Sun Young Kim, Jung Kim, Mark Ramos, Jeremy S. Haley, Diane T. Smelser, H. Shanker Rao, Uyenlinh L. Mirshahi, Katherine L. Nathanson, Barry I. Graubard, Hormuzd A. Katki, David J. Carey, Douglas R. Stewart, Aris Baras, Gonçalo R. Abecasis, Adolfo A. Ferrando, Giovanni Coppola, Andrew Deubler, Luca A. Lotta, John D. Overton, Jeffrey G. Reid, Alan R. Shuldiner, Katherine Siminovitch, Jason Portnoy, Marcus B. Jones, Lyndon J. Mitnaul, Alison Fenney, Jonathan Marchini, Manuel Allen Revez Ferreira, Maya Ghoussaini, Mona Nafde, William Salerno, Cristen J. Willer, Lourdes Crane, Christina Beechert, Erin Fuller, Laura M. Cremona, Eugene Kalyuskin, Hang Du, Caitlin Forsythe, Zhenhua Gu, Kristy Guevara, Michael Lattari, Alexander Lopez, Kia Manoochehri, Prathyusha Challa, Manasi Pradhan, Raymond Reynoso, Ricardo Schiavo, Maria Sotiropoulos Padilla, Chenggu Wang, Sarah E Wold, Manan Goyal, George Mitra, Sanjay Sreeram, Rouel Lanche, Vrushali Mahajan, Sai Lakshmi Vasireddy, Gisu Eom, Krishna Pawan Punuru, Sujit Gokhale, Benjamin Sultan, Pooja Mule, Mudasar Sarwar, Muhammad Aqeel, Xiaodong Bai, Lance Zhang, Sean O’Keeffe, Razvan Panea, Evan Edelstein, Ayesha Rasool, Evan K. Maxwell, Boris Boutkov, Alexander Gorovits, Ju Guan, Lukas Habegger, Alicia Hawes, Olga Krasheninina, Samantha Zarate, Adam J. Mansfield, Joshua Backman, Kathy Burch, Adrián Campos, Liron Ganel, Sheila M. Gaynor, Benjamin Geraghty, Arkopravo Ghosh, Salvador Romero Martinez, Christopher E. Gillies, Lauren Gurski, Eric Jorgenson, Tyler Joseph, Michael D. Kessler, Jack A. Kosmicki, Adam E. Locke, Priyanka Nakka, Karl Landheer, Olivier Delaneau, Anthony Marcketta, Joelle Mbatchou, Arden Moscati
Importance: There is clear evidence that deleterious germline variants in CHEK2 increase risk for breast and prostate cancers; there is limited or conflicting evidence for other cancers. Objective: To quantify the prevalence of as well as cancer risk and survival associated with CHEK2 germline pathogenic and likely pathogenic variants using genomic ascertainment. Design, Setting, and Participants: This case-control study used 2 electronic health record-linked and exome-sequenced biobanks: UK Biobank (n = 469 765) and Geisinger MyCode (adults only; n = 167 050). Variants were classified according to American College of Medical Genetics and Genomics and the Association for Molecular Pathology criteria. Cases were defined as individuals with heterozygous CHEK2, harboring pathogenic or likely pathogenic variants; controls as individuals with a benign or likely benign CHEK2 variation or wildtype CHEK2. Cancer registry (MyCode since approximately 1943; UK Biobank since approximately 1970) and demographic data were retrieved; to adjust for relatedness, association analysis was performed with SAIGE-GENE+ with Bonferroni correction. Main Outcomes and Measures: Prevalence of as well as cancer risk and survival in adults with CHEK2 germline variants. Results: Of 469 765 individuals in the UK Biobank, there were 3232 case participants (mean [SD] age, 70.8 [8.0] years; 3139 [97.1%] White; 1744 [54.0%] women); of 167 050 individuals with MyCode, there were 3153 case participants (mean [SD] age, 60.5 [17.8] years; 3123 [98.8%] White; 1935 [61.5%] women). In case participants in both MyCode and UKBB, there was a significant excess risk of all cancers (odds ratio [OR], 1.33 [95% CI, 1.18-1.49]; OR, 1.41 [95% CI, 1.26-1.59], respectively), breast (OR, 1.54 [95% CI, 1.18-2.00]; OR, 1.84 [95% CI, 1.49-2.27], respectively), prostate (OR, 1.62 [95% CI, 1.27-2.07]; OR, 1.78 [95% CI, 1.48-2.16], respectively), kidney (OR, 1.58 [95% CI, 1.03-2.41]; OR, 1.84 [95% CI, 1.22-2.77], respectively), and bladder (OR, 1.50 [95% CI, 1.01-2.23]; OR, 1.64 [95% CI, 1.17-2.31], respectively) cancers as well as lymphoid leukemia (OR, 2.08 [95% CI, 1.17-3.69]; OR, 2.21 [95% CI, 1.19-4.08], respectively). Compared with control participants, time to cancer in case participants was significantly shorter in both cohorts; no significant difference was observed between the age-dependent penetrance of truncating and missense variants for cancer in either cohort. Overall survival was significantly decreased in case participants in UK Biobank; however, the primary consequence was seen after 75 years. There was no statistical difference in survival in MyCode. There were no differences in survival between case participants with cancer and control participants with cancer. Conclusions and Relevance: In this case-control study of genomic ascertainment of individuals with heterozygous CHEK2 pathogenic or likely pathogenic variants in 2 population-scale cohorts, there was a significant excess risk of breast, prostate, kidney, bladder, and lymphoid leukemia cancer. The conferred excess mortality and cancer risk was low (ORs <2). This has clinical implications for individuals ascertained this way (vs with a family history of cancer).