Differences in historical operational biomarker positivity between two plasma-based comprehensive genomic profiling platforms in Japan: a nationwide C-CAT study
A nationwide Japanese study reveals that the historically large disparity in operational biomarker positivity between the FoundationOne Liquid CDx and Guardant360 CDx platforms is primarily driven by differences in Tumor Mutational Burden (TMB) reporting rather than patient characteristics or facility composition.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you are a detective trying to solve a mystery inside a patient's body. Instead of looking for clues in a dusty attic, you are hunting for tiny genetic "typos" in the DNA floating in their blood. This is the world of liquid biopsy, a modern way to check a patient's genetic code without needing a painful surgery to cut out a piece of their tumor. Scientists use these tests to find specific "biomarkers"—like finding a specific missing puzzle piece—that tell doctors which medicine might work best. Sometimes, the test also counts the total number of typos in the DNA, a score called Tumor Mutational Burden (TMB). If the score is high, it's like finding a whole room full of broken puzzles, which might mean the patient's immune system can be tricked into attacking the cancer.
But here is the tricky part: not all detective kits are the same. Some kits are bigger and report more clues than others. In Japan, researchers wanted to know why two popular blood-test kits seemed to give very different answers about how many patients had these helpful clues. Was it because the patients were different? Was it because the tests were used at different times? Or was it simply that one kit was reporting a specific clue that the other kit was ignoring? Understanding this is crucial because if a doctor relies on a test that misses a clue, they might miss a life-saving treatment option.
The Great Blood-Test Showdown
In a massive nationwide study across Japan, researchers took a deep dive into the records of nearly 17,000 patients who had their blood tested for cancer mutations. They compared two heavy-hitters in the world of liquid biopsies: FoundationOne Liquid CDx (F1L) and Guardant360 CDx (G360).
Think of these two tests as two different news reporters covering the same big story. The researchers were looking at a specific list of seven "headline" genetic clues that are linked to approved cancer treatments in Japan. They wanted to see how often each reporter found at least one of these seven headlines.
The results were a bit shocking. When the researchers looked at the raw numbers, the F1L reporter found a headline in about 14.3% of patients (that's roughly 2,131 out of 14,878). But the G360 reporter only found a headline in about 4.7% of patients (95 out of 2,042). That is a huge gap! It's like one reporter saying, "Hey, we found a treasure map!" while the other says, "I don't see anything."
Was it the Patients or the Reporters?
Naturally, the researchers asked: "Maybe the G360 reporter was just sent to a different neighborhood where there are fewer treasures?" They checked if the patients were different in age, gender, or the type of cancer they had. They also checked if the tests were run at different times of the year.
They ran the numbers through a complex statistical "filter" to level the playing field, making sure they were comparing apples to apples. Even after adjusting for all these differences, the gap didn't shrink. The F1L test still reported a positive result about 10 percentage points more often than the G360 test. It turned out that the patients weren't the problem; the difference wasn't because one group of patients was "luckier" than the other.
The Missing Clue: The TMB Mystery
So, what was causing the massive difference? The answer lay in one specific clue: Tumor Mutational Burden (TMB).
During the time this study covered, the F1L test included TMB in its report. If a patient had a high TMB score, it counted as a "win" for the test. However, the G360 test used in Japan during this period did not report TMB at all. In the data records, the TMB fields for G360 were completely blank. It wasn't that the test found zero TMB; it was that the test simply didn't tell the doctors the number.
When the researchers looked closely at the F1L results, they found that 65.6% of the patients who tested positive only because of TMB. In other words, if you took away the TMB clue from the F1L list, the two tests suddenly looked almost identical.
When they removed TMB from the equation and compared the remaining six clues, the gap between the two tests almost vanished. The difference dropped from a massive 9.7 percentage points down to a tiny 0.4 percentage points, a difference so small it could easily be due to random chance.
The Takeaway
The study concluded that the huge historical difference between these two tests wasn't because one test was "better" at finding cancer mutations or because the patients were different. It was simply because one test was reporting a specific piece of information (TMB) that the other test was not allowed to report at that time.
This is a vital lesson for doctors and patients: the information a test gives you depends heavily on what the test is allowed to say. If a test doesn't report TMB, it doesn't mean the patient doesn't have a high TMB; it just means that specific clue is missing from the report. The study didn't prove that one test is biologically superior, but it did show that the "reporting rules" of the time created a massive illusion of difference. The mystery wasn't in the blood; it was in the paperwork.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.