Fig. 3
From: RECIST 1.1 and lesion selection: How to deal with ambiguity at baseline?
Mathematical simulation of the proportional change of real malignant target lesions (TL) to trigger progression or response according to RECIST 1.1 thresholds in accordance with the proportion of non-malignant inactive lesions within the tumor burden. a For the Sum of Diameter thresholds corresponding to a progressive disease + 20% (red) and to a partial response − 30% (blue), we plotted the proportional change of the malignant part of the tumor burden at a given time point with respect to the proportion of inactive tumors in the tumor burden at baseline. We noted that the curve steepens from a proportion of about 50% of inactive lesions while the impact is limited to 20% of NMTL pooled in the tumor burden. b For the Sum of Diameter thresholds corresponding to a progressive disease, we simulated the needed change of malignant target to trigger a progression from baseline (red) (similar to a) and nadir20 (dark red) after a 20% PR. This scenario decreases even more the sensitivity of the RECIST assessment to capture progression