How one good story may trump a plethora of good data

Some readers may be interested in reviewing an article that appears in Time magazine this week entitled “Why people stick with cancer screening, even when it causes harm.” The article reviews how our experiences and perceptions may color the way we make our individual decisions about things like screening for breast and prostate cancers.

7 Responses

  1. I followed the link to the editorial by the discoverer of PSA (Richard Albin). I’m a bit confused. He says the test is little better than a coin toss in detecting cancer. Also says the test was approved based on a study showing that it could detect 3.8% of prostate cancers. I’m struggling a bit with that statement vs. common sense. We have pretty widespread screening in place. Is it possible that 96.2% of prostate cancers are slipping through, undetected?

  2. Dear Doug:

    First, we should be very clear that Dr. Albin did not, in fact, “discover” the PSA test at all. That claim is unjustified. Dr. Albin and some colleagues did, indeed, first isolate the protein now known as prostate-specific antigen, but he had nothing whatsoever to do with developing it’s use as a method to assess risk for or progression of prostate cancer, nor did he ever suggest either of those potential uses.

    Second, several of the other statements made by Dr. Albin in the editorial you refer to have been aggressively disputed by the members of the research team at the Roswell Park Cancer Center who did, actually, first develop the PSA test as a means to assess the progression of prostate cancer. (You need to bear in mind that the originally approved use of the PSA tests was as a tool to assess whether men who had been treated for prostate cancer had progressive disease.)

    Dr. Albin is, naturally, free to express his opinions on anything he likes, but as a reader you need to be aware that just because something appears in newspapers and magazines (let alone on the Internet) does not immediately make it truthful or factually accurate.

  3. Doug:

    I should also add — because this is very important — that no one has ever suggested that the PSA test can be used to “detect prostate cancer.” The PSA test can’t do that. It can, at best, detect risk for prostate cancer.

    Dr. Albin’s criticism that the PSA test can’t be used to accurately detect prostate cancer is therefore correct … but no knowedgeable person has ever considered that the PSA test could detect prostate cancer anyway. The key issue has always been whether the risk for prostate cancer suggested by the PSA test was sufficiently high to justify the downstream effects of biopsy and treatment — which is not the same thing at all.

  4. Sitemaster:

    I must disagree with you and Dr. Albin. Even before the abnormal PSA result I had, I was already at risk because of my brother’s prostate cancer. I had acceptable PSA results (< 4.0 at that time) for a number of years after his diagnosis. Then in the span of 4 months, my PSA went from 4.0 to 5.1. An ultrasound showed a suspicious shadow in my gland. This led to the biopsy which revealed a Gleason 9 cancer where the shadow was.

    I know the test is not specific, but that does not mean it can never detect prostate cancer. How else do you explain my case?

    My wife thinks I'm special, but I cannot imagine that I am unique out of the hundreds of thousands of men diagnosed.

    This test result certainly justified the treatment I received: external radiation plus hormonal therapy. That was over 8 years ago and there is no sign of recurrence to date. Without early detection by the PSA test, it's likely I wouldn't be here. I believe it saved my life.


  5. Let me ask the question a bit differently. Dr. Albin seems to be saying the PSA test is worthless for any other purpose than to detect the progression of prostate cancer in someone who has already been diagnosed. Is it likely (or even ever happened) that someone who has been regularly screened using PSA, consistently had low scores (say < 2), and suddenly been discovered to have advanced prostate cancer even though the PSA score remained < 2? (My current understanding of the PSA test would be that something like that rarely, if ever, happens). On the other end, how likely is it that someone who has had a "normal" score, say 2, and finds their score climbing up to 6 or 8 (not a single outlier reading, but a couple of tests) and then come up with a negative 12 needle biopsy? (My current understanding is that would be extremely rare.) I do realize various sorts of trauma to the prostate can raise the PSA score.

    By the way, I saw what I believe to be a bit odd in the risk chart depicted in the Time article. There was a large category for "men treated unnecessarily for prostate cancer". I didn't see a category for "lives saved or prolonged by timely treatment". The number of deaths from prostate cancer was the same in both charts (8). If I understood it correctly, the chart was claiming that screening 1,000 men would not save a single life. Even the USPSTF admits that some lives would be saved. So, it seems to me that the scale selected for the chart has to illustrate that, or the chart will be misleading. The number of men screened in order to save at least one life should be the minimum scale to make the chart useful. Or, to put it another way, shouldn't the scale be large enough so that the number of deaths from PC in the screened sample be less than the number in the unscreeened sample?

  6. But Manny … (a) I don’t agree with Dr. Albin and (b) with a brother already diagnosed and a PSA close to 4.0 ng/ml you were already at significant risk.

    I don’t think you understand what I am saying. I would clearly acknowledge that you had two risk factors and needed regular PSA tests. I, on the other hand (now at 64 years of age), have never ever had a PSA value > 1.2 ng/ml except in association with a clear case of a urinary tract infection, and I have no history of prostate cancer in my family. Frankly, my risk for a diagnosis of clinically significant prostate cancer in my lifetime is probably as close to zero as it gets (although I may well be found to have indolent prostate cancer cells in my prostate if anyone bothers to autopsy me when I die; most men do by the time they get well into their 80s).

    As I keep (gently) pointing out, if 1 in 6 men get diagnosed with prostate cancer in their lifetimes (whether it is clinically significant or not), this also means that 5 in 6 men do not! Surely a goal that is just as important as identifying clinically significant prostate cancer in those who do need treatment is to avoid inappropriate testing those men who are at no risk for clinically significant prostate cancer (and who therefore are at potentially very high risk for inappropriate treatment and complications of treatment if clinically insignificant cancer is identified).

  7. Dear Doug:

    Can we please distinguish carefully between (a) opinion; (b) journalism; and (c) facts?

    Dr. Albin is indeed saying that, in his opinion, the PSA test is worthless. That is his opinion. I disagree with him. However, I also do not think that screening every man over 40 or 50 on an annual basis is either scientifically justifiable or medically appropriate.

    Fact: It is actually not that unusual for men to suddenly be diagnosed with advanced or even metastatic prostate cancer and a PSA value of 4 or less (even less than 2). It is certainly not “common” but arguably it is also not “rare.” I would guess I regularly come across at least a couple of cases like this a year, most commonly in younger men. Kinda depends how you want to define “rare.”

    Fact: It is also not that unusual to see men have a relatively rapid rise in their PSA level (say from 2.5 to 6 or 8 ng/ml over a few months) and then to have a biopsy that is negative for prostate cancer. Most of these men likely have either some form of urinary tract infection or an inflammation of the prostate which may become chronic.

    Fact: The diagram presented by Time magazine is (arguably) about as accurate as the data reported by the PLCO trial (i.e., highly inaccurate) because it is based on the data from the PLCO trial, which was a highly flawed study.

    As I keep trying to point out to people, the PSA test can only give an indication of possible risk for prostate cancer. It is not and never has been specific for prostate cancer.

    We really ought to have understood by now that the PSA test does have a value in regular testing of those men known to have some degree of risk for prostate cancer (like Manny), but it has little value as a mass screening tool. All that anyone can do with the PSA test from a diagnostic point of view is use it to decide whether a biopsy might be appropriate in selected patients. It is not a test that can accurately determine risk for prostate cancer in all men — let alone distinguish between those men who will need treatment and those who do not.

    Finally, just because I report something doen’t mean I agree with it. We offer a resource for patients and other readers in a very complex field of medicine where even the “experts” commonly have divergent opinions. It is critical for patients to appreciate just how divergent these opinions can be. There are few real and definitive “rights” and “wrongs” when it comes to the diagnosis and management of prostate cancer … but there is a vast range of opinions.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: