PhD ceremony Ms. Y. Zhao: Lung nodule assessment in low-dose CT lung cancer screening. Validation of detection and volumetric measurement

PhD ceremony: Ms. Y. Zhao, 14.30 uur, Academiegebouw, Broerstraat 5, Groningen

Dissertation: Lung nodule assessment in low-dose CT lung cancer screening. Validation of detection and volumetric measurement

Promotor(s): prof. M. Oudkerk

Faculty: Medical Sciences

Early detection by computed tomography (CT) could improve the survival in lung cancer. The Dutch-Belgian Randomized Lung Cancer Screening Trial (Dutch acronym: NELSON). investigates whether screening by low-dose multi-detector CT in high risk subjects will decrease lung cancer mortality. The NELSON trial is the first in which lung nodule management is based on nodule volume at first detection, and volume doubling time on follow-up examinations.

As described in this thesis, over 7500 participants at high risk of lung cancer underwent CT screening in the NELSON trial. In the baseline round, 2.6% of the participants had a positive test result. 0.9% had lung cancer, the majority being an early stage of disease. A negative test ruled out the probability of lung cancer. Secondly, computer-aided detection (CAD) software detected more lung nodules than double reading. The combination of CAD and nodule size cut-off considerably reduces the false positive rate of CAD. Semi-automated software tools for volumetry yielded different nodule volumes. Variations between software results may lead to false-positive or false-negative screening results. Different software tools should not be used in a single screening trial. Lastly, the 3 month follow-up CT for intermediate-size nodules greatly reduced false-positive findings as many nodules show no or slow growth, or have disappeared. The NELSON strategy diminishes the need for further (invasive) work-up in participants with an intermediate-size nodule.

In conclusion, CT lung cancer screening based on volume and volume-doubling of lung nodules, with short-term repeat CT for intermediate-size nodules, is efficient to detect lung cancer.