Time to scrutinise prostate cancer tests

A lack of standardisation among PSA tests for prostate cancer is worrying medics and saw a patient at Connolly Hospital receive an artificially elevated reading.

THERE was a time when referring men for a biopsy on foot of concerns about the presence of prostate cancer was largely down to the discretion of the treating GP.

As Peter Ryan, consultant urologist at the Bon Secours Hospital in Cork points out, it is well recognised that the test is controversial and that different PSA (Prostate Specific Antigen) testing systems can give different results.

This was highlighted in a paper by Dr JC Forde, Department of Urology, Beaumont Hospital, published in 2011. After comparing six PSA testing methods in nine hospitals, Dr Forde found PSA values were inflated across eight of the nine hospitals when compared to the reference hospital.

However, when standardised test methods were used, variability in PSA values reduced. Dr Forde concluded that the lack of standardisation of PSA tests “leads to significant variability in the measured PSA values for the same patient samples”.

The findings, he said, “stress the need for nationwide standardisation of PSA testing”. Unfortunately, as Mr Ryan points out, such standardisation does not currently exist here. If it did, the problem that has occurred at Connolly Hospital where a specific PSA test kit has produced readings higher than the international reference standard, may not have occurred.

Mr Ryan added that a PSA standardisation committee is addressing this issue under the auspices of the NCCP (National Cancer Control Programme). The PSA test measures the blood level of PSA, a protein that is produced by the prostate gland. A high PSA level may be an indicator of cancer but Mr Ryan is keen to highlight there are other reasons for elevated results. “PSA levels can be raised for a number of reasons, not just the presence of prostate cancer. These include benign prostate enlargement and prostatitis (inflammation of the prostate) Mr Ryan said.

Because PSA is subject to non-cancer related fluctuations, Mr Ryan said doctors rarely act off the results of a single PSA test, and a rising PSA is also important.

“It’s more the change of PSA with time rather than a reading in isolation that GPs and urologists look out for, the focus is on a rising trend.”

Where clinicians did observe a rising trend, the next step was biopsy, but prior to the introduction in 2011 of Rapid Access Prostate Clinics (RAPC), there was “never a rigid cut-off point” above which doctors would automatically send patients forward for further diagnostic investigations.

“GPs had significant leeway in that regard. However that has changed since the introduction of the RAPCs,” Mr Ryan says.

The game changer in terms of referral criteria came with the opening of the RAPCs, operational in eight hospitals since 2011.

“Now we do have more specific cut-offs. GPs are expected to refer a patient to a RAPC if two successive tests come in above a certain cut-off point,” Mr Ryan says. “In the era of the RAPCs, patients with a 23% (PSA level) mark-up on two successive tests could end up having biopsies they might not have had if the lower PSA result had been obtained.

This 23% mark-up was the “average positive bias” produced by the IMMULITE systems test kits manufactured by Siemens Healthcare Diagnostics — and used to measure PSA levels in patients attending Connolly Hospital in Dublin.

When Siemens discovered the “bias” — and how it came to light is unclear — it implemented a global voluntary recall of the kits in question, conceding that the high readings “may impact clinical interpretation of test results and in some cases, increase the likelihood of a decision to initiate additional diagnostic tests”.

Siemens has advised its laboratory customers to discontinue use of the affected kits — released from Feb 2012 through May 2013 — and to discard any remaining in inventory.

According to Siemens, all affected healthcare facilities have been notified.

A patient who contacted the Irish Examiner having undergone PSA testing in Connolly — where his PSA level were recorded as 40% higher than a subsequent test using a different kit at Beaumont Hospital — said he contacted the newspaper out of anger and frustration. No-one had contacted him to tell him there was a problem with the test kits. As a result he endured “weeks of stress”.

“I only found out when I took it upon myself to collect the PSA test results from my GP as I was due to meet my consultant the following week. I called to the surgery and picked it up from reception and decided to open it. It said my results at Beaumont were 40% lower than when the Siemens method was used.” When he questioned his GP about what the report meant, he said he was told there had been “problems with the PSA Siemens machine in Connolly for months” and that a lot of results had come back with high PSA levels.

The man said he was fed up with a lack of transparency in the health service, that his own experience (three PSA tests) had cost him a fortune and considerable stress.

“I mean if the problem at Connolly had continued, my GP would probably be recommending a biopsy by now,” the man said.

As Mr Ryan says, the risks associated with biopsies range from mild to fatal. At the mild end of the scale patients face urinary tract infection, bleeding, prostatitis. At the upper end, complications can be as serious as septicaemia, even death.

It seems hard to understand why the Health Service Executive (HSE) did not take it upon itself to write directly to patients affected by the Connolly kit problem. Instead the HSE sent a written memorandum on Jul 1, a couple of days after the Siemens kit recall, to GPs and urologists in the Dublin area. When asked why the HSE had not notified patients directly, a HSE spokesperson said it was up to their referring doctors to pass the message on.

Siemens has recommended a two-month look-back period from test results for certain patients. “This would allow clinicians to review all factors contributing to previous decisions regarding further diagnostic testing,” Siemens said.

However, Mr Ryan says: “If this test has been withdrawn, and it’s recognised that there needs to be a look-back at the outcome of tests, I would be concerned that the two month look-back might not be sufficient.”