Skip to main content

Table 2 Characteristics of early phase studies and associations with phase III study results

From: Analysis of early phase and subsequent phase III stroke studies of neuroprotectants: outcomes and predictors for success

Study characteristics

All early phase studies (n = 59)

True neutral (n = 37)

False positive (n = 22)

aOR [95%-CI]

P

No. of subjects, median (IQR)

92.0 (46.0-176.0)

100.0 (46.0-157.0)

91.5 (49.3-197.5)

1.00 [0.99-1.00]

0.713

Industry sponsored, n (%)

36 (61.0)

25 (67.6)

11 (50.0)

1.73 [0.40-7.54]

0.465

Multicentered, n (%)

41 (69.5)

28 (75.7)

13 (59.1)

1.48 [0.28-7.87]

0.643

Randomized, n (%)

57 (96.6)

35 (94.6)

22 (100.0)

0.00 [0.00-∞]

1.000

Blinded outcome assessment, n (%)

52 (88.1)

34 (91.9)

18 (81.8)

3.95 [0.26-60.30]

0.323

Dose–response investigated, n (%)

16 (27.1)

12 (32.4)

4 (18.2)

1.25 [0.26-6.07]

0.786

Use of imaging endpoint, n (%)

15 (25.4)

8 (21.6)

7 (31.8)

0.46 [0.12-1.81]

0.263

Duration of follow-up for endpoints*, median (IQR), days

87.0 (30.0-90.0)

90.0* (30.0-90.0)

41.0 (28.0-90.0)

1.00 [0.99-1.01]

0.722

Same characteristics in phase I/II studies as in subsequent phase III studies

     

 Same therapeutic time window, n (%)

19 (32.2)

10 (27.0)

9 (40.9)

0.69 [0.18-2.68]

0.588

 Same dose, n (%)

29 (49.2)

18 (48.6)

11 (50.0)

0.94 [0.28-3.14]

0.916

 Same route of administration, n (%)

50 (84.7)

34 (91.9)

16 (72.7)

3.33 [0.59-18.86]

0.174

  1. IQR, interquartile range, aOR (adjusted odds ratio) for comparison of study characteristics between “false positive” and “true neutral” studies in a multivariable logistic regression analysis, CI, confidence interval, *one study did not report the exact duration of follow-up as endpoints were determined at discharge (mean length of hospital stay in this study was 9.5 days in the verum group and 11.2 days in the placebo group).