Loss of the Y chromosome (LOY) is one of the more common cytogenetic abnormalities in myelodysplastic syndromes (MDS); it has been associated with a lower risk prognosis within the IPSS-R.1,2 Additionally, LOY is a common age-related phenomenon in men. Cytogenetic abnormalities signify underlying clonal hematopoiesis (CH), and the finding of LOY within healthy individuals was one of the initial indications that CH occurred outside the context of malignancy in the 1970s.3 More recently, next-generation sequencing (NGS) has led to the recognition that CH is very common, increases in prevalence with age, and is a risk factor for both hematologic and nonhematologic diseases.4 Given the parallels between LOY and CH as determined by NGS, Dr. Madhu Ouseph and colleagues aimed to elucidate the relationship between these phenomena by describing the landscape of somatic mutations in patients with isolated LOY.
In this study, they identified bone marrow samples with documented isolated LOY, defined as three or more metaphases, from patients who also had NGS performed. All 73 patients included in the analysis had undergone evaluation for abnormal blood counts. Patients were categorized into four groups of LOY burden according to the number of metaphases with isolated LOY: less than 25 percent metaphases, 25 to 49 percent, 50 to 75 percent, and 75 percent or higher. Associations between LOY burden and age, diagnosis, and coincident pathogenic mutations in leukemia driver genes were explored. They performed multivariate analysis to determine whether LOY burden is an independent predictor of myeloid neoplasia and reported on trends of LOY burden over time in patients for whom they had sequential samples.
Consistent with prior studies, there was no association between LOY burden and age, affected lineage, or severity of cytopenias. However, patients with a higher LOY burden were more likely to have or develop myeloid neoplasms, particularly MDS. Furthermore, the LOY burden correlated with a higher likelihood of concurrent CH-associated mutations and a greater quantity of CH-associated mutations. The variant allele frequencies of those mutations also correlated with LOY burden. The most commonly mutated genes were TET2, SF3B1, U2AF1, ZRSR2, and ASXL1. In patients with cytopenias of unknown significance, the presence of CH-associated mutations, the number of mutations, and their variant allele frequencies have all previously been shown to predict the development of myeloid neoplasms.5 The researchers performed a multivariate analysis controlling for CH-associated mutations and found that LOY remained a significant, independent predictor of developing a myeloid neoplasm. Furthermore, patients with LOY were significantly more likely to bear CH-associated mutations than the age-matched general population. In fact, the prevalence of CH-associated mutations in patients with the highest quartile of LOY burden was comparable to that observed in patients with true MDS. Lastly, they analyzed patients with sequential samples and found that some patients who developed myeloid neoplasia developed increased LOY burden over time, though the sample size was too small for statistical analysis.
In Brief
In summary, the data here show that LOY burden is associated with the burden of CH-associated mutations and, after controlling for the presence of these mutations, remains an independent predictor of myeloid neoplasia. This study highlights the striking similarities between clonal hematopoiesis of indeterminate potential (CHIP), as defined by the presence of pathogenic somatic mutations and isolated LOY. Both entities are markers of CH, increase in prevalence with age, and are associated with an increased risk of developing myeloid neoplasms. Interestingly, LOY is also associated with a variety of nonhematologic disease states including cardiovascular disease and inferior survival,6 much like CHIP. This article contributes to the growing evidence that LOY is analogous to CHIP and should be included in the definitions of CHIP in clinical and research settings. LOY is detected in up to 20 percent of older men, a prevalence that is comparable to that of all other CHIP-associated mutations combined. Overlooking cytogenetic data could result in misclassifying a large subset of male patients who are at higher risk for developing MDS than their female counterparts. Furthermore, the clustering of LOY with CHIP-associated mutations suggests a shared role in leukemogenesis. Understanding how these various genetic lesions develop and interact with one another is critical to better understanding this process.
These findings are also clinically relevant. In a patient with cytopenias and bone marrow morphology that is nondiagnostic for myeloid neoplasia, how should the finding of isolated LOY be interpreted? Should it raise concern for evolving myeloid neoplasia or be disregarded as an incidental, age-related finding? Prior to the article by Dr. Ouseph and team, it might have been tempting to rely solely on the results of NGS testing, which have shown a high positive predictive value for patients developing myeloid neoplasia5 ; however, the article suggests that the burden of LOY can provide additional insight into whether cytopenias are related to a developing myeloid neoplasm. The possibility that longitudinal trends in LOY might be even more informative is provocative but should be assessed further. Overall, these findings reinforce the continued relevance of cytogenetic analysis, even in the age of NGS, to the diagnostic process and to prognostication at the bedside.
References
Competing Interests
Dr. Ambinder and Dr. DeZern indicated no relevant conflicts of interest.
