A FATal Combination: Fibroblast-Derived Lipids and Cancer-Derived Autotaxin Promote Pancreatic Cancer Growth

Pancreatic ductal adenocarcinoma (PDAC) is highly lethal, with a 5-year survival rate of less than 8%. Although several therapeutic advances have improved patient survival in other cancer types, no major breakthrough has been made in PDAC treatment. Efforts to develop novel therapies have largely focused on targeting cancer cells, despite the presence of nonneoplastic stromal components that substantially contribute to the progression of the disease. Cancer-associated fibroblasts (CAF), which are in part derived by the activation of resident pancreatic stellate cells (PSC), comprise the most abundant cell population in PDAC and secrete growth factors and inflammatory ligands that promote tumor proliferation, therapy resistance, and immune exclusion. Although these observations provide the rationale for targeting CAFs, recent preclinical and clinical studies indicate that these cells may also have tumor-restraining functions. Therefore, rather than deplete CAFs as a whole, blocking specific pathways will be key in order to design successful combination therapies that affect cancer cells and the tumor-promoting stromal components of the PDAC microenvironment. Auciello and colleagues take on this new approach in targeting a specific network of the cancer-associated stroma by focusing on one aspect of stromal metabolism that supports pancreatic cancer progression. The authors identify a novel lipid-based cross-talk between CAFs and pancreatic cancer cells and provide insights into the potentially beneficial effect of blocking this interaction to tackle PDAC progression (1).

PDAC cells have been shown to scavenge metabolites, sugars, amino acids, proteins, and lipids from the surrounding microenvironment and the circulation to support tumor growth (2–5). In particular, CAFs have been reported to promote pancreatic cancer progression by providing growth factors and amino acids as sources of mitogenic signals and biomass production to cancer cells (6). The contribution of the stroma in supporting tumor progression by releasing amino acids and metabolic intermediates, such as lactate, has been shown for a number of other cancer types. Auciello and colleagues take a further step in the understanding of the metabolic cross-talk between stromal cells and cancer cells and identify a previously unappreciated role of CAF-secreted lipids in promoting pancreatic cancer proliferation through activation of mitogenic pathways. Whereas in the normal pancreas PSCs are quiescent, fat-storing cells, in pancreatic cancer their activation to CAFs leads to the release of PSC intracellular lipids. Extracellularly, these lipids in turn function as signaling molecules and provide substrates for biomass production to cancer cells. Among these stromal-derived lipids, lysophosphatidylcholines (LPC) are hydrolyzed by the stroma- and cancer-secreted lysophospholipase autotaxin to produce lysophosphatidic acid (LPA). Although the source of LPA in PDAC was previously unclear, Auciello and colleagues identify activated PSCs as a major source of the LPA precursor LPC within pancreatic tumors. Extracellular LPA binds to several specific receptors that activate PI3K and AKT pathways, thereby stimulating the proliferation and migration of PDAC cells. The genetic or therapeutic targeting of autotaxin inhibits this signaling axis and impairs PDAC growth in subcutaneous and orthotopic models (Fig. 1).

Similar to the metabolic cross-talk described by Auciello and colleagues, adipocytes have previously been shown to provide lipids to fuel tumor cell growth in melanoma and ovarian cancer (7, 8). In addition, various classes of lipids, including LPA, have been shown to promote metastasis dissemination and formation in melanoma, rhabdomyosarcoma, and oral cancer, in some cases by modulating the recruitment of macrophages and other immunosuppressive cell populations (9, 10). These observations indicate that the role of stromal-derived lipids in supporting cancer growth is a common feature in different cancer types and should be further investigated.

Although the work of Auciello and colleagues highlights the relevance of activated PSC-derived lipids in affecting cancer cells at the primary site, it remains to be established whether stroma-derived extracellular lipids play a role in the healthy and inflamed pancreas, and in preneoplastic and metastatic carcinoma. In addition, a key question that needs to be addressed is what function these molecules have on other stromal cell types, such as immune cells, present in PDAC. Recent literature indicates that extracellular lipids might have pleiotropic effects depending on the context. At homeostasis, such as in the healthy colon, and in some inflammatory conditions, extracellular short-chain fatty acids have been shown to provide sources of biomass or signals to immune cells. Free fatty acids have also been shown to bind to Toll-like receptors and promote inflammation by activating the NF-κB and JNK1 signaling pathways. On the other hand, omega-3 and -6 fatty acids have been shown to have anti-inflammatory effects, highlighting specific roles of distinct classes of lipids.

Extracellular lipids that play various roles in healthy or cancerous tissues are not exclusively derived by the stroma. Indeed, lipids secreted by microbes have been shown to be important for colon health, but also to be a source of biomass in colon cancer. Altogether, these studies highlight the need to further investigate the contribution of extracellular lipids during homeostasis, inflammation, and neoplasia.

Although historically considered a homogeneously tumor-promoting population, CAFs in PDAC have been recently revealed as a heterogeneous population composed of distinct subtypes, named iCAFs and myCAFs depending on their inflammatory or myofibroblastic phenotype (11). Whereas single-cell RNA-sequencing studies have confirmed distinct CAF subsets in PDAC and other cancer types, including colon and lung cancers, the extent of stromal complexity has yet to be fully understood. In particular, new questions emerge regarding the roles and mechanisms of distinct CAF subtypes in affecting cancer progression. Auciello and colleagues start to address this by investigating potential metabolic differences between iCAFs and myCAFs in PDAC. The authors show that in vitro iCAFs and myCAFs secrete similar levels of LPCs, and they both induce autotaxin expression in cancer cells in a paracrine manner. However, the contribution of these distinct populations in vivo remains to be assessed. Moreover, iCAFs and myCAFs show a different spatial distribution in the tumor relative to cancer cells (11). Therefore, considering the importance of proximal LPA availability for uptake, the impact that LPCs secreted from either CAF subtype have on cancer cells might be different. In addition, the distinct phenotypes of iCAFs and myCAFs might indirectly affect this metabolic cross-talk. For example, IL6, which is produced by the iCAFs, has been shown to induce lipolysis in neighboring adipocytes and thus mediate the liberation of fatty acids. Further dissection of the metabolic cross-talk between distinct fibroblast subtypes and cancer cells in vivo will be key to understanding the potentially differential impacts of proteins and lipids in fueling cancer progression. Such work will lead to comprehensive knowledge of the metabolic interactions between cancer cells and the surrounding tumor niche.

Overall, Auciello and colleagues present a new paracrine pathway in the PDAC microenvironment, revealing a previously unappreciated metabolic vulnerability for this cancer. This work strengthens the hypothesis that although complete ablation of the PDAC stroma may be counterproductive, the selective targeting of tumor-promoting stromal pathways remains an attractive option. Moving forward, it will be important to assess the effects of long-term inhibition of this cross-talk in order to prevent metabolic rewiring and design effective combination therapies.

D.A. Tuveson is a member of the scientific advisory board at Surface Oncology and Leap Therapeutics, reports receiving commercial research grants from ONO and Fibrogen, and has ownership interest (including stock, patents, etc.) in Leap Therapeutics and Surface Oncology. No potential conflicts of interest were disclosed by the other author.

G. Biffi has been supported by the Human Frontiers Science Program (LT000195/2015-L) and EMBO (ALTF 1203-2014). D.A. Tuveson is supported by the Lustgarten Foundation, the Cold Spring Harbor Laboratory and Northwell Health Affiliation, and the NIH (5P30CA45508, 5P50CA101955, P20CA192996, U10CA180944, U01CA210240, U01CA224013, 1R01CA188134, and 1R01CA190092).