The issue of PSA screening has been in the news lately. For instance, an article in USA Todayreported the latest recommendations of the US Preventive Services Task Force (USPSTF): doctors should no longer offer the PSA screening test to healthy men, because the associated risks are greater than the benefits. The story was accurate and explained the reasons for that recommendation. The comments on the article were almost uniformly negative. Readers rejected the scientific evidence and recounted stories of how PSA screening saved their lives.
It’s not surprising that the public fails to understand the issue. It’s complicated and it’s counterintuitive. We know screening detects cancers in an early stage when they are more amenable to treatment. Common sense tells us if there is a cancer present, it’s good to know about it and treat it. Unfortunately, common sense is wrong. Large numbers of men are being harmed by over-diagnosis and unnecessary treatment, and surgery may not offer any advantage over watchful waiting.
The natural course of the disease
The natural course of prostate cancer is different from that of most cancers. It can be aggressive with early metastasis and death; but more often it is small, slow growing, and non-invasive. In the latter case, the patient dies of something else before the prostate cancer ever causes any harm. Autopsy studies of men who died of other causes found a prevalence of prostate cancer that increased with age and was as high as 80%. Obviously those 80% of men would not have benefited from pre-mortem diagnosis or treatment of their cancers. But prostate cancer does kill. Death rates per 100,000 population vary by race and ethnicity from around 10 for Asians to around 50 for blacks. It would be nice if we could clearly differentiate those cancers likely to kill from those that are harmless. We can’t. So we have felt compelled to treat every cancer we diagnose.
The PSA test doesn’t give simple yes/no answers. Using a cut-off of 4.0 ng/mL, only about 25% of men who have biopsies for elevated PSA levels are found to have cancer. And 15% of biopsies in men with lower levels of PSA also detect cancer. PSA levels can be elevated for non-cancer reasons like an enlarged prostate (benign prostatic hypertrophy or BPH), inflammation of the prostate, urinary retention, a recent rectal exam, a recent biopsy, or even recent ejaculation. Most doctors look at repeated tests, the degree of elevation, and increases over time before deciding to biopsy.
Making a diagnosis
If PSA results are worrisome, the next step is a needle biopsy, usually done through the rectum with ultrasound guidance. Biopsies are not yes/no tests either. Typically 12 areas are sampled. This leaves 99% of the prostate unexamined, so a cancer could easily be missed. The more sites biopsied, the more likely a cancer will be diagnosed and the less likely that the cancer will be one of those likely to kill the patient, so it’s difficult to decide how many biopsies are optimal. When initial biopsy results are negative, repeat biopsies are positive in 25-30% of cases. When cancers are detected, they are assigned a Gleason score based on their microscopic appearance. Gleason scores range from 2 to 10, with 10 having the worst prognosis.
For the typical newly diagnosed patient the treatment options are surgery or radiation. There are other treatment options for advanced or metastatic cancer, including hormones and chemotherapy. New treatments being studied include cryosurgery and focused ultrasound.
Radiation can cause impotence, urinary problems, and an increased risk of other cancers. Surgery can cause impotence, incontinence of urine and stool, and for some unknown reason the penis may be 1-2 cm shorter after radical prostatectomy. Despite pre-operative counseling, men tend to be surprised and disappointed when they develop impotence and incontinence. A new study in the Journal of Urology found that / a year after surgery, 46% reported that incontinence was worse than they had expected and 44% said sexual function was worse than expected.
The no-treatment option
A recent randomized controlled study in The New England Journal of Medicine followed a group of men with localized prostate cancer for 12 years and compared the outcomes from radical prostatectomy versus observation alone. They found that overall, surgery did not significantly reduce deaths from prostate cancer or all-cause mortality, although for a subgroup of men with a very high PSA level (over 10 ng/ml) there was a clear benefit from surgery. The study also collected data on the side effects of surgery. Within 30 days of surgery, 1/5 of patients had serious complications including deaths. Two years after surgery, 17% were incontinent, 81% had erectile dysfunction, and 12% had bowel dysfunction.
The USPSTF recommendations
I won’t review the studies the USPSTF based their recommendations on. They did a thorough review of the literature and explained their findings in detail. They acknowledged that the evidence was conflicting but concluded that PSA screening detects more cases of prostate cancer but results in small to no reduction in mortality while resulting in substantial harm to many men. They recommended against PSA screening.
Not everyone agrees. Catalona et al. wrote about “What the U.S. Preventive Services Missed in Its Prostate Cancer Screening Recommendations” in the Annals of Internal Medicine. The American Urological Association has also spoken out. Urologists and family physicians may have different perspectives because they see a different patient population.
A new article in The New England Journal of Medicine, “Quality-of-Life Effects of Prostate-Specific Antigen Screening,” revisits a large European study that showed a 29% reduction in prostate cancer deaths in men who underwent PSA screening. They used quality-adjusted life-years to measure both harms and benefits. Their findings indicate that the net effect of prostate-cancer screening could be a loss or a gain, depending on patients’ feelings about the downstream consequences of screening (such as the long-term consequences of impotence and incontinence from treatments that may have been unnecessary). Their findings support a shared decision-making process.
Most medical groups have accepted the USPSTF recommendations. If patients request testing, they recommend discussing the pros and cons and jointly making individual decisions based on risk factors and patient preferences. (Risk factors include age, ethnicity, family history, obesity, diet, and possibly smoking.) There are no simple answers, and we wouldn’t want to miss men with PSA levels above 10 ng/mL where treatment would make a difference.
Medicine is complicated.
This article was originally published in the Science-Based Medicine blog.