Table 1.
Case–control studies.
| Publication | Population, dataset | Period | Study design | Selection | Cases average age | Controls average age | Age-adjustment | Gene | Considered PVs | % PVs located in BRCA2 OCCR | Cases PV carriers/total (%) | Controls PV carriers/total (%) | OR (95% CI)a |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Johannesdottir [1] | Iceland | 1983–1992 | Cases vs controls from the same population | Cases: men diagnosed with PCa at age <65 at a single clinic (University Hospital of Iceland). Controls: participants in an unrelated public health study. | Not stated (all <65) | Not stated | None | BRCA2 | c.771_775del | 0% by design | 2/75 (2.67%) | 2/499 (0.40%) | 6.6 (0.81–56.9) |
| Hubert [3] | Israel | Not stated | Cases vs controls from the same population | Cases: unselected men diagnosed with PCa at a single clinic (Sharett Institute, Hadassah Hebrew University Hospital). Controls: recruited from homes for the elderly. | Median: 71 | Median: 72 | None, but cases and controls were of comparable ages | BRCA1 | c.68_69delAG | 2/87 (2.30%) | 2/87 (2.30%) | Not reported | |
| BRCA2 | c.5946delT | 100% by design | 1/87 (1.15%) | 1/87 (1.15%) | Not reported | ||||||||
| Vazina [4] | Israel | 1998 | Cases vs historical controls | Cases: unselected men diagnosed with PCa at three clinics (Rabin, Sheba or the Wolfson Medical Centers). Controls: historical US Ashkenazi controls [50]. | Median: 66 | Not stated (historical controls) | None | BRCA1 | c.68_69delAG | 4/87 (4.60%) | 61/5318 (1.15%) | Not reported | |
| BRCA2 | c.5946delT | 100% by design | 1/86 (1.16%) | 59/5087(1.16%) | Not reported | ||||||||
| Giusti [6] | Israel | 1994–1995 | Cases vs historical controls | Cases: unselected men diagnosed with PCa at 16 clinics. Controls: historical controls from the US Ashkenazi population [50] and an Israeli colorectal cancer case–control study [51]. | Mean: 73.6 | Not stated (historical controls) | None | BRCA1 | c.68_69delAG and c.5266dupC | 16/940 (1.70%) | 11/1344 (0.82%) | Not reported | |
| BRCA2 | c.5946delT | 100% by design | 14/940 (1.49%) | 10/1344 (0.74%) | 2.02 (0.89–4.56) | ||||||||
| Hamel [7]b | Canadian Ashkenazi | 1991–2002 | Cases vs historical controls | Cases: unselected Ashkenazi men diagnosed with PCa at three clinics in Montreal. Controls: historical controls from five studies with Ashkenazi general population or study control groups. | Mean: 67.9 | Not stated (historical controls) | None | BRCA1 | c.68_69delAG and c.5266dupC | 0/146 (0.00%) | 109/9371 (1.16%) | Not reported | |
| BRCA2 | c.5946delT | 100% by design | 2/146 (1.37%) | 119/9514 (1.25%) | Not reported | ||||||||
| Agalliu[10] | USA (predominantly European ancestry) | 1993–1996, 2002–2005 | Cases vs population frequency estimate | Cases: men diagnosed with PCa at age <55 in two case–control studies. No controls; comparison to a previous population BRCA2 frequency estimate for US Caucasians. | Median: 49.5 | -- | None | BRCA2 | c.3847_3848del and c.4398_4402del | 2/2 (100%) | 2/257 (0.78%) | Population frequency: 0.1% | 7.78 (1.80–9.37) |
| Agalliu [11] | US Ashkenazi | 1998–2005 | Cases vs controls from the same population | Cases and controls: self-selected Ashkenazi volunteers who were recruited through advertisements. The participants provided self-reported case/control status. | Mean: 69.4 | Mean: 68.3 | Covariate adjustment for age | BRCA1 | c.68_69delAG and c.5266dupC | 12/978 (1.23%) | 11/1247 (0.88%) | 1.39 (0.60–3.22) | |
| BRCA2 | c.5946delT | 100% by design | 18/969 (1.86%) | 12/1240 (0.97%) | 1.92 (0.91–4.07) | ||||||||
| Gallagher [12] | US Ashkenazi, MSKCC | 1988–2007 | Cases vs controls from the same population | Cases: unselected Ashkenazi men treated with PCa at a single clinic (Memorial Sloan-Kettering Cancer Center, New York). Controls: Ashkenazi healthy volunteers from a prospective study in New York. | Median: 68 | Median: 42 | Covariate adjustment for age | BRCA1 | c.68_69delAG | 6/832 (0.72%) | 4/454 (0.88%) | 0.38 (0.05–2.75) | |
| BRCA2 | c.5946delT | 100% by design | 20/832 (2.40%) | 3/454 (0.66%) | 3.18 (1.52–6.66) | ||||||||
| Fachal [13] | Spain | 2006–2009 | Cases vs controls from the same population | Cases: unselected men treated for PCa at one clinic (Santiago de Compostela). Controls: healthy men aged >44 (selection unclear). | Median: 68 | Median: 60 | Covariate adjustment for age | BRCA1 | c.211 A > G | 1/905 (0.11%) | 3/936 (0.32%) | 0.27 (0.01–2.36) | |
| Kote-Jarai [14] | UK, UKGPCS | Not stated | Cases vs population frequency estimate | Cases: men with PCa recruited nationwide due to being diagnosed with PCa at age <65 years (87% of the study sample), or due to having a family history of PCa (13% of the study sample). No controls; comparison to a previous UK population BRCA2 frequency estimate. | Not stated (87% <65) | -- | None | BRCA2 | Any pathogenic variant | 11/19 (58%) | 19/1832 (1.04%) | Population frequency: 0.16% | 8.6 (5.1–12.6) |
| Leongamornlert [15] | UK, UKGPCS | Not stated | Cases vs population frequency estimate | Cases: men with PCa recruited nationwide due to being diagnosed with PCa at age <65 years (90% of the study sample), or due to having a family history of PCa (10% of the study sample). No controls; comparison to a previous UK population BRCA1 frequency estimate. | Not stated (90% <65) | -- | None | BRCA1 | Any pathogenic variant | 4/886 (0.45%) | Population frequency: 0.12% | 3.75 (1.02–9.60) | |
| Cybulski [17] | Poland | 1999–2012 | Cases vs controls from the same population | Cases: unselected men with PCa from 14 centres. Controls: population controls from four sources (a random clinic record sample, a population-based study, PSA-screen negative men, colonoscopy screening participants). | Mean: 68.8 | Mean: 61.2 | None | BRCA1 | c.181 T > G, c.4035del and c.5266dupC | 14/3750 (0.37%) | 17/3956 (0.43%) | 0.9 (0.4–1.8) | |
| Akbari [18] | Canada (predominantly European ancestry) | 1998–2010 | Cases vs controls from the same population | Cases and controls: unselected men who had a biopsy because of elevated PSA or abnormal DRE at two clinics; cases were those biopsy-positive, controls were those biopsy-negative. | Mean: 65 | Not stated | None, but cases and controls were likely of comparable ages | BRCA2 | Any pathogenic variant | Not specified | 26/1904 (1.37%) | 9/2283 (0.39%) | 3.5 (1.63–7.48) |
| Pritchard [19] | UK and USA (predominantly European ancestry) | Not stated | Cases vs population frequency estimate | Cases: men with metastatic PCa from seven case series. No controls; comparison to carrier frequency in the Exome Aggregation Consortium database. | Not stated | Not stated (external estimate) | None | BRCA1 | Any pathogenic variant | 6/692 (0.87%) | 104/53105 (0.20%) | 3.9 (1.4–8.5) | |
| BRCA2 | Any pathogenic variant | 24/37 (65%) | 37/692 (5.35%) | 153/53105 (0.29%) | 18.6 (13.2–25.3) | ||||||||
| Matejcic [21] | US African Americans (AA) | 1993–2015 | Cases vs controls from the same population | Cases: men with PCa “overselected for high stage and Gleason score” from incident cases from a US prospective cohort study and two US case series of African American participants. Controls: unaffected African American participants in the US prospective cohort study. | Mean: 66.71 | Mean: 71.52 | Covariate adjustment for age and genetic ancestry | BRCA1 | Any pathogenic variant | 3/1447 (0.21%) | 1/995 (0.10%) | 2.84 (0.26–30.59) | |
| BRCA2 | Any pathogenic variant | Not stated | 9/1447 (0.62%) | 3/995 (0.30%) | 1.91 (0.48–7.59) | ||||||||
| Uganda | 2010–2016 | Cases vs controls from the same population | Cases: men with prostate cancer from 13 clinics in Uganda. Controls: patients recruited from non-urologic clinics in Uganda. | Mean: 70.77 | Mean: 65.04 | Covariate adjustment for age and genetic ancestry | BRCA1 | Any pathogenic variant | 2/651 (0.31%) | 1/486 (0.21%) | 1.11 (0.09–13.54) | ||
| BRCA2 | Any pathogenic variant | Not stated | 12/651 (1.84%) | 1/486 (0.21%) | 10.30 (1.28–82.58) | ||||||||
| Momozawa [22] | Japan, BioBank Japan | 2003–2018 | Cases vs controls from the same population | Cases: unselected men with PCa from a nationwide hospital-based biobank. Controls: male non-cancer patients from the same biobank older than 60 and with no personal or family history of cancer in first- or second-degree relatives. | Mean: 71.0 | Mean: 70.4 | None, but cases and controls were of comparable ages | BRCA1 | Any pathogenic variant | 14/7636 (0.18%) | 10/12366 (0.08%) | 2.27 (0.94–5.71) | |
| BRCA2 | Any pathogenic variant | Not specified | 83/7636 (1.09%) | 24/12366 (0.19%) | 5.65 (3.55–9.32) | ||||||||
| Oak [24] | USA (predominantly European ancestry), The Cancer Genome Atlas | 2005–2013 | Cases vs controls from the same population | Cases: men with PCa from a nationwide biobank. Controls: patients with non-prostate cancers from the same biobank. | Not stated | Not stated | Covariate adjustment for age and genetic ancestry | BRCA1 | Any pathogenic variant | 3/409 (0.73%) | Not stated (total: 7711 non-PCa patients) | 2.20 (0.62–7.83) | |
| Wokolorczyk [25] | Poland | 2000–2017 | Cases vs controls from the same population | Cases: men with PCa who had a family history of PCa in first- or second-degree relatives (three or more relatives with PCa, or two affected relatives of whom at least one was diagnosed before age 60). Controls: participants in an unrelated population-based study. | Mean: 61.6 | Mean: 59.4 | None, but cases and controls were of comparable ages | BRCA1 | Any pathogenic variant | 5/390 (1.28%) | 1/308 (0.32%) | 4.0 (0.5–34.3) | |
| BRCA2 | Any pathogenic variant | 0/4 (0%) | 4/390 (1.03%) | 0/308 (0.00%) | -- | ||||||||
| Nguyen-Dumont [26] | Australia | Not stated | Cases vs controls from the same population | Cases: men with aggressive prostate cancer (T4, M1, N1 or Gleason score≥8) from four cohort studies and case series. Controls: male participants in an unrelated trial.c | Median: 65–69 | Not stated (all ≥70) | Covariate adjustment for age | BRCA1 | Any pathogenic variant | 5/833 (0.60%) | 10/5356 (0.19%) | 2.9 (0.66–12.5) | |
| BRCA2 | Any pathogenic variant | 6/21 (29%) | 19/833 (2.28%) | 17/5356 (0.32%) | 3.9 (1.1–13) |
PCa prostate cancer, PV pathogenic variant, OCCR ovarian cancer cluster region, OR odds ratio, CI confidence interval.
aWhen available, the table includes adjusted odds ratio estimates (as indicated in the “age-adjustment” field). Otherwise, the reported unadjusted odds ratio estimates by each study are included. For studies that did not report odds ratios, unadjusted odds ratio estimates calculated from the frequencies of case and control PV carriers were used in the meta-analysis (not shown in this descriptive table but included in the forest plots).
bReported on both BRCA1 and BRCA2, but is not included in the BRCA1 meta-analysis due to observing no BRCA1 PVs in the cases which hence did not enable estimation of a 95% CI for the RR.
cThe main analysis in this study compared cases with aggressive prostate cancer to a combined comparison group comprising cases with non-aggressive prostate cancer and unaffected men. The meta-analysis includes the supplementary analysis of cases with aggressive prostate cancer versus unaffected men.