Tumor Response Evaluation Using iRECIST: Feasibility and Reliability of Manual Versus Software-Assisted Assessments

Area of Science:

• Radiology and Medical Imaging
• Oncology
• Cancer Treatment Response Assessment

Background:

• Accurate assessment of tumor response is crucial for guiding cancer treatment decisions.
• Immune-based therapies have revolutionized cancer treatment, necessitating precise response evaluation criteria like iRECIST.
• Manual assessment of imaging data can be time-consuming and prone to inter-reader variability.

Purpose of the Study:

• To compare the feasibility and reliability of manual versus software-assisted assessments of computed tomography (CT) scans.
• To evaluate these methods using the immune RECIST (iRECIST) criteria in patients undergoing immune-based cancer treatment.
• To determine the impact on reading time, accuracy, and inter-reader agreement.

Main Methods:

• CT scans from 30 cancer patients were assessed by four radiologists at baseline and two follow-ups.
• Tumor burden and response status were evaluated manually and via software-assisted methods.
• Reading time, sum of longest diameter (SLD), and response were recorded; consensus reading established a reference standard.

Main Results:

• Software-assisted assessments were significantly faster than manual assessments at follow-ups (p < 0.001).
• Manual assessments showed higher error rates (FU1: 3.3% vs. 0%; FU2: 10% vs. 1.7%) and lower inter-reader agreement (ICC) for SLDs.
• Software-assisted approach improved reliability and reduced misclassifications.

Conclusions:

• Software-assisted assessment streamlines iRECIST response evaluation in clinical practice.
• This approach decreases reading time and minimizes response misclassifications for cancer patients.
• It enhances the reliability and consistency of tumor response assessment in immune-based cancer therapy.