br p However analyses verified the previously
(−0.046 ± .042, p = .274). However, analyses verified the previously observed effects of the MAPs intervention on reducing expression of the 19-gene pro-inflammatory subcomponent of the CTRA score (mean changen = −.084 log2 mRNA expression units ± 0.030 SE, p = .013; Bower et al., 2015). Consistent with the observed decrease in the a priori-specified pro-inflammatory composite score, ancillary analyses of all 78 genes that empirically showed 1.15-fold change in average ex-pression from pre- to post-intervention (15 up-regulated and 63 down-regulated) indicated decreased activity in NF-κB/Rel family transcrip-tion factors (0.34-fold ratio of Fer-1 in up- vs. down-regulated genes: log2 ratio = −1.56 ± 0.60, p = .010), consistent with our earlier trial (Bower et al., 2015). Further, Transcript Origin Analyses identified pro-inflammatory CD16- “classical monocytes” as the pri-mary cellular origin of down-regulated gene transcripts (cell diag-nosticity score: 0.66 ± 0.27, p = .009).
3.3. Primary analyses: Associations between changes in well-being, depressive symptoms and CTRA gene expression
Our primary interest was evaluating whether the magnitude of change in eudaimonic well-being was associated with the magnitude of change in CTRA gene expression. Although the 53-gene CTRA compo-site score did not show significant pre-post intervention change, there was sufficient variability in the magnitude of change to test for corre-lations with change in eudaimonic well-being. Results showed that women who reported greater increases in eudaimonic well-being over the 6-week intervention showed greater reduction in CTRA scores over the same period (change in log2 RNA abundance per SD change in eudaimonic well-being: b = −0.140 ± SE 0.040, p = .0009; see Fig. 1a), controlling for age, BMI, time since cancer diagnosis, tumor stage, and use of endocrine therapy. Similar results emerged in analyses that additionally controlled for changes in hedonic well-being (b = −0.136 ± SE 0.039, p = .0009) and analyses not controlling for any covariates (b = −0.070 ± .035, p = .0492; see supplement page 2 for additional analysis of MHC-SF subscales). Exploratory analyses of spe-cific components of the overall CTRA composite score indicated that increases in eudaimonic well-being were associated with an increase in the inversely weighted antiviral/antibody-related CTRA subcomponent in both unadjusted analyses (b = .127 ± .053, p = .0228) and cov-ariate-adjusted analyses (b = .220 ± .062, p = .0012; Fig. 1b). How-ever, changes in eudaimonic well-being were not significantly corre-lated with changes in the pro-inflammatory subcomponent (Fig. 1b; p = .9215).
In secondary analyses, regression analyses were also conducted to examine the association between changes in hedonic well-being alone and CTRA gene expression and between changes in depressive symp-toms alone and CTRA gene expression. In these models, neither the magnitude of change in hedonic well-being nor depressive symptoms were associated with change in overall CTRA gene expression, in either unadjusted or covariate-adjusted analyses (all p’s > 0.20; see Fig. 1a). Further, changes in hedonic well-being and depressive symptoms were not related to changes in either of the CTRA sub-components (all p’s > .05).
Change in CTRA gene expression (% change in RNA per SD change in predictor)
Fig. 1. Relationship of change in CTRA and change in well-being and depressive symptoms. Greater increases in eudaimonic well-being were associated with greater reduction in the 53-gene CTRA composite score from pre- to post-intervention (A). Increases in eudaimonic well-being were associated with an increase in the antiviral/antibody-related CTRA subcomponent, but not the pro-inflammatory subcomponent, from pre- to post-intervention (B). Data represent regression coefficients from linear regression models controlling for age, BMI, time since cancer diagnosis, tumor stage, and use of endocrine therapy.
Change in CTRA subcomponent gene expression (% change in RNA per SD change in Eudaimonic well-being)
The present data are consistent with the hypothesis that the link between eudaimonic well-being and health may involve alterations in molecular signaling pathways that control immune function. In the present study, levels of eudaimonic well-being increased after partici-pation in a 6-week mindfulness intervention, and those who showed the greatest increases in eudaimonic well-being showed the greatest de-creases over time in expression of the CTRA transcriptome profile. The meditation intervention was also associated with decreases in depres-sive symptoms, but these changes were not significantly associated with changes in CTRA gene expression. These findings are consistent with previous cross-sectional studies reporting an association between in-dividual differences in eudaimonic well-being and individual differ-ences in CTRA gene expression under basal conditions, and these data show that such effects extend to intra-individual reductions in eu-daimonic well-being as breast cancer survivors initiated a mindfulness meditation practice.