Treatment with Monoclonal Antibodies against ...

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

c or r e sp ondence

Treatment with Monoclonal Antibodies against Clostridium difficile Toxins To the Editor: The findings by Lowy and col leagues (Jan. 21 issue)1 demonstrate the value of fully human monoclonal antibodies in prevent ing the recurrence of Clostridium difficile infection. The prevalence of C. difficile infection in hospital ized patients is greatest among patients older than 85 years of age,2 and recurrent C. difficile infection is both more frequent and more problematic in older adults.3 Although the oldest patient in the trial con ducted by Lowy et al. was 101 years old, the mean age was 64 years. We recently reported that the natural history of C. difficile infection in frail, elder ly patients is different from that in younger pa tients. In randomized trials of metronidazole and vancomycin, the “cure” rate among patients (mean age, approximately 60 years) was typically higher than 80% within the first week after treatment; in contrast, at least half the elderly patients with acquired C. difficile infection in our geriatrics department (mean age, 86 years) re mained symptomatic at 2 weeks.4 Consequently, we are not satisfied that it will be valid to ex trapolate the results from the study by Lowy et al. to the patients at greatest risk for C. difficile infec tion. Issues regarding the inclusion of elderly patients in clinical trials have been raised previ ously.5 Specific evaluation of antibiotic and bio this week’s letters 1444 Treatment with Monoclonal Antibodies against Clostridium difficile Toxins 1446 Genomewide Association Study of Leprosy 1448 Barrett’s Esophagus 1449 Williams–Beuren Syndrome 1444

logic agents for C. difficile infection in older adults is needed. Tom Parks, M.B., B.Chir. Oxford Radcliffe Hospitals Oxford, United Kingdom [email protected]

Effrossyni Gkrania-Klotsas, M.D., M.P.H. Claire Nicholl, M.B., B.S. Cambridge University Hospitals Cambridge, United Kingdom No potential conflict of interest relevant to this letter was re ported. 1. Lowy I, Molrine DC, Leav BA, et al. Treatment with mono

clonal antibodies against Clostridium difficile toxins. N Engl J Med 2010;362:197-205. 2. Smyth ET, McIlvenny G, Enstone JE, et al. Four country health care associated infection prevalence survey 2006: overview of the results. J Hosp Infect 2008;69:230-48. 3. Kyne L. Clostridium difficile — beyond antibiotics. N Engl J Med 2010;362:264-5. 4. Parks T, Wallis S, Wilson J, Gkrania-Klotsas E, Nicholl C. Continuing diarrhoea after ten days of oral metronidazole or oral vancomycin for presumed, hospital-acquired Clostridium difficile colitis in elderly hospital patients. J Hosp Infect 2010; 74:403-5. 5. Siu LL. Clinical trials in the elderly — a concept comes of age. N Engl J Med 2007;356:1575-6.

To the Editor: The treatment of C. difficile diar rhea has increasingly become a challenge.1 Neu tralizing antitoxins prevent deaths from C. difficile diarrhea in animals.2 Lowy et al. report a reduced recurrence of C. difficile diarrhea after the admin istration of neutralizing, fully human monoclo nal antibodies against C. difficile toxins A (CDA1) and B (CDB1) in addition to standard antimicro bial therapy. However, this adjuvant antibody therapy did not improve the severity of diarrheal illness, the duration of hospitalization, or the time to resolution of the diarrhea. Mucosal inflammation mediated by C. difficile toxin A and loss of epithelial barrier function

n engl j med 362;15 nejm.org april 15, 2010

Downloaded from www.nejm.org by TOM PARKS on June 28, 2010 . Copyright © 2010 Massachusetts Medical Society. All rights reserved.

correspondence

requires application of the toxin to the apical cell surface.3 Serum immunoglobulins were not expected to alter the function of the intralumi nal toxin. The mechanism for the reduced recur rence of C. difficile diarrhea mediated by CDA1– CDB1 is unclear; a plausible explanation may involve an ancillary influence on the intestinal cytokine milieu. T-cell–derived interferon-γ orches trates infiltration of neutrophils into the tissue, epithelial disruption in the intestines, and highvolume fluid secretion after exposure to toxin A.4 Low levels of interferon-γ mitigate toxin A–induced expression of the tumor necrosis factor α gene, neutrophil-chemotactic chemokines (e.g., macro phage inflammatory protein 1α), intercellular adhesion molecule 1, and other mediators of inflammation.4 Neutralizing serum CDA1–CDB1 may negate the intestinal type 1 helper T-cell immune response during subsequent infections. Amar Safdar, M.D., M.B., B.S. M.D. Anderson Cancer Center Houston, TX [email protected] No potential conflict of interest relevant to this letter was re ported. 1. Loo VG, Poirier L, Miller MA, et al. A predominantly clonal

multi-institutional outbreak of Clostridium difficile–associated diar rhea with high morbidity and mortality. N Engl J Med 2005; 353:2442-9. [Erratum, N Engl J Med 2006;354:2200.] 2. Babcock GJ, Broering TJ, Hernandez HJ, et al. Human mono clonal antibodies directed against toxins A and B prevent Clos tridium difficile-induced mortality in hamsters. Infect Immun 2006;74:6339-47. 3. Sutton PA, Li S, Webb J, Solomon K, Brazier J, Mahida YR. Essential role of toxin A in C. difficile 027 and reference strain supernatant-mediated disruption of Caco-2 intestinal epithelial barrier function. Clin Exp Immunol 2008;153:439-47. 4. Ishida Y, Maegawa T, Kondo T, et al. Essential involvement of IFN-gamma in Clostridium difficile toxin A-induced enteritis. J Immunol 2004;172:3018-25.

To the Editor: Lowy et al. report that a single injection of monoclonal antibodies prevented re current C. difficile infections. We were surprised, however, that this acute intervention would alter the long-term relapse rate with no effect on the acute infection (unlike, for example, antibody treatment of botulism infections).1 Indeed, the original protocol registered at ClinicalTrials.gov states that the primary outcome was to describe “the course of resolution of illness . . . plus a number of explanatory analyses.” The protocol was modified later to state that the primary out come was to test “the proportion of subjects with recurrent CDAD [C. difficile–associated diarrhea] . . . at 3 months.” We appreciate the positive role of serendipity in science.2 Is it possible, how

ever, that this trial involved multiple-hypothesis testing and capitalized on chance? If so, the article should clarify why the primary outcome changed. Ainsley E. Dawson, M.D. Steven L. Shumak, M.D. Donald A. Redelmeier, M.D. University of Toronto Toronto, ON, Canada [email protected] No potential conflict of interest relevant to this letter was re ported. 1. Arnon SS, Schechter R, Maslanka SE, Jewell NP, Hatheway

CL. Human botulism immune globulin for the treatment of in fant botulism. N Engl J Med 2006;354:462-71. 2. Redelmeier DA. Multiple attacks from multiple perspectives. J Clin Epidemiol 2006;59:871-2.

The Authors Reply: Parks et al. raise the con cern that the course of C. difficile infection in the elderly is distinct enough that the benefit of monoclonal antibodies for this population war rants further study. In our study, 53 of the 200 patients enrolled (26.5%) were 75 years of age or older and 18.5% were 85 years of age or older at enrollment. Age is an established risk factor for recurrent disease,1 and the median age of patients with recurrent C. difficile infection in our study was 73 years, suggesting the relevance of the study population to persons at high risk. We agree that future studies should be large enough to pro vide a more robust analysis of the effect of age on the recurrence of C. difficile infection in patients treated with monoclonal antibodies in addition to antibiotic agents. In response to Safdar: our data from in vitro cell-culture cytotoxicity assays provide direct evi dence of toxin neutralization by the monoclonal antibodies and loss of cytopathogenicity,2 and they suggest that this is probably the mechanism of protection. Although not evaluated in this study, it has been shown that antibodies admin istered systemically do reach the gut lumen, re sulting in prevention of colonization with certain bacteria and viruses.3,4 We agree with Safdar that the effect of the antibodies on the develop ment of subsequent cellular and humoral immu nity is not known. Finally, Dawson et al. request clarification of the timing of the decision regarding the primary outcome. Recurrence as the primary outcome was decided after discussions with the Food and Drug Administration during the first few months of study enrollment and before any unblinding oc curred.



1445

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

1. Kyne L, Warny M, Qamar A, Kelly CP. Association between

Israel Lowy, M.D., Ph.D. Bristol-Myers Squibb Lawrenceville, NJ

Deborah C. Molrine, M.D., M.P.H. Donna M. Ambrosino, M.D. MassBiologics Boston, MA [email protected] Since publication of their article, Dr. Lowy reports having equity in Bristol-Myers Squibb. No further potential conflict of interest relevant to this letter was reported.

antibody response to toxin A and protection against recurrent Clostridium difficile disease. Lancet 2001;357:189-93. 2. Babcock GJ, Broering TJ, Hernandez HJ, et al. Human mono clonal antibodies directed against toxins A and B prevent Clos tridium difficile-induced mortality in hamsters. Infect Immun 2006;74:6339-47. 3. Westerman LE, McClure HM, Jiang B, Almond JW, Glass RI. Serum IgG mediates mucosal immunity against rotavirus infec tion. Proc Natl Acad Sci U S A 2005;102:7268-73. 4. Santosham M, Reid R, Ambrosino DM, et al. Prevention of Haemophilus influenzae type b infections in high-risk infants treated with bacterial polysaccharide immune globulin. N Engl J Med 1987;317:923-9.

Genomewide Association Study of Leprosy To the Editor: Zhang and colleagues (Dec. 31 issue)1 report that genes in the nucleotide-bind ing oligomerization domain containing 2 (NOD2)– mediated signaling pathway are associated with susceptibility to infection with Mycobacterium leprae in China. India has the world’s greatest lep rosy disease burden. We therefore genotyped the single-nucleotide polymorphisms (SNPs) that were implicated by Zhang and colleagues in two Indian case–control cohorts (492 patients in New Delhi and 382 in Kolkata).2,3 We also genotyped 273 cases and 221 controls from Mali, West Africa.4 We observed associations between disease and SNPs at C13orf31 (the gene encoding chromosome 13 open reading frame 31) (rs3764147, P = 6.1×10−8) and CCDC122 (the gene encoding coiled-coil do main containing 122) (rs9533634, P = 1.1×10−5) (Table 1); both genes were of unknown function. We did not, however, observe associations be tween disease and the other four non–major his tocompatibility complex (MHC) genes related to the NOD2 pathway (NOD2, RIPK2 [the gene encod ing receptor-interacting serine–threonine kinase 2], TNFSF15 [the gene encoding tumor necrosis factor (ligand) superfamily member 15], and LRRK2 [the gene encoding leucine-rich repeat ki

nase 2]) (Table 1 in the Supplementary Appendix, available with the full text of this letter at NEJM. org), despite reasonable power to detect the ef fect sizes observed by Zhang et al.1 An analysis of 27 additional SNPs at NOD2 in the New Delhi cohort provided support for the absence of a con sistent association at this locus (Table 2 in the Supplementary Appendix). These results indicate heterogeneity among populations and suggest that future functional studies should focus on populations in which the relevant genetic asso ciation may occur. Nevertheless, a robust asso ciation of the Crohn’s disease chromosome 13q14.11 locus5 containing C13orf31 and CCDC122 with leprosy in China, India, and Mali provides support for a molecular link between mycobacte rial infection and Crohn’s disease. Sunny H. Wong, M.B., Ch.B. Adrian V.S. Hill, D.M., D.Phil. Fredrik O. Vannberg, D.Phil. Wellcome Trust Centre for Human Genetics Oxford, United Kingdom [email protected]

for the India–Africa–United Kingdom Leprosy Genetics Consortium

Table 1. Associations with Leprosy for Replicated Single-Nucleotide Polymorphisms on Chromosome 13q14.11, According to Case–Control Population.* SNP

Gene

Original Study

Present Study

New Delhi

Kolkata

odds ratio (95% CI)

Mali

Combined Populations

P value −3

rs3764147

C13orf31

1.68 (1.57–1.80)

1.59 (1.34–1.89)

2.7×10

6.4×10−2

1.1×10−5

6.1×10−8

rs9533634

CCDC122

0.76 (0.70–0.82)

0.70 (0.59–0.82)

1.5×10−3

3.6×10−1

1.1×10−5

1.1×10−5

* C13orf31 denotes the gene encoding chromosome 13 open reading frame 31, CCDC122 the gene encoding coiled-coil domain containing 122, CI confidence interval, and SNP single-nucleotide polymorphism.

1446

