Choice in Conventional Cancer Therapies
The Debate over Conventional Cancer Therapies
There is an important difference between a conventional cancer therapy that has proven efficacy in achieving cures or significant life extension at acceptable cost in terms of quality of life and a conventional therapy that does not have such a clearly superior track record. This difference, as we saw in chapter 3, is well recognized in England, where much of medical opinion is on the side of more conservative, less aggressive, and less toxic treatments. Yet even though American cancer medicine is far more aggressive, an intense debate over the efficacy of mainstream cancer treatments continues within the United States. This chapter is about that debate.
The American cancer establishment today is seriously divided between those who believe that aggressive therapies are overutilized without adequate scientific evidence and those who believe that aggressive treatments have begun to yield superior results which will be upheld by future studies, even if they have not been demonstrated effective so far.
In November 1985, an eminent Harvard researcher, John Cairns, published an article in Scientific American entitled “The Treatment of Diseases and the War Against Cancer.” His article signaled an intensification of the ongoing debate over the efficacy of conventional cancer therapies. Cairns described how, since World War II, cancer registries have been set up in a number of American states and several nations that chart changing trends in cancer incidence and mortality. “These registries yield a rather precise picture of the natural history of cancer, and that is a necessary starting point for any discussion of treatment.”
A group of patients can be considered cured of their cancers if they die at about the same rate as the general population, which they would if, thanks to their treatment, they had been returned to the common pool. …. The survival rate of Norwegian women who have colon cancer, for instance, has been compared with the survival rate of the general population of women. …. Most of the patients die rather soon after diagnosis, but a sizeable minority, about 30%, die at the same rate as the general population (that is, behave as if they had been cured). …. About a third of all Norwegian cancer patients suffered no loss of life span as a result of their disease.1
The Norwegian statistics, Cairns noted, came from the 1950s and 1960s: “We are looking at the results of treatment by surgery, occasionally backed up by X-irradiation when the primary tumor was inaccessible to surgery. It is the picture of what used to happen before the advent of screening programs, chemotherapy and numerous clinical trials.” He notes that “the major ancillary aids to surgery, such as blood transfusions, antibiotics, and improved forms of anesthesia” had already been developed and disseminated. The deciding issue, for nearly every cancer patient, then, was “the extent of spread of the cancer at the time of surgery.”2
Cairns explores the advantage that early screening programs had brought for cancer prevention. He found different results for different cancers. For breast cancer, he found that about a fourth of total mortality could be prevented if all women over 50 were offered a free breast examination every 1 to 3 years. For cervical cancer, while the Pap (Papanicolaou) smear, as Cairns explains, has never been fairly evaluated since it was introduced in the United States at a time when cervical cancer had begun its marked decline–“presumably because during this period the average levels of hygiene, affluence, and education have gone up”–available population comparisons indicate that “the decline in mortality from cervical cancer invariably accelerated at the time testing became widespread,” although the causal connection is disputed by many authorities. For lung cancer, on the other hand, a large-scale trial of its early diagnosis “indicated that no great benefit comes from having the disease detected by chest X-rays before it has produced any symptoms. …. To summarize, screening programs for earlier diagnosis sometimes bring benefits and sometimes do not.”3
Cairns then turns to the subject of adjuvant therapies–hormonal therapies, x-irradiation, and chemotherapy: “It remains a depressing truth that fewer than 50 percent of cancer patients can be cured by surgery. A tremendous effort has therefore gone into discovering adjuvant forms of treatment that can be given following surgery.”4 For a number of cancers, he concludes, impressive gains have been made in the use of adjuvant therapies. Hormonal treatment has been beneficial for some cancers of the breast and prostate. And, according to the National Cancer Institute (NCI), radiation therapy has been particularly useful in the treatment of Hodgkin’s disease, for which mortality has fallen 61% between 1950 and 1985, accompanied by a doubling of the 5-year survival rate. New chemotherapeutic techniques hold out the promise of even further improvements in the treatment of Hodgkin’s disease in coming years.5 Radiation treatment has also brought about improved survival in cervical cancer and in one kind of testicular cancer, and in combination with other modalities it has become one option in breast cancer.
Advances in chemotherapy have contributed to gains in overall survival rates for leukemia, from 10% for those diagnosed in the period from 1950 to 1954, to 32% for patients diagnosed between 1979 and 1984. Among the various leukemias, patients with acute lymphocytic leukemias experienced the most dramatic increase in 5-year survival, from virtually zero for those diagnosed in the period from 1950 to 1954 to a high of 54% for those diagnosed in 1977 and 1978. Five-year survival rates among children with acute lymphocytic leukemia similarly jumped 73% between 1950 and 1978. Slight declines in the survival rates in subsequent years were attributed by the NCI to the use of less aggressive therapies after the late 1970s.6
Most of the gains achieved by adjuvant therapies, according to Cairns, have come with cancers normally occurring in children or young adults:
With suitable combinations of chemotherapy it is now possible to cure many kinds of childhood cancer that would otherwise be rapidly fatal …. The reduction in the annual mortality of older children and young adults has been less spectacular, with the following notable exceptions. Hodgkin’s disease used to be inevitably fatal, but now most patients can be cured. …. About 35 percent of testicular cancers were fatal before chemotherapy, but now roughly a third of these deaths can be prevented. …. Finally, choriocarcinoma, a rare cancer of the placenta …. can now be cured by chemotherapy. [Most oncologists would add non-Hodgkin’s lymphoma to this list of potentially chemotherapy-curable malignancies.]
Despite this improvement, Cairns nevertheless concludes that overall, “the gains [from the use of chemotherapy] have been limited.” Under the age of 30, the gains have been substantial: “The latest figures for the U.S. show about 7,000 deaths per year from cancer under the age of 30, compared with the 10,000 we would have expected if the death rate had remained unchanged since the 1950s.”7
However, and this is a key point, Cairns points out that only 2% of cancer patients in the United States are under 30. “For the vast majority of cancers, which arise in older patients, the results of chemotherapy are much more controversial. …. Apart from the success with Hodgkin’s disease, childhood leukemia and a few other cancers, it is not possible to detect any sudden change in death rates for any of the major cancers that could be credited to chemotherapy [emphasis added].”
Those who organize cancer centers and supervise the many clinical trials of chemotherapy look for ways to circumvent these relentless statistics. Sometimes they explain away the unchanging statistics for mortality by pointing out that the national statistics are inevitably a few years behind the times and therefore do not reflect the most recent advances in treatment. Although this point is absolutely correct, it has been made repeatedly in the past 10 years but has never been vindicated by national statistics when these eventually became available. For the most part, however, the organizers disregard the figures for mortality and simply point out that the fraction of patients who are alive five years after diagnosis has been steadily increasing for nearly every kind of cancer. They attribute this increase in five-year survival to steady improvements in methods of treatment.8
Cairns explores the origins of this increase in 5-year survival, and argues that it is due to more accurate classification and not to actual treatment. In prostate cancer, for example, it turns out that a quarter of all American men who die over age 70 show, on postmortem examination, evidence of small prostate cancers, but only 10% of these cancers ever produce symptoms and a still smaller percent prove fatal. The increase in 5-year survival in prostate cancer, he finds, appears to be due largely to intensified surveillance. “The survival rate has therefore increased not because fewer men are dying from prostate cancer but because more men are being classified as having prostate cancer …. Similar artifacts probably affect the survival rates for many other types of cancer, particularly cancer of the breast.”9
“The role of chemotherapy in the treatment of the other major cancers of adults is much less well documented,” Cairns says. Ovarian cancers sometimes respond to chemotherapy. Chemotherapy and irradiation can shrink cancers at inaccessible sites, such as certain areas of the head and neck. “Overall, however, in terms of duration of survival, the results have been more often negative than positive.” Then Cairns renders his considered, powerful, and unforgettable judgment:
In spite of these rather sobering findings several cytotoxic drugs are now commonly employed. The Connecticut Cancer Registry, for instance, reports that about a fourth of all cancer patients are recorded as having some form of chemotherapy during their initial stay in the hospital. The National Cancer Institute estimates that more than 20,000 patients receive chemotherapy in the U.S. each year. For a dangerous and technologically exacting form of treatment these are disturbing figures, particularly since the benefit for most categories of patients has yet to be established. Furthermore, the number of patients who are cured can hardly amount to more than a few percent of those who are treated [emphasis added].10
In the end, Cairns holds to his judgment. He estimates that adjuvant treatments may avert “perhaps 2% or 3% of the 400,000 deaths from cancer that occur each year in the U.S.” He then puts his conclusion in a broader context:
These are very real gains and a fitting memorial to the many thousands of patients who took part in the early trials of chemotherapy. The fortitude and altruism of these patients have not, however, been matched by any comparable sense of responsibility on the part of those who determine national policies. By the 1960s, cigarette smoking had been established as the major cause of lung cancer …. Unfortunately, there are huge financial incentives for nations to sit back and do nothing. The cigarette is a readily taxable commodity; in the U.S. it provides the Federal government with about $6 billion a year. More important (at least for the British government, and possibly also in the eyes of the U.S. Government), smoking cuts down the bill for old age benefits because it reduces life span. At the price of a slight increase in costs for health care, the current smokers in the U.S. on the average each have saved the U.S. Government about $35,000 in Social Security payments …. The loss of life span represents a total saving of some $10 billion a year over the next half century or so. Some countries have banned all tobacco advertising, and this has had an almost instant effect on tobacco sales. The failure of the U.S. Government to take such a step far outweighs all the advances made in the treatment of cancer since the advent of modern surgery [emphasis added].11
In conclusion, Cairns notes, “It seems bad cost-accounting for the Federal Government to subsidize chemotherapy through research and not to subsidize the screening of women for breast cancer. Worse, it is surely an act of folly to pour hundreds of millions of dollars every year into giving a growing number of patients chemotherapy while doing virtually nothing to protect the population from cigarettes.”12
The Debate Intensifies
One would have thought that such an extensive and careful critique of cancer therapies published in Scientific American would receive some substantial response. That it did not says a great deal about the politics of medical and scientific journals. It was not until May 8, 1986, that a remarkably similar critique, this time published in the New England Journal of Medicine, initiated a major public debate. The article, “Progress Against Cancer?,” was written by Cairn’s Harvard colleagues John C. Bailar III and Elaine M. Smith.
There is “no evidence that some 35 years of intense and growing efforts to improve the treatment of cancer have had much overall effect on the most fundamental measure of clinical outcome–death,” Smith and Bailar wrote. In fact, the age-adjusted annual rate of cancer deaths increased in recent decades from 170 per 100,000 in 1962 to 185 per 100,000 in 1982, an age-adjusted rise of 8.7%. “We are losing the war against cancer, notwithstanding progress against several uncommon forms of the disease, improvements in palliation and extension of productive years of life.” The authors conclude: “Some thirty-five years of intense effort focused largely on improving treatment must be judged a qualified failure.”13
The Bailar and Smith article provoked a tremendous scientific and public debate. Vincent T. DeVita, Jr., then Director of the NCI, criticized the “glaring weakness” of using “age-adjusted mortality as the sole measure of progress.” This, he said, led to the “erroneous view” that the war on cancer was being lost.14 Another critic, Ezra M. Greenspan of Mount Sinai Medical Center in New York, noted that most oncologists treating breast cancer, for example, were still using the mild convenience regimen consisting of cyclophosphamide, methotrexate and fluorouracil (CMF) for premenopausal women and less than 35% employ the more aggressive (Cooper-type) regimen consisting of cyclophosphamide, methotrexate, fluorouracil, vincristine, and prednisone or regimens containing doxorubicin. …. I estimate that 10,000 lives could be saved by the early aggressive use of polychemotherapy in breast cancer, as compared with the negligible number of lives, perhaps several thousand, now being saved. …. Before condemning current treatment as futile one needs to examine the extent to which improved treatments are actually being employed.”15
Lawrence Garfinkel of the American Cancer Society took a different tack: “There’s no doubt that the reason the overall death rate continues to go up is because of lung cancer. If you take away lung cancer, instead of having an 8 percent increase, you have a 13 percent decrease,” he said.16 “That only proves my point,” Bailar responded.
Most every lung cancer death could be avoided if people quit smoking. We know how to prevent lung cancer deaths and we’re not doing enough to prevent them. We should not build a long-range research program on the assumption we’ll find fully effective cures. We gave that a good shot, and now it’s time to get serious about another approach–prevention. If we can convince the American people that their hope lies in preventing the disease [through reducing diet, lifestyle, and environmental risks] we will save more lives than any of these drugs ever will.17
One more important criticism was leveled at the Bailar-Smith analysis by Lester Breslow and William Cumberland writing in the Journal of the American Medical Association:
The problem with reliance on a single measure of progress is that the impression conveyed can vary dramatically when the measure is changed. For example, another view of cancer’s impact over the years, and one that paints a different picture from that given by Bailar and Smith, may be obtained by determining years of potential life lost (YPLL) due to cancer mortality …. The YPLL tend to emphasize reduction in cancer mortality in younger age groups, while the age-adjusted statistic counts a death at age 75 years essentially the same as a death at 5 years. …. How much the picture will change if we use YPLL rather than age adjusted mortality can be seen from the following calculations. In 1980, cancer was responsible for 1.824 million lost years of potential life in the United States to age 65. If, however, the cancer mortality rates of 1950 had prevailed, 2.093 million years of potential life would have been lost …. We are not advocating replacing the age-adjusted mortality with YPLL; rather, our purpose in making these calculations is to indicate how sensitive one’s conclusions are to the choice of measure.18
Public Policy and Individual Decisions
The debate over the Bailar-Smith assertion that the war on cancer was a qualified failure and that the national policy emphasis should turn toward prevention was so important for public policy that U.S. House member Ted Weiss of New York asked the General Accounting Office (GAO) to prepare a report on the extent of progress in cancer treatment for the Congress. The results of the GAO study were summarized in the Journal of the American Medical Association:
The (congressional) General Accounting Office has just presented a similar [to Bailar-Smith’s] dismal view, evidently based largely on Bailar and Smith’s analysis. It said that “Of the three major empirical indicators–incidence, mortality, and survival rates–the only indicator that improved since 1950 was the survival rate.” The General Accounting Office then attributed the reported improvement in survival largely to a statistical artifact [e.g., the fact that cancers were being diagnosed earlier and that the net of diagnosis was being broadened to include more nonfatal or slower-growing cancers].19
This debate indicates how deeply the question of whether or not there has been any meaningful progress in cancer therapy since the 1950s cuts to the heart of the concerns of the mainstream medical community and national policymakers. It is important for cancer patients facing choices about toxic adjuvant therapies to be aware of the intensity of the debate within the medical mainstream.
This much can be concluded. The great victories since the perfection of surgery in the 1950s have been in adjuvant treatment for the cancers of childhood and Hodgkin’s disease, which afflict a very small proportion of the population. For the common cancers of adult years–lung, breast, colon, and prostate cancer–adjuvant therapies have a mixed and more controversial record. Some, such as Cairns, would argue that chemotherapy in particular is widely and disturbingly overused, with great cost in quality of life and little gain in cancer cure. Others argue that more rigorous chemotherapies are being demonstrated to convey additional survival advantage, and that the major problem is that practicing oncologists are not moving quickly enough to provide their patients with the gains to be achieved with more rigorous chemotherapies.
1 John Cairns, “The Treatment of Diseases and the War Against Cancer,” Scientific American 253(5):52-3 (1985). Copyright © 1985 by Scientific American, Inc. All rights reserved.
2 Ibid., 53.
3 Ibid., 54-5.
4 Ibid., 56.
5 United States Department of Health and Human Services, Public Health Service, National Institutes of Health, National Cancer Institute, 1987 Annual Cancer Statistics Reviews (Bethesda, MD, February 1988), II.149.
6 Ibid., 174-5.
7 Cairns, “The Treatment of Diseases and the War Against Cancer,” 57.
10 Ibid., 59.
13 J.C. Bailar III and E.M. Smith, “Progress Against Cancer?” New England Journal of Medicine 314:1226-32 (1986).
14 Vincent T. DeVita, Jr., letter to the editor, New England Journal of Medicine 315(15):964 (1986).
15 Ezra M. Greenspan, letter to the editor, ibid., 963.
16 “Time for New Tactics Against Cancer,” Harvard Magazine July-August, 1986:7-8.
18 Lester Breslow and William G. Cumberland, “Progress and Objectives in Cancer Control,” Journal of the American Medical Association 259(11):1690-1 (1988).