Abstract LB-162: Causal role of body mass index in lung cancer in mendelian randomization analysis accounting for genetic pleiotropism
Bibliographic record
Abstract
Abstract Background: Body mass index (BMI) is inversely associated with lung cancer in observational studies, despite obesity increases risk for many other cancers. Mendelian Randomization (MR) analysis was previously employed to clarify the causal role of BMI in lung cancer. However, the results of MR were potentially biased by BMI-SNPs with pleiotropic effects on smoking and hence violating the assumption of MR. Objective: To examine the causal role of BMI in lung cancer using MR analysis after excluding BMI-SNPs with potential pleiotropic effects on smoking. Methods: The OncoArray data from the Transdisciplinary Research in Cancer of the Lung (TRICL) team of the International Lung Cancer Consortium (ILCCO) were used. To obtain a valid genetic instrument for MR analysis, we first identified a set of 241 BMI-SNPs (P<10-5) from the GIANT consortium, then filtered out SNPs with potential pleiotropic effects on smoking using causal network inference, and obtained a final set of 46 SNPs to calculate the genetic risk score for BMI (GRS46). Logistic regression model tested its association with lung cancer adjusting for age, sex, and genetic principle components. We compared results of GRS46 with those from GRS241 that included pleiotropic SNPs. Results: A total of 32,240 subjects of European descent (17,976 lung cancer cases and 14,264 controls) were included in the analysis. The average age was 63.0 (SD: 10.5) years, 61.2% were male, 59.5% were overweight (BMI: 25.0-29.9 kg/m2) or obese (BMI≥30 kg/m2), and 79.3% were ever smokers. GRS46 was positively associated with lung cancer (OR: 1.57, 95% CI: 1.20-2.07) in both never smokers (OR: 2.18, 95% CI: 1.10-4.31) and ever smokers (OR: 1.40, 95% CI: 1.02-1.92). GRS46 was significantly associated with increased risk of non-small cell lung cancer (OR: 1.69, 95% CI: 1.21-2.36), adenocarcinoma (OR: 1.76, 95% CI: 1.19-2.60), and squamous cell carcinoma (OR: 1.98, 95% CI: 1.18-3.33), but not with small cell lung cancer (SCLC) (OR: 1.29, 95% CI: 0.63-2.64). In contrast, GRS241 was significantly associated with SCLC (OR: 1.57, 95% CI: 1.21-2.05), which is highly attributable to smoking, but not associated with lung cancer subgroups that are not or less attributable to smoking: never smokers (OR: 1.00, 95% CI: 0.78-1.29) and adenocarcinoma (OR: 1.07, 95% CI: 0.92-1.24). The pleiotropic smoking-associated SNPs in GRS241 might have confounded the results in these subgroups. Conclusion: Increased BMI is a risk factor for lung cancer in both never and ever smokers and for major histological subtypes after accounting for genetic pleiotropism. We demonstrated that previous MR analyses that used a BMI-GRS including SNPs with pleiotropic effects on smoking could yield spurious results. Funding: This work was support by R21 CA202529 (Wang/Ho). Confirmation: These data were not published previously and will not publish prior to the dates of the AACR Annual Meeting 2018. Citation Format: Yiqun Wu, Jee-Young Moon, Christopher I. Amos, Rayjean J. Hung, Gloria Y. Ho, Tao Wang. Causal role of body mass index in lung cancer in mendelian randomization analysis accounting for genetic pleiotropism [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr LB-162.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.020 | 0.048 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".