PSA AND RISK OF PROSTATE CANCER
In a landmark study by Thompson et al published in 2004, 2950 men in the placebo arm of the Prostate Cancer Prevention Trial (PCPT) underwent end-of-study prostate biopsies regardless of PSA and digital rectal examination (DRE) findings. For those with PSA equal to or less than 4.0 ng/mL at the time of biopsy, about 15% were diagnosed with cancer. The risk increased with PSA levels but was present at even low levels. One critical question concerns the prevalence of insignificant cancers found at biopsy and in the pathological examination of prostates surgically removed. The PCPT data has now been analyzed to examine this question. PSA categories were 0-1.0, 1.1-2.5, 2.6-4. Tumors found at biopsy were categorized as potentially insignificant if they were stage T1c (detected by elevated PSA), negative DRE, PSA density less than 0.15 ng/mL/g, and no evidence of Gleason pattern 4 or 5 in the biopsy cores, Gleason score equal to or less than 6, tumor limited to less than 3 cores with no more than 50% involvement of any core. A second similar set of criteria were also applied. With the above definition of insignificant, the percentages of cancer meeting the criteria for the three strata of PSA were 51.7%, 33.7% and 17.8%. In the PSA range from 0 to 2.5, 39% had insignificant cancer or put another way 61% has significant cancer. The percentages of detected cancers that were significant increased with PSA level and for those with 2.6-4.0 mg/dL, the number was over 82%. These figures did not change much when the data is restricted to the placebo group with a normal DRE.
A subgroup underwent radical prostatectomy. Adverse pathology observed on the prostates removed surgically was defined as any of the following: Gleason score of equal to or greater than 7, extraprostatic extension, seminal vesicle invasion, positive surgical margin, or lymph node metastasis. In men with PSA between 1.1 and 2.5, 37.9% had one or more adverse pathologic features and this increased to 49.1% for the group between 2.6 and 4.0 mg/dL. Extraprostatic extension went up strongly with PSA, as did positive surgical margins. Even when the PSA was below 1.0, 15% had one adverse feature (in this case a Gleason equal to or greater than 7). The presence of adverse pathologic features indicates an enhanced risk of treatment failure.
These results strengthen the view that there is no cut-off value for PSA that can provide a high level of comfort if one is worried about having prostate cancer. Furthermore, the real risk of significant disease and disease with adverse features increased monotonically with PSA with no apparent threshold. These results also support the view that the only way to answer the question of the presence of prostate cancer is with a biopsy. But even this statement needs qualification since biopsies also miss cancers. The need for a simple diagnostic test for prostate cancer with a high level of sensitivity and specificity appears urgent. Those who think it is the PSA test are misguided or deceived.
MULTIPLE BIOPSIES EXPOSE LIMITATIONS OF PROCEDURE
A limitation with this study is that not all those with a negative first biopsy went on to have additional procedures until either cancer is detected or four or five have been performed. In this study, out of 2457 first biopsies, 810 were positive but of the remaining 1647, only 306 went on to the second biopsy which incidentally had a 29% yield of positive diagnosis. Those encouraged having repeat biopsies presented with risk patterns suggesting that cancer had been missed. This pattern repeated with the majority of individuals having negative results not having additional biopsies. On the third round, only 61 out of 220 had biopsies, with 15 positive results. For the fourth round, only 14 out 146 had biopsies, with 5 positive results, and finally in the fifth round, only 2 had biopsies with one positive result. No information was provided on the prevalence of strong indications in those with a negative biopsy at each stage in the series. Nevertheless, the results are suggestive of large numbers of cancers being missed even in repeat biopsies. The design of this study is quite different from a study reported in 2001, where all participants had agreed to up to 4 biopsies if no cancer was found after the first. The first biopsy yielded 22% positives, the second 10%, the third 5% and the fourth 4%. This study used an 8-needle protocol. Thus in the Spanish study, presumably there was some selection for high-risk candidates for each additional biopsy since they found much higher percentages of positive results.
Presumably, when a patient is told that his first biopsy is negative, he is also told that while it is unlikely that he has cancer, the probability is still significant. When a second biopsy is suggested, it is because there are clinical indications that there is a high probability cancer is present, frequently because of a persistently elevated PSA. It is nevertheless interesting that it can be missed in multiple biopsies. Thus what is held up as the gold standard in prostate cancer diagnosis is not quite 24-caret. We all, including the urology community, eagerly await both screening and diagnosis protocols that come closer to perfection. What is unfortunate is that so many individuals think that perfection has already been attained.
PROSTATE GROWTH AND PSA VELOCITY - A SOURCE OF CONFOUNDING?
% FREE PSA AND RISK OF PROSTATE CANCER IN MEN WITH A PSA EQUAL TO OR LESS THAN 2.5 ng/L
A study by Waltz et al has just been published in the journal Cancer which examines the ability of the % free PSA to stratify patients with PSA equal to or less than 2.5ng/mL in order to aid in the decision regarding the recommendation of a biopsy. Since the probability of prostate cancer in this group is about 17%, the objective was to find a way of identifying those most likely to actually have the disease. Between 1999 and 2006, 7880 consecutive men underwent an initial biopsy at two European centers (Hamburg and Milan). Of the total cohort, 1036 had total PSA equal to or less than 2.5 ng/mL. For patients in this subgroup with an unremarkable DRE, 19% had prostate cancer at biopsy whereas overall the rate in the subgroup was 23%. The researchers comment that this is somewhat higher than would be predicted from the PCPT and that this may be due to the nature of the cohort which was referred on the basis of abnormal or suggestive PSA behaviour. The present study also used more biopsy cores as compared to the PCPT study.
The % free PSA was found to provide considerable discrimination. For men with an unremarkable DRE, those with % free PSA equal to or less than 14%, 49% had prostate cancer, whereas for those with %free PSA equal to or greater than 28%, only 9% had prostate cancer. This unfortunately leaves a grey area of greater than 14% to less than 28% free PSA. For the entire group which included men with suspicious DRE, these same two cut-offs found 59% and 13% with prostate cancer. The authors point out that the finding that the % free PSA was highly useful in identifying patients with prostate cancer was also observed in a study by Catalona et al in a group with total PSA between 2.5 and 4.0 ng/mL. Also in studies that examined the performance of free PSA at initial biopsy, repeat biopsy and saturation biopsy, % free PSA was the strongest predictor of prostate cancer. However, a Korean study found that for men 50 to 65 years of age with an unremarkable DRE and PSA between 4.0 and 10.0, % free PSA did not add diagnostic advantage.
Walz et al also examined the fraction of cancers detected in the equal to or less than 2.5 ng/mL group that were deemed clinically significant among the patients undergoing a radical prostatectomy. Despite the favourable total PSA range of this cohort, 16% of the cancers had established extracapsular extension or seminal vesicle invasion and 35.6% had a pathological Gleason score of 3 + 4. It was thought that this subgroup undergoing surgery was representative of the total cohort. The authors comment that the surprisingly elevated proportion of unfavourable pathologic characteristics indicates that low total PSA values are neither indicative of nonexistent cancer nor a favourable pathology. Obviously, the need to replace PSA and even its variants with a definitive yes or no marker for prostate cancer is urgent, but it must be recognized that even the biopsy misses a significant number of cancers. We have come a long way in this field, but we appear to have a long way to go.
Prostate cancer screening in men over 75 years
A rational approach to prostate cancer treatment
Dr. Klotz believes that the time it takes for the PSA level to double is a good indicator of the aggressiveness of the cancer. If the PSA level doubles in less than 2 years or progression is noted on repeat biopsy or digital rectal examination, then the patient should be offered radical intervention. The PSA doubling time should be based on 3 separate measurements during a minimum of 6 months and the final value should be greater than 8 ng/mL before radical intervention is considered.
The median PSA doubling time (DT) among the 299 patients was 7 years and 42% of the men had
a DT of
greater than 10 years. Only 21% had a DT less than 3 years. Only 2 patients died from prostate
the first 5 years and after 8 years of follow-up the prostate cancer specific mortality in the group
was only 1%.
About 60% of the patients remained on active surveillance after 55 months, with the remaining
opting for radical
therapy either because of diagnosed progression or by personal preference. Dr. Klotz concludes
surveillance with selective delayed intervention in patients whose PSA DT is less than 3 years may
be a practical
compromise between across-the-board radical therapy for all patients with localized prostate cancer
watchful waiting with palliative therapy only.
Hand-held scanner for cancer.
New test predicts the risk of prostate cancer
SEATTLE, WASHINGTON. Researchers at the Pacific Northwest Research Institute and Baylor College of Medicine report the development of a promising new method for determining prostate cancer risk and the risk of metastasis. The researchers compared the DNA spectra (Fourier transform-IR spectra) of prostate tissue taken from younger men (aged 16 to 36 years) and older men (aged 55 to 80 years) as well as prostate tissue spectra from older men with prostate cancer and older men without prostate cancer. They found that there was a highly significant increase in DNA damage (8-hydroxypurine lesions) with advancing age. This damage could be estimated by measuring the concentration of 8- hydroxyguanine in prostate tissue from young and older men. It is known that 8-hydroxyguanine is formed when DNA is attacked by free radicals, specifically hydroxyl radicals.
The researchers also found that they could easily distinguish DNA spectra from older men without prostate cancer from those of older men with prostate cancer. They speculate that the new test may be able to predict the risk of a man developing prostate cancer by determining the extent of free radical damage at any particular point in time. Perhaps the most exciting discovery was the ability of DNA spectra analysis to determine whether the cancer had metastasized well before any damage to other organs actually had occurred.
The researchers conclude that prostate cancer is largely caused by free radical attacks on both the base
and backbone structure of DNA. They suggest that an increase in the intake of such dietary antioxidants
as vitamin E, lycopene and polyphenols may inhibit the development of prostate cancer.
Editor's comment: Vitamin-C (ascorbic acid) is also an effective scavenger of hydroxyl radicals and helps prevent the formation of 8-hydroxyguanine.
Thyroid hormone and prostate cancer
Researchers at Mount Sinai School of Medicine now report that the thyroid hormone, triiodothyronine (T3), may be a useful marker for prostate cancer aggressiveness. It is well established that T3 is required for the growth of prostate cancer cells in vitro. The researchers measured T3 levels in 208 men aged 46 to 96 years. Twenty of the men had an enlarged prostate (benign prostatic hyperplasia or BPH), 161 had localized prostate cancer, and 27 controls had neither BPH nor prostate cancer. The researchers found that men with BPH had significantly higher levels of T3 than did the controls and the prostate cancer patients. Prostate cancer patients also had significantly higher levels than the controls, but lower levels than the men with BPH.
The researchers conclude that T3 may be a useful biomarker for prostate cancer, but that more work is
required to definitely establish this. They also suggest that new therapies for BPH and prostate cancer
could perhaps be directed towards inhibiting the mitogenic (cell dividing) effects of T3.
Prostate biopsies often unnecessary
PSA Screening revisited
BOSTON, MASSACHUSETTS. Dr. Michael Barry of the Harvard Medical School and the Massachusetts General Hospital provides an excellent review of the current status of PSA screening for prostate cancer. Dr. Barry starts out by posing the question "Should a 65-year-old man with no risk factors for prostate cancer except his age and with a normal digital rectal examination undergo a PSA (prostate-specific- antigen) test?" Dr. Barry points out that whether or not to have a PSA test is controversial because of the following:
He also emphasizes that the PSA test is not that accurate. A recent large-scale trial showed that using a
cut-off point of 4.0 ng/mL would pick up 46 per cent of cancers that would occur within the next ten years
with an accuracy of 91 per cent. The average age of the test group was 63 years. Among older men with
benign prostatic hyperplasia (enlarged prostate) the accuracy may be as low as 54 per cent leading to
many unneeded biopsies and much unwarranted anxiety. It is estimated that 75 per cent of men
undergoing a prostate biopsy because they have PSA levels between 4 and 10 ng/mL do not have
cancer. On the other hand, there is also a 10 per cent chance of harbouring cancer even though the
biopsy shows nothing.
Several studies have shown that providing this information significantly reduces the proportion of men
who decide to be tested.
Should men be screened for prostate cancer?
Prostate cancer reaching epidemic proportions
Is screening for prostate cancer worthwhile ?
Are prostate cancer rates really growing?
Screening for prostate cancer not recommended
Cancer screening comes under increasing scrutiny
Accuracy of PSA test questioned
More tests lead to more surgery
Screening for prostate cancer comes under fire - again!
In a separate research paper Dr. Brian Cox of the Otago Medical School in New
Zealand echoes the Australians' recommendation and points out that no proper
trials have ever shown that screening has an effect on prostate cancer
mortality. He concludes that recommending asymptomatic patients have a PSA
test is unethical.
New early warning signal for prostate cancer