One could carp about several shortcomings in Clifton Leaf's book about cancer - eg. he gets himself confused on determining appropriate statistics to use measuring progress in the War on Cancer, too many pages are devoted to anecdotal accountings, and the book's flow can be confusing. However, the more I read this book, the more I realized how complex cancer and the War on Cancer are, and that Leaf has done a marvelous job of portraying both, as well as making a compelling case that drastic revisions are required to make substantial improvements in this 44-year effort. Meanwhile, thousands of physicians and scientists conduct a mostly uncoordinated campaign (eg. 2,786 IRS recognized cancer-focused charities in 2010, plus universities, drug companies, and many of the 50 states) that has mostly succeeded only in finding tiny improvements instead of breakthroughs (overall cancer death rates in 2004 compared with 1990 in men and 1991 in women decreased by 18.4% - 80% from lung, prostate, and colorectal, and 10.5% - 60% from breast and colorectal, respectively), rewards academic achievement over everything else (published about 2 million papers), emphasizes models that have consistently proven poor predictors ('instant tumors' researchers cause in mice can't mimic human cancer's most critical trait - quick-changing DNA; tumor shrinkage is a poor predictor of cancer outcomes), and is rife with redundant work.
Leaf's digging into the topic was first demonstrated 2004 in a cover-page Fortune article - 'Why We're Losing the War on Cancer - and How to Win It'. At that time, we'd already spend close to $200 billion (now $300 billion), in inflation-adjusted dollars. Turns out, little has changed since then and the percentage of Americans dying from cancer is about the same as in 1970 and even 1950. (Age-adjusted death rates have improved slightly, but the bulge of aging baby-boomers has more than made up for that.) And most of this modest gain comes not from all the publically-funded research, but from quitting smoking and increased colonoscopies and pap smear detection that identifies pre-cancerous cells that can be relatively easily removed.
Meanwhile, optimistic data are reported regularly about improved survival rates. Turns out that most of the 'improvement' is due to earlier detection, most of the life extensions attributed to new drugs are trifling, most announcements about new drugs approved for cancer treatment arise from simple extensions of FDA approval for existing drugs in added situations, and 75% of FDA approvals are not for prolong life but temporary tumor shrinkage.
So why isn't tumor shrinkage a big accomplishment? Localized tumors are not what kill most people with cancer - its metastasis (90%), in which aggressive cells spread around the body. Yet, less than 0.5% of study proposals focused primarily on metastasis.
There is some good news - involving childhood leukemia and other childhood malignancies. The bad news, however, is that the NCI says the incidence of childhood cancers has risen 0.6%/year from 1975 to 2006 and nobody knows why. Pediatric cancer survivors also have a 6X greater chance of a second malignancy vs. those w/o the first case.
Gleevec is another bright spot, for those with chronic myeloid leukemia (CML). Unfortunately, many have generalized its success as evidence that more 'magic bullets' can be found. Leaf points out that CML involves a single gene; most cancers, especially solid tumors, are heterogeneous. Not only are any two eg. breast cancer patients liable to have different mutational signatures, they're also likely to have widely different mutations in different cells w/I the same tumor or other tumors in their bodies. Another problem is that new mutations arise as a result of treatment, and then create subsequent relapse.
Breast cancer death rates began falling in 1990, falling nearly 25% in the next ten years. (Death rates from heart disease and stroke have declined about half.) Careful analysis, however, concluded drug treatment was responsible for a bit more than half the drop, and tamoxifen (developed decades ago prior to the War on Cancer, and more recently approved for more breast cancer situations) that major contributor to that gain. The rest came from earlier detection - yet, 30% of early-stage breast cancers return, usually in other locations, though allowing the women involved to get past the initial five-year survival rate. In a 2008-published study reporting on 3,000 early- and mid-stage breast cancer patients who had completed therapy at one of the world's top cancer research hospitals and assessed cancer-free at the five-year mark, within the following ten years 20% had a recurrent lesion.
More substantive good news lies in the impact of increased Pap smears (cut the cervical cancer death rate by 80%) and colonoscopies (reduced death rates there more than any drug or therapy). Since the origin of cancer, per Michael Sporn, is an uncertain and slow-motion process, 'pre-treatment' such as done for diabetes and heart disease (lower LDL, blood pressure and blood-sugar levels, could be much more effective against cancer.) Yet, even today, half of all cervical cancers in the U.S. and an even greater share of those in the colon or rectum are not discovered until relatively advanced. And per a 2008 NCI 'progress report,' there are no validated molecular biomarker tests for the early detection of any cancer' - despite 150,000 published studies of various candidates.
The USPHS grant application guide is 260 pages long; completing the review/award process takes about a year, discourages innovation, and has brought innumerable repeated studies in numerous areas (eg. 65,000 papers on the most famous cancer-related gene - p53, since its 1979 discovery, and tens of thousands of other studies for each of a number of others). The result - researchers can spend half their time simply filling out research grants; absent their doing so and succeeding, they risk unemployment.
What to do? Leaf recommends easing the process of applying for grants - this would include reducing the layers of review and the number of funding mechanisms within the Public Health Service. Secondly, he'd like to have greater encouragement for collaboration, a more permissive environment for breakthrough ideas, and standardization of various measures of patient status and outcomes. Additionally, Leaf is supportive of greater publicity for the fact that the cancer burden is growing (the accumulation of incremental improvements notwithstanding) because of our aging population, and that its usual heterogeneity precludes simple solutions, as well as improved emphasis on attacking it earlier as is done with heart disease, stroke, diabetes, etc.