Skip to content

FDA Approval Pathways in Oncology

Definition

FDA offers four primary expedited programs and two standard approval routes for oncology drugs:

Regular (Traditional) Approval: Based on substantial evidence of clinical benefit from adequate and well-controlled trials — demonstrated via clinical benefit endpoints (OS, DFS, EFS, symptom improvement) or validated surrogates known to predict clinical benefit (PFS in select settings, durable ORR with high CR fraction).

Accelerated Approval: "Based on an effect on a surrogate endpoint that is reasonably likely to predict clinical benefit," or on an intermediate clinical endpoint. Requires post-marketing confirmatory trial verifying clinical benefit. Governed by 21 CFR part 314, subpart H and 21 CFR part 601, subpart E. (FDA 2018, Final)

Breakthrough Therapy Designation (BTD): For serious/life-threatening conditions with preliminary clinical evidence suggesting substantial improvement over existing therapy. BTD does not change approval standards but provides intensive FDA guidance, organizational commitment, and rolling review.

Fast Track Designation: For serious conditions filling unmet medical need. Provides rolling review and more frequent FDA interactions but does not alter approval standards or endpoint requirements.

Priority Review: 6-month review vs. standard 10-month review. Does not change approval standards.

Orphan Drug Designation: For diseases affecting <200,000 persons in the US. Provides tax credits, 7-year market exclusivity, and FDA fee waiver — not an approval pathway per se.

Regulatory Position

Regular Approval — Endpoint Requirements

Setting Acceptable Primary Endpoints
Metastatic/advanced (all tumor types) OS (preferred); large/durable PFS + acceptable safety; durable ORR with high CR rate
Adjuvant solid tumor DFS or OS (DFS when OS impractical)
Neoadjuvant solid tumor EFS; pCR (accelerated only, then EFS/DFS confirmation)
Hematologic malignancy (traditional) CR rate (AML, lymphoma); PFS (myeloma, large-effect); OS
Symptom-driven Validated PRO instrument as primary

"The most reliable method for demonstrating efficacy is to show a statistically significant improvement in a clinically meaningful endpoint in randomized controlled trials (i.e., superiority)." (FDA 2018, Final)

Accelerated Approval — Endpoint Requirements

Common surrogate/intermediate endpoints accepted for accelerated approval:

  • ORR (most common): must be reasonably likely to predict clinical benefit; single-arm or randomized
  • CR rate (hematologic malignancies): AML, ALL, lymphoma
  • pCR (neoadjuvant high-risk breast cancer)
  • PFS (select settings, less common — more often supports traditional approval)
  • MRD-negative CR (myeloma, per January 2026 FDA draft at ≥10⁻⁵ sensitivity)
  • DFS/EFS (as surrogate in adjuvant when magnitude is large but OS immature)
  • MFS (metastasis-free survival) (non-metastatic CRPC)

"FDA may grant accelerated approval based on an effect on a surrogate endpoint that is reasonably likely to predict clinical benefit. Such surrogate endpoints are less well-established than those used for traditional approval." (FDA 2018)

Post-marketing confirmatory trial required to verify clinical benefit. If confirmatory trial fails or sponsor does not demonstrate due diligence, FDA may withdraw approval.

Confirmatory Trial "Underway" Requirements (2025 Draft)

Per FDA January 2025 accelerated approval draft guidance:

Three-part test for "underway":

  1. Trial has a target completion date consistent with diligent conduct
  2. Sponsor's plans provide sufficient assurance of timely completion
  3. Patient enrollment has begun prior to accelerated approval being granted

Full enrollment at approval may be required when market availability of the drug would make post-approval enrollment challenging.

Benchmarks required: Site activation statistics, accrual rates, retention projections, event rate assumptions.

Oncology benchmark: Median time from accelerated approval to benefit verification ~3 years (including FDA review); trending shorter — FDA expects continued reduction.

FDORA authority (2022): FDA can require confirmatory trial be underway before approval is granted. This closed the loophole where products maintained accelerated approval for years without meaningful confirmatory trial progress.

Status: January 2025 accelerated approval draft = Draft (comment period closed March 2025)

When to Use

Regular approval strategy:

  • Phase 3 randomized superiority trial with pre-specified clinical benefit endpoint
  • OS as primary when median OS < 24 months expected and crossover is absent
  • PFS as primary when: large targeted therapy effect (HR ≤ 0.60), acceptable safety, OS trend supportive
  • DFS as primary when: adjuvant setting, prolonged OS follow-up impractical

Accelerated approval strategy:

  • Unmet medical need: ≥2L setting with no effective standard therapy; or rare subtype
  • Pre-approval evidence: compelling Phase 1/2 ORR data with durable responses
  • Confirmatory trial must be underway (enrollment begun) at time of BLA/NDA submission per 2025 draft
  • Preferred endpoint: ORR (solid tumors), CR/MRD (hematologic), pCR (neoadjuvant breast)

Breakthrough Therapy Designation triggers:

  • Substantial improvement: typically requires preliminary evidence of >50% reduction in events vs. available therapy, or exceptional ORR in population with no effective options
  • Recommended endpoint discussion: BTD meetings provide venue for FDA to align on endpoint strategy before Phase 3 initiation

Key decision matrix:

Situation Recommended Pathway Primary Endpoint
Unmet need, strong single-arm ORR data (>30%, durable), ≥2L Accelerated approval ORR (single-arm)
Phase 3 vs. SOC, HR expected 0.65–0.75, OS impractical Regular approval PFS (+ OS hierarchical)
Phase 3 vs. SOC, HR expected 0.70–0.80, OS maturity <5 years Regular approval OS
Adjuvant, high recurrence rate, long OS follow-up Regular approval DFS
Neoadjuvant high-risk breast Accelerated → Traditional pCR → EFS/DFS
AML induction Regular or Accelerated CR/CRi rate
Myeloma relapsed/refractory, high unmet need Accelerated MRD-negative CR (2026 draft)

Design Considerations

Single-arm studies

"In settings where there is no available therapy and where major tumor regressions can be presumed to be attributed to the tested drug, the FDA has sometimes accepted ORR and response duration observed in single-arm studies as substantial evidence supporting accelerated approval." (FDA 2018)

  • Single-arm trials do NOT adequately support time-to-event endpoints (OS, PFS, DFS) without randomization
  • Historical control rate must be well-documented, contemporaneous, and reflect same population
  • Sample size: typically 50–100 patients for ORR-based accelerated approval

Randomized vs. single-arm

FDA strongly prefers randomized controlled trials. Single-arm is the exception for accelerated approval only. "Because of variability in the natural history of many forms of cancer, a randomized study is necessary to evaluate time-to-event endpoints." (FDA 2018)

Noninferiority trials

"A noninferiority (NI) trial should demonstrate the new drug's effectiveness by showing that the new drug is not less effective than a standard regimen by a prespecified amount." NI trials are problematic in oncology because: (1) active control effects often not well-characterized; (2) constancy assumption frequently violated; (3) crossover confounds analysis. "NI trials with endpoints other than overall survival and ORR are problematic." (FDA 2018)

FDORA / confirmatory trial compliance

Post-2022 FDORA enforcement: FDA has begun invoking its authority to require confirmatory trials be fully enrolled at approval or to withdraw accelerated approvals when confirmatory trials fail or are delayed without justification. Sponsors planning accelerated approval filings should:

  • Pre-specify confirmatory trial at IND stage
  • Provide enrollment metrics at every FDA meeting
  • Expect FDA to require enrollment to begin before BLA submission

Intercurrent Events by Pathway

Accelerated approval (ORR-based):

  • IE handling is simpler — ORR is a point-in-time measure, not time-to-event
  • Treatment discontinuation before response: count as non-responder (composite strategy)

Regular approval (PFS/OS):

Regulatory Precedent

Historical patterns from FDA approval database and CTG dataset:

Pathway Indication Typical Endpoint Representative Approval
Accelerated NSCLC (rare mutation, ≥2L) ORR (single-arm, IRC) Multiple TKI approvals (crizotinib, ceritinib initial)
Regular NSCLC (1L, targeted) PFS by IRC Osimertinib FLAURA (HR 0.46); alectinib ALEX (HR 0.47)
Regular Colorectal OS FOLFOX, bevacizumab combinations
Accelerated → Regular Neoadjuvant breast pCR → iDFS Pertuzumab neoadjuvant (pCR) → APHINITY (iDFS)
Accelerated AML (novel targeted) CR/CRi rate Ivosidenib IDH1, enasidenib IDH2
Regular Myeloma (1L–3L) PFS Daratumumab combinations (HR 0.4–0.6)

Limitations and Pitfalls

Accelerated approval withdrawal risk:

Post-FDORA, FDA has significantly increased scrutiny. Multiple drugs previously approved via accelerated approval have been withdrawn when confirmatory trials failed (bevacizumab breast cancer, pembrolizumab accelerated approvals in specific settings).

Confirmatory trial failure rate:

~25–30% of accelerated approvals have not been confirmed or have been withdrawn. The endpoint used for accelerated approval (ORR) does not always predict the confirmatory endpoint (OS/PFS). Early FDA alignment on confirmatory trial design is critical.

PFS vs. OS disconnect:

Approval based on PFS may be undermined when OS shows no benefit or trends against treatment (informative censoring, subsequent therapy). FDA expects OS data to be mature or trending supportively at time of regular approval submission.

BTD does not guarantee approval:

BTD provides intensive FDA interaction but does not lower the evidentiary bar. Sponsors sometimes over-interpret BTD as a near-guarantee of approval.

Single-arm bias:

Historical control comparisons are subject to selection bias, changing SOC, improved supportive care, and stage migration (Will Rogers effect). FDA may not accept single-arm data if the historical control population is not rigorously characterized.

Source: FDA Cancer Endpoints 2018 (Final); FDA Accelerated Approval Confirmatory Trial Draft Guidance (January 2025, Draft); FDA MRD/CR in Multiple Myeloma Draft (January 2026, Draft) Status: Cancer Endpoints 2018 = Final; January 2025 confirmatory trial draft = Draft; January 2026 MRD/MM draft = Draft Compiled from retrieved FDA chunks + ClinicalTrials.gov records