Sunday, September 12, 2010

Prostate Cancer, The Times, and Naivety

Today the NY Times had an editorial regarding the costs of health care. They specifically say:

Consider the prostate-specific antigen test, which is widely used to screen men for possible prostate cancer. In an Op-Ed piece in The Times in March, Richard J. Ablin, the doctor who discovered prostate-specific antigen, described the test as “hardly more effective than a coin toss” at distinguishing who is at risk, and lamented that the test’s popularity has led to “a hugely expensive public health disaster.”

Each year some 30 million American men undergo the test at a cost of at least $3 billion, and many go on to have surgery, intensive radiation or other damaging treatments that may not have been necessary.

We have already argued that the PSA test saves lives if properly interpreted. It requires long term tracking and specifically tracking of velocity as well as age specific values. Despite the regrets of the good doctor who discovered it, yes it may be over used and yes it may result in excess biopsies but yes it saves lives. Imagine what happened when the Government panel decided to reduce coverage on breast exams, women revolted. Well, perhaps the editorial was written by some woman who desires to get back at men. Despite the concerns PSA tests save lives. $3 B is peanuts compared to $300 B for fattys! So where is the outcry from the Times, the Government, just look at the Surgeon General!

The Times continues:

The Obama administration started the process, committing $1.1 billion from stimulus funds to finance comparative studies. The new reform law will move that ahead, setting up a nonprofit, independent institute to organize the work. The comptroller general will appoint a governing board of 19 members, representing patients, doctors, manufacturers and others, including two designated federal health officials.

If the institute works the way it is supposed to, patients, doctors and the government will have better information about what works and what does not, what may be worth the extra cost and what does not make sense. Even then, the legislative language is so convoluted there is no guarantee that even the most credible findings will help ensure that patients get the best and most cost-effective treatment.

The CER, comparative effectiveness research, and the like look at averages. There are NO AVERAGES, only for accountants. Each patient is different. As I have said many times, when a patient asks, "Doc, what are my chances?". the only answer is, "I have no idea." The patient will live or die, a priori one cannot say what the end point is and chances apply only to large groups!

The current Administration is looking at only one way to reduce costs, ration. Pure and simple, that is what the Soviets did and that is what they in the little minded manner will do. It will result in death to many, no via death panels but via ignorance, stupidity and arrogance.

In a recent NEJM article on the mammogram issue they demonstrate an interesting analytical measure of treatment versus no treatment. Specifically:

More generally, the net benefit of all medical treatments is a continuous function of three factors: the risk of morbidity or mortality if untreated (RiskNoRx), the treatment’s relative risk reduction (RRRRx), and the treatment’s risk of harm (HarmsRx):

Net Benefit = (RiskNoRx×RRRRx) – (HarmsRx).

As the risk of no treatment (RiskNoRx) decreases, the net benefit of treatment will decrease, even if the treatment’s relative benefit (RRRRx) remains constant. Indeed, for many interventions, if the risk of no treatment is low enough (e.g., if we lower the threshold for treatment too far or if a patient’s life expectancy is relatively limited for other reasons), then the side effects and risks of treatment will dominate, and the treatment will result in net harm.

Since the risk of no treatment varies dramatically among patients for almost every disease or condition, even a highly effective intervention will show a gradient of net benefit in a given population.

Now how would this apply to the case of prostate cancer. It is a bit complex since it depends upon what stage. Let us look at a simple but useful model.

1. Assume the Patient, Pt, is under 50, and has a PSA of 3.0 with a three year velocity greater than 0.75 per year. One would suggest a biopsy no matter what. The treatment is a biopsy, and its relative risk is quite low, possible infection, modest discomfort. The relative risk reduction is quite high, namely if we find a Gleason 7 or greater than we have possibly given the patient a significant upside. The harm of the treatment is minimal. So do we do it?

2. Assume we did 1 above and then we find a Gleason 7. What then? Now we have a real level of potential harm, impotency and or incontinence. As to survival, well that is a good question since we really nee to understand the status of the cancer pathways which for the most part we do not look at. Is PTEN knocked out? etc. Thus we have a real problem telling what the upside would be.

3. Assume another patient. Say a 70 year old health man with a slowly rising PSA. Say it is 2.5 but the velocity has suddenly hit 0.75. Should we biopsy? Let us further assume the Pt had a father who dies of PCa and the father went from 4 to 40 in 2 years and 40 to dead in 2 years. Then we would most likely do a biopsy. Then assume we have upon a biopsy not PCa but PIN, prostatic intraepithelial neoplasia, high grade but focal, then what, do more biopsies? The answer is yes. Is it a wast of money, no, and if the Pt pays for the biopsy it would cost the system nothing.

4. Assume a third Pt has no family history of PCa and had a PSA of 9 and a biopsy performed. Gleason 5 is detected. Now what? Watchful waiting. How long? If the Pt is 78 years old and has had Colon cancer and has Type 2 Diabetes, with HbC1A of 7.9, should we follow, operate, what? This is the difficult case.

Thus as I have argued, each patient is different and developing general and far reaching guidelines is absurd! Thus it is naive to think that any general set of rules will cover all cases, and in fact it may not cover any. Each human is different and each physician approaches their patient recognizing that there is a person before them not some algorithm ready to be decodes according to a set of rules. Naive, stupid, reckless, arrogant, whatever one may want to call it, it results in a disservice to the patient and in turn humanity.