Abstract Introduction. Pediatric germ cell tumors (GCT) are a rare form of childhood cancer that originate from the primordial germ cell. Little is known about etiology and few risk factors have been identified. Both pediatric GCT and adult testicular GCT (TGCT) appear to initiate in utero and thus may share etiologic factors. Several SNPs have been associated with adult TGCT in genome-wide association studies (GWAS). Here we test whether these SNPs are associated with pediatric GCT. Methods. This case-parent triad study includes cases diagnosed with GCT at age 0-19 years between July 1, 2008 and December 31, 2013 who were identified through the Children's Cancer Research Network (CCRN). Buccal cell DNA was collected from cases and their biological parents. We identified 21 SNPs at 15 independent loci from GWAS of adult TGCT and estimated main effects for pediatric GCT using the transmission disequilibrium test (TDT). We used Estimation of Maternal, Imprinting and interaction effects using Multinomial modelling (EMIM) to evaluate maternal effects in non-Hispanic white trios. A Bonferroni correction was used to account for multiple comparisons. Results. DNA was available for 364 complete trios. The SPRY4 SNP rs4624820 was significantly associated with reduced risk of GCT (OR [95% CI]: 0.69 [0.56, 0.85]; Table). BAK1 SNP rs210138 was associated with an increased risk estimate for GCT (OR [95% CI]: 1.70 [1.32, 2.18]), with a particularly strong estimated effect for testis tumors (OR [95% CI]: 3.31 [1.89, 5.79]). Nominal associations (p<0.05) were noted for 5 additional loci. Maternal effects were observed for KITLG SNP rs4474514 (OR [95% CI]: 1.87 [1.29, 2.71]) and rs7221274 (OR [95% CI]: 2.07 [1.43, 2.99]), near TEX14, RAD51C, and PPM1E. Conclusions. We observed significant associations between SNPs in SPRY4 and BAK1 and pediatric GCTs, with a particularly strong association between the SNP in BAK1 and testis tumors. These SNPs may impact risk of GCT through apoptosis pathways. The results of this analysis support that there are common genetic risk factors for pediatric and adult GCT. Associations between adult TGCT GWAS variants and risk of pediatric GCTMain effectsMaternal effectsMarker NameOriginal GWAS markerLocusGeneMAFRisk alleleOR (95% CI)p-valueOR (95% CI)p-valuers2072499rs20724991q22SLC0.3578C1.21 (0.96, 1.51)0.098990.84 (0.60, 1.18)0.31475rs3790672rs37906721q24.1UCK20.3104C1.01 (0.81, 1.26)0.911501.19 (0.70, 1.39)0.32380rs3773832rs105104523p24.3DAZL0.2953T0.92 (0.73, 1.14)0.431000.99 (0.70, 1.39)0.93238rs8336rs170214634q22.2HPGDS0.3977G0.93 (0.75, 1.15)0.478800.85 (0.59, 1.21)0.35992rs2720460rs27204604q24CENPE0.3606G1.09 (0.88, 1.35)0.416600.83 (0.59, 1.17)0.28210rs4624820rs4624820/rs68978765q31.1SPRY40.432A0.69 (0.56, 0.85)0.000470.89 (0.63, 1.28)0.54166rs474853rs38056635q31.1PITX10.4133A0.75 (0.60, 0.93)0.007601.03 (0.73, 1.45)0.87347exm-rs210138rs2101386p21.31BAK10.2445G1.71 (1.34, 2.20)0.000021.25 (0.88, 1.77)0.20662exm599133rs126994777p22.3MAD1L10.4092T1.28 (1.04, 1.58)0.019361.08 (0.76, 1.53)0.65794rs10275045rs102750457p22.3MAD1L10.4684T1.22 (0.99, 1.50)0.059871.10 (0.76, 1.57)0.61265rs6972549rs37788917p22.3MAD1L10.2905T0.95 (0.76, 1.20)0.683601.03 (0.73, 1.46)0.85882rs11986044rs70101628q13.3PRDM140.3228T0.76 (0.60, 0.97)0.023640.77 (0.54, 1.11)0.15983exm-rs7040024rs70400249p24.3DMRT10.2198A0.73 (0.56, 0.95)0.018100.81 (0.56, 1.17)0.26198rs7553831rs7553839p24.3DMRT10.2816T1.03 (0.81, 1.30)0.811701.33 (0.93, 1.90)0.12198exm-rs4474514rs447451412p22KITLG0.2043A0.76 (0.59, 0.99)0.040511.87 (1.29, 2.71)0.00085exm-rs995030rs99503012p22KITLG0.1964G0.80 (0.61, 1.05)0.102501.69 (1.16, 2.45)0.00603rs8046148rs804614816q12.1MFSD10.2074G0.69 (0.53, 0.89)0.003490.91 (0.63, 1.33)0.64298rs4888262rs488826216q22.3RFWD30.4993T0.92 (0.74, 1.13)0.420701.13 (0.79, 1.63)0.49973rs7221274rs722127417q22TEX14, RAD51C, PPM1E0.3606A0.92 (0.74, 1.13)0.422002.07 (1.43, 2.99)0.00008rs11656666rs990570417q22TEX14, RAD51C, PPM1E0.2699A0.85 (0.68, 1.07)0.174601.48 (1.05, 2.10)0.02651rs2250213rs283918621q22.3MCM3AP0.454A1.11 (0.90, 1.37)0.314600.87 (0.61, 1.23)0.43117 Citation Format: Erin L. Marcotte, Nathan Pankratz, James Amatruda, A. Lindsay Frazier, Mark Krailo, Stella Davies, Jacqueline Starr, Erica Langer, Caroline Hallstrom, Anthony Hooten, Jenny N. Poynter. Variants in BAK1 and SPRY4 are associated with pediatric germ cell tumors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr LB-374.
Read full abstract