 |
Animal Health Trust |
 |
Oncology Research Group
The value of tumour tissue for cancer research - what we could do if we
were able to collect more samples of dog tumours
The Animal Health Trust (AHT) Oncology Research Group is engaged in research into different aspects of common cancers in dogs. We do not use experimental animals in our research, but instead we study samples (biopsies) of naturally occurring tumours that are collected from patients by veterinary surgeons as part of the normal therapeutic and/or diagnostic clinical procedures. Tumour samples are only used for research purposes where an owner has signed a consent form to say that excess material (to that collected for diagnosis) can be used for research studies.
The objectives of our research are varied, but predominant are developing tests which improve the accuracy of predicting how tumours will behave and respond to treatment, and identifying the underlying causes of tumour development and progression.
The ability to predict the prognosis for a dog with cancer is crucial for clinicians to be able to select the most appropriate therapy for their patients. It enables a clinician to decide whether an animal may benefit from an aggressive treatment, thus ensuring that an animal that requires treatment receives it and that an animal that will not benefit is spared the rigours and possible hazards associated with treatment.
The conventional approach for predicting how a tumour will behave, and thus deciding on an appropriate form of treatment, relies on the tumour being classified on the basis of ‘grade’ and ‘clinical stage’. Tumour grade is determined by a pathologist via microscopic examination of a tumour biopsy to identify cancer cells and assess the degree to which the cancer cells have lost the specialist structure that is characteristic of the tissue in which the tumour is present. Staging is performed by a clinician and establishes the size and location of a tumour, and whether the tumour has spread from its original position to nearby locations and/or more distant parts of the body.
The limitation of these ‘grading’ and ‘staging’ schemes is that typically at best they assign a dog’s tumour to a category associated with a certain probability of having a particular outcome. However, within such a category tumours may still exhibit significant variation in behaviour.
We are attempting to develop assays that accurately predict the outcome for individual tumours. The way in which a tumour behaves is dependent upon the entities (known as ‘molecules’) within tumour cells that control events within a cell, thereby determining the functions of a cell and ultimately the fate of a cell. These ‘control molecules’ are of three sorts:
The chromosomes, which are the genetic material passed on from one generation to the next. The chromosomes (39 pairs in the dog) are composed of DNA (DeoxyriboNucleic Acid), and in the dog are organised into about 30,000 different genes. The genes (the ‘genetic code’) contain the instructions that specifies what cells look like and the functions that they perform.
Messenger RNAs are effectively copies of genes and carry the information contained within the genes to all parts of the cell. Each gene is copied to make a different messenger RNA species.
Proteins, which include enzymes and hormones, are the molecules in cells that actually perform functions. Each protein is made according to the instructions specified by one messenger RNA species, and therefore according to the information contained within one gene.
The basis for the design of a new generation of assays that predict the outcome for individual tumours is the observation that early in the development of a tumour, and/or prior to treatment, there are differences between the control molecules present in the cells of tumours that go on to behave differently. Consequently, by monitoring the control molecules present within tumour cells at the point of diagnosis, in principle it should be possible to predict how a tumour wilt proceed to behave. The monitoring of control molecules is known as ‘molecular profiling’, and may take the form of identifying structural changes to chromosomes that are often present in tumour cells, and/or measuring the number of copies of particular messenger RNAs, or proteins, in tumour cells.
Molecular profiling of a brain tumour from a bulldog
The picture shows a comparison between the chromosomes in the cells of a glial cell tumour and the chromosomes in ‘normal’ cells from a ‘healthy’ dog. The majority of the length of each chromosome appears yellow/orange in colour indicating that there is no difference between the number of copies of each part of the chromosome in the tumour cells and the normal cells. However, small sections of certain chromosomes appear green, or red, in colour. Green sections represent chromosome segments that are represented by more copies (‘amplifications’) in the tumour cells than in the normal cells, whilst red sections represent chromosome segments that are represented by fewer copies (‘deletions’) in the tumour cells than in the normal cells. One objective of our research is to establish whether the chromosome amplifications and deletions are predictive of how the tumour will respond to radiotherapy.
In order to develop assays that predict the outcome for individual tumours it is necessary to collect tumour biopsies from large numbers of affected dogs. It is only by relating the ‘molecular profiles’ of large numbers of tumours to their behaviour is it possible to identify specific chromosome changes, messenger RNAs, or proteins (‘molecular markers’) that reliably predict, for example, whether a mast cell tumour will metastasise, or whether a lymphoma will respond to a particular chemotherapeutic protocol.
However, the reality is that even at the AHT small animal veterinary hospital (Centre for Small Animal Studies), which sees approximately 500 new, referred cases of cancer in cats and dogs every year, the number of cases of each type of tumour is relatively small. The situation in practice is even more pronounced - an individual vet may see less than a handful of cases of even the most common canine cancers in a year. The difficulty in collecting tumour tissue for molecular profiling is further compounded by the need to collect tissue in a special ‘molecular preservative’ (known as ‘RNAIater’), which protects the chromosomes, messenger RNAs and proteins.
Tumour biopsies excised for diagnosis by a pathologist are typically collected in formalin, a ‘tissue fixative’ that maintains the morphology of the cells in the tissue. This ensures that the biopsy is a faithful representation of the tumour from which it is derived, and that the cells are recognisable by the pathologist. Unfortunately, formalin degrades chromosomal DNA, messenger RNAs and many proteins, to all intents and purposes making them unusable for molecular profiling. Consequently, in order to collect tumour biopsies for molecular profiling research, it is necessary to divide a tumour biopsy between formalin (for conventional histopathological diagnosis) and molecular preservative. However, whilst formalin will commonly be a fixture in a veterinary surgery, a bottle of molecular preservative will not.
A crucial aspect of our work is therefore to liaise with vets to seek their participation in projects that require the collection of samples of biopsies from particular dog tumours. Moves are also afoot within the veterinary cancer research community to establish a nationwide network of groups, with shared interests and objectives, who are willing to share precious tumour tissue and work together. However, there is little doubt that these initiatives would be greatly assisted if the owners of dogs with tumours were to encourage their vets to save surplus (to that collected for diagnosis) samples of tumours in a molecular preservative (that could be provided by a research group such as the AHT Oncology Research Group).
The identification of molecular markers that predict how tumours will behave has value beyond the development of prognostic assays. Knowledge of the consequence of specific chromosome abnormalities and the function of particular messenger RNAs, or proteins, will increase our understanding of the process of cancer progression, and this can only help researchers to devise improved treatment strategies. The value of tumour samples for research studies that are of direct benefit to dogs with cancer cannot be overemphasised.
Web: www.aht.org.uk
Lymphoma is the most frequent life-threatening cancer in dogs, accounting for up to 20% of all tumours and affecting as many as 24 out of every 100,000 dogs. In the most common form of the disease, cells (lymphocytes) derived from the bone marrow become cancerous and travel to one or more lymph glands (and possibly to the liver, thymus, spleen and tonsils) where they form solid tumours called lymphomas (or lymphosarcomas). Lymphoma may occur in dogs of any age, but is most common in dogs between 6-9 years old. If untreated, death can result within 12 weeks of diagnosis. However, lymphoma is very sensitive to chemotherapy and up to 80% of dogs treated will go into remission. On average, dogs receiving combination chemotherapy can survive for 12 months.
Development of a prognostic assay for canine lymphoma
We are endeavouring to develop a test that can be routinely used in the clinic to predict whether a dog with lymphoma is likely to enter remission following a specific course of chemotherapy, or whether the lymphoma is not likely to respond to this treatment. If we are successful, such a test will assist the veterinary clinician to choose an appropriate form of treatment and will enable the clinician to inform a dog owner of the likely outcome.
In studies of diffuse large B cell lymphoma in man, researchers have discovered that differential survival following chemotherapy concords with different patterns of gene expression prior to treatment. Furthermore, researchers demonstrated that it was possible to predict the post- chemotherapy outcome for an individual lymphoma on the basis of the levels at which a small number of genes were expressed at the time of diagnosis. We hope to demonstrate that sensitivity to chemotherapy has a similar molecular basis in dogs.
In order to undertake our study, it is necessary for us to collect lymphoma tissue from as many patients as possible, and subsequently follow how the patients respond to particular chemotherapy protocols.
Animal Health Trust, Lanwades Park, Kentford, Newmarket, Suffolk, CB8 7UU, UK Tel: +44 (0)8700 502424 Fax: +44 (0)8700 502425
Email: info@aht.org.uk Web: www.aht.org.uk
mike.starke
| ALL TEXT AND IMAGES © DOBERMANN BREED COUNCIL. |
|---|