The risk of prostate cancer increases with age. Men aged 50 or less have a risk of less than one in 30000. This rises to one in 625 for men above the age of 75 years. The corresponding figures for the USA are one in 476 and one in 9 and Japan are one in 31250 and one in 444. The US rates are about 70 times more than Mumbai or Japan.
Race and environmental factors appear to play an important role in prostate cancer. The table below shows the prostate cancer incidence in the US in 2010. The prostate cancer incidence (per 100,000 population) in the US increased from 94.02 in 1975 to 237.21 in 1992 before settling down to 163.06 in 2006. In the same period the annual death rates increased from 30.97 to 39.22 before finally settling down to 23.56. The five year survival which was 68.9 in 1975-1977 increased to 99.9 in 1999-2005. What were the reasons for this apparent epidemic of prostate cancer?
Early diagnosis of cancer means better cure. Screening tests, tests that are performed on individuals with no symptoms to detect occult cancer, are used with the hope of improve the survival of cancer patients. The Pap smear for carcinoma cervix, mammography for breast carcinoma and stool occult blood testing for early detection of colorectal carcinoma are few such tests. Prostatic specific antigen (PSA) is a protein exclusively produced by the prostate gland. PSA estimations with digital rectal examination are used to screen for prostate cancer.
Before the introduction of prostate cancer screening, a large number of patients with prostate cancer presented with locally advanced or metastatic disease. Only palliative treatment could be offered to these patients. Use of PSA has increased the detection rate of prostate cancer and has been responsible for the increase in the incidence of prostate cancer seen during and after the 1990s. Patients whose prostate cancer is detected by PSA screening have smaller and lower grade disease. The disease is more often restricted to the prostate allowing curative treatment. There is also a belief that some cancer so detected would never have caused a problem.
Is there evidence from clinical trials to support the contention that early diagnosis of prostate cancer means better treatment? Why do you need evidence to support something that is so obvious and common sense? It is necessary to understand the cost of a positive screening test.
Sensitivity is the number of patients with a disease who have a positive test. A sensitivity of 100% means no individual with disease has a negative result. The primary aim of a screening test is not to miss out any individual with disease. An individual who is declared to be free of disease by a screening test would become complacent, take symptoms lightly and is likely to present with advanced cancer. A screening test needs to be very sensitive to ensure that no individual with disease is declared disease-free.
Sensitive tests overdiagnose. Specificity of a test determines how many individuals the test diagnoses incorrectly as having disease. A specificity of 100% means there is no incorrect diagnosis. In practice there are tradeoffs between sensitivity and specificity. Sensitive tests tend to be less specific. A disease like cancer needs to be diagnosed with certainty before treatment is initiated. A confirmatory test needs to be performed in patients with a positive screening test to make a definitive diagnosis of cancer. The confirmatory tests come with costs, discomfort and risks.
A positive PSA is followed by a rectal ultrasound and multiple transrectal biopsy of the prostate. Even if the results of confirmatory tests are negative, patients worry about being diagnosed prostate cancer (J Gen Intern Med. 2006 Jul;21(7):715-21). Prostatic biopsy is associated with fever, pain, hematospermia, hematuria, positive urine cultures, and rarely sepsis (Urology. 2002 Nov;60(5):826-30). Finally, radical proctectomy, the treatment for localized prostate cancer, is associated with incontinence and sexual dysfunction. The critical question given the risks, complication and cost of confirmatory tests and therapeutic intervention is, does prostate cancer screening reduce mortality?
Many trials evaluating the ability of PSA and digital rectal examination to reduce prostate cancer mortality have been conducted, with conflicting results. One of the recently published studies, the ERSPC study, one of the studies with a positive result, concluded that 1410 men would have to be screened, and an additional 48 men would have to be treated (N Engl J Med 2009; 360:1320-1328) to prevent one prostate cancer related death. In societies with a lower incidence prostate cancer the number to be screened would be higher. This trial showed a 20% reduction in prostate cancer related mortality by screening. All experts agree on two issues. The benefits of prostate cancer screening, if any, are seen after a prolonged period (10 years or more) and the harms are apparent immediately.
Some prostate cancers, the low grade and small ones, will never cause problem and may not need treatment. Prostate cancer is a disease of the elderly, a group that suffers from other chronic illnesses like diabetes, hypertension, heart disease and stroke. Many patients with low grade prostate cancer would die with the prostate cancer rather than of prostate cancer. The focus in these patients needs should be the co-morbidity. As oncologists understand indolent prostate tumours better and are able to target treatment to those who are likely to succumb to prostate cancer, the impact of prostate cancer screening will become more evident. We are moving towards that goal but are not there.
What do we tell our patients? The American Cancer Society recommends a digital rectal examination and PSA beginning from the age of 50 years. Having a brother or a father with prostate cancer doubles the risk of prostate cancer and having two or more first degree relatives increases the risk seven to eight fold. These patients need to be screened from an earlier age (40 or 45 years). The patients must be informed that screening will result in further tests that can cause morbidity. The benefits if any would not be seen for at least 10 years after starting screening. The morbidity of the tests will on the other hand be immediate. Patients with less than 10 years to live should not be screened.
Finally, with so much controversy about prostate cancer screening, one wonders if what physicians practice and preach are different. Yes, physicians practice what they preach. About 95% of male urologists and 78% of primary care physicians who are 50 years of age or older report that they have had a PSA test themselves (J Gen Intern Med. 2006 Mar;21(3):257-9).