USPSTF Draft Report

[Updated April 17, 2017]

Since the material below was written the USPSTF has revised their recommendations.  See this blog post for a summary:

https://prostatecancerinfolink.net/2017/04/11/new-draft-uspstf-guidance-on-screening-for-risk-of-prostate-cancer-the-details/

and USPSTF's own web site here:

https://screeningforprostatecancer.org/

The US Preventive Service Task Force (USPSTF) has produced a draft recommendation on PSA screening available [here] based on the evidence presented [here] and [PMID: 22171385] and the head of the Task Force added this comment in the Philadelphia Inquirer. The USPSTF's gave PSA testing a grade of D which means that it recommends against PSA screening and discourages the use of it in practice. Note that this only refers to using the PSA test on healthy men with no suspicion of disease and does not refer to the use of PSA testing in other scenarios.

New Jersey has since passed legislation opposing the USPSTF recommendation (see [news]) and Kathleen Sebelius, Secretary of Health and Human Services, has issued [this letter] (from PC Infolink site) indicating that the US government does "not intend to eliminate coverage of this screening test under Medicare at this time" although that leaves open what future actions they might take.

Partly due to the controversy that this decision has generated, a new bipartisan bill HR1151 designed to reform the USPSTF process is under consideration. The bill is supported by the American Urological Association (AUA). A summary of some of the key points is available in this [June 3, 2013 Urology Times article] and this [March 11, 2015 Medscape article]. A 36 minute presentation by Peter C. Albertsen largely in support of the USPSTF recommendations is found at [January 14, 2015 UBC Urology Rounds]

Pro USPSTF (i.e. arguments against screening)

The report and other evidence supporting it includes:

• Minimal Benefit/High NNT. Although there were benefits shown from PSA screening the absolute level of these benefits was low (absolute mortality advantage of less than 1 percentage point) and the number needed to treat (NNT) to save one life was high (48 men) according to [PMID: 19297566] [Full Text]


• Significant Overdiagnosis and Overtreatment. Jemal et al [PMID: 19474385] [Full Text] present statistics that one in six men is diagnosed with prostate cancer whereas only one in 34 die of prostate cancer suggesting a large amount of overdiagnosis and treatment. Another 2009 study indicated that 1.3 million additional men had been diagnosed with prostate cancer since 1986 and 1 million of those had been treated. Even if one assumes that the significant reduction in prostate cancer deaths in the nineties was all due to screening (as opposed to other improved methods) the authors show that more than 20 men must be diagnosd with prostate cancer to save one life and this suggests a large degree of overdiagnosis and treatment. See [PMID: 19720969]. The magnitude of the problem is enormous when one considers that in 2007 there were 2.2 million Americans alive who had been diagnosed with prostate cancer according to SEER statistics referenced in [PMID: 21925305].


• Significant Harms. "Radiotherapy and surgery result in adverse effects, including urinary incontinence and erectile dysfunction in at least 200 to 300 of 1,000 men treated with these therapies. Radiotherapy is also associated with bowel dysfunction." [USPSTF draft report] and [PMID: 21984740] [Full Text] Since screening leads to significant overtreatment (as per last point) screening leads to harming men who never needed the treatment in the first place.


• Independent. Regarding criticism that there were no urologists, radiologists or medical oncologists on the USPSTF task force (see AUA Foundation Dec 1, 2011 press release and task force composition here) it has been suggested that it was necessary to have "independent scientists who are better able to objectively evaluate the literature without bias" [link] and to combat the harm that urological practice may be doing to the population by not adequately disclosing that the PSA test can lead to unnecessary treatment. Also see [PC Infolink post] for additional articulation of this viewpoint. The American Cancer Society is also moving to a system of generating guidelines that exclude specialists in the area from voting to eliminate conflict of interest (although input from them will be sought). See [WSJ Blog Dec 14, 2011] [PC Infolink blog Dec 16, 2011], JAMA article: [PMID: 22166609], Institute of Medicine [Clinical Practice Guidelines We Can Trust] and Institute of Medicine [Finding What Works in Health Care Standards].


• Excessive Cost. USPSTF did not consider cost but if they did that then it is estimated that it would cost $5 million per life saved and that amount of money could save more than a single life if redirected in other ways. Allan S. Brett and Richard J. Ablin (the developer of the PSA test) [PMID: 22029759] [Full Text]


• Other government and professional organizations. Other government and professional organizations have also concluded that the value of PSA screening is inconclusive such as in the [UK] or outright recommended against it such as the RACGP position in Australia (link no longer works but see 1st column of page 599 also on the RACGP site) and [Ontario, Canada] where PSA screening is not reimbursed.


• Inflammatory Language. It has been argued that the screening debate is unreasonably emotionally charged due to the labeling of both lethal and non-lethal forms of prostate cancer with the word "cancer" and that it would be better to have language which clearly distinguishes the two. See [NY Times].
• Broader Trend. It is believed that the USPSTF decision is part of a broader trend toward fewer screenings for any disease in healthy people. See [USA Today] .

Con USPSTF (i.e. arguments in favor of screening)

In contrast to government organizations, doctor and patient groups widely support PSA screening and have been vocal in criticizing the report. The American Urological Association (AUA) which is the primary body representing urologists wrote: "The AUA stronly opposes this position" [press release] (referring to USPSTF's grade D) and recommending its own AUA Practice Statement instead. John Hopkins University James Buchanan Brady Urological Institute stated that (referring to the USPSTF draft report) "this decision sets the clock back to before the 1990s, when “healthy” men were diagnosed with cancer that was palpable and often, too late to cure. Is this about progress, or saving money?" See [link]. A coalition of 10 prostate cancer patient groups, the Prostate Cancer Roundtable, was also critical [formal letter] [media release] as was Michael Miliken's Prostate Cancer Foundation (PCF) [PCF's USPSTF page].

Criticisms Of the USPSTF report included:

• Deaths Declining. Since the 1991-1994 time period when deaths from Prostate Cancer in the US peaked at an annual rate of 39 deaths per 100,000 men the rate has declined to approximately 24 deaths per 100,000 men [statistics cited by National Cancer Institute]. This time period correlates well with the introduction of PSA screening.


• Studies too short. Prostate cancer is a slow growing disease. The benefit is increasing over time and as the studies progress it appears that the trends show an increasing benefit. The eventual benefit to screening is expected to be much higher than that shown so far. In particular the "use of only a 10-year time frame is a significant flaw with their analysis. Urologists and oncologists know that prostate cancer is a slow-growing cancer and one that thankfully responds to hormonal manipulations. Looking at only 10 years’ worth of data completely ignores the fact that the real benefit to screening would be seen in the second 10-year span, when the cancer becomes hormone-resistant and more deadly. Goldstein comment


• Important Information Omitted. the USPSTF did not consider all the information. See Catalona Comments and Martin Goldstein Comment for a list of pertinent facts that were not considered. The latter noted, in particular, the exclusion of [PMID: 20598634] [Full Text] which was published in July 2010 and showed a 44% decrease in the death rate over 14 years.


• Morbidity Ignored. (1) "The panel makes virtually no mention of prostate cancer morbidity (sickness from the disease, including urinary obstruction and extraordinarily debilitating bone pain from cancer spreading to pelvis, ribs, back, and legs)." (See [link]) Metastatic disease, and not just death, should have been considered but were not. Fritz H. Schröder [PMID: 22029756] [Full Text]. (2) "Murray Feldstein, MD, from Phoenix, Arizona commented to The Annals of Internal Medicine following the USPSTF recommendations, “As an elderly urologist who spent nearly half of his career in the pre-PSA era, I can personally attest to another and perhaps even more important factor that is being overlooked—suffering from advanced prostate cancer. No longer do I see patients with bulky cancer who bleed and obstruct their urinary tracts.” He pointed out that painful prostate cancer that had spread to bones was now rare, a situation undoubtedly attributable to the widespread use of PSA screenings." (Quoted in [link] or [link]) (3) Similarly, Dr. Gomella, chair of the Dept. of Urology at the Kimmel Cancer Center at Thomas Jefferson University in Philadelphia, recalls that in the 1980s when he was a resident that, unlike today, patients typically presented with metastatic disease. “Our only option was to surgically castrate these men or treat them with estrogens, and they died a year or two later. I don’t want to go back to those days." Quoted in Aug 13, 2012 American Medical Association's news web site [here]. (4) A July 2012 paper [PMID: 22847578] estimated that in the year 2008 one would have expected 25,000 cases of metastatic prostate cancer based on data from the pre-screening era but only 8,000 were observed. That is metastatic prostate was 3x more prevalent prior to PSA screening than after. If this were due to screening then the effect of screening would be a reduction in metastatic prostate cancer of 17,000 cases in 2008. The study author was quoted as saying: "The reason our study has some meaning is that all a screening test can give you is a shift to lower-stage disease. ... It can't cure the disease and it can't prevent the disease. All it can do is allow you to catch it earlier and give appropriate treatment for the stage of disease you have caught it at." Even though this is not a randomized study, the author also pointed out that the randomized European study used by the USPSTF showed consistent data reinforcing the conclusions of this study. [from WebMD]. "The USPSTF will be criticized for not having done analysis such as this," Dr. Martin Sanda, director of the Prostate Center at Beth Israel Deaconess Medical Center in Boston, said of the study findings. [Fox News]. (5) "But what of outcomes other than death? Epidemiologic data show a 75% decrease in the number of men presenting with advanced prostate cancer since the introduction of PSA screening. And in the European Randomized Study of Screening for Prostate Cancer (ERSPC), the incidence of locally advanced and metastatic cancer was 40% higher in the control group than in the PSA-screened group. How do we balance the possibility of a later life with advanced prostate cancer marked by bone pain, pathologic fractures, and urinary obstruction against the more immediate symptoms of incontinence and impotence that often follow surgical or radiation treatment of early-stage prostate cancer? Is it possible to put numbers on the “utility” or impact of these conditions on a man's life?" Quoted in: Harzband and Groupman, NEJM 2012.


• Flawed PLCO Analysis. The USPSTF concluded that the PLCO trial showed no benefit but a re-analysis of that trial by comorbidity strata concluded that that data actually shows that "Selective use of PSA screening for men in good health appears to reduce the risk of PCSM with minimal overtreatment." [PMID: 21041707]. Also the PLCO study was heavily contaminated as discussed further down in the No Adjustment for Contamination point.


• Flawed ESPRC Analysis. A re-analysis of the ESRPC trial in [PMID: 20956725] showed that that trial underestimated the benefit by "(i) including in the 20% the years before the impact of the first screen becomes manifest, and (ii) not having full information for the follow-up years where the effects of the screening are most apparent."


• Flawed Handling of Low Quality Studies. The USPSTF relied on a meta-analysis which equally weighted 5 randomized controlled studies but the studies varied in quality and they ought to have down weighted or excluded the poorer quality studies. "This would be akin to using two fresh and three spoiled bananas to create a smoothie, then concluding that bananas should no longer be used in smoothies because the resultant mixture tastes bad."
  Catalona Press Release
and Fritz H. Schröder [PMID: 22029756] [Full Text].


• Final Assessment Wrong. Even without debating the merits of the USPSTF's report in other ways it has been suggested that they should have at least rated PSA testing as C rather than D since they did find evidence of benefit from PSA testing and they have no right to make a decision which implies that a certain number of deaths is acceptable. A grade of C would have put more of the onus on patients to make the decision whereas D is too strong as it seemingly imposes the USPSTF's value judgement on patients. Mary F. McNaughton-Collins and Michael J. Barry [PMID: 22029758] [Full Text]


• No Adjustment for Contamination. Intention-to-treat statistics should have been supplemented with statistics adjusting for contamination as such statistics show a large benefit to PSA screening. (Contamination refers mainly to the situation where subjects in the control group are actually screened. Adjusting for contamination means comparing those who were actually screened to those who were not screened as opposed to comparing those who originally intended to be screened vs. those who originally intended not to be screened. Using Intention to Treat is well established but both set of statistics should have been shown.) Fritz H. Schröder [PMID: 22029756] [Full Text].
Later congressional testimony pointed out that 90% of the subjects in the control arm of the PLCO study were screened so it was comparing patients who were screened against patients who were also screened. See [congressional testimony - youtube]

• No Real Experts on Panel. There were no urologists, radiologists or medical oncologists on the USPSTF task force. See composition of task force here: here. Because they had no experts in prostate cancer they have been accused of just manipulating the numbers without understanding their meaning or implication. In particular the biological process prostate cancer goes through is a symptomless phase (sometimes referred to as the curability window). If it is not caught at that time then it will not be curable. Given the significant death rate due to prostate cancer its important to catch it when its most addressable. The fact that the USPSTF conclusions seem to violate biological plausibility would appear to contravene the well known Bradford Hill Criteria which are commonly used to assess studies.


• Conclusions Not Based on Practice. The USPSTF misjudged the effectiveness of PSA testing by not considering how it is used in practice. In particular, higher Gleason scores (discovered on biopsy after a PSA test) and fast rising PSA after excluding infection are both reliable indicators that treatment is needed.


• Many Men Live to their Nineties. It has been suggested that even if one rejects the report that it should still apply to older men but men age 75 have a 50% chance of living past 90 so its not appropriate to exclude those men from screening. See Dr. Charles Myers video


• Screening Has Saved Lives. Two independently produced statistical models suggest that its plausible that 45% and 70% of the decline in prostate cancer mortality during the 1990s was due to screening. See [PMID: 18027095] [Full text] so its effect has saved many lives.
• Would Deprive Patients of Useful Info. Another criticism is that without the test, patients would be prevented from using all the knowledge that could be available to make decisions about their health. As pointed out by New Jersey physician, Adam Perzin, "The federal task force misdirects the very real medical concerns about the “risks of treatment” onto screening. But screening is not treatment. It is not even diagnosis. There are essentially no risks to screening. Denying patients the opportunity to participate in decisions regarding their own health care because of concerns regarding treatment they may never get is a stick-your-head-in-the-sand approach to medicine that in my view will cost lives." See [link]
.

• Significant Death Rate. Prostate cancer is one of the largest killers of men among cancers so its prudent not to reject the only test that can catch it early.


• Government Health Care Rationing. It has been pointed out that "if accepted by the government, this recommendation would essentially become healthcare rationing, in which only people who can pay out-of-pocket, would receive this life-saving test." Catalona Press Release According to a New York Times article the USPSTF's "work is mandated by the government and is administratively supported by the government". Further suspicion that the report is being commissioned to justify government health care rationing has been aroused by (1) the fact that the draft report was supposed to be issued for public comment but according to the same New York Times article the task force is not making the comments public indicating that "public comment" was intended to only mean "public input" and (2) according to the [boxed information on the USPSTF site] the USPSTF "lost" some of the public comments through a technical glitch. In 2009 the USPSTF recommended "that mammographic breast cancer screening before age 50 should be optional, and decided by a woman and her physician." At that time the "mammography controversy attracted congressional attention from legislators responding to constituents’ outrage and demanding greater transparency in the USPSTF decision making process. [Wall Street Journal Blog May 21, 2010]. HHS secretary Kathleen Sebelius eventually issued a statement emphasizing that the USPSTF does not set health policy." [USToo Hotsheet November, 2011]. . Despite such statements, a fear has been articulated by The American Enterprise Institute, a conservative think tank, in reference to the USPSTF, that the under the "Patient Protection and Affordable Care Act (PPACA), a previously obscure government advisory body has acquired vast authority to decide which health care services Americans will have access to." [The Bleeding Edge of Rationing] and 43 members of the House of Representatives [list of signatories via PC Infolink] consisting of both Democrats and Republicans have sent a [letter via PC Infolink] to Secretary of Health and Human Services Kathleen Sebelius asking for continued funding of PSA tests. The letter states that the USPSTF conclusions are based on studies which are "disputed by experts" and they further write that failing to cover the test would have "potentially dire" consequences. In another development, it has been charged that one particularly insidious approach to rationing health care is an attempt to simply rename cancer to something else so that funding bodies can get out of paying for treatment of it: [NY Times].
• Practice Not Affected. Based on a survey of primary care physicians [PMID: 22517310] [Science News] it seems that the USPSTF guidelines are not having much effect on practice.

Readers may also wish to review the PSA Test Consent forms by [Dr. Chodak] and [Ralph Valle (patient)].