Imagine that instead of setting out to invent a better lightbulb, Thomas Edison had announced his intention to invent a light-emitting diode that you could use to illuminate your kitchen. This isn't completely far-fetched: the first examples of light-emitting diodes (LEDs) began to appear as early as 1907. But it wasn't until the 1960's and 70's that useful, visible-spectrum LEDs began to appear, and LEDs haven't been used to light kitchens until very recently. Thomas Edison, had he set out to make a useful, household LED, would have been doomed to failure beacause it would be years before basic science made the necessary technologies possible.

When Richard Nixon declared the conquest of cancer "a national crusade" in 1971, cancer researchers were inevitably set up to be viewed as failures. Although at the time the recent molecular biology revolution led people to think that disease conquest was just around the corner, now we can look back and see that the War on Cancer had no hope of achieving its goals in the 1970's. Scientists are being punished for that hubris now, in the form of misguided news pieces such as Newsweek's current exposé: "We Fought Cancer...And Cancer Won".

Cancer won? That would be news to my physician friends at the Washington University Medical School, who treat problems ranging from childhood brain cancer to adult leukemia. Their jobs are satisfying, they tell me, because today more cancers are treatable, and even curable, than ever before. Since 1977, the 5-year US cancer survival rate (covering all cancers, both genders, and all races) have gone from less than 50% to more than 65%, and in specific cancers the success rate has increased even more dramatically.

Data from the National Cancer Institute SEER Cancer Statistics Review

And yet, because we haven't "cured cancer" researchers continue to get slammed by news pieces that confuse the role of basic science and clinical treatments. The hope in the 1970's was that basic research would quickly lead to cures, but that hope was misplaced, in part because the basic research necessary was not in place. In contrast with the World War II Manhattan project to build the first atomic bomb, which succeeded because the basics of nuclear physics had already been worked out beforehand, War on Cancer advocates were making promises about treatments based on science that had not yet been done. For decades, we didn't know the molecular players involved, and it hasn't been until the recent, genome-era advances that we've had a real hope of identifying all of the important cancer-related genes and understanding how they work together.

One major problem was that scientists severely underestimated the complexity of cancer. While it's true that all cancers are diseases of uncontrolled cell division, the molecular details are very different for each cancer type. Newsweek scoffs at this idea, but it's a fact: cancer is really many diseases, and there will never be a single breakthrough cure applicable to all cancers. There is no magic bullet. As one scientist in the Newsweek piece put it, you don't need to wait for the home run - "sometimes you get runs by hitting singles and doubles."

Because of the lack of home runs, Newsweek is suggesting that we've spent a whole lot of money to get very little:

"Indeed, it is possible (and common) for cancer researchers to achieve extraordinary acclaim and success, measured by grants, awards, professorships and papers in leading journals, without ever helping a single patient gain a single extra day of life. There is no pressure within science to make that happen. It is no coincidence that the ratio of useful therapy per basic discovery is abysmal. For other diseases, about 20 percent of new compounds arising from basic biological discoveries are eventually approved as new drugs by the FDA. For cancer, only 8 percent are."

This is an incredibly misplaced criticism which demonstrates a complete misunderstanding of how science works, and how it is related to medicine. Most researchers don't treat patients, because treating patients does not advance basic research. It should be obvious that to understand the molecular details of cancer we need people who are trained to study molecular biology, not people who diagnose patients. I am funded by the American Cancer Society, and I haven't helped "a single patient gain a single extra day of life." But will my research help treat or cure cancer? Not in the next 5 years. But if it's successful it will: it will help us model how the control machinery goes awry in cancer cells, and thus design better, multi-pronged strategies to shut down those cells. I don't see how you can design comprehensive, rational strategies to tackle cancer without such knowledge, knowledge which has only become possible within the last few years.

The other ridiculous notion put forward by Newsweek is that we have some quantified, optimal "ratio of useful therapy per basic discovery", and that the ratio we're getting is not what we should be getting. How do you measure such a ratio? Not by looking at FDA approved "new compounds", because most basic biological research isn't aimed at producing new compounds. Biotech companies, drug companies, and some university labs are screening libraries of chemicals to find new drugs, but this work is not about "basic discovery", and is thus a poor metric of how fruitful basic research really is. It is hard to find good new cancer drugs, in part because so many candidates are extremely toxic. Our current success rates are not optimal because people are still dying, but you can't say that we're not getting enough for our money without some objective way of determining how much we should be getting for each dollar spent.

The implication made here by Newsweek is that basic research is either a wasteful way to achieve cancer cures, or basic researchers aren't choosing the right problems to work on. It's unlikely however that cancer treatments will get better over the long-term if we start spending basic research money elsewhere - basic research is the only rational strategy we have; the alternative is sheer trial and error.

The key to making use of our rapidly growing molecular understanding of cancer is to design a coherent, multi-pronged chemical strategy. Previously we only had the carpet-bombing techniques of radiation and non-specific chemotherapy cocktails. Recently, new-generation cancer drugs such as Tarceva and Gleevec have become available, which target individual proteins known to play a role in cancer. For some cancers these drugs have worked quite well. But as Newsweek points out, "cancer cells are like brilliant military tacticians: when their original route to proliferation and invasion is blocked, they switch to an alternate, marching cruelly through the body without resistance."

This means that an effective drug treatment for many cancers will have to be based on a mixed strategy, such a drug regimen targeting not just one protein, but a combination of molecular players. Strategies like this can only be rationally designed when we actually understand how those molecular players work together. Basic researchers have been making steady progress towards that goal, but as I noted above, it is only within the last few years that genomics and computational biology have made such an understanding possible. And this understanding is not only relevant to cancer - the lessons we learn will be relevant to most disease.

The basic research will get there. It needs to be funded but not forced into preconceived notions of what success will look like. It's true that basic research can be frustratingly slow. The science of nuclear physics was worked on for decades before the first reactor was ever built, and cancer is more complex than a nuclear reactor.

In the mean time, it's clear that basic research isn't the only solution. Epidemiology has scored some big successes by linking smoking and hormone-replacement therapy to cancer. We can continue to significantly improve cancer prevention by identifying environmental factors, but in the end there will still be a large subset of cancers that are only weakly linked to environmental factors. Those cancers will not be solved by campaigns to quit smoking or use sunscreen.

Is our basic research enterprise wastefully working in the wrong direction? Have scientists missed critical ideas that a Newsweek journalist has discovered? Are we just coming up with more molecular details that will "join the long list of those that are interesting but irrelevant to patients"?

Don't count on it. While we should keep in mind exaggerated promises made by scientists 35 years ago, we need to also see that those promises were made long before any of the basic science was in place. Edison wasn't about to produce a kitchen LED, and no scientist in 1970 had a clue about the complex molecular machinery underlying cancer. But we eventually did build LEDs, and we now know an amazing amount about the genetics of cancer. Cancer survival rates are up. While there may never be an Armistice Day in the War on Cancer, the progress is real, and basic research, while not hitting home runs, can still win the game with singles and doubles.