Lung cancer screening with MRI: Evaluation of MRI for lung cancer screening by comparison of LDCT- and MRI-derived Lung-RADS categories in the first two screening rounds.

Abstract

To evaluate MRI for lung cancer screening comparing LDCT- and MRI-derived Lung-RADS categories in the first two screening rounds. 224 participants in a lung cancer screening study were examined with MRI and low-dose CT (LDCT). Acquired MRI sequences were T2, balanced, T1 and DWI. MRI was prospectively analysed regarding nodules. Minimum nodule size was 4 mm. Nodules were assigned a Lung-RADS score based on appearance and size at baseline and after 3, 6 and 12 months. MRI findings were correlated with LDCT. The early recall rate dropped from 13.8% at baseline to 1.9% in the second screening round with biopsy rates of 3.6% in the first round and 0.5% in the second round. Histology revealed lung cancer in 8/9 participants undergoing biopsy/surgery. All eight cancers were accurately depicted by MRI. The following categories were assigned on MRI (results of LDCT in parentheses): 4B/4X in 10 (10) cases, 4A in 16 (15) cases, 3 in 13 (12) cases, 2 in 77 (92) cases and 1 in 140 (126) cases. Lung-RADS scoring correlated significantly between MRI and CT. The score was overestimated by MRI in one case for category 4A, in two cases for category 3 and in five cases for category 2. MRI-based Lung-RADS score was underestimated for category 1 in 20 cases. Lung-RADS might be applied for lung cancer screening with MRI, since findings correlated with LDCT. Relevant findings with a Lung-RADS score of 3 and higher were never missed or underestimated by MRI KEY POINTS: • MRI performed comparably to low-dose CT in a lung cancer-screening programme. • Lung-RADS might be applied for lung cancer screening with MRI. • Lung-RADS findings score of 3 and higher were never missed by MRI.