Join the 200th Anniversary Celebration

Perspective

Stratifying Risk — The U.S. Preventive Services Task Force and Prostate-Cancer Screening

Fritz H. Schröder, M.D.

N Engl J Med 2011; 365:1953-1955November 24, 2011

Comments and Poll open through November 30, 2011

Article

On October 11, 2011, the U.S. Preventive Services Task Force (USPSTF) issued a draft report on prostate specific antigen (PSA)–based screening for prostate cancer, giving it a grade D recommendation. Grade D means that “the USPSTF recommends against the service” because it has concluded that “there is moderate or high certainty that the service has no net benefit or that the harms outweigh the benefits.”

This recommendation contradicts the view that PSA-based screening saves lives by reducing the risk of death from prostate cancer. The task force acknowledges that “clinical decisions involve more considerations than evidence alone” and that “clinicians should understand the evidence but individualize decision making to the specific patient or situation.” The latter recommendation, which I support, is in line with several guidelines recommending individualized decisions after the provision of balanced information on the risks and benefits of screening. Such information is unfortunately not currently available in the form of an internationally accepted document, though a proposal for such a document is part of the Prostate Cancer Risk Calculator, which is based on data from the European Randomized Study of Screening for Prostate Cancer (ERSPC, for which I am the international coordinator), and is available online (www.prostatecancer-riskcalculator.com).

The current draft USPSTF recommendation is meant to apply to the general population of men who might be at risk for clinically relevant prostate cancer. The review and recommendations presented by the task force attempt to balance the risks and benefits of screening against those of treatment. The document presents a high-level review of current knowledge. The recommendations are in line with several sets of U.S. guidelines, as well as with the recommendations made in 2009 by the ERSPC group,1 which state that any deliberation over introducing population-based screening must take into account unresolved questions about potential harms — mainly, the problems of overdiagnosis and overtreatment.

But the USPSTF report has a number of key weaknesses. First, it relied heavily on a meta-analysis that combined higher- and lower-quality evidence. The Cochrane recommendations for establishing scientific truth state that “a systematic review of all relevant randomized controlled trials is the highest level of evidence.” Yet clearly the randomized, controlled trials included in the PSA-screening meta-analysis are of varying levels of quality and relevance and should be weighted accordingly. In an article by Chou et al. that seems central to the USPSTF report, the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial, the ERSPC, and the Göteborg screening trials are weighted equally as “fair”2 — an assessment that seems unjustified, given the report's criticisms of the PLCO trial. That trial had an excessive contamination rate (52%) in the control group; more than 40% of subjects had undergone PSA testing before randomization; the biopsy rate was only about 40% among men who screened positive; follow-up was only 7 years with 98% complete mortality data; and the power calculation was changed in 1995, with serious effects — all limitations that the task force seems not to have considered in crafting its recommendations. Moreover, according to the revised evaluation plan for the study published in 2000, a 13-year follow-up was to be required after the anticipated completion of recruitment in 2001.

The USPSTF report also notes that “overall mortality was the primary outcome addressed in all prostate-cancer screening studies assessed by the Task Force.” This statement is incorrect, at least as far as the ERSPC and Göteborg trials are concerned. Overall mortality provided a parameter for assessing randomization rather than serving as an end point in itself. Randomized, controlled trials of screening are meant to contribute to reducing cancer-related mortality in general, which is a recognized goal in the United States and around the world; reducing prostate-cancer–related mortality is an important part of that effort.

The task force report considered, methodologically correctly, only the results of the intention-to-screen analysis. The absolute and relative differences between study groups change, however, when analyses are adjusted for subjects' noncompliance with screening. The resulting data are important because they show the effects in men who are actually screened. In the ERSPC as a whole and the Göteborg substudy,3 such adjusted analyses showed relative differences of 27% after 9 years of follow-up and 56% after 14 years of follow-up in prostate-cancer mortality in favor of screening. In addition, the high number needed to screen and the number needed to treat reported in the ERSPC are likely to improve with longer follow-up, as the data of the Göteborg study show.3 And in the ERSPC, the proportion of screening-detected cancers that were considered insignificant because they were indolent is estimated to be in the range of 30%; the remainder of estimated overdiagnosis can be considered to be due to patients' life expectancy. Finally, prostate-cancer morbidity was an important secondary end point of the ERSPC, which showed that screening has a pronounced effect on the relative risk of metastatic disease, reducing it by 41%, according to preliminary analyses.1

The authors of the USPSTF document, in their search for risk factors that may increase the likelihood of aggressive prostate cancer and may therefore outweigh the potential harmful effects of screening (mainly the diagnosis of indolent disease), considered free PSA levels as well as racial and family risk factors. They failed, however, to consider published and well-documented information on readily available clinical risk modifiers, such as the results of digital rectal examination (DRE) and transrectal ultrasonography, which have been included in risk calculators as a result of multivariate analyses. For example, according to the ERSPC-based risk calculator, in a 65-year-old man who decides to undergo prostatic biopsy on the basis of a PSA level of 4.0 ng per milliliter, there is a 21% chance that cancer will be found on sextant biopsies. If the same man undergoes routine urologic examinations that include a DRE and a volume estimate of his prostate by either transrectal ultrasonography or other means, the expected outcomes of the biopsy change dramatically. If he is found to have a normal DRE, no suspicious lesions on transrectal ultrasonography, and a prostatic volume of 55 ml, his risk of a positive biopsy result decreases to 8%. With abnormal results on DRE and transrectal ultrasonography and a prostatic volume of 25 ml, however, his risk of a positive biopsy would be 65%. Thus, the inclusion of other results strongly modifies the PSA-driven indication for biopsy. The effect on a population basis has been described by Roobol et al. as a dramatic reduction in “unnecessary” biopsies and a substantial increase in the positive predictive value of PSA.4

Risk stratification is currently also the only mechanism to reduce overdiagnosis that can be applied before undertaking biopsies. For example, if we decide not to perform biopsies in men in whom the probability of finding cancer is 12.5% or lower, we will avoid performing 33% of PSA-indicated biopsies and miss only a few aggressive cancers. Until there are marker substances or imaging technologies that permit selective detection of aggressive cancers, such risk stratification should be routinely used by clinicians.4

Overtreatment can be at least temporarily avoided by offering active surveillance to men who have a low risk of prostate cancer. These men would be taking a small risk of having progressive disease that goes unnoticed.5

All these considerations suggest that PSA screening should not be dismissed as uniformly nonbeneficial. Rather, decisions about screening should be made on an individual basis, by an informed patient and his clinician, after weighing that patient's particular risk factors.

Disclosure forms provided by the author are available with the full text of this article at NEJM.org.

This article (10.1056/NEJMp1112140) was published on October 26, 2011, at NEJM.org.

Source Information

From Erasmus University Medical Center, Rotterdam, the Netherlands.

References

References

  1. 1

    Schroder FH, Hugosson J, Roobol MJ, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med 2009;360:1320-1328
    Full Text | Web of Science | Medline

  2. 2

    Chou R, Croswell JM, Dana T, et al. Screening for prostate cancer: a review of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 2011 October 7 (Epub ahead of print).

  3. 3

    Hugosson J, Carlsson S, Aus G, et al. Mortality results from the Göteborg randomised population-based prostate-cancer screening trial. Lancet Oncol 2010;11:725-732
    CrossRef | Web of Science | Medline

  4. 4

    Roobol MJ, Steyerberg EW, Kranse R, et al. A risk-based strategy improves prostate-specific antigen-driven detection of prostate cancer. Eur Urol 2010;57:79-85
    CrossRef | Web of Science | Medline

  5. 5

    Klotz L, Zhang L, Lam A, Nam R, Mamedov A, Loblaw A. Clinical results of long-term follow-up of a large, active surveillance cohort with localized prostate cancer. J Clin Oncol 2010;28:126-131
    CrossRef | Web of Science | Medline

Citing Articles (3)

Citing Articles

  1. 1

    Schröder, Fritz H., Hugosson, Jonas, Roobol, Monique J., Tammela, Teuvo L.J., Ciatto, Stefano, Nelen, Vera, Kwiatkowski, Maciej, Lujan, Marcos, Lilja, Hans, Zappa, Marco, Denis, Louis J., Recker, Franz, Páez, Alvaro, Määttänen, Liisa, Bangma, Chris H., Aus, Gunnar, Carlsson, Sigrid, Villers, Arnauld, Rebillard, Xavier, van der Kwast, Theodorus, Kujala, Paula M., Blijenberg, Bert G., Stenman, Ulf-Hakan, Huber, Andreas, Taari, Kimmo, Hakama, Matti, Moss, Sue M., de Koning, Harry J., Auvinen, Anssi, . (2012) Prostate-Cancer Mortality at 11 Years of Follow-up. New England Journal of Medicine 366:11, 981-990
    Full Text

  2. 2

    Maciej Kwiatkowski, Laurence Klotz, Jonas Hugosson, Franz Recker. (2012) Comment on the US Preventive Services Task Force's Draft Recommendation on Screening for Prostate Cancer. European Urology
    CrossRef

  3. 3

    W. G. Nelson. (2011) Profiling prostate cancer. Proceedings of the National Academy of Sciences 108:52, 20861-20862
    CrossRef

Comments (48)

48 Reader's Comments

Page

Data by Profession and Location
JAMES CAVE, MBBS | Physician | Disclosure: None
NEWBURY United Kingdom
November 27, 2011

Missing the point

Just getting your prostate cancer diagnosed does not mean a doctor has saved your life, or even given you longer to live. The science in the studies simply has not shown that to be true. I am pleased that those who have had a diagnosis are glad they went through the ordeal and feel it has made a difference for them, but I worry when they go around preaching that their life has been saved by having a PSA and getting their colleagues to get one too.
So what are the alternatives to PSA you all ask?
Get on with life, spend less time in a doctor's room and more time walking and enjoying life, and if you get symptoms, go and see a doctor. You will live just as long and perhaps with all that extra exercise a little bit more.

JEANNE LENZER | Other | Disclosure: None
KINGSTON NY
November 27, 2011

Poll query linked to article improperly framed

As a journalist, I've found that many providers who believe the USPSTF Grade D recommendation, also believe in initiating a discussion with the patient by explaining the controversy, the reasons for the Grade D recommendation, and letting them decide based on that information. By shoehorning every provider who agrees with the Grade D recommendation into an option that states they would not initiate a conversation with the patient (according to the 3 poll options), NEJM introduces an element of bias, potentially forcing some providers to check off option B (support Grade C recommendation and initiate a conversation).

Ferdinand Frauscher | Physician | Disclosure: None
Austria
November 27, 2011

Do we need imaging for successful prostate cancer screening?

One of the major problems I see in the discussion to screen or not for prostate cancer, is the lack of an imaging method, which allows for detection, localisation and size and number of prostate cancer lesions. In Innsbruck we have already introduced PSA screening in 1988. The acceptance of PSA testing was good, however the approach when PSA is elevated, the next step is systematic biopsy has shown to be not effective. Systematic biopsy is an old technique with a lot of disadvantages including inaccuracy, invasiveness and no detailed information about the cancer size. Furthermore systematic biopsy has shown often to result in undergrading of prostate cancer, which may result in inadequate therapy. New imaging techniques - including contrast-enhanced US and sono-elastography - have shown the potential to detect cancer. Furthermore multiparametric MRI is also useful. In addition the combined use of US and MRI with new imaging fusion techniques is promising. Therefore I believe that imaging will play a major role in prostate cancer screening, since the current data show that imaging may allow for diagnosis of "relevant cancers", where therapy will improve patient survival.

WILLIAM FULLER | Student | Disclosure: None
BROOKLYN NY
November 27, 2011

Missing the Point

I am disheartened by several of the comments on this page that seem to miss a main point of the discussion. Even if we grant that there is some small absolute prostate cancer-specific mortality reduction associated with PSA screening, what is the cost, both financial and in terms of personal harm, associated with one death prevented by PSA screening? Those on this forum having undergone prostate biopsy will agree that it was a painful procedure, and the costs associated with it and its complications are high. How many patients will undergo this hardship and possibly suffer lasting harm in the hopes of preventing a disease that their PSA, due to its lack of specificity, does not reliably indicate that they have?

The question is not whether PSA might find cancer, but whether ordering it will be in the interests of an individual patient. If we refuse to believe the results of two controlled trials indicating that it truly isn't, we have given up any pretense of practicing evidence-based medicine. In short, just because it is the best test we have does not mean it's a good test, or that it should be used in a screening capacity for patients without a history of prostate cancer.

IAN HOLLIDAY | Other | Disclosure: None
GRAVESEND United Kingdom
November 24, 2011

A question of values?

Although a professional, I have read these comments as a patient looking for some facts before opting for biopsy following a superb PSA velocity which should be studied. Interesting as it is that the "ding dongs" a someone calls us, are arguing about research limitations and findings but there are several anecdotes of patients who are now alive because of screening, whether via DRE, PSA or biopsy. In the face of those anecdotes, I find myself saying "Regardless of the statistics, here are people - able to speak because of life saving intervention. OK, only a few but, I am minded to say I'd like to be one of those whose life is extended and who can continue to read this web site for a few more years." Maybe I am wrong but, then again, those of you who advocate patient discussion might glean something from my, maybe twisted, mind set. Alternatively, maybe someone should do some decent research on why patients have trouble deciding, rather than, again, anecdotal stuff.? Dr Siguel above seems to be the only one making sense to me, as a patient, when he says that patients need more data and more depth.

Timothy Dayton | Other | Disclosure: None
November 24, 2011

PSA Testing

The PSA testing I had done over a number years showed a gradual rise in the last 3 years until it reached > 4.0. At that point I did not have any symptom detectable by DRE. I decided to have a biopsy and was presented with cancer in several pin samples with a Gleason of 6 or 7. I decided to have surgery, I was 55. I had successful surgery and post operative testing showed a Gleason 7 and a relatively aggressive sort of cancer. I am very glad I did what I did and I am satisfied with the results so far. Had I not had the testing I am not sure how long it would have been to have a DRE indicate a problem and then what would be my choices several years further into cancer growth. Until there is a better test the PSA is the best we have even if it isn't perfect. Early detection permitted prostate removal with wide margins and with the use of nerve sparing surgery. I have had excellent results and am not worried about this problem anymore. I do get a PSA test every 6 months just to make sure it hasn't spread.

HENRY FOURCADE, MD | Physician | Disclosure: None
Blaine WA
November 23, 2011

Retired Anesthesiologist

My father had annual H&P's and was in good health until his first PSA test result returned at 79. He had external beam radiation for invasive prostate cancer followed by a complicated medical course. I would like to continue my annual PSAs & DRE, and know when my PSA is rising before it hits 79. Is someone in this free country going to tell me that is unethical?

MR ERNEST KRAUS, RPH | Other | Disclosure: None
WOODBURY NJ
November 23, 2011

My PSA <4.0

I too had increasing PSA's but never hit that magic .4.0. Digital exam found no enlargement, firmness and a small nodule. My urologist scheduled me for a biopsy with the comment " I'm sure it will be negative, but let's be sure." An evening phone call from him said that 12 of 16 biopsies were positive with Gleasons og 6/7. I opted for surgery with continuing PSA follow up. Some years later, the PSA went from <0.01 to detectable levels. Consultation with both my surgeon and the radiation oncologist came up with a diagnosis of recurence. I also found out that I'm part of the 35% that do reoccur. I opted for XRT and continue to monitor PSA. I can only speculate what would have happened had I not had the surgery; would I still be alive? Would I have mets? No one knows. I don't care about the studies, the statistics, the cost. To me, it was worth it.

JAI JOSHI, MD | Physician | Disclosure: None
LAREDO TX
November 23, 2011

Answer to Poll Question

The four articles in this issue illustrate, there is no right answer to the poll question. The seemingly easy answer is that you initiate a discussion, but that’s the wrong choice.

john williams | Other | Disclosure: None
November 23, 2011

PSA alternatives

I keep waiting for the medical "experts"to tell me what alternative tests are available for determining the presense of prostate cancer?????

GIOVANNI SALAMANO, MD | Physician - Endocrinology | Disclosure: None
VERCELLI Italy
November 06, 2011

Prostatic cancer screening: what is the most useful screening procedure?

The hypothesis that the pathogenesis of prostatic cancer is linked to an inflammatory process seems sufficiently demonstrated by both hystological and clinical evidences supporting the presence of a precancerous lesion called "Proliferative inflammatory atrophy."
Molecular genetic supports this idea demonstrating that hypermethylation of CpG island of Gluthatione Methyl Transferase (GTP1) gene, on chromosome 11q13, is the cause of reduced transcriptional and antioxidant activity of this enzyme, causing an increased genomic instability at prostatic epithelial basal cells.
The "gatekeeper" gene for "prostatic intraepithelial neoplasia" development is NKX3.1, located on chromosome 8p21, encoding a prostatic specific homeobox gene, essential for normal prostate development, inactive in cancer, and able to repress the expression of Prostatic Specific Antigen (PSA).
PSA, also called Kallicrein 3, is itself a protein involved into inflammatory processes, present at increasing concentrations during prostatitis.
Bearing these pathogenetic considerations in mind, the measurement of PSA total, free, velocity and density for screening pourpose is unjustified in otherwise healthy patients.

john williams | Other | Disclosure: None
November 05, 2011

PSA tests

My prostate cancer was diagnosed indirectlly through a PSA test >4 followed by a free PSA and a biopsy. It was very localized but a gleeson 9. What alternative do you ding dongs have to offer in lieu af a PSA test( the DRE was clear so you can't use that answer!!!)?

SERGIO STAGNARO, MD | Physician | Disclosure: None
RIVA TRIGOSO Italy
November 05, 2011

PSA Evaluation only in presence of Protate Oncological Inherited Real Risk.

To ascertain PSA in all men, without knowing prostate cancer Inherited Real Risk, is neither scientific nor useful.

MR EDWARD JAGO | Other | Disclosure: None
DUBLIN OH
November 04, 2011

Just a survivor's comment

It was suggested that I have a PSA test by my physician some years ago. The test showed a very small level. The test was repeated first at two year intervals and than at one year intevals over a period of about 10 years. My physician plotted the data each year on a time line and we watched first a steady very low increase then an accelerating rate until the last increase showed about a thirty % increase. At that point a biopsy indicated a cancer. I elected surgery and the prostate was removed and the margin cleared. I elected no further treatment be performed. Continued PSA monitoring for the last 12 years has shown no significant PSA so I consider my self prostate cancer free. So I say montitor and watch the trend liine.

MR MARIO GUARNIERI | Other | Disclosure: None
PRATO Italy
November 04, 2011

PSA Screening does help ....

Understand that more prostate volume and more inflammation gets more PSA . However PSA high value does not mean a prostate cancer. There is a helpful new blood test being done in Milan and Firenze, Italy and named P.H.I. (Prostate Health Index) which in my case and others did indicate high-risk of cancer, proven by biopsy. I had surgery on August 31 and am feeling good. PSA will help again this month for deciding whether radiotherapy applications may be advisable or not.

NORMAN SOLOMON, MD | Physician | Disclosure: None
LOS ANGELES CA
November 04, 2011

Advice On Counseling

The dilemma is clear and has been for years. All the anxious waiting for these studies to help clarify what to do have not really changed much. You can explain this to patients till you're blue in face or give them handouts, and they still ask you the same questions. The bottom line is what psychological burdens an individual patient can handle. If learning that ten years from now, they have prostate cancer and if they had only had a psa, maybe things would have been better is something they couldn't handle, then PSA screening is right for them. If they have a poor tolerance for making difficult decisions and would have difficulty knowing that they have a "cancer," but that they are just watching it, then PSA testing would probably be the wrong choice for them. The decision has more to do with what they can handle than it has to do all the facts we bombard them with. I've yet to have a patient NOT ask me at the end of my discussion, what would you do, which shows you their inability in many cases to process the information in a manner that is best suited for themselves.

Jim Thayer | Other | Disclosure: None
November 03, 2011

Missing a most relevant issue

As a researcher supporting the medical and pharmaceutical industry, and a prostate cancer (PC) survivor, I am appalled that so much effort is applied to this issue without considering the most obvious benefit of REPEATED PSA testing. While I did not know that my grandfather had died from metastasized PC until after I had been diagnosed, I had my PSA tested periodically (2-3 year intervals). It was fairly consistent until last year when it increased by several fold (over a 3-year period). I had the biopsy and discovered the PC in time to take action.
While a single PSA test may not be very instructive for all the reasons listed by others, periodic PSA tests are very helpful as they establish a baseline for subsequent comparisons. Because of the current controversy, my physician did not intend to perform the PSA test during my last physical, but I insisted. In my case, it probably saved my life. To me, all this controversy over the value of the PSA test seems ridiculous without consideration of repeated tests to establish a baseline, perhaps best initiated during the age of 45-50.

HANS-HERMANN DUBBEN | Other | Disclosure: None
Hamburg Germany
November 03, 2011

Prostate cancer screening: A controversy about random numbers

Studies on prostate cancer screening are principally not qualified to provide scientifically valid information about a benefit in terms of (cancer specific) mortality. Such trials have too low statistical power because of low cancer-specific mortality combined with long history of disease, lack of compliance, contamination, and unclear causes of death. Trials with sufficient power require millions of participants and tolerate no loss from follow up. Their results become available only decades after trial initiation. Then they are necessarily antiquated due to changes of the population, treatment options and diagnostics. Indispensible prerequisites of science are testable hypotheses and replicability. Neither is a given for early detection of (prostate) cancer. Underpowered studies simply generate random numbers. Thus it is not surprising that the results of the ERSPC and PLCO study are contradictory. Alone the disadvantages are evident because the number of overtreatments are much higher than the number of allegedly prevented PCa-deaths.
Cancer screening studies appear unethical and contradict §17 of the Declaration of Helsinki.

John Costello | Physician | Disclosure: None
November 02, 2011

Internist

The US Task Force stance is correct, but - it changes nothing . The trial in the US may be too compromised to prove the value , and the European trial suggest- with varying criteria for entry, and followup- that screening helps about 1 in 1, 000 or so /year- about the same as mammography, for those likely to live 10 years , between 50 and 65 -70. It's still up to the individual to decide, with the help of a physician/family / friend what is best for them, based on imperfect data. The larger ? is -should a health system pay for such screening?

CHUAN PHING TANG, MD | Physician - Dermatology | Disclosure: None
SITIAWAN Malaysia
November 02, 2011

Find a better treatment

It is agreeable that DRE with PSA screening if done properly will help to pick up early prostate cancer with better accuracy coupled with TRUS biopsy. It is the treatment options to the patient and its complications that put off the patients. PSA screening itself is not a big issue. It is time for the urologist to find a better and less invasive way of diagnosing and treating the Ca prostate. Man is worried of ED,incontinence,retrograde ejaculation... Man wants better quality of life.

JEFFREY HOROWITZ, MD | Physician | Disclosure: None
FALL RIVER MA
October 30, 2011

What is the code for "PSA discussion with patient"?

I can't help thinking about where are we going to find the extra 10-15 minutes to properly discuss the option of PSA screening with our patients. It seems impossible to tack that on to a routine office visit, do a good job and not feel rushed and fall behind schedule all day. Realistically, forget it unless there is a code for it!

GEORGE ADAMS, MD | Physician | Disclosure: None
October 29, 2011

Whenever You Like

It's not yet illegal to check your PSA. I check mine q6 months as that's what Im confortable with. So do what you're comfortable with. Check it whenever you like. You will be as right as anyone else as the final answers are not in and may never be. When it becomes illegal to check it frequently in the US go to Costa Rica on vacation and check it there.

EDUARDO SIGUEL, MD | Physician | Disclosure: None
GAITHERSBURG MD
October 29, 2011

Need Better data, not longer philosophical decisions

Change research and data to offer better data on choices. Build mathematical models probability distributions of types of cancer and treatment outcomes given values of variables such as age, weight, race, nutrient status (such as Vit D and essential fats), family history, prostate size, rate of increase of PSA, treatment type, etc. And chart probability distributions and changes in PSA as a function of biological variability, exercise, lab variability, diet, weight, etc. Develop accurate models of the causes of prostate cancer, the factors that increase or decrease cancer rates. This is feasible today. Instead, we spend millions on useless clinical trials. Current financial incentives discourage better mathematical/biophysical/biological models.
When the data become available, the optimal decisions will usually be obvious. Now, with ignorance, with poorly designed clinical trials, with focus on profits, we don't have data to make optimal decisions and we must obfuscate recommendations.

CLYDE SCHECHTER, MD | Physician | Disclosure: None
CYPRESS CA
October 28, 2011

The PSA Dilemma -- A Result of Entrepreneurial Medicine

So the major trials were seriously flawed. Those in the US suffered from severe contamination in the control group, and screening prior to randomization. Why is that? Because PSA screening was so deeply entrenched in medical practice before it was ever subjected to serious testing. This is typical of entrepreneurial systems: products and services are launched, marketed, and hyped before anyone has time to figure out if they are any good. When it's ordinary consumer goods, probably that doesn't matter much. But here we see that the US undertook a mass campaign of genital mutilation for decades, and the practice became so widespread that it was impossible to mount a valid study of whether we had anything good to show for it. We still don't know if even a single life was saved, and we never will.

The late Thomas Chalmers used to say "Randomize the first patient." His dictum, still ignored, is the wisest advice I have ever heard from any health professional.

James Mohler | Physician | Disclosure: None
October 28, 2011

Clarification on development of the PSA test

While Dr. Ablin did discover a "prostate-specific antigen" that is confined to the normal prostate, he neither developed the PSA test nor discovered the PSA on which the current test is based. That credit goes to Roswell Park Cancer Institute researcher T. Ming Chu, PhD, DSc, and his colleagues, who identified and purified PSA and later developed the simple PSA blood test that is used today for the early detection and management of prostate cancer. The team published its first major paper in 1979 in Investigative Urology, and in 1984, a patent was issued to New York and RPCI and the technology was transferred to the biomedical industry for preparing test kits. The PSA test received FDA approval in 1986 as a monitor for treatment response and disease recurrence and in 1994 as a screening tool for diagnosis. Dr. Chu received the Presidential Award from the AUA in 1993 and was featured in the April 1998 issue of Cancer Research for his research on the use of tumor cell products in the diagnosis and treatment of cancer and for his leadership role in the discovery of PSA and development of the PSA test.

Page

POLL
The U.S. Preventive Services Task Force recently issued draft recommendations on prostate-cancer screening. In light of the USPSTF report, how would you deal with a 50-year-old, non-Hispanic white man who comes to you for a new-patient appointment? He has no family history of prostate cancer and says that he does not have any lower urinary tract symptoms.
You order a PSA test (you believe a USPSTF Grade A or B recommendation to be appropriate).
15%
You initiate a discussion of prostate-cancer screening with the patient (you believe a USPSTF Grade C recommendation to be appropriate).
46%
Given the modest benefit and potential harms of screening, you do not initiate a discussion (you believe a USPSTF Grade D recommendation to be appropriate).
38%
Poll closed December 1, 2011 (2114 total Responses)