Estimating survival prognosis after treatment of choroidal melanoma
Prof. Bertil Damato
Prof. Sarah Coupland
Collaborators:
Azzam Taktak
Antonio Eleuteri
It is useful to know which patients will die of their uveal melanoma and which ones will survive. Such prognostication makes it possible to reassure those with a normal life-expectancy while targetting special care to those who are less fortunate. Prognostication also makes it possible to understand the significance of laboratory findings on tissue specimens without waiting many years to see if the patient dies of the melanoma or not.
Conventionally, survival is predicted according to the size and extent of the intraocular melanoma (i.e, the 'clinical stage') and the microscopic appearances (i.e., the 'histological grade of malignancy'). We have contributed significantly to the TNM staging system ('Tumour, Node, Metastasis') of the American Joint Committee on Cancer (AJCC). However, survival prediction according to clinical stage or histological grade is not precise enough to be relevant to individual patients. Since 1999, we have been typing uveal melanomas according to their genetic abnormalities and we now have data on more than a thousand patients. This information has enabled us to develop mathematical methods for estimating the survival prognosis according to clinical stage, histological grade and genetic type, also taking the patient's age and sex (and hence the normal life-expectancy) into account. These methods have resulted in an online tool that, for the first time, provides survival estimates that are accurate enough to be relevant to individual patients. We are now refining this tool as we gather more data.
Estimating patient numbers needed to perform clinical trials evaluating systemic adjuvant therapy of uveal melanoma - John Whitehouse, Sarah Coupland & Bertil Damato
Now that we can identify the patients who have invisible, systemic micrometastases it would be ideal if we could give such patients treatment aimed at delaying or preventing the onset of metastatic disease. To do this, it is first necessary to find out whether such 'systemic adjuvant therapy' is actually effective. This is done by randomising patients between treatment and non-treatment and comparing survival between the two groups of patients. If this 'randomised clinical trial' shows that treated patients live longer than untreated patients then the treatment is effective. With a conventional randomised trial design, it would be necessary to enrol about a thousand patients having a high risk of developing metastatic disease. Since less than 50% of all patients have such a high-risk tumour and because many patients will be unable to participate, for one reason or another, it would taken many years to collect enough data. This research project is aimed at investigation different ways in which the required number of patients can be reduced. Initial statistical investigations suggest that if the results are assessed at various stages of the trial, it will be possible to stop the study early, after investigating fewer patients for a shorter time, because the results either indicate that the treatment is useless or because the data prove that the treatment is highly effective.
Prof. Sarah Coupland
Collaborators:
Azzam Taktak
Antonio Eleuteri
It is useful to know which patients will die of their uveal melanoma and which ones will survive. Such prognostication makes it possible to reassure those with a normal life-expectancy while targetting special care to those who are less fortunate. Prognostication also makes it possible to understand the significance of laboratory findings on tissue specimens without waiting many years to see if the patient dies of the melanoma or not.
Conventionally, survival is predicted according to the size and extent of the intraocular melanoma (i.e, the 'clinical stage') and the microscopic appearances (i.e., the 'histological grade of malignancy'). We have contributed significantly to the TNM staging system ('Tumour, Node, Metastasis') of the American Joint Committee on Cancer (AJCC). However, survival prediction according to clinical stage or histological grade is not precise enough to be relevant to individual patients. Since 1999, we have been typing uveal melanomas according to their genetic abnormalities and we now have data on more than a thousand patients. This information has enabled us to develop mathematical methods for estimating the survival prognosis according to clinical stage, histological grade and genetic type, also taking the patient's age and sex (and hence the normal life-expectancy) into account. These methods have resulted in an online tool that, for the first time, provides survival estimates that are accurate enough to be relevant to individual patients. We are now refining this tool as we gather more data.
Estimating patient numbers needed to perform clinical trials evaluating systemic adjuvant therapy of uveal melanoma - John Whitehouse, Sarah Coupland & Bertil Damato
Now that we can identify the patients who have invisible, systemic micrometastases it would be ideal if we could give such patients treatment aimed at delaying or preventing the onset of metastatic disease. To do this, it is first necessary to find out whether such 'systemic adjuvant therapy' is actually effective. This is done by randomising patients between treatment and non-treatment and comparing survival between the two groups of patients. If this 'randomised clinical trial' shows that treated patients live longer than untreated patients then the treatment is effective. With a conventional randomised trial design, it would be necessary to enrol about a thousand patients having a high risk of developing metastatic disease. Since less than 50% of all patients have such a high-risk tumour and because many patients will be unable to participate, for one reason or another, it would taken many years to collect enough data. This research project is aimed at investigation different ways in which the required number of patients can be reduced. Initial statistical investigations suggest that if the results are assessed at various stages of the trial, it will be possible to stop the study early, after investigating fewer patients for a shorter time, because the results either indicate that the treatment is useless or because the data prove that the treatment is highly effective.
