Lung cancer: computerized quantification of tumor response--initial results

Area of Science:

• Radiology and Imaging Science
• Oncology
• Medical Informatics

Background:

• Accurate assessment of tumor response is crucial for guiding lung cancer treatment.
• Traditional unidimensional and bidimensional measurements may not fully capture volumetric changes.
• Advancements in imaging analysis offer potential for more precise tumor response quantification.

Purpose of the Study:

• To prospectively evaluate a semiautomated algorithm for quantifying tumor response in lung cancer patients.
• To compare the efficacy of tumor volume calculation versus traditional measurements (unidimensional and bidimensional) using thin-section CT.
• To assess tumor response or progression by analyzing changes in tumor volume and other parameters.

Main Methods:

• A semiautomated three-dimensional lung cancer segmentation algorithm was developed and applied to CT scans.
• Tumor volume, greatest diameter (unidimensional), and product of diameters (bidimensional) were calculated.
• 15 non-small cell lung cancer patients' scans before and after gefitinib treatment were analyzed; Exact McNemar tests compared measurement techniques.

Main Results:

• The algorithm accurately segmented 14 of 15 tumors, with one requiring manual adjustment.
• Semiautomated volume analysis identified significantly more patients with at least 20% tumor volume change (73%) compared to unidimensional (7%) and bidimensional (27%) measurements.
• Higher thresholds also showed greater sensitivity with volume: 47% of patients had >=30% volume change versus 0% (unidimensional) and 13% (bidimensional).

Conclusions:

• Semiautomated tumor segmentation and volume calculation provide a more sensitive method for assessing lung cancer response.
• This technique identifies a greater number of patients experiencing significant tumor volume changes (>=20% and >=30%) compared to traditional methods.
• Volumetric assessment using CT offers improved accuracy in quantifying tumor response or progression in lung cancer patients.