In our previous column, we discussed the results of several studies from Touch Research Institute (TRI) in Miami, Fla., showing positive effects of massage therapy on immune function, anxiety, and depression in subjects diagnosed with human immunodeficiency virus (HIV). Similar results have also been documented by TRI researchers for breast cancer patients, expanding the potential application of this modality to support and enhance healing in life-threatening illnesses.

Current statistics from the American Cancer Society indicate that breast cancer is second only to lung cancer as the leading cause of death in women. It is predicted that one of every seven women (13.4 percent) will be diagnosed with breast cancer during her lifetime.¹ Those diagnosed will likely face a grueling treatment regimen and uncertain outcome. Among the psychological risks noted for this group are increased depression, higher stress and anxiety levels, and anger.² Research has shown that psychological stress can negatively impact the immune system of breast cancer patients. Stress has been associated with lower counts of natural killer (NK) cells and decreased NK cell activity, which in turn increases opportunity for tumor growth.³ Thus, reducing stress would seem to bolster the body’s ability to mount an immune response. TRI’s studies of massage therapy for HIV, as noted, are supportive of this premise. Additionally, the institute’s decades of research on a variety of medical conditions have shown the positive impact of massage on neuroendocrine function and associated benefits for improved mood and decreased depression.

The two randomized controlled trials (RCT) presented here are a combination of an initial pilot project (Hernandez-Reif et al., 2004) and its extension into a second study incorporating a relaxation comparison group (Hernandez-Reif et al., 2005). In the 2004 study, researchers found breast cancer patients receiving massage therapy showed increased NK cells and lymphocytes, and reduction in self-reported depression, anxiety, and anger as compared to a control group with standard medical care. The addition of the comparison group in the second TRI study served to address whether results were due to relaxation benefits of massage or if some additional aspect of the treatment, such as applied pressure to the skin, might be influencing NK and lymphocyte counts.⁴

Several psychological interventions, including guided imagery
and cognitive-behavioral and stress management approaches have been shown to boost immune response in cancer patients.⁵ Benefits have also been documented for breast cancer patients when these approaches are combined with relaxation techniques. Walker et al. (1999) reported increased relaxation and enhanced quality of life for women receiving combined relaxation training with guided imagery during chemotherapy treatment.⁶ In another study using relaxation and visualization, with cognitive-behavioral intervention added to the initial protocol for a second subject group, findings were significant for reduction of pain (Arathuzik, 1994).⁷ Combining relaxation, guided imagery, and biofeedback, Gruber et al. (1993) reported significant increase in NK cell activity along with reduction on anxiety scores for women with Stage 1 breast cancer.⁸

It appears that relaxation plays some part in both psychological and physiological benefits for breast cancer patients, but within these multi-modality studies the emphasis is on combined effect, not single effect.⁹ Based on this research, as well as positive results from their previous studies of HIV and other medical conditions, TRI researchers surmised that massage therapy might provide yet another alternative approach “for reducing psychological stress and enhancing immune function in breast cancer patients.” Utilizing an experimental protocol of five weeks of massage therapy for early stage breast cancer patients, the team’s targeted potential outcomes in the first study were reduction in stress and depression levels, corresponding changes in lab values for neuroendocrine function, and an increase in NK cell count and activity.¹⁰

The pilot population included 34 women diagnosed with Stage 1 or 2 breast cancer within the past three years and who were three months beyond chemotherapy, radiation, or surgical treatment. Authors note the three-month marker is because of the effect of these treatments on immune measures and NK cell cytotoxicity (the degree to which NK cells are toxic to target cells). Subjects were randomized to a control group with no treatment (n=16) or massage therapy group (n=18). Members of the control group continued standard medical care, but were also offered massage therapy at the end of the study period. The experimental group received a 30-minute massage, three times weekly over five weeks, along with routine medical care. “The massage sessions followed a standardized protocol that was similar to the protocol that had been effective in reducing stress hormones and increasing NK cells in our HIV study,” researchers say. The routine, consisting of Swedish, Trager, and acupressure techniques, was administered by a different female therapist for each session.¹¹

On the first and last day of the study, a variety of measurements were taken to assess short- and long-term effects. To determine short-term (immediate) effects, self-report scales for anxiety and mood (depression, anger, and vigor) were administered before and after the massage session, or in the case of controls, before and after a 30-minute control period. Longer term effects were evaluated through a symptom checklist in which subjects reported distress related to depression, anxiety, and hostility. The Life Events Questionnaire was utilized to identify any confounding factors, such as a major loss or lifestyle change, that might have influenced treatment outcome. At the beginning and end of the study period, blood and urine samples were also taken for neuroendocrine, neurotransmitter, and immune function measures.¹²

For the most part, results replicated the HIV research outcome, with the massage group showing an increase in NK cell and lymphocyte count at trial end. But unlike the HIV studies, there was no change in NK cytotoxicity. This discrepancy might be related to frequency and duration, as the HIV subjects received more frequent and longer duration treatment than in this study. Unlike other TRI massage studies, there was no decrease in neuroendocrine measures (cortisol stress hormone, norepinephrine, and epinephrine), which researchers say also may be related to frequency and duration of treatment or to the breast cancer groups’ higher initial cortisol and potentially higher stress levels. However, that these measures did not increase could indicate massage administered several times weekly might keep stress levels from rising for this population.¹³

Positive results were also found in the massage group on self-report scales for reduction of depression, anxiety, and anger and hostility. These findings were complemented by increases in serotonin and dopamine levels. Previous massage studies have shown a similar association between increase in these neurotransmitters and decreases in psychological distress. In contrast, the control group showed little or no change in biochemical and immune measures, with the exception of an unexplained significant increase in norepinephrine.¹⁴

In this first pilot study, the team concluded “massage therapy was found to be a safe treatment” and “to positively impact the psychology, immunology, and biochemistry of women with breast cancer.”¹⁵ The investigation continued into a second RCT with the addition of four subjects to the massage group and 20 women assigned to a relaxation group. Researchers incorporated the control group data from the original pilot into the second study as a “no treatment” comparison. The additional massage subjects received treatment according to the same protocol and schedule as the pilot subjects. Participants in the relaxation group were given a 30-minute progressive muscle relaxation tape to follow for treatment sessions at home, according to the same timetable as massage (three times weekly for five weeks), with their first and last sessions conducted at TRI for collection of pre/post measurements.¹⁶

A step effect was found, as predicted by researchers, with massage showing greater benefits than relaxation, which, in turn, was of greater benefit than standard care. Short-term effects of decreased depression and anxiety were reported for both the massage and relaxation groups, with a greater decrease in depression for massage subjects. In long-term anxiety and depression measures, massage again showed greater effect. This step effect also held for measures of increased vigor and decreased anger.¹⁷

Although pain perception and pain level were assessed in both trials, these measures were not reported until completion of the second RCT. Change scores on the first day of treatment did not differ between relaxation and massage groups, but both showed a greater decrease than controls. These results held for the final day of treatment, with massage showing a greater decrease.¹⁸

As in the pilot study, only the massage group showed an increase in dopamine and serotonin levels by trial end. These improved values may relate to the findings of improved mood and vigor, as well
as reduced pain perception. But researchers say the increase in dopamine might also contribute to improved immune measures in the massage group.¹⁹ Recent studies with mice have suggested a possible correlation.²⁰

While some immune-boosting benefits were found for the relaxation group, effects were greater in the massage group, most notably the increase in NK cells and lymphocytes. The team states this was a pivotal finding that further supports results of their previous studies in HIV and cancer. But a conundrum arises in that a significant increase in NK cytotoxicity was found for relaxation, but not massage. Conversely, while the increase in NK cell count for massage reached statistical significance, no similar increase in NK or lymphocyte count was documented for relaxation.²¹ According to TRI’s director, Tiffany Field, Ph.D., researchers have no explanation for this outcome. As noted earlier, in previous HIV studies with men and adolescents, an increase in cytotoxicity accompanied the increase in NK cell count for massage groups.

It remains, however, that the findings for massage and relaxation therapy offer evidence of benefits for women with breast cancer. Although the massage effects appear to be greater, both of these complementary therapies may be helpful in reducing depression, anxiety, and pain in this population. As NK cells are known to destroy tumor cells, authors say, an improvement in clinical condition would be expected for breast cancer patients receiving massage.²² The increases in neurotransmitters documented for massage may also bolster the body’s defense system in fighting this disease.

While more research is called for in determining the mechanisms and interactions involved, TRI researchers suggest a possible explanation for the increase in NK and lymphoctye count for the massage group. It may be that massage enhances the immune function through the stimulation of pressure receptors that in turn “may decrease sympathetic and increase parasympathetic activity.”²³ In previous studies, TRI teams have reported on the physiological changes associated with moderate pressure of massage (see “Moderate vs. Light Pressure in Massage: New Studies from Touch Research Institute,” Massage & Bodywork, April/May 2005, page 134).

The neuroendocrine results also leave room for speculation. Initially, the team anticipated a decrease in cortisol stress hormone for the massage group, as this has been a repeated finding in TRI literature.²⁴ As noted, breast cancer patients are at increased risk for stress, and higher cortisol levels in this population have been reported in research.25 In their pilot report, the team surmises that a higher baseline level of cortisol stress hormone in their subjects might account for the lack of change in NK cytotoxicity, “in that cortisol is known to negatively impact NK cell activity.”26 And yet, cytotoxicity did increase for the relaxation group. As with many studies, there are always more questions to be explored.