Assessment of the Usefulness of the Quantitative Methods for the Response Evaluation of Solid Tumors: Analysis Using Four Cancer Types

Abstract

Background: The evidence provided by medical imaging techniques for the staging and follow-up is relevant in oncology. Objectives: The aims were (i) to compare the monitoring methods, (ii) to analyze the response variability between different tumors, and (iii) to decipher a general response curve that is independent of tumor type and drug treatment. Methods: We analyzed the response variability in four cancer types, looking for a general response curve independent of the tumor type and drug treatment. We compared the response of different types of lesions within each cancer type via an intra-class correlation coefficient, determining the minimum number of lesions suitable for monitoring. Results: The tested metrics allowed an objective evaluation of the response of solid tumors. The response was homogeneous between different cancer types. The intra-class correlation was high, allowing the monitoring of the response with a low number of lesions (2-4). The currently used metrics misrepresent the changes in the lesion volumes. Indeed, we observed non-linear overestimations of the RECIST and WHO values, which were more pronounced for the intermediate values. Additionally, the inclusion of lymphadenopathy among the target lesions produced a distortion in the evaluation of the response. Conclusion: The quantitative counts allowed an objective evaluation of the response of the solid tumors to therapy, showing that the response was homogeneous but variable between different types of tumors. Although the currently used metrics lead to misrepresentations of the changes in the lesion volume, they allowed setting a response pattern for tracking these lesions.