VG is an 82-year-old man who presented with abdominal pain. Imaging revealed widespread metastatic cancer with omental studding. Before a tissue diagnosis, tumor markers — CEA, CA19-9, PSA, AFP — were sent. The PSA was elevated at 569 ng/mL (normal: 0-4). After a biopsy, he was definitively diagnosed with metastatic prostate cancer.
VG had received prostate cancer screening, in the form of yearly PSA tests, for 14 years, from age 56 to 70. During this time, his PSA ranged from 0.8 to 1.2; the last was 1.0. At a clinic visit 12 years before the admission, when he was 70, his primary care doctor documented a discussion about discontinuing screening. The doctor had explained that PSA screening was probably no longer worthwhile, that the harms outweighed the benefits. The doctor reasoned that VG’s life expectancy (given his age and other medical problems) was less than 10 years and noted that the USPTF does not recommend PSA screening in patients over 70. VG said he understood the reasoning and was fine with discontinuing PSA checks.
VG is an amalgam of a few patients that I have been planning to write about. Then, on May 18th, we got news that makes this essay seem “ripped from the headlines.”
Anyone who has followed Sensible Medicine knows we are obsessed with cancer screening.1 Theoretically, screening for disease makes sense. It is better to treat something early than late, you know, an ounce of prevention is worth a pound of cure. But screening is complicated. We probably save thousands of people every year with screening tests, but:
We falsely reassure some people by telling them that they do not have disease when they do (false negatives, related to <100 % sensitivity of tests);
We tell some people that they have disease when they do not (false positives, related to < 100% specificity of tests and low disease prevalence in screened populations), and this leads to harm through anxiety and follow-up testing to prove false positives;
We diagnose and treat some people for disease that would never have caused them harm (overdiagnosis).
So, how should we think about someone like the vaguely fictional VG or the very real former president, a patient who gets sick with a disease after the screening period is over. I think there are three ways of looking at the situation.
Our guidelines failed the patient
This is the most simplistic way to look at the situation, but not an incorrect way. A patient and his doctor followed our screening recommendations, and the patient was still diagnosed with metastatic cancer. This is to be expected, even during the years of active screening (55-69 years old for prostate cancer, 21-65 for cervical cancer, 40-74 for breast cancer), people still die of these diseases. None of our screening tests are 100% sensitive, perfect for ruling out disease, and some cancers are lethal even when detected at an early stage. This is reflected in the 0.5% of men in this USPSTF prostate cancer infographic who die of prostate cancer despite screening (or the .4% who die of breast cancer despite mammography).
Our guidelines sacrificed the patient
This way of looking at the situation is similar to the first. Yes, our screening failed this patient, but it was for the greater good. What would happen if we continued to screen men for prostate cancer throughout their 70s? We would find a lot more prostate cancer. But remember, the goal of cancer screening is not to find cancer. The goal of screening is to decrease mortality.2 Every medical student knows that more people die with prostate cancer than of prostate cancer. If we continued to screen people late into life, we would harm a lot of men while helping a few. The harm would come in the form of anxiety, unpleasant and dangerous biopsies, and treatment of people destined never to be harmed by their disease. Not screening men in their 70s saves a lot of harm, but means that a few people, whose early-stage disease could have been cured, will progress to metastatic disease or death.
Our guidelines served the patient well
The final way of interpreting this case is that the guidelines led to perfect management. VG was screened for disease when he was younger, when early detection was most likely to help him. If he had developed prostate cancer at 55, it would have been more likely to be a high-grade cancer. His youth would have also meant that he had years to be affected by the disease. In his 70s, when he was no longer being screened, he was spared a diagnosis for which he didn’t need to be treated early. Early treatment of preclinical disease would not have improved his outcome. He was able to enjoy his 70s, not knowing he was a cancer patient and not having to spend time with urologists or radiation oncologists.
After diagnosis, VG was started on an antiandrogen. He tolerated therapy and, 3 months after diagnosis, his PSA was < 1.
Cancer screening and presidents
I hesitate to write too much about Joe Biden’s health. There are facts that we do not, and probably will not, know. We can ask the same questions we asked about VG. Did our screening guidelines hurt him or help him? With Biden, we can also ask, would screening in his 70s have changed the course of history? Although differential treatment for those in positions of power is usually anathema to me, would we consider different guidelines for presidents and presidential candidates? We have special health requirements for other jobs.
Conclusion
For any individual patient, we truly don’t know which of the three interpretations offered above is correct. Thinking about screening is challenging because we generally lack good data about the effectiveness of our screening interventions. The data we have speaks to populations rather than individuals. While this is true for evidence-based medicine in general, it is more important when you consider screening. We never know the counterfactual. Would VG have done better with early detection, or would he have suffered more?
If you want a deeper dive into the issues around cancer screening, here is some Sensible Medicine content. There is a video debate that Vinay, John, and I had about colon cancer screening. We also posted two follow-up articles after that debate. I have written six different articles about screening. Lastly, John and I covered a recent screening article in the This Fortnight in Medicine Sensible Medicine Podcast.
There is a debate about whether our goal for screening should be to decrease disease-specific mortality (prostate cancer screening should decrease mortality from prostate cancer) or overall mortality.
For me, it boils down to a simple question: if I undergo the screening, will there be a measurable increase in my life expectancy? According to the best science we have (thanks, Sensible Medicine), the answer is no. I find this kind of liberating. By following the evidence, I worry a lot less and can spend more time enjoying life, rather than hanging out in healthcare clinics. Avoiding cancer screening is not a matter of gambling; it’s part of maximizing happiness.
This was absolutely spot on! I’m a 72 year old whose PCP tried to coerce me into having a PSA in my late 60’s by threatening to fire me as a patient if I didn’t. So much for patient counseling, informed consent, and mutuality of decision making. Thank you for confirming my choice of treatment, or should I say non-treatment, that I arrived at myself, in spite of the PCP.