Omega-3 fatty acid supplementation – A possible dietary adjunct to enhance immune checkpoint inhibition therapy in cancer?

Abbreviations

1 Introduction

Colorectal cancer (CRC) is one of the most common cancers worldwide [1]. In CRC, the concept of an anti-tumor immune response is well established [2]. This inflammatory reaction in the tumor-microenvironment is thought to represent the host’s local immune response against tumor cells. Studies found that the inflammatory infiltrate in and around the tumor correlates with improved survival [2]. Recently, several studies have been published, which offer the prospect to a novel treatment of CRC by targeting cancer-immune checkpoints, such as the programmed cell death 1 (PD-1) pathway [3]. These drugs, such as the monoclonal antibody pembrolizumab can overcome immune resistance of subtypes of CRC and thus enhance the host’s local immune response against infiltrating tumor cells [3, 4]. Interestingly, Zelenay et al. and others provided evidence for a possible synergistic effect of PD-1 immune checkpoint blockade and NSAID-mediated antitumor pathways [5, 6]. Moreover, several epidemiological studies show beneficial effects of long-term intake of acetylsalicylic acid (ASA, aspirin) and other NSAIDs on incidence and survival of CRC [7, 8]. These beneficial effects are associated with reduced conversion of the omega-6 polyunsaturated fatty acid (n-6 PUFA) arachidonic acid (AA) into biologically active eicosanoids, such as PGE₂ [8, 9]. By reducing PGE₂ and other tumor-promoting lipid mediators, NSAIDs, such as aspirin, can help reverse immune evasion of CRC [5, 9] (Fig. 1).

OPEN IN VIEWER

Fig.1

Effect of aspirin on PGE₂-concentration in the microenvironment of colorectal cancer. Figure adapted and modified from (5).

These and other observations also point towards a possible role of polyunsaturated fatty acids (PUFA) in modulating the host’s local immune response in the context of colon cancer. PUFA are fatty acids, characterized by at least two carbon-to-carbon double bonds. Their classification is based on the position of the first double bond, counting from the methyl (omega) end [10]. In general, it appears that the n-6 PUFA AA promotes a predominantly pro-inflammatory state, whereas EPA and DHA exert an inflammation-dampening effect on immune cells [11]. Several of these effects are believed to be mediated through the action of lipid mediators [11, 12]. Indeed, AA-derived leukotrienes and prostaglandins can act as potent pro-inflammatory lipid metabolites (depending on cell type and receptor) [11, 13]. AA-derived PGE₂, in particular, has shown to be instrumental in tumor immune evasion of tumors [5, 8]. PGE₂ inhibits phagocytosis and the TLR-dependent activation of TNF-α secretion via the IL-1R–associated kinase-M [14–16]. EPA and DHA, on the other hand, inhibit synthesis of AA-derived, pro-inflammatory eicosanoids such as PGE₂ [17]. Moreover, n-3 PUFA are also precursors of anti-inflammatory lipid mediators, such as resolvins (RVs), protectins (PDs), and maresins (MaRs) with their pathway indicators 18-hydroxyeicosapentaenoic acid (18-HEPE), 14-hydroxydocosahexaenoic acid (14-HDHA) and 17-hydroxydocosahexaenoic acid (17-HDHA) [12, 18].

In line with the possibility to transfer the anti-tumor paradigm stated by Zelenay et al. [5] for aspirin (Fig. 1) to n-3 PUFAs, an increase in the dietary n-3 to n-6 PUFA ratio was found to not only correlate with higher TNF-α secretion but also with significantly lower levels of AA-derived PGE₂ in immune cells [16]. Moreover, in a previous study we demonstrated the ability of DHA to reduce formation of PGE₂ as well as proliferation in CRC cells [19]. Indeed, several studies on n-3 PUFA emphasize their inhibitory action on the synthesis of PGE₂ [15, 20], supporting the hypothesis of increased tumor immune surveillance due to this PGE₂-suppressive n-3 PUFA effect. In line with this, incubation of murine peritoneal macrophages with AA potently inhibited LPS-induced TNF-α production [21]. It was observed that concomitant treatment of these macrophages with AA and indomethacin (inhibiting the synthesis of PGs) restored 90% of the TNF-α concentration, which indicates that AA exerts an inhibitory effect on TNF-α secretion via increased PG-levels.

Due to the demonstrated effects of n-3 and n-6 PUFA on the immune system, these fatty acids could affect the immune response to tumors. Particularly in light of the previously published findings of immune-based therapies in CRC patients and the paradigm of immune-activation as an anti-cancer treatment approach, it is now pertinent to reassess the possible effects of n-3 and n-6 PUFA on immune cell activity: Do these fatty acids suppress immune function including anti-tumor immune reactions, or could they even have immune-stimulatory effects in the tumor microenvironment, supporting Cancer Immune Therapy?

To directly test for effects of n-3 and n-6 PUFA on immune cells with regard to CRC we assessed differentially induced cytokine secretion by human PBMCs, derived from healthy donors. Cells were tested with two different stimuli, [1] LPS to mimic activation by bacterial products and [2] colon tumor cell conditioned media to mimic activation by tumor cells (for detailed Methods see Supplementary Material). In these experiments DHA significantly reduced LPS-triggered IL-10 secretion by PBMCs (Fig. 2a). Interestingly, DHA had a more pronounced effect on cytokine secretion that was induced by conditioned media derived from human colorectal adenocarcinoma HT-29 cells (CM): DHA reduced secretion of IL-10 while increasing TNF-α levels (Fig. 2b). The n-6 PUFA AA, on the other hand, reduced TNF-α secretion by PBMCs stimulated with LPS as well as with CM (Fig. 2a and b, respectively). In analysis of variance (ANOVA), when compared to PBMCs treated with AA, LPS- as well as CM-induced TNF-α secretion was significantly higher in cells incubated with EPA or DHA (p < 0.05). TNF-α is a typical pro-inflammatory cytokine, while IL-10 has been shown to exert effects limiting cytotoxic T-cell action [22, 23]. Our results thus suggest that incubation of PBMCs with DHA results in a more aggressive immunological response against tumor cells, while AA could be associated with an immunosuppressive effect in this context.

OPEN IN VIEWER

Fig.2

Effect of three major n-3 and n-6 PUFA (EPA, DHA, and AA) on cytokine secretion by PBMCs. (a) LPS-induced cytokine secretion; (b) Cytokine secretion induced by conditioned media derived from human colorectal adenocarcinoma HT-29 cells (CM). PBMCs were isolated from leukocyte depletion filters, acquired from adult blood bank donors. After 24h-preincubation with 50μM EPA, DHA, or AA, PBMCs were stimulated with LPS (a) or CM (b) for another 24 h. TNF-α, IL-6, and IL-10 secretion were then measured in the supernatant using ELISA. For controls PBMCs stimulated with LPS or CM, w/o prior incubation with PUFA, were used (for a detailed description of Materials and Methods used refer to Supplementary Material). Additional controls included: (1) untreated PBMCs (w/o PUFA-incubation or stimulation) and (2) PBMCs incubated with PUFA but w/o subsequent stimulation (no significant difference was observed when comparing cytokine levels found in additional control groups (1) and (2) - data not shown). Data is expressed as the relative mean±SEM of 5 PBMC donors as compared to LPS and CM control, respectively. *p < 0.05, **p < 0.01.

The primary prevention of CRC by long-term NSAID-intake, in particular aspirin, is believed to be caused by a reduced conversion of AA into biologically active eicosanoids such as PGE₂, which has been shown to contribute to immune evasion by tumor cells [5, 8]. Additionally, recently published studies on immune checkpoint inhibitors demonstrate the clinical effectiveness of increasing an anti-tumor immune response as a novel treatment approach in CRC [3, 4]. In this context, published data indicate a possible supporting effect of aspirin in the context of immune checkpoint inhibitor use for cancer therapy [5, 6].

Our findings on DHA-induced changes in cytokine secretion of PBMCs raise the possibility of a pro-immunogenic effect of n-3 PUFA via reduced PGE₂ in the context of the immune system’s response to cancer (Fig. 3). This could be of particular interest with the advent of immune checkpoint inhibitor therapy in oncology as this implies the possibility of an enhancing effect of n-3 PUFA in the context of these therapeutic interventions. However, the implications of the changes in cytokine secretion observed here are not entirely clear for several reasons: For one, TNF-α has shown to exert ambivalent effects on cancer cells, depending on the activation of intracellular pathways [24]. Also, the role of IL-10 in the context of cancer is controversial: While many data show that Il-10 can reduce antigen-specific T-cell activation and induce T-cell anergy and might thus be a pro-tumorigenic inflammatory mediator, recent data demonstrate an important role for IL-10 in effective immune surveillance of tumor cells [25].

OPEN IN VIEWER

Fig.3

Possible effect of n-3 PUFA in colorectal cancer. In analogy to the effect of aspirin, increasing the anti-tumor immune response (5), n-3 PUFA might have a similar effect.

We therefore propose future studies, in experimental (animal) models as well as in the clinical setting, to test for an enhanced anti-tumor effect of the combination of high n-3 PUFA supplementation with cancer immunotherapy as compared to immunotherapy in the context of a high n-6 PUFA environment.

Conflicts of interest

None.