In the United States, the blaCMY-2 gene contained within incompatibility type A/C (IncA/C) plasmids is frequently identified in extended-spectrum-cephalosporin-resistant (ESC(r)) Escherichia coli strains from both human and cattle sources. Concerns have been raised that therapeutic use of ceftiofur in cattle may increase the prevalence of ESC(r) E. coli. We report that herd ESC(r) E. coli fecal and hide prevalences throughout the residency of cattle at a feedlot, including during the period of greatest ceftiofur use at the feedlot, were either not significantly different (P ≥ 0.05) or significantly less (P < 0.05) than the respective prevalences at arrival. Longitudinal sampling of cattle treated with ceftiofur demonstrated that once the transient increase of ESC(r) E. coli shedding that follows ceftiofur injection abated, ceftiofur-injected cattle were no more likely than untreated members of the same herd to shed ESC(r) E. coli. Pulsed-field gel electrophoresis (PFGE) genotyping, antibiotic resistance phenotyping, screening for presence of the blaCMY-2 gene, and plasmid replicon typing were performed on 312 ESC(r) E. coli isolates obtained during six sampling periods spanning the 10-month residence of cattle at the feedlot. The identification of only 26 unique PFGE genotypes, 12 of which were isolated during multiple sampling periods, suggests that clonal expansion of feedlot-adapted blaCMY-2 E. coli strains contributed more to the persistence of blaCMY-2 than horizontal transfer of IncA/C plasmids between E. coli strains at this feedlot. We conclude that therapeutic use of ceftiofur at this cattle feedlot did not significantly increase the herd prevalence of ESC(r) E. coli.