A E Hosman1, C L Shovlin2,3. 1. Department of Pulmonology, St Antonius Hospital, Nieuwegein, The Netherlands. 2. Respiratory Medicine, Imperial College Healthcare NHS Trust, London, UK. c.shovlin@imperial.ac.uk. 3. NHLI Cardiovascular Sciences, Imperial Centre for Translational and Experimental Medicine, Imperial College London, Hammersmith Campus, Du Cane Road, London, W12 0NN, UK. c.shovlin@imperial.ac.uk.
Abstract
OBJECTIVE: To examine associations between cancer incidence and hereditary haemorrhagic telangiectasia (HHT). METHODS: Two studies with contrasting conclusions were compared. The first had used a registry-based, matched-pairs approach, while the second utilised HHT family-based, survey methodology. RESULTS: The first manuscript captured data on cancer incidence in a total of 316,581 matched cancer patients-non-cancer controls, which included 431 HHT cases. No association was found between HHT and pooled cases of lung, breast, prostate, and colorectal cancer (adjusted OR 0.978, 95% CI [0.795, 1.204]). The second, which was powered to examine these four cancers individually, captured data from 2161 HHT cases and 2817 related controls. Fewer HHT-affected individuals had cancer (398/2161, [18.4%]) compared to 668/2817 (23.7%) related controls (p = 0.0012). Of the four most common cancers, prostate and colorectal cancer rates were equivalent, but lung cancers were significantly less frequent in HHT (adjusted OR 0.48 [0.30, 0.70], p = 0.0012), and breast cancer was more frequent (adjusted OR 1.52 [1.07, 2.14] p = 0.018). CONCLUSIONS: The respective studies had different methodological strengths and weaknesses. Potential reasons for the discrepant conclusions include study power, particularly important to dissect specific cancers where differential contributions from HHT genotypes and environmental confounders might be predicted.
OBJECTIVE: To examine associations between cancer incidence and hereditary haemorrhagic telangiectasia (HHT). METHODS: Two studies with contrasting conclusions were compared. The first had used a registry-based, matched-pairs approach, while the second utilised HHT family-based, survey methodology. RESULTS: The first manuscript captured data on cancer incidence in a total of 316,581 matched cancerpatients-non-cancer controls, which included 431 HHT cases. No association was found between HHT and pooled cases of lung, breast, prostate, and colorectal cancer (adjusted OR 0.978, 95% CI [0.795, 1.204]). The second, which was powered to examine these four cancers individually, captured data from 2161 HHT cases and 2817 related controls. Fewer HHT-affected individuals had cancer (398/2161, [18.4%]) compared to 668/2817 (23.7%) related controls (p = 0.0012). Of the four most common cancers, prostate and colorectal cancer rates were equivalent, but lung cancers were significantly less frequent in HHT (adjusted OR 0.48 [0.30, 0.70], p = 0.0012), and breast cancer was more frequent (adjusted OR 1.52 [1.07, 2.14] p = 0.018). CONCLUSIONS: The respective studies had different methodological strengths and weaknesses. Potential reasons for the discrepant conclusions include study power, particularly important to dissect specific cancers where differential contributions from HHT genotypes and environmental confounders might be predicted.
Authors: A M Jelsig; P M Tørring; A D Kjeldsen; N Qvist; A Bojesen; U B Jensen; M K Andersen; A M Gerdes; K Brusgaard; L B Ousager Journal: Clin Genet Date: 2015-12-21 Impact factor: 4.438
Authors: Christine W Duarte; Kimberly Murray; F Lee Lucas; Kathleen Fairfield; Heather Miller; Peter Brooks; Calvin P H Vary Journal: Cancer Epidemiol Biomarkers Prev Date: 2013-11-05 Impact factor: 4.254
Authors: Inês G Mollet; Dilipkumar Patel; Fatima S Govani; Adam Giess; Koralia Paschalaki; Manikandan Periyasamy; Elaine C Lidington; Justin C Mason; Michael D Jones; Laurence Game; Simak Ali; Claire L Shovlin Journal: PLoS One Date: 2016-02-11 Impact factor: 3.240