https://orcid.org/0000-0002-8171-4106
In this issue of Blood Advances, Stewart et al1 provide a superb report of 12 individuals with rare (gnomAD frequency, 1.0 × 10–5 to 7.0 × 10–5) germ line pathogenic/likely pathogenic (P/LP) variants at the 564th residue of JAK2 (p.R564Q and p.R564L) who developed hereditary thrombocytosis and/or hereditary hematopoietic malignancies (HHMs). The authors describe multiple pedigrees with highly penetrant thrombocytosis and a spectrum of hematopoietic malignancies.
The first pedigree includes a proband diagnosed as having essential thrombocythemia at age 29 years. Her father developed chronic lymphocytic leukemia, and her 2-year-old son was subsequently diagnosed as having hereditary thrombocythemia. Each patient carried the germ line JAK2 p.R564Q variant. The proband of a second pedigree was diagnosed as having thrombocytosis at age 50 years and essential thrombocythemia at 53 years. The proband’s 27-year-old son also had thrombocytosis. Both carried the germ line JAK2 p.R564L variant. The proband from a third pedigree was diagnosed as having thrombocytosis at age 36 years and essential thrombocythemia at 49 years. She also carried the germ line JAK2 p.R564L variant. Notably, her father, for whom no genetic data were available, was diagnosed as having non-Hodgkin lymphoma at 52 years.
The authors queried their institutional molecular database and identified 7 additional singletons with the germ line p.R564L variant. These individuals had both myeloid and lymphoid conditions: polycythemia vera (diagnosed in 3 singletons), essential thrombocythemia (1), chronic lymphocytic leukemia (1), acute myeloid leukemia (1), and monoclonal gammopathy of uncertain significance (1). These findings, paired with in silico modeling data, strongly implicate the p.R564L variant in aberrantly activated Janus kinase-signal transducer and activator of transcription (JAK-STAT) signaling, as occurs with the p.R564Q variant.2 Notably, in this series, most patients with hematopoietic malignancies lacked additional driver mutations, suggesting that second hits are not essential for disease progression in patients with germ line p.R564L/Q variants.
This publication from Stewart et al is timely and has significant clinical implications when taken in the context of other work from the past 6 months. Ceyhan-Birsoy et al3 recently showed that patients with myeloproliferative neoplasm (MPN) had the highest germ line diagnostic yield (26%) of any blood cancer type in a large cohort (n = 240) of relatively unselected patients. Notably, the next-generation sequencing (NGS) panel in Ceyhan-Birsoy’s study omitted JAK2, which immediately raises the question of whether their germ line diagnostic yield was underestimated. Universal germ line testing is generally recommended for tumors with a germ line yield of >10%, such as pancreatic cancer (15%), metastatic prostate cancer (12%-17%), and others.4 Therefore, the work from Ceyhan-Birsoy et al3 bolstered by Stewart et al begs the question: should universal germ line genetic testing be offered to all patients with MPNs? The answer seems to be a resounding “yes,” based on established standards in genetic medicine, cancer risk, and oncology.
The benefits of universal germ line genetic testing would extend to patients with MPNs and their family members. First, stem cell transplants continue to serve a crucial role for many patients with MPNs. During the initial diagnostic evaluation, universal germ line genetic testing would streamline donor searches by enabling transplant teams to determine which potential related donors are a genetic match. The definition of a genetic “match” would extend beyond traditional HLA typing by including HHM-related variants. This approach would avoid the limitations of an urgent genetic evaluation immediately before a transplant.5 Identifying patients with HHMs would also inform cascade genetic testing in their family members and the development of personalized solid cancer screening strategies for these individuals. Some patients with HHMs in Ceyhan-Birsoy’s study had P/LP variants in genes related to “pure” hematopoietic phenotypes, such as ETV6. However, other patients had variants for which personalized solid cancer screening approaches were available, including APC, ATM, BRCA1, BRCA2, CHEK2, MUTYH, and POT1.3 Knowledge of these variants would inform personalized solid tumor screening approaches for the proband and their family members.
A universal germ line genetic screening recommendation for patients with MPNs would also help avoid delays in care related to insurance denials. In our clinical experience, the absence of clear HHM testing guidelines leads to the inappropriate denial of genetic testing coverage, even for patients with clear HHM phenotypes. Current clinical guidelines contain significant heterogeneity regarding germ line genetic testing,6 which confuses treating teams and third-party payers. A universal germ line genetic testing approach for patients with MPNs would be an important first step in addressing this problem.
Technical deficiencies, however, will also need to be addressed as the field moves toward more comprehensive germ line genetic testing practices for patients with blood cancers. First, patients must receive high-quality sequencing of germ line DNA (from cultured skin fibroblasts, hair follicles, or nails) that analyzes the full spectrum of genes relevant to HHMs. The assay and bioinformatics pipelines must be capable of detecting HHM-related variants (ie, insertions/deletions, structural variants, and others). Unfortunately, many NGS assays currently advertised for diagnosing HHMs are broadly deficient in technical quality and have remained so despite an increased understanding of the genetic basis of many blood cancers.7,8 Notably, in a recent study, only 1 of 8 HHM assays (13%) sequenced JAK2, meaning most commercial assays would misdiagnose the patients in this series from Stewart et al. In addition to false negative results, these types of deficiencies place patients at risk of potentially catastrophic complications, most notably donor-derived malignancies, if a patient receives stem cells from a matched related donor who unknowingly carries the same HHM-related germ line variant. Some of this risk of false negatives may be mitigated by a broader shift from targeted NGS panels to exome- and genome-based sequencing approaches in the field. However, even less biased sequencing approaches must be paired with contemporary bioinformatics pipelines that are frequently updated to identify the full spectrum of variants responsible for HHMs.
In conclusion, Stewart et al should be commended for this latest addition to the rapidly evolving HHM literature. This work implicates additional germ line JAK2 variants as a risk factor for HHMs. Furthermore, it reiterates the need for refined contemporary approaches to accurately diagnosing these diseases in a timely fashion so that the care of patients with MPNs and HHMs can be optimized.
Conflict-of-interest disclosure: M.W.D. reports consulting or advisory roles (Argenx), honoraria for educational writing (American Society of Hematology Self-Assessment Program: Bone Marrow Failure chapter), and honoraria (Novartis). J.C. declares no competing financial interests.
