“Should I get a PSA?”
It depends. Who is asking, and why?
If the question is being asked by a 45-year-old man, the answer is going to be different than if it’s being asked by a 75-year-old. If a primary care clinician is asking, he/she may be confused about current recommendations regarding screening for prostate cancer. The answer is complicated and fraught with bias. Unfortunately, the controversy over PSA screening for prostate cancer still persists after 40 years of experience.
I practiced urology for 35 years. I started my residency in urology at Oregon Health Science University in 1979, eight to 10 years before urologists began using PSA.
I have witnessed the evolution of prostate cancer “screening” using PSA. I was there before the controversy started, and I’m still here in the middle of the fight. The “fight?” Yes, the fight. In one corner are the doubters, the “non-screeners,” physicians who take their cues from the US Preventive Task Force’s 2012 Level D recommendation (since revised to Level C in 2018) – “we don’t need to screen for prostate cancer by doing PSAs on our patients because most prostate cancers are indolent and slow-growing, and men aren’t going to die from prostate cancer. The treatment (surgery, radiation, hormones) is worse than the disease.”
In the other corner are urologists, oncologists, and radiation oncologists who believe that early detection of prostate cancer using PSA as an initial marker is not only possible but also saves lives. More importantly, early detection prevents morbidity from advanced and/or metastatic disease.
In 1980, PSA was still in research labs. By the mid-80s, doctors realized that PSA was a more sensitive test for detecting prostate cancer than acid phosphatase (a blood test). Acid phosphatase was elevated in advanced/metastatic prostate cancer but not as sensitive for prostate cancer as PSA. Prior to PSA, the only way to detect localized prostate cancer was by a digital rectal exam (DRE) or by investigating symptoms of bladder outlet obstruction or symptoms of advanced/metastatic disease.
Before the mid-80s, if a nodule was felt, a blind biopsy using only finger guidance was performed. A positive biopsy, normal acid phos, negative bone and CT scan meant “localized prostate cancer.” That patient was usually offered surgery (a radical retro-pubic prostatectomy) or external beam radiation therapy. This was rare. From 1979 to 1982, I assisted in two radical prostatectomies. After PSA had been around for 15 years, I was involved in performing two radical prostatectomies a week, roughly 100 a year.
Half of the prostate cancer patients in the early 80s had advanced or metastatic disease. If the patient had locally advanced disease but no metastases (Stage ‘C’), he would be treated with radiation. 25% of patients were found to have prostate cancer after a TURP, stage A1 (fewer than 5% of chips +) or A2 (greater than 5% of chips +). At the Portland VA, typically half of our patients on the ward were being treated for metastatic prostate cancer with IV stilphosterol and/or radiation therapy for distant mets, or painful bone mets in danger of pathologic fracture.
What happened over the next four decades with respect to PSA in the US (things are different in other countries) is both phenomenal and tragic. With millions of PSAs and prostate biopsies being done, the incidence of prostate cancer increased. However, the percentage of patients presenting with advanced and metastatic disease fell from 50% to less than 10%. Mortality from prostate cancer declined.
We learned that PSA was an excellent test to assess response to treatment. An undetectable PSA after treatment was indicative of an excellent response to treatment (surgery, radiation, even hormone ablation for the treatment of metastatic disease). A rising PSA after treatment was indicative of recurrence.
We learned about PSA velocity (change in PSA over time), PSA density (PSA/prostate volume), % free PSA, PCA3 (prostate cancer antigen urine test), isoPSA, and recently newer tests like ExoDx (urine biomarker). All of these variations and tests have been helpful in refining the sensitivity and specificity of PSA-based screening. Add to those tools multi-parametric MRI.
Prostate cancer incidence increases 10% each decade. A 50-year-old man has a 50% chance of having “prostate cancer” in his prostate, a 60-year-old, 60%, etc. Obviously, not all prostate cancers are “clinically significant,” meaning they are slow-growing, indolent, and will never cause any problems for the duration of that man’s life. This is why many doctors will tell patients, “You are more likely to die with prostate cancer than from prostate cancer.”
This created problems because many clinically insignificant cancers were unnecessarily treated. With overtreatment came significant complications – impotence, incontinence, and post-surgical and radiation complications.
The USPTF issued a Level D recommendation against PSA-based screening (2008 and 2012). The recommendation was based on three studies (PLOC, ERSPC, and Gothenburg). All three studies are flawed for a variety of reasons and were misinterpreted by the USPFT. Not one urologist or oncologist was on the USPTF committee. The USPFT recommendation has faced extensive criticism in the literature. Morbidity from metastatic disease and treatment is not part of their analysis. As a result of the flawed recommendation by the USPTF, there was an increase in advanced/metastatic disease, which prompted a revision of their recommendation to Level C in 2018.
Today, the two camps are still far apart on the early detection of prostate cancer. Like other conflicting beliefs in 2023, this conflict makes no sense and doesn’t have to persist. Thousands of men are still presenting with advanced and metastatic prostate cancer. It is estimated that 35,000 men will die of prostate cancer in 2023.
“Who are these men? Have they ever had a PSA or digital rectal exam?” In the mid-90s, we reviewed 100 consecutive patients who presented with advanced or metastatic prostate cancer. Many had PSAs over 100. We asked, “How many had health care insurance during the ten years prior to presentation?” 85. We then wondered, “How many had a PSA and/or a DRE within ten years of being diagnosed with metastatic/advanced disease?” Zero! (0!). And, “How many had a documented ‘shared decision-making’ conversation regarding the ‘potential harms/risks vs benefits’ of getting a PSA?” Again, zero.
The April 2023 AUA Guideline Early Detection of Prostate Cancer: AUA/SUO Guideline (2023) states, “Clinicians should engage in shared decision-making (SDM) with people for whom prostate cancer screening would be appropriate and proceed based on a person’s values and preferences (Clinical Principle).”
From the May 2018 USPTF recommendation: “Based on a review of the evidence, the Task Force recommends that men aged 55 to 69 years make an individual decision about whether to be screened after a conversation with their clinician about the potential benefits and harms. For men 70 years and older, the potential benefits do not outweigh the expected harms, and these men should not be routinely screened for prostate cancer.”
Most clinicians do not have the time to conduct a robust, factual, non-biased “shared decision-making” conversation. Moreover, it’s not the patient’s responsibility to be informed enough to engage in a conversation about whether or not to have a PSA.
The disconnect between primary care and urology need not persist. Urologists are not interested in performing biopsies on patients with PSAs (and other indicators – MRI Pi-RAD of 3-5, ExoDx > 15, and more) indicative of low-risk, indolent prostate cancer. Urologists are not interested in treating cancers that don’t need treatment. In fact, a full third of patients who have positive biopsies are offered active surveillance. Likewise, primary care physicians are not interested in conducting numerous PSAs, which leads to unnecessary biopsies and unnecessary treatment causing impotence, incontinence, and worse, only to benefit a small number of those men.
We know enough about PSA (and DRE) to inform a man of his risk of having clinically significant versus indolent CaP PRIOR to TRUS/bx. Risk calculators are readily available, and a man can then decide, based upon his risk, whether or not to have a biopsy.
If the biopsy is positive, urologists have the necessary information to have a robust SDM conversation about whether or NOT treatment is indicated, based on individual preferences, and what each treatment (potential harms, risks, benefits) entails.
For four decades, urologists have improved the specificity of PSA as an initial early detection tool. The days of a normal PSA being < 4 have long passed. Every man has an individual risk of developing clinically significant and potentially dangerous prostate cancer based on tons of good studies and evidence (mpMRI, enhanced PSA interpretation using % free PSA, PSA density, PSA velocity, isoPSA, urine biomarkers (PCA3, ExoDx), genetic markers, and more! If you want to know your risk, get a PSA and then find someone who can tell you what your risk is.
Stephen Lieberman is a urologist.