INFLUENCE: individualized follow-up for breast cancer

In contrast to significant progress in precision medicine and the tailoring of therapeutic strategies, current follow-up after breast cancer is still consensus-based and not dependent on the risk of recurrence of the individual patient. Better and personalized follow-up based on individual risk can reduce the burden for both the patients and the health care capacity due to the increasing patient volume requiring follow-up. To determine the risk of recurrence, data from the nation-wide population-based Netherlands Cancer Registry (NCR) was used from the incidence years 2003-2006 (N=37 230). The risk of recurrence was modelled using different techniques using baseline clinical and pathological characteristics. A model for time-dependent risk of recurrence was implemented in the INFLUENCE nomogram (www.utwente.nl/influence). Also, insight in long-term risk and risk of subsequent recurrence was acquired. Using a retrospective cohort study, guideline adherence was determined. Results showed no personalization on known risk factors and lower risk patients were actually given more frequent follow-up compared to higher risk patients. Additionally, the age-based recommendations after five years of follow-up do not relate to the risk of recurrence. Multiple risk factors should be taken into account when estimating recurrence risk. Subsequently, a partially observable Markov decision process (POMDP) was developed that takes into account the risk, as well as benefits and harm of follow-up to provide optimized follow-up schedules. This will enable clinicians to make informed decisions and focus resources on patients with higher risk, while avoiding unnecessary and potentially harmful follow-up visits for women with very low risks.