Bezlotoxumab (BEZ) for Prevention of Clostridium difficile Infection (CDI) Recurrence (rCDI): Distinguishing Relapse from Reinfection with Whole Genome Sequencing (WGS)

BACKGROUND: Bezlotoxumab (BEZ) and actoxumab (ACT) are monoclonal antibodies against C. difficile toxins B and A, respectively. Patients receiving a single infusion of BEZ alone or with ACT in the MODIFY I/II trials showed an absolute 10% (relative ~40%) reduction in rCDI over 12-weeks compared with placebo (PBO). The addition of ACT did not improve efficacy. This post hoc analysis investigated whether BEZ prevented relapse with the same strain and/or reinfection with a new strain. METHODS: C. difficile strains isolated from patient stool samples were typed by PCR ribotyping, PCR free library construction and Illumina whole genome sequencing (WGS). rCDI was defined as diarrhea with toxigenic C. difficile in stool. Reinfection and relapse were differentiated by comparing ribotype (RT) and pair-wise single-nucleotide WGS variations (PWSNV). Relapse was assigned if the baseline RT and the RT isolated during rCDI were the same and PWSNVs were ≤2. Reinfection was defined as rCDI cases with a different RT compared with baseline or the same RT with >10 PWSNVs. Patients receiving BEZ or ACT+BEZ were pooled and patients receiving PBO or ACT were pooled. The effect of BEZ on the cumulative incidence of relapse and reinfection was estimated by Fine & Gray’s competing risks survival model. RESULTS: Among 514 patients with rCDI in MODIFY I/II, 259 (50.4%) had a baseline and a post-baseline C. difficile isolate. There were 198 (76.4%) relapse and 50 (19.3%) reinfection cases (Table). Among rCDI cases, proportions of reinfection and relapse were similar between treatments. Proportion of relapses was higher for RT 027. Significant differences in crude cumulative incidence for relapse (P < 0.001) were observed for BEZ and ACT+BEZ groups compared with PBO and ACT groups. Similar changes were observed for reinfection but results were not significant. Cumulative incidence curves showed that relapses occurred earlier and at a higher rate than reinfections, but the reduction in rCDI was similar (Figure). CONCLUSION: The BEZ-induced reduction in rCDI observed in MODIFY I/II reflects the prevention of relapses due to the same strain. A reduction in reinfections was also observed, but likely due to a smaller number of reinfection cases, the difference was not significant. DISCLOSURES: M. B. Dorr, Merck & Co., Inc.: Employee and Shareholder, may own stock/hold stock options in the Company; Z. Zeng, Merck & Co., Inc.: Employee, May own stock/hold stock options in the Company ; M. Wilcox, Merck & Co., Inc.: Consultant, Consulting fee; Cubist: Consultant, Grant Investigator and Speaker’s Bureau, Consulting fee, Grant recipient and Speaker honorarium; Alere, Actelion Pharma, Astellas, Optimer, sanofi pasteur, Summit Pharma, bioMerieux, Da Volterra, Qiagen, Cerexa, Abbott, AstraZeneca, Pfizer, Durata Therapeutics, Seres Therapeutics, Valneva, Nabriva Therapeutics, Roche, The Medicines Company, Basilea P: Consultant, Consulting fee; Alere, Actelion Pharmaceuticals, Pharmaceuticals, Astellas, Optimer Pharmaceuticals, sanofi pasteur, Summit Pharmaceuticals, bioMerieux, Da Volterra, Qiagen, Cerexa, and Abbott: Grant Investigator, Grant recipient; J. Li, BGI-Shenzhen: Employee, Salary; H. Zhao, BGI-Shenzhen: Employee, Salary; X. Li, BGI-Shenzhen: Employee, Salary; D. Guris, Merck & Co., Inc.: Employee, may own stock/hold stock options in the Company; P. Shaw, Merck & Co., Inc.: Employee, May own stock/hold stock options in Company