Comparison of 1D, 2D, 3D and volumetric parameters in measuring tumor response in high-grade gliomas in adults

Abstract

1523 Background: : The RECIST criteria (1D measurement) is widely used to measure response in solid tumors but there are few studies validating this criteria in brain tumors. In a small study in childhood brain tumors, Warren et al. (JNCI 2001;92:205) found a high concordance between 1D, 2D and 3D measurements in detecting partial responses but not disease progression. More recently, Galanis et al (ASCO Abst 423, 2003) found that tumor responses assessed by RECIST and 2D measurements were equally associated with survival improvement. In this study we compared the 1D, 2D, 3D and computer assessed tumor volume (CV) measurements in adult brain tumors. Methods: : Tumor size on MRI scans in 80 patients with high-grade gliomas treated on clinical protocols were measured using 1D, 2D, 3D and CV. A total of 278 T1 post-gadolinium MRI scans (80 baseline and 198 follow up scans) were evaluated. Volumetric analysis was performed using VITREA. Results: Pearson correlation coefficients (ρ) with 95% confidence intervals (CI) are presented in the table for each measurement modality. All ρ values are greater than 0.70 with 95% CI no wider than 14%. Correlation was best between the 1D vs 2D and 2D vs 3D group. The poorest correlations were between the 1D and 3D and 1D and volumetric criteria. Median progression-free survival (MPFS) and 6-month progression free survival (6m-PFS) were 5.9 months (95% CI=5.0–7.6) and 45% (SD 6%) for 1 D, 6.0 m (5.1–7.6) and 47% (6%) for 2D, 6.0 m (5.1–7.7) and 47% (6%) for 3D, 6.2 m (5.1–8.2) and 49% (6%) for volumetric analysis. Conclusions: These results suggest that there is good concordance between the RECIST criteria and 2D measurements in determining response, MPFS and 6m-PFS in adult brain tumor patients. The correlations between 1D or 2D criteria and volumetric measurements are not as strong, but still high. No significant financial relationships to disclose.