June 05, 2006
Price Competition in Cancer Drugs?
The American Society of Clinical Oncology meeting underway in Atlanta is generating lots of excitement about the next generation of targeted cancer therapies. It’s the most interesting development in cancer therapeutics since the arrival of drugs like Gleevec for one form of leukemia and Herceptin for breast cancer, which were the first targeted drugs.
While David Hamilton in Saturday's Wall Street Journal (subscription required) raised the specter these new drugs will lead to price competition, the real use for these new options will either be in combination with the older drugs or on their own after first line chemotherapy has failed.
Cancers are, by definition, the rapid proliferation of mutant cells. Simple population genetics inside the body dictates that when cancer cells are subjected to a change in the external environment – for instance, the presence of a chemotherapy drug – the mutant cells that survive and thrive become the dominant strain. Patients then need a different drug to keep the cancer at bay. That's why most chemotherapy agents when tested in controlled clinical trials extend life by a few months, and are almost never a cure.
So this new generation of cancer drugs are not like the “me too” drugs that populate the blood pressure, anti-cholesterol, or antiacid classes of medicines. They are badly needed follow-on options for desperately ill patients who, if they’re willing to put up with the side effects, deserve a chance at a few more months of life.
That’s also why the analysts quoted in Hamilton’s article who raised the specter of price competition are raising false hopes for the people or insurers who will pay for these drugs. As Andrew Pollack suggested in his story in the New York Times, many of these drugs will be used in combination with older medicines, so even if their prices come down slightly, the total bill for patients will be higher.
In any case, I doubt there will be significant price competition. It certainly won’t be at the same level as generics, which often come to market at half the price of a brand name drug going off patent. A study a few years ago by the drug-industry funded Tufts Center for Drug Development suggested new patented entries coming to market in a class previously dominated by a single patented drug led to significant price competition. But if you looked closely at the data in the study, most drugs actually came in within 10 percent of the price of the innovator’s product. When you’re talking $100,000 per year, which is the price of many of the new targeted cancer therapies, that’s no break at all.
The drug industry should be congratulated for coming up with these new options for cancer patients. But as cancer moves closer to becoming a manageable disease using a succession of drugs that extend life for six months or a year and each commands a $10,000 a month fee, one has to wonder how the health care system will be able to afford it.
Cancer therapeutics may well be the wedge for developing mechanisms for divorcing the way we fund innovation from the way we fund health care.
Indeed, one should be skeptical about claims of price competition. And one should be skeptical of anything coming from Tufts.
The Tufts study was done essentially in response to the recognition that when therapeutic competitors enter the market, there isn't much sign that the price of the first drug comes down. They wanted to provide a little bit of ammunition to counter that.
Later entrants come in at a lower price perhaps to shave a little market share from the few price-sensitive consumers...like some formularies. So, in a sense, "me-too competition" could be viewed as market segmentation as much as it is competition.
Posted by: cb at June 5, 2006 11:24 PMMerrill,
I would hope that you post this comment and allow me to correct the mischaracterizations of my study that have been fostered by your book, again in your posting here, and, more egregiously above by the posting from "cb." Your book cites my study simply as something coming from the Tufts Center for the Study of Drug Development (CSDD). This is wrong on two counts. First, it suggests to the reader that it is a vanity publication. In fact, it is published on a govermnet website (http://aspe.hhs.gov/health/reports/Drug-papers/dimassi/dimasi-final.htm). Your book contains lots of full citations to other articles, including many with just web addresses. Why was my paper not appropriately cited? Second, as indicated on the title page of the paper, while my affiliation is with Tufts CSDD, this was NOT a Tufts CSDD paper. It was done on my own time under contract from the government (Department of Health and Human Services - Clinton administration) and used data acquired by the Drug Value Group in the Schneider Institute for Health Policy at Brandeis University, with whom I consulted for several years. As far as I know, the Drug Value Group had not received any drug industry money. DHHS had asked the group to examine, using these data, drug pricing trends for therapeutic classes. The launch price analysis to which you refer is only one of many analyses in the paper.
Your poster, cb, professes with certitude knowledge of the source and genesis of this paper. The fact that cb is so clearly wrong suggests that we should "be skeptical" of anything that cb writes.
With regard to the substance of the paper, I don't know what you mean by "new patented entities coming to market in a class previously dominated by a single patented drug." This appears to suggest that all of the data points examined were for second entrants to a class. In fact, only four of the 20 drugs examined were second entrants (the rest being later entrants).
The average price reduction for the new entrants in my analysis was 26% relative to the class price leader and 14% relative to the weighted mean price for all drugs in the class that were on the market at the time that the new entrant was launched. These figures did not include rebates (which were unavailable). It is widely recognized that rebating (for the insured sector) has been significant once there are more than a couple of drugs in a class on the market.
Particular launch prices will depend on conditions in individual markets and what evidence exists on the differential attributes of the new entrants. Five of the 20 drugs were introduced essentially at parity (which was defined as being within 3% of existing prices). However, four of them were in just one class. The fifth was Vioxx, but it was introduced just five months after the first entrant (Celebrex). These Cox-2 inhibitors, incidentally, were introduced at prices that were essentially the same as those for older branded NSAIDs already on formularies (even though they had purported, but not then proven, GI side effect advantages). Using generic drug prices as the bar by which price competition is to be judged is inappropriate. No one argues that brand price competition would result in generic-level pricing. I would venture to say that there isn't any industry where brand products are priced the same as generic, or no-name, products.
I was constrained by the length of the period for which we had data (1995-1999) from doing something like what Lu and Comanor did in a study cited in my paper that used a longer (but much older) period ("Strategic Pricing of New Pharmaceuticals," Review of Economics and Statistics 1998 80(1):108-18). The paper provided a rigorous statistical analysis controlling for a number of factors. In their abstract the authors wrote: "In addition, the number of branded substitutes has a substantial negative effect on launch prices, which reflects the importance of competitive pressures." In the text of the article they wrote: "increasing the number of direct substitutes from one to two leads on average to a 38% decline in the ratio of a product’s launch price to the average price of its predecessors; while increasing the number of substitutes from two to three leads to a 19% decline." Finally, the authors also found that: "More numerous rivals have the expected effect of slowing price increases." Since discounting and rebating increased over time, the effects that the authors found could be more pronounced with more recent data. As the authors noted: "The period covered in this study largely predates the rapid growth of managed care, which is relevant because it led to much greater price discounting. For this reason, the results presented here may not fully represent current conditions in pharmaceutical markets."
Joseph A. DiMasi, Ph.D.
Posted by: Joseph A. DiMasi at June 6, 2006 04:12 PMprovocative question about funding care. what's precedent now for people who require long-term expensive chronic care -- or is there any? my guess is that people run out their lifetime benefit limit. is that happening to any significant number of people now and, if so, is there any public response?
Posted by: jim jaf at June 6, 2006 04:46 PMDr. DiMasi deserves a response. But first a correction. I did not check carefully to ascertain that it was the government that backed his study on price competition from new entrants. However, the Tufts Center, with which he is associated, is financed in part by the drug industry, and I presume the original article (which I don't have in front of me) disclosed those conflicts of interest. Also, my claiming that the new entrants used in his study were all in markets dominated by a single, patented drug was simply incorrect. However, I don't think that changes the essence of my argument.
On the merits, I would suggest a careful reading of his note does not refute my position. He says the "average" price reduction on new entrants was 27 percent compared to the price leader and 14 percent compared to the mean of all the drugs in the class when there were more than one. Averages tell us very little. My point was that more than half were under 10 percent, and my larger point was that in cancer drugs used for follow-on therapy (sometimes called salvage therapy) or in combination with older drugs, we can expect the new entrants to be very close to or even higher than the older drugs. If Vioxx and other Cox-2s with "purported" benefits (good word to describe something not proved in clinical trials or to the satisfaction of the FDA) enter at the same price as other patented NSAIDs, what do you think will happen with new cancer drugs that have very tangible benefits for patients failing older therapies?
It was economic historian Daniel Kevles' original studies of drug industry behavior that first showed that the pharmaceutical industry competes on anything but price. I believe it is as true today as it was in the early 1960s, when Sen. Estes Kefauver held antitrust hearings into the pricing practices of what he called "the antibiotic cartel."
Posted by: Merrill at June 7, 2006 07:20 AMI apologize forgetting the asterisk on Dr. DiMasi's paper. But the point remains the same. The question is whether me-too drugs provide price competition as we observe in markets for other products, including generic drugs? Or are we just observing penetration pricing that has only a small impact on the price of the leader. Do we see substantial competition or very limited competition?
I think we could reach a consensus that Lu and Comanor find, at best, evidence of limited competition.
I apologize if I have unfairly characterized the genesis of this particular paper. But it isn't a secret that industry funds CSDD (although not this particular paper) because the work is used to create industry talking points. And these talking points are frequently misleading.
http://mednet3.who.int/prioritymeds/report/append/8342.pdf
NIHCM Quote: “This pattern suggests that when there are several new NME’s [New Molecular Entities] in a therapeutic class, price competition among them is limited.”
REALITY:
In its report, NIHCM suggests that having several NMEs in a therapeutic class does not lead to price competition. However, a study by Dr. Joseph A. DiMasi of Tufts University found that new drugs in a class are often priced lower than existing drugs in the class.6....
Does Dr. DiMasi's paper really refute the contention that competition among me-toos is "limited"?