128 Experimental Oncology 32, 128–131, 2010 (September)

Components of the pancreatic cancer micro-

environment. The activated stroma of pancreatic

cancer comprises several different cell types including

stellate cells, endothelial cells, nerve cells, immune

cells such as macrophages, lymphocytes, dendritic

cells and the extracellular matrix.

Pancreatic stellate cells. In 1998 Bachem and Apte

isolated and cultured pancreatic stellate cells (PSC)

[1, 2]. Morphologic, functional and gene expression

studies revealed that PSC resemble HSC characteristics

and therefore may possibly share a common origin [3].

However, the origin of stellate cells is still controversially

debated. Mesenchymal, endodermal as well as neuroec-

todermal origins are suggested. Further, it is postulated

that in the diseased organ, stellate cells are transformed

from their quiescent precursors, or recruited from local

fibroblasts, bone marrow derived cells or generated via

epithelial-mesenchymal transformation [3]. Pancreatic

stellate cells (PSC) are myofibroblast-like cells found

in the periacinar spaces in the normal pancreas and have

long cytoplasmic processes that encircle the base

of the acinus like the pericytes around the breast acini [2].

In their quiescent state, pancreatic stellate cells com-

prise approximately 4% of the pancreatic cell popula-

tion. They can also be found in perivascular, periductal

and periacinar regions of the pancreas and serve as key

participants in the pathobiology of the major disorders

of the exocrine pancreas, i.e., chronic pancreatitis and

pancreatic cancer [1, 2, 4, 5]. In these disorders, PSC

participate in disease pathogenesis after transforming

from a quiescent state into an «activated» state, also

known as a «myofibroblastic» state. This activation

process is accompanied by a loss of the characteristic

retinoid containing fat droplets in their cytoplasm with

a concomitant expression of alpha smooth muscle ac-

tin [1, 2] (Fig. 1a, b). Although in acute pancreatitis PSC

activity is transient, their persistent activity in chronic

inflammation and pancreatic ductal adenocarcinoma

(PDAC) can impair organ function due to their excessive

contraction and abundant extracellular matrix protein

deposition especially in the periacinar spaces [5]. It is

also becoming clearer that myofibroblast like cells found

in the activated stroma of pancreatic cancer significantly

impact tumor behavior [6].

Endothelial cells / neoangiogenesis. In in vi-

tro- and animal-experiments, pancreatic cancer cells

are shown to induce angiogenesis by producing an-

giogenic substances like vascular endothelial growth

factor (VEGF), fibroblast growth factor (FGF) and IL-8

[7–10]. Nevertheless, in vivo, pancreatic cancer is a

hypoxic and hypovascular tumor (Fig. 2a, b) with an

80% reduction of the microvascular density compared

to the normal pancreas [5, 11, 12]. This discrepancy

between experimental data and the clinical reality

results mostly from the inefficiency of our current ex-

perimental setups in recapitulating the tumor microen-

vironment. First, pancreatic cancer cells produce also

potent anti-angiogenic substances like angiostatin,

endostatin and thrombospondin-1, which in fact ren-

ders them dominantly antiangiogenic [5, 13]. Second,

when cancer cells interact with pro-angiogenic PSC,

the dominant effect of this «system» -resembling

pancreatic cancer microenvironment- remains anti-

angiogenic. This cumulative antiangiogenic effect is

due to increased production of endostatin cleaved

from collagen 18 (produced by cancer cells) by matrix

metalloproteinase-12 (MMP-12) secreted by cancer

and stellate cells [5].

TUMOR MICROENVIRONMENT AND PROGRESSION

OF PANCREATIC CANCER

M. Erkan, C. Reiser-Erkan, C.W. Michalski, J. Kleeff

Department of Surgery, Technische Universität München, Ismaningerstrasse 22, Munich 81675, Germany

Pancreatic ductal adenocarcinoma is characterized by «tumor desmoplasia», a remarkable increase in connective tissue that pen-

etrates and envelopes the neoplasm. It is becoming clear that this desmoplastic microenvironment of pancreatic cancer — which

is forming approximately eighty percent of the tumor mass — is not a passive scaffold for the tumor cells but an active player in

carcinogenesis. Several chemotherapeutic agents and novel molecular targeted therapies against epithelial tumor cells — although

showing antitumor activity in cell culture and mouse experiments — have failed to show significant effects in the clinic. Thus,

targeting pancreatic tumor cells alone seems unlikely to improve the dismal prognosis of pancreatic cancer. It has recently been

shown that the activated stroma of pancreatic cancer is an independent prognostic marker with an impact on patient survival as

much as the lymph node status of the cancer. Several primarily benign conditions associated with expansion of stromal and inflam-

matory components, such as chronic pancreatitis or hereditary pancreatitis are believed to increase the risk of pancreatic cancer.

Similar observations have been made in other cancer types such as chronic hepatitis-liver cancer, Barrett dyplasia-esophageal

cancer, and inflammatory bowel disease-colon cancer. The common denominator of all these conditions is; chronic inflammation

leads to increased incidence of cancer. In this review the impact of the activated stroma on pancreatic carcinogenesis is discussed.

Key Words: microenvironment, pancreatic cancer, extracellular matrix.

Received: July 20, 2010.

*Correspondence: Fax: +49 (894) 140-48-70;

E-mail: erkan@chir.med.tu-muenchen.de

Abbreviations used: ECM — extracellular matrix; FGF — fibroblast

growth factor; MMP — matrix metalloproteinase; PDAC — pancreatic

ductal adenocarcinoma; PSC — pancreatic stellate cells; VEGF — vas-

cular endothelial growth factor.

Exp Oncol 2010

32, 3, 128–131

Experimental Oncology 32, 128–131, 2010 (September) 129

a

b

Fig. 1. Pancreatic stellate cells: pancreatic stellate cells are acti-

vated myofibroblasts that typically express alpha-smooth muscle

actin (a, immunofluorescence analysis shows the intracellular

filaments). Pancreatic cancer cells activate the stellate cells

around them (b, immunohsitochemistry staining of stellate cells

by alpha smooth muscle actin).

Nerve cells / Pancreatic neuropathy / Intrapan-

creatic perineural invasion. The pancreas has an

abundant nerve supply composed of various myelinated

and unmyelinated nerve fibers and aggregates of neu-

ral cell bodies known as intrapancreatic ganglia [14].

The autonomic nervous system regulates the secretory

functions of the pancreas, constriction and relaxation

of the blood vessels and excretory ducts. The pancreas

has a sensory nerve supply that is involved in signal trans-

mission to the central nervous system [14]. The nerves

are usually involved when pathologic changes occur

in an organ. Although the nerve fiber density of the pan-

creas decreases due to periacinar fibrosis, the number

of pathologically enlarged nerves increase [15]. Pain is

reported by 75–80% of patients at the initial evaluation.

The pain in pancreatic cancer is believed to be a kind of

neuropathic pain. Neuropathic pain is caused by injury

of peripheral or visceral nerves, rather than stimulation

of pain receptors and is a result of nerve compression

by tumor, or direct infiltration of nerves by cancer- or

inflammatory cells [15–19]. Interestingly, the size of nerve

fibers is increased within the areas of pancreatic fibrosis

that characterizes chronic pancreatitis, which further

supports the theory that stromal changes within the pan-

creas may precede the development of malignancy [17].

Logically, intrapancreatic nerve invasion should precede

extrapancreatic perineural invasion, which causes pain

and precludes curative resection. Nonetheless, con-

sidering the high frequency of intrapancreatic nerves

(Fig. 3a, b), if analyzed in detail, it should be impossible to

find any case of PDAC without intrapancreatic perineural

invasion, therefore the prognostic value of this finding is

highly questionable.

a

b

Fig. 2. Pancreatic stellate cell overactivity leads to the fibrotic

microenvironment of the cancer. The fine periacinar capillary

network of the normal pancreas (a, CD31 staining of the endo-

thelial cells) is lost in pancreatic cancer. Mason trichrom staining

of the pancreatic cancer (b) depicts the collagen-rich (blue) mi-

croenvironment of the tumor with notable scarcity of the vessels.

Inflammatory cells. A relationship between in-

flammation and cancer was hypothesized by Rudolph

Virchow back in the 1850s [20]. Epidemiological, clinical

and experimental evidence has then contributed to sup-

port and further elucidate the functional role of inflamma-

tory cells in tumor progression. Accordingly, malignant

tumors are considered as «wounds that do not heal» [21],

where microenvironment-derived growth promoting

factors sustain the survival and proliferation of initiated

cells. All components of the innate (e.g., macrophages,

dendritic cells, mast cells, granulocytes) and adaptive

immunity (B- and T-lymphocytes) are present in the tu-

mor microenvironment, although in different proportions

and with different — and sometimes probably opposite

functions [22]. Cells of the innate immunity are recruited

to tumor sites by growth factors and chemokines that are

often produced by the cancer cell themselves, including

colonystimulating factor-1 (CSF-1), CC chemokines

(CCL2, CCL3, CCL4, CCL5, CCL8), transforming growth

130 Experimental Oncology 32, 128–131, 2010 (September)

factor ß-1 (TGFß-1), vascular endothelial growth factor.

In return, inflammatory cells produce mediators that

contribute to cancer growth, invasion and metasta-

sis [23]. Tumor-associated macrophages exhibit a dis-

tinct phenotype that share many characteristics with

M2-polarized macrophages and influence every step of

tumor progression [24]. Inflammatory cells are part of the

stromal reaction that characterizes pancreatic cancer,

a fact that highlights the close relation between a chronic

inflammatory condition and this tumor [25]. Indeed,

similarities between the stroma composition in chronic

pancreatitis and pancreatic cancer further emphasize the

pathogenetic link between them. Macrophages, mast

cells, neutrophils, dendritic cells, B- and T-lymphocytes

have all been described in the stroma of pancreatic

cancer [26, 27]. Interestingly, the number of mast cells

and macrophages showed a positive correlation with

the microvessel count and mast cell/macrophage-rich

tumors with high microvessel density displayed a ten-

dency for a worse prognosis [26]. In addition to their

role in promoting an angiogenic phenotype, mast cells

are also probably involved in the establishment of the

stromal reaction in primary and metastatic pancreatic

cancer. An invasive front of neutrophil granulocytes has

been reported at the invasive edge of pancreatic ductal

adenocarcinoma, where they produce TGFβ-1, which

further affects the stromal reaction that accompanies

pancreatic cancer [28]. Taken together, recent evidence

supports an active protumorigenic role of inflamma-

tory cells in the development and progression of human

pancreatic cancer.

Extracellular matrix. The extracellular matrix

(ECM) exerts both pro- and anti-tumorigenic effects

in pancreatic cancer. Although a fibrotic matrix was ini-

tially regarded as a host barrier against tumor invasion,

it has become evident that it can modulate and even

initiate tumorigenesis[4, 6, 29, 30]. In vitro, extracel-

lular matrix proteins influence growth, diffe rentiation,

survival and motility of cancer cells both by providing

a physical scaffold and by acting as a reservoir for

soluble mitogens. Cancer cells are believed to exploit

this tumor-supportive microenvironment [4, 29].

The replacement of the normal parenchyma by exces-

sive desmoplastic tissue rich in collagen, fibronectin

and periostin ostensibly plays a role in the aggressive

behavior of PDAC. The initial stellate cell activity ap-

parently starts at the periacinar spaces on the invading

front of the activated stroma [5]. Therefore, on the

invasive front of the activated stroma, continuous acti-

vation of PSCs and the subsequent fibrosis of the peri-

acinar spaces may physically cause tissue hypoxia and

parenchymal loss by hindering the blood circulation,

oxygen diffusion and damaging fine innervation. Nor-

mally, the collagen type IV-rich basement membrane,

which separates epithelial cells from the interstitial

matrix, exerts inhibitory effects on both cancer cells

and stellate cells. In various tumors, including PDAC,

breaching of the basement membrane is a critical step

in tumor progression. This step brings malignant cells

into direct contact with ECM proteins such as fibrillary

collagens, supporting their growth and contributing to

their chemo-resistance [4, 29].

a

b

Fig. 3. Replacement of the fine periacinar nerve fibers of the

normal pancreas with pathologic nerves in pancreatic cancer.

Similar to the capillary network, periacinar spaces of the normal

pancreas harbour fine nerve fibers (a, GAP-43 staining without

counterstaining). In the fibrotic parts of the pancreas the number

of the pathologically enlarged nerves increase (b, GAP-43 stain-

ing without counterstaining).

On the other hand the amount of collagen deposited

in the stroma of pancreatic cancer is a positive prog-

nostic factor supporting the idea that a fibrotic matrix is

also a host barrier against tumor invasion [6]. To better

understand which components of the ECM has pro- and

anti-tumorigenic effects further studies are warranted.

CONCLUSION

Activated stroma of the pancreatic cancer compris-

ing activated stellate cells, inflammatory cells impacts

on carcinogenesis significantly. Therefore, targeting

the activated stroma in order to uncouple epithelial-

stromal interactions may interrupt multiple aberrant

autocrine and paracrine pathways that promote

pancreatic cancer cell growth, invasion, metastasis,

and angiogenesis.

REFERENCES

1. Apte MV, Haber PS, Applegate TL, et al. Periacinar

stellate shaped cells in rat pancreas: identification, isolation,

and culture. Gut 1998; 43: 128–33.

2. Bachem MG, Schneider E, Gross H, et al. Identification,

culture, and characterization of pancreatic stellate cells in rats

and humans. Gastroenterology 1998; 115: 421–32.

Experimental Oncology 32, 128–131, 2010 (September) 131

3. Erkan M, Weis N, Pan Z, et al. Organ-, inflammation-

and cancer specific transcriptional fingerprints of pancreatic

and hepatic stellate cells. Mol Cancer 2010; 9: 88.

4. Erkan M, Kleeff J, Gorbachevski A, et al. Periostin cre-

ates a tumor-supportive microenvironment in the pancreas

by sustaining fibrogenic stellate cell activity. Gastroenterology

2007; 132: 1447–64.

5. Erkan M, Reiser-Erkan C, Michalski CW, et al. Cancer-

stellate cell interactions perpetuate the hypoxia-fibrosis cycle in

pancreatic ductal adenocarcinoma. Neoplasia 2009; 11: 497–508.

6. Erkan M, Michalski CW, Rieder S, et al. The activated

stroma index is a novel and independent prognostic marker

in pancreatic ductal adenocarcinoma. Clin Gastroenterol

Hepatol 2008; 6: 1155–61.

7. Brown LF, Berse B, Jackman RW, et al. Expression of

vascular permeability factor (vascular endothelial growth fac-

tor) and its receptors in adenocarcinomas of the gastrointestinal

tract. Cancer Res 1993; 53: 4727–35.

8. Compagni A, Wilgenbus P, Impagnatiello MA, et al.

Fibroblast growth factors are required for efficient tumor

angiogenesis. Cancer Res 2000; 60: 7163–9.

9. Matsuo Y, Ochi N, Sawai H, et al. CXCL8/IL-8 and

CXCL12/SDF-1alpha co-operatively promote invasiveness and

angiogenesis in pancreatic cancer. Int J Cancer 2009; 124: 853–61.

10. Carmeliet P. Angiogenesis in health and disease. Nat

Med 2003; 9: 653–60.

11. Koong AC, Mehta VK, Le QT, et al. Pancreatic tumors

show high levels of hypoxia. Int J Radiat Oncol Biol Phys

2000; 48: 919–22.

12. Reiser-Erkan C, Erkan M, Pan Z, et al. Hypoxia-induc-

ible proto-oncogene Pim-1 is a prognostic marker in pancreatic

ductal adenocarcinoma. Cancer Biol Ther 2008; 7: 1352–9.

13. Schwarte-Waldhoff I, Volpert OV, Bouck NP, et al.

Smad4/DPC4-mediated tumor suppression through suppression

of angiogenesis. Proc Natl Acad Sci U S A 2000; 97: 9624–9.

14. Salvioli B, Bovara M, Barbara G, et al. Neurology and

neuropathology of the pancreatic innervation. Jop 2002; 3: 26–33.

15. Ceyhan GO, Schafer KH, Kerscher AG, et al. Nerve

growth factor and artemin are paracrine mediators of pan-

creatic neuropathy in pancreatic adenocarcinoma. Ann Surg

2010; 251: 923–31.

16. Farrar JT, Portenoy RK. Neuropathic cancer pain: the

role of adjuvant analgesics. Oncology (Williston Park) 2001;

15: 1435-42, 1445; discussion 1445, 1450–3.

17. Ceyhan GO, Bergmann F, Kadihasanoglu M, et al. The

neurotrophic factor artemin influences the extent of neural dam-

age and growth in chronic pancreatitis. Gut 2007; 56: 534–44.

18. Ceyhan GO, Deucker S, Demir IE, et al. Neural fractal-

kine expression is closely linked to pain and pancreatic neuritis

in human chronic pancreatitis. Lab Invest 2009; 89: 347–61.

19. Michalski CW, Oti FE, Erkan M, et al. Cannabinoids

in pancreatic cancer: correlation with survival and pain.

Int J Cancer 2008; 122: 742–50.

20. Balkwill F, Mantovani A. Inflammation and cancer:

back to Virchow? Lancet 2001; 357: 539–45.

21. Dvorak HF. Tumors: wounds that do not heal. Simi-

larities between tumor stroma generation and wound healing.

N Engl J Med 1986; 315: 1650-9.

22. de Visser KE, Eichten A, Coussens LM. Paradoxical

roles of the immune system during cancer development. Nat

Rev Cancer 2006; 6: 24–37.

23. Pollard JW. Tumour-educated macrophages promote tu-

mour progression and metastasis. Nat Rev Cancer 2004; 4: 71–8.

24. Mantovani A, Sozzani S, Locati M, et al. Macrophage

polarization: tumor-associated macrophages as a paradigm

for polarized M2 mononuclear phagocytes. Trends Immunol

2002; 23: 549–55.

25. Michalski CW, Maier M, Erkan M, et al. Cannabinoids

reduce markers of inflammation and fibrosis in pancreatic

stellate cells. PLoS One 2008; 3: e1701.

26. Esposito I, Menicagli M, Funel N, et al. Inflammatory

cells contribute to the generation of an angiogenic phenotype

in pancreatic ductal adenocarcinoma. J Clin Pathol 2004; 57:

630–6.

27. Michalski CW, Gorbachevski A, Erkan M, et al. Mono-

nuclear cells modulate the activity of pancreatic stellate cells

which in turn promote fibrosis and inflammation in chronic

pancreatitis. J Transl Med 2007; 5: 63.

28. Aoyagi Y, Oda T, Kinoshita T, et al. Overexpression of

TGF-beta by infiltrated granulocytes correlates with the ex-

pression of collagen mRNA in pancreatic cancer. Br J Cancer

2004; 91: 1316-26.

29. Bissell MJ, Radisky D. Putting tumours in context.

Nat Rev Cancer 2001; 1: 46–54.

30. Lunevicius R, Nakanishi H, Ito S, et al. Clinicopatho-

logical significance of fibrotic capsule formation around liver

metastasis from colorectal cancer. J Cancer Res Clin Oncol

2001; 127: 193–9.

Copyright © Experimental Oncology, 2010