1) The consequence of the 2012 USPSTF recommendation to cease PSA testing;
2) Consideration of PSA testing for those men most likely to have cancer and possibly require a biopsy;
3) Targeting biopsies to those men most likely to harbor aggressive disease.
A substantial number of men currently are diagnosed with advanced cancer, i.e. Gleason score 7(4+3)-10. High grade cancer has a substantial prevalence (i.e., present but undiagnosed) in the population even at a “low” PSA levels.
In the control arm of the Prostate Cancer Prevention Trial, of the 2950 men whose PSA never was above 4 ng/ml, 15.2% had a positive biopsy, of which 14.9% had high grade.
Autopsy studies in the age group 40 – 50 years finds a prevalence of 20 – 30% of men unknowingly harboring the disease.
This figure increases by approximately 10% in each succeeding decade with an increasing proportion having high-grade disease.
There is an urgent need to develop diagnostic tools sufficiently focused so as to identify these men as early as possible and provide them the best chance for cure with primary therapy, … and at the same time avoid a biopsy in men whose cancer does not threaten them.
1) Consequence of the 2012 United State Preventive Services Task Force recommendation to “not test.”
Clinicians who actually diagnose and treat prostate cancer, from the very start regarded the USPSTF dictum to cease PSA testing as terribly misguided — and harmful. This recommendation was a very gross tool designed to prevent “overdiagnosis and overtreatment.” However, the adverse consequences were easily predicted. And now at the January meeting of the American Society of Clinical Oncology Olson et al. reported a study by the largest urology group in the US. They accumulated data to confirm the adverse trend and its outcome: i.e., a greater proportion of men with advanced disease are being diagnosed at initial biopsy. The ultimate outcome of this trend is a predicted increase in cancer related deaths in 5-10 years.
The 2016 ASCO abstract #68. This study compared prostate biopsy results during the 2 years prior to the recommendation (3429 men) with pathology findings in the subsequent 3 years (total, 7117 men). It found that the percentage of cancers diagnosed with Gleason score 8-10 increased over time.
In the period 2010/11 of 16,271 cores, 14.8% of positive cores had Gleason score 8-10.
In 2013, 19.7% of positives cores were Gleason score 8-10;
in 2014, 26.1%; and in
2015 25.4 % of positive cores were Gleason score 8-10. “Gleason 6 scores were stable over time.”
The authors’ conclusion: “ … there has been a persisting increase in positive cores as well as cores found with higher Gleason scores. This increases patients presenting with higher risk disease and portends a possible reversal of gains made in reducing prostate cancer mortality.”
2) Which men are candidates for PSA testing and further evaluation?
It is widely recommended that men 40 years old be tested for a baseline PSA test. Without a baseline PSA value it is not possible to individualize a schedule for further testing.
Numerous schemas for further testing have been suggested. This subject was discussed in detail in the PCa COMMENTARY Vol. 88 (click link) Many men will have in mind the off-repeated, and incorrect, myth that concern for prostate cancer begins at a PSA level above 4 ng/mL. They will be surprised that the median PSA level of men 40 – 49 years old is between 0.68 and 0.85 ng/mL.
A study by Heidegger et al. (PLoS ONE, Jul 2015) addressed: “Age-adjusted PSA Levels in Prostate Cancer Prediction: Updated Results of the Tyrol Prostate Cancer Early Detection Program.” They combined PSA values and percent-free PSA (<21%) to suggest which men should be further evaluated.
“We found that in combination with a free PSA <21% the following PSA cut-offs had the best cancer specificity for finding cancer on a biopsy:
>1.75 ng/ml for men <49 years and also for men 50-59 years;
>2.25 ng/ml, 60 – 69 years; and
>3.25 ng/ml for men older than 70 years.
“Using these adjusted PSA cut-off values all significant tumors are recognized in all age groups, yet the number of biopsies is reduced.” The adjustment of risk related to age takes its relevance from the 4-5% increase in PSA yearly in men due to the gradual enlargement of the prostate gland over time.
There is strong evidence suggesting that retesting is not indicated for men with an initial PSA of <1.0 at any age. A more conservative approach would suggest that after an initial PSA of <1.0 ng/ml the frequency of retesting can be less frequent, such as every 5, 8, or 10 years.
Dr. Ian Thompson, et al. J UROL, 2015, reported on the outcome of 2923 men followed for a median of 7.5 years. Cancer was diagnosed in 289. Those men “with baseline PSA of 1.0 ng/ml or less were at a 3.4% 10-year risk of cancer and 90% of the cancers were low risk.” The authors suggested that these men could be tested again at 10 years. By comparison, those men with PSA level of 3-10 were at a 39% risk for cancer in 10 years.
In the Commentary article PCa COMMENTARY Vol. 88 the various approaches for further testing for men with PSA values > 1 ng/ml was discussed in detail. Drs. Wilt and Scardino crafted an evidence-based schema, J Natl Cancer Inst, 2014:
- Age 45, PSA <1 ng/ml — screening every 5 years to age 50, 55, and 60; and if <1 ng/ml at age 60 no further testing;
- Age 45-69, PSA between 1 and 3 ng/ml — screening every 2-4 years;
- Age 70: for “previously screened men older than 70 years and in all men with short life expectancy (i.e. 10–15 year longevity), and men with serious medical conditions — no screening is recommended.
PSA levels above 3 warrant further evaluation (see last section below) and consideration for biopsy.
The NCCN guideline of Prostate Cancer Early Detection, Version 1.2015 NCCN.org (click link) is a bit more conservative. All the recommendations, however, are best efforts to match risk against testing frequency and harm (i.e. infection, needless follow-up studies, anxiety, and unnecessary biopsies). The general theme in all the recommendations is that by considering the patient’s age and PSA level, testing frequency can be reduced. The message clearly is that indiscriminate yearly testing is inappropriate as a general rule.
3) Targeting biopsies to those men likely to harbor aggressive cancer (e.g., Gleason score 7 (4+3), and those with scores of 8-10).
The venerable PSA test has been in service since the early 1990’s, but is now known to be a very blunt tool for the diagnosis of prostate cancer. For men with negative DRE exams whose PSAs are in the range of 4-10 ng/ml the specificity of the test for diagnosing any cancer is between 25 -35%. The majority of men will have low-grade cancer, which might better not have been diagnosed, or, if diagnosed, qualify those men as candidates for active surveillance.
This effort to hone the testing to aggressive cancers is a work in progress. Its successful future resolution will likely involve genomic analysis based on blood samples. The current best tests for identifying aggressive cancers are still based on permutations and combinations of members of the kallikrein family of serine proteases — i.e. total PSA, (human Kallikrien 3; hK3); % free PSA; two isoforms of PSA, i.e., [-2] pro-PSA and intact PSA; and the less known serine kinase, kalikrein-related peptidase-2 (hK2).
What are the currently available tests to guide clinicians in targeting biopsies to those men at greatest risk for aggressive cancer and avoid biopsies in men with slow-growing cancers?
The Prostate Cancer Prevention Trial Risk Calculator – Version 2 (cost, free) (PCPT) Prostate Cancer Risk Calculator – Individualized … is based on data from 6664 biopsies on men in the control arm of the trial. The 2012 update, version 2, incorporates %free PSA and [-2]proPSA and is designed to differentiate the risk of no cancer, and low- versus high-grade disease on biopsy. It is convenient to use, incorporating easily available data: PSA, digital rectal exam, age, race (African-American, Caucasian, Hispanic, other), family history, and prior prostate biopsy. In an article on the new version, Ankerest et al. (Urology 2014 Jun) conclude: “By differentiating risk of low- verses high-grade disease on biopsy, PCPTRC 2.0 better enables physician-patient counseling concerning whether to proceed to biopsy.”
The PCA3 urine-based test (Progenza; cost $385) predicts for cancer on biopsy on the strength of the ratio of RNA from the cancer-related PCA3 gene to PSA in the cells shed into the urine after prostatic massage. The test is a weak predictor of grade of cancer and not serviceable to guide biopsies toward aggressive disease. It is useful, however, in predicting cancer on repeat biopsy after an initial negative first biopsy. The test is performed in the physician’s office. This test is approved only in men >50 years with a prior negative biopsy and where a physician is considering another biopsy.
The Prostate Health Index (Beckman Coulter Corp; cost $125) combines the measurement of PSA; %free PSA; and [-2]pro-PSA, an isoform of PSA that is preferentially made in high-grade cancer cells. Results are reported on a scale of 0 – 100 based on a complex formula. Its performance in indicating any cancer vs benign conditions has been multiply verified in studies comparing it to PSA and %free PSA ( Loeb S, et al., J Urol. 2014 Nov). Biopsies could be avoided in 31% of men in the PSA range of 4 – 10 ng/mL (and in some studies 2 – 10) with a PHI score of < 27 (set at 90% sensitivity) ( ibid, Loeb). “A prostate health index cutoff of 24 led to 36% biopsies avoided and very few aggressive cancers missed,” (de la Calle C et al., J Urol. 2015, Jan). However, “Using a phi cutoff of 21.3% (95% sensitivity), 25% of missed cancers were Gleason score >7; therefore, careful surveillance is necessary” (Catalona et al., J Urol. 2011 May).
For informative about ordering a PHI test contact 1-855-420-7140, ext 3, or check http://www.myinnovativelab.com/ (click link). The test can be obtained in a physician’s office.
The 4Kscore Test (OPKO Diagnostics, LLC; cost ~$400) is a blood-based test that combines four elements: the PSA (hK3) and two PSA isoforms (%free PSA and Intact PSA) and human kallikrein-2 (kallikrein-related peptidase-2 [hK2]) correlated with the basic parameters of age, DRE, and prior biopsy to arrive at the 4Kscore. The isoform Intact PSA differentiates benign from malignant tumors. A low %free PSA, i.e. <20, is associated with cancer, but also carries additional prognostic implication for later metastatic cancer. hK2 is independently associated with cancer risk as compared to PSA, particularly when used in relation to %free PSA. (Haese and Partin, Current Clinical Urology: Management of Prostate Cancer, Second Edition)
The 4Kscore predicts the probability between <1% – >95% for aggressive cancer, e.g. Gleason >7 on biopsy and the likelihood of distant metastases within 20 years.
A 4Kscore value of < 7.5% predicts at a 95% accuracy the probability of metastases at 5, 10, 15, 20 years of 0%, 0.2% 1.0%, and 1.8%. Whereas a value above 7.5% predicts for metastases of 2.4%, 5.6%, 9.9%, and 16.4% for those same years. (J Clin Oncl 32, 2014, abstr 5081) The abstract bases its findings on data from the European screening trial where biopsies were done on all men with PSA 3 ng/mL or more. By restricting biopsies to only those men with a 4Kscore test level of >7.5% in men at age 60 biopsies could have been avoided in 38%. The abstract findings were peer reviewed in Stattin et al., Eur Urol, 2015.
In a study of current urological practice in 26 European centers (Parekh et al. Eur Urol. 2014) the performance of the 4Kscore was evaluated in 1012 men referred for consideration of biopsy.
Fifty-four percent of biopsies were negative and 46% positive, of which 23% had Gleason scores 7 or more. “The 4Kscore showed near-perfect calibration, with the predicted probabilities of Gleason
> 7 cancer accurately describing the true risk in the cohort.” Depending on the cutoff, 30 -58% of biopsies could have been avoided, while 1.3% – 4.7% men would experience a delay in diagnosing cancer of Gleason score > 7.
A poster presentation at the 2016 AUA meeting reported that “the 4Kscore Test can predict high grade (Gleason >7) prostate cancer even in patients with a tPSA 7 prostate cancer … .”
In a study of 611 men referred for evaluation, urologists at a consortium of 35 academic and community centers utilized the 4Kscore test and reduced the number of biopsies by 64% (Konety et al., reviews in Urology, 2015). “Results for the patients were stratified [on the basis of the 4Kscore] into low risk (<7.5%), intermediate risk (7.5% – 19%), and high risk (> 20%) for aggressive cancer.” In these groups respectively biopsies were reduced by 94.0%, 52.9% and 19.0%. “The 4Kscore risk category category is strongly associated with biopsy pathology.”
The 4K test should not be used alone, but rather as a follow-up test to an abnormal PSA and/or DRE in order to restrict biopsies to the men most likely to have aggressive cancer.
Blood draw location sites for this test can be found at www.4Kscore.com, click on Patient Service Center Locator. The test can be obtained in a physician’s office.
The Future — Next Generation Genomic Tests: With the refinements built into the PHI and the 4Kscore tests the predictive power of permutations of the kallikrein family may have been maximized. To accurately address the extensive heterogeneity of prostate cancer, microarray gene panels based on genomic signatures are likely to be the basis of the next generation of tests— and these are in development.
One example of the many that will likely be forthcoming: “Diagnostic and prognostic utility of a DNA hypermethylation gene signature in prostate cancer,” Goh et al. PLoS One, 2014. The panel differentiated benign, low-grade, and high-grade prostate cancer with an overall accuracy of 96.4%.
PSA Testing for Aggressive Prostate Cancer
BOTTOM LINE: Refinements in PSA testing have advanced the art significantly, leaving the 2012 edict to “not test” as a misguided relic. Easily available information and new tests can provide guidance in selecting which men should be considered for biopsy and which of those men are likely on biopsy to have the aggressive cancer that merits treatment.