Table 2

Baseline, with treatment, and change from baseline in efficacy parameters for randomly assigned patients in the treatment cohort

nBaseline*Week 12 (LOCF)*Change from baseline and differences in change vs. placebo + insulin*
A1C (%)
    Placebo + insulin198.3 ± 0.88.5 ± 0.80.09 (−0.2 to 0.4)
    10 mg dapagliflozin+ insulin238.4 ± 0.77.8 ± 0.7−0.61 (−0.9 to −0.4)
    Difference in change vs. placebo + insulin−0.70 (−1.1 to −0.3)
    20 mg dapagliflozin+ insulin238.5 ± 0.97.8 ± 0.6−0.69 (−0.9 to −0.4)
    Difference in change vs. placebo + insulin−0.78 (−1.2 to −0.4)
FPG (mg/dl)
    Placebo + insulin22166.8 ± 52.6180.8 ± 56.917.8 (1.4 to 34.2)
    10 mg dapagliflozin + insulin23155.7 ± 39.8160.5 ± 38.72.4 (−13.6 to 18.3)
    Difference in change vs. placebo + insulin−15.4 (−38.4 to 7.5)
    20 mg dapagliflozin + insulin23157.9 ± 53.0149.4 ± 32.0−9.6 (−25.6 to 6.3)
    Difference in change vs. placebo + insulin−27.4 (−50.3 to −4.6)
Body weight (kg)
Placebo + insulin22101.3 ± 16.799.4 ± 16.7−1.9 (−2.9 to −0.9)
10 mg dapagliflozin + insulin23102.8 ± 9.998.2 ± 9.4−4.5 (−5.5 to −3.5)
Difference in change vs. placebo + insulin−2.6 (−4.0 to −1.2)
20 mg dapagliflozin + insulin23102.1 ± 15.097.8 ± 14.1−4.3 (−5.3 to −3.3)
Difference in change vs. placebo + insulin−2.4 (−3.8 to −1.0)
TDDI (units)
    Placebo + insulin2254.1 ± 27.355.7 ± 26.51.7 (−3.8 to 7.2)
    10 mg dapagliflozin + insulin2452.4 ± 24.451.3 ± 20.1−1.4 (−6.6 to 3.9)
    Difference in change vs. placebo + insulin−3.1 (−10.7 to 4.6)
    20 mg dapagliflozin + insulin2454.5 ± 36.353.5 ± 32.1−0.8 (−6.1 to 4.5)
    Difference in change vs. placebo + insulin−2.5 (−10.2 to 5.1)
PPG (mg/dl)
    Placebo + insulin15312.6 ± 82.2331.3 ± 46.818.7 (−13.5 to 50.9)
    10 mg dapagliflozin + insulin19320.2 ± 51.4286.0 ± 55.1−34.3 (−67.5 to −1.1)
    20 mg dapagliflozin + insulin18314.5 ± 71.8272.6 ± 51.2−41.9 (−74.8 to −8.9)
  • Data are means ± SD or mean (95% CI). n = 71.

  • *Excludes data after insulin up-titration, except for TDDI.

  • †Adjusted change from baseline based on an ANCOVA model with treatment group as an effect and baseline value as a covariate.

  • ‡Number of subjects with a nonmissing baseline and a week 12 LOCF value.