Reining in a Killer Disease

For the past century, since learning that radium treatments can decimate tumors, researchers have accelerated their efforts to cure cancer. A savvy, adaptable, and resilient killer, cancer—in its approximately 200 forms—has persisted despite highly toxic regimens, massive public education programs, and armies of researchers working worldwide. "

By | May 27, 2002

For the past century, since learning that radium treatments can decimate tumors, researchers have accelerated their efforts to cure cancer. A savvy, adaptable, and resilient killer, cancer—in its approximately 200 forms—has persisted despite highly toxic regimens, massive public education programs, and armies of researchers working worldwide. "The history of cancer therapy is that the cells are much smarter than the clinicians, and [they] quickly evolve pathways that can bypass the treatment," says Ray Warrell Jr., CEO of Genta, Berkeley Heights, NJ.

With the help of new technologies, new target discoveries, a deepening understanding of the human genome, and especially the proteome, many researchers now confront cancer with a different mindset. Without losing sight of a cure, scientists are exploring whether it can be treated as a chronic disease, one to subdue and control. This, they say, is a more reachable target.

"In 1990, you rarely heard about [disease management]," says Sudhir Agrawal, president and chief scientific officer at Hybridon in Cambridge, Mass. "But in the middle of the '90s, we started hearing about antibodies and specific drug approaches, like EGFR" [Epidermal Growth Factor Receptor]. An American Association for Cancer Research spokesman says that the disease-management approach became a discussion topic first at AACR's 1998 annual meeting. Then in April, Judah Folkman, of Children's Hospital and Harvard Medical School in Boston, declared that disease management "is what is emerging from" the 2002 AACR conference.

Warrell, who conducted clinical cancer research at New York's Memorial Sloan-Kettering Cancer Center for 22 years before joining Genta, says this concept has two perspectives. First, trying to control cancer is smart, based on current knowledge of cancer and availability of new agents. The second, more jaundiced view is that while this is an admission of failure, it acknowledges the researchers' limited understanding of the disease and the available drugs. "I think that both ways are accurate. Both forces are very much in play," says Warrell.

Some scientists say they understand the logic, but their eyes are still on the prize. Said researcher Robert Schwall of Genentech, South San Francisco, Calif., "I see that, [but] I don't agree. We should cure it. That's the goal for me." Yet he conceded that management is "better than nothing. It's hard to make the thing that kills cancer." Said another researcher, who like Schwall, was presenting posters at the AACR meeting, "you don't look at the target and say can we get a cure or improve? We start with high hopes."

Paul Durda, an assistant biochemist at Massachusetts General Hospital, Boston, is working through his Needham, Mass.-based company, CytoCure, on various compounds that upregulate antigen expression; these antigens are then targeted by T cells. Cytocure's goal is ablation, but control, he says, is acceptable. Durda adds, there is "usually no way to permanently keep it under control."

Folkman, often credited with starting the whole disease-management idea with his angiogenesis work, said that in pursuing the conversion to this approach, scientists do not feel they are lowering their expectations. "No, they see it breaking down into steps. There are molecular diagnostics now that are opening doors."

Emotional reasons factor into the chronic-disease focus: The possibility of telling patients that real hope might exist, or even that more time could be bought, is on oncologists' minds. "We are all real tired of telling people bad news," says Louis M. Weiner, chair of the Department of Medical Oncology at Fox Chase Cancer Center in Philadelphia. "If you are a patient with three half-inch lung lesions, and we can keep those lesions that way forever, that's good," says Louise Grochow, chief of the Investigational Drug Branch of the Cancer Therapy Evaluation Program at the National Cancer Institute. And, researchers say, these patients are willing to take these drugs until they die—which is sometimes sooner than later. Says Jill O'Donnell-Tormey, executive director of the Cancer Research Institute, "People live with arthritis. I think [cancer patients] would be very happy to carry on an active life."

Same Beast, Different Stripes

New drugs have helped the disease-management concept, Warrell says. "The broad characterization of these drugs is that ... [they are] likely to be cytostatic, rather than cytotoxic." So, the expectation is that these drugs will either prevent the tumor's growth or slow it down. "These drugs will never eradicate tumors by themselves."

Finding drugs that control cancer is arguably as hard as discovering a cure. Doctors had high hopes for imatinib mesylate (Gleevec), for example, because it so precisely targets the molecular cause of chronic myelogenous leukemia. Gleevec is now falling prey to the same resistance problems of traditional chemotherapy: patients with advanced disease do great initially, but then cancer finds a way to evade the drug.1

"Whenever you develop a new treatment for cancer, it takes 10 to 15 years to know if it is going to improve survival," says Kent Osborne, director of the Breast Center at Baylor College of Medicine in Houston, explaining that breast cancer survival statistics have just recently begun improving when subjects use tamoxifen, even though this drug was launched many years ago.

Cancer cells mutate quickly and are genetically unstable to boot. So when it comes to solid tumors, frustration levels reach new heights because these growths are less accessible than blood cancers; they must be removed through surgery or biopsy. Frank Rauscher III, deputy director of Wistar Cancer Center at the Wistar Institute in Philadelphia, says that it is "highly unlikely [that] we are going to find disease-specific targets for [the major killers]. Once you have diagnosed lung or colon cancer, 10 more changes [have occurred] in that cell already. Targeting all of them is going to be very difficult." Says Folkman: "One way to bypass all the targets is to simply target the endothelial cells, which are genetically stable." This cell, he says, provides more range.

Angiogenesis inhibitors may be the poster-child drugs for turning cancer into a manageable disease. Folkman tells a story—albeit a possible anomaly—of a cancer patient, who had no options left, living a normal life despite the relatively large tumors still within her. "She had been told she was terminally ill, and her heart had begun to fail from the adriamycin. After two months on [endostatin] she starts to feel better. She develops a tremendous appetite, and goes eight months with stable disease," he says. The tumors are still there, but they seem to be slowly receding. Says Folkman, "At 1.5 years, she's at 46% regression." Ironically, a 46% tumor regression would not be considered a partial response in a standard cancer trial; yet this patient is working and living a relatively normal life.

Highly targeted drugs that promise the low side effects necessary for chronic treatment frequently only work for a subset of patients, says Fox Chase's Weiner. Using a hypothetical model, Weiner says that a target might be present or accessible in only 25% of the patients. "You have to recognize, if you treat everybody, that only 25% will benefit, 75% won't, and you might have side effects." The challenge, he says, is to hone in on the population and increase the probability that they can be treated more effectively. Trastuzumab (Herceptin) is one such drug. It is useful only in the approximately 40% of breast cancer patients who have tumors that overexpress its target, HER2/neu. This kind of personalized medicine can lead to smaller markets for pharmaceutical companies, reducing their incentives to develop drugs. On the other hand, knowing ahead of time which patients are likely to respond can allow clinical trials to be smaller, reducing development costs. Says Hybridon's Agrawal, "We are taking short steps."

Courtesy of Genentech

Dealing with a Devil: A Herceptin-responsive HER2-transgenic mouse breast tumor line developed resistance to Herceptin through somatic mutations in HER2 that abrogated trastuzumab binding. Whether similar mutations occur in human breast cancers that are non-responsive to Herceptin is not yet known.

History Shorts

In the early 1970s, a published article pointed toward a virus as the cancer-causing culprit, thereby initiating changes in the cell's metabolism and allowing for the tumor's uncontrolled growth. In the same article,2 Harvard Medical School's Jerome Groopman wrote that "abnormalities in the genes of the cancer cell were thought to be incidental, rather than fundamental, to the disease."

And, in remarks to a US House of Representatives health committee about this time, renowned pediatrician Sidney Farber said that it was not necessary "to have the full solution of all the problems of basic research.... The history of medicine is replete with examples of cures obtained years, decades, and even centuries before the mechanism of action was understood."2 Researchers also were just becoming aware that tumors needed nutrients to survive.

Genes and their protein products became focal points—the first proto-oncogene was found in 1975, and the tumor suppressor p53 was discovered a few years later. About the same time, researchers created hybridoma technology, designed to create monoclonal antibodies, and tamoxifen received US Food and Drug Administration approval in 1978.3

While researchers were exploring new areas, clinicians were still primarily using toxic and invasive treatments. In 1996, Scientific American reported that for the top 12 types of cancer, radiation, chemotherapy, and surgery were still de rigeur.4 "That is still today true," Folkman says. "Those are standards that allow best possible cures."

Today, approximately 400 cancer-targeted drugs are under consideration;5 many are designed to work with established toxic treatments. Pharmaceutical Research and Manufacturers of America reports that 175 of the 375 biotechnical medicines in clinical trials are anticancer. The worldwide anticancer market is projected to be near $30 billion (US) by 2003, nearly double what it was in 1998.6

So Many Paths

At a recent, private meeting held in France, researchers lamented that more recent advancements in cancer therapy were not happening, says attendee A. Donny Strosberg, CEO of Hybrigenics in Paris. The problem, he says, is that different schools of thought are focused on different pathways, inhibitors, and other targets. A more global approach—integrating these various schools to try and see the whole picture—is necessary, he adds.

With so many molecules to target, some researchers, such as Strosberg and National Cancer Institute director Andrew von Eschenbach, are rethinking how cancer is defined—not by its site of origin, but by the mechanism responsible for the cell's development, maintenance, and survival. "It changes the way oncologists think about treating cancer," says Fox Chase's Weiner, whose group is studying the "electrical wiring" of the cell. "While we certainly hope that these treatments will lead to eradication ..., it may get to the point to induce long-term control."

Using protein-protein interactions, Hybrigenics researchers want to unravel the various pathways in cells and discover novel therapeutic targets. If the company can find ways to manipulate these pathways, Strosberg says, "it doesn't matter what kind" of organ-related cancer it is. Discovering these pathways, he says, will eventually require fewer targeted drugs, and will be more cost-efficient.

The other problem, Strosberg says, is lack of research integration. Databases are needed, a common language must be agreed to—he says some proteins have different names among international researchers—and bioinformatics must be mastered. "One day we have to put it all together," he says. At the AACR conference, Stephen Chanock, of NCI's pediatric oncology branch, said, "We know the way to get the answer is for six different continents to get together."

As time goes on, says J. Clifford Murray, of the Cancer Research Campaign, Department of Clinical Oncology at City Hospital in Nottingham, UK, management will become more important because cancer patients are living longer. "It becomes a chronic illness almost by definition.... With more sophisticated diagnostic techniques, we will be able to detect very small numbers of cells so we will know if the patient already has disseminated disease that is in a very early stage, so we then treat them and keep those tumors, which may be spread all over the body, under control."

Christine Bahls can be contacted at
Mignon Fogarty is a freelance science writer in Santa Cruz, Calif.


2. J. Groopman, "The thirty years' war," The New Yorker, June 4, 2001, p. 52-63.


4. "Twelve major cancers," Scientific American, 275:126-32, 1996.



Supplemental Materials


Popular Now

  1. Unstructured Proteins Help Tardigrades Survive Desiccation
  2. What Budget Cuts Might Mean for US Science
    News Analysis What Budget Cuts Might Mean for US Science

    A look at the historical effects of downsized research funding suggests that the Trump administration’s proposed budget could hit early-career scientists the hardest.  

  3. Opinion: On “The Impact Factor Fallacy”
  4. Inflammation Drives Gut Bacteria Evolution
Business Birmingham