FDA Drug Approval Rates by Indication
The actual probability data behind the Submarine Catalyst PoA model — Phase 3 to approval rates by therapeutic area, sourced from peer-reviewed literature and updated with real outcomes.
Why These Numbers Matter
When a biotech stock trades before a PDUFA date, the stock price already reflects the market's estimate of approval probability. If the market is pricing in 90% odds and the actual historical rate for that indication is 49%, someone is wrong — and that mispricing is where the edge comes from.
The Submarine Catalyst probability of approval (PoA) model starts with these indication-specific base rates, then applies seven additional signal layers: review pathway, AdCom vote, CMC risk, resubmission history, special designations, clinical trial status, and insider activity. The base rate is the foundation everything else is built on.
Source Data
The base rates below are sourced from two primary studies and updated with Submarine Catalyst's own outcome tracking:
Hay et al. 2014 — "Clinical development success rates for investigational drugs." Nature Biotechnology. Analyzed 835 companies and 4,451 drugs across FDA phases from 2003–2011. The most cited source for Phase 3 to approval rates by indication.
Wong et al. 2019 — "Estimation of clinical trial success rates and related parameters." Biostatistics. Updated the Hay analysis with a larger dataset (2000–2015), introducing important corrections for indication-level variance.
Both studies measure the rate at which drugs that enter Phase 3 ultimately reach FDA approval. This is the closest proxy for "what is the probability this drug gets approved at the PDUFA date" given that most PDUFA events involve drugs that have completed Phase 3 trials.
FDA Approval Rates by Therapeutic Area (Phase 3 → Approval)
| Indication / Area | Approval Rate | Rate | Notes |
|---|---|---|---|
| Hematology / Blood | High | Strong endpoint clarity; unmet need historically high | |
| Oncology / Cancer | High | Accelerated approval pathways available; surrogate endpoints accepted | |
| Rare / Orphan Disease (general) | High | High unmet need; smaller trial sizes accepted; PRV incentive | |
| Gene Therapy | High | Clinical rates high but CMC risk is elevated; manufacturing complexity | |
| Metabolic / Endocrine | High | Diabetes, obesity, thyroid — well-defined endpoints | |
| Gastroenterology | High | IBD, gastroparesis — strong need; endpoints evolving | |
| Infectious Disease | High | HIV, hepatitis, antibiotics — clear endpoints; FDA prioritizes | |
| Respiratory | High | COPD, asthma — large datasets; pulmonary endpoints well-defined | |
| Dermatology | High | Psoriasis, atopic dermatitis — objective scoring systems | |
| Ophthalmology | High | Retinal disease, macular degeneration — measurable BCVA endpoints | |
| Immunology / Autoimmune | High | RA, lupus, Crohn's — endpoints established by precedent | |
| Cardiovascular | Mid | Large CV outcome trials required; high bar for mortality benefit | |
| Musculoskeletal | Mid | Osteoporosis, arthritis — functional endpoints sometimes contested | |
| Neuromuscular (DMD, ALS, SMA) | Mid | Complex endpoints; CMC risk elevated for biologics; FDA has high bar | |
| Psychiatry / Mental Health | Mid | Schizophrenia, depression — subjective endpoints; placebo effect high | |
| Neurology / CNS (general) | Low | Alzheimer's, Parkinson's, epilepsy — hardest endpoints in medicine | |
| Pain | Low | Opioid epidemic policy scrutiny; abuse potential concerns; subjective endpoints | |
| Duchenne MD (DMD) specifically | Low | Multiple prior CRLs in space; biomarker endpoints contested; CMC complexity | |
| ALS specifically | Low | Fastest-progressing terminal disease; functional endpoints difficult to demonstrate |
Sources: Hay et al. 2014 (Nature Biotechnology), Wong et al. 2019 (Biostatistics), FDA PDUFA VII performance reports, Submarine Catalyst outcome tracking. DMD/ALS rates reflect category-specific analysis added to the model in April 2026.
How Review Pathway Modifies the Base Rate
The indication base rate is only half the input. The FDA review pathway the drug received also significantly affects approval probability. The base rates above assume standard review. These pathway multipliers are blended 50/50 with the indication rate in Submarine Catalyst's PoA model:
| Review Pathway | Base Probability | Why |
|---|---|---|
| Breakthrough Therapy | 90% | Highest FDA confidence, intensive guidance |
| Priority Review (NME) | 81% | FDA priority; usually significant unmet need |
| sNDA / Supplement | 80% | Drug already approved; incremental label change |
| Standard Review | 75% | No special status; standard 10-month review |
| Accelerated Approval | 72% | Surrogate endpoint; confirmatory trial pending |
| Post-CRL Resubmission | 65% | Varies by CRL type; see CRL page for breakdown |
The REPL Case Study — When Resubmission Risk Was Underweighted
RP1 (vusolimogene oderparepvec) + nivolumab for advanced melanoma had strong surface-level inputs: oncology base rate (85.5%), Breakthrough Therapy designation (90%), and a 34% ORR with 24.8-month median DOR in the IGNYTE trial. The model output was 84% PoA. But REPL was a BLA resubmission after a prior CRL (July 2025) where FDA had raised the same fundamental objection — that the single-arm IGNYTE trial could not establish RP1's contribution of effect in the combination. The resubmission pathway modifier (65%) was applied, but the model did not adequately penalize for a second submission facing the same unresolved trial design concern.
The fix: Submarine Catalyst's scoring model now applies a compounding penalty when a resubmission faces the same class of FDA objection as the prior CRL. A second CRL for efficacy/trial design concerns is fundamentally different from a first CRL for CMC — the resubmission modifier alone doesn't capture that distinction. The model also now weights single-arm combination studies more conservatively when contribution-of-effect has been challenged.
What This Means for Evaluating Biotech Stocks
When a biotech trades at a stock price that implies 85% approval odds but the drug is in CNS (historically 49%) with no special designations, that's a setup where the market is significantly overpricing the approval probability. The trade isn't just about whether the drug gets approved — it's about whether the market has priced approval correctly.
Conversely, a drug in oncology with breakthrough designation and a depressed stock price (post-CRL resubmission, for example) might be priced as if FDA approval is 50/50 when the historical rate for that setup is closer to 85%. That's where the real edge is.
The Submarine Catalyst scanner runs these calculations automatically for every active PDUFA event — blending indication rate, pathway rate, AdCom vote, CMC signals, and six other layers into a single PoA score. The data on this page is the foundation.
See how these rates are applied: How Institutions Trade PDUFA Events →
CRL rates by type: Complete Response Letter Guide →
Full track record with actual outcomes: Track Record →
Access the Scanner — $29.99/monthPDUFA scanner · PoA scores · STN Risk · Dilution Radar · 870+ companies