Previous studies suggested that caspofungin dose escalation against Candida species is more beneficial than currently used lower daily doses. Thus, we determined in vitro and in vivo activity of caspofungin against six wild-type C. albicans clinical isolates, the ATCC 10231 strain and an echinocandin resistant strain. MIC ranges of clinical isolates in RPMI-1640 with and without 50% serum were 0.125-0.25 and 0.015-0.06 mg/L, respectively. Two and three isolates showed paradoxical growth in MIC and time-kill tests, respectively, in RPMI-1640 but not in 50% serum. Caspofungin killing rate (k) in RPMI-1640 at 1 mg/L was higher than at 16 and 32 mg/L for all isolates (p<0.001). Killing rates for five of six isolates were concentration independent between 1-32 mg/L in 50% serum (p>0.05 for all comparisons), but for one isolate k value at 32 mg/L was significantly lower than at 1-16 mg/L. Although k values at 1-32 mg/L showed a great variability in 50% serum (the lowest and highest k value ranges were 0.085-0.109 and 0.882-0.985 1/h, respectively), daily 3, 5 and 15 mg/kg caspofungin was effective in a neutropenic murine model against all isolates, without significant differences between the effective doses. This study confirms that paradoxical growth does not affect the in vivo efficacy of caspofungin. We demonstrated that dose escalation did not increase the efficacy of caspofungin against C. albicans either in vitro or in vivo. These results are in concordance with the clinical experience that efficacy of echinocandins does not increase at larger doses.