How to Read Cancer Clinical Trial News

Why trial coverage is hard to read

Cancer trial results are constant news. A drug "shows promise," a study "extends survival," a treatment is called a breakthrough. Behind each one sits a specific study that asked a specific question, and the answer is usually narrower than the headline.

This page is about the handful of things that decide how much weight a result can carry.

The four kinds of trial

NCI groups cancer clinical trials by what they are trying to find out.

• Treatment trials test new treatments, or new ways of using existing ones. Most cancer trials are treatment studies.
• Prevention trials look at ways to prevent cancer. Most participants do not have cancer but are at high risk, or have had cancer and are at high risk of a new one.
• Screening trials test ways to find cancer before it causes symptoms.
• Supportive and palliative care trials look at ways to improve quality of life, especially for side effects of cancer and its treatment.

Knowing which type produced a headline tells you what it can and cannot claim. A screening trial says nothing about how to treat a cancer once found.

What the phases actually mean

NCI describes trials of new cancer treatments as moving through phases, with each phase depending on what came before.

Phase 1 asks whether a treatment is safe, what its side effects are, what people can tolerate, and the highest tolerable dose. NCI puts these at roughly 15 to 30 people. They also check whether the treatment affects the cancer at all.

Phase 2 includes more people, around 50 to 100, and asks whether the treatment seems to work: shrinking tumors, or slowing their growth. Safety study continues.

Phase 3 compares the new treatment against current standard therapy, on effectiveness and on side effects. Participants are randomly assigned. NCI puts these at 100 to several thousand people. FDA describes phase 3 studies as running one to four years and providing most of the safety data, because they are large and long enough to reveal rarer effects.

Phase 4 happens after approval, tracking long-term safety and effectiveness in large, diverse populations.

Most treatments do not finish the journey. FDA reports that roughly 70 percent of drugs move from phase 1 to phase 2, about 33 percent move from phase 2 to phase 3, and about 25 to 30 percent move from phase 3 onward.

Our page on clinical trial phases sets this out in more detail.

Randomization, and the bias it removes

Randomization means assigning people to groups by chance, usually by computer. NCI explains why it exists with a concrete example: if doctors could choose who went into which group, some might place healthier patients in the treatment group and sicker patients in the control group, without meaning to. The treatment would then look better than it is.

Randomization removes that. It is also why neither you nor your doctor can choose your group in a randomized trial.

A study without randomization is not worthless. It simply cannot separate the effect of the treatment from the differences between the people who got it.

Placebos are rarer than people think

NCI states that placebos are rarely used in cancer treatment trials. In most cases one group receives the new treatment and another receives the already approved standard treatment.

Placebos appear where no standard treatment exists, or in a design comparing standard treatment plus placebo against standard treatment plus the study drug. If a placebo is used, participants are told in advance and it is covered in the consent form.

Eligibility criteria shape who a result applies to

Every trial has requirements for joining, called eligibility criteria. NCI lists common ones: health, medical history, family medical history, risk factors, age, previous treatment and the tumor's genetic changes.

Those criteria reduce medical differences among participants, limit risk, and narrow the group to those most likely to benefit. That is what makes a result interpretable. It is also what makes it narrow.

So when a result is reported, the useful question is who was in the trial. A finding in fit 55-year-olds with one previous treatment may not carry over to an 80-year-old with heart disease.

When to read past the headline

Some practical prompts:

• Was this a press release, a conference abstract or a peer-reviewed paper? They carry different amounts of scrutiny, and the first two are often early.
• How many people took part, and for how long?
• Was there a comparison group, and what did it receive?
• What was measured: how long people lived, how long before scans showed growth, or how many tumors shrank?
• Is the improvement reported in relative terms, such as "30 percent lower risk," or in absolute terms, such as "from 8 percent to 6 percent"? The same result can be described either way.
• Were the side effects reported alongside the benefit?

What this study coverage cannot tell you

• It cannot tell you whether a trial result applies to your situation. Eligibility criteria decide that, and only your care team knows your details.
• It cannot show that an early, promising result will hold. Most treatments that look good in phase 1 or 2 do not survive phase 3.
• It cannot substitute for the published paper. Headlines routinely drop the confidence intervals, the follow-up time and the harms.
• It cannot support starting, stopping or changing treatment.

If a trial sounds relevant, our pages on what clinical trials are and joining a clinical trial explain how enrollment actually works.

Sources

• https://www.cancer.gov/research/participate/clinical-trials/what-are-clinical-trials
• https://www.cancer.gov/research/participate/clinical-trials/how-trials-work
• https://www.fda.gov/patients/drug-development-process/step-3-clinical-research

An AI-assisted editorial system helped prepare this page. No named medical reviewer has reviewed it unless one is listed.

See an error, old source, or unclear wording? Tell us.