Semi-automatic volumetric measurement of response to chemotherapy in lung cancer patients: How wrong are we using RECIST?

Objectives Lung cancer typically starts as a near-spherical lesion, but as it grows it may acquire an irregular radiologic formation. RECIST is based on the assumption that tumors are spherical, and consequently, proportional changes of tumor volume and parallel changes in tumor diameter, and vice versa. Hence, a 30% decrease in diameter (2r) implies a 65% decrease of volume, and a 20% increase in diameter implies a 73% increase of volume. Materials and methods We compared volumetric measurement based on multi-detector CT technology with calculated volume (CV) according to RECIST in a cohort of 43 patients with advanced, non-squamous cell type, lung cancer treated with a combination of platinum and pemetrexed. Results CV was larger than SMV in most patients both at baseline and at best overall response (BOR). The difference between the sum of volumes based on volumetric measurement (SMV) and CV was larger for higher sum of diameters. The Lin's concordance correlation coefficient between the percent changes in SMV and CV at BOR was 0.757. Of note, four patients (4/43, 9.5%) were categorized as PD according to the method of CV, but SD according to the method of SMV. Conclusion Our study highlights the importance of volumetric measurement for assessing response to treatment in lung cancer patients particularly showing large, irregular lesions.