Fallout from elimination of PSA data from the SEER database


On April 29 this year, the National Cancer Institute (NCI) announced a decision to eliminate all PSA data from their current data files because of apparent inaccuracies in the ways that such data had been accumulated over time and administered. The full ramifications of this are — as yet — unclear.

If you want to read exactly why the NCI made this decision, click here. (There was no media release about this, so most of us didn’t hear about this until the last couple of days.)

If you want to be able to read what a well-respected specialist in urology thinks about it, click here.

If you want to read what Medscape had to say on the topic (complete with quotes from Dr. Penson’s article and the opinions of at least one other well-respected urologic oncologist), then click here.

On the whole, The “New” Prostate Cancer InfoLink is not as worried about this decision as others seem to be, and our reasons are threefold:

  • Few really major studies about the effectiveness and safety of prostate cancer diagnosis or treatment have been dependent on PSA data in the SEER and SEER-Medicare databases. There have been a few important ones, but not many. Frankly, it is unlikely that any of these studies have been critical to decisions about how to treat men with prostate cancer.
  • The SEER data on PSA had little or nothing to do with any decisions about whether the PSA test is or isn’t a good tool for screening men for prostate cancer. While they were certainly looked at by the U.S. Preventive Services Task Force in making their evaluation of the merits and non-merits of PSA screening, those data were not the ones that drove the USPSTF decision (one way or another).
  • The biggest problem with the withdrawal of these data is that they call into question the merits of research done by lots of relatively young urologists and others during their residencies and fellowships as they started publishing in the scientific and medical literature. Publishing studies based on large data sets like the SEER data is a “rite of passage” for many scientific and medical researchers early in their careers today. Being the first author on a paper whose results may now be called into question may not be good for some careers — but probably very few.

The “New” Prostate Cancer InfoLink congratulates the National Cancer Institute for having to guts to purge poor data from the SEER database. At the same time, this only emphasizes the potential value of the evolving AQUA Registry database being put together under the auspices of the American Urological Association, which will use data drawn directly from the electronic records of numerous urology practices around the country over time.

There is little doubt that some studies published in the past will now be discredited because they were overly dependent on PSA data from the SEER and the SEE-Medicare databases. We shall keep our eyes out for any mention of such studies. At present, however, we believe that the NCI’s decision has been the correct one, and that there will be little immediate or serious “fallout” in terms of the questioning of the types of studies which we seek to focus on in bringing prostate cancer news to our readers on a regular basis.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.