Comparing quantitative imaging biomarker alliance volumetric CT classifications with RECIST response categories.

Area of Science:

• Radiology and Medical Imaging
• Oncology
• Biomarker Development

Background:

• Accurate tumor response assessment is critical for cancer treatment evaluation.
• Quantitative Imaging Biomarker Alliance (QIBA) CT volumetry offers a detailed approach to measuring tumor changes.
• The Response Evaluation Criteria in Solid Tumors (RECIST) is a widely used standard for tumor response assessment.

Purpose of the Study:

• To evaluate the agreement between tumor change classifications derived from QIBA CT volumetry and RECIST criteria.
• To identify and quantify discrepancies between QIBA CTvol and RECIST classifications.
• To determine the nature of disagreements, distinguishing between substantive differences and those due to enhanced sensitivity of volumetry.

Main Methods:

• Analysis of target lesions (lung, liver, lymph nodes) from 10 historical cancer clinical trials.
• Independent segmentation of lesions by three radiologists using two software tools at baseline and follow-up.
• Classification of disagreements into Type I (substantive) and Type II (volumetry sensitivity-related).

Main Results:

• Overall agreement between QIBA CTvol and RECIST was 66.6%.
• Discrepancies occurred in 33.4% of interpretations, with substantive disagreements (Type I) in only 1.5%.
• Larger lesions (≥50 mm) showed significantly more disagreements; scanner vendor, segmentation tool, and lesion type were not significant predictors.

Conclusions:

• QIBA CTvol classifications demonstrate substantial agreement with RECIST categories.
• The sensitivity of CT volumetry does not frequently lead to clinically significant disagreements with RECIST.
• QIBA CTvol is a reliable method for assessing tumor response, comparable to RECIST.