Tsur N, Kogan Y, Avizov-Khodak E, et al. - Researchers sought to develop a new personalization algorithm that may assist in predicting time to disease progression in metastatic melanoma under pembrolizumab treatment. A simple mathematical model for the interactions of an advanced melanoma tumor with both the immune system and the immunotherapy drug, pembrolizumab was developed and was implemented in an algorithm that, in conjunction with clinical pretreatment data, may allow prediction of the personal patient response to the drug. Clinical data of 54 patients with advanced melanoma, who had been treated by pembrolizumab, were obtained and personal pretreatment measurements were correlated to the mathematical model parameters, to develop the algorithm. They applied the algorithm in conjunction with the longitudinal tumor burden of each patient in order to assess the personal mathematical models, which were simulated to predict the patient’s time to progression. The Leave-One-Out cross-validation methodology was used to validate the prediction capacity of the algorithm. The activation rate of CD8+ T cells and the net tumor growth rate were noted to be significantly correlated with the baseline tumor load, the Breslow tumor thickness, and the status of nodular melanoma, among the analyzed clinical parameters. For the first time, it was identified that a comparatively simple mathematical mechanistic model, implemented in a personalization algorithm, can be personalized by clinical data, assessed before immunotherapy onset. The algorithm had moderate accuracy for predicting individual patients’ response to pembrolizumab and can be improved by training on a larger number of patients.