Background

Previous studies suggest polygenic risk scores (PRS) may improve melanoma risk stratification. However, there has been limited independent validation of PRS-based risk prediction, particularly assessment of calibration (comparing predicted to observed risks).

Objectives

To evaluate PRS-based melanoma risk prediction in prospective UK and Australian cohorts with European ancestry.

Methods

We analysed invasive melanoma incidence in UK Biobank (UKB; n=395,647; 1,651 cases) and a case-cohort nested within the Melbourne Collaborative Cohort Study (MCCS, Australia; n=4,765; 303 cases). Three PRS were evaluated: 68 SNPs at 54 loci from a 2020 meta-analysis (PRS68); 50 SNPs significant in the 2020 meta-analysis excluding UKB (PRS50); 45 SNPs at 21 loci known in 2018 (PRS45). 10-year melanoma risks were calculated from population-level cancer registry data by age group and sex, with and without PRS adjustment.

Results

Predicted absolute melanoma risks based on age and sex alone underestimated melanoma incidence in UKB (ratio expected/observed cases E/O=0.65, 95% confidence interval (95%CI) 0.62-0.68) and MCCS (E/O=0.63, 95%CI 0.56-0.72). For UKB, calibration was improved by PRS-adjustment, e.g. PRS50-adjusted risks E/O=0.91, 95%CI 0.87-0.95. Discriminative ability for PRS68- and PRS50-adjusted absolute risks was higher than for risks based on age and sex alone (ΔAUC 0.07-0.10, p<0.0001), and higher than for PRS45-adjusted risks (ΔAUC 0.02-0.04, p<0.001).

Conclusions

A PRS derived from a larger, more diverse meta-analysis improves risk prediction compared to an earlier PRS, and might help tailor melanoma prevention and early detection strategies to different risk levels. Re-calibration of absolute risks may be necessary for application to specific populations.