Biobehavioral Factors Mediate Exercise Effects on Fatigue in Breast Cancer Survivors

ROGERS, LAURA Q.1; VICARI, SANDRA2; TRAMMELL, RITA3; HOPKINS-PRICE, PATRICIA3; FOGLEMAN, AMANDA1; SPENNER, ALLISON4; RAO, KRISHNA5; COURNEYA, KERRY S.6; HOELZER, KAREN S.7; ROBBS, RANDALL4; VERHULST, STEVEN4

Medicine & Science in Sports & Exercise:
doi: 10.1249/MSS.0000000000000210
Clinical Sciences
Abstract

Purpose: This study aimed to examine mediators of fatigue response to an exercise intervention for breast cancer survivors in a pilot randomized controlled trial.

Methods: Postmenopausal breast cancer survivors (n = 46; ≤stage 2), off primary treatment, and reporting fatigue and/or sleep dysfunction were randomized to a 3-month exercise intervention (160 min·wk−1 of moderate-intensity aerobic walking, twice weekly resistance training with resistance bands) or control group. Six discussion group sessions provided behavioral support to improve adherence. Fatigue, serum cytokines, accelerometer physical activity, cardiorespiratory fitness, sleep dysfunction, and psychosocial factors were assessed at baseline and 3 months.

Results: The exercise intervention effect sizes for fatigue were as follows: fatigue intensity d = 0.30 (P = 0.34), interference d = −0.38 (P = 0.22), and general fatigue d = −0.49 (P = 0.13). Using the Freedman–Schatzkin difference-in-coefficients tests, increase in fatigue intensity was significantly mediated by interleukin 6 (IL-6) (82%), IL-10 (94%), IL-6/IL-10 (49%), and tumor necrosis factor-α (TNF-α):IL-10 (78%) with reduced sleep dysfunction increasing the relationship between intervention and fatigue intensity rather than mediating intervention effects (−88%). Decrease in fatigue interference was mediated by sleep dysfunction (35%), whereas IL-10 and pro–anti-inflammatory cytokine ratios increased the relationship between intervention and interference (−25% to −40%). The reduction in general fatigue was significantly mediated by minutes of physical activity (76%), sleep dysfunction (45%), and physical activity enjoyment (40%), with IL-10 (−40%) and IL-6/IL-10 (−11%) increasing the intervention–fatigue relationship. In the intervention group, higher baseline fatigue, anxiety, depression, and perceived exercise barrier interference predicted a greater decline in fatigue interference and/or general fatigue during the intervention.

Conclusions: Biobehavioral factors mediated and enhanced intervention effects on fatigue, whereas psychosocial factors predicted fatigue response. Further study is warranted to confirm our results and to improve understanding of relationships that mediate and strengthen the intervention–fatigue association.

Author Information

1Department of Nutrition Sciences, University of Alabama at Birmingham, Birmingham, AL; 2Department of Psychiatry, SIU School of Medicine, Springfield, IL; 3Department of Internal Medicine, SIU School of Medicine, Springfield, IL; 4Center for Clinical Research, SIU School of Medicine, Springfield, IL; 5Department of Medicine, SIU School of Medicine, Springfield, IL; 6Department of Physical Education and Recreation, University of Alberta, Edmonton, Alberta, CANADA; and 7Department of Hematology Oncology, Springfield Clinic, Sprinfield, IL

Address for correspondence: Laura Q. Rogers, M.D., M.P.H., University of Alabama at Birmingham, Webb 326, 1720 2nd Avenue South, Birmingham, AL 35294-3360; E-mail: rogersl@uab.edu.

Submitted for publication May 2013.

Accepted for publication October 2013.

© 2014 American College of Sports Medicine