The Asco Post

Do We Need the USPSTF?

By Derek Raghavan, MD, PhD
November 1, 2011, Volume 2, Issue 16

Like most of the folks reading this commentary, I’m a taxpayer. Although I sometimes become impatient with the strategic games on Capitol Hill, I basically appreciate that government helps many things to work, and some of them even work well.

However, there are aspects of government function that do trouble me. I was more than a little perturbed when I had to deal with the extraordinary time-wasting nonsense that followed the encyclical from the U.S. Preventive Services Task Force (USPSTF) a couple of years ago, in which they unilaterally announced a change in recommendations on screening for breast cancer.

It wasn’t that I was personally opposed to the specific recommendations, as I don’t specialize in the care of early breast cancer and didn’t know the data. But I was troubled that their constituency did not seem to include an expert specifically in breast cancer care, and more troubled that they felt it appropriate to issue their bulletin and press releases apparently without consulting the various specialist bodies and societies, and without alerting the people who take care of cancer patients (ie, us) that they were about to create chaos in the patient and nonpatient populations. As it turned out, some of their interpretations of data probably had merit, but their recommendations were widely condemned, several government leaders disowned them, and the whole episode ended up in confusion and uncertainty—a questionable use of my tax dollars.

USPSTF Strikes Again….Twice

Now the USPSTF has done it again…twice! First, its members decided to assess the utility of screening for testicular cancer. In a new guideline published by Annals of Internal Medicine,1 the Task Force announced that they still do not recommend screening for testicular cancer, as there are no new data, and gave this recommendation a D grading (defined as: “The USPSTF recommends against routinely providing [the service] to asymptomatic patients. The USPSTF found at least fair evidence that [the service] is ineffective or that harms outweigh benefits”). They went on to note that the disease has a low incidence and that it is associated with favorable treatment outcomes, thus vitiating the need for any screening approach or further attention to the issue.

Had there been real experts involved, after noting the lack of level 1 data on screening, they might have commented on the many papers that have given useful information on the topic—for example George Bosl’s time-honored observation that delay in diagnosis correlates directly with stage and inversely with outcome.2 They might have spoken about the importance of primary tumor stage and size with regard to outcome of local treatment,3 and the fact that uneducated populations present usually later than those with good health education. They might have quoted the NCI statement4 noting that patients with a history of testicular cancer are at increased risk of developing a second testicular cancer, and that this requires lifelong follow-up.

Responsibility for Context

I’m not attempting to endorse screening for testicular cancer, because I think that the Task Force is probably correct that there are no level 1 or level 2 data to support its routine use. However, I do think that if you grab the bully pulpit (and expend taxpayer dollars), you have a responsibility to tell the complete story, applying the pulpit in a useful fashion that helps educate patients and their medical attendants.

Young men should be made aware that testicular cancer exists and is curable, and that achieving cure of testicular cancer by chemotherapy for an advanced presentation (while an excellent advance in the past 25 years) is harder and associated with more morbidity than achieving cure by an uncomplicated inguinal orchiectomy for stage I disease. Whether screening and testicular self-examination are effective or not, it is clear that educated men who know about testicular cancer will have a better chance of cure, and with less morbid treatment. When a learned body makes a statement on the topic, if there are no level 1 data, one could at least expect a modicum of common sense in providing the big picture.


Not content with the above non-event, they now have reconsidered the data that have been published on randomized screening trials for prostate cancer. This has been—and will be—covered ad nauseam in the medical and lay press, and I won’t delve into the details. It was appropriate for an independent body to review this issue for the government, although disappointing that none of the Task Force committee members apparently has specific expertise in this domain. They concluded that there is no role for the use of prostate-specific antigen (PSA) tests in routine screening for prostate cancer in the community.5 

Based on the published trials, they are probably correct, as it is hard to endorse a screening test that does not produce an overall survival benefit. The reduction of deaths from prostate cancer per se in some of the trials is potentially very important, but is negated by the fact that the overall death rates are apparently the same in screened and unscreened patients. Some members of the urology community have dealt with this observation by questioning the constitution and execution of the trials—but it actually doesn’t work that way! You can’t just dismiss major randomized trials because you don’t like the results. If screening is to be used routinely in a community struggling to pay its medical bills, there needs to be published level 1 data that show an overall survival benefit at some time point. In fact, that may eventually be demonstrated, just by following the published trials for a longer period, as the natural history of prostate cancer from early detection is long.

So why am I concerned? Once again, it is the pathway that is the problem. A group of non-experts have done a re-analysis of well published data, and have chosen to release draft guidelines, apparently with no consultation with any serious expert body—American College of Physicians, American Urological Association, perhaps even the ASCO Prostate Cancer Panel. The Task Force has repeatedly argued that such expert groups have conflicts of interest, and they don’t wish their pristine thoughts to be compromised by venal self-interest. Fair enough—so why could they not send their draft review and recommendations for a response from professional societies, consider those responses, identify any possible flaws in their own arguments, and take advice on framing the information for public distribution? Instead they have, again, acted unilaterally, causing chaos among the medical profession, and more importantly, in the patient community.

Of particular importance, potentially the most at-risk groups—African-Americans and men with family histories—are now even more confused about their appropriate management, and whether screening should have an ongoing role. These populations were not analyzed by our friends in the U.S. Preventive Services Task Force, but the blanket statements in the draft report seem to cover these patients nonetheless.

Dubious Benefit

The U.S. Preventive Services Task Force seems, again, to have shown a capacity for ineptitude that has led to another waste of the money that we pay the government each year, without obvious benefit to the community at large. We are moving into a time of budgetary restraint by government. I wonder if congressional budgetary planners might consider an investigation of the function and composition of the USPSTF, and whether this panel really needs to exist. Moreover, the editors of Annals of Internal Medicine might consider whether government committees should have a free ride to publication. ■

Disclosure: Dr. Raghavan reported no potential conflicts of interest.


1. U.S. Preventive Services Task Force: Screening for testicular cancer: U.S. Preventive Services Task Force reaffirmation recommendation statement. Ann Intern Med 154:483-485, 2011.

2. Bosl GJ, Vogelzang NJ, Goldman A, et al: Impact of delay in diagnosis on clinical stage of testicular cancer. Lancet 2(8253):970-973, 1981.

3. Raghavan D, Vogelzang NJ, Bosl GJ, et al: Tumor classification and size in germ-cell testicular cancer: Influence on the occurrence of metastases. Cancer 50:1591-1595, 1982.

4. National Cancer Institute: Testicular cancer screening (PDQ®). Available at

5. 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. October 7, 2011 (early release online).

More on Twitter