Michael J. Day, BSc, BVMS(Hons), PhD, DSc, DECVP, FASM, FRCPath, FRCVS
In recent years there has been increased understanding of the mechanisms involved oncogenesis--such as the role of genes regulating the cell cycle and apoptosis (e.g., p53) and the sequential progression of malignancy (initiation, promotion and progression). The development of neoplasia is clearly multifactorial, with both genetic background and a range of environmental or lifestyle factors influencing cellular transformation. One potential trigger for the neoplastic process is preceding inflammatory pathology, perhaps in the presence of altered (impaired) host immune defenses. There is also now recognition that inflammation may be a precursor to particular types of tumour arising in domestic animal species--and this appears to be a particular phenomenon in the cat. This presentation will review the proposed mechanistic basis for inflammation-associated neoplasia, with reference to two specific entities in feline medicine.
Mechanisms of Inflammation-Associated Neoplasia
Inflammation may act to enhance any of the three phases of tumour development.1 Local tissue inflammation might initiate neoplasia when inflammatory mediators such as reactive oxygen or nitrogen species (ROS, RNS) interact with genomic DNA in local tissue cells, leading to mutations. Inflammation likely has the most significant role in tumour promotion and the molecular pathways by which this occurs are now well-defined. Key inflammatory cytokines and chemokines (e.g., IL-1, IL-6, TNF, CXCL8 [IL-8]) are released from inflammatory cells and can bind to specific receptors on the surface of tumour cells. This in turn initiates an intracytoplasmic signaling pathway which generates molecules which enter the cell nucleus and activate gene transcription through the transcription factor NF-κB (nuclear factor κB). The activated genes encode anti-apoptotic products which enhance tumour cell survival, such as: BCL-XL, GADD45β, BFL1 and SOD2. Other products from inflammatory cells have additional effects on tumour cell survival or the growth of the tissue stroma in which the tumour is located (e.g., vascular endothelial growth factor [VEGF], colony stimulating factor [CSF-1]).
The roles of these various factors have been studied in murine models of neoplasia. For example, MDR2-/- mice spontaneously develop cholangitis which progresses to hepatocellular carcinoma; and this process can be blocked by inactivating NF-κB. Similarly, a model of inflammation-associated colon cancer involves pre-treatment of mice with a pro-carcinogen which is activated in enterocytes, followed by induction of colitis by local instillation of the irritant substance DSS. The inflammation leads to reduced barrier function, exposure of lamina propria macrophages to luminal bacteria, production of pro-inflammatory cytokines which signal through the NF-κB pathway to promote neoplasia amongst the transformed enterocytes. Blockade of the NF-κB pathway in enterocytes in this model impairs tumour formation.
Finally, inflammation might also stimulate the stage of tumour progression and this is thought to relate to a specific population of macrophage--the tumour associated macrophage (TAM). TAMs may have a specific repertoire of cytokine production, including: IL-10, TGFβ, VEGF, iNOS and IL-1receptor agonist. Regulatory cytokines such as IL-10 and TGFβ may inhibit the anti-tumour cytotoxic immune response, thus permitting increased tumour growth. Additionally, matrix metalloproteinases (MMP1, MMP3 and MMP9) derived from macrophages may allow disruption of surrounding matrix to encourage infiltrative growth and vascular metastasis, and VEGF may enhance tumour angiogenesis.
Recent research has indicated a key role for the cytokine IL-23 in the promotion of tumour growth.2 IL-23 is related to the cytokine IL-12 as both heterodimers share the same p40 subunit, with IL-23 linking this to p19 and IL-12 to p35. Similarly, the receptors for IL-12 and IL-23 share a common IL-12Rβ1 subunit and both are predominantly produced by dendritic cells and macrophages. Despite this close structural relationship, the functions of IL-12 and IL-23 are distinct. IL-12 drives the production of IFNγ secreting Th1 cells that enhance cytotoxicity and tumour destruction. By contrast, IL-23 induces Th17 cells--whose signature cytokine (IL-17) promotes tumour angiogenesis, enhances MMP production and inhibits the infiltration of tumours by CD8+ cytotoxic T cells. Inhibition of IL-23 expression within tumours has been shown to inhibit their growth.
Inflammation-Associated Neoplasia in Humans
There are several examples of inflammation-associated neoplasia in human medicine. Some are directly associated with infectious agents (e.g., papillomavirus, hepatitis B and C viruses, Epstein-Barr virus, Helicobacter pylori) and others are related to non-infectious causes of chronic inflammation (e.g., cigarette smoke, asbestos, silica and other environmental carcinogens). Progression of chronic inflammation to neoplasia most often occurs in the gastrointestinal tract, lungs, bladder, oesophagus and pancreas.
Feline Alimentary Lymphoma
One well-recognized, but poorly defined, example of inflammation-associated neoplasia is the proposed progression from inflammatory bowel disease (IBD) to alimentary lymphoma. This progression is documented in human patients with coeliac disease, and although reported in both the dog and cat, this appears to be far more of an issue in the feline species. In fact, the distinction between IBD and alimentary lymphoma on histopathological examination of gut biopsies still provides one of the most challenging interpretations for the diagnostic pathologist. Determining whether a mononuclear cell infiltration of the intestinal lamina propria is reactive or neoplastic has now been aided by the techniques of immunohistochemistry and clonality testing. Immunohistochemical labeling for the expression of molecules restricted to T (e.g., CD3) or B (e.g., CD21, CD79a) lymphocytes can determine whether an infiltrate is mixed in nature or monomorphic--the latter being more likely associated with neoplasia.3 In a recent investigation, we reviewed 32 cases diagnosed as alimentary lymphoma by routine examination of HE-stained intestinal biopsies taken from cats with chronic diarrhoea.4 Immunohistochemistry defined these as predominantly T cell tumours, with fewer B cell or null cell lymphomas. Of greater interest, was the group of 5 cases in which immunohistochemistry was able to demonstrate a mixed infiltrate of morphologically-normal cells more consistent with chronic inflammation than neoplasia. The recent introduction of 'clonality testing' (at least in the United States) has provided a second means of making this distinction.5 This methodology determines whether the population of lymphocytes infiltrating the mucosa is monoclonal (i.e., carries a single type of T or B cell receptor) or polyclonal (i.e., a mixed population with numerous different T and B cell receptors)--the former associated with neoplastic transformation. Clonality testing has to date been most widely applied to blood samples or fresh lymphocyte suspensions, but can also be performed with fixed tissue biopsies.
In the context of lymphoma, a recent epidemiological study has suggested an association between passive exposure to cigarette smoke and the development of feline lymphoma--which if confirmed may be an additional example of inflammation-associated neoplasia.6 It is also of note that feline lymphoma may arise in tissues which are often affected by chronic lymphoplasmacytic inflammation (e.g., nasal mucosa, intestine, renal interstitium).
Feline Vaccine-Associated Sarcoma
The second example of a likely inflammation-associated neoplasm is the range of sarcomata that develop at the site of previous injection--most often, but not exclusively, with vaccine. Repeated injection (and perhaps also application of topical medication) to a single cutaneous site (generally the dorsal cervical region) may act as a trigger for the subsequent development of an aggressive, infiltrative sarcoma which provides a major surgical challenge. These tumours are most often recorded as fibrosarcoma, but a broad spectrum of mesenchymal neoplasms induced in this way is documented.7
Epidemiological studies have suggested that these tumours are more frequently associated with the use of adjuvanted vaccine products (in particular rabies and feline leukaemia virus vaccines). The most widely used adjuvant is alum, although some products contain a lipid-based adjuvant. The current proposed pathogenesis of vaccine-associated sarcoma is that adjuvant-induced chronic inflammation with localized release of pro-inflammatory cytokines initiates genetic mutation and transformation of cutaneous spindle cells. Although definitive proof for this mechanism is not reported numerous studies have shown compatible molecular changes in vaccine-associated versus non-vaccine associated sarcomas by either immunohistochemistry or RT-PCR.8,9 For example, there is expression of platelet-derived growth factor [PDGF], epidermal growth factor [EGF] and its receptor, basic fibroblast growth factor [FGF-b], TGFβ, matrix metalloproteinase [MMP2, MT-MMP16] and the proto-oncogene c-jun (which encodes the transcriptional protein AP-1) by tumour cells, and expression of PDGF by infiltrating lymphocytes and macrophages. Abnormalities in expression of the pro-apoptotic tumour suppressor gene p53 have also been described in vaccine-associated sarcoma.10 It has also been suggested that cats which develop vaccine-associated sarcoma subsequently have a higher incidence of lymphoma.11 Recently, there have been several reports of sarcoma developing in association with microchip implantation in the dog.12
We have recently reported an investigation of the histopathological changes that occur in the subcutis of kittens given a single injection of multicomponent vaccine--either: non-adjuvanted, alum-adjuvanted or lipid-adjuvanted.13 There are striking differences in the extent and severity of inflammatory change induced by adjuvanted compared with non-adjuvanted products. A single injection of alum-adjuvanted vaccine leads to extensive tissue necrosis and inflammation which extends from the panniculus muscle, through the subcuticular adipose tissue to the dorsal cervical musculature. This reaction increases in intensity from day 7 to day 21 post injection--and at day 62 post-injection, although there is tissue repair and fibrosis, there are persistent granulomatous aggregates of alum-laden macrophages and giant cells. In this study, there was no evidence for early neoplastic transformation--but it might be expected that this would take a longer time to develop, and would be enhanced by repeated injection and inflammatory insult to the same anatomical location.
1. Karin M, Greten FR. 2005. NF-κB: linking inflammation and immunity to cancer development and progression. Nature Reviews Immunology 5: 749-759.
2. Langowski JL, Zhang X, Wu L et al., 2006. IL-23 promotes tumour incidence and growth. Nature 442: 461-465.
3. Waly NE, Stokes CR, Gruffydd-Jones TJ, Day MJ. 2004. Immune cell populations in the duodenal mucosa of cats with inflammatory bowel disease. Journal of Veterinary Internal Medicine 18: 816-825.
4. Waly NE, Gruffydd-Jones TJ, Stokes CR, Day MJ. 2005. Immunohistochemical diagnosis of alimentary lymphomas and severe intestinal inflammation in cats. Journal of Comparative Pathology 133: 253-260.
5. Moore PF, Woo JC, Vernau W, Kosten S, Graham PS. 2005. Characterization of feline T cell receptor gamma (TCRG) variable region genes for the molecular diagnosis of feline intestinal T cell lymphoma. Veterinary Immunology and Immunopathology 106: 167-178.
6. Bertone ER, Synder LA, Moore AS. 2002. Environmental tobacco smoke and risk of malignant lymphoma in pet cats. American Journal of Epidemiology 156: 268-273.
7. Richards JR, Starr RM, Childers HE et al., 2005. The current understanding and management of vaccine-associated sarcomas in cats. JAVMA 226: 1821-1842.
8. Nieto A, Sanchez MA, Martinez E, Rollan E. 2003. Immunohistochemical expression of p53, fibroblast growth factor-b, and transforming growth factor-α in feline vaccine-associated sarcomas. Veterinary Pathology 40: 651-658.
9. Sorensen KC, Kitchell BE, Schaeffer DJ, Mardis PE. 2004. Expression of matrix metalloproteinases in feline vaccine site-associated sarcomas. Am J Vet Res 65: 373-379.
10. Hershey AE, Dubielzig RR, Padilla ML, Helfand Sc. 2005. Aberrant p53 expression in feline vaccine-associated sarcomas and correlation with prognosis. Veterinary Pathology 42: 805-811.
11. Madewell BR, Gieger TL, Pesavento PA et al., 2004. Vaccine site-associated sarcoma and malignant lymphoma in cats: a report of six cases (1997-2002). JAAHA 40: 47-50.
12. Vascellari M, Melchiotti E, Mutinelli F. 2006. Fibrosarcoma with typical features of postinjection sarcoma at site of microchip implant in a dog: histologic and immunohistochemical study. Veterinary Pathology 43: 545-548.
13. Day MJ, Schoon H-A, Magnol J-P, Saik J, Devauchelle P, Truyen U, Gruffydd-Jones TJ, Cozette V, Jas D, Poulet H, Thibault J-C, Pollmeier M. 2007. A kinetic study of histopathological changes in the subcutis of cats injected with non-adjuvanted and adjuvanted multi-component vaccines. Vaccine (in press).