INFORME: coupling information-theoretic experimental design with nonlinear mixed-effects modeling for efficient observation scheduling
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Mathematical models of treatment response can inform individualized therapy, but their calibration often requires longitudinal measurements that are costly, burdensome, and collected on fixed schedules. Such schedules may be inefficient, over-sampling patients whose response is already well characterized while delaying informative measurements for those whose model parameters remain uncertain. We present INFORME (INFORmation-theoretic design with Mixed Effects), a framework that combines Bayesian information-theoretic experimental design with nonlinear mixed-effects modeling to adaptively select each patient’s next measurement time. Population and response-subgroup parameter distributions learned from an existing cohort provide informative priors, allowing candidate measurement times to be ranked by their expected reduction in patient-specific parameter uncertainty. As observations accumulate, priors can be updated to reflect the response subgroup most consistent with the patient’s data. We evaluate INFORME in two radiotherapy datasets: 150 synthetic tumor volume trajectories from a hybrid cellular automaton model of prostate cancer spheroids (HD1) and longitudinal tumor volumes from 39 patients with head-and-neck cancer (HD2). In HD1, population priors allowed omission of both pretreatment scans, while adaptive scheduling reduced the protocol from nine scans to three or four, with the response group identified from a single post-treatment scan on day 27. In HD2, the adaptive schedule used three scans instead of six and improved prediction by delaying the first on-treatment scan from week 1 to week 2, avoiding transient dynamics that produced false-positive and false-negative response projections. Across both datasets, the adaptive schedules used a mean of 2.7 scans in stead of seven and advanced completion of the patient-specific prediction by a mean of 15.5 days (95% CI, 6.7–24.3) relative to the equidistant protocol, while treatment duration remained unchanged. INFORME therefore reduces measurement burden and accelerates patient-specific prediction by concentrating observations at times that are most informative for model calibration.
Author summary
When a patient is treated for cancer, repeated measurements of tumor state track its response to treatment. Mathematical models can use these measurements to predict treatment response, but they need enough data to make reliable predictions. Measurements, such as imaging, are also costly, time-consuming, and burdensome for patients. In many studies, they are collected on a fixed schedule that does not incorporate the new information that has already been learned about an individual patient. We developed a method that answers when each patient’s next measurement would be most useful to inform model calibration. It first learns from previously treated patients how tumors typically respond. Then, as measurements are collected from a new patient, it identifies when another measurement would provide the most useful information about that patient’s response. We tested this approach using simulated tumors and measurements from patients with head-and-neck cancer. By concentrating scans at more informative times, our method could reach predictions earlier while using substantially fewer measurements. In the simulated study, a patient’s response group could be identified from the first scan after treatment began. Our results show that better-timed measurements can reduce scanning burden while providing useful information about treatment response sooner.
