Abstract
Mathematical models of cancer–immune dynamics rely on uncertain parameters, limiting predictive reliability. We apply Approximate Bayesian Computation (ABC) to calibrate key parameters using observed tumor response data. By sampling random parameters from uniform distribution and simulating with them, we can obtain the likelihood of the parameters based on the distance to the observed data. The posterior distributions based on the distances show that some parameters are strongly constrained, while others remain broadly distributed, indicating potential non-identifiability. This framework suggests model applicability and personalized medicine of immunotherapy in cancer treatment.