For doctors and patients, the Holy Grail of medicine would be a simple blood or saliva test to detect all types of cancer before symptoms or sickness appears. Doctors could screen and treat patients earlier in the course of disease. As Dr. Lisa Stempel, director of the high-risk cancer screening program at Rush University Medical Center, told the Tribune recently, “The goal of all screening is to find cancer early when we can treat it.”
But as with the Holy Grail of ancient Christian legend, an early-detection multicancer test has long eluded all who have pursued it. Like the Grail, such a test may not even exist, and if it does, it might actually be quite different from what is being sought. Despite ongoing efforts to create such an early detection test, no regulatory body in the U.S. has yet to approve of one; there are still too many pitfalls in the results of tests currently available.
But the search continues, and despite all those pitfalls, hospitals and physicians are using unapproved tests that are available to screen patients, often at exorbitant costs. For those understandably concerned about their health and undeterred by cost, there are things they should know about early-detection multicancer testing — questions to ask before consenting to be tested:
Has this test been evaluated in my specific medical population, and am I an appropriate candidate for it?
Screening tests depend on the possibility you could have the diseases those tests are designed to discover (epidemiologists call this “pretest probability”). If you are unlikely to have a certain disease, it would not be helpful to screen for it; the best screening test for a tropical disease such as malaria is pointless in Scandinavia. Because cancer incidence increases as people age, a multicancer screening test may be more appropriate for old people than for young ones.
Does this test fail to identify many people who have cancer?
A cancer test should be positive in most people with cancer (the test “sensitivity”). The test then has value even when negative — a negative test in these situations can be reassuring that you do not have cancer. But a test missing a good percentage of those with treatable cancers is less useful.
Is the test often positive in people who don’t have cancer?
A cancer test that is positive in many people without cancer can do harm medically, emotionally or financially for obvious reasons. It should be specific for cancer (the test “specificity”). There is usually a tradeoff between the sensitivity and specificity of a screening test. Those tests that pick up most cancers tend to pick up other unimportant things as well — and those screening tests that are limited to discovering cancer often miss many people with the disease.
What are the false positive and false negative rates for this test?
These rates, from 0% to 100%, are the practical clinical expressions of test sensitivity and specificity. A false positive test is one that is positive when the person tested does not have cancer. A false negative test is one that is negative in a person who has cancer. The actual rates of false positive and false negative depend on how many people in a population have the disease. A false positive test for prostate cancer is much more likely in a 20-year-old than in a 75-year-old because virtually no 20-year-olds have prostate cancer, so any positive test is likely to be false positive.
A new early-detection multicancer screening test may be described in glowing terms if it has a 1% false positive rate, but if 1 million people are screened, 10,000 will be told they have cancer when they don’t. If you are one of those 10,000, this matters to you, your family and those who care about you.
Will having this test prolong my life?
Stempel’s quote about the goal of screening being to find cancer earlier is incomplete. The ultimate goal of screening is to prolong life. It is often taken for granted that through screening, earlier diagnosis and treatment will improve survival. This is usually but not always true. If a cancer is fast-growing or treatment is ineffective, survival will not improve. That’s why large-scale evaluation of these screening tests is needed to demonstrate not simply earlier diagnosis but also better patient survival.
To illustrate, consider a concept called lead-time bias, in which earlier diagnosis makes it appear survival has improved when it hasn’t. If a patient is diagnosed with cancer through conventional testing in 2025, then lives until 2030, he or she is said to have a five-year survival after diagnosis. If an early screening test diagnoses the same patient with cancer in 2024, but the patient still only lives until 2030, the patient has not lived longer but appears to have an improved six-year survival with the test.
Cancer-specific screening tests have helped prolong the lives of patients with selected cancers: breast, cervical, colon, prostate and lung. So far, no multicancer screening test meets this standard. Anecdotes of patients who have undergone multicancer screening and had their lives lengthened by early discovery of cancer should be taken seriously. But the plural of anecdote is not data. What’s more, data is not truth, and truth is not wisdom.
When considering an early-diagnostic multiscreening cancer test, wisdom is what patients actually seek. Without good answers to these questions, the journey from data to truth to wisdom ends before it begins.
Dr. Cory Franklin is a retired intensive care physician and the author of “The COVID Diaries 2020-2024: Anatomy of a Contagion as it Happened.”
Submit a letter, of no more than 400 words, to the editor here or email letters@chicagotribune.com.