Fig 1.
CBL mutants are associated with proliferative features, increased BM blast percentage, leukocytosis and splenomegaly.
(A-F) Clinical characteristics of PREACH-M cohort at baseline, stratified based on the detection of CBL mutation. (G) MD-, MP-CMML classification based on WCCα (H) CMML-0, -1, -2 classifications based on BM blast percentageβ. 2016 WHO Classification: αBased on WCC: MD-CMML WCC<13×109/L, MP-CMML WCC>13×109/L. β Based on BM blast %: CMML-0 PB <2%, BM <5%; CMML-1 PB 2–4%, BM 5–9%, CMML-2 PB>5%, BM 10–19%. Bars represent mean ± standard error of mean. Mann-Whitney test used to determine statistical significance, where P<0.05 was statistically significant. [BM bone marrow; WCC white cell count; CRP C-reactive protein; MD-CMML myelodysplastic-CMML; MP-CMML myeloproliferative-CMML, wt wildtype, mut mutant].
Table 1.
Clinical characteristics, complete blood examination and bone marrow analyses of CMML patients in the PREACH-M trial stratified as RAS pathway (KRAS, NRAS, PTPN11, CBL) mutant vs. wildtype.
Table 2.
Clinical characteristics, complete blood examination and bone marrow analyses of CMML patients in the PREACH-M trial stratified as CBL mutant vs. RAS pathway wildtype.
Fig 2.
CBL mutants frequently co-occur with TET2 mutants and are associated with a complex subclonal architecture.
(A) Oncoplot for the PREACH-M cohort (n = 24). Mutation groups are shown in rows with each individual patient represented by a column. The presence of a mutation is indicated by the red or blue colored bars. Age category of the patients indicated by the black and grey bars and sex of patients by the green and gold bars. (B) Number of CBL mutant cases where TET2 mutations (n = 9) and other RAS pathway mutations (n = 5) were detected, where variation in the VAF of CBL vs. TET2 or RAS pathway mutant clones were ≤10% (dark blue) or >10% (light blue) (C) Number of cases where more than one variant of CBL, NRAS, KRAS or PTPN11 mutation was detected. (D) Details of CBL variants detected in each patient with CBL mutation. [VAF variant allele frequency].
Fig 3.
CMML have an increased percentage of CD116 and CD131 positive CD34+ stem and progenitor cells.
(A) Flow cytometry analysis of a representative CMML sample and healthy control stained for CD45, CD34, CD14 and CD16, and gating strategy used to define CD45+ mononuclear cells, CD34+ stem and progenitor cells and CD14+ monocytes. (B) Percentage of CD34+ progenitors and CD14+ monocytes in CMML samples (n = 4) vs. healthy control (n = 2). (C) Illustration of the cluster of differentiation (CD) markers where CD114 is a marker for G-CSFR, CD116 GMRα and CD131 βc. G-CSFR is homodimeric while GM-CSFR is heterodimeric receptor consisting of GMRα and βc. The expression of CD114, CD115, CD116 and CD131 in CMML samples (n = 4–6) vs. control (n = 2–3) in CD45+, CD34+ and CD14+ subpopulations, expressed as percentage of positively stained cells (D) and MFI (E) compared to control (cord blood or peripheral blood mononuclear cells from healthy donors). Bars represent mean ± standard deviation in (B). Box and whiskers graphs were plotted with min and max in (C) and (D). Unpaired Student’s t-test between CMML vs. healthy control used to determine statistical significance, where P<0.05 was statistically significant. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. [MFI mean fluorescence intensity; G-CSFR granulocyte-colony stimulating factor receptor; GMRα granulocyte-macrophage colony stimulating factor receptor subunit α; βc beta common subunit].
Fig 4.
CBL mutation hotspots in CMML cluster in the RING domain, unlike in JMML where they more commonly occur within the LHR.
(A) Table of CBL variants detected in our PREACH-M cohort combined with data sourced from COSMIC. Variants include nonsense or missense substitutions, frameshift and in-frame insertions or deletions within the coding sequence of CBL, filtered for all hematopoietic and lymphoid malignancies including CMML and JMML (n = 549), and CMML only (n = 137) or JMML only (n = 46) (B) Contingency analysis of CBL mutation hotspots within the LHR and RING domain of Cbl in CMML and JMML (C) Heat map representation of all sites within the LHR (amino acid residues 353–380) and RING domain (amino acid residues 381–435) where mutations have been reported. Numbers within the figure and on the scale depict counts (D) Tertiary protein structure of native wildtype Cbl (PDB ID 2Y1M) in inactive, closed conformation. The TKBD is colored beige, LHR blue and RING domain red. Amino acid residues of the top 6 mutation hotspots are indicated in inset; Tyrosine 371 (Y371), Leucine 380 (L380), Cysteine 384 (C384), Cysteine 396 (C396), Cysteine 404 (C404) and Arginine 420 (R420). (E) X-ray structures of wildtype Cbl in unphosphorylated, inactive state and in closed conformation (PDB ID 2Y1M), wildtype Cbl in Y371 phosphorylated, active state and in open conformation (PDB ID 4A4C), mutant Cbl Y371E (PDB ID 5HKX) and mutant Cbl Y371F (PDB ID 5J3X). The TKBD is colored beige, LHR of wildtype blue, LHR of mutant cyan, RING domain of wildtype red, RING domain of mutant pink. RMSD values between various Cbl conformations are shown in table. Statistical analysis was performed using two-sided Fisher’s exact test, where P<0.05 was statistically significant. [TKBD tyrosine kinase binding domain; LHR linker helix region; RING RING domain; H+L all hematopoietic and lymphoid malignancies; WT wildtype; RMSD root mean square deviation (distanced-based measure of protein structure similarity)].