Performance of Lung Cancer Low-Dose CT Screening With Increasing Screening Interval in NELSON Trial
The Dutch and Belgian NELSON trial is assessing the effect of increasing interval low-dose computed tomography (CT) screening for lung cancer on lung cancer mortality. In an analysis of screening test performance among Dutch participants reported in Lancet Oncology, Horeweg and colleagues found high rates of specificity and sensitivity and a low incidence of interval cancers.
Study Details
The NELSON trial is a randomized trial assessing the effect of low-dose CT screening with screening intervals of 1, 2, and 2.5 years vs no screening on lung cancer mortality. Of 15,822 participants enrolled in the trial, 7,915 are in the screening group and 7,907 in the no-screening group. The current analysis included 7,155 Dutch participants who underwent at least one round of screening.
Eligible participants were aged 50 to 75 years, had smoked ≥ 15 cigarettes per day for > 25 years or ≥ 10 cigarettes per day for > 30 years and were still smoking or had stopped smoking < 10 years previously. People with self-reported moderate or bad health, inability to climb two flights of stairs, body weight ≥ 140 kg, current or past renal cancer, melanoma, or breast cancer, lung cancer diagnosed < 5 years previously, or chest CT examination < 1 year previously were excluded.
Screening results were based on volumetry; initial results were classified as negative, indeterminate, or positive based on nodule presence and volume. Participants with an initial indeterminate result had follow-up screening to classify final results as negative or positive based on nodule volume-doubling time.
Data are from the first three rounds of screening and an additional 2 years of follow-up from the national cancer registry. Median follow-up was 8.16 years.
Screening Test Performance
In total, 187 (3%) of 7,155 screened participants were diagnosed with 196 screen-detected lung cancers. A total of 34 (< 1%), including 19 in the first year after screening and 15 in the second year after screening, were diagnosed with 35 interval cancers.
For the three screening rounds combined and with 2-year follow-up, screening had a sensitivity of 84.6% (95% confidence interval [CI] = 79.6%–89.2%), specificity of 98.6% (95% CI = 98.5%–98.8%), positive predictive value of 40.4% (95% CI = 35.9%–44.7%), and negative predictive value of 99.8% (95% CI = 99.8%–99.9%).
Interval Cancers
Retrospective evaluation of the last screening CT and clinical CT in the 34 patients with interval cancers showed that the cancers were not visible in 12 cases (35%). In the remaining 22 cases, cancers were visible but remained undiagnosed due to radiologic detection and interpretation errors (17 cases, 50%), misclassification by the protocol (2 cases, 6%), participant noncompliance (2 cases, 6%), and nonadherence to protocol (1 case, 3%).
Compared with screen-detected cancers, interval cancers were diagnosed at more advanced stages (83% vs 22% stage III or IV, P < .0001), were more frequently small-cell carcinomas (20% vs 4%, P = .003), and were less frequently adenocarcinomas (26% vs 52%, P = .005).
The investigators concluded: “Lung cancer screening in the NELSON trial yielded high specificity and sensitivity, with only a small number of interval cancers. The results of this study could be used to improve screening algorithms, and reduce the number of missed cancers.”
The study was funded by Zorgonderzoek Nederland Medische Wetenschappen and Koningin Wilhelmina Fonds.
Nanda Horeweg, MD, of Erasmus University Medical Center, is the corresponding author for the Lancet Oncology article.
The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
