Docosahexaenoic acid (DHA) supplementation attenuates changes in the concentration, phenotype, and response of immune peripheral blood cells in breast cancer patients undergoing neoadjuvant therapy. Secondary findings from the DHA-WIN trial

Abstract

Background: Breast cancer neoadjuvant therapy may negatively impact the immune system. As a secondary outcome of the docosahexaenoic acid (DHA) for women with breast cancer in the neoadjuvant setting (DHA-WIN trial), we sought to assess the effects of an intervention with DHA on parameters of immune function of women undergoing neoadjuvant therapy.

Methods: Women with early-stage breast cancer in the neoadjuvant setting were recruited for the DHA-WIN trial and randomly assigned to receive either 4.4 g/day of DHA or a placebo for 18 weeks in conjunction with their neoadjuvant chemotherapy for breast cancer. Venous blood was collected to isolate peripheral blood mononuclear cells. Immune parameters were assessed by measuring white blood cell concentration, flow cytometry, and cytokines concentration after mitogen-stimulated immune response.

Results: In the placebo group the proportion of T cells (CD3 +), and functionally active monocytes (CD14 + HLA-DR +) was reduced at the last cycle of chemotherapy (15 weeks) but remained constant in the DHA group (P interaction < 0.05). The neutrophil-to-lymphocyte ratio (NLR) was maintained in the DHA group but increased in the placebo at the end of chemotherapy (P-interaction = 0.02). An increase in this ratio was associated with lower chance of achieving pathological complete response (OR = 0.32, 95% CI [0.14,0.16], P = 0.01). After 15 weeks of therapy, the DHA-supplemented group had higher concentrations of stimulated cytokines IL-4, IL-10, and the T helper type 1 cytokine IFN-γ after phytohemagglutinin (PHA) challenge, and higher concentrations of TNF-α and IFN-γ cytokines after lipopolysaccharide exposure (P < 0.05).

Conclusion: Supplementing DHA during breast cancer neoadjuvant chemotherapy improved systemic immune function by attenuating changes in blood cell concentrations, preventing depletion of immune cells, and enhancing ex vivo cytokine secretion after stimulation.

Keywords: Cytokines; Immune phenotype; Immune response; Long-chain polyunsaturated fatty acid; Pathological complete response.

Fig. 1

Change in the concentration of blood cells during treatment in the DHA-WIN trial. A Values are represented as mean and 95% of the confidence interval. Generalized estimating equations were performed with the main effects and interaction between time and groups of intervention. The neutrophils to lymphocytes ratio was stratified by placebo and DHA groups. *denotes the value is different from the baseline within the group based on post hoc Bonferroni adjustment (P < 0.05). B-C Changes in the concentration of blood cell concentrations during neoadjuvant treatment categorized by therapeutic response (pCR). The concentration of blood at 18 weeks was subtracted from values at baseline to determine the change during treatment. B Forest plot of univariate logistic regression analysis (n = 25 non- pCR and n = 14 pCR). C Mann–Whitney test comparing changes in the neutrophils-to-lymphocytes ratio between participants who achieved pCR and patients who did not achieve pCR at the end of neoadjuvant chemotherapy (bars represent median and range). DHA, docosahexaenoic acid; RBC, red blood cells; WBC, white blood cells; pCR, pathological complete response

Fig. 2

Changes in the proportion of T cell phenotypes in the PBMCs of participants in the DHA-WIN trial. Values are represented as mean and error bars as the 95% of the confidence interval. Generalized estimating equations were performed for each parameter analyzed with the main effects and interaction between time and groups of intervention. A * Indicates the value is different from the baseline within the group based on post hoc Bonferroni adjustment, * P < 0.05. B-F Labelled means without a common letter differ (P < 0.05) based on post hoc with Bonferroni adjustment. DHA, docosahexaenoic acid; Th1, T helper type 1

Fig. 3

Changes in the proportion of cell phenotypes in the PBMCs of participants in the DHA-WIN trial. Values are represented as mean and 95% of the confidence interval. Generalized estimating equations were performed for each parameter analyzed with the main effects and interaction between time and groups of intervention. A * Indicates the value is different from the baseline within the group based on post hoc Bonferroni adjustment, * P < 0.05. P values indicate the comparison between groups at the same time point (based on post hoc Bonferroni adjustment). B-H Labelled means without a common letter differ (P < 0.05) based on post hoc with Bonferroni adjustment. PBMCs, peripheral blood mononuclear cells; DHA, docosahexaenoic acid; NK, natural killer; DCs, dendritic cells

Fig. 4

Cytokine concentration in the supernatant from isolated peripheral blood mononuclear cells after stimulation with PHA for 48 h. Values are represented as mean and 95% of the confidence interval. Generalized estimating equations were performed for each parameter analyzed with the main effects and interaction between time and groups of intervention. A-D * Denotes the value is different from the baseline within the group based on post hoc Bonferroni adjustment * P < 0.05, **P < 0.01, *** P < 0.001. P values indicate the comparison between groups at the same time point (based on post hoc Bonferroni adjustment). E–H Labelled means without a common letter differ (P < 0.05) based on post hoc with Bonferroni adjustment. Although a statistically significant time effect was found for the production of IL-2, pairwise comparisons did not result in statistically significant differences (represented by common letters). DHA, docosahexaenoic acid; PHA, phytohemagglutinin; LPS, lipopolysaccharide; IL, interleukin; IFN-γ, interferon-gamma; TNF- α, tumour necrosis factor-alpha

Fig. 5

Changes over time in cytokines concentrations in the supernatant from isolated peripheral blood mononuclear cells after stimulation with LPS for 48 h (n = 49). Values are represented as mean and 95% of the confidence interval. The P values represent the interaction between the different groups and time points using Generalized estimating equations (GEE). Multiple comparisons with Bonferroni adjustment. A-B * Indicates the value is different from the baseline within the group based on post hoc Bonferroni adjustment, * P < 0.05. P values indicate the comparison between groups at the same time point (based on post hoc Bonferroni adjustment). C-E Labelled means without a common letter differ (P < 0.05) based on post hoc with Bonferroni adjustment. PHA, phytohemagglutinin; LPS, lipopolysaccharide; IL, interleukin; IFN-γ, interferon-gamma; TNF- α, tumour necrosis factor-alpha

Fig. 6

Human epidermal growth factor receptor 2 (HER2) status and changes in different immune parameters analyzed in the trial. The values at the end of therapy (18 weeks, RBCs concentration) or cycle 6 (15 weeks, LPS IL10, % of Th17⁺ cells, mature and naïve T helper CD4 cells), were subtracted from values assessed at baseline (before intervention). Groups were stratified based on the expression of HER2 and Mann–Whitney tests were performed to evaluate differences between groups. A Changes in red blood cell concentration. B Change in the concentration of the cytokine IL-10 in the supernatant of LPS-stimulated PBMCs. C Changes in the proportion of Th17⁺ cells in CD3⁺CD4⁺ from the flow cytometry analysis. D Changes in the proportion of memory (CD45RO⁺) in CD3⁺CD4⁺ from the flow cytometry analysis. E Changes in the proportion of naïve (CD45RA⁺) in CD3⁺CD4⁺ from the flow cytometry analysis; LPS, lipopolysaccharide; IL, interleukin; PBMCs, peripheral blood mononuclear cells