Targeted Therapy in CMML: Complete Molecular Response to Sorafenib in a Patient with a FLT3-ITD Malignant Hematopoiesis

Chronic myelomonocytic leukemia (CMML) is a rare clonal hematopoietic stem cell disorder whom biology remains unclear.CMML is associated with many different somatic mutations in genes involved in key cellular processes including signaling (N/K-Ras, CBL, JAK2); differentiation (RUNX1, NPM1, CEBPa); epigenetic regulation (TET2, ASXL1, IDH1/2, EZH2, DNMT3A); and RNA splicing (SRSF2, U2AF1, SF3B1 and ZRSR2). FLT3 mutations are very rare but provide the rationale for FLT3 tyrosine kinase inhibitor use to treat this disease. We report the case of a 60-years-old patient diagnosed with a hepatocarcinoma metastasis which legitimized the introduction of anti-angiogenic therapy using the VEGF-R2 inhibitor, sorafenib.

The patient was addressed to the hematology department with a myeloproliferative-like CMML in transformation. We show here that the molecular analyses of this hematological disorders allow us to use sorafenib as a targeted therapy to inhibit the consequences of a FLT3-ITD mutation.

Cytogenetic analysis and genome wide array-based comparative genomic hybridization (aCGH) were performed at diagnosis. The reference standard used for aCGH was matched genomic constitutional DNA (CD3+ T cells sorted from a blood sample). Serum samples collected from the patient before or under treatment with sorafenib were assessed for their plasma inhibitory activity by western blotting analyses of signaling molecules downstream the FLT3-ITD mutation. Genomic DNA samples extracted from BMMCs and peripheral blood (PB) cells at diagnosis were screened for mutations in 18 classical genes. To monitor the FLT3-ITD mutation, the exon 15 of FLT3 was amplified by a specific PCR using a 6FAM-labeled forward primer.

The patient developed hyperleucocytosis (48.2 G/L) with neutrophilia (30.4 G/L), monocytosis (11.6 G/L) and basophilia (0.5 G/L) in January 2011. The BM was hypercellular with granulocytic and monocytic proliferation, dysgranulopoiesis and dysmegacaryopoiesis. Blast cells plus promonocytes accounted for 30% of the nucleated BM cells, leading to a diagnosis of AML secondary to CMML in the WHO classification

BM karyotype identified no clonal abnormalities and aCGH analysis of BMMCs produced normal findings. BM and PB cells were screened for mutations in 18 CMML-associated genes. Only two abnormalities were identified: a 27 base pair (bp) insertion FLT3-ITD mutation (exon 15) detected in BM cells with near complete disappearance of the wild type (WT) FLT3 allele (FLT3-ITD/FLT3-WT ratio at 9.62) and a classical heterozygous mutation (dupG) was found in the exon 12 of ASXL1. ASXL1 and FLT3-ITD mutations were not detected in purified CD3+ T lymphocytes.

Five months after sorafenib introduction, PB was strictly normal and BM examination demonstrated normal richness, blast cells and promonocytes accounting for 2% of the nucleated BM cells but persistent moderate dysgranulopoiesis and dysmegakaryopoiesis, indicative of complete remission. At this time, the FLT3-ITD/FLT3-WT ratio was 1.66 and 0.58 in the BM and PB, respectively. In January 2012, the WBC profile was still normal and a BM smear only showed moderate dysgranulopoiesis. On the molecular side, FLT3-ITD mutation was undetectable, indicative of complete molecular response. But ASXL1 mutation was evident at all time points.

The serum of the patient, obtained before and under sorafenib was tested on cell line harboring FLT3-ITD mutation. Constitutive FLT3 Y⁵⁹¹, Akt S⁴⁷³, STAT5 Y⁶⁹⁴ and ERK1/2 T²⁰²/Y²⁰⁴ phosphorylations were fully inhibited in the presence of the serum extracted under sorafenib treatment.

Our patient clearly had a myeloproliferative-CMML driven by an homozygous FLT3-ITD mutation. This is the first report of such a CMML patient achieving sustained CR and CMR after treatment with an FLT3-ITD tyrosine kinase inhibitor.

In this case, the ASXL1 mutation remained detectable upon sorafenib treatment after the suppression of FLT3-ITD-driven malignant hematopoiesis, suggesting that it arose from a FLT3-WT subclone that contributed to the CMML phenotype with some dysplastic feature.

In conclusion, we propose that mutations in the FLT3 gene should be examinated in all CMML cases, even their low frequency because FLT3 TKI may induce dramatic and sustained responses without significant toxicity and eventually allow for allogenic transplantation.

No relevant conflicts of interest to declare.