It is widely accepted that physical activity improves QOL in cancer survivors [6
]. This study uses a large and racially/ethnically diverse cohort to extend our knowledge about physical activity to the early phases of active adjuvant treatment. We found that women who reported the highest levels of moderate and vigorous activity had the highest QOL in this period. Increasing BMI was also independently but inversely associated with QOL, but did not explain the physical activity findings. There were some differences between activity levels and race/ethnicity and relationship to QOL but differences by type of hormonal agent were eliminated after considering age.
This group of breast cancer patients had a higher level of moderate and vigorous activity than reported in other past studies of breast cancer patients [10
]. This is likely due to the more detailed physical activity assessment in our study, resulting in possible overestimation of energy expenditure. Also, since activity levels have increased over time, our cohort may have been more physically active than groups reported in the past. Increasing awareness of the benefits of exercise may have also led to higher self-reported activity levels. Our sample was comparable to other similar breast cancer populations, however, in ratings of QOL during treatment [22
The magnitude of observed differences in QOL scores between the lowest and highest quartile of physical activity was small but in the range reported as important clinically in other studies (2 points on sub-scales and 5 points for the overall score) [22
]. Similar magnitude of benefits have been observed in the few published [1
] studies; a few clinical trials [28
] and observational studies [30
] of exercise during active treatment are currently in process and shown provide additional evidence to support using active treatment as a “teachable moment” for lifestyle changes in the lives of breast cancer patients [31
The association of physical activity was not explained by BMI, although BMI was also associated with QOL. Higher BMI may be associated with poorer body image, although we did not measure this construct [32
]. High BMI, especially BMI of 30 or more, has been linked to increased risk of breast cancer, diminished treatment effectiveness and lower survival [33
]. It is thought that these effects are in part mediated through insulin-like growth factor-1 (IGF-1) axis and altered production of proinflammatory cytokines [36
]. These same paths have been linked to cancer fatigue [37
], which could, in turn, lower QOL. These biological mechanisms are also postulated to partially explain the pathways whereby physical activity improves QOL [38
]. We will be examining inflammatory markers in this cohort in our future study to better understand these pathways in breast cancer QOL and prognosis.
In this sample, White women were more likely to report the highest levels of activity and better QOL compared to minority women. A similar result has been noted in other breast cancer studies [39
]. Of note, physical activity was associated with QOL for Whites but not for minority women. Moreover, even after considering activity level, social support, comorbidity, and clinical factors, Whites maintained a higher QOL during active treatment than minority women. These results suggest that minority women may have different experiences during active therapy than Whites, or that there are unmeasured differences in correlates of QOL by race/ethnicity. Alternatively, minority women may be judging their cancer diagnosis against a different contextual backdrop that leads them to rate QOL at lower levels during the active treatment phase of care (e.g., feeling that “cancer is like a death sentence”) [40
]. Others have noted poorer QOL in Blacks (vs. Whites) among longer-term survivors [41
]. It will be important to explore treatment experiences further, especially when considering physical activity interventions for minority groups.
While aromatase inhibitors have equivalent or better survival benefits than tamoxifen [42
], they have different side-effect profiles, many with the potential to affect physical activity, and QOL. For instance, aromatase inhibitors have been noted to cause greater fatigue, arthralgias, and myaglias than tamoxifen, side effects that could affect both activity level and QOL [43
]. We noted that women on aromatase inhibitors were less active than women on tamoxifen, but this effect disappeared after considering age and the activity-QOL association did not differ by type of hormonal therapy received. It is possible, however, that activity levels were affected by side effects of specific non-hormonal agents (e.g., taxanes and muscle pain and neuropathy). Unfortunately, we do not have data on specific agents and doses or self-reported toxicities; this will be important to consider in future investigations. Although, depression may influence both activity levels and QOL; depression did not alter the association of physical activity with QOL.
Surprisingly, while they had lower levels of activity, older women reported higher
QOL during active therapy than younger women despite having higher risks of toxicity and decreased functional status during chemotherapy than younger women [31
]. It is possible that older women were more accustomed to chronic illness so that health decrements during treatment were viewed less negatively than by younger women. It is unclear, however, if increasing activity levels can enhance physiological reserve and QOL in those with comorbidities (e.g., by diminishing fatigue or increasing oxygen consumption). This will be an important area for future research.
There are several caveats that should be noted in considering our results. First, we measured activity and QOL at the same time, so we cannot make any inferences about causality. It is logical that greater activity levels would lead to improved QOL, yet it is also plausible that women with greater QOL felt more energetic or had greater exercise self-efficacy and, thus, were able to be more active [45
]. Also, a related issue is that our measure of activity asks about the past 6 months, a period that spans the treatment period, but may be skewed toward higher activity levels from pre-diagnosis and does not capture decrements due to breast cancer. We are collecting follow-up assessments of both physical activity and QOL, and will be able to examine temporal relationships in future analyses. Next, we rely on self-reported activity levels, although the instrument that our tool was based on has been validated against doubly labeled water [11
]. Our sample was drawn from a managed care population and findings may therefore be limited in generalizability to other care settings or the uninsured. However, it is unlikely that type of health care coverage affects physical activity patterns per se, although there may be effects on QOL. In addition, women who are diagnosed with breast cancer in this setting appear to be largely similar demographically and diagnostically to the breast cancer population overall. Despite these limits, the study has several strengths, including the very large and diverse sample, use of standard, validated measures of QOL and physical activity, and assessment of important covariates that captured a very high proportion of variance in QOL.
Physical activity, improves QOL for survivors, may decrease breast cancer mortality [36
] and diminishes all-cause mortality rates [10
]. Our study demonstrates that physical activity is also associated with QOL during the active treatment phase of breast cancer care. If causal relationships are demonstrated in future research, our results suggest that physical activity “prescriptions” might be considered early in cancer therapy and continue into survivorship care planning [48
]. Value for minorities and for the growing older population with comorbidities remains to be established and should receive priority for future investigation.