FDA’s Project Optimus is an initiative of the agency’s Oncology Center of Excellence (OCE) to reform how oncology drug developers select the dose that moves forward into registration trials and, eventually, the marketed product. Its core argument is that decades of relying on the maximum tolerated dose (MTD) from early-phase oncology trials produced approved doses that were often higher than needed for efficacy and less tolerable than they should have been — a problem specific to oncology because traditional dose-finding methods borrowed from cytotoxic chemotherapy, where “more drug is better,” don’t hold for targeted therapies and immunotherapies with different dose-response relationships.
What Project Optimus actually changed
Historically, oncology dose selection leaned on the 3+3 dose-escalation design: small cohorts of patients receive successively higher doses in a Phase 1 trial until intolerable toxicity defines the MTD, which is then typically carried forward largely unchanged through Phase 2 and Phase 3 and into the approved label. That approach answers “what’s the highest dose patients can tolerate,” not “what’s the dose that gives the best balance of benefit and risk” — and for many modern oncology drugs, especially targeted small molecules and biologics, those are different doses.
Project Optimus asks sponsors to instead identify an optimal biological dose (or optimal dose, sometimes referenced as OBD) — a dose regimen supported by pharmacokinetic, pharmacodynamic, safety, and preliminary efficacy evidence, rather than toxicity alone. In practice, FDA’s expectation (reflected in the guidance below) is that sponsors typically evaluate more than one dose in randomized, comparative fashion before selecting the dose to carry into pivotal trials, rather than defaulting to the MTD identified in Phase 1.
The guidance documents behind the initiative
Project Optimus is not itself a regulation — it’s an FDA program (housed in the OCE) that has been implemented primarily through guidance documents and stakeholder engagement rather than a rule change:
- Draft guidance (January 2023): “Optimizing the Dosage of Human Prescription Drugs and Biological Products for the Treatment of Oncologic Diseases” laid out FDA’s expectations for dose-optimization strategies earlier in development, before pivotal trials begin.
- Oncology Dosing Tool Kit: a companion resource from the OCE intended to help sponsors evaluate the pharmacokinetic, pharmacodynamic, and exposure-response data that support a chosen dosage, published on FDA’s Oncology Center of Excellence site.
- Final guidance (August 2024): FDA finalized the dosage-optimization guidance for oncology drug and biological products, formalizing the shift away from MTD-only dose selection. The final guidance describes a general recommendation that sponsors evaluate more than one dose — commonly the MTD (or maximum administered dose) and at least one lower dose — in a randomized, parallel design before selecting the dose(s) to move into later-phase development, with the comparison ideally informed by both safety and efficacy signals rather than tolerability alone.
Sources: FDA’s Oncology Center of Excellence program overview (fda.gov), the Oncology Dosing Tool Kit (fda.gov), and the dosage-optimization guidance document (fda.gov/media/164555).
Why this matters for trial design and research administration
For anyone administering, budgeting, or monitoring an oncology clinical trial, Project Optimus has concrete downstream effects on protocol design and study operations:
- Dose-finding cohorts get more complex. A single-arm 3+3 dose-escalation cohort followed by a fixed Phase 2 dose is no longer the default expectation. Sponsors increasingly build randomized, multi-arm dose-comparison cohorts into the same early-phase protocol — which has staffing, IRB, and budget implications distinct from a traditional Phase 1 design.
- Bayesian and model-informed designs have become mainstream. Because comparing two or more doses efficiently, with limited patient numbers, is a harder statistical problem than a simple escalation rule, adoption of Bayesian adaptive dose-finding designs in oncology has increased substantially since Project Optimus began — reported industry surveys put Bayesian design adoption in oncology dose-finding trials at roughly 48% in 2021 rising to around 75% by 2024, alongside separate survey findings that the large majority of oncology drug developers say Project Optimus has changed their dose-optimization strategy. See CASRAI’s Bayesian Adaptive Design entry for the underlying statistical concept.
- It interacts with, but is distinct from, other FDA oncology programs. Project Optimus is a dose-selection requirement that applies regardless of whether a product is also pursuing an expedited pathway. A sponsor seeking Accelerated Approval or another designation under CASRAI’s FDA expedited programs guide still has to satisfy Project Optimus’s dose-optimization expectations — expedited review changes the review timeline, not the evidentiary bar for dose selection.
- It’s discussed early, often at the Pre-IND stage. FDA has encouraged sponsors to raise dose-optimization strategy during early regulatory interactions rather than after Phase 1 data exist — see CASRAI’s FDA Pre-IND meeting guide for how that conversation typically fits into an Investigational New Drug (IND) program timeline.
Optimal dose vs. maximum tolerated dose: the core distinction
| Concept | What it measures | How it was historically used | Project Optimus expectation |
|---|---|---|---|
| Maximum Tolerated Dose (MTD) | The highest dose a defined proportion of patients can tolerate without unacceptable dose-limiting toxicity, typically identified in a Phase 1 dose-escalation cohort | Carried forward largely unchanged as “the” dose for Phase 2/3 and the eventual label | Still identified as one reference point, but no longer assumed to be the right dose to advance on its own |
| Optimal (Biological) Dose | A dose regimen supported by pharmacokinetic/pharmacodynamic, safety, and preliminary efficacy evidence, reflecting benefit-risk rather than tolerability alone | Not formally required as a distinct concept before Project Optimus | The dose sponsors are expected to identify and justify, typically via a randomized comparison of two or more doses |
Scope: which products and trials does it apply to?
Project Optimus and its guidance are specific to oncology drug and biological products under FDA’s Center for Drug Evaluation and Research and Center for Biologics Evaluation and Research, administered through the Oncology Center of Excellence. It does not apply as a formal requirement to non-oncology indications, though the general principle it embodies — that dose selection should reflect a full benefit-risk evaluation rather than tolerability alone — has been echoed more broadly in FDA’s dosage-optimization thinking, including a related, broader-scope guidance on optimizing dosage for the treatment of any serious disease. Sponsors developing combination oncology regimens, biomarker-selected populations, and novel modalities (e.g., antibody-drug conjugates, bispecifics, cell therapies) are all within scope and have been specific areas of FDA-industry discussion, including joint FDA workshops with ASCO and AACR on dose optimization in combination cancer therapies.
Frequently asked questions
When did FDA launch Project Optimus?
Project Optimus originated within FDA’s Oncology Center of Excellence and became publicly visible through its January 2023 draft guidance, though the underlying policy shift had been signaled by OCE leadership in the preceding years. The initiative reached a major milestone with the final guidance issued in August 2024.
Is Project Optimus a binding regulation?
No. It operates through FDA guidance documents, which represent the agency’s current thinking on dose optimization and are not themselves legally binding — but guidance documents set clear regulatory expectations that reviewers apply during IND, NDA, and BLA review, so in practice sponsors that ignore them risk significant review delays or requests for additional dose-finding data.
Does Project Optimus require two doses in every oncology trial?
The final guidance’s general recommendation is that sponsors evaluate more than one dose — often the MTD (or maximum administered dose) and at least one lower dose — in a randomized fashion before selecting a dose for later-phase development, but FDA has indicated flexibility where a sponsor can otherwise justify its dose-selection approach with adequate pharmacokinetic, pharmacodynamic, safety, and efficacy data.
How does Project Optimus relate to ICH guidelines on trial design?
Project Optimus is an FDA-specific oncology program, but the dose-finding and dose-optimization studies it calls for still have to be designed, randomized, and analyzed consistent with the general trial-design and statistical principles in ICH E6(R3) Good Clinical Practice and the broader ICH E8/E9 trial-design and biostatistics framework — Project Optimus changes what dose-selection strategy FDA expects, not the underlying GCP and protocol-design standards a trial has to meet.
Related CASRAI resources
- What Makes Oncology Clinical Trials Different: RECIST, Endpoints, and the NCTN
- FDA Expedited Programs: Fast Track, Breakthrough Therapy, Accelerated Approval, and Priority Review Compared
- FDA Pre-IND Meeting: Process, Timeline, and Briefing Package
- Investigational New Drug (IND)
- Phase 1 Trial
- Clinical Trial Protocol
- Clinical Research Administration
