Galleri: The Cancer Test Everyone Is Talking About

Let’s start with some good news: cancer death rates in the United States are falling. Since 1991, the age-adjusted cancer mortality rate has declined by about 34% (reference). This can be credited to many factors, including fewer people smoking, earlier detection, and better treatments. That translates into millions of lives saved.

However, cancer remains the 2nd leading cause of death in the United States. Death rates from colon and breast cancer among those under age 50 are increasing. The idea of a simple, non-invasive blood test that could detect many cancers early is incredibly appealing

The Galleri test by GRAIL

With just a single blood draw, the Galleri test looks for DNA fragments shed by tumors—circulating tumor DNA (ctDNA)—across more than 50 different cancers, including pancreatic and ovarian, which are notoriously difficult to screen and often deadly when finally caught.

So how good is the Galleri test at finding cancer?

Based on published studies conducted by the company, the overall sensitivity of the test is about 50%. That means if you have cancer, this test can find it about half the time. For a single blood draw looking for many different cancers, that is a remarkable result. But when we look further, we find that the results vary widely depending on cancer type.

The highest sensitivity is for detecting stage 3 and 4 cancers which are more likely to release tumor DNA in the bloodstream. Unfortunately, these late stage cancers may not be curable. For stage 1 cancers—the ones we are most interested in finding—the sensitivity drops to an overall 16%. There are some exceptions - for early stage pancreas and ovarian cancers, the sensitivity is about 50%. But the overall low sensitivity for early cancers is disappointing.

Another important metric is the accuracy of a positive test (i.e., false positive rate). The PATHFINDER study evaluated the Galleri test in 6,621 asymptomatic adults aged 50 years or older in the United States. Ninety-two participants (1.4%) had a positive test and further testing confirmed cancer in approximately half of these cases. The positive predictive value was 49%, meaning a positive result was correct half the time. This again stacks up well against some other screening tests; for example, mammograms done annually for 10 years also result in a 50% false positive rate. The big unanswered question, however, is the most important one: do people who test positive and undergo earlier treatment actually live longer?

Why Earlier Isn’t Always Better

Imagine two 60-year-olds develop the same cancer at the same time.

• One gets screened, is diagnosed early, treated, goes into remission, but dies at 65 from recurrence.

• The other doesn’t get screened, is diagnosed later with stage 4 disease, and also dies at 65.

In this scenario, screening did not extend lifespan—though the first person endured years of anxiety and treatment side effects. This is known as lead-time bias, and it reminds us: earlier detection ≠ longer survival.

The only way to prove a test saves lives is through large, randomized trials with long follow-up, looking at mortality reduction

The Ongoing NHS Trial

GRAIL has partnered with the National Health Service in the UK to conduct a randomized controlled clinical trial of 140,000 patients aged 50-77 years old. Half of the patients get an annual Galleri blood test for 3 years and half the patients do not. The primary endpoint is not decrease in mortality, but the decrease in the rate of stage 3 and 4 cancers. The study results are expected in February 2026 and I am eager to see what they find.

For now, the Galleri test does not replace current cancer screening tests. Current guidelines suggest screening for just 5 cancers - cervical, breast, colon, prostate, and lung (only in smokers). The data for decreasing mortality is strongest for cervical cancer screening.

The Galleri test represents a promising new frontier—especially for cancers like pancreas and ovary, where existing screening tools are inadequate. But until the NHS trial reports out, it remains unclear whether this test reduces late-stage diagnoses, much less mortality. In the meantime, it may cause unnecessary anxiety, invasive testing, and overtreatment.

I’m hopeful about the future of blood-based cancer detection. But today, my advice is simple: stick with established screening tests and wait for the data.