California Department of Health Services (CDHS). The project director or principal investigator would then transport the masked, letter-coded study vial containing either BIG-IV or placebo to the hospital, review the medical record, examine the patient, obtain written informed consent for enrollment from the parents, and administer the contents of the vial. The study protocol was approved by the 62 institutional review boards that represented 90 hospitals in California, the Centers for Disease Control and Prevention (CDC), and the State of California Health and Human Services Agency.
The open-label study protocol was approved by the institutional review boards of the California Health and Human Services Agency and 155 participating hospitals, and written informed consent was obtained from the parents of the patients. Patients in the open-label study were eligible to receive BIG-IV any time up to three weeks after hospital admission.
Study Groups
The therapeutic intervention was a single intravenous infusion of either BIG-IV (50 mg per kilogram of body weight [1 ml per kilogram]) or an identical-appearing, conventional, licensed intravenous immune globulin (Gammagard or Gammagard S/D, Baxter International) that did not neutralize botulinum toxin in the mouse bioassay.1 BIG-IV is a lyophilized powder reconstituted to contain approximately 5 percent human immune globulin with at least 15 IU of neutralizing antibodies against toxin type A and at least 4 IU of neutralizing antibodies against toxin type B per 50 mg. BIG-IV was produced for the CDHS for this study in December 1991 by the Massachusetts Public Health Biologic Laboratories from source plasma collected by the CDHS. Plasma donors were CDHS laboratory workers and colleagues immunized for occupational safety with pentavalent botulinum toxoid (Food and Drug Administration [FDA] Investigational New Drug application 0161) and given a single booster immunization before plasmapheresis.
Randomization
Each patient was randomly assigned by the study statistician to one of the two study groups with the use of a printed random-number table and a master sequential list that was unavailable to the study investigators. The use of a block size of two accomplished an approximately equal allocation of patients with illness caused by type A toxin and those with illness caused by type B toxin to the two groups of the study as a natural consequence of the random occurrence of type A or type B illness during the five years of the clinical trial. Each patient's sequential study-enrollment number was linked to one of eight letter codes stamped on the study vials; this identified the vial to be used in treatment. Serologic studies to determine the half-life of BIG-IV that were performed after the trial was unblinded confirmed that all vials were administered as assigned.
Outcome Measures
The primary safety outcome for all study subjects, regardless of their eventual status with regard to botulism, was the occurrence of adverse events, including possible allergic reactions and blood-borne infections. The primary efficacy outcome was the duration of hospitalization required, as determined by the following standardized criteria for discharge: no further need for inpatient care for infant botulism or its complications, no need for mechanical ventilation or supplemental oxygen for at least three days, no worsening of paralysis in the previous three days and a demonstrated improvement in motor and bulbar function, and three days of intake by tube feeding of 25 percent or less of maintenance volume and calories, with the remainder consumed by mouth.
Secondary efficacy outcomes included the durations of intensive care, mechanical ventilation, and tube feeding, the number of adverse events, and the total charges for the hospital stay. Before the study was unblinded, adverse events and relevant dates were verified by means of hospital and clinic notes and confirmed by parents, who were informed after unblinding of the treatment allocation of their child.
We designed the study to enroll 120 patients with laboratory-confirmed infant botulism, the projected number of cases in a three-year period (the maximum length of an initial FDA funding award for the study of orphan drugs). However, the trial took five years to complete because of lower-than-expected incidence and a temporary suspension that followed the recall of unrelated lots of the placebo.15 This sample size provided 80 percent power at a 5 percent significance level to detect a 42 percent decrease in the mean length of the hospital stay.
Hospital charges were used as a surrogate for the cost of the illness. They do not include the fees of the attending physicians, unless these were billed through the hospital; the costs of transferring the patient by ambulance and, in the open-label study, occasionally by aircraft between states; or indirect costs to parents such as lost work time and hotel bills. Hospital charges were adjusted annually into current-year dollars according to the lowest percentage increase in medical costs in the previous year (obtained from the Web site of the U.S. Bureau of Labor Statistics, http://data.bls.gov/cgi-bin/srgate) in the San Francisco metropolitan area, the Los Angeles metropolitan area, or all cities in the United States, and also, in the open-label study, in the New York–New Jersey metropolitan area or the Philadelphia–New Jersey metropolitan area.
Laboratory Testing
Fecal and enema samples were tested by established methods for identifying and typing C. botulinum toxin and organisms.1 To determine the half-life of BIG-IV, serum samples obtained from patients in the randomized trial before infusion and after infusion at regular intervals for a period of 28 weeks were measured for levels of IgG against botulinum toxin type A and type B. We developed at the CDC a double-sandwich enzyme-linked immunosorbent assay that used purified botulinum toxin type A or type B as the capture antigen, in order to measure the sequential anti–A toxin antibody concentrations in the patients with type B illness and the sequential anti–B toxin antibody concentrations in the patients with type A illness. The results of serologic testing were withheld from the CDHS investigators until the study was unblinded.
Statistical Analysis
Outcome measures for the randomized trial were analyzed by the nonparametric Kolmogorov–Smirnov test16 (requested by the FDA after the conclusion of enrollment) and by the t-test (preplanned), as permitted on the basis of the relatively large sample sizes (
20 patients). Statistical analyses with the use of Stata software, version 8.0 (StataCorp), were based on the intention-to-treat population as initially assigned. Seven patients without laboratory-confirmed infant botulism were excluded prospectively from the efficacy analysis but were included in the safety analysis. A single, prespecified interim analysis was performed after 60 patients had been enrolled and was adjusted for in the final analysis. Linear regression techniques were used to adjust for the random differences between the placebo and BIG-IV groups.
Results
Randomized Clinical Trial
A total of 180 patients were evaluated for eligibility, and 129 were enrolled in the clinical trial (Figure 2). Of these, 122 participants had laboratory-confirmed infant botulism and could be evaluated for all outcome measures. Sixty-three patients received the placebo immune globulin, and 59 patients received BIG-IV. The baseline characteristics of the two groups were similar, except that infants in the BIG-IV group were older and weighed more than infants in the control group. The proportions of patients with type A illness and patients with type B illness in the two groups were similar (Table 1). Before the study was unblinded, determination of the length of the hospital stay on the basis of prespecified discharge criteria resulted in the exclusion of 18 of 3603 hospital days (0.5 percent; 12 in the placebo group and 6 in the BIG-IV group).
|
|
|
The safety analysis included 129 infants. No patient had bronchospasm or an anaphylactic reaction from the study infusion. The only adverse event perhaps related to BIG-IV was a transient, blush-like rash that also occurred in patients with untreated infant botulism (Table 3). The frequency of new urinary tract infections and anemia requiring therapy was lower in the BIG-IV group than in the control group (P<0.05) (Table 3). One death occurred five months after treatment with placebo in a patient whose weakness was eventually diagnosed as resulting from spinal muscular atrophy (Werdnig–Hoffmann disease).
|
Open-Label Study
BIG-IV was given on an open-label basis to 382 patients with laboratory-confirmed infant botulism in 128 hospitals located in 37 states. For the 366 patients treated within 7 days of hospital admission, the mean length of hospital stay was 2.2 weeks (2.5 weeks for 146 patients with type A illness, and 2.1 weeks for 220 patients with type B illness) (Table 4). BIG-IV treatment given within 3 days of hospital admission (to 287 patients) shortened the mean length of the hospital stay to 2.0 weeks, which was significantly shorter than the mean stay of 2.9 weeks when treatment was given 4 to 7 days after hospital admission (to 79 patients; P<0.001) (Table 4). Mean hospital charges per patient for the 366 patients were $57,900, and a total of 20.3 years of hospital stay and $34.2 million in hospital charges were avoided through open-label use of BIG-IV (Table 4).
|
Discussion
Treatment of patients with infant botulism type A or type B with BIG-IV in a five-year, randomized, controlled clinical trial reduced the mean length of the hospital stay for all patients by 3.1 weeks. The mean length of stay in the intensive care unit, the mean duration of mechanical ventilation, the mean duration of tube or intravenous feeding, and the mean hospital charges per patient were also significantly reduced in infants treated with BIG-IV. The only adverse effect perhaps related to treatment with BIG-IV was a transient, blush-like erythematous rash. No serious adverse events occurred more commonly in infants treated with BIG-IV than in those treated with placebo.
The reduction in the mean length of the hospital stay in patients treated with BIG-IV and the safety attributes of BIG-IV in the randomized trial were confirmed in the subsequent six-year nationwide, open-label study. In that study, treatment with BIG-IV given within 3 days after hospital admission shortened the mean length of the hospital stay by approximately 1 week more than did treatment with BIG-IV given 4 to 7 days after admission (2.0 vs. 2.9 weeks). The total length of hospital stays and total hospital charges avoided during the nationwide open-label study were 20.3 years and $34.2 million (in 2004 U.S. dollars), respectively.
Keen clinical awareness may be needed to recognize this rare disease, since infant botulism was suspected at admission in only half the infants enrolled in the randomized trial. Also, infant botulism may present as, or quickly may become, a life-threatening illness. Almost three quarters of the infants in the randomized trial needed intensive care, approximately half of them needed mechanical ventilation, and many had serious complications. However, treatment with BIG-IV did not always result in a shortened hospital stay. One patient with type A illness who received BIG-IV treatment acquired a nosocomial C. difficile infection that was complicated by pseudomembranous colitis, toxic megacolon, and shock; this patient was hospitalized for 12.4 weeks.19 Acquired C. difficile complicating infant botulism may occur more frequently than is generally appreciated.20 One patient with type B illness who received BIG-IV treatment required surgical repair of tracheal stenosis that followed endotracheal intubation; her several hospital stays totaled 15.4 weeks.
BIG-IV has a half-life of approximately 28 days in vivo and a large capacity to neutralize botulinum toxin. A single infusion will neutralize for at least six months all botulinum toxin that may be absorbed from the colon of an infant. This feature is particularly important because it renders moot the concern that antibiotics used to treat secondary bacterial infections may lyse C. botulinum vegetative cells in the intestinal lumen and thereby increase the amount of botulinum toxin available for absorption.21
Although commercial equine botulinum antitoxin has been available in the United States since 1940,22 its efficacy has never been evaluated in a controlled trial. In retrospective and observational studies, early administration of equine antitoxin to adult patients with foodborne and wound botulism was associated with improved outcomes.23,24 However, approximately 6 percent of adults with foodborne botulism had anaphylaxis or serum sickness when treated with one or two vials of equine botulism antitoxin13; such treatment would not be suitable for infants. Because botulinum toxin is now categorized as a category A (maximum threat) bioweapon,25 a larger supply of fully human-compatible botulism antitoxin than can be obtained by plasmapheresis is needed,26 and a suitable recombinant product is under development.27
We conclude that BIG-IV, now licensed by the FDA to CDHS as BabyBIG, is a safe and effective treatment for infant botulism type A and type B. Treatment should be given as soon as possible after hospital admission and should not be delayed for confirmatory testing of feces or enema. BabyBIG is available as a public-service orphan drug in the United States. (Information on this drug may be obtained at www.infantbotulism.org and by telephone from the CDHS Infant Botulism Treatment and Prevention Program at 510-231-7600.)
Supported by a cooperative agreement (FD-U-000476) between the FDA Office of Orphan Products Development and the CDHS and by the CDHS.
Dr. Arnon reports having received salary support from the CDHS. Dr. Schechter reports having received salary support from the CDHS and from a cooperative agreement between the FDA Office of Orphan Products and the CDHS. No other potential conflict of interest relevant to this article was reported.
We are indebted to the parents and children who participated in these studies and to the volunteer donors of botulism hyperimmune plasma who made the creation of BabyBIG possible; to the participating investigators in California and other states who represented the study to their hospital colleagues and institutional review boards; to the data and safety monitoring committee (William Silverman [chair], Parvin Azimi, John Bastian, Philip Brunell, and Nicholas Jewell); to Alan Hubbard for additional statistical expertise; to Edward Gomperts for arranging for Baxter Health Care's donation of Gammagard placebo; to Eric Johnson, University of Wisconsin–Madison, for donating purified botulinum toxin types A and B; to Karen Adler, Christine Ashley, Lynn Bruni, Ellen Doyle, Terry Floyd, Teresa Frankovich, Ingrid Friberg, Somayeh Honarmand, Lorena Juarez, Jessica Khouri, Edith Lee, Karen Mena, Melissa Pedlar, Mary Rodriguez, Patricia Schwarz, Tania Tang, and Claudine Woo for research assistance and support at the CDHS; to Thaddeus Midura, Jason Barash, Raymond Bryant, and James McGee at the CDHS and Loretta McCroskey and Janet Dykes at the CDC, for laboratory diagnostic studies; to Mary Alleman, Ketino Kobaidze, Rebekah Starner, Michelle Claggett, and colleagues for assistance with the enzyme-linked immunosorbent assays; to Jeanne Leszczynski, Catherine Hay, George Siber, and colleagues for manufacturing BIG-IV; to Ted Murphy, Mara Casar, James Cahill, Jason Horn, David Shoemaker, Lynda Sutton, Allen Cato, and colleagues for regulatory expertise; and to David Clissold and Frank Sasinowski for legal expertise.
Source Information
From the Infant Botulism Treatment and Prevention Program (S.S.A., R.S.), California Department of Health Services, Richmond; the National Botulism Surveillance and Reference Laboratory, Centers for Disease Control and Prevention, Atlanta (S.E.M., C.L.H.); and the Division of Biostatistics, School of Public Health, University of California, Berkeley (N.P.J.).
Dr. Hatheway is deceased.
Address reprint requests to Dr. Arnon at the California Department of Health Services, 850 Marina Bay Pkwy., Rm. E-361, Richmond, CA 94804, or at sarnon{at}dhs.ca.gov.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
