Direct Answer

A drug candidate moves through a sequential set of clinical stages before it can reach the market: preclinical testing (lab and animal studies), Phase 1 (small trials primarily testing safety), Phase 2 (larger trials testing efficacy and dosing), Phase 3 (large trials confirming efficacy and safety versus a control, typically required for regulatory approval), and finally regulatory review and approval. Each stage generates data that either supports advancing the candidate to the next stage or ends the program.

Historical approval rates vary meaningfully by clinical stage and therapeutic area. As a general pattern, later-stage candidates that have already cleared earlier trials carry lower remaining development risk than earlier-stage candidates, simply because more of the scientific and regulatory hurdles have already been passed. Specific approval probabilities vary by drug class and should be sourced from current industry data rather than treated as fixed constants.

Key Takeaways

  • Five sequential stages: preclinical, Phase 1, Phase 2, Phase 3, and regulatory review/approval, each building on evidence gathered in the stage before it.
  • Preclinical work happens before human testing: lab and animal studies assess basic safety signals and biological activity before any human trial begins.
  • Phase 1 is primarily about safety: small trials establish tolerability and dosing in a limited number of participants, not whether the drug works.
  • Phase 2 starts testing whether the drug works: larger trials evaluate efficacy and refine dosing in the target patient population.
  • Phase 3 is the confirmatory stage: large trials compare the drug against a control group and typically produce the evidence regulators require before approval.
  • Risk generally declines at later stages, but it never reaches zero, a candidate can still fail in Phase 3 or during regulatory review even after clearing every prior stage.
  • Approval odds are not fixed constants, they vary by therapeutic area and drug class, and should be checked against current industry data rather than assumed.

What Are the Clinical Stages?

Preclinical: Lab and Animal Testing

Before a drug candidate is ever given to a human, it goes through preclinical testing, laboratory experiments and animal studies designed to assess basic biological activity and identify early safety signals. This stage generates the data that supports (or fails to support) an application to regulators for permission to begin human trials. Most candidates that enter early research never make it past this point.

Phase 1: Primarily Testing Safety

Phase 1 trials are small, typically enrolling a limited number of participants, and are primarily designed to test safety, tolerability, and appropriate dosing ranges. The central question at this stage is not yet whether the drug works, but whether it can be given to humans at a given dose without causing unacceptable harm.

Phase 2: Testing Efficacy and Dosing

Phase 2 trials are larger than Phase 1 and begin to test efficacy, whether the drug actually produces the intended clinical effect, alongside continued refinement of dosing in the target patient population. A candidate that shows encouraging safety in Phase 1 but fails to demonstrate meaningful efficacy in Phase 2 is commonly discontinued at this stage.

Phase 3: Confirming Efficacy and Safety Versus a Control

Phase 3 trials are large-scale studies that confirm efficacy and safety versus a control group, and are typically the stage of evidence required for regulatory approval. Because these trials are larger, longer, and more expensive than earlier stages, a Phase 3 failure represents a significant loss of invested capital and time, but a successful Phase 3 result is the core evidence package a regulator reviews when deciding whether to approve a drug.

Regulatory Review and Approval

Once Phase 3 data is complete, the sponsor submits an application to the relevant regulator for review. The regulator evaluates the full body of evidence from all prior stages before deciding whether to approve the drug for marketing, request additional data, or decline the application. This stage carries its own distinct risk, separate from the clinical trial risk of the stages before it, a scientifically successful Phase 3 trial does not guarantee a positive regulatory decision.

How It's Used: Reading a Pipeline by Stage

Hypothetical example, for education only.

  1. Identify each candidate's current stage. A biotech company's pipeline disclosure might show Drug A in preclinical, Drug B in Phase 2, and Drug C in Phase 3, three assets at three very different points in development.
  2. Rank remaining development risk by stage. All else equal, Drug C (Phase 3) has already cleared preclinical, Phase 1, and Phase 2 testing, so it generally carries lower remaining development risk than Drug B (Phase 2), which in turn generally carries lower remaining risk than Drug A (still in preclinical).
  3. Recognize that risk is lower, not absent, at later stages. Drug C could still fail its Phase 3 trial or be declined at regulatory review, so "further along" does not mean "safe". It means fewer remaining hurdles rather than zero remaining hurdles.
  4. Check the therapeutic area before assuming a generic probability. Because historical approval rates vary meaningfully by stage and therapeutic area, an investor comparing Drug B to a Phase 2 oncology candidate at another company should look up current, area-specific approval-rate data rather than apply one universal number to both.
  5. Watch for stage transitions as the key events. A candidate advancing from Phase 2 to Phase 3, or from Phase 3 into regulatory review, is a data point that the market treats as a meaningful reduction in remaining risk, and a candidate discontinued at any stage removes that asset's expected value from the pipeline entirely.

Limitations and Common Mistakes

Treating "later stage" as "guaranteed to succeed"

A Phase 3 or regulatory-review-stage candidate has cleared meaningfully more hurdles than a preclinical one, but it has not cleared all of them. Phase 3 trials fail, and regulators decline or delay applications even after a scientifically encouraging trial result. Reading stage progression as a certainty rather than a reduction in risk is a common and costly error.

Elderly female scientist in PPE examining samples with a microscope in a modern lab.
Photo by kaboompics.com via Pexels

Applying one approval-rate assumption across every drug class

Historical approval rates vary meaningfully by clinical stage and therapeutic area, an oncology candidate and a candidate for a common chronic condition do not share the same historical pattern of success at each stage. Using a single generic probability across all candidates, rather than checking current industry data specific to the drug class and stage in question, produces a misleading picture of pipeline risk.

Ignoring that a stage label alone doesn't convey trial design quality

Two Phase 2 trials are not automatically comparable. Trial size, endpoint definitions, and control group design all affect how much a given result should move an assessment of the candidate's odds. The stage tells you roughly how much of the process has been completed; it does not by itself tell you how strong the underlying evidence is.

Confusing scientific success with commercial outcome

Even a drug that clears every clinical stage and receives regulatory approval still faces commercial risk after launch, reimbursement decisions, competition, and adoption by prescribers are separate from the clinical development process this page describes. Clinical stage measures development risk, not the eventual size or certainty of commercial revenue.

FAQ

What are the clinical stages a drug candidate goes through?

A drug candidate typically moves through preclinical testing (lab and animal studies), Phase 1 (small trials primarily testing safety), Phase 2 (larger trials testing efficacy and dosing), Phase 3 (large trials confirming efficacy and safety versus a control, typically required for regulatory approval), and finally regulatory review and approval. Each stage builds on data from the one before it, and a candidate can be discontinued at any point if the data does not support moving forward.

Why does a drug's clinical stage matter for valuing a biotech company?

Later-stage candidates that have already cleared earlier trials generally carry lower remaining development risk than earlier-stage candidates, because they have survived more rounds of safety and efficacy testing. A Phase 3 asset is priced with less discount for failure risk than a preclinical or Phase 1 asset with the same eventual market opportunity, since fewer clinical hurdles remain between it and a potential approval decision.

What is the difference between Phase 1, Phase 2, and Phase 3 trials?

Phase 1 trials are small and primarily test safety, tolerability, and dosing in a limited number of participants. Phase 2 trials are larger and begin testing whether the drug actually works (efficacy) at various doses in the target patient population. Phase 3 trials are large-scale studies that confirm efficacy and safety against a control group (often a placebo or existing standard treatment) and are typically the evidence regulators require before approving a drug for market.

Are approval rates the same for every clinical stage and drug class?

No. Historical approval rates vary meaningfully by clinical stage and therapeutic area. The general pattern is that odds of eventual approval improve at each later stage, but the specific probabilities are not fixed constants, they vary by drug class, disease area, and time period, and should be sourced from current industry data rather than assumed to be uniform across all biotech programs.

Can a drug candidate fail after reaching Phase 3?

Yes. Reaching Phase 3 substantially reduces but does not eliminate remaining risk. A Phase 3 trial can still fail to show the required efficacy or safety margin versus its control, and even a successful trial does not guarantee regulatory approval, since agencies can request additional data or decline to approve for reasons specific to the application. This is why later-stage candidates carry lower, not zero, remaining risk.

Where can I find primary data on clinical trial phases and drug approvals?

ClinicalTrials.gov (run by the National Institutes of Health) lists individual trial registrations, phases, and status. The FDA's drug development and review pages describe the regulatory process in detail. Company 10-K and 10-Q filings on SEC EDGAR disclose pipeline stage and material trial updates for publicly traded biotech and pharmaceutical companies.

What is an adaptive trial design and how does it blur the phase boundaries?

An adaptive design allows pre-specified changes to a trial while it runs, such as dropping a dose arm, adjusting enrolment, or expanding a promising cohort. Combined protocols labeled Phase 1/2 or Phase 2/3 run what would traditionally be two separate studies as one continuous trial. This can shorten development time, but it means a stage label conveys less about how much evidence exists. Reading the protocol design rather than the phase label is what identifies the actual evidentiary standard a result has met.

What is a Phase 4 study and why does it happen after approval?

Phase 4, or post-marketing, studies run after a product is approved and available. They collect safety and effectiveness information in a much larger and more varied population than a controlled trial can enrol, and regulators sometimes require specific post-marketing commitments as a condition of approval. Findings can lead to label changes, new warnings, or in some cases withdrawal. For an investor, this is the stage where a commercial product can still face regulatory risk long after the approval decision.

What is the difference between a primary endpoint and a surrogate endpoint?

A primary endpoint is the outcome a trial is designed and powered to test, and success or failure is judged against it. A surrogate endpoint is a measurable marker used to stand in for a clinical outcome that would take much longer to observe, such as a laboratory value substituting for long-term disease progression. Regulators accept certain surrogates in defined settings. A trial that misses its primary endpoint but reports favorable secondary results is a different outcome from one that met the endpoint it was designed around.

References

Disclaimer

This article is for educational and informational purposes only and does not constitute personalized investment, financial, medical, or legal advice. Clinical development outcomes and historical approval rates change over time and vary by drug class and therapeutic area; always verify current data from primary regulatory and company sources before making investment decisions. Trading involves risk, including the possible loss of principal.