Assessing the value of genomic medicine in oncology

Over the past decade there have been remarkable advancements in the understanding of the molecular underpinnings of malignancy, leading to the development of a number of innovative and highly effective treatments. The success of these treatments can largely be attributed to an increased ability to deliver cancer therapies on a more individualized level based on the genetic makeup of patients’ tumours. Methods of testing such as fluorescent in situ hybridization, polymerase chain reaction and direct sequencing , which can identify the genetic architecture of tumours have improved to the point where there is an increased desire to incorporate the genomic information derived from such tests into treatment selection for cancer patients. There is, however, uncertainty with regards to the impact that genomic testing prior to treatment will have on not only health outcomes but also the costs it will incur. There is therefore a need to develop economic evidence to support the implementation of genomic testing and targeted treatment strategies in oncology to ensure the resources allocated to these approaches offer value for money. This thesis primarily investigates the most appropriate approaches to assessing the value of genomic medicine in oncology. First, factors that are important to consider when undertaking economic evaluations of genomic test and treat strategies are identified and a discussion is presented on how existing methods can be adapted to account for the unique aspects of genomic technologies. Second, one of the first model-based economic evaluations of genomic testing and targeted treatment is developed to provide initial evidence of cost-effectiveness in a metastatic lung adenocarcinoma population. Third, efforts to improve the translation of the economic evidence of genomic medicine in oncology are undertaken to ensure appropriate and timely access to these innovative and potentially beneficial technologies. Finally, the thesis addresses another important issue in the economic evaluation of oncology treatments; the estimation of health state utility values for use in cost-utility analyses. To limit bias in future evaluations, existing estimation approaches are compared and the most reliable approach is identified. The findings from this thesis suggest that a number of test-related characteristics are important to consider and can have large impacts on the results of economic evaluations of genomic approaches to treatment. More complex testing approaches are likely to make the application of traditional methods for assessing clinical and economic value more difficult, suggesting the need to adapt existing methods in the future. The results of the modelled economic evaluation highlight the difficulty in assessing the value of genomic testing when results dictate the use of high cost targeted therapies associated with only small improvements in clinical outcomes. Potential future advancements in testing, however, offer the opportunity for improvements in cost-effectiveness. To facilitate this, a greater understanding and flexibility from all stakeholders involved will be required. With respect to estimating health state utility values in oncology, response mapping is shown to result in the most accurate and reliable predictions and should therefore be used in future economic evaluations, where appropriate. To move the science forward, this program of research would suggest an iterative approach to the economic evaluation of genomic test and treat strategies in oncology. This will greatly improve our understanding of their impact on health outcomes and the overall health budget, thus informing evidence-based decision making.