Prostate Cancer STHLM3 Screening Algorithm Put to the Test

Kate Johnson

March 22, 2016

MUNICH — A new assessment of the multiplex Stockholm 3 (STHLM3) prostate cancer screening algorithm suggests that, compared with current clinical screening, the test can significantly reduce the rate of unnecessary biopsies without compromising the identification of high-risk prostate cancer.

"We were quite astonished by this result," said Henrik Grönberg, MD, the study's lead investigator, who reported the results here at the European Association of Urology 2016 Congress.

Applying the STHLM3 algorithm to a cohort of about 56,000 men who had been assessed in regular clinical practice would have resulted in a 55% reduction in prostate biopsies with either benign or low-risk results, reported Dr Grönberg, who is from the Karolinska Institute in Stockholm, Sweden.

The STHLM3 algorithm includes a combination of plasma protein biomarkers (prostate-specific antigen [PSA], free PSA, intact PSA, human kallikrein 2, beta-microseminoprotein, and macrophage inhibitory cytokine 1), 232 single-nucleotide polymorphisms, and clinical variables (age, family history, previous prostate biopsy, and prostate examination results).

As previously reported by Medscape Medical News, the prospective STHLM3 study published last year (Lancet Oncol. 2015;16:1667-1676) showed that the STHLM3 algorithm performed significantly better than PSA alone for the detection of cancers with a Gleason score of at least 7 (P < .0001).

Specifically, among men 50 to 69 years of age, the STHLM3 test could reduce the total number of biopsies by 32%, compared with PSA testing only, and it could reduce the number of benign biopsies by 44%.

The results come from a study that compared STHLM3 screening with the more realistic standard of current clinical practice, which Dr Grönberg defined as a combination of total PSA, free PSA, family history, age, earlier prostate biopsies, prostate volume, and digital rectal exam.

The analysis involved 56,282 men 50 to 69 years from the 2011 Stockholm PSA and Biopsy Register. All men had undergone at least one PSA test in 2011.

The STHLM3 algorithm was then applied to the cohort, giving them an STHLM3 risk score.

On the basis of this score, the men were then matched with the STHLM3 study cohort, so that both groups had an equal number of prostate cancers that had a Gleason score of 7 or higher.

Comparing the number of biopsies in both groups, the analysis showed that the men screened under current clinical practice underwent 2796 biopsies — a number that could have been reduced by 55% with the STHLM3 test.

Additionally, 79% of the 1794 biopsies with benign results could have been prevented with the STHLM3 test, as could 27% of the 466 biopsies with a Gleason score of 6.

"Overall, today 19% of men aged 50 to 69 who undergo biopsy will have a cancer diagnosis that we most likely will treat," he commented. "If we use the STHLM3 test instead, that number will go from 19% to 43%, which I think is quite an astonishing result," he said.

"Because, if you think about it, in current clinical practice, we use a lot of the things that are included in the STHLM3 test, but what this really shows is that we don't use them effectively," Dr Grönberg said.

He said the strength of the STHLM3 test is the combination of all factors. For men with PSA levels of 1 ng/mL to 3 ng/mL, the STHLM3 test would find that 30% of biopsies would yield cancers with a Gleason score of 7 or higher, compared with just 8% among men biopsied on the basis of PSA levels alone.

"This is really showing the power of combining both biomarker and clinical assessments," he said.

Dr Grönberg said his group has also conducted a thorough health economic evaluation and has estimated that the STHLM3 would cost €200 (US$224) per test. At this price, "this test will actually save money from day 1 with the savings in the number of biopsies."

His group plans to calibrate this test in clinical practice this spring, and then validate it in other populations around the world later this year.

Chair of the session Felix Chun, MD, from University Medical Center Hamburg-Eppendorf in Germany, said he was impressed with the findings.

"I think the study is a fantastic example of how we should proceed with screening strategies. It's prospective, generalizable, and it has been validated," he commented.

Approached for comment, Alastair Lamb, MBChB, PhD, senior fellow in urology at the Peter MacCallum Cancer Centre in Melbourne, Australia, who cowrote the comment (Lancet Oncol. 2015;16:1579-1580) that accompanied the initial publication of the STHLM3 results, welcomed the new results.

The increased diagnosis rate of biopsies "is, indeed, an impressive statistic," he said.

However, Dr Lamb told Medscape Medical News that "the field is moving too much in a number of directions for this to be the final answer."

"This is a promising step forward in the quest for a next-generation screening tool for prostate cancer, but...there will be a need to assess this alongside MRI scanning, which has been shown to increase the hit rate on biopsies to 80%," he said. "There are also other noninherited genetic changes that give a stronger risk of disease and should probably be included in the panel."

In addition, he pointed out that given the slow-growing nature of prostate cancer, it will be important to see in the longer term whether the STHLM3 algorithm reduces mortality for men with prostate cancer.

Dr Grönberg, Dr Chun, and Dr Lamb have disclosed no relevant financial relationships.

European Association of Urology (EAU) 2016 Congress: Abstract 86. Presented March 12, 2016.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.