Cancer trial phases explained — phase 1, 2 and 3

Each phase answers a different question. Phase 1 asks whether a treatment is safe enough in people, and at what dose. Phase 2 asks whether it does anything useful in one specific cancer. Phase 3 compares it against the current standard treatment. Phase 4 watches for rarer effects after approval.

The phase number describes the question the study was built to answer. It does not describe how good the treatment is, and it is not a ranking. A phase 3 study is not a better treatment than a phase 1 study — it is a study at a later stage of testing, asking a different thing.

That distinction is the whole point of reading this page. Families are usually told “there is a trial” and then asked to decide. Knowing what the phase means lets you weigh the offer on its own terms instead of accepting it out of hope or refusing it out of fear.

Source: the phase definitions used in NCCN, ASCO and ICMR patient-education material. The participant numbers are typical ranges, not rules — individual studies vary widely, and the protocol for the specific study is the only accurate source for that study.

Phase 1 carries the most unknowns. The dose is still being established, and the full side-effect profile in humans has not been described yet. That is why phase 1 participants are monitored most intensively. It does not follow that later phases are risk-free — they are simply better characterised.

Risk in a trial has three separate parts, and they are worth separating before you weigh an offer.

• How much is already known about the treatment. This is what the phase number tells you. In phase 1 the dose is being worked out in small steps, so the study team is deliberately looking for the point at which side effects become unacceptable. By phase 3 the common side effects have usually been described in earlier studies, and the consent document should list them.
• What kind of treatment it is. An immune-based treatment does not carry the same risks as a chemotherapy drug or a targeted tablet. Immune-related side effects can appear late, sometimes after treatment has finished, and can involve the gut, lungs, thyroid, liver, skin or heart. Ask specifically which organs the study team will be watching, and how often.
• Who you are. Your other illnesses, your current medicines, your organ function and your performance status change the risk more than most people expect. Two people in the same study do not carry the same risk, which is one reason eligibility criteria are written so narrowly.
• What happens when something goes wrong. Under the New Drugs and Clinical Trials Rules, 2019, a sponsor is required to provide free medical management for a trial-related injury, and compensation is payable in defined circumstances. Ask for that section to be shown to you in the consent document, not summarised verbally.

Source: New Drugs and Clinical Trials Rules, 2019 (CDSCO) on medical management and compensation; general trial-safety framing follows NCCN and ASCO patient-education material. No figures for risk or response are given here, because they differ for every study.

No phase is designed to guarantee benefit to an individual. Phase 1 exists to establish safety and dose, so personal benefit is not its purpose. Phase 3 studies usually compare a new treatment against accepted standard care, so participants in every arm receive an active treatment rather than nothing.

That is the whole honest answer, and it is worth sitting with for a moment, because it is the opposite of how a trial is usually described to a family.

• A trial is research, not a treatment plan with a known result. If the answer were already known, the study would not be permitted. Anyone who tells you what the outcome will be for you is not describing a trial.
• Phase 3 is where the comparison is with real standard care. In most cancer phase 3 designs the control arm receives the accepted standard treatment, sometimes with a dummy added on top so that the comparison stays fair. This is why oncologists often discuss phase 3 studies with patients who are eligible for them.
• Close monitoring is real, and it is not the same as benefit. Trial participants are seen more often, scanned on a fixed schedule and assessed against a written protocol. Many families value that. It is a genuine feature of taking part, but it is not evidence that the treatment is working.
• Standard care remains the comparison, always. Before agreeing to any study, ask what your treatment would be if you did not join, and what is known about that. If nobody has told you the non-trial option in plain terms, the conversation is not finished.

Ask for the study title and its registration number, look it up in the registry yourself, then read the phase and what the study measures. After that, ask what is paid for, take the consent document home, and get an independent opinion before you sign. Six steps, and none of them need medical training.

1. 1

   Ask for the study title and registration number

   Every Indian study should have a CTRI number, and most international studies carry a ClinicalTrials.gov identifier too. Ask for it in writing, and expect it without asking twice.

2. 2

   Look the study up yourself

   Search that number on the CTRI website or ClinicalTrials.gov. You will see the official title, the phase, the sponsor, the participating sites, the eligibility criteria and what the study is measuring. If a study cannot be found anywhere, stop and ask why before going further.

3. 3

   Read the phase, then read what it measures

   The phase tells you which question is being asked. The primary endpoint tells you what result would count as an answer. Read those two lines before any talk of how promising the drug is.

4. 4

   Ask what is paid for, and by whom

   Ask for a written list: study drug, study scans, study blood tests, routine care, travel, and what happens if you leave the study or it closes early. Also ask what happens to your treatment after the study ends. These answers should exist on paper (indicative, as of August 2026).

5. 5

   Take the consent document home and read it

   You are entitled to time. Consent is a process, not a signature, and should be in a language you read comfortably — see what informed consent means.

6. 6

   Get an independent opinion before deciding

   An oncologist who is not connected to the study can tell you how the trial compares with the standard options for your cancer, stage and biomarker results. Asking is not disloyalty to the doctor who offered it, and any reasonable investigator expects it.

One thing this page cannot do. It cannot tell you whether a particular study is right for you, and it is not an offer of a place in any study. Enrolment is decided by the trial site against the written protocol.

Most of the confusion in a trial conversation comes from eight or nine words. Here is what each one means in practice, so you can read the document you were handed instead of nodding through it. Ask the study team to point to the line where each of these appears.

• Arm. One of the treatment groups in a study. A two-arm study has two groups being compared.
• Randomised. A computer decides which arm you go into, not you and not your doctor. It is done to keep the comparison fair.
• Blinded. You do not know which arm you are in. In a double-blind study your treating team does not know either, until the study is unblinded.
• Placebo. A dummy with no active drug. In cancer studies it is usually added to standard care rather than given instead of treatment.
• Primary endpoint. The one result the study is designed to measure. It tells you what the researchers count as an answer to their question.
• Dose escalation. A phase 1 method. Small groups receive rising dose levels, with a safety review between each level before the next group starts.
• Eligibility criteria. The written list of who may and may not join. It is not negotiable and it is why screening failure happens.
• Expanded access, or compassionate use. A separate route, outside a study, that can allow an unapproved medicine in defined situations. It is not the same as joining a trial and it has its own approval process.
• Withdrawal. Leaving the study. You may leave at any time, for any reason, and your usual care must continue. Our page on leaving a clinical trial once you have started covers what changes and what does not.

Source: standard clinical-trial terminology as defined in ICMR and CDSCO documentation and in NCCN and ASCO patient-education glossaries.

Usually not. In most cancer phase 3 studies every arm receives an accepted standard treatment, and the study tests whether adding or changing something improves on it. Earlier-phase studies are more often considered when standard options have been used, or when a patient is not a candidate for them.

The useful way to hold the decision is as a comparison, not as a leap. On one side is what standard care would give you, with what is known about it. On the other is what the study would give you, with what is not yet known. A trial is worth considering when you understand both sides, and worth postponing when you understand neither.

• Ask what happens if you say no. There should be a clear answer describing your treatment outside the study. If saying no leaves you with nothing, that itself is important information about where you are.
• Ask whether the study delays anything. Screening takes time. Where a treatment decision is urgent, the timeline matters as much as the science, and your oncologist can tell you whether the wait is safe in your situation.
• Ask who stays in charge of your care. Trial participation does not remove your treating oncologist. Find out who you call for a side effect, who arranges your routine scans, and how the two teams communicate.
• Tell both teams everything you are taking. Ayurvedic and homeopathic preparations, supplements, and anything bought privately. Some affect liver, kidney or blood results, and several studies exclude particular preparations outright. Disclosure protects you; it is never used against you.

A last word on tone. Nobody should be pushed into a study, and nobody should be shamed out of one. Trials are how every treatment now considered standard was established, including the immunotherapies used today. Taking part is a genuine contribution as well as a personal decision, and it is entirely reasonable to decline.

• What Informed Consent for a Trial Actually Means — what the document must contain, what you are entitled to ask, and why signing it is a step in a conversation rather than the end of one.
• Risks of Joining an Immunotherapy Trial — the risks specific to immune-based study treatments, including the side effects that can appear late, and what monitoring should be in place.
• Can You Leave a Clinical Trial Once You Have Started? — the answer is yes; this page covers what happens to your treatment and your follow-up when you do.
• Immunotherapy at CION Cancer Clinics — what immunotherapy is, how it is assessed and delivered as day care here, and who is and is not a candidate for it outside a study.

This page is general information about how clinical trials are structured. It is not a recruitment page, it is not an offer of a place in any study, and it makes no claim about how effective any investigational treatment is. CION Cancer Clinics does not promise or arrange enrolment in any trial. Decisions about joining a study belong to you, your treating oncologist and the investigator running that study.

Nobody should sign a study document they do not understand

Phase, arm, randomised, endpoint — the vocabulary alone stops families from asking the obvious questions. We will translate it in plain language, at no cost.