How do mind-body interventions affect breast cancer?

Some of the greatest hopes in the field of psychoneuroimmunology have been in breast cancer treatment. The recent replication of a key study dimmed some of those hopes when it found that a group therapy intervention made no difference in the survival rates of women with metastatic breast cancer compared to controls, as first found in 1989 by Stanford University's David Spiegel, MD.

But the news isn't all discouraging: The replication study did find that women who engaged in group therapy experienced quality-of-life benefits, including reduced pain and improved psychological symptoms, especially among women who were distressed to begin with.

"We were able to confirm clear evidence of psychological benefits in terms of mood and pain control, and those benefits were greatest in women who had psychological distress or poorly controlled pain at the time they entered the study," says Pamela Goodwin, MD, a medical oncologist at the University of Toronto's Mount Sinai Hospital who led the team that sought to replicate Spiegel's findings.

Goodwin's study is not, of course, the last word, say psychologists and others who are continuing to explore mind-body links in cancer.

Psychologists also emphasize that the medical community should not assume from one study that group therapy doesn't help prolong life and, moreover, should look beyond length of life and consider therapy's contributions to quality of life.

"It would be really unfortunate if the survival test is the only one people use when judging the value of these groups," says Geoffrey M. Reed, PhD, the APA Practice Directorate's assistant executive director for professional development. "While Goodwin very clearly articulates that psychotherapy groups have benefits other than survival, we do have some concern that her findings will be interpreted by health systems or hospitals as suggesting that psychotherapy is of negligible value for breast cancer patients."

The Spiegel and Goodwin studies

The study by Spiegel and colleagues was published in the Oct. 14, 1989, issue of The Lancet (Vol. 2, No. 8668). It found that women with metastatic breast cancer who received a year of intensive supportive-expressive group therapy lived 18 months longer on average from study entry than controls. The findings raised the hopes of many that mind-body interventions could make a significant difference in the health outcomes of very ill patients.

But the results also have been criticized for being potentially misleading and building false hope among patients. Goodwin's study, published in the Dec. 13, 2001, issue of the New England Journal of Medicine (Vol. 345, No. 12), studied 158 women with metastatic breast cancer who received the type of therapy Spiegel administered in his study, compared with 77 controls who didn't get the treatment. Although the Goodwin study found quality-of-life benefits, "to our disappointment, we were not able to confirm a survival benefit; there wasn't even a trend toward it," she says.

Spiegel notes that Goodwin's findings shouldn't stop people from using empirically validated treatments with obvious benefits for participants. "It's clear by now that this kind of psychotherapy helps cancer patients," Spiegel says. "How it helps them is a matter of legitimate disagreement. That it helps them is not."

And, says cancer-intervention researcher Barbara Andersen, PhD, of The Ohio State University, the Goodwin study "is just one study, just as Spiegel's was. One of two discrepant findings can't invalidate the other. What Goodwin's finding suggests is the need for additional tests."

Indeed, psychologists emphasize that replication efforts such as Goodwin's are crucial because aspects of the 1989 study had methodological problems. For one, Spiegel's original sample may have been unintentionally biased, notes James L. Spira, PhD, who heads the division of health psychology at the Naval Medical Center in San Diego. Psychologist Bernard Fox, PhD, for example, argued in Psycho-oncology (Vol. 7, No. 5) that Spiegel's control group died at a faster rate than the normal population of recurrent breast cancer patients, while the treatment group lived as long as would be expected in a general population of these patients. In addition, notes Spira, Goodwin's study has a large sample size of 235 women, compared with Spiegel's of 86.

Medical advances in breast cancer treatment may also help explain the discrepancy between Spiegel's and Goodwin's findings, speculates Gregory Miller, PhD, of Washington University. "Many breast cancer patients will survive because of these outstanding medical advances, so there isn't as much room for a psychological intervention to make a difference," Miller points out.

And, he adds, the culture of social support has changed dramatically since Spiegel's original study. "Breast cancer was much more stigmatized when David conducted his seminal study," Miller says. "His intervention provided an incredible outlet for women who didn't have anywhere else to go to discuss this very stigmatized condition. Today, however, people are finding their own sources of support and they don't necessarily need to go to a group to get it."

Psychologists and Spiegel himself concede that he took on a tall order by trying to show survival effects in such an ill population. "Obviously, metastatic cancer is a very powerful biological process," says APA's Reed. "In some ways, it was the 1989 finding by Spiegel and his colleagues that was surprising: that a psychological intervention could affect survival outcomes in this patient group."

Like other earlier studies in this area, Spiegel set out to look only at psychosocial variables. Only later did he conduct analyses of disease outcomes, notes Andersen. "It's only in the last five years that studies have been designed to look directly at survival outcomes," she says.

New intervention trends

Although research shows that group therapy can enhance patients' quality of life, the psychological community is concerned about how the new data will be used. Indeed, psychologists have seen the medical establishment ignore quality-of-life findings before.

"Twenty years of research documents that psychosocial interventions impact a wide range of quality-of-life variables, including ability to sleep, fatigue reduction, improved mood, increased vitality, pain reduction and more general functional capacities like one's ability to exercise," says Spira. "Unfortunately, this news has been known for a while and it won't make the front pages unless there's a physical-health outcome benefit."

Meanwhile, a trend emerging in cancer-intervention research is replacing the focus on survival rates: testing which interventions work best with what populations.

It's an important line of inquiry because research shows that shotgun approaches are spotty at best, observes Margaret Kemeny, PhD, director of the health psychology and behavioral neuroscience program at the University of California at San Francisco.

"What we tend to do now is to take these psychological techniques used in psychotherapy--like the supportive-expressive technique that David Spiegel uses, or cognitive-behavioral techniques--and use them in all contexts with all patients," Kemeny says. "What might be more valuable is to characterize the people we're putting in interventions, select people with a particular psychological state or vulnerability that might warrant some kind of intervention, then target the intervention to them."

Research is pointing to who some of these patients might be, adds Andersen, who is conducting a long-term psychological intervention study of 240 women with Stage-II and Stage-III breast cancer. Research in the area, she explains, is confounded by the fact that many people's stress levels plummet from time of diagnosis on, regardless of whether they've entered an intervention or not. "You have these big, whopping time effects, and oftentimes they wipe out the intervention effects," she notes.

Yet studies consistently find that if you split the sample of treated patients into groups that initially experienced high versus low distress, "lo and behold, there are all of these [positive] changes in the high-distressed groups," says Andersen, whose review of research on outcomes in cancer patients following psychological treatment will appear in this month's issue of the Journal of Consulting and Clinical Psychology (see sidebar, on previous page). Hence, she notes, clinical researchers may want to start thinking about targeting interventions toward these more distressed women.

A group-therapy treatment manual to be published this fall by APA, "Group Psychotherapy for Women with Breast Cancer: A Treatment Manual," by Spira and Reed, addresses some of these distinctions. For instance, the manual provides different protocols for working with first-occurrence and recurrent breast cancer patients, since research shows that these two groups benefit from different types of interventions.

A companion volume to the manual, "The Breast Cancer Notebook: The Healing Power of Reflection," likewise taps into recent research findings, in particular that writing about trauma can be good for your health. Written by Ava Stanton, a licensed clinical social worker in Santa Monica, Calif., and Reed, the book is designed to help women with breast cancer understand and take charge of their own experience.

Other promising avenues for intervention include studying the psychoneuroimmunology of cancer and testing interventions based on those findings (see A bright future for PNI) and examining the benefits of other forms of psychosocial and alternative therapies in addition to group therapy. Few studies, for example, look at the potential benefits of individual therapy for cancer patients or at alternative techniques such as acupuncture, meditation or yoga, says Spira.

As researchers examine which interventions are most helpful to breast cancer patients, there's a group of people they should keep in mind: the patients themselves, emphasizes Reed.

"We have to be careful about concluding a lack of benefit from any intervention without asking participants what the intervention meant to them," he says. "Typically, participants in these groups describe them as being terribly important in terms of finding meaning, connecting with others and expressing their deepest fears in a way that helps them feel they can control these feelings. These are important experiences, even if they're not well-reflected in standardized outcomes scales."