ABSTRACT

In metastatic cancer, the growth and drug-induced shrinkage of individual tumor lesions may behighly dependent on the microenvironment of the hosting organ, and individual lesions cancontribute differently to overall disease progression and survival. In the traditional analysis of tumorresponse to treatment, the overall measurement of patient tumor burden is used, i.e. the sum of thelongest diameters (SLD) of up to 10 lesions (RECIST criteria v 1.0). Information carried byindividual lesions and their impact on the disease outcome is hence ignored. This analysis aimed todevelop a population model to better understand and characterize the differences in tumor dynamicsbetween lesions and different metastatic sites. Using lesion measurement data from 183 subjectswith metastatic HER2-negative breast cancer receiving docetaxel, a kinetic/pharmacodynamic modelhas been developed. The framework model was able to describe metastatic breast cancer data on thelevel of lesion, organ, and patient. Besides, a logistic regression dropout model allowed to accountfor the loss of follow-up based on tumor size and identify influencing covariates. A new lesionappearance model was able to predict the development of new metastasis and find predictors.Finally, an overall survival model enabled exploring the relationship between tumor measurementand survival expectancy. This developed framework may provide a better understanding of therelationship between drug exposure and tumor response on the level of lesion, organ, and patient.Thus, it may help drive decisions in clinical studies and increase our knowledge about metastasis.