Cutting the Cost of Cancer

Someone has to pay for all the research. But who can?

A very expensive needle in a haystack.

Photographer: Dan Kitwood/Getty Images/Cancer Research UK

How much should we pay for cancer drugs?

The answer, in the past, has frequently been "How much you got in your pockets? That, plus 10 percent." Insurers and patients have been pushing back on the high cost of drugs, negotiating lower rates and switching to generic treatments. But cancer is special. It's hard to be a discriminating shopper when someone tells you that you have 18 months to live.

Interestingly, cancer drugs have been (relatively) immune to the perennial complaints about Big Pharma. I'm not saying you never see anyone ask whether it's really OK to spend tens of thousands of dollars to marginally extend the life of someone with pancreatic cancer. But whenever I start an article about how awful Big Pharma is, I know that I'm much more likely to hear about depression, diabetes or statins than about cancer treatments. Who wants to suggest that patients should just give up?

And yet, the complainers do seem to be focusing these days more on treatments for things like hepatitis C and cancer -- killer diseases for which we suddenly seem to be getting better treatments. Gilead's Sovaldi, which treats the former, attracted a lot of attention, and now oncologist Leonard Saltz is saying that pricing for cancer treatments is completely unsustainable.

What's interesting is that these complaints are an inversion of the most common charge leveled at pharmaceutical companies: that they're charging us nosebleed prices for snake oil, or for something an older drug would do for a lot less money. The problem with these new drugs, however, seems to be that they work too well, and therefore it's reprehensible to charge prices that will make it impossible to deliver enough of them.

Look at these results for melanoma using a new drug combination that costs almost $300,000 a year. This is not quite penicillin-level miraculous -- the side effects are serious, it doesn't work in all patients, we don't know how long the results last -- but in a field like cancer research, and particularly with something like melanoma, this is probably the next best thing.

Matthew Herper has some very useful thoughts on this. I don't myself know what the "right" price for these drugs is. The closest I can get is to say that if these drugs turn out to only extend life for a year in 60 percent of patients, the price is probably too high. On the other hand, if unleashing the immune system on tumors can actually produce years of healthy remission, it's probably not too high a price to pay. And in the long run, it won't matter.

There are lots of reasons that cancer drugs cost so much, but one reason is, ironically, that they don't work very well. Because so many cancers don't have particularly effective treatments, doctors tend to keep throwing one thing after another at them. And new drugs keep getting developed in the hopes that this one might do the trick. Contrast that with blood pressure drugs or antibiotics, where there are lots of great ones, so you're going to have a hard time getting enough people to buy your $10,000-a-year hypertension drug (even if you can get it approved).

If the new round of immunology-based treatments are dramatically more effective than their predecessors, then come 2026 or so, some very good treatments will become available for many previously intractable cancers. Those treatments won't be as cheap as aspirin, but they will be a limiting factor for anyone who wants to put their new 80-squillion-dollar melonoma treatment on the market.

Of course, even if that happy scenario works out, we will still have a big short-term problem: How do we finance an explosion of very expensive, very effective treatments for deadly diseases? How do we push back on costs without cutting off the supply of effective new drugs? These therapies are very expensive to develop, and someone has to pay for all that research.

I think that many of Herper's suggestions, such as considering how prices should be adjusted when drugs turn out to have a wider market than initially anticipated, will help. Getting governments involved with more alternative incentives, such as prizes, could also offer part of the solution.

But the broader answer is that we are probably not going to find a perfect answer. We often talk about the purpose of research as being "finding a cure for cancer" -- but we rarely ask if that wouldn't create problems of its own. 

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

    To contact the author on this story:
    Megan McArdle at

    To contact the editor on this story:
    Brooke Sample at

    Before it's here, it's on the Bloomberg Terminal.