Low-Dose Computed Tomography for Lung Cancer Screening: Comparison of Performance Between Annual and Biennial Screen

Aims and objectives The Multicenter Italian Lung Detection (MILD) was a lung cancer screening trial by low-dose computed tomography (LDCT) with peculiar design. Participant in MILD trial were randomized to two different LDCT arms, namely annual LDCT (LDCT1) and biennial LDCT (LDCT2). The aim of this study was to compare the performance metrics between two different strategies of lung cancer screening by LDCT. Methods and materials Randomization of study population Consenting participants in the MILD trial were randomly assigned into two arms: a) control arm undergoing primary prevention program with pulmonary function test and blood sample collection; b) LDCT arm undergoing the same primary prevention program and LDCT. The LDCT arm was further randomized into two subgroups: LDCT1 with annual LDCT scan; LDCT2 with biennial LDCT scan. Results Participants of LDCT arms underwent: 6893 LDCTs in LDCT1 4715 LDCTs in LDCT2 A median of 6 ± 1.7 follow-up LDCT1 scans and 3 ± 1.7 follow-up LDCT2 scans over a median follow up of 7.3 years were obtained. The LDCT arms were similar for nodules number: 1418 nodules detected in LDCT1 1392 nodules detected in LDCT2. Participants adherence was maintained susbstantial untill T6 and T3 rounds for the LDCT1 and LDCT2 participants, respectively (Fig. 1, 2). Conclusion Recall rate and detection rate are high priority targets for optimization of lung cancer screening, along with reduction of screens. This study shows that a biennial screening may be at least as efficient as the annual screening in terms of detection rate, sensitivity, specificity, PPV and NPV. The slightly higher total recall rate in the LDCT2 arm was counterbalanced by the lower number of scans, notably, about two thirds of scans performed in LDCT1 arm.