Tamoxifen calms down the distressed PDAC stroma

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest human cancers and is associated with extensive desmoplastic changes in the tumor microenvironment. In this issue of EMBO Reports, two studies by Cortes et al 1, 2 identify the G‐protein‐coupled estrogen receptor (GPER) as an important regulator of the PDAC‐associated stroma, modulating tissue stiffness, hypoxic responses, and desmoplasia. Intriguingly, the authors find that tamoxifen, which is widely used for its antagonizing effect on nuclear estrogen receptor (ER)‐positive breast cancers, acts as GPER agonist to normalize the PDAC microenvironment. The two studies thus open up new opportunities to explore tamoxifen as potential anti‐stromal therapy in PDAC.

For the past decades, the nonsteroidal drug tamoxifen has been standard of care as endocrine therapy against breast cancer. Originally developed in the 1960s as a contraceptive that eventually failed, tamoxifen was found to be effective for the treatment of luminal breast cancers and has subsequently benefited millions of breast cancer patients 3. Tamoxifen antagonizes the nuclear estrogen receptors (ER) ERα and ERβ. However, it can also act as agonist of the G‐protein‐coupled estrogen receptor (GPER), which is expressed by many normal and malignant cells, and is commonly localized at intracellular membranes such as the endoplasmic reticulum 4. Thus, tamoxifen may exert additional effects on a number of cell types that do not express ER. However, the biological functions mediated by GPER signaling are generally poorly understood.

The tumor microenvironment is widely recognized as an important determinant of cancer development and progression. In pancreatic ductal adenocarcinoma (PDAC), one of the deadliest malignancies in humans, the stromal contribution is substantial and desmoplasia is a prevailing feature 5. Desmoplasia is characterized by excessive deposition of collagen and other extracellular matrix (ECM) proteins. In addition, matrix crosslinking and increased contractility are observed, caused by activated pancreatic stellate cells (PSCs), the main drivers of the desmoplastic reaction 5. Activated PSCs exhibit a contractile myofibroblastic phenotype, inducing a significant stiffening of the ECM. The stiff matrix impedes drug delivery to the tumor, which is further hindered by a dysfunctional vasculature that generates a hypoxic environment commonly observed in PDAC 5, 6. However, despite emerging appreciation for the importance of desmoplastic and hypoxic responses as contributors to PDAC, these functions have thus far been difficult to target.

In this issue of EMBO Reports, Cortes et al 1, 2 present two complementing studies that unveil tamoxifen–GPER axis as an essential regulator of the PDAC microenvironment (summarized in Fig 1). They show that tamoxifen has pleiotropic effects, targeting PSCs, macrophages, and cancer cells within tumors of KPC mice (Kras^G12D/+; p53^R172H/+; Pdx‐1‐Cre), an established model of PDAC. Tamoxifen treatment of tumor‐bearing KPC mice reduces signs of fibrosis, such as deposition of fibrillar collagen and fibronectin that form thick fibers in the ECM of desmoplastic tissues. The authors find that tamoxifen not only represses collagen and fibronectin deposition, but also alters fiber alignment that likely contributes to reduced tissue stiffness 1, 2. In line with these changes, alpha smooth muscle actin (α‐SMA) expression is reduced in tamoxifen‐treated KPC mice, indicating impaired activation of PSCs 1. Using magnetic bead displacement, gel contraction, and matrix remodeling assays, the authors find that tamoxifen indeed inhibits the myofibroblastic differentiation of PSCs, and that this is GPER dependent 1. Tamoxifen‐treated PSCs lose their ability of actomyosin contraction by reduced activation of the mechanotransducers RhoA and myosin light chain 2 (MLC‐2). As potential regulators of matrix architecture, the authors identify tamoxifen‐dependent repression of ECM‐modifying enzymes such as lysyl oxidase homolog 2 (LOXL2) and other LOX family proteins as well as matrix‐metalloproteinase 2 (MMP‐2) in PSCs 2. These enzymes have been shown to modify collagen and to play an important functional role in cancer progression 7. Importantly, these matrix changes have a significant effect on other cell types within pancreatic tumors. For example, invasion of pancreatic cancer cells into decellularized matrices from tamoxifen‐treated PSCs is restrained. Moreover, reduced ECM stiffness limits macrophage attachment and cell spreading in vitro, which is in line with decreased macrophage infiltration in KPC mouse tumors upon tamoxifen treatment 1. Interestingly, reduced levels of CD204, a marker of M2 macrophage polarization, are also observed, indicating decreased infiltration of immunosuppressive macrophages that may affect tumor progression 5. In addition, tamoxifen can affect macrophages directly via GPER, reducing macrophage attachment, spreading and invasion in vitro 1.

Figure 1. PDAC desmoplasia is suppressed by tamoxifen modulation of the tumor stroma.

(A) PDAC is characterized by a pronounced fibrotic stroma termed desmoplasia. The altered stroma promotes infiltration of macrophages and invasion of cancer cells. Tamoxifen treatment inhibits the function of reactive stellate cells, infiltrating macrophages, including M2 macrophages, and the cancer cells themselves in the tumors. This leads to reduced cancer cell invasion and proliferation as well as increased apoptosis. Tissues become more vascularized, potentially improving drug delivery to tumors, and hypoxic responses are reduced. (B) Pancreatic stellate cells are the key cell type involved in the fibrotic response and regulate tumor biomechanical properties via contraction and ECM modulation. Tamoxifen acts as agonist of GPER in pancreatic stellate cells, repressing genes that drive contraction and ECM remodeling in PDAC.

Using proteomic profiling of pancreatic tumor tissues from mice and RNA sequencing of tamoxifen‐treated PSCs, the authors observe a reduction in hypoxia response genes and an induction of genes regulating blood vessel morphogenesis 2. These observations are consistent with an increase in vascularization and reduction in hypoxic responses in tamoxifen‐treated KPC mice. Indeed, hypoxia‐inducible factor 1 alpha (HIF‐1α), a key mediator of hypoxic responses that regulates LOXL2 and MMP‐2 and which is frequently induced in pancreatic cancer, is repressed by tamoxifen. However, tamoxifen‐mediated HIF‐1α repression is not solely due to changes in oxygen levels, but can also be regulated mechanically by myosin‐dependent GPER signaling in PSCs 2. Hypoxia‐independent regulation of HIF‐1α signaling observed in PSCs also occurs in pancreatic cancer cells, where tamoxifen treatment decreases HIF‐1α, promoting apoptosis and impairing cancer cell proliferation.

Taken together, the studies by Cortes et al identify tamoxifen‐activated GPER as a key regulator in the PDAC microenvironment and reveal an interesting link between tissue mechanics—as part of the desmoplastic PDAC stroma—and adaptive survival responses to hypoxia. The results pave the way for future investigations on tamoxifen as potential anti‐stromal therapy for PDAC. LOX/hypoxia responses and tissue stiffening play an important role in pancreatic cancer progression in mouse models and associate with poor clinical outcome in PDAC patients 8, 9. Further functional studies in mice will be useful to shed light on the role of tamoxifen‐mediated stromal normalization in cancer progression and metastasis. This is especially important in the light of evidence suggesting that components of PDAC stroma may restrain cancer progression 5. Clinically, pancreatic cancer is difficult to manage and responds poorly to therapy, which is at least in part due to the desmoplastic nature of the tumors and poor drug delivery. Thus, combination treatment with tamoxifen and standard of care therapies against pancreatic cancer, such as FOLFIRINOX or gemcitabine, will be interesting to study in preclinical models. It will also be valuable to explore whether the effect of tamoxifen on PSCs is transferrable to fibroblasts within other solid cancer types that exhibit tissue fibrosis. Interestingly, this may be the case in breast cancer models, where tamoxifen treatment leads to substantial changes in matrix composition 10. However, it is important to note that due to the high doses of tamoxifen applied in the presented work, the authors acknowledge the need for additional inquiry using clinically relevant doses of tamoxifen. The work of Cortes et al underscores the importance of careful investigation of pleiotropic tissue responses to cancer drugs, even for those drugs that are considered targeted therapies. We are beginning to uncover the manifold effects of several drugs that are currently used in the clinic, yet our understanding of the mode of action of some of these drugs is still rudimentary. Thus, dissecting the complex and often broad effects of cancer drugs, as presented in the two studies, may reveal novel opportunities to expand usage of therapeutic compounds and improve treatment efficacy with combination therapy.

Conflict of interest

The authors declare that they have no conflict of interest.

EMBO Reports (2019) 20: e47334