Article Figures & Data
Figures
- Figure 1.
Selection of studies for the systematic review.
- Figure 2.
Meta-analysis of studies investigating prostate cancer risk for men with Lynch syndrome stratified by different study designs. Horizontal lines represent 95% CI. Each box represents the RR point estimate, and its area is proportional to the weight of study (the inverse variance of each study's effect). The diamond represents the combined summary estimated RR, with 95% CI given by the width of the diamond. The unbroken vertical line is at the null value (RR = 1). CRC, colorectal cancer; FDR, first-degree relative.
- Figure 3.
Funnel plot for the meta-analysis of studies investigating prostate cancer risk for men with Lynch syndrome. The vertical line in the funnel plot indicates the fixed-effects summary estimates (using inverse-variance weighting). The sloping lines indicate the expected 95% CIs for a given standard error assuming no heterogeneity between studies. Asymmetry in the funnel plot indicates a systematic underreporting of smaller and negative studies (publication bias) or a systematic difference between smaller and larger studies that arises from inherent between-study heterogeneity.
Tables
- Table 1.
Summary of molecular studies that investigated MMR deficiency status of prostate cancer tumors of men with Lynch syndrome
No. of MMR-deficienta prostate cancer/No. of confirmed MMR gene mutation carriers No. of MSI-H tumor/No. of tumor tested for MSI No. of tumor with absence of MMR protein expression in IHC/No. of tumor tested by IHC MMR gene mutated Single case study Soravia et al. (16) 1/1 1/1 1/1 MSH2 Wagner et al. (17) 1/1 1/1 1/1 MSH2 TOTAL 2/2 Case series Bauer et al. (21) 2/2 2/2 NT MSH2 Buttin et al. (18) 1/2 0/1 1/2 MSH6 Grindedal (19) 7/8 NT 7/8 MSH2, MSH6 Rosty et al. (20) 22/32 5/10 22/32 MLH1, MSH2, MSH6 TOTAL 32/44 Abbreviation: NT, not tested.
-
↵aDisplaying high MSI (MSI-H) and/or absence of MMR protein expression in immunohistochemistry.
-
- Table 2.
Summary of risk studies that investigated risk of prostate cancer for men with Lynch syndrome
Author Year Country Study population Study design Results Comments Zhang et al. (25) 2005 China 53 families fulfilling the Amsterdam-I (40) and four families fulfilling the Amsterdam-II criteria (41). Cross-sectional. O = 2 Unable to calculate SIR. E = not reported Age at diagnosis = not reported Ascertained families fulfilling the Amsterdam-I or II criteria from 51 regions of China. Goecke et al. (26) 2006 Germany 988 members (418 confirmed carriers, 22 obligate carriersa, and 548 first- or second-degree relatives) from 281 families with MLH1 (n = 124) or MSH2 (n = 157) mutations. Retrospective cohort. O = 10 (eight in confirmed or obligate MSH2 mutation carriers, two in untested relatives of MSH2 families, and no prostate cancer observed in MLH1 families) Unable to calculate SIR. E = not reported Families fulfilling the Amsterdam-II Criteria (41) or the original Bethesda guidelines (55) were ascertained through the six German centers. Mean age at diagnosis = 59 y Barrow et al. (28) 2009 UK 839 carriers (249 confirmed carriers, 90 obligate carriersb, 331 putative carriersc, and 169 assumed carriersd) from 121 MMR gene mutation-carrying families (51 MLH1, 59 MSH2, and 11 MSH6). Retrospective cohort. O = 6 Unable to calculate SIR. E = not reported Age at diagnosis = not reported Ascertained from families fulfilling the Amsterdam (41) or Bethesda criteria (42) who attended the Manchester Regional Genetics Service. Stupart et al. (29) 2009 South Africa 200 (102 male) confirmed carriers of MLH1 c.C1528T mutation from 17 families. Prospective cohort. O = 1 (19 extracolonic cancers in both men and women). Unable to calculate SIR. E = not reported Ascertainment of families was not described. Age at diagnosis = not reported da Silva et al. (30) 2010 Brazil 2,095 (1,040 male) members from 60 families fulfilling the Amsterdam-I (40) or -II criteria (41). Retrospective cohort. O = 16 (21% of 77 extracolonic cancers observed in men) Unable to calculate SIR. E = not reported Used families registered in the Oncotree database of the Hereditary Colorectal Cancer Registry of the Pelvic Surgery Department of Hospital AC Camargo, Sao Paulo. Age at diagnosis = >70 y in “most of the cases” Author stated, “the age of diagnosis was over 70 years and because of this, prostate cancer is unlikely to be part of LS spectrum of tumors.” Talseth-Palmer et al. (31) 2010 Australia 78 MSH6 mutation carriers from 29 families and seven PMS2 mutation carriers from six families. Retrospective cohort. O = 7 in relatives of MSH6 carriers Unable to calculate SIR. E = not reported Ascertained from families referred to Hunter Area Pathology Service in Newcastle, New South Wales, for genetic testing due to fulfilling the Amsterdam-II (41) or Bethesda Criteria (42) or MSI or IHC absence of MSH6/PMS2 expression. Mean age at diagnosis (SD) = 65 (7.6); calculated from the reported ages (70, 60, 75, 56, and 64, and two missing). Scott et al. (24) 2001 Australia 95 families (22 MLH1, 12 MSH2, 61 unknown) fulfilled the Bethesda Criteria (55) (n = 63) or the Amsterdam Criteria (40) (n = 32). Retrospective cohort. For carriers only. For untested relatives, this cohort may contain noncarriers and therefore likely to underestimate the true risk. O = 1 (MSH2 carrier) E = not reported SIR = 1.02 (95% CI, 0.1–13.6) Ascertainment of families was not described. For both carriers and relatives, O = 11 E = not reported SIR = 2.69 (95% CI, 1.2–5.8) Age at diagnosis = not reported Maul et al. (27) 2006 U.S. 65 CRC cases and 509 FDRs from 18 families fulfilling Amsterdam-I criteria (40). Retrospective cohort. For CRC cases, No ascertainment bias. O = 4 E = 1.62 SMR = 2.5 (95% CI, 0.67–6.34) For CRC cases, the estimates may reflect only for those with a prior diagnosis of CRC. Identified through the Utah Population Database. For FDRs, For FDRs, this cohort may contain noncarriers and therefore likely to underestimate the true risk. O = 19 E = 8.73 SMR = 2.18 (95% CI, 1.31–3.39) Age at diagnosis = not reported Barrow et al. (35) 2013 UK 292 male confirmed carriers (130 MSH2, 141 MLH1, and 21 MSH6) and 529 male untested FDRs. Retrospective cohort. For MSH2 carriers, O = 4 For confirmed carriers, estimate of association may be upwardly biased.e E = 0.38 RR = 10.41 (95% CI, 2.8–26.65) Ascertained from the Manchester Regional Genetics Service where referral was due to clustered colorectal cancer and extracolonic cancers. For untested FDRs of MSH2 carriers, For untested relatives, this cohort may contain noncarriers and therefore likely to underestimate the true risk. O = 3 E = 1.24 RR = 2.41 (95% CI, 0.48–7.03) For untested FDRs of MSH6 carriers, O = 1 E = 0.29 RR = 3.36 (95% CI, 0.04–18.67) Median age at diagnosis (range) = 67.3 (54.2–76.0) y Aarnio et al. (22) 1999 Finland 360 carriers (265 confirmed and 95 obligateb carriers) from 50 families with MLH1 (n = 47) or MSH2 (n = 3) mutations. Retrospective cohort. O = 4 No ascertainment bias. E = not reported SIR = 2.9 (95% CI, 0.8–7.4) Ascertained through the population-based nationwide Finnish Cancer Registry. Age at diagnosis = not reported Grindedal et al. (19) 2009 Norway 106 confirmed or obligatef male carriers (68 MSH2, 19 MLH1, 13 MSH6, and 6 PMS2) from 34 families. Retrospective cohort. O = 9 (6 MSH2, 2 MSH6, and 1 MSH6) Estimate of association may be upwardly biased.e E = 1.52 SIR = 5.9 (95% CI, 4.1–17.1) Ascertained from families referred to the Section for Hereditary Cancer at The Rikshospitalet University Hospital in Oslo. Cumulate risk to age 70 y = 30% (SE 0.088) Mean age at diagnosis (range) = 60.4 (53–68) y Expected age at diagnosis = 66.6 y Engel et al. (32) 2012 Germany, Netherlands 2,118 (1,107 male) confirmed carriers (806 MLH1, 1,004 MSH2, 308 MSH6). Retrospective cohort. E = not reported Estimate of association may be upwardly biased.e O = not reported SIR = 2.5 (95% CI, 1.4–4.0) Ascertained from families fulfilling the Amsterdam-II (41) or Bethesda criteria (42) through the German HNPCC Consortium and the registry of the Netherlands Foundation for the Detection of Hereditary Tumours. Cumulative risk to 70 y: 9.1% (95% CI, 4.4–13.8) Median age at diagnosis (range) = 59 (50–74) y Pande et al. (34) 2012 U.S. 368 (152 male; 165 probands and 203 relatives) confirmed carriers (152 MLH1, 197 MSH2, 16 MSH6, 3 PMS2) from 176 families. Retrospective cohort. For both probands and relatives, Estimate of association may be upwardly biased, especially for estimates of both probands and relatives.e O = 3 E = 3.22 SIR = 0.93 (95% CI, 0.19–2.7) Probands ascertained from the gastroenterology and gynecologic oncology clinics and Clinical Cancer Genetics Clinics at MD Anderson Cancer Center. For relatives only, O = 2 E = 1.52 SIR = 1.3 (95% CI, 0.15–4.7) Median age at diagnosis = 65 y Raymond et al. (36) 2013 U.S.A. 8,314 (4,127 male) members from 198 mutation-carrying families (74 MLH1, 101 MSH2, and 23 MSH6). Retrospective cohort. Overall HR = 1.99 (95% CI, 1.31–3.03) Ascertainment bias corrected by conditioning on the proband's genotype and phenotype status. For ages 20–59 y, HR = 2.48 (95% CI, 1.34–4.59) Ascertained through family cancer clinics at the Dana-Farber Cancer Institute, Boston, and University of Michigan Comprehensive Cancer Centre, Ann Arbor. Cumulative Risk to age 70 y = 17.39 (95% CI, 10.10–24.07) The estimates may reflect only for carriers from family cancer clinics or carriers with a relative with CRC or endometrial cancer. Median age at diagnosis (range) = 65 (38–89) y Dowty et al. (15) 2013 Australia, USA, Canada 17,576 members from 166 MLH1 and 224 MSH2 mutation-carrying families. Retrospective cohort. For MLH1 carriers, Ascertainment was corrected by statistical methods conditioning the likelihood for each pedigree. O = 45 HR* = 0.79 (95% CI, 0.25–2.5) Ascertained from population cancer registries or family cancer clinics of the Colon Cancer Family Registry. For MSH2 mutation carriers, O = 60 HR = 1.0 (95% CI, 0.47–2.3) Mean age at diagnosis (SD) = 69.6 (10.4) y for MLH1 and 68.6 (11.5) y for MSH2 The estimates may reflect only for MLH1 or MSH2 carriers. Win et al. (14) 2012 USA, Canada, Australia, New Zealand 446 confirmed carriers (161 MLH1, 222 MSH2, 47 MSH6, and 16 PMS2). A median 5-year follow-up. Prospective cohort. O = 3 No ascertainment bias. E = 1.21 SIR = 2.49 (95% CI, 0.51–7.27) Ascertained from population cancer registries or family cancer clinics of the Colon Cancer Family Registry. Median age at diagnosis (range) = 54 (50–62) y Hemminki et al. (23) 2001 Sweden 4,794 CRC cases from 88 families fulfilling the Amsterdam (40, 41) or modified Bethesda criteria (23). Retrospective cohort. O = 1 No ascertainment bias. E = not reported SIR = 3.52 (95% CI, 0.00–13.81) Identified through the Swedish Family-Cancer Database. Age at diagnosis = not reported The estimates may reflect only for those with a prior diagnosis of CRC. Win et al. (33) 2012 USA, Canada, Australia, New Zealand 764 confirmed carriers (316 MLH1, 357 MSH2, 49 MSH6, and 42 PMS2) with a prior diagnosis of CRC. Retrospective cohort. For all carriers, No ascertainment bias. O = 19 E = 9.25 SIR = 2.05 (95% CI, 1.23–3.01) The estimates may reflect only for carriers with a prior diagnosis of CRC. Ascertained from population cancer registries or family cancer clinics of the Colon Cancer Family Registry. Median age at diagnosis (range) = 64 (55–77) y For MLH1 carriers, O = 3 E = 3.44 SIR = 0.87 (95% CI, 0.00–2.19) For MSH2 carriers. O = 15 E = 4.15 SIR = 3.62 (95% CI, 2.07–5.36) For MSH6 carriers. O = 1 E = 1.16 SIR = 0.86 (95% CI, 0.00–3.03) Abbreviations: LS, Lynch syndrome; O, observed number of prostate cancers; E, expected number of prostate cancers; SMR, standard morbidity ratio; FDR, first-degree relative; CRC, colorectal cancer.
-
↵aObligate carrier was not defined (26).
-
↵bObligate carrier was defined because of their position in the pedigree in relation to a confirmed MMR gene mutation carrier (22, 28).
-
↵cPutative mutation carriers were first-degree relatives of a confirmed mutation carrier with a Lynch syndrome–related cancer (colorectal, endometrial, ovarian, gastric, brain, biliary, small bowel, and sebaceous adenocarcinoma; ref. 28).
-
↵dAssumed mutation carriers were half of the untested first-degree relatives with non–Lynch syndrome spectrum cancers; or a proportion of untested, unaffected first-degree relatives with no cancers calculated on the basis of the proportion of unaffected relatives who tested positive of the total number of individuals actually tested for each age group (28).
-
↵eEstimates of prostate cancer risk are likely to be upwardly biased if any of the family members attended family cancer clinics because of a family history of prostate cancer (see text for details).
-
↵fObligate carrier was defined as being in between relatives who were confirmed mutation carriers and/or had shown loss of the relevant gene product by immunohistochemistry of a Lynch syndrome–associated tumor if mutation testing had not been possible (19).
-
Volume 23, Issue 3
