The predictive value of mosaic chromosomal alterations for lymphoid malignancies varies by monoclonal B-cell lymphocytosis status

Abstract Background Mosaic chromosomal alterations (mCAs) are large-scale somatic gains, losses, or copy-neutral loss of heterozygosity (CNLOH) detectable by genotyping of peripheral blood DNA. They represent a form of clonal hematopoiesis and are observed in 0.5-5% of cancer-free individuals. The presence of any autosomal mCA has been associated with >3-fold risk of hematological malignancies. In particular, mCAs occurring in regions related to lymphoid malignancies have an 11-fold increased risk of lymphoid malignancies and a 68.6-fold risk for chronic lymphocytic leukemia (CLL). Monoclonal B-cell lymphocytosis (MBL), the precursor state to CLL, is defined by <5×10⁹/L clonal B cells in otherwise healthy individuals and is classified as low-count (<0.5x109/L) or high-count (0.5-5x10⁹/L) based on clonal B-cell count. MBL is also associated with increased risk of lymphoid malignancy (beyond CLL), with 4.3-fold and 74-fold risk in individuals with low- and high-count MBL, respectively. We previously reported a significant enrichment of autosomal mCAs in MBL, demonstrating a 1.7-fold association in those with low-count MBL and a 50.5-fold association in those with high-count MBL. Whether mCAs independently predict lymphoid malignancy risk after accounting for MBL status is unknown. Here, we evaluated the impact of mCAs on risk of incident lymphoid malignancies in individuals with MBL and in individuals screened negative for MBL. Methods CLL-type MBL (CD5 positive) was identified using flow cytometry assay on peripheral blood. Participants were age ≥40, had no prior hematologic malignancy, and were enrolled through the Mayo Clinic Biobank or the Division of Hematology at Mayo Clinic, Rochester. Participants were followed for incident lymphoid malignancy, including CLL, with diagnoses confirmed by medical record review. mCAs with cell fraction >1% were detected from array-based genotyping of whole blood DNA. Cox proportional hazards models, adjusted for age and sex, estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for risk of developing an incident lymphoid malignancy in individuals with autosomal mCA versus those without an autosomal mCA. Results In 4,331 participants with available MBL and mCA status, the median age was 67 years, and 45% were male. Flow cytometry identified 3,231 individuals who screened negative for MBL, 754 individuals with low-count MBL, and 346 individuals with high-count MBL. Autosomal mCAs were present in 6.5% of individuals without MBL, 7.6% with low-count MBL, and 71.1% with high-count MBL. Over a median follow-up of 6.9 years, 237 individuals developed an incident lymphoid malignancy. The most common types were CLL, multiple myeloma, and diffuse large B-cell lymphoma. Ignoring MBL status, the presence of an autosomal mCA was associated with increased risk of lymphoid malignancy (HR = 2.65; 95% CI: 1.45–4.86; P = 0.002), replicating previous results. However, after excluding individuals with MBL, this association attenuated; those with mCA alone did not have a significantly elevated risk (HR = 1.60; 95% CI: 0.57-4.52; P = 0.38) compared to individuals without MBL or mCA. We next examined the effect of mCAs among individuals with MBL. In those with low-count MBL, the most frequent mCA events were 13q loss (2.1%), 12 gain (1.9%), and 4q CNLOH (1.2%), and individuals with an mCA had a borderline significant risk of developing a lymphoid malignancy (HR = 2.78; 95% CI: 1.01–7.67; P = 0.049) compared to those without an mCA. Similarly, in those with high-count MBL, the most frequent mCA events were 13q loss/CNLOH (31.2%), 12 gain (13.0%), and 11q loss (4.9%), and the presence of an mCA conferred an increased risk compared to those without mCA (HR = 1.51; 95% CI: 1.09-2.08; P = 0.012). Conclusions Autosomal mCAs were more frequent in individuals with high-count MBL compared to those with low-count or no MBL. mCAs showed limited predictive value for lymphoid malignancy among individuals without MBL. In contrast, in individuals with low-count MBL, autosomal mCAs emerged as a potential biomarker for identifying those at risk of developing a lymphoid malignancy. In high-count MBL, mCAs were also a biomarker, but were driven by known chromosomal abnormalities identified through clinical testing. These findings suggest that the utility of mCAs as a predictive marker for lymphoid malignancy is contingent on the presence of MBL and may inform risk stratification and surveillance strategies in this population.