Table 2.
Results from testing the association between polygenic risk scores based on UK data, and a phenotype status, based on Icelandic data
| PRSs | Phenotype | Effect (β) | P-value | PSA_increase/PRS_SD (%) | 95% c.i. (%) |
|---|---|---|---|---|---|
| (a) Separately | |||||
| PC | PSA levels | 0.089 | 9.8 × 10−68 | 16.3 | (14.3, 18.3) |
| BPH/LUTS | PSA levels | 0.071 | 6.0 × 10−45 | 12.9 | (10.9, 14.8) |
| (b) Jointly | |||||
| PC | PSA levels | 0.074 | 4.1 × 10−43 | 13.3 | (11.3, 15.3) |
| BPH/LUTS | PSA levels | 0.049 | 3.0 × 10−20 | 8.6 | (6.7, 10.5) |
Shown are results from testing the association of polygenic genetic risk scores (PRSs), based on effect estimates from the UK for: prostate cancer (PC) and benign prostatic hyperplasia/lower urinary tract symptoms (BPH/LUTS), for correlation with serum levels of PSA (PSA levels) in 18,929 Icelandic males. Shown are the effect estimates (β), the two-sided P-values calculated using logistic regression in R (v3.5), the percentage increase in PSA levels for each standard deviation (SD) increase in the PRSs, and the 95% confidence intervals (c.i.)
In section a the results are shown separately for the PRSs of prostate cancer (PC) and BPH/LUTS, whereas in section b the results are shown jointly (i.e. after being conditioned for each other)