A pill that achieved a 31% reduction in mortality due to a particular condition, but only at a 48 to 1 over-prescription rate and leaving between 30% to 80% (depending on whose numbers you believe) of those who took the pill impotent, incontinent, or with bowel disorders, would never make it past stage 1 trials. You simply cannot base a decision on a reduction of mortality percentage without considering the associated morbidity.

It has been clear from the scientific literature for some time that PSA level is only a bit better than chance in indicating prostate cancer. Even without age corrrection, an elevated PSA increases the likelihood of detection of prostate cancer under biopsy from about 15-20% to only about 25%-30% (again depending on whose numbers you believe). Since both the incidence of prostate cancer and PSA level are strongly correlated with age, the difference is even less when the comparison is limited to men of the same age. That was Albin’s point. The increased biopsy rate resulting from screening (even if the selection of men for biopsy was only a little better than chance) has probably contributed to the reduction in mortality, but as I’m sure you are aware there are a number of other effects unrelated to screening (better treatment, attribution bias, etc.) that also likely contributed as well. This is reflected by decreasing mortality in countries where screening has been much less prevalent. There is little point in discussing that now, though. Only well-designed, controlled studies can effectively separate out the effects of screening alone, and the data is finally becoming available.

I read your earlier post as saying there is another, larger, set of men who also benefit from screening by avoiding the QOL issues of late-stage metastatic disease that are missed when only mortality rates are considered. I was saying that a 25:1 need-to-treat ratio to prevent any detectable metastasis is consistent with those who suffer late-stage metastatic disease being the same as those who ultimately die of the disease (at a 48:1 need-to-treat ratio to treat one), so by tracking only mortality rates you identify essentially all who have benefitted from a screening/treatment program. Your citation further demonstrates that point. I apologize if I misunderstood your initial post on the subject. If you were saying that not only death, but the QOL of late-stage disease should also be considered as part of the cost of lethal disease in making a decision, I certainly do not disagree with that. But the need-to-treat ratio is still on the order of several tens to one.

If a pharmaceutical company had a pill that reduced mortality by 31% or even by 20%, what you call inconsequential would be a major issue. In the absence of a better diagnostic tool, PSA identifies prostate gland problems. It is not cancer-specific, but is prostate specific and fundamentally important to identify other prostate gland maladies along with prostate cancer. As mentioned before, the problem is not with the test, the problem is with how it has been used.

You really believe that “when corrected for age, PSA is only a slightly better indicator for the occurrence of prostate cancer (lethal or not) than chance?” You think that the 40% reduction in mortality in the USA happened by chance?

In your previous post you said: “that “metastatic disease” does not necessarily equate to late-stage, symptomatic disease” and I responded saying that once metastasis developed, the median time to death was 5 years in the Pound study and the QOL of those men at that time is not great as the disease undergoes progression. Now, you say that the study supports your point … how?

Some may cherry-pick the 20% versus 31%, but that difference is ultimately of small consequence (representing a 0.3% lifetime mortality change), and that number alone is not sufficient to make a rational decision in the first place. The cost-to-benefit ratio matters, which is reflected in a 48 to 1 need-to-treat ratio to achieve that benefit. Hypothetically it might even be possible to virtually eliminate prostate cancer mortality by treating all men at age 30. No one would seriously consider that suggestion due to the overwhelming cost in both morbidity and dollars, but even then the need-to-treat ratio would be about 33 to 1 (based on a 3% lifetime mortality rate), lower than what has been observed in the screening trial to date. Are ratios that high due to screening acceptable because the men are treated at a later age?

And a core question is why is the need-to-treat is so high in the first place. The reason is that PSA is fundamentally flawed as a screening test. It was first implemented based on the most basic of logical fallacies: men who died of prostate cancer had a highly elevated PSA implies that men with an elevated PSA have lethal prostate cancer. A implies B does not mean that B implies A. And here the fallacy was taken another step: an only moderately elevated PSA implies not just the presence of lethal disease, but the presence of lethal disease that is still at a curable stage. Dr. Albin pointed out the reality in his recent editorial: when corrected for age, PSA is only a slightly better indicator for the occurrence of prostate cancer (lethal or not) than chance. But with the preconceived notion that it must be an indicator of lethal disease, the result has been to look and look until something can be found to justify treatment. Screening provides an excuse to biopsy, not an indication of lethal but curable disease.

What has the European trial demonstrated thus far? That if you biopsy and treat enough men, eventually you are bound to cure some who actually need curing, not that screening has identified men who actually need treatment. The real control that is needed is for every man who is biopsied due to screening, another man of similar age, health, family history, etc., who has not been screened is randomly selected for biopsy as well. The outcomes in the two groups would likely not differ greatly.

The Pound et al. study makes the point I was trying to make, that the set of men who suffer the quality of life issues due to late stage metastatic disease and the set of those who die of the disease are essentially the same, so you can effectively evaluate the performance of prostate cancer screening/treatment programs by considering the mortality rates.

It is a statistical no-no to cherry pick data. When a slight of hand is used with “intent to screen” as compared to ”actually screened,” the mortality reduction is 20% instead of 31% in the ERSPC study. Add that screening was done every 4 years and the study reported after 9 years of follow-up and things might look that bad in another 5-10 years. Consider that, until now, after 20-odd years of PSA use, no screening study that reported a survival benefit has been “acceptable” to the anti-screening crowd.

As far as your statement that metastatic disease does not necessarily equate to late stage, read the study by Pound et al. that reports that the median time to death after the development of metastasis was 5 years. And those tend to be 5 years of not so good QOL.

Again, the fact that many men jump to treatment because of misinformation, diagnostic uncertainty, and fear is a reality. Let’s be fair about the information that is provided. Let’s establish a strict guideline to force practitioners to provide unbiased information for men to decide. The fault is not with the PSA test, but how it has been used. The actual uncertainty of our diagnostic tools needs to be better understood. For most patients diagnosed with what is perceived as early disease, a period of learning should be always prescribed whenever possible. Acting out of fear instead of knowledge is wrong and should be prevented. Easier said than done, but we should all be working to that end in favor of those currently diagnosed with prostate cancer.

Terry, we are talking of symptoms of progression in men with early-stage prostate cancer. How were men diagnosed prior to the PSA era? The great majority were symptomatic (urinary obstruction, bone pain, etc). The point I am making is that given enough time and in the absence of other comorbidities, prostate cancer is a progressive disease that is lethal. To ignore that there is progression overn time is wishful thinking, and that is what I object to when I read about “crystal ball” statements that overtreated men would not be affected in their lifetimes. Where are the studies of untreated prostate cancer that support that?

The very low risk of dying from prostate cancer is not low when men survive cardiovascular disease and other more lethal cancers such as lung cancer. To depend on other causes of higher mortality and ignore that prostate cancer is a progressive cancer is a poor excuse to support “low risk.”

Look, I am not trying to instill fear at all. Anyone should decide what to do based on available information.

The reality is that symptoms of progression in untreated prostate cacner are always ignored. I have read enough studies to support that statement and have always marveled on how those that claim a high rate of overtreatment “know” that those overtreated men would have never be affected by the disease in their lifetime. The authors of the study are well-known supporters of watchful waiting who followed this cohort of 223 men for 21 years, but were honest enough to state:

“In conclusion, our data indicate that the probability of progression to a more aggressive and lethal phenotype may increase after long-term follow-up of prostate cancers that are diagnosed at an early stage and initially left without treatment. These findings argue for early radical treatment of patients with long life expectancy. Not only would such surgical intervention potentially prevent deaths, it would also convey prevention from disability caused by local tumor growth.”

According to Loeb and Partin, the number of patients who needed to be treated to prevent one case of metastatic disease in the European ERSPC screening trial was 25. Those are still terrible odds, particularly considering that “metastatic disease” does not necessarily equate to late-stage, symptomatic disease. (Loeb and Partin also say that the number of patients who needed to be treated to avoid one case of metastatic disease is “only” 15 when you cherry-pick the data and compare them only to uncscreened men in Northern Ireland, but that is a statistical no-no.)

Maybe that is because there are hardly ever any? After all, time and again we are told that in most cases prostate cancer is asymptomatic.

I was surprised at the start of my journey to read one Scandanavian study where some of the men were said to have died within 3 months of diagnosis — and were therefore excluded from the study.

My thought then was that perhaps the threat of a long and painful death might be overstated and might be part of the “fear factors” that drive men into early unwarranted treatment. That is still my view.

I am not denying that for some men the disease can progress and cause very bad symptoms and a painful death. However, I do believe that they are in the substantial minority and the chances of a man diagnosed with what is now termed a “very low risk” form of the disease progressing into that arena is also very low.

You asked: “One piece of information I have yet to learn is what percent of patients who choose active surveillance suffer from high stress or actual pain? Shouldn’t these additional factors be assessed, not just death, when weighing risks or benefits before deciding what treatment to choose?”

No question that proper informed consent should prevail for patients to decide what to do with a prostate cancer diagnosis. Active surveillance is an excellent tool to slow down precipitous decisions and avoid unnecessary treatment. That said, information should be more complete so that better decisions can be made.

One hardly ever reads about the symptoms of untreated prostate cancer progression. It is as if this cancer is always assumed to be indolent in nature and to never or hardly ever progress in a patient’s lifetime. Why is this so? I believe there is important information about this (of the true natural history of untreated prostate cancer) that is being ignored to the detriment of the decision making process.

It is as if prostate cancer progression, although typically slow and potentially variable in nature, doesn’t exist. Invariably and rightfully so we hear about the side effects of treatments, but not about the symptoms of progression. So, what is the natural history of untreated prostate cancer progression?

There is a rich history of untreated prostate cancer in Scandinavia. When citing such history, instantly one is reminded that this was the era preceding PSA and those cancers were different from the present. Were they really different? It is very possible that those cancers were diagnosed at more advanced stages, but there is ample evidence that studies exist to demonstrate what is the nature of progression from earlier stages of untreated disease.

For the sake of brevity here, I am citing the study by Johansson et al. This is the original study on watchful waiting. What follows was extracted from the referenced article:

“Our data may be important for counseling and clinical management of individual patients. Postponement of death is not the only treatment objective because local progression may create substantial suffering. Indeed, many of our patients experienced symptomatic local growth without generalized disease (Table 1), requiring treatment with estrogens or orchidectomy.

“If our data reflect a real phenomenon, they would imply that the probability of progression from localized and indolent to metastatic mortal disease increases
markedly after long-term followup. This progression is not restricted to cancers diagnosed due to clinical symptoms but includes also tumors detected incidentally at transurethral resection due to presumed benign prostatic hyperplasia.

“Our survival data, supported by biopsy specimens taken during follow-up, would further imply that these latter lesions are either incompletely removed or multifocal with malignant clones left at a transurethral resection. Contrary to emerging views,^11,12 our data also suggest that metastases may arise as a consequence of late mutations rather than being determined already by the early mechanisms of malignant transformation.

“In conclusion, our data indicate that the probability of progression to a more aggressive and lethal phenotype may increase after long-term follow-up of prostate cancers that are diagnosed at an early stage and initially left without treatment. These findings argue for early radical treatment of patients with long life expectancy. Not only would such surgical intervention potentially prevent deaths, it would also convey prevention from disability caused by local tumor growth.”

Ignoring the harms of prostate cancer progression while highlighting the harms of treatment misinforms patients at decision time.

I think that’s the point of the work being done to which Mike refers — how decisions are made. The way I process the data and my background — my paradigm — lead me to a conclusion that may well be very different from yours.

At age 50 you have a life expectancy of between 14 years and 43 years with a median of 29 years, according to one set of US tables. Now that’s a long time to try and predict anything. To get some idea of this, I suggest to men that they look back (in your case) 14, 29, and 43 years and try honestly to evaluate just what has happened over the intervening years that could have been predicted in that time frame. Just what did you plan for your life at age 21 — and has everything turned out as you expected? Most of us just couldn’t have predicted what we landed up doing and where, so how can we expect an accurate forecast of the progression or failure to progress of a notoriously complex disease?

But we try to do so and if I was in your position now, knowing what I do, I’d try to calculate the odds of dying within that time span, using whatever information I could find. There are various studies calculating disease-specific mortality at 10 years and some of these extend to 15 or even 20 years. There seems to be some kind of consensus that at 10 years the disease-specific mortality with a diagnosis like yours is less than 5%, treated or not. There is less agreement on the figures beyond that but at 20 years it is probably less than 10% to 15%. If I were really conscientious I’d then look at the specific mortality risks for the eight causes of death which are ahead of prostate cancer in the pantheon of fatal occurrences for men in the US. This would give me a relative risk for prostate cancer vs. “the rest.”

Then I would try to look at the relative risks over time. For example the basic data on the SEER site shows half the men who died from prostate cancer between 2002 and 2006 were more than 80 years of age. The figures also show that over 90% of the men who died were over the age of 65 — 1.4% of the deaths were in men between 45 and 54 and 7.2% between 55 and 64.

The next question is to ascertain, if possible, the diagnostic criteria and treatment choices (if any) of the unfortunate men who died younger than the median. If you do this, you will find that they do not match yours. Bill, another young man, in his post categorizes himself as an insignificant minority whose situation is trivialized and who says he often finds his position to be a lonely place. I can understand his position and his view, but believe that he is not correct when he says that he is insignificant and his diagnosis trivialized. He isn’t and it isn’t, at least as far as I am concerned, but he is not typical of young men diagnosed and it is important for other young men to understand that — if you haven’t read his story you can find it here.

Against this collection of information, I’d try and set the “cost.” There is no curative treatment that does not carry side effects. The degree of these side effects, their effect on the man and his family, and the long-term outcome will all vary considerably — and unpredictably, so the available relevant data would be scoured to try to obtain some objective viewpoint and to consider the cumulative effect of these over a period of up to 43 years.

It is then that the individual cost-benefit analysis must be made, weighing up the various ingredients and coming to a conclusion that suits the individual man and his diagnosis.

But that’s my way of dealing with things — and it might not be yours.

Sure many people have “unnecessary” treatment, in the sense that if they had not had treatment, their prostate cancer would not have killed them. So they suffered side effects for “no reason.” But, of course, there is presently no way of knowing who has the cancer that won’t kill you and who has the cancer that will. (And even some of those tests that could potentially shed some modest amount of light on this question — PAP, ploidy analysis, color doppler, Combidex scan, re-evaluation of Gleason score by expert pathologist, etc., many/most don’t do).

And if patient defers treatment (AS, or whatever), the chance of missing the window of curability is greater than 0%.

So, say you’re a 50-year-old guy, in great health, diagnosed with a PSA of 4.5 (up from 3.5 the prior year), Gleason 6 (confirmed by an expert pathologist, and no tertiary pattern observed), 2/12 cores positive, 10% in each core. What do you do? “Talk with your doctor about the various options” is not a good answer. I mean, of course you do that, but then what? Guys like this get surgery because they figure, “The doc can’t tell me that this definitely won’t advance and kill me — I know there’s side effects, but I have 3 young kids and I don’t want to die.” Should this guy be thinking differently?

From the University of Texas MD Anderson Cancer Center: “Your best chance for surviving prostate cancer is detecting it early. When found early, there is nearly a 100% chance for cure.”

From the ProstRcision.com main page: “The 10-year documented cure rate for ProstRcision® is 88% for early and intermediate cancer — higher than any other prostate treatment option available today.”

From a “Letter from the Director” of the Uro-Radiology Prostate Institute: “In our 10 years experience, our biochemical results are excellent with 89%, 94%, and 95% disease-free survival rate in high, intermediate, and low-risk patients, respectively.”

From the main prostate cancer page at the Mayo Clinic: “One finding: The 10-year mark cancer-specific survival rate for Mayo Clinic patients is more than 90 percent.”

One of the problems that we all share is “statistical confusion.” So the point that the good Dr. McHugh made above is absolutely correct: how one presents the data affects how it is interepreted by differing groups of people.

In the piece above, I very carefully included two pieces of information. The first, early in the piece, was that 27,000+ men die of prostate cancer each year. The second was that only 3% of all the men diagnosed with localized prostate cancer actually die from it. But you are absolutely correct because most people don’t seem to be able to process these two pieces of information simultaneously. I have no idea why not. It doesn’t seem like a particularly difficult pair of ideas to handle.

This is also one of the reasons why I am uncomfortable with Terry’s desire to emphasize that prostate cancer is only the second highest cause of cancer death for men in their 80s. It is certainly true, but it marginalizes even further the men (like you) who are diagnosed young with aggressive forms of the disease.

I greatly appreciate your discussion on how the complete or incomplete disclosure (= quality) of the message (= ,the doctor’s communication) that precedes or follows a PSA screening and biopsy will have an impact on a patient’s choice of a prostate cancer treatment option.

As a prostate cancer survivor, patient educator and author (Conquer Prostate Cancer: How Medicine, Faith, Love and Sex Can Renew Your Life), I have only one quibble: You’ve understated the outcome of treatments like surgery and radiotherapy in responding to Shimon that “In many men the side effects of radical prostatectomy are a significant problem.”

Rather than “many men”, it’s MOST men that will experience serious treatment side effects. After considerable research for my book and its companion blog it is hardly a “stretch” to conclude that some 2/3 of men (plus or minus 10%) who go through radiotherapy or surgery will experience ED, and 5% to 15% will end up with permanent incontinence, apart from other effects like radiation burn.

This is the heart of the matter which has led men like Dr. Otis Brawley, the American Cancer Society’s Chief Medical Officer, to oppose routine PSA screenings and their aftermath. His concern, based on his public statements and in a personal comment to me months ago, is that the PSA and biopsy are dangerous for patients IF the risks and benefits are not thoroughly discussed with them in advance of a PSA and biopsy and treatment decisions.

I don’t agree with him and others at the ACS, or with other leading groups and scholars who totally dismiss the value of PSA screenings until a more reliable biomarker is discovered. However, I concur with him, you, and the social scientists you’ve mentioned, that being ill-informed is a terrible basis for determining what treatment is in a man’s best interest, — be it surgery of the robotic kind I had (RALP), or AS, or radiotherapy, etc.

The likelihood of getting ED and the possibility of becoming incontinent are serious outcomes that every man must weigh, in contrast to the 3% chance of dying from prostate cancer. I know that is your view, but statistically the majority of men are likely to face such side effects, and patients need to know this in advance of opting for whaever treatment they select, including active surveillance.

One piece of information I have yet to learn is what percent of patients who choose active surveillance suffer from high stress or actual pain? Shouldn’t these additional factors be assessed, not just death, when weighing risks or benefits before deciding what treatment to choose?