Table 1.
Main characteristics of the public databases of predictive biomarkers
| GDKD | CIViC | TARGET | ||
|---|---|---|---|---|
| Number of genes | 170 | 290 | 135 | |
| Number of predictive genes | Total | 170 | 213 | 111 |
| Exclusive | 46 | 105 | 10 | |
| Common | 79 | |||
| Number of variant–drug associations | 618 | 1931 | 111 | |
| Number of cancer types | 65 | 177 | Not specified | |
| Biomarker types | Predictive | Predictive | Predictive | |
| Prognostic | Prognostic | |||
| Diagnostic | Diagnostic | |||
| Clinical significance levels | Response | Sensitivity | Free text | |
| Sensitivity | Resistance or non-response | |||
| Increased benefit | ||||
| No response | ||||
| No sensitivity | ||||
| Reduced/decreased sensitivity | ||||
| Resistance | ||||
| Evidence levels | NCCN/FDA | A: clinical routine | None | |
| Late trials | B: clinical trials | |||
| Early trials | C: case reports | |||
| Case report | D: preclinical | |||
| Preclinical | E: inferential | |||
| Variant specific | Yes | Yes | No | |
| References provided | Yes | Yes | No | |
| Version | v19 | 1 June 2017 | v3 | |
| Source | [20] | [21] | [12] | |
Each column summarizes the specificities of each database: GDKD Gene Drug Knowledge Database, CIViC Clinical Interpretation of Variants in Cancer, TARGET Tumor Alterations Relevant for Genomics-driven Therapy