Role of lymphangiogenesis in cancer.
BIOLOGY OF NEOPLASIA
J Clin Oncol. 2007 Sep
Sundar SS, Ganesan TS.
Department of Gynaecological Oncology, Cheltenham General Hospital,
Gloucestershire Hospitals Foundation Trust, Gloucestershire, United
Kingdom. sudhasun…@minox.demon.co.uk
Regional lymph node metastasis is a common event in solid tumors and
is considered a marker for dissemination, increased stage, and worse
prognosis. Despite rapid advances in tumor biology, the molecular
processes that underpin lymphatic invasion and lymph node metastasis
remain poorly understood. However, exciting discoveries have been made in the field of lymphangiogenesis in recent years. The identification of vascular endothelial growth factor ligands and cognate receptors involved in lymphangiogenesis, an understanding of the embryology of the mammalian lymphatic system, the recent isolation of pure populations of lymphatic endothelial cells, the investigation of lymphatic metastases in animal models, and the identification of markers that discriminate lymphatics from blood vessels at immunohistochemistry are current advances in the field of
lymphangiogenesis, and as such are the main focus of this article.
This review also evaluates evidence for lymphangiogenesis (ie, new
lymphatic vessel formation in cancer) and critically reviews current
data on the prognostic significance of lymphatic vascular density in
tumors. A targeted approach to block pathways of lymphangiogenesis
seems to be an attractive anticancer treatment strategy. Conversely,
promotion of lymphangiogenesis may be a promising approach to the
management of treatment-induced lymphedema in cancer survivors.
Finally, the implications of these developments in cancer therapeutics
and directions for future research are discussed.
INTRODUCTION
The importance of lymphatic metastases is well recognized in cancer
staging and treatment, with lymph node status determining
multimodality treatment in patients with solid tumors such as breast
cancer, colorectal cancer, and head and neck cancer. However, the
process of lymphatic invasion and metastasis to regional lymph nodes,
and whether tumors promote lymphangiogenesis (ie, new lymphatic vessel growth) in a manner similar to angiogenesis, remain poorly understood. Recent advances in the biology and pathology of lymphangiogenesis have provided new insights into the field of lymphatic vascular biology.
ANATOMY AND PHYSIOLOGY OF THE LYMPHATIC SYSTEM
The lymphatic system functions as a conduit for protein-rich fluid
extravasated from the cardiovascular system, has a critical role in
maintaining tissue homeostasis and fat absorption from the gut, and
also plays an important role in immune surveillance.1 It consists of a
hierarchical network of vessels, starting from capillaries formed by
single thin-walled lymphatic endothelial cell (LEC) layers, through to
larger diameter vessels that eventually connect to the venous system.
Unlike blood vessels, lymphatic capillaries lack a continuous basement
membrane and pericytes, demonstrate large infrequent endothelial gaps, and are anchored to the extracellular matrix by elastic fibers (anchor filaments). These features prevent the vessels from collapsing during changes in interstitial pressure and facilitate the uptake of soluble tissue components, even in high-pressure environments.1 The collecting lymphatic vessels contain valves that prevent lymph backflow and have a coating of perivascular smooth muscle cells that allow propulsion of lymph through the vessels.2
EMBRYOLOGY OF THE LYMPHATIC SYSTEM
The earliest event in lymphatic development is the polarized
expression of the homeobox transcription factor Prox-1 in a
subpopulation of endothelial cells in the cardinal vein, which through
budding and sprouting, give rise to the lymphatic system.3 The
vascular phenotype is the default fate of budding embryonic venous
endothelial cells; on expression of Prox-1, cells adopt a lymphatic
vascular phenotype.4 Prox-1 overexpression in human vascular
endothelial cells suppresses blood vessel–specific genes and
upregulates lymphatic endothelial specific cell transcripts.5,6
Homozygous disruption of the Prox gene in mouse embryos is lethal at
embryonic day E14 to E15, with the embryos showing severe chylous
ascites and no lymphatic vasculature.3 Functional inactivation of a
single allele of the homeobox gene Prox1 leads to adult-onset obesity
due to abnormal lymph leakage from mispatterned and ruptured lymphatic vessels; Prox1 heterozygous mice are a model for adult-onset obesity.7.
Vascular endothelial growth factor C (VEGF-C) is an essential
chemotactic and survival factor during lymphangiogenesis and is
required for the sprouting of the first lymphatic vessels from
embryonic veins.8 Homozygous deletion of VEGF-C leads to the complete absence of lymphatic vasculature in mouse embryos; heterozygous VEGF-C mice display severe lymphatic hyperplasia.8 Deletion of VEGF-D does not affect development of lymphatic vasculature, although exogenous VEGF-D rescues the impaired vessel sprouting in VEGF-C–/– embryos.8,9
Vascular endothelial growth factor receptor 3 (VEGFR-3) signaling may
confer lymphatic endothelial-like phenotypes to endothelial cells10;
VEGFR-3 deletion leads to defects in blood vessel remodeling and
embryonic death at mid-gestation, indicating an early blood vascular
function.11 VEGFR-2 signaling is required for the endothelial
differentiation of mouse embryonic stem cells induced by VEGF-C.10
The forkhead transcription factor FOXC2 may control the maturation
stage of lymphatic vascular development and is important in the
genesis of lymphatic valves.12 There may be complex roles for ephrin
B2 and the Eph receptors; a valine deletion of PDZ binding site of
ephrin B2 causes mice to have normal blood vascular structure but
disturbed maturation of lymphatic vessels and valves.13 Deficiency of
podoplanin leads to abnormally dilated and nonfunctioning superficial
lymphatic vessels in the skin of newborn mice.14 Neuropilin 2, which
was originally shown to act as a semaphorin receptor in the nervous
system, binds to VEGF-C in addition to VEGFR-3.15 Neuropilin 2–
deficient mice showed severe hypoplasia of lymphatic capillaries from
E13 to birth, but had normal collecting vessels16; these defects were
transient and surviving mice regenerated capillaries similar to
heterozygous VEGF-C mice. Mice deficient in angiopoietin-2 (Ang 2)
also show defects in lymphatic vasculature, which could be rescued by Ang 1.17 Integrin IXβ1 binds directly to VEGF-C/D, suggesting
crosstalk between VEGFR-3 signaling and integrin-mediated adhesion and migration of lymphatic endothelial cells.18 The syk/slp-76 signaling
pathway may be involved, given that mice with mutations in the protein receptor tyrosine kinase Syk or its substrate (the adaptor protein SLP-76) exhibit edema and ascites as well as arteriovenous
malformations.19 The secondary lymphoid organs (lymph nodes and
Peyer’s patches of the small intestine) are superimposed on the
lymphatic vessels from the primitive lymph sacs formed by migrating
Prox-1–positive endothelial cells.20 Thus, although the development of
the lymphatic system is not understood completely, it is now agreed
that Prox-1 and VEGF-C are essential, with a varied contribution from
other proteins.
LYMPHANGIOGENIC FACTORS VEGF-C/D AND RECEPTOR VEGFR-3
VEGF-C and -D are ligands for the receptor tyrosine kinase VEGFR-3
(Fig 1 ). The VEGF-C/VEGF-D/VEGFR-3 axis constitutes the signal
transduction system for lymphatic endothelial cell growth, migration,
and survival.26-31 These growth factors are secreted as full-length
inactive forms consisting of NH2- and COOH-terminal propeptides and a central VEGF homology domain containing receptor binding sites.32
Proteolytic cleavage with plasmin removes the propeptides to generate mature forms, consisting of dimers of the VEGF homology domain, that bind receptors with much greater affinity than the full-length forms. 33 VEGFR-3 stimulation alone protects the lymphatic endothelial cells from serum deprivation-induced apoptosis, and induces their growth and migration. These signals are transduced via a protein kinase C–dependent activation of the p42/p44 mitogen-activated protein kinase signaling cascade and via a wortmannin-sensitive induction of Akt phosphorylation.31
The specific biologic effects of VEGF-C are critically dependent on
its proteolytic processing in vivo. Proteolytically processed VEGF-C/
VEGF-D also activates VEGFR-2 and can induce blood vessel growth.
27,28,34,35 Conversely VEGF-A, which is the primary angiogenic factor binding to VEGFR-2, can induce lymphatic hyperplasia but cannot substitute for VEGF-C in lymphatic development.8,36 Lymphatic endothelial cells express VEGFR-236-41 and VEGF-A is able to induce lymphangiogenesis potently at the cellular level on lymphatic
endothelial cells, in xenografts, and in skin during inflammation and
tissue repair.36,38,42
The VEGF-C/VEGFR-3 axis, through upregulation of contactin-1 and
activation of the src-p38 mitogen-activated protein kinase C/enhancer binding protein-dependent pathway may regulate the invasive capacity in different types of cancer cells and contribute to the promotion of cancer cell metastasis.43 Insulinlike growth factors 1 and 2 also induce lymphangiogenesis in vivo.44 Hepatocyte growth factor (HGF) is a lymphangiogenic factor that may contribute to lymphatic metastasis when overexpressed in tumors.45 The growth of HGF-induced lymphatic vessels can be partially blocked by a soluble VEGFR-3, suggesting that HGF may stimulate lymphatic vessel growth through an indirect mechanism. Other novel lymphangiogenic factors include Ang 1 and 2.17,46 Cyclooxygenase-2 may have a regulatory role in VEGF-C synthesis.47 Thus, the VEGF-C/VEGF-D/VEGFR-3 axis is the main signal transduction system in lymphatics; other novel lymphangiogenic factors may have direct or indirect influences on this system.
Experimental models have enhanced our understanding of lymphatic
vascular biology. These include the isolation of LECs and the
establishment of LEC lines31,40,42,48 using lymphatic markers such as
VEGFR-3, podoplanin, lymph vessel endothelial hyaluronic acid receptor 1 (LYVE-1). The Xenopus laevis tadpole has also been identified as an animal model that can be genetically manipulated to identify new lymphangiogenesis candidates (similar to zebrafish for angiogenesis research).49,50
LYMPHANGIOGENIC FACTORS PROMOTE TUMOR METASTASIS
Data supporting the premise that expression of lymphangiogenic factors promotes metastasis comes from in vitro and in vivo work.21 First, immunohistochemical studies show that overexpression of VEGF-C in breast, colorectal, gastric, thyroid, and prostate cancers is
associated with poor prognosis.51,52 Similarly, the expression level
of VEGF-D is an independent prognostic factor in ovarian cancer53 and stimulates lymphangiogenesis and lymphatic metastasis in human ductal pancreatic cancer54; expression of VEGFR-3 by lymphatic endothelial cells is associated with lymph node metastasis in prostate cancer.55 Expression levels of VEGF-C (and less often VEGF-D) also strongly correlate with lymph node metastasis in more than 30 studies.56,57 However, these observations in clinical tumor samples are largely correlative.
Experimental studies in animal models demonstrate that the VEGF-C/VEGF- D/VEGFR-3 signaling axis can promote tumor lymphangiogenesis and the metastatic spread of tumor cells.58 Two approaches have been used:
either to overexpress these factors in cell lines and study the
effects in vitro and in vivo, or to block the signaling with
inhibitors of the signaling cascade and observe the effects on
lymphatic and distant organ metastases.58 VEGF-C overexpression in
breast cancer cells in mouse experiments potently increased
intratumoral lymphangiogenesis, resulting in significantly enhanced
metastasis to regional lymph nodes and to lungs.59 Both VEGF-C and
VEGF-D enhance tumor lymphangiogenesis and lymphatic metastasis in
xenotransplant and transgenic models, and this promotes sentinel node metastasis.30,59-61 Moreover, in a chemically induced skin
carcinogenesis model, VEGF-C–overexpressing tumors induced the
expansion of lymphatic vessels within sentinel lymph nodes before the
onset of metastasis and promoted cancer metastasis beyond the sentinel lymph nodes to distal lymph nodes and lungs.62 VEGF-C–overexpressing
human melanomas in nude mice also showed enhanced tumor angiogenesis, indicating a coordinated regulation of lymphangiogenesis and angiogenesis in melanoma progression.63 Furthermore, VEGF-C induced chemotaxis of macrophages in vitro and in vivo, revealing a potential function of VEGF-C as an immunomodulator.63,64 Tumor- associated macrophages express lymphatic endothelial growth factors and VEGFR-3, and are related to peritumoral lymphangiogenesis, which may play a role in tumor cell dissemination.65
Transgenic mice that overexpress VEGF-A strongly promote multistep
carcinogenesis on chemical stimuli and demonstrate active
proliferation of VEGFR-2–expressing tumor–associated lymphatic vessels as well as tumor metastasis to the sentinel and distant lymph nodes.37 VEGF-A–overexpressing primary tumors induced sentinel node lymphangiogenesis even before metastasizing and maintained their lymphangiogenic activity after metastasis to draining lymph nodes.37 Primary tumors may prepare their future metastatic site by producing lymphangiogenic factors that mediate their efficient transport to the sentinel lymph node.37,62 These effects of VEGF-A on lymphatic vessels may be secondary to the induction of lymphatic vascular permeability or to recruitment of the inflammatory cells that produce VEGF-C/VEGF-D. 66
Conversely, blocking VEGFR-3 signaling with gene transfection or
recombinant adenoviruses suppressed tumor lymphangiogenesis and lymph node metastasis67 in xenografts established with a highly metastatic lung cancer cell line. However, lung metastasis was not affected by the blockade. Expression of a soluble VEGFR-3 antibody in a highly metastatic mammary cancer cell line suppressed metastasis formation in regional lymph nodes and lungs of rats.68 Similar successful approaches using recombinant adenoviruses have been used in a melanoma model,69 VEGFR-3–blocking antibodies in gastric cancer,70 and RNA interference in mouse mammary models.71
The cellular mechanisms of lymphangiogenesis in human diseases are
currently unknown, and could involve division of local preexisting
endothelial cells or incorporation of circulating progenitors. In
renal transplants, potential lymphatic progenitor cells derive from
the circulation, transmigrate through the connective tissue stroma,
presumably as macrophages, and incorporate into the growing lymphatic vessel.72
LYMPHATIC VASCULAR MARKERS
The recent identification of novel lymphatic markers that can
accurately discriminate lymphatic vessels from blood vessels in tissue
sections (LYVE-1, podoplanin, β-chemokine receptor D6, macrophage
mannose receptor, desmoplakin, and D2-4073-76) has yielded new insight into mechanisms of metastasis. Of these, LYVE-1, D2-40, and podoplanin have been most commonly used in studies to assess the significance of lymphatic vessel density/lymphatic area as a prognostic variable in survival and as a tool for predicting lymph node status in cancer (Fig 2).
LYVE-1 is a lymph-specific receptor for hyaluronan (HA).74 LYVE-1
sequesters HA on lymph vessel endothelium, colocalizes with HA on the luminal surface of lymphatic vessels, and binds both soluble and
immobilized HA exclusively.77 However LYVE-1 is also expressed in
normal liver blood sinusoids in mice and humans,78 and has been
identified on macrophages.79 The mucin-type transmembrane glycoprotein podoplanin is a highly expressed lymphatic-specific gene in cultured human LECs,42,73 and D2-40 is a novel monoclonal antibody that reacts with an oncofetal antigen present in fetal germ cells75,80; both are highly reliable lymphatic endothelial markers. These antibodies seem equally efficient in identifying lymphatic vessels in formalin-fixed tissue sections.81,82
These markers have been used to assess the significance of lymphatic
vessel density, to determine its prognostic significance to predict
nodal metastases and survival, and to investigate lymphangiogenesis in primary human tumors, with conflicting results (Table 1). Certainly in head and neck cancer, convincing evidence exists to support new
lymphatic vessel proliferation,100 and high intratumoral lymphatic
vessels were clearly associated with a higher risk for local relapse
as well as with poor disease-specific prognosis.83 However, in breast
cancer84,101,102 intratumoral lymphatic vessels are absent and the
significance of peritumoral vessels is unclear. Contradictory results
exist in melanoma and cervical cancer, in which two large studies
suggest improved survival with high lymphatic counts and attribute
this effect to immune responses triggered by inflammatory stromal
reaction.81,85 However, two smaller studies in melanoma suggest that
increased lymphatic density was associated with sentinel node
metastasis and poor survival, and that lymphatic vessel density could
discriminate between melanomas that metastasized and those that did
not.86,87 To date, no published studies in human cancer have commented on the presence of intratumoral lymphatics within metastases.
There may be several explanations for these conflicting results.
First, the studies differed with respect to antibodies used and
methods of evaluation of lymphatic vessel density. Lymphatic vessel
density was assessed either in hot spots or in areas of the highest
concentration of lymphatic vessels, or across the whole tissue
section. The proceedings of the First International Consensus
Conference on the Methodology of Lymphangiogenesis Quantification have been published recently, and recommend double immunostaining with the D2-40 monoclonal antibody and proliferation marker anti–Ki-67 antibody to assess lymphangiogenesis and details methods of quantification of lymphatic vessel density.103 This should result in standardization of results in future studies.
In some solid tumors such as ovarian cancer, the tissue sections may
not represent the invasive front of the tumor.88 Lymphangiogenesis may be more relevant in some cancers (eg, head and neck cancers) and more subtle changes in lymphatic surface area may be contributory in other cancers (eg, melanoma).87 Finally, lymphatic metastases may be an early event, which loses its prognostic significance in advanced cancers; more focused studies of early stage cancers or preinvasive lesions to identify a potential lymphangiogenic switch may be relevant. Interestingly, lymphovascular invasion identified in tissue sections of tumors clearly has been documented to be a reliable prognostic variable predicting nodal metastasis and survival in breast and cervix cancer, and influences decision making for therapy in testicular cancer.104-106 Although some of the conflicting data on the prognostic significance of lymphatic vessel density in human cancers can be attributable to varying methods of immunohistochemistry, antibodies, and modes of counting58 used in these studies, lymphatic metastases in human cancers may be complex and may reflect changes in
surface area of lymphatics or tumor cell–LEC interaction rather than a
simple increase in number of draining capillaries.107
Finally, the debate continues about whether tumor lymphangiogenesis
exists in human cancer or whether tumor cells invade preexisting
lymphatics at the periphery of the tumors,108 given that increased
interstitial fluid pressures may result in collapsed intratumoral
lymphatics. Additional studies in other tumor types and investigation
of lymphatic vessel density and proliferation markers may clarify the
role of new lymph vessel formation and the significance of lymphatic
vessel density in cancer progression.
MECHANISMS OF TUMOR CELL ENTRY INTO LYMPHATICS IN CANCER
He et al109 investigated how tumor cells gain access into lymphatic
vessels and at what stage tumor cells initiate metastasis. VEGF-C
produced by tumor cells induced extensive lymphatic sprouting toward
the tumor cells, dilation of the draining lymphatic vessels, and a
significant increase in lymphatic vessel growth between 2 and 3 weeks
after tumor xenotransplantation, with concurrent lymph node
metastasis. These processes were blocked dose dependently by
inhibition of VEGFR-3 signaling. However, lymph node metastasis was
not suppressed if soluble antibody to VEGFR-3 was started at a later
stage after the tumor cells had already spread out, suggesting that
tumor cell entry into lymphatic vessels is a key step during tumor
dissemination via the lymphatics. Whereas lymphangiogenesis and lymph node metastasis were inhibited significantly by antibody to VEGFR-3, some tumor cells were still detected in the lymph nodes in some of the treated mice. This indicates that complete blockade of lymphatic metastasis may require the targeting of both tumor lymphangiogenesis and tumor cell invasion.
However, Wong et al110 demonstrated that xenografts established in the prostate cancer model using stable short interfering RNA inhibiting
VEGF-C resulted in reduction of tumor lymphangiogenesis by 99%, but
this did not affect lymph node metastasis, indicating that tumor-
secreted VEGF-C is necessary for lymphangiogenesis, but
lymphangiogenesis was unnecessary for lymph node metastasis.
Tumor cells may also use physiologic chemokine receptor ligand
interactions for metastasis. Human breast cancer express chemokine
receptors CXCR4 and CCR7, and their respective ligands, CXCL12
(stromal-cell derived factor 1) and CCL21 (secondary lymphoid
chemokine) are highly expressed in the target organs of breast cancer
metastasis.111 Isolated LECs also express stromal-cell derived factor
1 and secondary lymphoid chemokine, suggesting they can attract tumor cells through secretion of chemokines.40 Similarly, lymphatic
endothelium secretes chemokine CCL21 (secondary lymphoid chemokine), which binds to CC chemokine receptor 7 leading to chemoattraction and migration of mature dendritic cells from skin to lymph nodes; CC chemokine receptor 7 is also expressed in some malignant melanoma cell lines.112
In summary, VEGF-C/VEGF-D, acting through their receptor VEGFR-3,
promote lymphatic metastasis. This could involve intratumoral or
peritumoral lymphangiogenesis, or more subtle alterations in tumor
cell–LEC interactions, and inhibition of this signaling causes
inhibition of lymph node metastasis (Fig 3).
TARGETING LYMPHANGIOGENESIS
Lymphedema
Lymphedema can be primary (caused by genetic conditions) or secondary to infection, malignancy, surgery, and/or radiotherapy. Filariasis in tropical countries and breast cancer treatment in the industrialized world are leading causes.113 Promising lymphedema treatment has been achieved in preclinical models using viral gene transfer vectors that induce lymphangiogenic factors.114 VEGFC gene transduction induces growth of functional lymphatic vessels,115 whereas the mature form of VEGF-D is a very powerful inducer of angiogenesis.35 VEGF-C gene therapy was effective in mouse models that demonstrate hereditary lymphedema.15 VEGF-C apparently can reverse the abnormalities in tissue architecture that accompany chronic lymphatic insufficiency.116 Ang1 gene transfer to mouse skin promotes lymphangiogenesis while inhibiting vascular permeability.46,117 Potential adverse effects from
prolymphangiogenic approaches include stimulation of lymphatic vessel
growth in cancer patients that may enhance metastatic spread118 and modulation of the immune system.
Targeting Tumor Lymphatics to Inhibit Metastases
Inhibition of VEGF-C/VEGF-D/VEGFR-3 axis in animal models can inhibit
tumor lymphangiogenesis and lymph node metastases.59 In addition,
blocking mouse VEGFR-3 with specific inhibitors has been shown to
block new lymphatic vessel growth exclusively, with no effect on
either blood angiogenesis or function of existing lymphatic vessels.
119 Akin to antiangiogenesis strategies, potential antilymphangiogenic
strategies could involve blocking antibodies or molecules that compete
for VEGF-C/VEGF-D/VEGFR-3 signaling, gene therapy to inhibit
lymphangiogenesis, small molecule tyrosine kinase inhibitors, and
inhibitors of other novel lymphangiogenic factors.58 Monoclonal
antibodies that inhibit lymphangiogenesis and angiogenesis exist,
120,121 although to our knowledge no studies exist yet in the clinical
setting. Blocking of lymphangiogenesis will need to be evaluated in
the adjuvant or neoadjuvant setting in combination with cytotoxics and surgery for primary therapy; cancers of interest on the basis of
proven lymphangiogenesis in animal models would appear to be breast
cancer, melanoma, colorectal cancer, and gastric cancer.58 The recent identification of novel lymphangiogenic factors, including hepatocyte growth factor122 and angiopoietin,17 indicate that efficient antilymphangiogenic strategies may need to target additional
lymphangiogenic molecules.123
Lymphatics in normal tissue play an important role in maintaining
tissue homeostasis and maintenance of interstitial fluid pressure.
Xenograft models of cancer demonstrate high intratumoral interstitial
fluid pressure, resulting in collapsed and nonfunctional intratumoral
lymphatics,108 whereas peritumoral lymphatics induced under the
influence of VEGF-C or VEGF-A seem to function poorly with incompetent valves.36,124 Increased interstitial fluid pressure forms a barrier to transcapillary transport and is an obstacle in tumor treatment; it results in inefficient uptake of therapeutic agents. Lowering the tumor interstitial fluid pressure might be a useful approach to improving anticancer drug efficacy.125 Indeed, normalization of tumor vasculature resulting in improved blood supply and drug transport to solid tumors has been proposed as the rationale behind combination treatment with antiangiogenesis agents and conventional cytotoxic therapy.126 It would be interesting to determine if a similar role can be envisaged for blocking lymphangiogenesis.
Several inhibitors of angiogenesis are being evaluated in trials127:
bevazucimab (recombinant human monoclonal antibody to VEGF) has been shown to improve survival in colorectal cancer. VEGF-A potentiates lymphangiogenesis and evaluation of inhibition of lymphangiogenesis in translational studies incorporated into trials with angiogenesis inhibitors may yield useful information. Combination treatment with the anti–VEGFR-2 and anti–VEGFR-3 antibodies seems to have cumulative effects on metastases compared with treatment with each antibody alone, suggesting that a combination therapy with antiangiogenic agents may be a promising approach for controlling metastases.128
ADDITIONAL DIRECTIONS OF RESEARCH AND UNANSWERED QUESTIONS
The growth of primary and metastatic tumors is unequivocally dependent on angiogenesis; similar proof does not currently exist for
lymphangiogenesis. Several unanswered questions exist in this field.
What are the mechanisms that control tumor cell interactions with
lymphatic endothelium? Is there a lymphangiogenic predisposition that
causes metastasis? Is there a correlation between histologic
differentiation/aggressive phenotypes on pathology and
lymphangiogenesis? Why do tumor cells promote lymphangiogenesis?129 Laboratory-based research with LECs, animal models, and translational studies should shed light on these questions.
In conclusion, the field of lymphangiogenesis has witnessed rapid
development after a long hiatus; the identification of lymphangiogenic
factors and their receptors, identification of lymphatic vascular
markers, and the implications of their activity in normal physiology
and pathology have improved our understanding of lymphatic vascular
biology. In the context of cancer, it is clear that tumor
lymphangiogenesis occurs in some tumors; blocking this process might
inhibit metastasis to lymph nodes, and lymphatic vascular markers may
be useful as a prognostic indicator of metastatic risk in cancer.
Additional research will clarify whether novel molecules targeting the
lymphangiogenic process are useful adjuvants to conventional
chemotherapy, and the extent to which existing antiangiogenic agents
may influence inhibition of lymphangiogenesis. Finally, a greater
understanding of these processes, particularly in the field of tumor
cell interactions with lymphatic endothelial cells, will pave the way
for harnessing this knowledge in cancer therapeutics.
ACKNOWLEDGMENTS
We thank Cancer Research UK and Oxfordshire Health Services Research;
Sanjeev Manek, John Radcliffe Hospital, United Kingdom, and Kathleen
Romain, MD, Cheltenham General Hospital, United Kingdom, for
assistance with immunohistochemistry and pathology images; and S.
Madhusudan, PhD, Churchill Hospital, United Kingdom for helpful
comments.
Journal of Clinical Oncology
