FDA approves 75 novel drugs in a year

Where things stand

• 2026 so far. FDA's 2026 table lists 46 novel approvals through Oct 9, 2026 (last entry: Hovilpri, 10/9). That is the fastest pace on record. By quarter, 2026 has run 10 / 13 / 22 / 1-so-far. The 22 in Q3 is the most ever in one quarter; the earlier peaks were 21 in Q3 2018 and 21 in Q4 2019.
• Past years. Full-year totals from FDA's year pages (2021–2026, plus archived 2018–2020) were 59, 48, 53, 50, 37, 55, 50 and 46 for 2018–2025. The record before 2026 was 59 (2018). The 2017–2025 counts average about 49, with a log standard deviation of 0.13.
• 2026 cannot plausibly reach 75. It would need 29 more approvals in about 12 weeks. No quarter has ever had more than 22, and the public decision calendar lists only about 18 novel CDER candidates due in Q4. I put this at about 0.2%. The likely 2026 total is roughly 60–65, a new record.

Reference class and base rate

• Annual counts. CDER novel approvals have sat on a plateau of about 50 a year since 2017. Year-to-year noise is about 13% in log terms.
• Past jumps. Annual counts reached 1.19× or more of the best of the prior five years in 3 of about 30 years (1996, 2012, 2018). All three came in growth phases or after a backlog was cleared.
• What 75 means. It is about 27% above the record before 2026 and about 20% above where 2026 is likely to finish. Today's higher counts have lower relative noise than the small counts of the 1990s, so I discount that past jump rate.
• Starting rate. This gives a per-year chance of a few percent before looking at the specifics of this case.

Evidence on the level for 2027–2031

Pointing up:

• FDA-TRACK PDUFA data (as of 6/30/2026) show 67 new-drug and original biologic applications received in the first 9 months of FY2026. Annualized, that is about 89, against 74–75 in FY2024 and FY2025 and a previous high of 83 in FY2020. The count covers both FDA drug centers (CDER and CBER), not CDER alone.
• Over 2019–2026, approvals in a calendar year have run at about 0.46–0.83 times the prior fiscal year's applications (mean about 0.67). That implies roughly 60 approvals in 2027, and 70 or more if the recent higher ratio holds.
• The regulatory climate favours approval:
  • FDA now treats a single pivotal trial as the default (announced February 2026; draft guidance revised July 2026).
  • Commissioner's National Priority Vouchers: 22 awarded and 9 approvals as of August 2026.
  • Leadership has stabilized, with Davis named CDER director on Sep 8, 2026.
  • A consultant quoted by BioSpace expects "a larger bump in 2027."

Pointing down:

• IQVIA expects only 50–55 new US medicine launches a year through 2030, and a flat 70–80 new active substances a year globally.
• Citeline reports the first drop in total pipeline size since the mid-1990s. Deloitte reports that late-stage pipelines at the top 20 companies shrank.
• FDA's own workload model in the FY2027 user-fee notice predicts 131 CDER original applications in FY2027, against 145 actual in FY2025.
• Record years have historically been followed by declines (1997, 2019).
• Priority vouchers pull some approvals forward into 2026.
• Other drags: closer scrutiny of foreign trial data, staff attrition, and political risk, including the PDUFA reauthorization due in September 2027.

Model

I simulated three scenarios with yearly noise of about 0.11–0.13 in log terms:

• Reversion to about 54–57 a year (weight about 50%): roughly 7% chance of any year reaching 75.
• A new plateau of about 60–63 (about 38%): roughly 30%.
• An uptrend toward about 70 (about 12%): roughly 70%.

The mixture gives about 22–26% that some year in 2027–2031 reaches 75. I shaded this to about 22% because of IQVIA's flat outlook and the pipeline data.

Chance of first reaching 75 in each approval year: 2026 0.2%, 2027 5.0%, 2028 4.8%, 2029 4.4%, 2030 4.0%, 2031 3.6%.

Timing: a year that reaches 75 would almost always do so in December. I put about 94% of each year's probability in that year's Q4 and about 5% in the next Q1, for cases where a late-December 75th approval is posted in January. I put about 1% in Q3, which would need a pace near 100 a year.

Strongest counterargument

Nobody has ever been near 75, the industry forecasters expect a plateau, and the FY2026 jump in applications may be one-off catch-up after the 2025 turmoil. A 10–12% forecast is defensible. I stay higher because the input data (record applications, a record 2026, a record quarter) are hard evidence of a shift. The noise in annual counts is also large enough that a 2018-style outlier on a level near 60 would reach 75.

Sources

• FDA, Novel Drug Approvals for 2026, accessed 2026-10-11: https://www.fda.gov/drugs/novel-drug-approvals-fda/novel-drug-approvals-2026
• FDA year pages for 2021–2025, and Wayback Machine copies for 2018–2020.
• FY2025 PDUFA Performance Report: https://www.fda.gov/media/191112/download
• FDA-TRACK PDUFA dataset: https://www.fda.gov/media/145457/download
• FY2027 PDUFA fee notice, Federal Register, 2026-07-30: https://www.federalregister.gov/documents/2026/07/30/2026-15334/prescription-drug-user-fee-rates-for-fiscal-year-2027
• IQVIA, 2026-06-16: https://www.iqvia.com/blogs/2026/06/us-medicine-use-climbs-amid-continued-affordability-and-access-challenges
• BioSpace, H1 2026 review: https://www.biospace.com/fda/despite-chaos-and-churn-fda-decisions-hold-mostly-steady-in-h1-2026
• PharmaVoice on the Citeline review, Sep 2026: https://www.pharmavoice.com/news/pharma-rd-pipeline-cancer-obesity-rare-disease/830134/
• HHS leadership release, 2026-09-08: https://www.hhs.gov/press-room/hhs-announces-key-fda-leadership-selections.html
• The Pharma Letter on priority vouchers, 2026-08-14: https://www.thepharmaletter.com/insights/national-priority-vouchers-a-year-of-reform
• Assyro decision calendars for Oct 2026–Feb 2027: https://www.assyro.com/tools/pdufa-calendar/2026/december

Question. By the last day of each quarter, has FDA's Center for Drug Evaluation and Research (CDER) publicly reported approving 75 or more novel drugs in a single calendar year?

Definitions.

• Novel drugs: the products counted in FDA's “New Molecular Entity and New Therapeutic Biological Product Approvals (CDER only)” series, published through its annual Novel Drug Approvals pages and reports. FDA's classification controls; this is not a count of all new drug applications or all FDA product approvals.
• Single calendar year: January 1 through December 31 of one year, using the approval dates assigned by FDA. A published year-to-date count of at least 75 qualifies without waiting for the year to end. Rolling twelve-month totals and fiscal-year totals do not qualify.
• Counting unit: one entry in CDER's novel-drug count, regardless of the number of ingredients, indications, dosage forms, or application numbers. Combination products count as FDA counts them. Both traditional and accelerated approvals count, including diagnostic drugs if included in the series. No AI contribution is required.

Resolves YES if a qualifying FDA publication states a total of at least 75 for one calendar year, or FDA's year-specific novel-drug table lists at least 75 distinct qualifying entries for that year. The Resolution window and Observation period conventions apply: the annual or year-to-date observation period must end after the opening date. Approvals earlier in 2026 may be included in a qualifying 2026 total. An FDA revision that first establishes the threshold qualifies on its publication date; a later downward revision or withdrawal does not reverse an established YES.

Resolves NO for a quarter if no qualifying publication has established the threshold by that quarter-end. Exclude products approved by the Center for Biologics Evaluation and Research (CBER), generics, biosimilars, supplemental indications, and reformulations unless FDA actually includes the product as an entry in CDER's novel-drug series. Emergency use authorizations, investigational permissions, pending applications, projected approvals, and totals combining different calendar years do not count.

Resolving source. FDA-authored publications only, a narrower source rule than Public confirmation (A2): the Novel Drug Approvals at FDA series, its year-specific pages (for example, 2026), the linked annual New Drug Therapy Approvals reports, or another FDA publication explicitly reporting the same CDER novel-drug statistic. A renamed successor series counts only if it retains the same CDER-only product scope. Use an explicitly reported total over an inconsistent table in the same publication, and the latest FDA correction available when first establishing YES. Apply Resolution date (A3) to the publication or page update first establishing the threshold, not to the date of the 75th approval; dated archived versions may establish when an updated table was public.