Soon Ho Yoon

M.D.

Seoul National University Hospital

Assistant professor

NLST (Learn more about this study)

NLST-465

Jan 2, 2019

Gender difference in benefits of lung cancer screening

Background
1) Current evidence of gender difference in a lung cancer CT screening
One of the main findings in the final result of the NELSON trial is that lung cancer screening may be more effective in female heavy-smokers than in male heavy-smokers, but the statistical significance is not available at the moment (Risk ratio of lung cancer-specific mortality in CT arm, 0.74 (0.60-0.91) vs 0.61 (0.35-1.04))(WCLC 2018). The similar gender difference was observed in the NLST trial, although the difference did not reach statistical significance (0.92 vs 0.73; P=0.08)(Cancer. 2013 Nov 15;119(22):3976-83).

2) Adenocarcinoma-specific survival benefit of lung cancer screening
Low dose chest CT screening tended to be more effective for adenocarcinoma than for squamous cell carcinoma. The screening did not affect the outcome of small cell lung cancer (Cancer. 2013 Nov 15;119(22):3976-83; Chest. 2018 Dec;154(6):1284-1290). Similarly, adenocarcinoma tended to have a more favorable outcome than squamous cell carcinoma in early-stage lung cancer (Jpn J Clin Oncol 2012;42(3)189–195).

3) Female lung cancer and adenocarcinoma
Adenocarcinomas account for approximately 45% in male lung cancer and 55% female lung cancer in 2010 in the United States SEER 9 registry (PLoS One. 2015 Mar 30;10(3):e0121323).

4) Favorable outcome in female lung adenocarcinoma
Among patients receiving platinum-based chemotherapy, women had a longer survival than men in adenocarcinoma pathology (Annals of Oncology 21: 2023–2028, 2010).

Summary of the findings
Women potentially have more benefits from lung cancer CT screening. It might result from the relatively higher prevalence of adenocarcinoma in women or a similar prevalence of adenocarcinoma but a more favorable outcome of early-stage adenocarcinoma in women. The gender difference may also originate from an earlier age, less smoking at the time of diagnosis of lung cancer, more favorable stage shift in female heavy-smoker. Furthermore, we should consider the presence of bronchioloalveolar carcinoma which tends to be more prevalent in women but can result in over-diagnosis.


Currently, six randomized-controlled trials on lung cancer screening agreed to participate in this meta-study: MILD, ITALUNG, DANTE, DLCST, LSS, and NLST (via CDAS).
A statistical model for the analysis should include several covariates including age, gender, smoking, lung cancer pathology, and stage. The gender-specific difference cannot be adequately modeled with study-level data due to the small number of studies (6 studies) but the relatively large number of covariates (5 covariates).
Thus, we decided to perform a meta-analysis based on the anonymized patient-level data of individual trials.

We ought 1) to meta-analytically evaluate whether a lung cancer screening is more beneficial for women than for men in heavy smokers across randomized-controlled trials, and 2) to investigate the potential reasons. We will apply the following inclusion criteria to determine the eligibility.
a) Lung cancer screening trial using a low dose chest computed tomography scan
b) A randomized controlled trial with a control arm of usual care or chest radiograph
c) Available final results of lung cancer trial with a median follow-up of 5 years or longer after the randomization
d) Trial with the participation of the principal investigator with a provision of anonymized individual data

The main outcome of this study is a relative risk reduction of lung cancer-specific mortality by low dose chest CT screening according to gender.