Economic evaluation of next generation sequencing in childhood cancer care

Cancer is the most common causes of death from chronic disease among children. Advances in genomic sequencing technology enables new treatment approaches, one of which is genomically guided precision medicine to be applied in patients with poor prognosis. Despite the promising benefits of precision medicine, evidence of this treatment approach being cost-effective is limited. The purpose of this thesis was to address gap in the current evidence using a meticulous and rigorous methodological approach. This thesis addressed two key objectives: (1) to provide comprehensive estimates of the hospitalisation utilisation and cost of childhood cancers; and (2) to develop a microsimulation model assessing the cost-effectiveness of precision medicine in childhood cancers. Modelling cost-effectiveness of precision medicine is complicated, designed to identify biomarkers that occur at either a low or unknown frequency across diseases. This heterogeneity of cohorts within such clinical trials makes modelling them using commonly used decision trees, or Markov Chain models challenging. Thus, microsimulation methods which are designed to capture heterogeneity were used. The hospitalisation costs of childhood cancers were determined via linking multiple administrative datasets. The linked datasets form the base population of a microsimulation model, Paediatric Cancer MOD (PeCanMOD), developed to evaluate the cost-effectiveness of applying nextgeneration sequencing and precision medicine in childhood cancers. Childhood cancers utilised a disproportionate amount of health care resources. The estimated median annual cost of hospitalization in the first-year post-diagnosis was AU$88,964 for patients diagnosed between 0-14 years old and AU$23,384 for patients diagnosed between 15-17 years old. Majority of the cost of hospital admissions occurred in the first-year post-diagnosis, accounting for more than 70% of hospital costs within 5 years post-diagnosis. PeCanMOD was demonstrated to be suitable for handling the complicated treatment algorithms commonly found in precision medicine trials. Output from PeCanMOD suggested that introducing precision medicine as last line of treatment in childhood cancers is unlikely to be cost-effective at this point. The overall incremental costs per QALY gained was over AU$300,000. However, a costeffective outcome could be achieved with a higher proportion of patients treated with precision medicine, combined with lower drug and sequencing costs and improved survival outcome.