Predicting Patient-Specific Tumor Dynamics: How Many Measurements Are Necessary?

Area of Science:

• Oncology
• Mathematical Modeling
• Biostatistics

Background:

• Accurate prediction of tumor growth dynamics is crucial for effective breast cancer treatment.
• The logistic growth model is frequently used to describe tumor volume changes over time.
• Data acquisition and noise levels can significantly impact the accuracy of predictive models.

Purpose of the Study:

• To determine the minimum number of breast tumor volume measurements required for accurate prediction of tumor growth dynamics using the logistic growth model.
• To investigate the influence of data noise on the number of measurements needed.
• To establish a metric for clinicians to assess data sufficiency for treatment decisions.

Main Methods:

• Calibrated the logistic growth model using tumor volume data from 18 untreated breast cancer patients.
• Interpolated varying numbers of measurements at clinically relevant timepoints with simulated noise (0-20%).
• Compared errors in model parameters and data to identify sufficient measurement counts.

Main Results:

• Three tumor volume measurements were sufficient for accurate parameter estimation in the absence of noise.
• Increased levels of clinical noise necessitated a higher number of measurements for reliable growth dynamics prediction.
• Accurate tumor growth dynamics estimation depends on growth rate, noise level, and acceptable parameter error.

Conclusions:

• The study provides a framework for determining data sufficiency in breast tumor growth modeling.
• Clinicians can use the identified relationships to confidently predict patient-specific tumor growth and guide treatment.
• Optimizing data acquisition based on noise and growth rate improves predictive model accuracy.