Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Prognostic and predictive markers in canine tumours: Rationale and relevance. A review

Prognostic and predictive markers in canine tumours: Rationale and relevance. A review Veterinary Quarterly 2005; 27(2): 52-64 Prognostic and predictive markers in canine tumours: Rationale and relevance. A review S. Mukaratirwa* Department of Paraclinical Studies, Pathology Section, Faculty of Veterinary Science, University of Zimbabwe TABLE OF CONTENTS Summary and keywords Abbreviations Introduction Proliferation markers Mitotic count/index Incorporation techniques (thymidine and BrDU) Proliferating cell nuclear antigen (PCNA) Ki-67 Argyrophilic nucleolar organiser region (AgNOR) proteins Predictive value of proliferation markers for therapy response of canine tumours 56 Apoptosis Apoptosis and prognosis in canine tumours Extracellular matrix and adhesion molecules Extracellular matrix molecules Matrix metalloproteinases Adhesion molecules Angiogenesis p53 protein Conclusion References . . . . 61 *Address: Department of Paraclinical Studies. Pathology Section, Faculty of Veterinary Science, University of Zimbabwe, P.O. Box MP 167, Mount Pleasant, Harare, Zimbabwe. E-mail: tmukaratirwa@vet.uz.ac.zw 53 Prognostic and predictive markers in canine tumours: Rationale and relevance. A review S. Mukaratirwa SUMMARY Cancer is one of the most common causes of death in dogs. The availability of multiple treatment modalities and novel therapeutic targets make the correct diagnosis, prognostic stratification, and the identification of treatment effect predictive factors an issue of major debate in cancer management. Selection of high and low risk patients and the type of systemic or local treatment is important in cancer management. The search for better prognostic markers and predictive factors is now focused on the molecular mechanisms which underlie tumour behaviour, such as altered cell cycle progression, proliferation, apoptosis, and angiogenesis. The ultimate goal is to identify reliable markers that can accurately diagnose and stage a tumour and predict tumour's clinical behaviour, prognosis and response to therapy. In this review, the current state of prognostication in canine tumours and promising new molecular markers are discussed. The markers are allocated to four groups according (" °proliferation markers, mapoptosis, extracellular to their function: matrix and cell adhesion molecules, (i")angiogenesis and °cell cycle regulators. Clinicopathological factors and histopathological grading remain the most practical parameters in decision-making. Although experimental research has shown that molecular markers have a good potential to be used as diagnostic, prognostic or predictive markers in canine tumours, insufficient evidence exists on their efficacy for routine use in veterinary oncology. Keywords: Adhesion molecules; Angiogenesis; Apoptosis; Canine neoplasms; Cell cycle; Dog diseases; Molecular markers; p53 protein; Prognosis; Prognostic and predictive markers; Proliferation; Proliferation markers; Tumour behaviour; Tumour grading; Tumour markers; Tumour staging; Tumours; Veterinary oncology. Abbreviations AgNOR: Argyrophilic nucleolar organiser region iMVD: Intratumoural microvessel density MMPs: Metalloproteinases BrDU: Bromodeoxyuridine ECM: Extracellular matrix PCNA: Proliferating cell nuclear antigen evidence of distant metastases have long INTRODUCTION been recognized, but this is often not enough to predict The demonstration of clinicopathological factors accurately the prognosis and response to therapy of and biological markers with prognostic significan- the tumour. Inconsistencies between histopatholo- ce has become one of the most important fields of gical features and biological behaviour or progno- study in medical and veterinary oncology. These sis have been documented in several canine factors to discriminate are used and markers tumours. Therefore, a considerable number of benign from malignant tumours, grade tumours molecular markers have been applied to diagnose according to aggressiveness and metastatic risk, and to predict the prognosis of neoplastic diseases and to predict the prognosis, disease-free survival in dogs. after surgical rates rates and overall survival The aim of this paper is to review biomarkers excision or therapy of the tumour. used, (uto differentiate histologically similar The grading of malignancy and prognosis of a tumours, (")to discriminate malignant from benign tumour based on the classical clinicopathological neoplastic diseases and classify some of the factors and histopathological criteria such as size tumours into grades .according to aggressiveness and presence of location of the tumour, and metastatic risk, predict the prognosis of microscopic regional lymph node metastases and neoplastic disease, and (1")to predict response to 54 therapy of neoplastic diseases. The rationale and cell size and the effect of time before sample relevance of these markers is also discussed. fixation on mitosis. In some instances pathologists Moreover, this review is intended to raise interest may have difficulties in discriminating mitoses and in the use of these markers in veterinary oncology. necroses (karyorrhectic and pyknotic nuclei), and In this review, the term diagnostic marker means a prophase and anaphase are difficult to identify in marker that can help to distinguish histologically haematoxylin and eosin stained sections. In spite similar tumours; the term marker prognostic of these shortcomings, determination of mitotic means a marker that correlates with malignancy, count/index on haematoxylin and eosin stained disease-free survival rate and overall survival rate; tumours has some clinical implications. The major the term treatment effect predictive marker advantage is that histopathologic diagnosis and indicates a marker that is capable of predicting mitotic count/index can be determined at the same tumour sensitivity or resistance to various time, making it the most convenient method of therapies evaluating cell proliferation. In veterinary oncology, mitotic count/index has PROLIFERATION MARKERS been used by histopathologists both as a diagnostic aid and prognostic indicator for several tumours. Cell proliferation is the increase in cell number Mitotic count/index can be useful in separating resulting from completion of the cell cycle (72). several benign tumours from their malignant The proliferation rate of tumour cells is a principal counterparts e.g. canine mammary tumours (76) determinant of the aggressiveness of a tumour and, and canine smooth muscle tumours of urinary therefore, an important prognostic factor. Many bladder (38), and may also be employed in different methods have been developed to detect separating tumours with similar histological cell proliferation in tumours. These methods differ appearance e.g. canine basal cell carcinomas from not only in what they detect, but also in the phases squamous cell carcinomas (15). of the cell cycle that they label the cell. Therefore, these methods often produce different results. Mitotic count/index has been shown to be of prognostic value in several canine tumours Methods used to estimate proliferation of tumour including melanomas (6), fibrous connective tissue cells include the early methods of measuring mito- (5), sarcomas haemangiopericytomas (28), mast tic count/index and S-phase labelling with thymidi- cell tumours (4), mammary tumours (94), and ne or bromodeoxyuridine (BrDU), and recently, splenic non-angiomatous and non-lymphomatous flow cytometry, in situ hybridization, histochemi- tumours (100). A high mitotic index in these cal staining methods and immunohistochemical tumours is associated with high tumour recurrence labelling methods. The most commonly used histo- after surgical excision and with a shorter survival chemical proliferation marker is argyrophilic time of the animal. nucleolar organiser region (AgNOR), and the most commonly used immunohistochemical markers are Incorporation techniques proliferating cell nuclear antigen (PCNA) and (thymidine and BrDU) Ki-67 antigen. Each of these methods has some practical problems in routine histopathology. In vitro 3H-thymidine incorporation technique was the earliest technique available for evaluating cell Mitotic count/index proliferation. However, 3H-thymidine incorpora- tion technique is impractical because it is time Mitotic count is the number of mitoses per high consuming and has the difficulty and expense of power field (x 400 magnification) and mitotic disposing radioactive material. Because of these index is the fraction of mitoses expressed as a drawbacks, bromodeoxyuridine (BrDU) was intro- percentage of the total number of cells present. duced for the visualization of DNA synthesis in Mitotic count or index can determined on the S-phase of the cell cycle. BrDU, an analogue haematoxylin and eosin stained sections of to 3H-thymidine, is incorporated in the nucleus tumours, and therefore, a routine part of during the S-phase and can be histopathological evaluation. detected The validity of by immunohistochemistry on formalin-fixed, paraffin- mitotic count or index remains controversial. It is embedded tissue or by flow cytometry with anti- not a standardized method and does not consider 55 BrDU monoclonal antibody (27). The need for pri- PCNA index was not associated with overall or administration of BrDU to the live animal is the survival time (42). In mast cell tumours, disease- free survival and overall survival time main drawback of BrDU incorporation technique. was longer in tumours with a low PCNA index (1,98). For the use of BrDU Few studies have analysed mammary tumours, one study showed no and irnmunohistochemistry the diagnosis in significant association of a high PCNA with prognosis of canine tumours. BrDU index could be metastasis, death from neoplasia, low disease-free used to distinguish acanthomatous epulides from survival rates and low overall survival rates (73), less aggressive epulides (114), osteosarcomas from while in another study, a low PCNA index was chondrosarcomas (70), and could be used to grade associated with low disease-free survival rates mast cell tumours according to their aggressiveness (52). In melanocytic tumours, PCNA index was of acanthomatous (92). in A high BrDU index no prognostic significance (88). epulides, cell carcinomas, and fibrosar- coma correlateswith their poor prognosis (114). Ki-67 Proliferating cell nuclear antigen (PCNA) Ki-67 protein is a non-histone nuclear protein present during all active phases of the cell cycle The proliferating cell nuclear antigen (PCNA) is a (G1, S, G2, and mitosis), but is absent from quies- nuclear protein which functions as a co-factor for cent or resting cells (23). During interphase, Ki-67 DNA-polymerase delta in both the G phase and in antigen can be exclusively detected within the DNA synthesis (8). PCNA is produced in the mid- nucleus, whereas in mitosis, most of the protein is it is 01 phase and throughout the S phase but, relocated to the surface of the chromosomes. This detectable throughout the cell cycle due to its long makes Ki-67 protein an excellent marker for half life (9). It is present in all active phases of the determining the growth fraction of a given cell cell cycle (GI, S, G2), manifesting a striking in- population. Unlike proliferating cell nuclear crease in concentration during the S-phase. There- antigen (PCNA), which not only plays a role in fore, its rate of synthesis is correlated directly with DNA replication, but also in DNA repair, Ki-67 is proliferation rate of cells. However, its expression virtually restricted in its role as a proliferation may be induced by DNA damaging agents. antigen, making it a more specific determinant of There are a few cases where PCNA index has growth fraction. For this reason, antibodies against been found to be of value in separating histologi- Ki-67 are increasingly being used in human and cally similar canine tumours e.g. lung adenosqua- veterinary oncology as prognostic markers in carcinoma cell mous carcinoma and squamous different types of neoplasms. MIB-1 is the (29), and cutaneous squamous cell carcinoma and monoclonal antibody developed against the Ki-67 basal cell carcinoma (56). However, the PCNA antigen that can be used on routinely formalin- index was unable to separate testicular tumours fixed, paraffin-wax embedded tissues after cell (Sertoli into types different histological microwave antigen retrieval (12). Ki-67 antigen tumour, seminoma and Leydig cell tumour) (93). can also be detected using wet-fixed cytology imprints (115), thus providing rapid information The value of PCNA index in separating benign has about malignancy and patient outcome. tumours from their malignant counterparts been assessed in several canine tumours. In most The Ki-67 index, determined by immunohistoche- of the tumours, including mammary tumours (22, mistry on tumour tissues has been used as a 73,76), melanocytic tumours (88), perianal gland potential diagnostic marker in several canine and spindle cell tumours (34), malignant tumours tumours. It has been shown that Ki-67 index can had significantly higher PCNA index than their be used to distinguish canine aortic tumours from benign counterparts. The PCNA index has been carotid body tumours (10), cutaneous histiocytoma found to be useful in grading mast cell tumours from transmissible venereal tumour (30), keratini- according to their degree of aggression (34). zing acanthoma from well differentiated squamous cell carcinoma of the Studies on PCNA index as a prognostic marker skin (13), and classify have been done on canine lymphoid, mast cell, spontaneous lung epithelial tumours into adeno- mammary and melanocytic tumours. In lymphomas squamous and squamous cell carcinomas (29). In canine testicular tumours, however, Ki-67 without surgical or therapeutic intervention, the index 56 was of no diagnostic value in separating Of particular interest is, that the amounts of seminomas from interstitial and Sertoli cell interphase AgNOR proteins are strictly related to tumours (93). Ki-67 index was also unable to rapidity of cell proliferation (99). Therefore, classify canine extramedullary plasmacy-tomas into unlike BrDU, Ki-67 or PCNA index which different grades (75). In canine mammary tumours measures the percentage of proliferating cell, (49) and melanocytic tumours (58), Ki-67 index AgNOR count gives information on the rapidity of was able to distinguish benign from malignant cell proliferation (proliferation rate). AgNOR tumours, with malignant tumours having a higher count can be determined using histochemical index than benign tumours. staining methods on histological sections of routinely processed issue samples. Most of the studies on Ki-67 antigen expression as prognostic markers have been performed on The AgNOR count was be able to separate several mammary, lymphoid, melanocytic and mast cell canine tumours into malignant and benign tumours tumours. Several researchers found Ki-67 index to including mammary tumours (7,14,39), urinary be of prognostic significance in canine mammary bladder smooth muscle cell tumours (38), perianal tumours, with a high index being positively gland tumours (34), fibroblastic tumours (34) and correlated with increased risk of metastasis, death epithelial skin tumours (13,15). In these studies, from neoplasia, low disease-free survival and low malignant tumours had a higher AgNOR count overall survival rates (67,73,94,115). However, than their benign counterparts. one study dealing with Ki-67 immunolabelling in The prognostic value of AgNOR has, unfortunate- canine mammary of tumours reported lack ly, been done on only a few canine tumours. Two prognostic significance of this marker (52). studies performed on the prognostic value of Conflicting results on the prognostic significance AgNOR count in mast cell tumours (4,98), have of Ki-67 index in human breast tumours have also shown that, tumours with a high AgNOR count been documented (41,90). The discrepancy in had a higher chance of recurring after surgical these results can be explained by methodological excision than tumours with a low count. In issues such as sampling strategies (criteria addition, the overall survival time of dogs with employed in dividing patients into groups) and mast cell tumours was found to be lower in dogs intratumoural heterogeneity (heterogeneity in the with tumours with a high AgNOR count (98). In prevalence and biological behaviour of various study an early of mammary tumours on the histological types of tumours). prognostic significance of AgNOR count, it could In one study of canine lymphomas, the Ki-67 not be used to predict survival time of dogs with index correlated with the histologic grade of the tumours (52). However, a more recent study on tumour (21), suggesting that Ki-67 index may be canine mammary tumours (94) revealed a signifi- useful for classification of lymphomas. Other cant association between a high AgNOR count and canine tumours in which Ki-67 index was found to a short one-year survival time. be of prognostic significance include melanocytic tumours (48,58,88) and mast cell tumours (1). Predictive value of proliferation markers for Tumours in which Ki-67 index was found to be of therapy response of canine tumours no prognostic significance includes canine Tumour cell proliferation rate is important neuroendocrine tumours of the pancreas (59). for tumour growth, invasion and metastasis. There- fore, pre-operative or pre-treatment proliferation Argyrophilic nucleolar organiser region rate of a tumour may be correlated with response (AgNOR) proteins to therapy and disease outcome. A rapidly prolife- Argyrophilic nuclear organizer region (AgNOR) rating tumour should respond poorly to therapy proteins are a set of nuclear proteins that accumu- compared to a slowly growing tumour. A high late in highly proliferating cells. In proliferating proliferation index is not, however, consistently cells, the quantity of interphase AgNOR proteins associated with a poor response to therapy in all progressively increases from early G I phase, tumours. More studies have been done on the reaches a maximum value at the end of S-phase diagnostic and prognostic value of proliferation and remains constant up to the late G2 phase (99). markers than on the predictive value of these 57 ting indexes using markers. A good prognostic indicator may not be some of the proliferation necessarily a good predictive factor for therapy, markers. This does not necessarily means, that these markers are not good indicators and vice versa. of cell proliferation. It may also indicate that there is in the The markers value of proliferation variability in the course of the disease in relation and prediction chemotherapy to of response to the proliferative activity. canine radiotherapy in has been studied lymphomas and transmissible venereal tumours. APOPTOSIS Pretreatment BrDU index was predictive of first remission in lymphomas (108). For the studies The rate of tumour growth is determined by done on the predictive value of Ki-67 index in tumour cell proliferation and tumour cell death. canine lymphomas, the results obtained by two An imbalance between tumour cell proliferation (43) al. research groups are different. Kuipel et and death is fundamental for tumour growth, found no correlation between Ki-67 index and progression and regression. Therefore, qualitative response to therapy of canine lymphomas, while and quantitative changes in cancer cell death along (74) found that Ki-67 index was Phillips et al. with proliferative alterations are essential relapse free predictive of the duration of first determinants in the pathogenesis of a malignant in interval In two studies after chemotherapy. disease and its responsiveness to therapy. canine lymphomas on the use of mitotic index Apoptosis is a genetically regulated form of cell (43,74) and the four studies done on the use of death, abolishing cell populations in embryological PCNA index (42,43,74,108) as a prediction of developments and in physiological and pathological response to chemotherapy, both mitotic index and processes including tumours. Apoptosis has PCNA index were not predictive for the duration characteristic biochemical and morphological of first tumour relapse free interval and overall criteria which distinguish it from necrosis (113). survival. In the studies done on the use of AgNOR Deregulated apoptosis can promote carcinogenesis count as a prediction of response of lymphomas to in two ways. First, impaired apoptosis, either due chemotherapy, the count was not predictive for the to an uncontrolled expression of oncogenes or duration of first tumour relapse free interval (42, tumour suppressing genes, leads to an uncontrolled 43,108). In canine transmissible venereal tumour, accumulation of malignant cells and eventually poor response to chemotherapy was associated formation of cancer cells (40). Apoptosis also with a high AgNOR count (31). In a study done limits the population expansion of tumour cells on the value of PCNA index in the prediction of early in the process of tumour growth (65). In response to radiotherapy of dogs with incompletely addition, tumours develop methods to evade resected meningiomas, a high PCNA index was elimination by immune system; one such associated with a low progression-free survival mechanism involves expressing Fas ligand, which rate (106). enables them to delete (by apoptosis) anti-tumour In summary, it is apparent that increased prolifera- lymphocytes (68). However, the accumulation of tion strongly with malignancy, poor neoplastic cells is not consistently associated correlates with prognosis and response to therapy irrespective of decreased rate of apoptosis. On the contrary, high literature apoptotic activity seems to be the methodology used. The available apparent particularly in clearly argues for the assessment of tumour cell aggressive tumours such as human urinary bladder carcinoma (51). proliferation in canine tumour prognostics. Mitoses counting provides the most simple and reproduci- Ki67 and PCNA/MIB-1 ble prognostic value, Apoptosis and prognosis in canine tumours index and AgNOR count are promising alternatives There are few published data on the value of that need methodological fine tuning and standardi- apoptosis as a diagnostic or prognostic factor in sation. In this context, it is worth mentioning that canine tumours. A study by Guvenc et al. (30) the diagnostic, prognostic or predictive value of showed that apoptotic indices are useful in proliferating indexes depends on the tumour type distinguishing canine cutaneous histiocytomas from and the availability of protocols for treatment. For transmissible venereal tumours. Malignant no correlation could be certain tumour types, melanocytic (87), and malignant mammary (22) demonstrated between prognosis and the prolifera- tumours had higher apoptotic indices than their 58 benign counterparts. However, in mammary Interaction between host cells, tumour cells and tumours, no significant correlations were noted ECM molecules, adhesion molecules and proteoly- between apoptotic index and clinical features such tic enzymes can have restrictive or permissive as metastasis and tumour diameter (22). influence on the behaviour (proliferation, invasion and metastasis) of tumour cells. Most anticancer chemotherapeutic drugs, as well Therefore, the level of expression of ECM molecules, proteolytic as radiation therapy, regardless of their distinct enzymes and adhesion molecules in tumours could mechanisms of action, appear to suppress tumour be of diagnostic or prognostic value growth mainly by induction of apoptosis of tumour and might be of value in predicting the effect of therapy cells (19), and there is now in of some vitro evidence malignancies. pointing to apoptosis playing an important role in the tumour response elicited by anticancer agents Extracellular matrix molecules (95). In dogs with invasive urinary bladder cancer, piroxicam, a cyclooxygenase inhibitor caused Extracellular matrix molecules are thought to be reduction in tumour volume, and this reduction in involved in numerous cellular functions including tumour size was strongly associated with induction cell adhesion, differentiation and proliferation. of apoptosis (60). Vincristine sulphate chemothera- These molecules are over-expressed in the stroma py of canine transmissible venereal tumour causes of several human (45) and canine tumours (16,18, tumour regression by apoptosis (25). Therefore, 32,61,63,64), and this led to the assertion that the extent of apoptosis in a tumour may be of these molecules may be associated with malignan- value in predicting and monitoring response to cy and poor prognosis. The most widely studied chemotherapy or radiotherapy. However, only one ECM molecules in neoplastic diseases are tenascin- study has been performed on predictive value of C, versican and hyaluronan. apoptosis in canine tumours. In this study, pretreatment apoptotic index of lymphomas was Tenascin-C, a large ECM glycoprotein, is predictive of the duration of first relapse free implicated in cell adhesion and proliferation, and interval (74). its overproduction is associated with poor prognosis in several human tumours including In conclusion, apoptosis seems to play an (46), colorectal breast (36) and gastric (111) important role in all the stages of carcinogenesis cancer. In canine tumours, no studies have been (initiation, promotion, progression and regression), done on the prognostic significance of tenascin-C. but the clinical and biological relevance of However, studies on canine mammary tumours apoptosis of canine tumours need to be studied. In (18) and canine gastrointestinal epithelial tumours human oncology, where numerous studies have (63) have demonstrated an increased immunohisto- been performed, the relationship between a high chemical expression of tenascin-C apoptotic index and good prognosis has not been in tumours compared to normal tissues. It is likely substantiated. that other factors in tumour progression, such as other forms of tumour a member of the Versican, large aggregating cell death, proliferation rate, angiogenesis and the sulphate proteoglycan chondroitin family, is host immune response, have a confounding effect recognized anti-cell adhesive molecule involved in on tumour behaviour and response to therapy. regulating cell motility on ECM components (17). Therefore, there is a need to do a multivariate The level of versican expression is of prognostic study on these factors. value in several human tumours including prostatic (86) and breast (85) tumours. Versican expres-sion EXTRACELLULAR MATRIX AND was shown to be increased in canine mammary ADHESION MOLECULES (16) and colorectal tumours (64). Although no studies have been done on the prognostic value of Tumour invasion and metastasis results from versican in canine tumours, the level of expression aberrant communication between tumour cells and of versican in colorectal epithelial tumours was microenvironment. The microenvironment consists associated with depth of tumour invasion and local of a variety of host cells (vascular cells, fibro- lymph node metastasis (64), suggesting that blasts, myofibroblasts and inflammatory cells), versican can be of prognostic value. extracellular matrix (ECM) molecules, cell adhesion molecules and proteolytic enzymes. Hyaluronan is a linear unsulphated glycosamino- 59 E-cadherin is glycan synthesized on the cytoplasmic side of the involved in cell-cell adhesion, plasma membrane. Hyaluronan has been associated whereas the intracellular domain connects to the actin cytoskeleton via catenins. Dysfunction with many different cellular processes including or loss of the intercellular cell adhesion, proliferation, E-cadherin-beta-catenin aggregation, complex is frequent locomotion (44). Hyaluronan in many human epithelial and migration tumours (103). In some human over-production is a diagnostic or prognostic factor carcinomas including breast (97) and prostate carcinomas (84), for several types of human tumours including E-cadherin and beta-catenin loss is associated colorectal (89) and breast (3) tumours. In canine with a poor prognosis. Down-regulation of E-cadherin colorectal tumours, there is increased hyaluronan expression has been demonstrated expression compared to normal tissues (64), but no in canine mammary (11,79,82) and colorectal carcinomas studies or its prognostic have been done on (57), but no studies have been done on the prog- predictive value. In canine transmissible venereal nostic significance of E-cadherin expression in tumours, strong stromal hyaluronan intensity and a canine tumours. high proportion of hyaluronan-positive tumour cells were significantly associated with progressing In summary, it is apparent that ECM molecules, tumours (61). MMPs and adhesion molecules play an important role in multistage carcinogenesis. The altered expression of these molecules Matrix metalloproteinases may be induced by genetic or epigenetic events and this altered state Degradation of the ECM is crucial for malignant may favour altered cell differentiation, prolifera- and invasion tumour growth, angiogenesis, tion, locomotion, invasion and metastasis. Whilst metastasis. Matrix metalloproteinases (MMPs) are there is no single ECM molecule, MMP or adhe- a family of zinc-dependent neutral endopeptidases sion molecule identified for routine clinical use in which catalyse degradation of diverse substrates in the prognostic evaluation of both human and in both the ECM. The MMPs are involved animal tumours, results from studies on human growth and promoting stimulating tumour cell tumours are promising. Therefore, there is need to invasion and metastasis (33). The levels of MMPs do more studies on the diagnostic, prognostic and can be determined in patient serum and urine or in predictive value of tumour microenvironment in tumour tissues by immunohistochemistry, Western canine tumours. blotting, Northern blotting or polymerase chain in tumour reaction. their role Consistent with ANGIOGENESIS of certain MMPs in progression, levels high cancer tissues correlate with poor prognosis in Angiogenesis, the development of new blood vessels (104), from breast the several human cancers including existing vasculature, is an essential component of solid colonic (69) and prostatic (101) carcinomas. tumour growth and metastasis. Without angiogenesis, tumour expan- Evidence for the role of MMPs in canine tumours sion cannot proceed beyond 1-2 mm because is sparse. The distribution of MMPs has been tumour proliferation is severely limited by nutrient studied by immunohistochemistry in various canine supply to the tumour and waste removal from the sarcomas and carcinomas (53), mammary gland tumour into the surrounding medium (20). There- tumours (71), osteosarcomas (47) and cutaneous fore, angiogenesis is considered an attractive In these studies, higher mast cell tumours (50). prognostic indicator. levels of MMPs were found in neoplastic tissues In human cancers, the degree of than in normal tissues. No studies have been done angiogenesis of a tumour, as assessed by intratumoural on the prognostic significance of MMPs in canine microvessel density (iMVD), is emerging tumours. as a powerful prognostic and a predictive marker. The iMVD Adhesion molecules has since been reported as a possible prognostic indicator in numerous human solid tumours Reduced tumour cell adhesion is associated with including breast (107), prostate (110) and invasive and prognosis. growth unfavourable colorectal (105) carcinomas. homophilic E-cadherin is a calcium-regulated adhesion molecule. The extracellular domain of In canine mammary tumours (26,80,83), squamous 60 cell carcinomas (55) and seminomas (81), angioge- useful in grading mast cell tumours (37). nesis is greater in less differentiated tumours, In the canine tumours, the prognostic value of malignant tumours and in tumours that have immunohistochemical expression of p53 protein metastasized than in benign tumours. In canine has been studied in a limited number of tumours. transmissible venereal tumour, progressing In mammary tumours (109), p53 protein overex- tumours had higher microvessel counts at the pression has been interpreted as a possible indica- invasive edges of the tumours than regressing tion of shortened survival time. Studies on canine tumours (62). These results suggest that increased epithelial colorectal (112) and cutaneous mast cell angiogenesis may be associated with increased (24,37) tumours, suggested that p53 protein tumour cell proliferation, invasion and metastasis expression is not a useful prognostic marker. in canine tumours. However, little research has been done on the prognostic and predictive value CONCLUSION of angiogenesis in canine tumours. In canine mast Ideally, a good diagnostic, prognostic cell tumours (77) angiogenesis was an independent or predictive marker should have a high specificity, prognostic factor. sensitivity, and reproducibility, and its assessment should be practical, easy and cost effective. As the progress p53 PROTEIN in molecular techniques in oncology advances, so Uncontrolled proliferation is a hallmark of cancer does the understanding of tumour biology cells. Therefore, genes and proteins regulating the (initiation, promotion, progression and regression), transition of the GI to S phase of the cell cycle and more new prognostic markers with high have been implicated in the development of human sensitivity and specificity will be found and used and animal cancers. One of the most studied cell in clinical assays. The data on the several markers cycle regulator in human and animal tumours is discussed in this review demonstrate, that many the p53 tumour suppressor gene and its protein molecular markers have a potential to be used for product. The protein product is vital for arresting prognostic evaluation and prediction of response to the cell cycle in cells with damaged DNA and therapy of some canine tumours. However, this is inducing transcription of DNA repair enzymes only true if the mean value of the level of during this arrest (2). expression of these markers is determined in series of tumours. Overlapping results have been Mutations of p53 gene occur in many human reported. Therefore, the data is not compelling tumours (78) and in some canine tumours (66,96). enough at this time to warrant routine use of these Mutant p53 protein is more stable and has an markers. Most of these prognostic markers are extended half-life time than the wild type p53 expensive and not standardized, and, therefore, are protein, and this allows detection of the mutant still limited to research. Before we think of clinical protein in tissues from tumours by immunohisto- of application these markers, additional chemistry. Therefore, immunohistochemical over- well-designed studies over large and diverse expression of p53 protein has been demonstrated samples are needed that will standardize assay in about half of the human tumours (2). method and interpretation criteria. In addition, In canine tumours, immunohistochemical overex- multivariate analysis of prognostic and predictive pression of p53 protein has been reported in some factors should be performed to test the combined tumours including osteogenic tumours (54,91), influence of several markers on prognosis and epithelial colorectal tumours (112), mast cell response to therapy. For now, no ideal molecular tumours (24,37), seminomas and Sertoli-cell prognostic marker has been discovered for any tumours, but not in Leydig-cell tumours (35), therefore, canine tumour, the combination of mammary tumours (49,109) and malignant features and histopathological clinicopathological lymphomas (102). Overexpression of p53 was grading and mitotic index from haematoxylin and found to be of diagnostic significance in eosin stained sections, seems to be more practical distinguishing osteosarcomas from chondrosarco- for the diagnostic and prognostic evaluation of mas or multilobular tumours of the bone (54), and canine tumours. 61 16. Erdelyi I, Nieskens DH, van Dijk JE Vass L and REFERENCES Nederbragt H. Immunohistochemical evaluation of versi- can, in relation to chondroitin sulphate, in canine 1. Abadie JJ, Amardeilh MA and Delverdier ME. Immuno- nuclear mammary tumours. Histology and Histopathology 2003; cell histochemical detection of proliferating 18: 1067-1080. . antigen and Ki-67 in mast cell tumors from dogs. Journal 17 Evanko SP, Angello JC and Wight TN. Formation of of the American Veterinary Medical Association 1999; hyaluronan- and versican-rich pericellular matrix is 215: 1629-1634. required for proliferation and migration of vascular 2. Agarwal ML, Taylor WR, Chernov MV, Chernova OB smooth muscle cells. Arteriosclerosis, Thrombosis, and and Stark GR. The p53 network. Journal of Biological Vascular Biology 1999; 19: 1004-1013. Chemistry 1998; 273: 1-4. 18. Faustino AM, van Garderen E, Schalken JA and 3. Auvinen P. Tammi R, Parkkinen J, Tammi M, Agren U, Nederbragt H. Tenascin expression in normal, hyperplas- Johansson R, Hirvikoski P, Eskelinen M and Kosma VM. tic, dysplastic and neoplastic canine mammary tissues. Hyaluronan in peritumoral stroma and malignant cells Journal of Comparative Pathology 2002; 126: 1-8. spreading and predicts associates with breast cancer 19. Fisher DE. Apoptosis 156; in cancer therapy: crossing the of Pathology 2000: survival. American Journal threshold. Cell 1994; 78: 539-542. 529-535. 20. Folkman J. What is the evidence that tumors are angioge- 4. Bostock DE, Crocker J, Harris K and Smith P. Nucleolar nesis dependent? Journal of the National Cancer Institute organiser regions as indicators of post-surgical prognosis 1990; 82: 4-6. in canine spontaneous mast cell tumours. British Journal 21. Fournel-Fleury C, Magnol JP, Chabanne L, Ghernati I, of Cancer 1989; 59: 915-918. Marchal T, Bonnefond C, Bryon PA and Felman P. Bostock DE and Dye MT. Prognosis after surgical exci- Growth fractions in canine non-Hodgkin's lymphomas as sion of canine fibrous connective tissue sarcomas. Veteri- determined in situ by the expression of the Ki-67 antigen. nary Pathology 1980; 17: 581-588. Journal of Comparative Pathology 1997; 117: 61-72. 6. Bostock DE. Prognosis after surgical excision of canine 22. Funakoshi Y, Nakayama H, Uetsuka K, Nishimura R, melanomas. Veterinary Pathology 1979; 16: 32-40. Sasaki N and Doi K. Cellular proliferative and telomerase 7. Bratulic M, Grabarevic Z, Artukovic B and Capak D. activity in canine mammary gland tumors. Veterinary Number of nucleoli and nucleolar organizer regions per Pathology 2000; 37: 177-183. nucleus and nucleolus--prognostic value in canine mam- 23. Gerdes J, Lemke H, Baisch H, Wacker HH, Schwab U mary tumors. Veterinary Pathology 1996; 33: 527-532. and Stein H. Cell cycle analysis of a cell proliferation- 8. Bravo R, Frank R, Blundell PA and Macdonald-Bravo H. DNA associated human nuclear antigen defined by the monoclo- protein of auxiliary Cyclin/PCNA is the nal antibody Ki-67. Journal of Immunology 1984; 133: polymerase-delta. Nature 1987; 326: 515-517. 1710-1715. Bravo R and Macdonald-Bravo H. Changes in the nuclear 24. Ginn PE, Fox LE, Brower JC, Gaskin A, Kurzman ID distribution of cyclin (PCNA) but not its synthesis depend and Kubilis PS. Immunohistochemical detection of p53 on DNA replication. EMBO Journal 1985; 4: 655-661. tumor-suppressor protein is a poor indicator of prognosis 10. Brown PJ, Rema A and Gartner F. Immunohistochemical for canine cutaneous mast cell tumors. Veterinary characteristics of canine aortic and carotid body tumours. Pathology 2000; 37: 33-39. Journal of Veterinary Medicine A. Physiology, Pathology, 25. Gonzalez CM, Griffey SM, Naydan DK, Flores E, Clinical Medicine 2003; 50: 140-144. Cepeda R, Cattaneo G and Madewell BR. Canine trans- 11. Brunetti B, Sarli G, Preziosi R, Leprotti S and Benazzi missible venereal tumour: a morphological and C. E-cadherin expression in canine mammary carcinomas immunohistochemical study of 11 tumours in growth with regional lymph node metastases. Journal of Veterina- phase and during regression after chemotherapy. Journal Medicine ry Medicine A. Physiology, Pathology, Clinical of Comparative Pathology 2000; 122: 241-248. 2003; 50: 496-500. 26. Graham JC and Myers RK. The prognostic significance of 12. Cattoretti G, Becker MH, Key G, Duchrow M, Schluter angiogenesis in canine mammary tumors. Journal of C, Galle J and Gerdes J. Monoclonal antibodies against Veterinary Internal Medicine 1999; 13: 416-418. recombinant parts of the Ki-67 antigen (MIB 1 and MIB 27. Gratzner HG. Monoclonal antibody in microwave-processed to 5-bromo- and 3) detect proliferating cells Journal of Pathology 5-iododeoxyuridine: A new reagent for detection of DNA sections. formalin-tixed paraffin replication. Science 1982; 218: 474-475. 1992; 168: 357-363. 28. Graves GM, Bjorling DE and Mahaffey 13. Della Saida L, Preziosi R, Mazzoni M and Marcato PS. E. Canine hemangiopericytoma: 23 cases (1967-1984). in canine infundibular Cell proliferation patterns Journal of the American Veterinary Medical Association keratinizing acanthorna and well differentiated squamous 1988; 192: 99-102. cell carcinoma of the skin. European Journal of Histo- 29. Griffey SM, Kraegel SA and Madewell BR. Proliferation chemistry 2002; 46: 165-172. indices in spontaneous canine lung cancer: proliferating 14. Destexhe E, Vanmanshoven P and Coignoul F. Compari- organizer regions by cell nuclear antigen (PCNA), Ki-67 (MIB1) and mitotic nucleolar . son of argyrophilic counts. Journal of Comparative Pathology counting and image analysis in canine mammary tumors. 1999; 120: 321-332. 1995; 56: of Veterinary Research American Journal 30. Guvenc T, Haligur M, Orman MN and Haziroglu 185-187. R. Mitosis and apoptosis in canine cutaneous histiocytoma 15. De Vico G, Agrimi U and Maiolino P. Nucleolar size and and transmissible venereal tumour. mitotic index in basal cell carcinomas (BCC) and squa- Acta Veterinaria Hungarica 2002; 50: 315-321. mous cell carcinomas (SCC) of canine skin. Zentralblatt fur Veterinarmedizin. Reihe A 1994; 41: 76-79. 31. Harmelin A, Zuckerman A and Nyska A. Correlation of 62 Ag-NOR protein measurements with prognosis in canine Bergstrom R, Gerdin B, Pahlman L and Glimelius B. transmissible venereal tumour. Journal of Comparative Stromal tenascin distribution as a prognostic marker in Pathology 1995; 1 12: 429-433. colorectal cancer. British Journal of Cancer 1997; 76: 32. Hinrichs U, Rutteman GR and Nederbragt H. Stromal 526-530. accumulation of chondroitin sulphate in mammary 47. Lana SE, Ogilvie GK and Hansen RA. Identification of tumours of dogs. British Journal of Cancer 1999; 80: matrix metalloproteinases in canine neoplastic tissue. 1359-1365. American Journal of Veterinary Research 2000; 61: 33. Hojilla CV, Mohammed FF and Khokha R. Matrix metal- 111-114. loproteinases and their tissue inhibitors direct cell fate 48. Laprie C, Abadie J, Amardeilh MF, Net JL, Lagadic M during cancer development. British Journal of Cancer and Delverdier M. MIB-1 immunoreactivity correlates 2003; 89: 1817-1821. with biologic behaviour in canine cutaneous melanoma. 34. Hung C, Pong VF, Cheng CR, Wong FI and Chu RM. Veterinary Dermatology 2001; 12: 139-147. An improved system for quantifying AgNOR and PCNA 49. Lee CH, Kim WH, Lim JH, Kang MS, Kim DY and in canine tumors. 20: Anticancer Research 2000; Kweon OK. Mutation and overexpression of p53 as a 3273-3280. prognostic factor in canine mammary tumors. Journal of 35. Inoue M and Wada N. Immunohistochemical detection of Veterinary Science 2004; 5: 63-69. p53 and p21 proteins tumours. in canine testicular 50. Leibman NF, Lana SE, Hansen RA, Powers BE, Fettman Veterinary Record 2000; 146: 370-372. MJ, Withrow SJ and Ogilvie GK. Identification of matrix 36. Ioachim E. Charchanti A, Briasoulis E, Karavasilis V, metalloproteinases in canine cutaneous mast cell tumors. Journal Tsanou H, Arvanitis DL, Agnantis NJ and Pavlidis N. of Veterinary Internal Medicine 2000, 14: Immunohistochemical expression of extracellular matrix 583-586. components tenascin, fibronectin, collagen type IV and 51. Lipponen PK and Aaltomaa S. Apoptosis in bladder laminin in breast cancer: their prognostic value and role . cancer as related to standard prognostic factors and in tumour invasion and progression. European Journal of prognosis. Journal of Pathology 1994; 173: 333-339. Cancer 2002; 38: 2362-2370. 52. Lohr CV, Teitke JP, Failing K, Weiss E. Characterizati- 37. Jaffe MH, Hosgood G, Taylor HW, Kerwin SC, Hedlund on of the proliferation state in canine mammary tumors by CS, Lopez MK, Davidson JR, Miller DM and Paranjpe the standardized AgNOR method with postfixation and immunohistologic M. Immunohistochemical and clinical evaluation of p53 in detection of Ki-67 and PCNA. canine cutaneous mast cell tumors. Veterinary Pathology Veterinary Pathology 1997; 34: 212-221. 2000; 37: 40-46. 53. Loukopoulos P. Mungall BA, Straw RC, Thornton JR and Robinson WF. Matrix 38. Johnson GC, Miller MA and Ramos-Vara JA. Compari- metalloproteinase-2 and -9 involvement in canine son of argyrophilic nucleolar organizer regions (AgNORs) tumors. Veterinary Pathology and mitotic index in distinguishing benign from malignant 2003; 40: 382-394. canine smooth muscle tumors and in separating inflamma- 54. Loukopoulos P, Thornton JR and Robinson WF. Clinical and pathologic relevance of p53 index in canine tory hyperplasia from neoplastic lesions of the urinary osseous bladder mucosa. Journal of Veterinary Diagnostic Investi- tumors. Veterinary Pathology 2003; 40: 237-248. gation 1995; 7: 127-136. 55. Maiolino P. Papparella S. Restucci B and De Vico G. 39. Juntes P and Pogacnik M. Morphometric analysis of Angiogenesis in squamous cell carcinomas of canine skin: AgNORs in tubular and papillary parts of canine an immunohistochemical and quantitative analysis. Journal mammary gland tumors. Analytical and Quantitative of Comparative Pathology 2001; 125: 117-121. Cytology and Histology 2000; 22: 185-192. 56. Maiolino P, Restucci B and De Vico G. Expression of 40. Kerr JFR, Winterford CM and Harmon BV. Apoptosis: proliferating cell nuclear antigen in basal cell carcinomas Its significance in cancer and cancer therapy. Cancer and in squamous cell carcinomas of canine skin: correlati- 1994; 73: 2013-20126. with mitotic index and on histological features. 41. Keshgegian AA and Cnaan A. Proliferation markers in Zentralblatt fur Veterinarmedizin. Reihe A 1995; 42: breast carcinoma. Mitotic figure count, S-phase fraction, 339-343. proliferating cell nuclear antigen, Ki-67 57. McEntee MF and Brenneman KA. Dysregulation of beta- and. MIB-1. American Journal of Clinical Pathology 1995; 104: 42-49. catenin is common in canine sporadic colorectal tumors. 42. Kiupel M, Bostock D and Bergmann V. The prognostic Veterinary Pathology 1999; 36: 228-236. significance of AgNOR counts and PCNA-positive cell 58. Millanta F, Fratini F, Corazza M, Castagnaro M, counts in canine malignant lymphomas. Journal of Zappulli V and Poli A. Proliferation activity in oral and Comparative Pathology 1998; 119: 407-418. cutaneous canine melanocytic tumours: correlation with 43. Kiupel M, Teske E and Bostock D. Prognostic factors for histological parameters, location, and clinical behaviour. treated canine malignant lymphoma. Veterinary Pathology Research in Veterinary Science 2002; 73: 45-51. 1999; 36: 292-300. 59. Minkus G, Jutting U, Aubele M, Rodenacker K, Gais P, 44. Knudson CB and Knudson W. Hyaluronan-binding pro- Breuer W and Hermanns W. Canine neuroendocrine tu- teins in development, tissue homeostasis, and disease. mors of the pancreas: a study using image analysis techni- FASEB Journal 1993; 7: 1233-1241. ques for the discrimination of metastatic versus nonmeta- 45. Koukoulis GK, Gould VE, Bhattacharyya Gould JE. static tumors. Veterinary Pathology 1997; 34: 138-145. Howeedy AA and Virtanen I. Tenascin in normal, reacti- 60. Mohammed SI, Bennett PF, Craig BA, Glickman W, ve, hyperplastic, and neoplastic tissues: biologic and Mutsaers AJ, Snyder PW, Widmer WR, DeGortari AE, pathologic implications. Human Pathology 1991; 22: Bonney PL and Knapp DW. Effects of the cyclooxygena- 636-643. se inhibitor, piroxicam, on tumor response, apoptosis, and 46. Kressner U, Lindmark G, Tomasini-Johansson B, angiogenesis in a canine model of human invasive urinary 63 Diagnostic Investigation 2000; 12: 111-117. bladder cancer. Cancer Research 2002; 62: 356-358. 75. Platz SJ, Breuer W, Pfleghaar 61. Mukaratirwa S, Chimonyo M, Obwolo M, Gruys E and S. Minkus G and Hermanns W. cells and extracellular matrix Prognostic value of histopathological Nederbragt H. Stromal grading in canine extramedullary components in spontaneous canine transmissible venereal plasmacytomas. Histology and Veterinary Pathology 1999; 36: 23-27. tumour at different stages of growth. 76. Preziosi R, Sarli G, Benazzi C and Marcato PS. Detecti- Histopathology 2004; 19: 1117-1123. on of proliferating cell nuclear antigen (PCNA) in canine 62. Mukaratirwa S. Chiwome T, Chitanga T, and Bhebhe E. and feline mammary tumours. Journal of Comparative Canine transmissible venereal tumour: assessment of mast Pathology 1995; 113: 301-313. cell numbers as indicators of growth stage. Veterinary 77. Preziosi R, Sarli G and Paltrinieri M. Prognostic value of Research Communications 2005; in press. van Ederen AM and intratumoral vessel density in cutaneous mast cell 63. Mukaratirwa E, S, de Witte tumors of the dog. Journal of Comparative Pathology 2004; 130: Nederbragt H. Tenascin expression in relation to stromal 143-151. tumour cells in canine gastrointestinal epithelial tumours. 78. Prokocimer M and Rotter V. Structure and function of Journal of Comparative Pathology 2003; 129: 137-146. p53 in normal cells and their aberrations in 64. Mukaratirwa S, van Ederen AM, Gruys E and Nederbragt cancer cells: projection on the hematologic cell H. Versican and hyaluronan expression in canine colonic lineages. Blood 1994; 84: 2391-2411. adenomas and carcinomas: relation to malignancy and 79 Reis AL, Carvalheira J, Schmitt of Comparative depth invasion. Journal FC, and Gartner F. of tumour Immunohistochemical study of the Pathology 2004; 131: 259-270. expression of E-cadherin in canine mammary 65. Naik P, Karrim J and Hanahan D. The rise and fall of tumours. Veterinary down- Record 2003; 152: 621-624. tumorigenesis: apoptosis multistage during from progression 80. Restucci B, De Vico G and Maiolino P. Evaluation of to tumor modulation contributes angiogenesis in canine mammary tumors by quantitative angiogenic progenitors. Genes and Development 1996; platelet endothelial cell adhesion molecule immunohisto- 10: 2105-2116. chemistry. Veterinary Pathology 2000; 37: 297-301. 66. Nasir L, Rutteman GR, Reid SW, Schulze C and Argyle 81. Restucci B, Maiolino P. Paciello 0, Martano M, De Vico DJ. Analysis of p53 mutational events and MDM2 ampli- G and Papparella S. Evaluation of angiogenesis in canine fication in canine soft-tissue sarcomas. Cancer Letters seminomas by quantitative immunohistochemistry. Journal 2001; 174: 83-89. of Comparative Pathology 2003; 128; 252-259. 67. Nieto A, Pena L, Perez-Alenza MD, Sanchez MA, Flores of detection 82. Restucci B, Papparella S. De Vico G and Maiolino P. E JM and Immunohistologic Castano M. cadherin expression canine mammary tumors: in in normal and neoplastic estrogen receptor alpha canine and prognostic mammary gland. Journal of Comparative Pathology 1997; associations clinical and pathologic 116: 191-202. significance. Veterinary Pathology 2000; 37: 239-247. 83. Restucci B, Papparella S, Maiolino P and De Vico G. Bennett MW, O'Sullivan GC, Roche D, 68. O'Connell J, Expression of vascular endothelial growth factor in canine Kelly J, Collins JK and Shanahan F. Fas ligand expressi- mammary that the tumors. Veterinary Pathology on in primary colon adenocarcinomas: evidence 2002; 39, 488-493. immune a prevalent mechanism of Fas counterattack is 84. Rhodes DR, Sanda MG, Otte AP, Chinnaiyan AM evasion in human colon cancer. Journal of Pathology and Rubin MA. Multiplex biomarker approach for determi- 1998; 186: 240-246. ning risk of prostate-specific antigen-defined 69. Ogura E, Senzaki H and Kiyozuka Y. Expression and recurrence of tissue and prostate cancer. Journal of the National Cancer Institute metalloproteinases localization of matrix 2003; 95: 661-668 inhibitors of metalloproteinases as a prognostic factor in advanced colorectal carcinomas. Oncology Reports 1997; 85. Ricciardelli C, Brooks JH, Suwiwat S, Sakko AJ, Mayne K, Raymond WA, Seshadri R, LeBaron RG and Horsfall 4: 701-705. DJ. Regulation of stromal versican expression by breast 70. Ohta G, Sakai H, Kachi S, Hirata A, Yonemaru K, Kitajima A, Yanai T and Masegi T. Assessment of cancer cells and importance to relapse-free survival in and patients osteosarcomas with node-negative of canine primary breast proliferative potentials cancer. Clinical Cancer Research 2002; 8: 1054-1060. chondrosarcomas by MIB-1 immunohistochemistry and 86. Ricciardelli C, Mayne K, Sykes bromodeoxyuridine incorporation. Journal of Comparative PJ, Raymond WA, McCaul K, Marshall VR and Horsfall D. Elevated Pathology 2004; 131: 18-27. levels of versican but not decorin predict disease progression 71. Papparella S, Restucci B, and Maiolino P. Immunohisto- in early-stage prostate cancer. chemical distribution of type IV collagenase in normal, Clinical Cancer Research 1998; 4: 963-971. dysplastic and neoplastic canine mammary gland. Journal 87. Roels S, Tilmant K and Ducatelle R. P53 of Comparative Pathology 1997; 117: 277-282. expression and proliferation. 72, Pardee AB. GI events and regulation of cell apoptosis in melanomas of dogs and cats. Research in Veterinary Science 2001; 70: 19-25. Science 1989; 246: 603-608. AI, Perez-Alenza D, Cuesta P and 88. Roels S. Tilmant K and Ducatelle R. PCNA 73. Pena LL, Nieto and Ki67 Castano M. Immunohistochemical detection of Ki-67 and proliferation markers as criteria for prediction of clinical PCNA in canine mammary tumors: relationship to clinical behaviour of melanocytic tumours in cats and dogs. and pathologic variables. Journal of Veterinary Diagnostic Journal of Comparative Pathology 1999; 121: 13-24. 89. Ropponen K, Tammi M, Parkkinen Investigation 1998; 10: 237-246. J, Eskelinen M, 74. Phillips BS, Kass PH, Naydan DK, Winthrop MD, Tammi R, Lipponen P, Agren U, Alhava E and Kosma VM. Tumor cell-associated hyaluronan Griffey SM and Madewell, BR. Apoptotic and proliferati- as an unfavorable of Veterinary on indexes in canine lymphoma. Journal prognostic factor in colorectal cancer. Cancer Research 64 1998; 58: 342-347. 55 cases, with observations on p53 immunoexpression. Rudas M, Gnant MF, Mittlbock M, Neumayer R, Journal of Comparative Pathology 2004; 131: 207-213. Kummer A, Jakesz R, Reiner G and Reiner A. Thymidi- 103. Takeichi M. Cadherins in cancer: implications for inva- ne labeling index and Ki-67 growth fraction in breast sion and metastasis. Current Opinion in Cell Biology cancer: comparison and correlation with prognosis. 1993; 5: 806-811. Breast Cancer Research and Treatment 1994; 32: 104. Talvensaari-Mattila A, Paakko P and Hoyhtya M. 165-175. Matrix metalloproteinase-2 immunoreactive protein. A Sagartz JE, Bodley WL, Gamblin RM, Couto CG, marker of aggressiveness in breast carcinoma. Cancer Tierney LA and Capen CC. p53 tumor suppressor 1998; 83: 1153-1162. protein overexpression in osteogenic tumors of dogs. 105. Tanigawa N, Amaya H, Matsumura M, Lu C, Kitaoka Veterinary Pathology 1996; 33: 213-2121. A, Matsuyama K and Muraoka R. Tumor angiogenesis Sakai H, Noda A, Shirai N, Iidaka T, Yanai T, Masegi and mode of metastasis in patients with colorectal T, Shirai N, Iidaka T, Yanai T and Masegi T. Prolifera- cancer. Cancer Research 1997; 57: 1043-1046. tive activity of canine mast cell tumours evaluated by 106. Theon AP, Lecouteur RA, Carr EA and Griffey SM. bromodeoxyuridine incorporation and Ki-67 expression. Influence of tumor cell proliferation and sex-hormone Journal of Comparative Pathology 2002; 127: 233-238. receptors on effectiveness of radiation therapy for dogs Sarli G, Benazzi C, Preziosi R and Marcato PS. Prolife- with incompletely resected meningiomas. Journal of rative activity assessed by anti-PCNA and Ki-67 mono- American Veterinary Medical Association 2000; 216: clonal antibodies in canine testicular tumours. Journal of 701-707, 684-685. Comparative Pathology 1994; 110: 357-368. 107. Toi M, Kashitami J and Tominaga T. Tumour angioge- Sarli G, Preziosi R, Benazzi C, Castellani G, Marcato nesis is an independent prognostic indicator of breast PS. Prognostic value of histologic stage and proliferative carcinoma. International Journal of Cancer 1993; 55: activity in canine malignant mammary tumors. Journal 37 1-374 . Veterinary Diagnostic Investigation 2002; 14: 25-34. 108. Vail DM, Kisseberth WC, Obradovich JE, Moore FM, Schweyer S, Soruri A, Meschter 0, Heintze A, London CA, MacEwen EG and Ritter MA. Assessment Zschunke F, Miosge N, The len P. Schlott T, Radzun HJ of potential doubling time (Tpot), argyrophilic nucleolar and Fayyazi A. Cisplatin-induced apoptosis in human organizer regions (AgNOR), and proliferating cell nucle- malignant testicular germ cell lines depends on MEK/ ar antigen (PCNA) as predictors of therapy response in ERK activation. British Journal of Cancer 2004; 91: canine non-Hodgkin's lymphoma. Experimental Hemato- 589-598. logy 1996; 24: 807-815. Setoguchi A, Sakai T, Okuda M, Minehata K, Yazawa 109. Wakui S, Muto T, Yokoo K, Yokoo R, Takahashi H, M, Ishizaka T, Watari T, Nishimura R, Sasaki N, Masaoka T, Hano H and Furusato M. Prognostic status Hasegawa A and Tsujimoto H. Aberrations of the p53 of p53 gene mutation in canine mammary carcinoma. tumor suppressor gene in various tumors in dogs. Anticancer Research 2001; 21: 611-616. American Journal of Veterinary Research 2001; 62: 110. Weidner N, Carroll PR, Flax J, Blumenfeld W and 433-439. Folkman J. Tumor angiogenesis correlates with metasta- Siitonen SM, Kononen JT, He lin HJ, Rantala IS, Holli sis in invasive prostate carcinoma. American Journal of KA and Isola JJ. is Reduced E-cadherin expression Pathology 1993; 143: 401-409. associated with invasiveness and unfavorable prognosis 111. Wiksten JP, Lundin J, Nordling S, Lundin M, Kokkola in breast cancer. American Journal of Clinical Pathology A, von Boguslawski K and Haglund C. Tenascin-C 1996; 105: 394-402. expression correlates with prognosis in gastric cancer. 98. Simoes JP, Schoning P and Butine M. Prognosis of Oncology 2003; 64: 245-250. canine mast cell tumors: a comparison of three methods. 112. Wolf JC, Ginn PE, Homer B, Fox LE and Kurzman ID. Veterinary Pathology 1994; 31: 637-647. Immunohistochemical detection of p53 tumor suppressor 99. Sirri V, Roussel P and Hernandez-Verdun D. The in gene protein canine epithelial colorectal tumors. AgNOR proteins: qualitative and quantitative changes Veterinary Pathology 1997; 34: 394-404. during the cell cycle. Micron 2000; 31: 121-126. 113. Wyllie AH, Kerr JFR and Currie AR. Cell death: the 100. Spangler WL, Culbertson MR and Kass PH. Primary of significance apoptosis. International Review of mesenchymal (nonangiomatous/nonlymphomatous) neo- Cytology 1980; 68: 251-304. plasms occurring in the canine spleen: anatomic classifi- 114. Yoshida K, Yanai T, Iwasaki T, Sakai H, Ohta J, Kati cation, immunohistochemistry, and mitotic activity cor- S, Ishikawa K, Lackner AA and Masegi T. Proliferative related with patient survival. Veterinary Pathology 1994; potential of canine oral epulides and malignant neo- 31: 37-47. plasms assessed by bromodeoxyuridine labeling. Veteri- 101. Stearns M and Stearns ME. Evidence for increased acti- nary Pathology 1999; 36: 35-41. vated metalloproteinase 2 (MMP-2a) expression associa- 115. Zuccari DA, Santana AE, Cury PM and Cordeiro JA. ted with human prostate cancer progression. Oncology Immunocytochemical study of Ki-67 as a prognostic Research 1996; 8: 69-75. t marker in canine mammary neoplasia. Veterinary Clini- 102. Sueiro FA, Alessi AC and Vassal lo J. Canine lympho- cal Pathology 2004; 33: 23-28. mas: a morphological and immunohistochemical study of http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Veterinary Quarterly Taylor & Francis

Prognostic and predictive markers in canine tumours: Rationale and relevance. A review

Veterinary Quarterly , Volume 27 (2): 13 – Jun 1, 2005

Loading next page...
 
/lp/taylor-francis/prognostic-and-predictive-markers-in-canine-tumours-rationale-and-GroAfwtOGl

References (117)

Publisher
Taylor & Francis
Copyright
Copyright Taylor & Francis Group, LLC
ISSN
1875-5941
eISSN
0165-2176
DOI
10.1080/01652176.2005.9695186
pmid
16022341
Publisher site
See Article on Publisher Site

Abstract

Veterinary Quarterly 2005; 27(2): 52-64 Prognostic and predictive markers in canine tumours: Rationale and relevance. A review S. Mukaratirwa* Department of Paraclinical Studies, Pathology Section, Faculty of Veterinary Science, University of Zimbabwe TABLE OF CONTENTS Summary and keywords Abbreviations Introduction Proliferation markers Mitotic count/index Incorporation techniques (thymidine and BrDU) Proliferating cell nuclear antigen (PCNA) Ki-67 Argyrophilic nucleolar organiser region (AgNOR) proteins Predictive value of proliferation markers for therapy response of canine tumours 56 Apoptosis Apoptosis and prognosis in canine tumours Extracellular matrix and adhesion molecules Extracellular matrix molecules Matrix metalloproteinases Adhesion molecules Angiogenesis p53 protein Conclusion References . . . . 61 *Address: Department of Paraclinical Studies. Pathology Section, Faculty of Veterinary Science, University of Zimbabwe, P.O. Box MP 167, Mount Pleasant, Harare, Zimbabwe. E-mail: tmukaratirwa@vet.uz.ac.zw 53 Prognostic and predictive markers in canine tumours: Rationale and relevance. A review S. Mukaratirwa SUMMARY Cancer is one of the most common causes of death in dogs. The availability of multiple treatment modalities and novel therapeutic targets make the correct diagnosis, prognostic stratification, and the identification of treatment effect predictive factors an issue of major debate in cancer management. Selection of high and low risk patients and the type of systemic or local treatment is important in cancer management. The search for better prognostic markers and predictive factors is now focused on the molecular mechanisms which underlie tumour behaviour, such as altered cell cycle progression, proliferation, apoptosis, and angiogenesis. The ultimate goal is to identify reliable markers that can accurately diagnose and stage a tumour and predict tumour's clinical behaviour, prognosis and response to therapy. In this review, the current state of prognostication in canine tumours and promising new molecular markers are discussed. The markers are allocated to four groups according (" °proliferation markers, mapoptosis, extracellular to their function: matrix and cell adhesion molecules, (i")angiogenesis and °cell cycle regulators. Clinicopathological factors and histopathological grading remain the most practical parameters in decision-making. Although experimental research has shown that molecular markers have a good potential to be used as diagnostic, prognostic or predictive markers in canine tumours, insufficient evidence exists on their efficacy for routine use in veterinary oncology. Keywords: Adhesion molecules; Angiogenesis; Apoptosis; Canine neoplasms; Cell cycle; Dog diseases; Molecular markers; p53 protein; Prognosis; Prognostic and predictive markers; Proliferation; Proliferation markers; Tumour behaviour; Tumour grading; Tumour markers; Tumour staging; Tumours; Veterinary oncology. Abbreviations AgNOR: Argyrophilic nucleolar organiser region iMVD: Intratumoural microvessel density MMPs: Metalloproteinases BrDU: Bromodeoxyuridine ECM: Extracellular matrix PCNA: Proliferating cell nuclear antigen evidence of distant metastases have long INTRODUCTION been recognized, but this is often not enough to predict The demonstration of clinicopathological factors accurately the prognosis and response to therapy of and biological markers with prognostic significan- the tumour. Inconsistencies between histopatholo- ce has become one of the most important fields of gical features and biological behaviour or progno- study in medical and veterinary oncology. These sis have been documented in several canine factors to discriminate are used and markers tumours. Therefore, a considerable number of benign from malignant tumours, grade tumours molecular markers have been applied to diagnose according to aggressiveness and metastatic risk, and to predict the prognosis of neoplastic diseases and to predict the prognosis, disease-free survival in dogs. after surgical rates rates and overall survival The aim of this paper is to review biomarkers excision or therapy of the tumour. used, (uto differentiate histologically similar The grading of malignancy and prognosis of a tumours, (")to discriminate malignant from benign tumour based on the classical clinicopathological neoplastic diseases and classify some of the factors and histopathological criteria such as size tumours into grades .according to aggressiveness and presence of location of the tumour, and metastatic risk, predict the prognosis of microscopic regional lymph node metastases and neoplastic disease, and (1")to predict response to 54 therapy of neoplastic diseases. The rationale and cell size and the effect of time before sample relevance of these markers is also discussed. fixation on mitosis. In some instances pathologists Moreover, this review is intended to raise interest may have difficulties in discriminating mitoses and in the use of these markers in veterinary oncology. necroses (karyorrhectic and pyknotic nuclei), and In this review, the term diagnostic marker means a prophase and anaphase are difficult to identify in marker that can help to distinguish histologically haematoxylin and eosin stained sections. In spite similar tumours; the term marker prognostic of these shortcomings, determination of mitotic means a marker that correlates with malignancy, count/index on haematoxylin and eosin stained disease-free survival rate and overall survival rate; tumours has some clinical implications. The major the term treatment effect predictive marker advantage is that histopathologic diagnosis and indicates a marker that is capable of predicting mitotic count/index can be determined at the same tumour sensitivity or resistance to various time, making it the most convenient method of therapies evaluating cell proliferation. In veterinary oncology, mitotic count/index has PROLIFERATION MARKERS been used by histopathologists both as a diagnostic aid and prognostic indicator for several tumours. Cell proliferation is the increase in cell number Mitotic count/index can be useful in separating resulting from completion of the cell cycle (72). several benign tumours from their malignant The proliferation rate of tumour cells is a principal counterparts e.g. canine mammary tumours (76) determinant of the aggressiveness of a tumour and, and canine smooth muscle tumours of urinary therefore, an important prognostic factor. Many bladder (38), and may also be employed in different methods have been developed to detect separating tumours with similar histological cell proliferation in tumours. These methods differ appearance e.g. canine basal cell carcinomas from not only in what they detect, but also in the phases squamous cell carcinomas (15). of the cell cycle that they label the cell. Therefore, these methods often produce different results. Mitotic count/index has been shown to be of prognostic value in several canine tumours Methods used to estimate proliferation of tumour including melanomas (6), fibrous connective tissue cells include the early methods of measuring mito- (5), sarcomas haemangiopericytomas (28), mast tic count/index and S-phase labelling with thymidi- cell tumours (4), mammary tumours (94), and ne or bromodeoxyuridine (BrDU), and recently, splenic non-angiomatous and non-lymphomatous flow cytometry, in situ hybridization, histochemi- tumours (100). A high mitotic index in these cal staining methods and immunohistochemical tumours is associated with high tumour recurrence labelling methods. The most commonly used histo- after surgical excision and with a shorter survival chemical proliferation marker is argyrophilic time of the animal. nucleolar organiser region (AgNOR), and the most commonly used immunohistochemical markers are Incorporation techniques proliferating cell nuclear antigen (PCNA) and (thymidine and BrDU) Ki-67 antigen. Each of these methods has some practical problems in routine histopathology. In vitro 3H-thymidine incorporation technique was the earliest technique available for evaluating cell Mitotic count/index proliferation. However, 3H-thymidine incorpora- tion technique is impractical because it is time Mitotic count is the number of mitoses per high consuming and has the difficulty and expense of power field (x 400 magnification) and mitotic disposing radioactive material. Because of these index is the fraction of mitoses expressed as a drawbacks, bromodeoxyuridine (BrDU) was intro- percentage of the total number of cells present. duced for the visualization of DNA synthesis in Mitotic count or index can determined on the S-phase of the cell cycle. BrDU, an analogue haematoxylin and eosin stained sections of to 3H-thymidine, is incorporated in the nucleus tumours, and therefore, a routine part of during the S-phase and can be histopathological evaluation. detected The validity of by immunohistochemistry on formalin-fixed, paraffin- mitotic count or index remains controversial. It is embedded tissue or by flow cytometry with anti- not a standardized method and does not consider 55 BrDU monoclonal antibody (27). The need for pri- PCNA index was not associated with overall or administration of BrDU to the live animal is the survival time (42). In mast cell tumours, disease- free survival and overall survival time main drawback of BrDU incorporation technique. was longer in tumours with a low PCNA index (1,98). For the use of BrDU Few studies have analysed mammary tumours, one study showed no and irnmunohistochemistry the diagnosis in significant association of a high PCNA with prognosis of canine tumours. BrDU index could be metastasis, death from neoplasia, low disease-free used to distinguish acanthomatous epulides from survival rates and low overall survival rates (73), less aggressive epulides (114), osteosarcomas from while in another study, a low PCNA index was chondrosarcomas (70), and could be used to grade associated with low disease-free survival rates mast cell tumours according to their aggressiveness (52). In melanocytic tumours, PCNA index was of acanthomatous (92). in A high BrDU index no prognostic significance (88). epulides, cell carcinomas, and fibrosar- coma correlateswith their poor prognosis (114). Ki-67 Proliferating cell nuclear antigen (PCNA) Ki-67 protein is a non-histone nuclear protein present during all active phases of the cell cycle The proliferating cell nuclear antigen (PCNA) is a (G1, S, G2, and mitosis), but is absent from quies- nuclear protein which functions as a co-factor for cent or resting cells (23). During interphase, Ki-67 DNA-polymerase delta in both the G phase and in antigen can be exclusively detected within the DNA synthesis (8). PCNA is produced in the mid- nucleus, whereas in mitosis, most of the protein is it is 01 phase and throughout the S phase but, relocated to the surface of the chromosomes. This detectable throughout the cell cycle due to its long makes Ki-67 protein an excellent marker for half life (9). It is present in all active phases of the determining the growth fraction of a given cell cell cycle (GI, S, G2), manifesting a striking in- population. Unlike proliferating cell nuclear crease in concentration during the S-phase. There- antigen (PCNA), which not only plays a role in fore, its rate of synthesis is correlated directly with DNA replication, but also in DNA repair, Ki-67 is proliferation rate of cells. However, its expression virtually restricted in its role as a proliferation may be induced by DNA damaging agents. antigen, making it a more specific determinant of There are a few cases where PCNA index has growth fraction. For this reason, antibodies against been found to be of value in separating histologi- Ki-67 are increasingly being used in human and cally similar canine tumours e.g. lung adenosqua- veterinary oncology as prognostic markers in carcinoma cell mous carcinoma and squamous different types of neoplasms. MIB-1 is the (29), and cutaneous squamous cell carcinoma and monoclonal antibody developed against the Ki-67 basal cell carcinoma (56). However, the PCNA antigen that can be used on routinely formalin- index was unable to separate testicular tumours fixed, paraffin-wax embedded tissues after cell (Sertoli into types different histological microwave antigen retrieval (12). Ki-67 antigen tumour, seminoma and Leydig cell tumour) (93). can also be detected using wet-fixed cytology imprints (115), thus providing rapid information The value of PCNA index in separating benign has about malignancy and patient outcome. tumours from their malignant counterparts been assessed in several canine tumours. In most The Ki-67 index, determined by immunohistoche- of the tumours, including mammary tumours (22, mistry on tumour tissues has been used as a 73,76), melanocytic tumours (88), perianal gland potential diagnostic marker in several canine and spindle cell tumours (34), malignant tumours tumours. It has been shown that Ki-67 index can had significantly higher PCNA index than their be used to distinguish canine aortic tumours from benign counterparts. The PCNA index has been carotid body tumours (10), cutaneous histiocytoma found to be useful in grading mast cell tumours from transmissible venereal tumour (30), keratini- according to their degree of aggression (34). zing acanthoma from well differentiated squamous cell carcinoma of the Studies on PCNA index as a prognostic marker skin (13), and classify have been done on canine lymphoid, mast cell, spontaneous lung epithelial tumours into adeno- mammary and melanocytic tumours. In lymphomas squamous and squamous cell carcinomas (29). In canine testicular tumours, however, Ki-67 without surgical or therapeutic intervention, the index 56 was of no diagnostic value in separating Of particular interest is, that the amounts of seminomas from interstitial and Sertoli cell interphase AgNOR proteins are strictly related to tumours (93). Ki-67 index was also unable to rapidity of cell proliferation (99). Therefore, classify canine extramedullary plasmacy-tomas into unlike BrDU, Ki-67 or PCNA index which different grades (75). In canine mammary tumours measures the percentage of proliferating cell, (49) and melanocytic tumours (58), Ki-67 index AgNOR count gives information on the rapidity of was able to distinguish benign from malignant cell proliferation (proliferation rate). AgNOR tumours, with malignant tumours having a higher count can be determined using histochemical index than benign tumours. staining methods on histological sections of routinely processed issue samples. Most of the studies on Ki-67 antigen expression as prognostic markers have been performed on The AgNOR count was be able to separate several mammary, lymphoid, melanocytic and mast cell canine tumours into malignant and benign tumours tumours. Several researchers found Ki-67 index to including mammary tumours (7,14,39), urinary be of prognostic significance in canine mammary bladder smooth muscle cell tumours (38), perianal tumours, with a high index being positively gland tumours (34), fibroblastic tumours (34) and correlated with increased risk of metastasis, death epithelial skin tumours (13,15). In these studies, from neoplasia, low disease-free survival and low malignant tumours had a higher AgNOR count overall survival rates (67,73,94,115). However, than their benign counterparts. one study dealing with Ki-67 immunolabelling in The prognostic value of AgNOR has, unfortunate- canine mammary of tumours reported lack ly, been done on only a few canine tumours. Two prognostic significance of this marker (52). studies performed on the prognostic value of Conflicting results on the prognostic significance AgNOR count in mast cell tumours (4,98), have of Ki-67 index in human breast tumours have also shown that, tumours with a high AgNOR count been documented (41,90). The discrepancy in had a higher chance of recurring after surgical these results can be explained by methodological excision than tumours with a low count. In issues such as sampling strategies (criteria addition, the overall survival time of dogs with employed in dividing patients into groups) and mast cell tumours was found to be lower in dogs intratumoural heterogeneity (heterogeneity in the with tumours with a high AgNOR count (98). In prevalence and biological behaviour of various study an early of mammary tumours on the histological types of tumours). prognostic significance of AgNOR count, it could In one study of canine lymphomas, the Ki-67 not be used to predict survival time of dogs with index correlated with the histologic grade of the tumours (52). However, a more recent study on tumour (21), suggesting that Ki-67 index may be canine mammary tumours (94) revealed a signifi- useful for classification of lymphomas. Other cant association between a high AgNOR count and canine tumours in which Ki-67 index was found to a short one-year survival time. be of prognostic significance include melanocytic tumours (48,58,88) and mast cell tumours (1). Predictive value of proliferation markers for Tumours in which Ki-67 index was found to be of therapy response of canine tumours no prognostic significance includes canine Tumour cell proliferation rate is important neuroendocrine tumours of the pancreas (59). for tumour growth, invasion and metastasis. There- fore, pre-operative or pre-treatment proliferation Argyrophilic nucleolar organiser region rate of a tumour may be correlated with response (AgNOR) proteins to therapy and disease outcome. A rapidly prolife- Argyrophilic nuclear organizer region (AgNOR) rating tumour should respond poorly to therapy proteins are a set of nuclear proteins that accumu- compared to a slowly growing tumour. A high late in highly proliferating cells. In proliferating proliferation index is not, however, consistently cells, the quantity of interphase AgNOR proteins associated with a poor response to therapy in all progressively increases from early G I phase, tumours. More studies have been done on the reaches a maximum value at the end of S-phase diagnostic and prognostic value of proliferation and remains constant up to the late G2 phase (99). markers than on the predictive value of these 57 ting indexes using markers. A good prognostic indicator may not be some of the proliferation necessarily a good predictive factor for therapy, markers. This does not necessarily means, that these markers are not good indicators and vice versa. of cell proliferation. It may also indicate that there is in the The markers value of proliferation variability in the course of the disease in relation and prediction chemotherapy to of response to the proliferative activity. canine radiotherapy in has been studied lymphomas and transmissible venereal tumours. APOPTOSIS Pretreatment BrDU index was predictive of first remission in lymphomas (108). For the studies The rate of tumour growth is determined by done on the predictive value of Ki-67 index in tumour cell proliferation and tumour cell death. canine lymphomas, the results obtained by two An imbalance between tumour cell proliferation (43) al. research groups are different. Kuipel et and death is fundamental for tumour growth, found no correlation between Ki-67 index and progression and regression. Therefore, qualitative response to therapy of canine lymphomas, while and quantitative changes in cancer cell death along (74) found that Ki-67 index was Phillips et al. with proliferative alterations are essential relapse free predictive of the duration of first determinants in the pathogenesis of a malignant in interval In two studies after chemotherapy. disease and its responsiveness to therapy. canine lymphomas on the use of mitotic index Apoptosis is a genetically regulated form of cell (43,74) and the four studies done on the use of death, abolishing cell populations in embryological PCNA index (42,43,74,108) as a prediction of developments and in physiological and pathological response to chemotherapy, both mitotic index and processes including tumours. Apoptosis has PCNA index were not predictive for the duration characteristic biochemical and morphological of first tumour relapse free interval and overall criteria which distinguish it from necrosis (113). survival. In the studies done on the use of AgNOR Deregulated apoptosis can promote carcinogenesis count as a prediction of response of lymphomas to in two ways. First, impaired apoptosis, either due chemotherapy, the count was not predictive for the to an uncontrolled expression of oncogenes or duration of first tumour relapse free interval (42, tumour suppressing genes, leads to an uncontrolled 43,108). In canine transmissible venereal tumour, accumulation of malignant cells and eventually poor response to chemotherapy was associated formation of cancer cells (40). Apoptosis also with a high AgNOR count (31). In a study done limits the population expansion of tumour cells on the value of PCNA index in the prediction of early in the process of tumour growth (65). In response to radiotherapy of dogs with incompletely addition, tumours develop methods to evade resected meningiomas, a high PCNA index was elimination by immune system; one such associated with a low progression-free survival mechanism involves expressing Fas ligand, which rate (106). enables them to delete (by apoptosis) anti-tumour In summary, it is apparent that increased prolifera- lymphocytes (68). However, the accumulation of tion strongly with malignancy, poor neoplastic cells is not consistently associated correlates with prognosis and response to therapy irrespective of decreased rate of apoptosis. On the contrary, high literature apoptotic activity seems to be the methodology used. The available apparent particularly in clearly argues for the assessment of tumour cell aggressive tumours such as human urinary bladder carcinoma (51). proliferation in canine tumour prognostics. Mitoses counting provides the most simple and reproduci- Ki67 and PCNA/MIB-1 ble prognostic value, Apoptosis and prognosis in canine tumours index and AgNOR count are promising alternatives There are few published data on the value of that need methodological fine tuning and standardi- apoptosis as a diagnostic or prognostic factor in sation. In this context, it is worth mentioning that canine tumours. A study by Guvenc et al. (30) the diagnostic, prognostic or predictive value of showed that apoptotic indices are useful in proliferating indexes depends on the tumour type distinguishing canine cutaneous histiocytomas from and the availability of protocols for treatment. For transmissible venereal tumours. Malignant no correlation could be certain tumour types, melanocytic (87), and malignant mammary (22) demonstrated between prognosis and the prolifera- tumours had higher apoptotic indices than their 58 benign counterparts. However, in mammary Interaction between host cells, tumour cells and tumours, no significant correlations were noted ECM molecules, adhesion molecules and proteoly- between apoptotic index and clinical features such tic enzymes can have restrictive or permissive as metastasis and tumour diameter (22). influence on the behaviour (proliferation, invasion and metastasis) of tumour cells. Most anticancer chemotherapeutic drugs, as well Therefore, the level of expression of ECM molecules, proteolytic as radiation therapy, regardless of their distinct enzymes and adhesion molecules in tumours could mechanisms of action, appear to suppress tumour be of diagnostic or prognostic value growth mainly by induction of apoptosis of tumour and might be of value in predicting the effect of therapy cells (19), and there is now in of some vitro evidence malignancies. pointing to apoptosis playing an important role in the tumour response elicited by anticancer agents Extracellular matrix molecules (95). In dogs with invasive urinary bladder cancer, piroxicam, a cyclooxygenase inhibitor caused Extracellular matrix molecules are thought to be reduction in tumour volume, and this reduction in involved in numerous cellular functions including tumour size was strongly associated with induction cell adhesion, differentiation and proliferation. of apoptosis (60). Vincristine sulphate chemothera- These molecules are over-expressed in the stroma py of canine transmissible venereal tumour causes of several human (45) and canine tumours (16,18, tumour regression by apoptosis (25). Therefore, 32,61,63,64), and this led to the assertion that the extent of apoptosis in a tumour may be of these molecules may be associated with malignan- value in predicting and monitoring response to cy and poor prognosis. The most widely studied chemotherapy or radiotherapy. However, only one ECM molecules in neoplastic diseases are tenascin- study has been performed on predictive value of C, versican and hyaluronan. apoptosis in canine tumours. In this study, pretreatment apoptotic index of lymphomas was Tenascin-C, a large ECM glycoprotein, is predictive of the duration of first relapse free implicated in cell adhesion and proliferation, and interval (74). its overproduction is associated with poor prognosis in several human tumours including In conclusion, apoptosis seems to play an (46), colorectal breast (36) and gastric (111) important role in all the stages of carcinogenesis cancer. In canine tumours, no studies have been (initiation, promotion, progression and regression), done on the prognostic significance of tenascin-C. but the clinical and biological relevance of However, studies on canine mammary tumours apoptosis of canine tumours need to be studied. In (18) and canine gastrointestinal epithelial tumours human oncology, where numerous studies have (63) have demonstrated an increased immunohisto- been performed, the relationship between a high chemical expression of tenascin-C apoptotic index and good prognosis has not been in tumours compared to normal tissues. It is likely substantiated. that other factors in tumour progression, such as other forms of tumour a member of the Versican, large aggregating cell death, proliferation rate, angiogenesis and the sulphate proteoglycan chondroitin family, is host immune response, have a confounding effect recognized anti-cell adhesive molecule involved in on tumour behaviour and response to therapy. regulating cell motility on ECM components (17). Therefore, there is a need to do a multivariate The level of versican expression is of prognostic study on these factors. value in several human tumours including prostatic (86) and breast (85) tumours. Versican expres-sion EXTRACELLULAR MATRIX AND was shown to be increased in canine mammary ADHESION MOLECULES (16) and colorectal tumours (64). Although no studies have been done on the prognostic value of Tumour invasion and metastasis results from versican in canine tumours, the level of expression aberrant communication between tumour cells and of versican in colorectal epithelial tumours was microenvironment. The microenvironment consists associated with depth of tumour invasion and local of a variety of host cells (vascular cells, fibro- lymph node metastasis (64), suggesting that blasts, myofibroblasts and inflammatory cells), versican can be of prognostic value. extracellular matrix (ECM) molecules, cell adhesion molecules and proteolytic enzymes. Hyaluronan is a linear unsulphated glycosamino- 59 E-cadherin is glycan synthesized on the cytoplasmic side of the involved in cell-cell adhesion, plasma membrane. Hyaluronan has been associated whereas the intracellular domain connects to the actin cytoskeleton via catenins. Dysfunction with many different cellular processes including or loss of the intercellular cell adhesion, proliferation, E-cadherin-beta-catenin aggregation, complex is frequent locomotion (44). Hyaluronan in many human epithelial and migration tumours (103). In some human over-production is a diagnostic or prognostic factor carcinomas including breast (97) and prostate carcinomas (84), for several types of human tumours including E-cadherin and beta-catenin loss is associated colorectal (89) and breast (3) tumours. In canine with a poor prognosis. Down-regulation of E-cadherin colorectal tumours, there is increased hyaluronan expression has been demonstrated expression compared to normal tissues (64), but no in canine mammary (11,79,82) and colorectal carcinomas studies or its prognostic have been done on (57), but no studies have been done on the prog- predictive value. In canine transmissible venereal nostic significance of E-cadherin expression in tumours, strong stromal hyaluronan intensity and a canine tumours. high proportion of hyaluronan-positive tumour cells were significantly associated with progressing In summary, it is apparent that ECM molecules, tumours (61). MMPs and adhesion molecules play an important role in multistage carcinogenesis. The altered expression of these molecules Matrix metalloproteinases may be induced by genetic or epigenetic events and this altered state Degradation of the ECM is crucial for malignant may favour altered cell differentiation, prolifera- and invasion tumour growth, angiogenesis, tion, locomotion, invasion and metastasis. Whilst metastasis. Matrix metalloproteinases (MMPs) are there is no single ECM molecule, MMP or adhe- a family of zinc-dependent neutral endopeptidases sion molecule identified for routine clinical use in which catalyse degradation of diverse substrates in the prognostic evaluation of both human and in both the ECM. The MMPs are involved animal tumours, results from studies on human growth and promoting stimulating tumour cell tumours are promising. Therefore, there is need to invasion and metastasis (33). The levels of MMPs do more studies on the diagnostic, prognostic and can be determined in patient serum and urine or in predictive value of tumour microenvironment in tumour tissues by immunohistochemistry, Western canine tumours. blotting, Northern blotting or polymerase chain in tumour reaction. their role Consistent with ANGIOGENESIS of certain MMPs in progression, levels high cancer tissues correlate with poor prognosis in Angiogenesis, the development of new blood vessels (104), from breast the several human cancers including existing vasculature, is an essential component of solid colonic (69) and prostatic (101) carcinomas. tumour growth and metastasis. Without angiogenesis, tumour expan- Evidence for the role of MMPs in canine tumours sion cannot proceed beyond 1-2 mm because is sparse. The distribution of MMPs has been tumour proliferation is severely limited by nutrient studied by immunohistochemistry in various canine supply to the tumour and waste removal from the sarcomas and carcinomas (53), mammary gland tumour into the surrounding medium (20). There- tumours (71), osteosarcomas (47) and cutaneous fore, angiogenesis is considered an attractive In these studies, higher mast cell tumours (50). prognostic indicator. levels of MMPs were found in neoplastic tissues In human cancers, the degree of than in normal tissues. No studies have been done angiogenesis of a tumour, as assessed by intratumoural on the prognostic significance of MMPs in canine microvessel density (iMVD), is emerging tumours. as a powerful prognostic and a predictive marker. The iMVD Adhesion molecules has since been reported as a possible prognostic indicator in numerous human solid tumours Reduced tumour cell adhesion is associated with including breast (107), prostate (110) and invasive and prognosis. growth unfavourable colorectal (105) carcinomas. homophilic E-cadherin is a calcium-regulated adhesion molecule. The extracellular domain of In canine mammary tumours (26,80,83), squamous 60 cell carcinomas (55) and seminomas (81), angioge- useful in grading mast cell tumours (37). nesis is greater in less differentiated tumours, In the canine tumours, the prognostic value of malignant tumours and in tumours that have immunohistochemical expression of p53 protein metastasized than in benign tumours. In canine has been studied in a limited number of tumours. transmissible venereal tumour, progressing In mammary tumours (109), p53 protein overex- tumours had higher microvessel counts at the pression has been interpreted as a possible indica- invasive edges of the tumours than regressing tion of shortened survival time. Studies on canine tumours (62). These results suggest that increased epithelial colorectal (112) and cutaneous mast cell angiogenesis may be associated with increased (24,37) tumours, suggested that p53 protein tumour cell proliferation, invasion and metastasis expression is not a useful prognostic marker. in canine tumours. However, little research has been done on the prognostic and predictive value CONCLUSION of angiogenesis in canine tumours. In canine mast Ideally, a good diagnostic, prognostic cell tumours (77) angiogenesis was an independent or predictive marker should have a high specificity, prognostic factor. sensitivity, and reproducibility, and its assessment should be practical, easy and cost effective. As the progress p53 PROTEIN in molecular techniques in oncology advances, so Uncontrolled proliferation is a hallmark of cancer does the understanding of tumour biology cells. Therefore, genes and proteins regulating the (initiation, promotion, progression and regression), transition of the GI to S phase of the cell cycle and more new prognostic markers with high have been implicated in the development of human sensitivity and specificity will be found and used and animal cancers. One of the most studied cell in clinical assays. The data on the several markers cycle regulator in human and animal tumours is discussed in this review demonstrate, that many the p53 tumour suppressor gene and its protein molecular markers have a potential to be used for product. The protein product is vital for arresting prognostic evaluation and prediction of response to the cell cycle in cells with damaged DNA and therapy of some canine tumours. However, this is inducing transcription of DNA repair enzymes only true if the mean value of the level of during this arrest (2). expression of these markers is determined in series of tumours. Overlapping results have been Mutations of p53 gene occur in many human reported. Therefore, the data is not compelling tumours (78) and in some canine tumours (66,96). enough at this time to warrant routine use of these Mutant p53 protein is more stable and has an markers. Most of these prognostic markers are extended half-life time than the wild type p53 expensive and not standardized, and, therefore, are protein, and this allows detection of the mutant still limited to research. Before we think of clinical protein in tissues from tumours by immunohisto- of application these markers, additional chemistry. Therefore, immunohistochemical over- well-designed studies over large and diverse expression of p53 protein has been demonstrated samples are needed that will standardize assay in about half of the human tumours (2). method and interpretation criteria. In addition, In canine tumours, immunohistochemical overex- multivariate analysis of prognostic and predictive pression of p53 protein has been reported in some factors should be performed to test the combined tumours including osteogenic tumours (54,91), influence of several markers on prognosis and epithelial colorectal tumours (112), mast cell response to therapy. For now, no ideal molecular tumours (24,37), seminomas and Sertoli-cell prognostic marker has been discovered for any tumours, but not in Leydig-cell tumours (35), therefore, canine tumour, the combination of mammary tumours (49,109) and malignant features and histopathological clinicopathological lymphomas (102). Overexpression of p53 was grading and mitotic index from haematoxylin and found to be of diagnostic significance in eosin stained sections, seems to be more practical distinguishing osteosarcomas from chondrosarco- for the diagnostic and prognostic evaluation of mas or multilobular tumours of the bone (54), and canine tumours. 61 16. Erdelyi I, Nieskens DH, van Dijk JE Vass L and REFERENCES Nederbragt H. Immunohistochemical evaluation of versi- can, in relation to chondroitin sulphate, in canine 1. Abadie JJ, Amardeilh MA and Delverdier ME. Immuno- nuclear mammary tumours. Histology and Histopathology 2003; cell histochemical detection of proliferating 18: 1067-1080. . antigen and Ki-67 in mast cell tumors from dogs. Journal 17 Evanko SP, Angello JC and Wight TN. Formation of of the American Veterinary Medical Association 1999; hyaluronan- and versican-rich pericellular matrix is 215: 1629-1634. required for proliferation and migration of vascular 2. Agarwal ML, Taylor WR, Chernov MV, Chernova OB smooth muscle cells. Arteriosclerosis, Thrombosis, and and Stark GR. The p53 network. Journal of Biological Vascular Biology 1999; 19: 1004-1013. Chemistry 1998; 273: 1-4. 18. Faustino AM, van Garderen E, Schalken JA and 3. Auvinen P. Tammi R, Parkkinen J, Tammi M, Agren U, Nederbragt H. Tenascin expression in normal, hyperplas- Johansson R, Hirvikoski P, Eskelinen M and Kosma VM. tic, dysplastic and neoplastic canine mammary tissues. Hyaluronan in peritumoral stroma and malignant cells Journal of Comparative Pathology 2002; 126: 1-8. spreading and predicts associates with breast cancer 19. Fisher DE. Apoptosis 156; in cancer therapy: crossing the of Pathology 2000: survival. American Journal threshold. Cell 1994; 78: 539-542. 529-535. 20. Folkman J. What is the evidence that tumors are angioge- 4. Bostock DE, Crocker J, Harris K and Smith P. Nucleolar nesis dependent? Journal of the National Cancer Institute organiser regions as indicators of post-surgical prognosis 1990; 82: 4-6. in canine spontaneous mast cell tumours. British Journal 21. Fournel-Fleury C, Magnol JP, Chabanne L, Ghernati I, of Cancer 1989; 59: 915-918. Marchal T, Bonnefond C, Bryon PA and Felman P. Bostock DE and Dye MT. Prognosis after surgical exci- Growth fractions in canine non-Hodgkin's lymphomas as sion of canine fibrous connective tissue sarcomas. Veteri- determined in situ by the expression of the Ki-67 antigen. nary Pathology 1980; 17: 581-588. Journal of Comparative Pathology 1997; 117: 61-72. 6. Bostock DE. Prognosis after surgical excision of canine 22. Funakoshi Y, Nakayama H, Uetsuka K, Nishimura R, melanomas. Veterinary Pathology 1979; 16: 32-40. Sasaki N and Doi K. Cellular proliferative and telomerase 7. Bratulic M, Grabarevic Z, Artukovic B and Capak D. activity in canine mammary gland tumors. Veterinary Number of nucleoli and nucleolar organizer regions per Pathology 2000; 37: 177-183. nucleus and nucleolus--prognostic value in canine mam- 23. Gerdes J, Lemke H, Baisch H, Wacker HH, Schwab U mary tumors. Veterinary Pathology 1996; 33: 527-532. and Stein H. Cell cycle analysis of a cell proliferation- 8. Bravo R, Frank R, Blundell PA and Macdonald-Bravo H. DNA associated human nuclear antigen defined by the monoclo- protein of auxiliary Cyclin/PCNA is the nal antibody Ki-67. Journal of Immunology 1984; 133: polymerase-delta. Nature 1987; 326: 515-517. 1710-1715. Bravo R and Macdonald-Bravo H. Changes in the nuclear 24. Ginn PE, Fox LE, Brower JC, Gaskin A, Kurzman ID distribution of cyclin (PCNA) but not its synthesis depend and Kubilis PS. Immunohistochemical detection of p53 on DNA replication. EMBO Journal 1985; 4: 655-661. tumor-suppressor protein is a poor indicator of prognosis 10. Brown PJ, Rema A and Gartner F. Immunohistochemical for canine cutaneous mast cell tumors. Veterinary characteristics of canine aortic and carotid body tumours. Pathology 2000; 37: 33-39. Journal of Veterinary Medicine A. Physiology, Pathology, 25. Gonzalez CM, Griffey SM, Naydan DK, Flores E, Clinical Medicine 2003; 50: 140-144. Cepeda R, Cattaneo G and Madewell BR. Canine trans- 11. Brunetti B, Sarli G, Preziosi R, Leprotti S and Benazzi missible venereal tumour: a morphological and C. E-cadherin expression in canine mammary carcinomas immunohistochemical study of 11 tumours in growth with regional lymph node metastases. Journal of Veterina- phase and during regression after chemotherapy. Journal Medicine ry Medicine A. Physiology, Pathology, Clinical of Comparative Pathology 2000; 122: 241-248. 2003; 50: 496-500. 26. Graham JC and Myers RK. The prognostic significance of 12. Cattoretti G, Becker MH, Key G, Duchrow M, Schluter angiogenesis in canine mammary tumors. Journal of C, Galle J and Gerdes J. Monoclonal antibodies against Veterinary Internal Medicine 1999; 13: 416-418. recombinant parts of the Ki-67 antigen (MIB 1 and MIB 27. Gratzner HG. Monoclonal antibody in microwave-processed to 5-bromo- and 3) detect proliferating cells Journal of Pathology 5-iododeoxyuridine: A new reagent for detection of DNA sections. formalin-tixed paraffin replication. Science 1982; 218: 474-475. 1992; 168: 357-363. 28. Graves GM, Bjorling DE and Mahaffey 13. Della Saida L, Preziosi R, Mazzoni M and Marcato PS. E. Canine hemangiopericytoma: 23 cases (1967-1984). in canine infundibular Cell proliferation patterns Journal of the American Veterinary Medical Association keratinizing acanthorna and well differentiated squamous 1988; 192: 99-102. cell carcinoma of the skin. European Journal of Histo- 29. Griffey SM, Kraegel SA and Madewell BR. Proliferation chemistry 2002; 46: 165-172. indices in spontaneous canine lung cancer: proliferating 14. Destexhe E, Vanmanshoven P and Coignoul F. Compari- organizer regions by cell nuclear antigen (PCNA), Ki-67 (MIB1) and mitotic nucleolar . son of argyrophilic counts. Journal of Comparative Pathology counting and image analysis in canine mammary tumors. 1999; 120: 321-332. 1995; 56: of Veterinary Research American Journal 30. Guvenc T, Haligur M, Orman MN and Haziroglu 185-187. R. Mitosis and apoptosis in canine cutaneous histiocytoma 15. De Vico G, Agrimi U and Maiolino P. Nucleolar size and and transmissible venereal tumour. mitotic index in basal cell carcinomas (BCC) and squa- Acta Veterinaria Hungarica 2002; 50: 315-321. mous cell carcinomas (SCC) of canine skin. Zentralblatt fur Veterinarmedizin. Reihe A 1994; 41: 76-79. 31. Harmelin A, Zuckerman A and Nyska A. Correlation of 62 Ag-NOR protein measurements with prognosis in canine Bergstrom R, Gerdin B, Pahlman L and Glimelius B. transmissible venereal tumour. Journal of Comparative Stromal tenascin distribution as a prognostic marker in Pathology 1995; 1 12: 429-433. colorectal cancer. British Journal of Cancer 1997; 76: 32. Hinrichs U, Rutteman GR and Nederbragt H. Stromal 526-530. accumulation of chondroitin sulphate in mammary 47. Lana SE, Ogilvie GK and Hansen RA. Identification of tumours of dogs. British Journal of Cancer 1999; 80: matrix metalloproteinases in canine neoplastic tissue. 1359-1365. American Journal of Veterinary Research 2000; 61: 33. Hojilla CV, Mohammed FF and Khokha R. Matrix metal- 111-114. loproteinases and their tissue inhibitors direct cell fate 48. Laprie C, Abadie J, Amardeilh MF, Net JL, Lagadic M during cancer development. British Journal of Cancer and Delverdier M. MIB-1 immunoreactivity correlates 2003; 89: 1817-1821. with biologic behaviour in canine cutaneous melanoma. 34. Hung C, Pong VF, Cheng CR, Wong FI and Chu RM. Veterinary Dermatology 2001; 12: 139-147. An improved system for quantifying AgNOR and PCNA 49. Lee CH, Kim WH, Lim JH, Kang MS, Kim DY and in canine tumors. 20: Anticancer Research 2000; Kweon OK. Mutation and overexpression of p53 as a 3273-3280. prognostic factor in canine mammary tumors. Journal of 35. Inoue M and Wada N. Immunohistochemical detection of Veterinary Science 2004; 5: 63-69. p53 and p21 proteins tumours. in canine testicular 50. Leibman NF, Lana SE, Hansen RA, Powers BE, Fettman Veterinary Record 2000; 146: 370-372. MJ, Withrow SJ and Ogilvie GK. Identification of matrix 36. Ioachim E. Charchanti A, Briasoulis E, Karavasilis V, metalloproteinases in canine cutaneous mast cell tumors. Journal Tsanou H, Arvanitis DL, Agnantis NJ and Pavlidis N. of Veterinary Internal Medicine 2000, 14: Immunohistochemical expression of extracellular matrix 583-586. components tenascin, fibronectin, collagen type IV and 51. Lipponen PK and Aaltomaa S. Apoptosis in bladder laminin in breast cancer: their prognostic value and role . cancer as related to standard prognostic factors and in tumour invasion and progression. European Journal of prognosis. Journal of Pathology 1994; 173: 333-339. Cancer 2002; 38: 2362-2370. 52. Lohr CV, Teitke JP, Failing K, Weiss E. Characterizati- 37. Jaffe MH, Hosgood G, Taylor HW, Kerwin SC, Hedlund on of the proliferation state in canine mammary tumors by CS, Lopez MK, Davidson JR, Miller DM and Paranjpe the standardized AgNOR method with postfixation and immunohistologic M. Immunohistochemical and clinical evaluation of p53 in detection of Ki-67 and PCNA. canine cutaneous mast cell tumors. Veterinary Pathology Veterinary Pathology 1997; 34: 212-221. 2000; 37: 40-46. 53. Loukopoulos P. Mungall BA, Straw RC, Thornton JR and Robinson WF. Matrix 38. Johnson GC, Miller MA and Ramos-Vara JA. Compari- metalloproteinase-2 and -9 involvement in canine son of argyrophilic nucleolar organizer regions (AgNORs) tumors. Veterinary Pathology and mitotic index in distinguishing benign from malignant 2003; 40: 382-394. canine smooth muscle tumors and in separating inflamma- 54. Loukopoulos P, Thornton JR and Robinson WF. Clinical and pathologic relevance of p53 index in canine tory hyperplasia from neoplastic lesions of the urinary osseous bladder mucosa. Journal of Veterinary Diagnostic Investi- tumors. Veterinary Pathology 2003; 40: 237-248. gation 1995; 7: 127-136. 55. Maiolino P. Papparella S. Restucci B and De Vico G. 39. Juntes P and Pogacnik M. Morphometric analysis of Angiogenesis in squamous cell carcinomas of canine skin: AgNORs in tubular and papillary parts of canine an immunohistochemical and quantitative analysis. Journal mammary gland tumors. Analytical and Quantitative of Comparative Pathology 2001; 125: 117-121. Cytology and Histology 2000; 22: 185-192. 56. Maiolino P, Restucci B and De Vico G. Expression of 40. Kerr JFR, Winterford CM and Harmon BV. Apoptosis: proliferating cell nuclear antigen in basal cell carcinomas Its significance in cancer and cancer therapy. Cancer and in squamous cell carcinomas of canine skin: correlati- 1994; 73: 2013-20126. with mitotic index and on histological features. 41. Keshgegian AA and Cnaan A. Proliferation markers in Zentralblatt fur Veterinarmedizin. Reihe A 1995; 42: breast carcinoma. Mitotic figure count, S-phase fraction, 339-343. proliferating cell nuclear antigen, Ki-67 57. McEntee MF and Brenneman KA. Dysregulation of beta- and. MIB-1. American Journal of Clinical Pathology 1995; 104: 42-49. catenin is common in canine sporadic colorectal tumors. 42. Kiupel M, Bostock D and Bergmann V. The prognostic Veterinary Pathology 1999; 36: 228-236. significance of AgNOR counts and PCNA-positive cell 58. Millanta F, Fratini F, Corazza M, Castagnaro M, counts in canine malignant lymphomas. Journal of Zappulli V and Poli A. Proliferation activity in oral and Comparative Pathology 1998; 119: 407-418. cutaneous canine melanocytic tumours: correlation with 43. Kiupel M, Teske E and Bostock D. Prognostic factors for histological parameters, location, and clinical behaviour. treated canine malignant lymphoma. Veterinary Pathology Research in Veterinary Science 2002; 73: 45-51. 1999; 36: 292-300. 59. Minkus G, Jutting U, Aubele M, Rodenacker K, Gais P, 44. Knudson CB and Knudson W. Hyaluronan-binding pro- Breuer W and Hermanns W. Canine neuroendocrine tu- teins in development, tissue homeostasis, and disease. mors of the pancreas: a study using image analysis techni- FASEB Journal 1993; 7: 1233-1241. ques for the discrimination of metastatic versus nonmeta- 45. Koukoulis GK, Gould VE, Bhattacharyya Gould JE. static tumors. Veterinary Pathology 1997; 34: 138-145. Howeedy AA and Virtanen I. Tenascin in normal, reacti- 60. Mohammed SI, Bennett PF, Craig BA, Glickman W, ve, hyperplastic, and neoplastic tissues: biologic and Mutsaers AJ, Snyder PW, Widmer WR, DeGortari AE, pathologic implications. Human Pathology 1991; 22: Bonney PL and Knapp DW. Effects of the cyclooxygena- 636-643. se inhibitor, piroxicam, on tumor response, apoptosis, and 46. Kressner U, Lindmark G, Tomasini-Johansson B, angiogenesis in a canine model of human invasive urinary 63 Diagnostic Investigation 2000; 12: 111-117. bladder cancer. Cancer Research 2002; 62: 356-358. 75. Platz SJ, Breuer W, Pfleghaar 61. Mukaratirwa S, Chimonyo M, Obwolo M, Gruys E and S. Minkus G and Hermanns W. cells and extracellular matrix Prognostic value of histopathological Nederbragt H. Stromal grading in canine extramedullary components in spontaneous canine transmissible venereal plasmacytomas. Histology and Veterinary Pathology 1999; 36: 23-27. tumour at different stages of growth. 76. Preziosi R, Sarli G, Benazzi C and Marcato PS. Detecti- Histopathology 2004; 19: 1117-1123. on of proliferating cell nuclear antigen (PCNA) in canine 62. Mukaratirwa S. Chiwome T, Chitanga T, and Bhebhe E. and feline mammary tumours. Journal of Comparative Canine transmissible venereal tumour: assessment of mast Pathology 1995; 113: 301-313. cell numbers as indicators of growth stage. Veterinary 77. Preziosi R, Sarli G and Paltrinieri M. Prognostic value of Research Communications 2005; in press. van Ederen AM and intratumoral vessel density in cutaneous mast cell 63. Mukaratirwa E, S, de Witte tumors of the dog. Journal of Comparative Pathology 2004; 130: Nederbragt H. Tenascin expression in relation to stromal 143-151. tumour cells in canine gastrointestinal epithelial tumours. 78. Prokocimer M and Rotter V. Structure and function of Journal of Comparative Pathology 2003; 129: 137-146. p53 in normal cells and their aberrations in 64. Mukaratirwa S, van Ederen AM, Gruys E and Nederbragt cancer cells: projection on the hematologic cell H. Versican and hyaluronan expression in canine colonic lineages. Blood 1994; 84: 2391-2411. adenomas and carcinomas: relation to malignancy and 79 Reis AL, Carvalheira J, Schmitt of Comparative depth invasion. Journal FC, and Gartner F. of tumour Immunohistochemical study of the Pathology 2004; 131: 259-270. expression of E-cadherin in canine mammary 65. Naik P, Karrim J and Hanahan D. The rise and fall of tumours. Veterinary down- Record 2003; 152: 621-624. tumorigenesis: apoptosis multistage during from progression 80. Restucci B, De Vico G and Maiolino P. Evaluation of to tumor modulation contributes angiogenesis in canine mammary tumors by quantitative angiogenic progenitors. Genes and Development 1996; platelet endothelial cell adhesion molecule immunohisto- 10: 2105-2116. chemistry. Veterinary Pathology 2000; 37: 297-301. 66. Nasir L, Rutteman GR, Reid SW, Schulze C and Argyle 81. Restucci B, Maiolino P. Paciello 0, Martano M, De Vico DJ. Analysis of p53 mutational events and MDM2 ampli- G and Papparella S. Evaluation of angiogenesis in canine fication in canine soft-tissue sarcomas. Cancer Letters seminomas by quantitative immunohistochemistry. Journal 2001; 174: 83-89. of Comparative Pathology 2003; 128; 252-259. 67. Nieto A, Pena L, Perez-Alenza MD, Sanchez MA, Flores of detection 82. Restucci B, Papparella S. De Vico G and Maiolino P. E JM and Immunohistologic Castano M. cadherin expression canine mammary tumors: in in normal and neoplastic estrogen receptor alpha canine and prognostic mammary gland. Journal of Comparative Pathology 1997; associations clinical and pathologic 116: 191-202. significance. Veterinary Pathology 2000; 37: 239-247. 83. Restucci B, Papparella S, Maiolino P and De Vico G. Bennett MW, O'Sullivan GC, Roche D, 68. O'Connell J, Expression of vascular endothelial growth factor in canine Kelly J, Collins JK and Shanahan F. Fas ligand expressi- mammary that the tumors. Veterinary Pathology on in primary colon adenocarcinomas: evidence 2002; 39, 488-493. immune a prevalent mechanism of Fas counterattack is 84. Rhodes DR, Sanda MG, Otte AP, Chinnaiyan AM evasion in human colon cancer. Journal of Pathology and Rubin MA. Multiplex biomarker approach for determi- 1998; 186: 240-246. ning risk of prostate-specific antigen-defined 69. Ogura E, Senzaki H and Kiyozuka Y. Expression and recurrence of tissue and prostate cancer. Journal of the National Cancer Institute metalloproteinases localization of matrix 2003; 95: 661-668 inhibitors of metalloproteinases as a prognostic factor in advanced colorectal carcinomas. Oncology Reports 1997; 85. Ricciardelli C, Brooks JH, Suwiwat S, Sakko AJ, Mayne K, Raymond WA, Seshadri R, LeBaron RG and Horsfall 4: 701-705. DJ. Regulation of stromal versican expression by breast 70. Ohta G, Sakai H, Kachi S, Hirata A, Yonemaru K, Kitajima A, Yanai T and Masegi T. Assessment of cancer cells and importance to relapse-free survival in and patients osteosarcomas with node-negative of canine primary breast proliferative potentials cancer. Clinical Cancer Research 2002; 8: 1054-1060. chondrosarcomas by MIB-1 immunohistochemistry and 86. Ricciardelli C, Mayne K, Sykes bromodeoxyuridine incorporation. Journal of Comparative PJ, Raymond WA, McCaul K, Marshall VR and Horsfall D. Elevated Pathology 2004; 131: 18-27. levels of versican but not decorin predict disease progression 71. Papparella S, Restucci B, and Maiolino P. Immunohisto- in early-stage prostate cancer. chemical distribution of type IV collagenase in normal, Clinical Cancer Research 1998; 4: 963-971. dysplastic and neoplastic canine mammary gland. Journal 87. Roels S, Tilmant K and Ducatelle R. P53 of Comparative Pathology 1997; 117: 277-282. expression and proliferation. 72, Pardee AB. GI events and regulation of cell apoptosis in melanomas of dogs and cats. Research in Veterinary Science 2001; 70: 19-25. Science 1989; 246: 603-608. AI, Perez-Alenza D, Cuesta P and 88. Roels S. Tilmant K and Ducatelle R. PCNA 73. Pena LL, Nieto and Ki67 Castano M. Immunohistochemical detection of Ki-67 and proliferation markers as criteria for prediction of clinical PCNA in canine mammary tumors: relationship to clinical behaviour of melanocytic tumours in cats and dogs. and pathologic variables. Journal of Veterinary Diagnostic Journal of Comparative Pathology 1999; 121: 13-24. 89. Ropponen K, Tammi M, Parkkinen Investigation 1998; 10: 237-246. J, Eskelinen M, 74. Phillips BS, Kass PH, Naydan DK, Winthrop MD, Tammi R, Lipponen P, Agren U, Alhava E and Kosma VM. Tumor cell-associated hyaluronan Griffey SM and Madewell, BR. Apoptotic and proliferati- as an unfavorable of Veterinary on indexes in canine lymphoma. Journal prognostic factor in colorectal cancer. Cancer Research 64 1998; 58: 342-347. 55 cases, with observations on p53 immunoexpression. Rudas M, Gnant MF, Mittlbock M, Neumayer R, Journal of Comparative Pathology 2004; 131: 207-213. Kummer A, Jakesz R, Reiner G and Reiner A. Thymidi- 103. Takeichi M. Cadherins in cancer: implications for inva- ne labeling index and Ki-67 growth fraction in breast sion and metastasis. Current Opinion in Cell Biology cancer: comparison and correlation with prognosis. 1993; 5: 806-811. Breast Cancer Research and Treatment 1994; 32: 104. Talvensaari-Mattila A, Paakko P and Hoyhtya M. 165-175. Matrix metalloproteinase-2 immunoreactive protein. A Sagartz JE, Bodley WL, Gamblin RM, Couto CG, marker of aggressiveness in breast carcinoma. Cancer Tierney LA and Capen CC. p53 tumor suppressor 1998; 83: 1153-1162. protein overexpression in osteogenic tumors of dogs. 105. Tanigawa N, Amaya H, Matsumura M, Lu C, Kitaoka Veterinary Pathology 1996; 33: 213-2121. A, Matsuyama K and Muraoka R. Tumor angiogenesis Sakai H, Noda A, Shirai N, Iidaka T, Yanai T, Masegi and mode of metastasis in patients with colorectal T, Shirai N, Iidaka T, Yanai T and Masegi T. Prolifera- cancer. Cancer Research 1997; 57: 1043-1046. tive activity of canine mast cell tumours evaluated by 106. Theon AP, Lecouteur RA, Carr EA and Griffey SM. bromodeoxyuridine incorporation and Ki-67 expression. Influence of tumor cell proliferation and sex-hormone Journal of Comparative Pathology 2002; 127: 233-238. receptors on effectiveness of radiation therapy for dogs Sarli G, Benazzi C, Preziosi R and Marcato PS. Prolife- with incompletely resected meningiomas. Journal of rative activity assessed by anti-PCNA and Ki-67 mono- American Veterinary Medical Association 2000; 216: clonal antibodies in canine testicular tumours. Journal of 701-707, 684-685. Comparative Pathology 1994; 110: 357-368. 107. Toi M, Kashitami J and Tominaga T. Tumour angioge- Sarli G, Preziosi R, Benazzi C, Castellani G, Marcato nesis is an independent prognostic indicator of breast PS. Prognostic value of histologic stage and proliferative carcinoma. International Journal of Cancer 1993; 55: activity in canine malignant mammary tumors. Journal 37 1-374 . Veterinary Diagnostic Investigation 2002; 14: 25-34. 108. Vail DM, Kisseberth WC, Obradovich JE, Moore FM, Schweyer S, Soruri A, Meschter 0, Heintze A, London CA, MacEwen EG and Ritter MA. Assessment Zschunke F, Miosge N, The len P. Schlott T, Radzun HJ of potential doubling time (Tpot), argyrophilic nucleolar and Fayyazi A. Cisplatin-induced apoptosis in human organizer regions (AgNOR), and proliferating cell nucle- malignant testicular germ cell lines depends on MEK/ ar antigen (PCNA) as predictors of therapy response in ERK activation. British Journal of Cancer 2004; 91: canine non-Hodgkin's lymphoma. Experimental Hemato- 589-598. logy 1996; 24: 807-815. Setoguchi A, Sakai T, Okuda M, Minehata K, Yazawa 109. Wakui S, Muto T, Yokoo K, Yokoo R, Takahashi H, M, Ishizaka T, Watari T, Nishimura R, Sasaki N, Masaoka T, Hano H and Furusato M. Prognostic status Hasegawa A and Tsujimoto H. Aberrations of the p53 of p53 gene mutation in canine mammary carcinoma. tumor suppressor gene in various tumors in dogs. Anticancer Research 2001; 21: 611-616. American Journal of Veterinary Research 2001; 62: 110. Weidner N, Carroll PR, Flax J, Blumenfeld W and 433-439. Folkman J. Tumor angiogenesis correlates with metasta- Siitonen SM, Kononen JT, He lin HJ, Rantala IS, Holli sis in invasive prostate carcinoma. American Journal of KA and Isola JJ. is Reduced E-cadherin expression Pathology 1993; 143: 401-409. associated with invasiveness and unfavorable prognosis 111. Wiksten JP, Lundin J, Nordling S, Lundin M, Kokkola in breast cancer. American Journal of Clinical Pathology A, von Boguslawski K and Haglund C. Tenascin-C 1996; 105: 394-402. expression correlates with prognosis in gastric cancer. 98. Simoes JP, Schoning P and Butine M. Prognosis of Oncology 2003; 64: 245-250. canine mast cell tumors: a comparison of three methods. 112. Wolf JC, Ginn PE, Homer B, Fox LE and Kurzman ID. Veterinary Pathology 1994; 31: 637-647. Immunohistochemical detection of p53 tumor suppressor 99. Sirri V, Roussel P and Hernandez-Verdun D. The in gene protein canine epithelial colorectal tumors. AgNOR proteins: qualitative and quantitative changes Veterinary Pathology 1997; 34: 394-404. during the cell cycle. Micron 2000; 31: 121-126. 113. Wyllie AH, Kerr JFR and Currie AR. Cell death: the 100. Spangler WL, Culbertson MR and Kass PH. Primary of significance apoptosis. International Review of mesenchymal (nonangiomatous/nonlymphomatous) neo- Cytology 1980; 68: 251-304. plasms occurring in the canine spleen: anatomic classifi- 114. Yoshida K, Yanai T, Iwasaki T, Sakai H, Ohta J, Kati cation, immunohistochemistry, and mitotic activity cor- S, Ishikawa K, Lackner AA and Masegi T. Proliferative related with patient survival. Veterinary Pathology 1994; potential of canine oral epulides and malignant neo- 31: 37-47. plasms assessed by bromodeoxyuridine labeling. Veteri- 101. Stearns M and Stearns ME. Evidence for increased acti- nary Pathology 1999; 36: 35-41. vated metalloproteinase 2 (MMP-2a) expression associa- 115. Zuccari DA, Santana AE, Cury PM and Cordeiro JA. ted with human prostate cancer progression. Oncology Immunocytochemical study of Ki-67 as a prognostic Research 1996; 8: 69-75. t marker in canine mammary neoplasia. Veterinary Clini- 102. Sueiro FA, Alessi AC and Vassal lo J. Canine lympho- cal Pathology 2004; 33: 23-28. mas: a morphological and immunohistochemical study of

Journal

Veterinary QuarterlyTaylor & Francis

Published: Jun 1, 2005

Keywords: Adhesion molecules; Angiogenesis; Apoptosis; Canine neoplasms; Cell cycle; Dog diseases; Molecular markers; p53 protein; Prognosis; Prognostic and predictive markers; Proliferation; Proliferation markers; Tumour behaviour; Tumour grading; Tumour markers; Tumour staging; Tumours; Veterinary oncology

There are no references for this article.