Skip to main content

Table 3 Summary of extracted data for the 97 included articles. The denominator for computing percentages (given to 1 decimal place) is 97 unless stated otherwise

From: Innovative trial approaches in immune-mediated inflammatory diseases: current use and future potential

Question

n (%)

What immune-mediated inflammatory disease(s) was the trial conducted in?a

 Rheumatoid arthritis

30 (30.9)

 Systemic lupus erythematosus

10 (10.3)

 Psoriasis

9 (9.3)

 Psoriatic arthritis

9 (9.3)

 Juvenile idiopathic arthritis

4 (4.1)

 Multiple sclerosis

4 (4.1)

 Sjögren’s syndrome

3 (3.1)

 Systemic sclerosis

3 (3.1)

 Other (see Supplementary Materials)

25 (25.8)

How many treatment arms were in the trial?

 1

5 (5.2)

 2

63 (64.9)

 3

13 (13.4)

 4

11 (11.3)

 5

2 (2.1)

 6

3 (3.1)

What was the total planned sample size according to the sample size calculation?

Median: 214

IQR: [113, 400]

Range: [10, 5400]

How was the trial funded?

 Industry

73 (75.3)

 Academic

16 (16.5)

 Mixed

4 (4.1)

 Not reported

3 (3.1)

 Charity

1 (1.0)

Was any innovative design used?b

 Yes

19 (19.6)

 Group-sequential design/futility interim analysis

7 (7.2)

 Sequential multiple assignment randomised trial design

6 (6.2)

 Bayesian methods used

4 (4.1)

 Sample size re-estimation

2 (2.1)

 Basket trial design

1 (1.0)

Use of an innovative design by trial funding

 Industry

16/73 (21.9)

 Other

3/24 (12.5)

Did the trial design report the involvement of any evidence from routinely collected data, cohorts, or biobanks?

 Yes

2 (2.1)

What was the length of patient recruitment (in weeks)?c

Median: 96

IQR: [55, 120]

Range: [16, 296]

What was the primary endpoint timepoint (in weeks)?d

Median: 24

IQR: [12, 48]

Range: [4, 240]

Did the exclusion criteria explicitly include the presence of another autoimmune disease?e

 Yes

58 (59.8)

Were any endpoints based on dichotomizing continuous information?

 Primary

66 (68.0)

 Secondary

81 (83.5)

How were dichotomized responder endpoints analysed?f

 Cochrane-Mantel-Haenszel test

27 (27.8)

 Logistic regression

21 (21.6)

 Chi-square test

9 (9.3)

 Cox model

9 (9.3)

 Fisher’s exact test

9 (9.3)

 Log rank

5 (5.2)

 Other

16 (16.5)

Any high-dimensional data collected at baseline (gene expression, GWAS, synovial biopsies etc.)?g

 Yes

8 (8.2)

  1. aA small number of articles included patients with more than one IMID in their trial, though none for the diseases named here
  2. bOne article used a sequential multiple assignment randomised trial design with an interim futility assessment
  3. cOne article did not report the recruitment period. To translate recruitment periods given inmonths to weeks, 4 weeks was taken to be equivalent to 1 month
  4. dFor two, one, and one article respectively the median, mean, and maximum follow-up times are used. To translate recruitment periods given in days and months to weeks, 30 days and 1 month were taken to be equivalent to 4 weeks
  5. eOne article reported “any serious illness” as an exclusion criteria and is considered as ‘No’ for the extraction
  6. fArticles may have utilised more than one method
  7. gThree articles that collected MRI imaging data, for which it was unclear as to whether this was high-dimensional, are listed as ‘No’ for the extraction