The stroma of pancreatic ductal adenocarcinoma (PDA) forms a major barrier to therapy and immune surveillance. Elahi-Gedwillo and colleagues demonstrate that halofuginone has potent antifibrotic activity in PDA by directly inhibiting the activation of pancreatic stellate cells, thereby reducing the deposition of extracellular matrix components, including collagen and hyaluronic acid. As a result, halofuginone improves drug delivery and the infiltration of favorable immune cells such as immune-stimulatory myeloid cells and cytotoxic T cells. Despite recent controversies regarding targeting stroma in PDA, this study highlights that modifying the stroma of PDA remains an attractive strategy to improve the efficacy of therapy.

See related article by Elahi-Gedwillo et al., p. 372

Pancreatic ductal adenocarcinoma (PDA) has a dismal prognosis and is projected to become the second leading cause of cancer-related deaths in the United States by the year 2030. This unfortunate outcome is mainly due to the aggressive behavior of cancer cells in the unique tumor microenvironment of PDA. PDA is characterized by a desmoplastic reaction that results in a dense stroma, consisting of abundant extracellular matrix (ECM) and stromal cells, including cancer-associated fibroblasts (CAF), immune cells, and endothelial cells (1). The epithelial/tumor cell component of PDA typically constitutes only 10% to 20% of the tumor volume, with the remaining volume comprised of stromal cells and ECM. This complex and remodeling microenvironment contributes substantially in promoting PDA development, therapy resistance, and metastasis. Classically, pancreatic stellate cells (PSC) are thought to be the major source of CAFs in PDA. Once activated, PSCs express the myofibroblast protein α-smooth muscle actin (αSMA) and secrete ECM constituents, including collagens, fibronectin, and hyaluronic acid (a.k.a. hyaluronan). Consequently, the stiffness of the tumor is enhanced, and interstitial fluid pressure (IFP) is elevated, factors that combine to blunt effective drug delivery and enhance resistance to treatment. The stroma of PDA is also considered to be a major contributor to immune evasion. To date, PDA is still highly refractory to immune therapies, including immune checkpoint blockade.

Elahi-Gedwillo and colleagues (2) recently reported that halofuginone, an analogue of the natural quinazolinone alkaloid febrifugine, has potent antifibrotic activity in a genetically engineered mouse model (GEMM) of PDA. Halofuginone directly inhibited the activation of PSCs and as a result, the deposition of key ECM molecules, including collagen and hyaluronic acid, was significantly reduced. Because of the disruption of the stroma, IFP was markedly reduced, resulting in improved delivery of small-molecule therapy, which the authors demonstrated by exploiting the autofluorescence of doxorubicin. Consistent with a disruption in physical barriers, halofuginone promoted a compelling change in the tumor immune landscape. Within 24 hours of treatment, immune cell infiltration in general increased >3-fold compared with control-treated tumors, with the most noticeable changes being in CD11b+ and iNOS+ myeloid cells, which were increased >5-fold and >10-fold respectively. Importantly, CD11b+ and iNOS+ cells remained elevated after 10 days of therapy, and the distribution of these myeloid cells shifted so that they were in closer proximity to cytokeratin+ tumor cells. CD8+ T cells were also significantly elevated after halofuginone therapy. These observations suggest that the alteration of ECM and antifibrotic therapy might be effective in increasing the efficacy of immunotherapy in immunologically cold, desmoplastic tumors, such as PDA. The study from Elahi-Gedwillo and colleagues (2) also highlights the importance and utility of GEMMs of PDA in the evaluation of potential new therapeutic strategies, especially those that target stroma. The evaluation of the effect of novel strategies on the tumor microenvironment in immunocompetent, spontaneous tumors is likely to be more predictive of potential microenvironmental changes observed in clinical studies.

Although the mechanism of halofuginone is not fully understood, this study highlights the potential of targeting the stromal component of PDA, an area that while still therapeutically attractive is complex and has seen challenges in recent years. For example, the inhibition of Hedgehog signaling was initially thought to be an effective method to limit fibroblast activation and reduce stromal accumulation in PDA (3). However, the efficacy of Hedgehog inhibitors in patients with metastatic PDA was not successful, with at least one trial suggesting that combining Hedgehog inhibition with chemotherapy can result in worse outcomes than chemotherapy alone (4). Additional challenges have resulted from preclinical studies. The genetic deletion of Sonic Hedgehog (Shh) in a GEMM of PDA resulted in the expected reduction of αSMA-positive CAFs and stroma formation, but the loss of Shh induced a more aggressive and undifferentiated tumor phenotype with shorter animal survival (5). While in another study, genetic deletion of αSMA-positive CAFs in a GEMM of PDA resulted in decreased survival with an invasive and undifferentiated tumor phenotype, characterized by epithelial-to-mesenchymal transition (EMT) and cancer stemness, as well as therapy resistance and increased tumor immune evasion (6). The results of Gedwillo and colleagues (2), taken in the context of these recent challenges to stromal targeting, emphasize the complexity of CAF and stroma biology in PDA and underscore the need for caution in developing stroma-targeting therapy.

Evidence of the existence of different subtypes of CAFs is emerging due to the distinct expression patterns of multiple fibroblast markers in PDA and the characterization of two CAF populations using a GEMM-based organoid coculture system (7). Of the two CAF populations, one appears to be inflammatory, characterized by the secretion of a variety of inflammatory chemokines, including IL6, whereas the other population is characterized by αSMA expression and typically considered a myofibroblast. The data presented thus far suggest that the two CAF populations are mutually exclusive and reversible. How these CAF populations contribute to the PDA microenvironment and immune evasion is currently being evaluated intensely by multiple groups. For example, the targeted inhibition of IL6, presumably produced by the inflammatory CAF population, was shown to enhance the efficacy of anti–PD-L1 in PDA. These data in aggregate suggest that the inflammatory CAF population might be a more relevant target, while the αSMA-positive CAF population is regulated by Hedgehog signaling but has a tumor-suppressive function.

An alternative to targeting PDA stroma by inactivating or depleting a certain population of CAFs might be to alter or target the products these CAFs produce. For example, enzymatically digesting hyaluronic acid by PEGPH20 has been shown to decrease IFP and improve drug delivery in PDA, and as a result, improve the efficacy of chemotherapy and immunotherapy preclinically (8). Clinical studies testing the efficacy of combining PEGPH20 with chemotherapy ± immune checkpoint blockade in multiple cancers is ongoing. An additional attractive CAF product is collagen. PDA is rich in fibrillar collagens, which support tumor cell survival and tumor progression and which signal through discoidin domain receptor 1 and 2 (DDR1 & 2) expressed on PDA tumor cells and stromal cells. Small-molecule–mediated selective inhibition of DDR1 reduces collagen deposition, decreases epithelial plasticity, and improves the response to chemotherapy in GEMMs of PDA (9). Preclinical studies to investigate the effect of DDR1 inhibition on the immune landscape of PDA are underway. Thus, it appears that although the biology of targeting stroma in PDA is complex, there are exciting opportunities that might enhance standard therapy and also kickstart the use of immunotherapy in patients with PDA. Furthermore, the therapeutic benefit of halofuginone might be due to the fact that it reduces collagen and hyaluronic acid in PDA.

A potential challenge of these approaches is that halofuginone also inhibits the activation of αSMA-positive CAFs. Thus far, the tumor-promoting effects of other strategies that target αSMA CAFs have not been observed with halofuginone therapy and the reason for the difference between the inhibitors is not clear. We suggest that future studies should investigate the effect of halofuginone on the inflammatory CAF population. The drug induces significant infiltration of immune-stimulatory macrophages and cytotoxic T cells, which could result from the inhibition of immune-suppressive chemokine secretion from inflammatory CAFs as well as a reduction in physical barriers. For example, another important molecule that is impacted by halofuginone is TGFβ, a potent inducer of desmoplasia, EMT, and immune suppression. Halofuginone can block TGFβ signaling by inhibiting Smad2 and Smad3 phosphorylation and also by reducing the expression of TGFβ receptor 2. Thus, TGFβ inhibition likely contributes to the consequences of halofuginone therapy. However, targeting TGFβ in PDA is complex in its own right. Canonical TGFβ signaling is tumor-suppressive in a context-dependent manner such that genetic and pharmacologic inhibition of TGFβ in PDA has resulted in reduced survival in GEMMs of PDA and other models (10). Thus, excitement regarding the development and use of halofuginone as a therapeutic strategy for PDA should be tempered with caution and rigor.

Stroma-targeting therapy is alive and well and remains a promising strategy to overcome therapy resistance in PDA. The existence of CAF heterogeneity suggests that stroma subtyping might be useful to stratify patients to achieve the most therapeutic benefit from stroma-targeting therapy. This is already being done in a sense in the clinical studies of PEGPH20, where patients with high levels of hyaluronic acid are chosen for therapy. Understanding whether particular subtypes of PDA might be more susceptible to halofuginone or other stroma-targeting strategies would be a significant step forward. The concept that modifying the stroma of PDA as a means to improve immunotherapy in this tumor type is important and the results with PEGPH20, and now halofuginone, are cause for cautious optimism.

R.A. Brekken is a co-founder of Tuevol Therapeutics, a company developing inhibitors of DDR1. No potential conflicts of interest were disclosed the other author.

1.
Neesse
A
,
Michl
P
,
Frese
KK
,
Feig
C
,
Cook
N
,
Jacobetz
MA
, et al
Stromal biology and therapy in pancreatic cancer
.
Gut
2011
;
60
:
861
8
.
2.
Elahi-Gedwillo
KY
,
Carlson
M
,
Zettervall
J
,
Provenzano
PP
. 
Antifibrotic therapy disrupts stromal barriers and modulates the immune landscape in pancreatic ductal adenocarcinoma
.
Cancer Res
2019
;
79
:
372
86
.
3.
Olive
KP
,
Jacobetz
MA
,
Davidson
CJ
,
Gopinathan
A
,
McIntyre
D
,
Honess
D
, et al
Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer
.
Science
2009
;
324
:
1457
61
.
4.
Catenacci
DV
,
Junttila
MR
,
Karrison
T
,
Bahary
N
,
Horiba
MN
,
Nattam
SR
, et al
Randomized phase Ib/II study of gemcitabine plus placebo or vismodegib, a Hedgehog pathway inhibitor, in patients with metastatic pancreatic cancer
.
J Clin Oncol
2015
;
33
:
4284
92
.
5.
Rhim
AD
,
Oberstein
PE
,
Thomas
DH
,
Mirek
ET
,
Palermo
CF
,
Sastra
SA
, et al
Stromal elements act to restrain, rather than support, pancreatic ductal adenocarcinoma
.
Cancer Cell
2014
;
25
:
735
47
.
6.
Ozdemir
BC
,
Pentcheva-Hoang
T
,
Carstens
JL
,
Zheng
X
,
Wu
CC
,
Simpson
TR
, et al
Depletion of carcinoma-associated fibroblasts and fibrosis induces immunosuppression and accelerates pancreas cancer with reduced survival
.
Cancer Cell
2014
;
25
:
719
34
.
7.
Ohlund
D
,
Handly-Santana
A
,
Biffi
G
,
Elyada
E
,
Almeida
AS
,
Ponz-Sarvise
M
, et al
Distinct populations of inflammatory fibroblasts and myofibroblasts in pancreatic cancer
.
J Exp Med
2017
;
214
:
579
96
.
8.
Wong
KM
,
Horton
KJ
,
Coveler
AL
,
Hingorani
SR
,
Harris
WP
. 
Targeting the tumor stroma: the biology and clinical development of pegylated recombinant human hyaluronidase (PEGPH20)
.
Curr Oncol Rep
2017
;
19
:
47
.
9.
Aguilera
KY
,
Huang
H
,
Du
W
,
Hagopian
MM
,
Wang
Z
,
Hinz
S
, et al
Inhibition of discoidin domain receptor 1 reduces collagen-mediated tumorigenicity in pancreatic ductal adenocarcinoma
.
Mol Cancer Ther
2017
;
16
:
2473
85
.
10.
Hezel
AF
,
Deshpande
V
,
Zimmerman
SM
,
Contino
G
,
Alagesan
B
,
O'Dell
MR
, et al
TGF-beta and alphavbeta6 integrin act in a common pathway to suppress pancreatic cancer progression
.
Cancer Res
2012
;
72
:
4840
5
.