Purpose: To determine whether 2D or 3D Choi and modified Choi (mChoi) criteria could assess the efficacy of everolimus against metastatic renal cell carcinoma (mRCC).
Methods: RECIST-1.1, Choi, and mChoi criteria were applied retrospectively to analyse baseline and 2-month contrast-enhanced computed tomography (CECT) images in 48 patients with mRCC enrolled in the everolimus arm of the French randomized double-blind multicentre phase III trial comparing everolimus versus placebo (RECORD-1). The primary endpoint was centrally reviewed progression-free survival (PFS) calculated from the initial RECORD-1 analysis. Mean attenuation was determined for 2D target lesion regions of interest drawn on CECT sections whose largest diameters had been measured, and for the 3D whole target lesion.
Results: The median PFS was 5.5 months. The median PFS for everolimus responders defined using 3D mChoi criteria was significantly longer than for non-responders (7.6 versus 5.4 months, respectively), corresponding to a hazard ratio for progression of 0.45 (95 % CI: 0.22-0.92), with respective 1-year survival rates of 31 % and 9 %. No other 2D or 3D imaging criteria at 2 months identified patients who would benefit from everolimus.
Conclusions: At 2 months, only 3D mChoi criteria were able to identify mRCC patients with a PFS benefit from everolimus.
Key points: Choi criteria could not identify everolimus-treated patients with significantly prolonged PFS. mCHOI enabled identification of everolimus-treated mRCC patients with a PFS benefit. 3D attenuation measurement criteria appeared to perform better than single-slice measurement.
Keywords: 3D imaging; Computed tomography; Neoplasm metastasis; Renal cell carcinoma; Tumour response.