Number of lesions measured and the relative percent change in tumor burden and response rate.

Abstract

e16504 Background: In early-phase clinical trials and clinical practice, response to treatment is assessed by the relative percent change in overall tumor burden (RC) from baseline to follow-up visits. Patients frequently have multiple metastatic lesions, and there is some debate over the number of target lesions to include when calculating the RC. The recently updated Response Evaluation Criteria in Solid Tumors (RECIST 1.1) recommends including a maximum of five lesions, a decrease from the ten lesions previously recommended in RECIST 1.0. Methods: We derive an analytic expression for the distribution of the RC from baseline to follow-up. Using data from eight clinical trials studying therapies with varying degrees of activity, we calculate the expected RCs and response rates (RRs) for a range of number of lesions measured. Results: Average RCs and RRs changed based on the number of lesions measured. The magnitude of this difference depends upon the situation. In some cases, fewer than five lesions may be sufficient while in other cases it may be preferable to measure more than five lesions. For instance, in situations where there was a low RR together with a high correlation between baseline and follow-up measurements, we found less variation in the average RCs and RRs across the number of lesions measured than in other situations. Two different scenarios are presented in the Table. Conclusions: Knowing the distribution of the RC allows investigators to obtain an idea of the number of target lesions to measure when planning clinical trials. This number depends upon a variety of factors which are inputs to our analytic expression. Relative percent change from baseline to follow-up measurement Number of lesions measured 1 2 3 4 5 6 7 8 9 10 Scenario 1: Moderate response rate, moderate correlation between baseline to follow-up measurements Average -1% -7% - 10% -11% -12% -12% -13% -13% -13% - 13% SD 0.62 0.44 0.36 0.32 0.29 0.28 0.27 0.25 0.24 0.23 RR = Prob (RC ≤ -30%) 38% 34% 32% 32% 30% 29% 28% 26% 25% 24% Scenario 2: Very low response rate, high correlation between baseline to follow-up measurements Average 17% 15% 15% 14% 14% 14% 14% 14% 14% 14% SD 0.38 0.30 0.27 0.26 0.25 0.24 0.23 0.23 0.23 0.22 RR = Prob (RC ≤ -30%) 7% 4% 2% 2% 1% 1% 1% 1% 1% 1% No significant financial relationships to disclose.