Circulating high sensitivity C reactive protein concentrations and risk of lung cancer: Nested case-control study within Lung Cancer Cohort Consortium
Muller, David C.; Larose, Tricia L; Hodge, Allison; Guida, Florence; Langhammer, Arnulf; Grankvist, Kjell; Meyer, Klaus; Cai, Qiuyin; Arslan, Alan A.; Zeleniuch-Jacquotte, Anne; Albanes, Demetrius; Giles, Graham G.; Sesso, Howard D.; Lee, I-Min; Gaziano, J. Michael; Yuan, Jian-Min; Hoffman Bolton, Judith; Buring, Julie E.; Visvanathan, Kala; Le Marchand, Loic; Purdue, Mark P.; Caporaso, Neil E.; Midttun, Øivind; Ueland, Per Magne; Prentice, Ross L.; Weinstein, Stephanie J.; Stevens, Victoria L.; Zheng, Wei; Blot, William J.; Shu, Xiao-Ou; Zhang, Xuehong; Xiang, Yong-Bing; Koh, Woon-Puay; Hveem, Kristian; Thomson, Cynthia A.; Pettinger, Mary; Engström, Gunnar; Brunnström, Hans; Milne, Roger L.; Stampfer, Meir J.; Han, Jiali; Johansson, Mikael; Brennan, Paul; Severi, Gianluca; Johansson, Mattias
Journal article, Peer reviewed
MetadataShow full item record
Original versionBMJ-BRITISH MEDICAL JOURNAL. 2019, 364:k4981 1-9. 10.1136/bmj.k4981
Objectives: Inflammation may play an important role in lung cancer pathogenesis. Previous studies have suggested an association between C-reactive protein (CRP) as a marker of inflammation, and risk of lung cancer, but have lacked power to comprehensively investigate associations by smoking status and histological subtype. Our objective was to conduct a comprehensive analysis of prospectively measured circulating high-sensitivity CRP (hs-CRP) concentration and risk of lung cancer overall, by smoking status (current, former and never smokers), and histological sub-type. Design: Pooled analysis of prospective case-control studies nested within 20 population cohort studies. Setting: Population-based cohort studies in the Asia, Europe, Australia, and the United States. Participants: 5299 incident lung cancer cases, with individually incidence-density matched controls. Exposure: Circulating hs-CRP concentrations in pre-diagnostic serum/plasma samples. Main Outcome Measure: Incident lung cancer diagnosis. Results: We observed a positive association between circulating hs-CRP concentration and risk of lung cancer for current (OR per doubling in hs-CRP [ORlog2CRP]: 1.09, 95% CI: 1.05 to 1.13) and former smokers (ORlog2CRP 1.09, 95% CI: 1.04 to 1.14), but not never smokers (p-interaction <0.01). This association was strong and consistent across all histological subtypes, with the exception of adenocarcinoma, which was not strongly associated with risk regardless of smoking status (ORlog2CRP adenocarcinoma overall: 0.97, 95%CI: 0.94 to 1.01). Among current and former smokers, the associations were strongest within the first two years of follow-up. Including hs-CRP in a risk model in addition to smoking-based variables did not improve risk discrimination overall, but slightly improved discrimination for cancers diagnosed in the first two years of follow-up. Conclusions: Ever smokers with higher circulating hs-CRP concentrations had higher risk of lung cancer overall. hs-CRP was not associated with risk of lung adenocarcinoma. Circulating hs-CRP concentration may be a pre-diagnostic marker of lung cancer rather than a causal risk factor.