https://orcid.org/0000-0002-6013-0785
-
PDF
- Split View
-
Views
-
Cite
Cite
Ya Zhang, Pengfei Lyu, RE: Genetic risk, health-associated lifestyle, and risk of early-onset total cancer and breast cancer, JNCI: Journal of the National Cancer Institute, Volume 117, Issue 4, April 2025, Pages 801–802, https://doi.org/10.1093/jnci/djae322
Close - Share Icon Share
We read with great interest the recently published article in the Journal by Zhang et al. titled “Genetic risk, health-associated lifestyle, and risk of early-onset total cancer and breast cancer.”1 This study provided valuable insights into the interaction between genetic predispositions and lifestyle factors, offering an important contribution to the field of cancer epidemiology.
The study notably used data from the UK Biobank, which enriched the research with comprehensive datasets, but it is important to acknowledge that cohort studies based on volunteers, such as the UK Biobank, have inherent biases. Only 5.5% of individuals initially invited participated in the UK Biobank, and these individuals tended to be healthier and better educated and to lead healthier lifestyles than the general UK population.2,3 This selection bias may have substantially affected the determination of health-related lifestyle scores in this study.
Moreover, the issue of participation bias is not limited to lifestyle assessments. It is equally critical to assess to what extent participation bias may have skewed genomic studies and subsequent analyses when determining genetic risk. Observational studies using UK Biobank data have shown that participation bias can distort the associations between phenotypic exposures and outcomes.4 A previous study applying genome-wide analyses to a more representative (weighted) sample of the UK Biobank demonstrated that selective participation could skew genomic results and downstream analyses.5 Correcting for participation bias by adopting samples that more accurately represent the target population could enhance the robustness of genetic findings.6
The authors used multivariable Cox proportional hazards models to estimate the associations between polygenic risk score and health-associated lifestyle score groups with early-onset total cancer and breast cancer, but this approach may underestimate the risks of these outcomes.7 To produce more robust results, we suggest conducting sensitivity analyses using competing-risk models, where death is treated as a competing risk event. Doing so would provide a clearer picture of the true risk estimates, especially in a cohort that may have substantial mortality due to cancer or other causes.
In conclusion, Zhang et al.’s work is an excellent and valuable contribution to the understanding of genetic and lifestyle influences on early-onset cancer. The thorough analysis and insightful conclusions offer a solid foundation for future research in personalized cancer prevention. Although the study holds notable clinical implications, however, it is essential to exercise caution when interpreting the results.
Author contributions
Ya Zhang, MD (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Validation; Visualization; Writing—original draft; Writing—review & editing) and Pengfei Lyu, MD (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Supervision; Writing—original draft; Writing—review & editing).
Consent for publication
All authors have read and approved the final manuscript.
Funding
None declared.
Conflicts of interest
None declared.
Data availability
Not applicable.
