Decision Tree — Drug Development Stage Gates (Phase II to Approval)
A pharmaceutical stage-gate tree with costs paid as each phase starts: funding Phase II rolls back to $70.7M against $35.0M for out-licensing now, with a Phase III go decision on a positive readout — and a risk profile showing the 83% chance the programme loses money. Illustrative values.
Make it your own.
title "Phase II oncology asset — fund in-house, out-license or stop?"
subtitle "Risk-adjusted NPV, US$ millions; each stage's cost is paid when it starts"
currency $
unit M
decimals 1
risk-profile on
note "Stage success rates are of the order of published oncology benchmarks: Phase II about 1 in 3, Phase III just over 1 in 2, approval about 9 in 10."
note "Illustrative sales NPVs for a fictional asset."
decision "Development strategy"
"Fund Phase II in-house" cost 45 -> chance "Phase II readout"
"Positive" p 0.32
decision "Phase III go / no-go"
"Run Phase III" cost 180 -> chance "Phase III readout"
"Meets primary endpoint" p 0.58 -> chance "Regulatory review"
"Approved" p 0.90 -> chance "Peak-sales scenario"
"Best-in-class uptake" p 0.25 -> 2,100
"Base case" p 0.50 -> 900
"Crowded market" p rest -> 250
"Complete response letter" p rest -> 0
"Misses primary endpoint" p rest -> 0
"Out-license after Phase II" -> 220
"Negative" p rest -> 0
"Out-license the asset now" -> 35
"Stop development" -> 0