3+3 Dose Escalation Design¶
Traditional rule-based dose escalation for Phase I first-in-human studies.
Design¶
- Type: 3+3 dose escalation
- Dose levels: 10, 25, 50, 100, 200 mg
- Cohort size: 3 subjects
- Max DLT rate: 33%
- Max subjects: 30
Escalation Rules¶
| DLTs in Cohort | Action |
|---|---|
| 0/3 | Escalate to next dose |
| 1/3 | Expand cohort (+3) |
| 0-1/6 | Escalate to next dose |
| 2+/3 or 2+/6 | Stop, previous dose is MTD |
Features Demonstrated¶
- Dose escalation design
- Rule-based decisions
- DLT probability modeling
- MTD determination
Files¶
| File | Description |
|---|---|
| python.py | Python implementation |
| julia.jl | Julia implementation |
| cli.json | CLI specification |
Expected Output¶
3+3 Dose Escalation Study
=========================
Dose levels: [10, 25, 50, 100, 200] mg
Starting dose: 10 mg
Max subjects: 30
Cohort 1 (10 mg): 0/3 DLTs -> Escalate
Cohort 2 (25 mg): 0/3 DLTs -> Escalate
Cohort 3 (50 mg): 1/3 DLTs -> Expand
Cohort 4 (50 mg): 0/3 DLTs (1/6 total) -> Escalate
Cohort 5 (100 mg): 2/3 DLTs -> Stop
Recommended MTD: 50 mg
Total subjects: 15