Prostate cancer test may do no good

A Column from a website called which is about the media talking about itself, has been making the rounds.

The column was written by New York physician George Lombardi, who was described as having "led the team that famously saved Mother Teresa's life in the late 1980s". It was entitled: "My Patient, Killed By The New York Times" (NYT).

Dr Lombardi's long-time patient who died from prostate cancer at the age of 73 had learnt from the NYT and other media reports that screening using prostate specific antigen (PSA) levels was inaccurate: Only 20 per cent with raised PSA actually have prostate cancer. So he declined the PSA test that Dr Lombardi urged him to have.

Despite his doctor's claims, the patient in fact has every reason to decline the test. Here's why.

Even if the test result was a true positive - showing a raised PSA from an actual prostate cancer - it can't tell if the tumour is going to turn aggressive or not. But it's critical to know this last bit, since most prostate cancers grow so slowly that patients die from other causes instead.

Studies show that 60 per cent to 80 per cent of men who die at the age of 80 or above from causes unrelated to prostate cancer do have microscopically detectable prostate cancer post-mortem. Had they been screened and biopsied in life, this non-invasive cancer would have been picked up - causing much anxiety - and treatment options considered.

Since doctors can't predict if it will become invasive, once diagnosed, prostate cancer is invariably treated. So PSA screening often leads to surgery or radiation which can result in loss of bladder control. Prostate radiation can cause collateral damage to the rectum, which means rectal pain, rectal bleeding and loss of bowel control that can go on for years. Impotence can also occur after prostate radiation or surgery as well.

In 2008, the United States government came out against routine PSA screening. In Singapore, the Health Ministry noted in 2010 that there is no evidence routine PSA screening improves death rates for prostate cancer.

It is suggested only for men aged 50 to 75, with an estimated life expectancy of over 10 years, who pass blood in the urine, have difficulty urinating, or have a family history of prostate cancer.

The life expectancy for a US male born in 1940 like Dr Lombardi's patient was 60.8 years; for comparison, it was 76.2 years for one born in 2010. When he retired five years ago at 68 and refused PSA testing, he had already exceeded his life expectancy.

True, he did have the cancer afterwards. But if he had had the PSA test, a biopsy - which is a traumatic and invasive procedure - would have ensued. That would have been followed by surgery, radiotherapy or chemotherapy, which would have led to more illness, suffering and discomfort than the cancer. Not having to go through those treatments likely gave him a better quality of life in his declining years. No normal aged man fancies living out his remaining years in adult diapers.

So while his patient applied the guidelines correctly, Dr Lombardi did not.

But a recent Journal of Clinical Oncology study shows the good doctor to be in good (or bad) company - for doctors are still screening nearly a quarter of US men aged 85 and older for PSA.

Many physicians - including my own - regard PSA screening to be an unadulterated good. However, frequent and early PSA screening is fraught with "lead time bias", which leads to overdiagnosis and over-treatment - all without improving death rates.

What is this bias? A cancer is usually diagnosed only after symptoms appear but if many apparently healthy people are screened, it can be detected before that.

Take prostate cancer which usually grows so slowly that many patients survive for decades, even without therapy. Say it takes 15 years for symptoms to appear after it sets in and another five years to kill after that.

If screening picks it up 15 years before symptoms ever develop, the "survival rate" which is always dated from the time of diagnosis is now 20 years instead of five years. But the cancer's biological behaviour has not changed one iota, so the man still dies five years after symptoms develop.

Clearly, screening can increase survival rates after diagnosis without any real improvement in cure rates. Screening gives the doctor years of "lead time" before the cancer manifests clinically, from which point the patient would survive just five years as before.

Mortality rates can be reduced only by a cure. So while survival rates may "improve" on paper if measured from the time of diagnosis, death rates may stay unchanged. Death rates may rise if increased public awareness leads to more men with symptoms coming forward for treatment, adding to the total number of prostate cancer deaths.

The five-year survival rate for prostate cancer in Singapore improved steadily from 40 per cent between 1973 and 1977 to 84.7 per cent between 2003 and 2007.

But its mortality rate rose from 1.3 per 100,000 males per year in the 1968-1972 period to 4.7 per 100,000 males per year in the 2003-2007 period.

Not only were more men diagnosed with prostate cancer, but more were also diagnosed as dying from the cancer in the latter period. This may have come about from greater awareness of prostate cancer, and encouraging more men with symptoms to seek diagnosis and treatment - who eventually died from the cancer.

Thus, prostate cancer survival rates here "improved" without more men being cured. This implies that PSA screening transformed more men into prostate cancer patients for more years of their lives, during which they likely lost bladder and bowel control, could not have sex and, generally, had a poorer quality of life.

In sum, PSA screening likely causes more harm than good for men at low risk. Only men between 50 and 75 who are at high risk - those with symptoms or a family history - should have it; males over 75 shouldn't at all.

Get a copy of The Straits Times or go to for more stories.