“Quick, easy, universal” cancer test — What could go wrong?

December 10th, 2018

An Australian team of scientists made headlines earlier this month by developing a quick, non-invasive blood test that can detect all types of cancer. The test identifies DNA that has been affected by cancer anywhere in the body. Some news sources hailed the test as the “holy grail” of cancer diagnostics and a scientific “breakthrough.”  

However, other doctors and researchers point out that the test has a long way to go before it could be used. And it is not yet clear exactly how the test would be used, and on what population, to identify a threat of cancer.

The first issue is accuracy – how often the test give a true positive response. The test correctly identifies cancer 90% of the time, which means that 10% of people who do have cancer would not have it diagnosed. This could give patients false reassurance in their health, which could delay their diagnosis.

The other issue is specificity – how often does the test give a false positive response. Because the test does not show where a cancer exists in the body or how bad the cancer is, this test could lead to a cascade of unnecessary tests and further incidental findings, especially if it is used to screen low-risk populations. A recent study in the Annals of Oncology found that most older healthy individuals have incidental cancer mutations in their DNA, which may lead to false positives if found on a liquid biopsy. Previous studies of autopsies find that at least one third of men over 70 have incidental prostate cancer. Would all of these incidental cancers be found on a DNA test? And would it benefit patients if these were treated?

One researcher quoted in The Guardian said that the test “could be a screening tool to inform clinicians that a patient may have a cancer, but they would require subsequent tests with other techniques to identify the cancer type and stage.” How exactly would that work? If the test were positive, would the clinician have to guess at where the mostly likely place for cancer would be and then scan every one of those body parts? Doctors previously discussed these same questions about a liquid biopsy for melanoma but did not get sufficient answers.

Being able to identify a universal cancer DNA marker is a great scientific achievement, but hyping the test before we know how to use it is not helpful. As Gary Schwitzer writes in Health News Review about the test, “Applaud the science, the pursuit, the goal. But hold off on making incredibly hyperbolic statements about an incredibly simple universal marker of cancer.”