Table 1.
Characteristics of the clinical trials whose datasets were included
| Metastatic disease17 [endocrinologic oncology] | Esophageal cancer20 [gastrointestinal oncology] | Hereditary hearing loss24 [otolaryngology] | Cardiac Amyloidosis25 [cardiology] (*) | |||||
|---|---|---|---|---|---|---|---|---|
| Training set | Test set | Training set | Test set | Training set | Test set | Training set | Test set | |
| Patient number | ||||||||
| Total [n] (with disease/without disease [%]) |
493 (34/66) |
295 (19/81) |
7899 (3/97) |
6698 (2/98) |
1209 (76/24) |
569 (77/23) |
1712 (50/50) |
430 (50/50) |
| Patient sex | ||||||||
| Female/male [%] | 49/51 | 57/43 | 0/100 | 0/100 | 49/51 | 39/61 | N/A | N/A |
| Patient age [years] | ||||||||
|
Median Mean ± standard deviation Range (minimum, maximum) |
42 42 ± 18 (4, 83) |
48 47 ± 16 (11, 82) |
56 56 ± 9 (39, 82) |
55 56 ± 9 (24, 86) |
N/A 18 ± 15 (N/A, N/A) |
N/A 34 ± 12 (N/A, N/A) |
N/A | N/A |
| Location of clinical trial | US, Netherlands | Germany, Poland, Netherlands | China | China | China | China | US | US |
(*) indicates that the original data split and, consequently, the external validation dataset was unavailable per the original study. In line with the published methodology, we randomly allocated 80% of patients and controls to the training set (n = 1712) and 20% to the test set (n = 430). N/A not available.