Variant CURation: A Key Clinical Need Toward CURes in Patients With Inherited Predisposition

In early 2020, The Hematologist published an interview with Drs. Lucy Godley and Jorge Di Paola, two of the researchers leading this effort, which is funded by ASH. Please see the March/April issue for that additional context.

In the current era for patients with myeloid neoplasia (MN), the goal of complete biologic understanding to improve outcomes is most often attempted through use of extensive implementation of next-generation sequencing and microarrays. This advanced testing can demonstrate genes associated with predisposition to MN.^1,2  Owing to the potential impact on clinical care, germline predisposition to MN was incorporated into the World Health Organization classification in 2016, and diagnostic recommendations were added to the most recent National Comprehensive Cancer Network practice guidelines,^3,4  which encourage clinicians to consider this testing as part of the standard work-up for appropriate patients with MN. Despite the emphasis on inherited predisposition in recent years, there remains significant complexity and heterogeneity in the reporting of results, which can limit patient outcomes. Thus in 2018, ASH and the Clinical Genome Resource cosponsored the Myeloid Malignancy Variant Curation Expert Panel (MM-VCEP) to develop rules for classifying gene variants associated with germline predisposition to MN.

The current article, authored by leaders in the field of inherited predisposition (and members of MM-VCEP), provides education on variant curation to further guide best available data reporting and clinical care. Among all the possible genes to focus on, the group opted to start with germline mutations in RUNX1, which predispose to familial platelet disorder (FPD) with predisposition to acute myeloid leukemia (AML). If a clinician suspected this disorder in a patient (who might have thrombocytopenia, hematologic malignancy, or platelet dysfunction [Table]) and had a RUNX1 mutation in their testing, the provider might access ClinVar, the free and public repository that archives the relationships among human variations and phenotypes. FPD/AML and RUNX1 is an illustrative model in ClinVar as there are as many as 325 germline RUNX1 variants listed for this autosomal dominant disorder. The challenge for a clinician attempting to translate mutation reports for RUNX1 patients is that more than half of these variants are currently reported in ClinVar as “uncertain significance,” which is often less than helpful at the bedside.

In their review article, Dr. David Wu and colleagues demonstrate the application of rules developed specifically for the RUNX1 gene via six case presentations to guide clinicians through the classification of this gene within the proposed framework. Each case has some clinical data as well as discussions of the phenotypic criteria for FPD/AML and summaries of molecular and functional roles of RUNX1 variants. These include single-nucleotide variations and indels such as missense, nonsense, frameshift, and splice site variants, and copy number variations such as whole-gene and intragenic deletions. This work is useful for illustrating the disease phenotypes and mutational challenges. Commentary by the authors highlights the need for a definitive germline sample, continuation of family studies with pedigrees and segregation analyses, use of a broad testing approach to avoid an “over-filtered” result early on, and consistent application of interpretation rules. Additionally, they advocate for ClinVar data deposition using these rules to improve the accuracy and consistency of variant curation. This collaboration resulted in the initial set of modified criteria for germline RUNX1 variants with the goal of improving clinical outcomes for patients with FPD/AML.

Given that inherited predisposition to myeloid disorders collectively represents a significant proportion of hematologic malignancies with at least 5 percent of cases having a germline etiology,⁶  and with a prevalence up to 10 to 15 percent in certain patient cohorts,^7-9  efforts directed at improved variant curation are critical for patients in the present era. Timely diagnosis of an underlying genetic predisposition has broad implications for treatment, transplant considerations, long-term surveillance, and family counseling. Potential avenues for cure are not possible if a predisposition is not diagnosed correctly. It is important to acknowledge that the interpretation of genetic testing remains subjective and complex, and the publicly available data could also be misleading at times. Additionally, mutations may need to be reconsidered and clarified as additional data emerge in the field (i.e., mutations initially deemed to be pathogenic may need to be reconsidered and reclassified as nonpathogenic or vice versa). Dr. Wu and colleagues highlight the fact that despite implementation of recent guidelines for sequence variant interpretation,⁵  the number of variants deposited in ClinVar classified as having uncertain significance (or being discordantly classified among clinical laboratories) can create clinical uncertainty or potentially adversely impact patient care. MM-VCEP will inform, educate, and advance the field through collaborative efforts. This group plans to set curation rules for other germline mutations such as GATA2, ANKRD26, ETV6, CEBPA, and DDX41 in the next few years, on the path to curing as many inherited disorders as possible.