Cancer Decisions - Free Newsletter

HERE AT THE MOSS REPORTS

The effect of media reporting on the public's awareness of, and demand for, specific medical treatments was the subject of a 2002 review by the prestigious Cochrane Collaborative. (Cochrane Reviews are designed to provide reliable information on the risks and benefits of medical treatments, based on careful analysis of the available scientific evidence.) The review found that favorable media publicity about a drug or treatment was strongly associated with an increased public demand for that treatment. Media coverage of medical developments, in other words, ultimately has a strong influence on medical practice and prescribing habits (Grilli 2002).

This, of course, is not news to the drug companies. They have long been aware that public demand can be used to drive prescribing habits. The way in which developments are reported is therefore of crucial importance. It is no accident that industry-sponsored press conferences have become the preferred venue for announcing clinical trial results, and that industry press releases are now one of medical journalism's principal sources.

This week I continue my three-part series on the recent announcement that the drug Celebrex may have a role in the prevention of colorectal cancer. Is this drug truly a prevention breakthrough? Or are reports of its promising potential in cancer prevention just another example of media hyperbole?

Over the past 30 years I have written and published extensively on the subject of cancer and its treatment. The fruit of this daily labor has been a comprehensive series of more than 200 individual reports on different cancer diagnoses – The Moss Reports – each one of which examines both the standard treatment options that are likely to be offered for a particular cancer diagnosis, and the possible alternative and complementary approaches to that disease.

If you would like to order a Moss Report for yourself or someone you love, you can do so directly from our website, www.cancerdecisions.com or by calling 1-800-980-1234 (814-238-3367 from outside the US).

Also downloadable from our website are our special reports on important topics in the field of cancer prevention and treatment. These reports are priced at $9.95 each. Currently available are the following:

I am also available for phone consultations with those clients who have purchased a Moss Report. Many patients have told us that such consultations can be enormously helpful in drawing up an effective treatment strategy. To schedule a phone consultation, please call 1-800-980-1234 (814-238-3367 from outside the US) or send an email to Jacquie@cancerdecisions.com.

We look forward to helping you.

DOES CELEBREX ACTUALLY CUT RISK OF COLON CANCER? PART TWO

Last week we began a consideration of the use of Celebrex (celecoxib) in colon cancer prevention. We continue that discussion this week.

According to Michael Santoro, MD, writing at medicinenet.com, "polyps in the colon are extremely common, and their incidence increases as individuals get older. It is estimated that 50 percent of the people over the age of 60 will harbor at least one polyp." Thus, the number of people who are potential candidates for a drug that prevents colon polyps is almost limitless. There are almost 50 million Americans aged 60 or older. And so just among seniors there is a huge potential customer base for a drug such as Celebrex.

Of course younger people can also develop polyps. But the current recommendation is to begin screening with a procedure called flexible sigmoidoscopy in those over the age of 50. That is because more than 90 percent of colon cancers are diagnosed in people over the age of 50 (although presumably most of these arose as colon polyps years earlier). This would add an additional 25 million Americans to the list of potential customers. But let us focus our attention on people who are around 60 years of age and stand a relatively high risk of colon cancer.

Let us, for the sake of argument, construct a little thought experiment, involving 2,000 of these older individuals. Their colon polyp status is unknown but – based on Dr. Santoro's figures – we can presume that half of them have some polyp formation. One thousand of these 2,000 people are given Celebrex, while the other 1,000 are not treated, or are given an inert placebo. Since about 20 percent of polyps will eventually turn into malignant tumors, we can project that, without treatment, there would eventually be 200 colon malignancies in that group of 1,000 (although this is a process that could take decades to manifest itself).

Let us further stipulate that every one of the treated individuals would benefit from Celebrex to the degree that was claimed in the recently reported APC clinical trial. Thus, by taking the high (800 mg) dose of Celebrex, these 1,000 individuals would have a reduction of 24 percent in their likelihood of developing colon polyps at three years. We can therefore project that instead of 200 out of these 1,000 people eventually developing colon cancer, only 152 of them would do so, because 48 people who might otherwise have gone on to develop colon cancer would have now have avoided the disease.

Let us now assume that all of these individuals had their tumors discovered in stage I (the earliest and most treatable stage). This would be likely if they all availed themselves of regular screening techniques, such as tests for fecal occult blood every year and/or endoscopic tests (sigmoidoscopy and colonoscopy) periodically as well. In that case, the great majority of them would be surgically cured of their tumors. That is because the five-year survival (or "cure") rate for stage I colon cancer treated surgically currently stands at 93.2 percent (O'Connell 2004). Under such ideal circumstances, if 48 people were treated for stage I colon cancer then three of them (3.26, to be exact) would die of this disease. Put another way, one would have to treat 1,000 patients with daily high does of Celebrex over a period of years in order to save 48 of them from having colon cancer with an ultimate saving of 3 lives.

But perhaps I am stacking the deck by assuming that all of these patients would be found to have cancers in its earliest stage? After all, not every one at risk of colon cancer has regular screening. So now let us imagine a more realistic scenario and say that these 1,000 patients would be no more vigilant about their colon health than the majority of the American population. In that case, the average five-year survival rate would decrease to 62.5 percent (O'Connell 2004). Eighteen individuals out of the original 1,000 would have their lives spared (or 1.8 percent of the total) over a period of three years.

There is certainly no denying that saving the lives of between 3 and 18 people per 1,000 at risk for colon cancer is a desirable goal (as is sparing a larger number the duress of a surgical operation). Let us say, as a rough figure, that around 10 people per 1,000 (1 percent) would have their lives saved from colon cancer by taking Celebrex.

But here's the kicker. In the current clinical trial, out of every 1,000 people who were administered Celebrex, 34 suffered heart attacks, strokes or other serious cardiovascular events. Since there were 10 such events in the placebo groups, we can infer that 24 of these catastrophic events were caused directly by the drug itself (2.4 percent).

How many lives would this cost? It is not certain. The mortality rate from heart attacks in the US is 36.8 percent; that for strokes somewhat less. Precise figures are impossible here because we do not know how these Celebrex-induced cardiovascular events would compare to naturally occurring heart attacks, strokes, etc. But it is certainly plausible that around 10 out of 1,000 people (i.e., 1 percent) would die within the first three years as a direct result of taking Celebrex. Nor do we know how many people would die in subsequent years, although it is a fair assumption that the damage is not confined to the three years that happened to be the timeframe of these clinical trials. Thus, although precise figures are difficult to obtain, the deaths avoided from colon cancer would be roughly equal to the deaths from cardiovascular disease caused by this drug. This is simply too high a price to pay for any sort of "chemoprevention."

I am sure that Pfizer would be thrilled if every person in America and the world who was at risk for colon polyps decided to take Celebrex. But the result could be catastrophic. If you were to treat, say, the 60 million Americans who because of their age or other factors were at greatest risk for colon polyps, giving each of them a daily 800 mg dose of this drug, you would, according to the results of these studies, create an excess of around 600,000 deaths from cardiovascular disease, or around 200,000 per year.

Among patients who already had a history of serious adverse cardiovascular events, the news is even worse: while a total 3 percent of them experienced some new event on placebo, this rose to an alarming 8.8 percent who experienced new events while on Celebrex. In fact, the NCI trial had to be halted because of all the heart attacks, strokes, etc. occurring in the treatment group. There are already 7.5 million heart attack survivors in the US and 6 million stroke survivors. Therefore one can estimate that there probably about 6 or 7 million Americans who have had a heart attack or stroke and who also have colon polyps.

One hopes they would be strictly excluded from any Celebrex treatment strategy. If all of them were to take Celebrex for three years, then half a million of them would suffer new heart attacks or strokes.

And then there is the economic consideration. Celebrex is rather expensive. At discount Web sites, 800 milligrams of Celebrex sells for around US $10 per day. It is unclear how long people would need to stay on the drug, but it seems likely that if the drug were withdrawn, the beneficial effect would diminish or disappear. Thus, patients might be expected to take the drug for the rest of their lives and to spend around US $3,500 per year on the medication. Life expectation in the US currently stands at 77.7 years. Thus if one took Celebrex from the time one reached 50 (assuming it didn't lead to premature death) one could expect to spend between $50,000 and $100,000 on the drug.

The cost to society would also be staggering. The total number of US seniors is currently around 60 million. If one treated 30 million or so individuals at high risk of colon polyps with Celebrex, the retail cost of the drug alone would be around $300 million per day, or $109.5 billion per year. For perspective, that is 25 times the entire budget of the National Cancer Institute! All this for a drug that has little, if any, impact on the overall death rate.

Because of nightmare calculations like this, it is noteworthy that few scientists have jumped on the Celebrex bandwagon. In fact, the authors of the study acted with proper circumspection, at a time when AACR as an organization (with its "breakthrough" rhetoric) seemed to lose its bearings.

"Celecoxib is an effective agent for colorectal adenoma chemoprevention, but it cannot be recommended for prevention of sporadic colorectal adenomas until issues regarding cardiovascular toxicity are addressed," said Monica Bertagnolli, MD, the surgeon at Brigham and Women's Hospital and the study's principal investigator, who presented the findings at the AACR meeting (Goldberg 2006).

Dr. Bertagnolli and other scientists do believe that Celebrex can be rescued as a potential preventive treatment for colorectal cancer by selecting patients in the future who are (a) at high risk for recurring polyps but also (b) at low risk for cardiovascular events. Specifically, a future trial might enroll patients who have had large, multiple polyps (a population believed to be at high-risk for the future development of colon cancer) but who also have relatively healthy hearts and cardiovascular systems.

Well perhaps this can be done. But once the drug is approved for this indication many individuals who are currently unaware of their own cardiovascular risk status might start taking it as well. It is difficult to know the actual physical condition of one's own heart or cardiovascular system. Many of us have multiple risk factors such as advancing age, excess weight, a sedentary life style, elevated blood pressure and cholesterol, etc. and yet have no outward signs of cardiac compromise.

Embracing Celebrex would mean taking a drug over many years that, while it might prevent the formation of polyps (some of which might possibly lead to colon cancer at some unspecified future date), would at the same time definitely increase your risk of having a heart attack, stroke or other dire event in the short to medium term. That doesn't seem like a very good bargain at all.

On a more theoretical level, there continues to be a great deal of uncertainty as to what effect Celebrex has on the more dangerous kind of polyps. For instance, it is astonishing to learn that people who have small polyps "probably aren't at any significant risk for colon cancer anyway," according to Raymond DuBois, MD, director of Vanderbilt-Ingram Cancer Center, a member of the scientific advisory board of the Pfizer trial and a discussant at the AACR presentation on the topic.

The straight-talking Dr. DuBois also raised a serious question about the clinical significance of the Celebrex and polyp finding. "The biggest question I have now is, 'Which polyps are we reducing by the treatment?'" DuBois asked. "Are we reducing ones that are going to progress on to cancer, or are we reducing the ones that wouldn't progress on, and we must design some way to answer that question. Not all polyps are alike" (The Cancer Letter, April 14, 2006). It would be necessary in future trials to "interrogate" these polyps at the molecular level to figure out which kind of polyp was actually being affected. If these are relatively benign polyps, it is incorrect to even describe them as "precancers," as almost every news story on this development has done.

What makes this question especially important was the little-noted fact that in the aforementioned APC trial more patients in the Celebrex arm developed colorectal cancer than in the placebo arm (1 patient in placebo arm versus 6 in the Celebrex arm). Since most polyps are ultimately harmless, there is a possibility that Celebrex is preventing the formation of less ominous polyps, while efficiently selecting for the more dangerous and aggressive ones.

This would be a terrible and ironic outcome for these trials, which were designed to prevent, not cause, colorectal cancers. While one can hardly generalize from such a relatively small number of excess tumors, it does pose the pressing question as to what the actual biological effect of these NSAIDs might be. On this score alone, caution is advised.

Perils of Surrogate Markers

Until there is a way of identifying precisely which polyps are truly dangerous, and which are not, any trial that simply demonstrates the ability of a drug to suppress polyp growth cannot be classed as a breakthrough, as AACR has done. At this stage, all we know is that some of these polyps are associated with an increased possibility of the eventual appearance of colorectal cancers. The general association of polyps with cancer was used, in these clinical trials, to make polyps a "surrogate marker"– that is, a substitute measurement that may, or may not, be useful in more rapidly judging the effects of a treatment.

Such surrogate markers are controversial in the field of cancer because they often do not correlate with actual patient benefit. But studies that would establish the relevance of preventing polyp formation to actually improving survival statistics from colon cancer would take decades to complete. The exaggeration and misrepresentation of trial results is almost inevitable when surrogate markers are used, and when mere associations are conflated with actual causation. The enthusiasm for Celebrex is thus a house built on sand.

To be concluded, with references, next week.

--Ralph W. Moss, Ph.D.

**NOTE** To view this page in a more printable format, please CLICK HERE.

IMPORTANT DISCLAIMER

The news and other items in this newsletter are intended for informational purposes only. Nothing in this newsletter is intended to be a substitute for professional medical advice.