AUTHORS
Brittany Stewart, CGC, PhD, Darren McAffee, PhD2, Erica Vaccari, MS3, Michelle Green, PhD4, Rebecca Previs, MD5, Jonathan Williams, PhD6, Grant Hogg, MS7, Josh Keyes, MS8, Jill Polk, MS9, Wendy McKinnon, CGC10, Gergi Mansour, MD11, Sakshi Jasra, MD12, Kara Landry, MD13, Hetty Carraway, MD, MBA14, Sarah M. Young, MS15
AFFILIATIONS
1 Leukemia Program, Taussig Cancer Institute, Cleveland Clinic, Cleveland, Illinois, United States
2 Labcorp (formerly Invitae Corp.), San Francisco, Illinois, United States
3 Labcorp (formerly Invitae Corp.), San Francisco, Illinois, United States
4 Labcorp, Durham, North Carolina, United States
5 Labcorp, Durham, North Carolina, United States
6 Labcorp, San Diego, California, United States
7 Labcorp, San Diego, California, United States
8 Labcorp (formerly Invitae Corp.), San Francisco, Illinois, United States
9 Labcorp (formerly Invitae Corp.), San Francisco, Illinois, United States
10 University of Vermont Cancer Center, Burlington, Vermont, United States
11 University of Vermont Cancer Center, Burlington, Vermont, United States
12 University of Vermont Cancer Center, Burlington, Vermont, United States
13 University of Vermont Cancer Center, Burlington, Vermont, United States
14 Leukemia Program, Taussig Cancer Institute, Cleveland Clinic, Cleveland, Illinois, United States
15 Labcorp (formerly Invitae Corp.), San Francisco, Illinois, United States
ABSTRACT
Background: Germline genetic testing for hereditary hematologic malignancies (HHM) genes is increasingly incorporated into broad multigene panel testing (MGPT) for cancer predisposition, both for patients with and without indications for HHM evaluation. We evaluated the yield of germline pathogenic variants (PVs) in HHM genes in a large, heterogeneous, indication-agnostic cohort of patients undergoing MGPT.
Methods: A retrospective review was performed of adult patients undergoing germline testing for at least nine “core” HHM genes associated with adult-onset myeloid malignancy as the primary malignancy risk (ANKRD26, CEBPA, DDX41, ETV6, GATA2, RTEL1, RUNX1, TERC, TERT), regardless of clinical indication for testing, from March 2015 to July 2026. Personal/family history of cancer and indications for genetic testing were determined by clinician-reported ICD-10 codes and keywords. Descriptive statistics were used to report the frequency of PVs in 124 genes, classified according to their estimated lifetime risk of hematologic malignancy as high (>50%), moderate (20-50%) or low (<20%). Multivariable logistic regression modeled the association of clinico-demographic characteristics on the odds of carrying a PV in a core HHM gene compared to having a negative or uncertain MGPT result.
Results: A total of 20,309 patients underwent testing for ≥ 9 HHM genes. Of these, 73% were female; 62% were White, and the mean (SD) age at testing was 53 (16). Most frequent test indications included: personal history of hematologic malignancy/disorder (23%) and/or solid tumor (38%), or family history of hematologic (28%) or other cancers (25%). Common ordering patterns included: 72% with >100 genes analyzed (mean [SD]: 170 [270]); 66% with acute leukemia/MDS panel; 54% with an additional solid tumor panel(s). Sample types included blood (57%), saliva (14%), buccal swab (18%), and gDNA (11%).
In total, 3,967 (20%) patients were positive for ≥ 1 PV in a HHM gene. However, 2,231 (11%) patients carried only a single PV in a gene associated with autosomal recessive inheritance and 118 (0.6%) had only a possibly mosaic result. 1,618 (8%) patients were identified to have at least one autosomal dominant or two autosomal recessive PVs in a gene associated with HHM risk, with 1686 PVs in 41 genes. Of these 1,618 patients, 451 (28%) were positive for a high- or moderate-risk HHM gene, most frequently DDX41 (208 patients), RTEL1 (47), POT1 (32), and RUNX1 (26), the majority (318/451, 71%) of which were a core HHM gene. Across the whole cohort, the most frequent genes with PV were CHEK2 (417 patients positive/19483 tested, 2.1%), BRCA1/2 (319/19538, 1.6%), DDX41 (208/20309, 1.0%) and ATM (183/20299, 0.9%). 10,428 (51%) patients had a variant of uncertain significance-only result.
Multivariable analysis showed increased odds of a PV in a core HHM gene was associated with older age at testing (odds ratio [OR]: 1.3, 95% confidence interval [CI] 1.2-1.4), male sex (OR: 2.0, 1.5-2.5), personal history of any HHM indication (OR: 2.1, 1.6-2.8), heme-only panel order (vs. additional solid tumor panel(s) ordered) (OR: 2.7, 1.5-5.0), and gDNA sample type (vs. blood) (OR: 1.9, 1.3-2.6). Conversely, decreased odds of a core HMM gene PV were associated with larger number of genes analyzed (OR: 0.92, 0.90-0.94), personal history of solid tumor (OR: 0.32, 0.24-0.44), and family history of hematologic malignancy or other cancer (OR: 0.70, 0.51-0.97, OR: 0.59, 0.43-0.79, respectively).
Conclusion: In this clinically diverse cohort of more than 20,000 patients undergoing MGPT, 2% had a PV in a high- or moderate-risk HHM gene and 8% had a PV in any HHM-associated gene. A personal history of hematologic malignancy/disorder, older age, male sex and gDNA sample increased the odds of identifying a PV in a core HHM gene, informing clinical implementation. Broad MGPT identified both HHM predisposition genes and genes associated with predominantly solid tumor risk (i.e. ATM, BRCA1/2, CHEK2). Limitations include limited clinical/histologic data, varying gene panel size, and ascertainment bias due to enriched personal/family history.
DISCLOSURES
Brittany Stewart, CGC, PhD: No relevant conflicts of interest to declare.