Questions Remain Over Role of PCA3 Assay in Prostate Cancer

Liam Davenport

January 12, 2015

The role of the prostate cancer antigen 3 (PCA3) assay (Hologic/Gen-Probe) in ruling out prostate cancer in men presenting for a repeat biopsy has been supported by a large American study.

However, the place of the assay in assessing men presenting for an initial biopsy is less clear. It remains to be seen whether the results should be used as a continuous variable or dichotomized with cutoff values, and the value of the PCA3 assay in combination with other markers and patient-related data is still being examined.

The study was published in the December issue of the Journal of Clinical Oncology.

The findings are welcomed by Andrew J. Vickers, PhD, attending research methodologist at the Memorial Sloan Kettering Cancer Center in New York City, who wrote an accompanying editorial.

"If we wanted to say to somebody tomorrow where [the PCA3] could be used, then the role would be in repeat biopsy, because you could reduce the number of repeat biopsies quiet substantially without missing many high-grade cancers at all," Dr Vickers told Medscape Medical News.

However, he added, "in the initial biopsy setting, the risk that you'll have a low PCA3 but still have an aggressive cancer is higher than most urologists would be comfortable with."

Study Details

The study was conducted by John Wei, MD, from the Department of Urology at the University of Michigan in Ann Arbor, and researchers from 10 other centers across the United States. The study involved 859 men (mean age, 62 years) who were scheduled for a diagnostic prostate biopsy. A family history of prostate cancer was reported by 29% of patients, and 16% had an abnormal digital rectal examination (DRE).

The PCA3 test had a positive predictive value of 80% for men who presented for an initial prostate biopsy. For PCA3 scores above 60, the sensitivity for prostate cancer was 0.42 and the specificity was 0.91.

The PCA3 test had a negative predictive value of 88% for men who presented for a repeat biopsy. For PCA3 scores below 20 in these men, the sensitivity for prostate cancer was 0.76 and the specificity was 0.52.

Importantly, the team found that adding the assay to an individual risk-estimation model that incorporated age, race/ethnicity, previous biopsy, prostate-specific antigen (PSA), and DRE significantly improved the stratification of any prostate cancer and high-grade cancer, increasing the area under the curve for both initial and repeat biopsies.

"These results empirically demonstrate that overdetection of low-grade cancer in the repeat biopsy setting and underdetection of high-grade cancer in the initial biopsy setting can be improved through the adoption of new biomarkers such as PCA3," Dr Wei and colleagues write.

Questions Remain

In his editorial, Dr Vickers describes the study as "large and rigorous."

However, he questions the dichotomization of the PCA3 assay, through the use of cutoff values, because it "does not allow individualization of care" and makes it difficult to integrate the data with other findings.

Dr Vickers suggests that the end point for markers studies such as this should be high-grade cancer, not a positive biopsy or any cancer, because the treatment of men with low-grade prostate cancer is of little benefit.

He said he thinks there should not be just one test for prostate cancer, and that integrating information from tests such as PCA3 into a risk-calculation score is not important. It is important, however, that "what the doctor and patient have in front of them is the risk that the patient actually has aggressive disease and that the decision is made on a single number representing risk, rather than a plethora of numbers saying this is above the threshold and this is below the threshold, and this one's close to the threshold and that one's not," he explained. "I just don't know how you can make a decision on that basis."

He provided the example of a patient who has not had a previous biopsy, has a PSA level that is quite high, a free/total PSA ratio in the mid-range, normal DRE results, and a low PCA3 score that just missed the threshold.

"We know that people are very bad at integrating quantitative information like that," Dr Vickers said. However, "you combine all that information quantitatively into a single statistical model and you give the doctor a single number, which is the risk that the patient has high-grade disease."

Reinforcing the Intended Use

"In the repeat biopsy setting, I would say the findings absolutely reinforce the current intended use for PCA3," said Jack Groskopf, PhD, senior director of oncology research and development at Hologic/Gen-Probe.

In the United States, the current intended use is for repeat biopsy, whereas in Europe, the indication "is a bit broader," Dr Groskopf told Medscape Medical News.

For an initial biopsy, this study "used a threshold for ruling in an initial prostate biopsy," he continued. "I think the unclear part is a result of the fact that they didn't use a threshold designed to 'safely' rule out an initial biopsy. That's the area where a little further investigation is warranted."

Dr Groskopf acknowledged that "if you are trying to do more personalized risk assessment, the continuous PCA3 score is better suited for that purpose" than the dichotomized score.

"We would all like things to be more black and white, of course, and in cancer they definitely aren't. But there is continued interest from clinicians in a threshold, for example, below which the negative predictive value would be quite high," he explained.

However, "there is some value in having a low threshold as a rule-out and a high threshold as a rule-in," he added.

Dr Groskopf said he believes that the assay has a role in risk calculators. "We have always viewed PCA3 as an independent piece of information to be used in conjunction other patient information or so-called standard-of-care algorithms to help with biopsy decisions," he said.

Intended-use statements in both the United States and the European Union "indicate that PCA3 should be used in that manner, and that clinicians don't use a single test in isolation," Dr Groskopf noted. However, "it can get a little bit tricky because there will be a lot of different calculators out there."

PCA3 'Not Clinically Useful'

A draft guidance on the role of PCA3 and another test — the Prostate Health Index (PHI) — was recently issued by the National Institute for Health and Care Excellence (NICE) in the United Kingdom, as reported by Medscape Medical News.

It states that the PCA3 assay does not improve the diagnosis of prostate cancer in patients with negative or inconclusive results on prostate biopsy sufficiently to be recommended for routine use in clinical practice. Specifically, the guidance states that adding either of these tests to clinical assessment plus MRI is unlikely to improve diagnostic accuracy.

Dr Groskopf noted that the findings of the current study were not available to NICE before the draft guidance was issued. "At the time, they looked at just using PCA3 in combination with clinical assessment, independent of MRI. In that case, their conclusion was there was some improvement in diagnostic accuracy."

"But these meta-analyses can get pretty complicated, so they also said there is too much uncertainty for the impact of using PCA3 in clinical practice to be clear," Dr Groskopf explained.

"I would add that there is a large body of published literature out there demonstrating that PCA3 and PHI can be helpful when using them to refine risk for biopsy decision-making," he said.

There are really only a couple of studies on PCA3 in the MRI setting, "so I would say the jury is still out, and I would say that the wording in the NICE report is pretty balanced," Dr Groskopf added.

Dr Vickers pointed out that MRI is not routinely used in the United States for diagnosing prostate cancer because it is markedly more expensive than in the United Kingdom. "That may be why there's more interest in MRI in the United Kingdom as a kind of routine test before biopsy."

The study was supported by US National Cancer Institute Grants, and some of the reagents for this trial were provided by Gen-Probe. Coauthor Ziding Feng, MD, from the Fred Hutchinson Cancer Research Center in Seattle, reports acting as a consultant for Gen-Probe. Coauthor Ian Thompson, MD, from the University of Texas in San Antonio, reports consulting for Exosome Diagnostics. Coauthor Adam Kibel, MD, from Harvard University in Cambridge, Massachusetts, reports consulting for sanofi-aventis, Dendreon, and Myriad Genetics. Dr Vickers reports owning stock in Opko, being coinventor of the 4Kscore test for prostate cancer, and being the potential recipient of royalties from commercial sales of the test.

J Clin Oncol. 2014;32(36):4033-4034, 4066-4072. Editorial, Abstract


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.
Post as: