Correspondence
Winning the War on Cancer
N Engl J Med 1997; 337:935-938September 25, 1997
- Article
To the Editor:
In their Special Article, provocatively entitled “Cancer Undefeated,” Bailar and Gornik (May 29 issue)1 acknowledge the substantial changes in mortality due to cancer during the past 20 years in men and women under the age of 55 years. The authors note that for this age group, there has been a 25 percent decrease in all cancer-related deaths, a 23 percent reduction in deaths from breast cancer, and a slight but definite decline in mortality from lung cancer; for persons of all ages, there has been approximately a 15 percent drop in deaths from colorectal cancer. Bailar and Gornik believe these changes reflect changes in the incidence of cancer or early detection, discount entirely the impact of therapeutic intervention, and argue that progress will occur only through a national commitment to prevention.
Nobody disputes the merits of cancer prevention. By now, eliminating the use of tobacco products — particularly among the young — has become a social and legislative issue; sufficient research has already been performed to justify the needed behavioral changes. Reducing exposure to known carcinogens such as ultraviolet light, hepatitis B and hepatitis C viruses, asbestos, and excess ethanol has received widespread attention, as has the importance of screening for breast, cervical, and colorectal cancers. The recent development of germ-line genetic-testing techniques will probably identify people at very high risk for breast, colorectal, and ovarian cancers in whom prophylactic medical or surgical interventions, or both, may be of value. A major component of the National Cancer Institute's budget is for cancer prevention, and in 1996, a distinguished panel of experts in this area was commissioned by the institute's director, Dr. Richard D. Klausner, to provide an external critique of this effort.
Bailar and Gornik reveal their underlying bias by choosing to ignore the influence of treatment on the reduction in cancer-related mortality among persons under the age of 55 years. During the past 25 years, previously fatal conditions, such as advanced testicular cancer,2 Hodgkin's disease,3 and childhood leukemia,4 have become curable in more than 70 percent of cases, and up to 50 percent of patients with non-Hodgkin's lymphomas may now be cured.5 Prospective, randomized trials have shown that postoperative (i.e., adjuvant) therapy leads to a 25 to 30 percent reduction in mortality among patients with locally advanced breast cancer 6 or colorectal cancer.7 Reductions in cancer-related mortality clearly have multifactorial explanations, but for Bailar and Gornik to dismiss widely used, well-accepted advances in treatment is not only absurd but also potentially damaging to patients with newly diagnosed malignant conditions, who may be influenced by the publicity surrounding this extreme view to reject life-saving treatment. . . .
7 ReferencesRobert J. Mayer, M.D.
Lowell E. Schnipper, M.D.
American Society of Clinical Oncology, Alexandria, VA 223141
Bailar JC III ,Gornik HL . Cancer undefeated. N Engl J Med 1997;336:1569-1574
Full Text | Web of Science | Medline2
Williams SD ,Birch R ,Einhorn LH ,Irwin L ,Greco FA ,Loehrer PJ . Treatment of disseminated germ-cell tumors with cisplatin, bleomycin, and either vinblastine or etoposide. N Engl J Med 1987;316:1435-1440
Full Text | Web of Science | Medline3
Canellos GP ,Anderson JR ,Propert KJ , et al. Chemotherapy of advanced Hodgkin's disease with MOPP, ABVD, or MOPP alternating with ABVD. N Engl J Med 1992;327:1478-1484
Full Text | Web of Science | Medline4
Pui C-H . Childhood leukemias. N Engl J Med 1995;332:1618-1630
Full Text | Web of Science | Medline5
Fisher RI ,Gaynor ER ,Dahlberg S , et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin's lymphoma. N Engl J Med 1993;328:1002-1006
Full Text | Web of Science | Medline6
Bonadonna G ,Valagussa P ,Moliterni A ,Zambetti M ,Brambilla C . Adjuvant cyclophosphamide, methotrexate, and fluorouracil in node-positive breast cancer -- the results of 20 years of follow-up. N Engl J Med 1995;332:901-906
Full Text | Web of Science | Medline7
Moertel CG ,Fleming TR ,Macdonald JS , et al. Fluorouracil plus levamisole as effective adjuvant therapy after resection of stage III colon carcinoma: a final report. Ann Intern Med 1995;122:321-326
Web of Science | Medline
To the Editor:
Bailar and Gornik express gratitude to me “for kindly suggesting” the title of their article, “Cancer Undefeated.” I did not suggest this title, and I respectfully decline the acknowledgment. I did, as part of a dialogue with Dr. Bailar, refer him to a 1960 article by Sir John Crofton, entitled “Tuberculosis Undefeated.”1 This article offers many parallels for our discussions about cancer today.
1 ReferencesSamuel Broder, M.D.
Miami, FL 33137To the Editor:
The results of treatments for cancer may be better than Bailar and Gornik suggest. Any improvement in survival will increase the age at death. The number of deaths below any given age will therefore fall, and the number above it will rise. This will cause a divergence between the mortality rates for the old and those for the young. Indeed, Figure 2 and Figure 3 in the article by Bailar and Gornik show a marked divergence in the rates at the age of 55 years — probable evidence of steadily increasing survival.
T.H. Hughes-Davies, F.R.C.P.
Breamore Marsh, Fordingbridge, Hampshire SP6 2EJ, United KingdomTo the Editor:
Although Bailar and Gornik clearly demonstrate the failure of current treatment efforts in many areas, I fail to see how their article supports the conclusion that more money should be spent on prevention, since prevention has not been very effective either. . . .
Tom S. Rand, M.D.
Wilson Orthopaedic Surgery and Neurology Center, Wilson, NC 27895-3148To the Editor:
Bailar and Gornik report that the age-adjusted rate of mortality from all cancers in the United States declined by 1 percent from 1991 through 1994. Our estimate for the same interval1 was 2.2 percent, and we also reported a 3.9 percent decline during the period from 1990 to 1995. The discrepancy in the data for the 1991–1994 period stems from the use of different populations for age adjustment. Bailar and Gornik used the relatively elderly 1990 U.S. population and by doing so, minimized striking reductions in mortality that occurred among young and middle-aged persons. We used the U.S. “standard million” population, the basis for all national reports. Use of this population, which is essentially the relatively young 1940 population, reveals the full downturn in cancer-related mortality. We used the standard million not for impact but of necessity to describe a current trend; the latest data are available only in this form. Thus, our 1996 report includes findings for 1995, whereas the report by Bailar and Gornik is limited to 1994. Data for the period from January to October 1996 2 show a further 0.7 percent reduction in mortality from cancer, bringing the decline for the period from 1990 to 1996 to 4.6 percent.
Our more important difference with Bailar and Gornik concerns their view that improvements in treatment resulted in little reduction in mortality from cancer. We reported that one half of the decline we observed reflected advances in medical care and access to it. This statement was based on data showing long-term gains in the survival of patients with cancer even after a correction had been made for the effect of earlier diagnosis.
There are three more reasons for our opinion. First, many aspects of the diagnosis and treatment of cancer have improved greatly, as Bailar and Gornik acknowledge. Second, virtually all oncologists believe that cures and long-term palliation of cancer are much more common now than previously. Finally, several national trends seem explicable only in terms of treatment gains. For example, the mortality rate for all cancers except lung cancer has declined since the mid-1970s, whereas the incidence has remained the same or increased.3
3 ReferencesPhilip Cole, M.D.
Brad Rodu, D.D.S.
University of Alabama at Birmingham, Birmingham, AL 352941
Cole P ,Rodu B . Declining cancer mortality in the United States. Cancer 1996;78:2045-2048
CrossRef | Web of Science | Medline2
National Center for Health Statistics. Births, marriages, divorces, and deaths for November 1996. Mon Vital Stat Rep 1997;45(11).
3
Devesa SS ,Blot WJ ,Stone BJ ,Miller BA ,Tarone RE ,Fraumeni JF Jr . Recent cancer trends in the United States. J Natl Cancer Inst 1995;87:175-182
CrossRef | Web of Science | Medline
To the Editor:
. . . The cavalier attitude of Bailar and Gornik toward the remarkable reduction of deaths due to childhood cancer is wrong. Although the numbers of cured children may be small, each child's life affects many people — the family, the school, the community, and the parent's workplace. Moreover, without a cost–benefit analysis of curative childhood cancer effects, there can be no complete evaluation of the “war on cancer.” The number of deaths from cancer is just one outcome to be reckoned with.
Seth Corey, M.D., M.P.H.
Children's Hospital of Pittsburgh, Pittsburgh, PA 15213To the Editor:
An analysis of the effect of the U.S. research effort on cancer that is based entirely on mortality rates, with no consideration of incidence rates, is flawed. Bailar and Gornik conclude that research funds should be diverted from treatment to prevention. They provide no analysis of the effectiveness of preventive strategies and summarily dismiss gains from nonpreventive approaches. Data from the Surveillance, Epidemiology, and End Results (SEER) study show that from 1973 to 1993, the increases in mortality and incidence rates for cancer were 5.8 and 27.3 percent, respectively.1 Mortality rates decreased in 14 of the 23 cancer sites assessed. For six of the remaining nine sites, the increase in mortality was smaller than the increase in the incidence of cancer.
There are many points of attack in the effort to defeat cancer, and we have seen valuable gains in the quality of life and knowledge of cancer biology, as well as reductions in mortality. For example, there have been major advances in preserving anatomy or function in treating cancers of the eye (uveal melanoma), esophagus, breast, larynx, anus and rectum, extremities, and prostate.
Significant improvement in survival has been demonstrated in recent phase 3 clinical trials for cancer of the testis, breast, rectum, colon, and esophagus, as well as osteogenic sarcoma and cancers in children. These improvements are largely due to the use of multidisciplinary treatment strategies (combinations of surgery, chemotherapy, and radiation therapy).
Finally, a remarkably rapid increase in our knowledge of cancer biology at the most basic level has occurred since 1970. Cancer is now known to be a genetic disease. We have gained great insight into the multistep process of cancer through research on tumor-suppressor genes, oncogenes, programmed cell death, DNA repair, angiogenesis, and the process of metastasis. These diverse research successes will make it possible for oncologists to begin using molecular diagnostics, individualizing management strategies, and planning gene therapy.
We support research on prevention, but it should not be undertaken at the expense of early detection and treatment.
1 ReferencesHerman Suit, M.D.
Kurt Isselbacher, M.D.
Bruce Chabner, M.D.
Massachusetts General Hospital, Boston, MA 02114To the Editor:
As a practicing medical oncologist, I agree with Bailar and Gornik that the progress we have made in the treatment of cancer over the past number of years is disappointingly small. However, as both an oncologist and a patient with cancer, I vehemently disagree with the widely publicized opinion of the authors that “in an age of limited resources this may well mean curtailing efforts focused on therapy.” Why? Are we putting up a white flag?
War is hell, including the war against the diseases called cancer. In war, progress may not be evident immediately. If the cause is just, one does not quit because of a few lost battles. Some wars last six days, others a hundred years. The war against cancer has been fought for a relatively short time and only very recently with the most modern laboratory techniques. Science is by nature a slow process with an occasional breakthrough. Twenty-five or 30 years is too short a period for a declaration of failure against such a difficult foe.
Alan Feldman, M.D.
Coney Island Hospital, Brooklyn, NY 11235To the Editor:
Bailar and Gornik state, “35 years of intense effort focused largely on improving treatment must be judged a qualified failure,” and they believe the emphasis should therefore be shifted toward a preventive approach. In truth, the effort has been far less than intense. As Donald Coffey, president of the American Association for Cancer Research, has noted, “A real war against cancer has never been mounted. To date, available federal funds have supported only a small, intense skirmish by a limited number of investigators.”
Although political leaders pay lip service to stopping a disease that will attack one of every four Americans alive today, the fact is that the government's commitment has not changed substantially. During the past 10 years, federal funding for research on cancer, adjusted for inflation, has increased by just 1 percent. Today, research on cancer represents just 0.1 percent of the federal budget. . . .
Richard N. Atkins, M.D.
Philip A. Arlen
Christopher L. Noxon
CaP CURE, Santa Monica, CA 90401To the Editor:
Cancer is “undefeated”; the “war against cancer” has not been won. These are military allusions, initially used by a political figure and too readily adopted as snappy media language by the medical and scientific communities in the United States and elsewhere.
In wars there are the victors and the vanquished (not always easily distinguished), and collateral damage is all too common. Wars delay and obfuscate problems but do not often solve them. Wars encourage simplistic and jingoistic attitudes — us versus them. But cancer is so much more complex than this. There is no invading army, no call to arms, no enemy — the trouble is within.
It is time to redefine the problem. Although we should continue our exploration of the biology of cancer, trials of new therapies, and population-based preventive strategies, we also need to face the inevitability of cancer. Cancer is the price we pay for being sophisticated organisms, and there are only so many times we can faithfully replicate the genome with each cell division before making a critical mistake. In addition, the rising incidence of some cancers needs to be seen in context: overall life expectancy in the Western world continues to increase. Although better prevention and early detection should reduce mortality, metastatic cancer will develop in many people and is likely to remain largely incurable. For these people, the emphasis should be on living with cancer rather than dying in battle.
The war-on-cancer metaphor distracts attention from the complexity of the disease and inevitably identifies winners and losers. We should tell the world that we are working at understanding cancer and that knowledge is power.
Nicholas Wilcken, M.D., Ph.D.
Westmead Hospital, Sydney, NSW, 2145, AustraliaAuthor/Editor Response
The authors reply:
To the Editor: Several letters and the Sounding Board article by Kramer and Klausner in this issue of the Journal 1 distort our position and divert attention from the critical issues. None of the authors, however, question the finding that cancer-related mortality is higher now than at the time of the National Cancer Act of 1971, even after adjustment for aging in the population and declines in other lethal diseases.
Kramer and Klausner1 charge that we extrapolate the future from the past. A very long history of great effort by great scientists, marked by great ballyhoo and very spotty progress, should engender some skepticism about today's claims of wonderful things to come. We acknowledged that there are successes, but not enough, in palliation and treatment for childhood cancers and some adult cancers. Our argument is that new efforts should be made to advance cancer prevention, already shown to be fruitful, and the examples Kramer and Klausner cite actually support our conclusion. Fleming's discovery of penicillin was a product of acute observation, not basic science; iron lungs disappeared because of prevention, not treatment; tamoxifen is indeed useful in treatment but also has potential for prevention; the delay in federal efforts to reduce smoking (prevention again) was due to political pressures, not lack of knowledge or will at the National Cancer Institute; and the discovery of the roles of human papillomavirus, Helicobacter pylori, and nicotine addiction reinforces the need for greater attention to cancer prevention.
Kramer and Klausner ask whether cancer is sufficiently homogeneous to emphasize a single path. We do not claim so, but note that some preventive approaches, such as chemoprevention and strengthening of internal defenses, may have a broader spectrum of benefit than specific treatment regimens. We are not content with a 0.6 percent decrease per year in cancer-related mortality, since at that rate, it would take 115 years for mortality to decline to half the present level. Kramer and Klausner also dispute our contention that the present program is lopsided but fail to mention that prevention and control accounted for about 6 percent of the National Cancer Institute's budget from 1973 (the first year the budget was presented in the current form) until 1994 and that the recent expansion to 10 percent was at the direction of Congress. Furthermore, some of that money is for improved screening and treatment, not prevention.
Mayer and Schnipper say we ignore the influence of treatment on cancer-related mortality in people under the age of 55 years. This is not so; Figure 2 of our article shows trends for people 55 years or older and for those younger than 55, and we specifically mention improved treatment for Hodgkin's disease and childhood neoplasms. Whether adjuvant therapy for breast and colorectal cancer will have effects demonstrable at the population level is not yet known.
The point raised by Hughes-Davies applies to trends in crude rates. We presented only age-adjusted rates to avoid such problems.
We refer Rand to the substantial decline in tobacco use among adults; the effective control of asbestos, benzene, and many other industrial carcinogens; reductions in radiation doses per exposure; and the dietary changes adopted by increasing numbers of Americans — all initiated with little support from basic-science investigators or the government. We need to know how much more we could achieve with a vigorous program of prevention encompassing research and practice.
If cancer-related mortality rates for people of different ages were moving in parallel, the choice of a standard for age adjustment would make little difference, but the rates are not parallel. Declines are greatest at the youngest ages, and increases are greatest at the oldest ages, with a gradual change between these extremes and a crossover from declines to increases at about the age of 55 years. We chose the 1990 standard as the midpoint of the critical recent period; the National Cancer Institute chose the 1970 standard, with somewhat more favorable results; and Cole and Rodu prefer the even more favorable findings with the 1940 standard. If we had used a medieval population, with half the population under the age of 6 years and almost nobody over the age of 50, the trend would have looked wonderful. But only the 1990 standard is appropriate for comparisons of U.S. trends over a period centered on the year 1990.
In response to Cole and Rodu and to Suit et al.: we gave good reasons for not using incidence rates or case survival rates. Furthermore, the argument that better treatment is balancing the rapid increase in incidence supports our conclusion that prevention — reversing the increases in incidence — is crucial.
1 ReferencesJohn C. Bailar, III, M.D., Ph.D.
Heather L. Gornik, M.H.S.
University of Chicago, Chicago, IL 60637-14701
Kramer BS ,Klausner RD . Grappling with cancer -- defeatism versus the reality of progress. N Engl J Med 1997;337:931-934
Full Text | Web of Science | Medline
