Triple-negative myelofibrosis makes up 10% to 15% of patients with myelofibrosis, but it is associated with higher rates of leukemic transformation and poorer survival. Investigators at the University of Michigan set out to better understand the disease and found that the clinical, cytogenetic, and molecular features of triple-negative myelofibrosis were heterogeneous.
Kristen M. Pettit, MD
Triple-negative myelofibrosis (MF) makes up 10% to 15% of patients with MF, but it is associated with higher rates of leukemic transformation and poorer survival. Investigators at the University of Michigan set out to better understand the disease and found that the clinical, cytogenetic, and molecular features of triple-negative MF were heterogeneous.
“These patients tend to present with high-risk clinical features,” said first author Kristen M. Pettit, MD, assistant professor at the University of Michigan, when presenting the findings at the 2018 SOHO Annual Meeting. “The mutation burden in triple-negative MF is low and includes genes commonly mutated in other myeloid malignancies. Select patients may exist with no clonal markers, and in those patients, inflammatory or immune pathways may contribute.”
MF is characterized by activation of the JAK/STAT signaling pathway driven by mutations in theJAK2,CALR, andMPLgenes. Patients with triple-negative MF do not have mutations in these driver genes.
Investigators enrolled patients in MI-ONCOSEQ, an integrative sequencing research program, to perform next-generation sequencing (NGS) on peripheral blood or bone marrow samples from patients with overt myelofibrosis treated at Michigan. The test was designed to identify nonsynonymous somatic point mutations in a panel of more than 1700 genes.
The assay included 24,774 deeply sequenced capture targets covering all or part of 1711 genes. These results include somatic insertions, deletions, and copy number aberrations, as well as gene fusions and outlier gene expression.
At baseline, investigators collected clinical data and followed those data longitudinally. Patients with clinical or pathologic suspicion of another fibrotic myeloid neoplasm or secondary marrow fibrosis were excluded.
Out of 97 patients enrolled, Pettit et al detected 7 cases (7.21%) of triple-negative MF. As of this presentation, 6 patients are still alive and none have experienced a leukemic transformation. One patient received allogeneic stem cell transplantation, but later died due to transplant-related complications.
Patients with triple-negative MF tended to be younger, with a median age of 59 years (range, 49-73), and 5 of 7 had a high DIPSS-Plus score. Five patients were women. Five patients had thrombocytopenia at presentation and 6 had anemia. Two patients had a history of autoimmune conditions. Five patients had low-level circulating blasts, and 2 have developed pulmonary hypertension.
Splenomegaly was present in 3 patients, absent in 3, and unevaluable in 1 due to a previous splenectomy.
Pettit said patients had hypercellular marrows with significant fibrosis, which was not surprising.
Three patients had normal cytogenetics; 1 had high-risk inv(9), i(17q); 2 had del(13q); and the karyotype was unknown in 1.
Even with this very broad sequencing platform, the median number of mutations detected per patient was 1 (range, 0-6). Recurring mutations includedASXL1,SRSF2,TP53, andU2AF1.
Investigators detected mutations in pathways involved in DNA damage repair (TP53in 1,ATMin 1, andRAD51in 1), epigenetic modifiers (ASXL1in 3,DNMT3Ain 1), and splicing factors (SRSF2in 2,U2AF1in 1).
Pettit et al also discovered the novel mutationsFAT3andADK, which has not been detected in cancer previously, and was considered to be a variant of unknown significance.
“FAT3is a tumor suppressor gene,” she said. “The particular mutation we found has been found to be mutated in many other cancers, but hasn't been reported in myelofibrosis to date.”
NGS could identify no mutations in 1 patient. That patient also had no cytogenetic or gene expression abnormalities.
"We couldn't find a clonal marker in that 1 patient," Pettit said. "Despite that, that patient had very characteristic marrow findings of primary myelofibrosis."
At a median duration of follow-up of 14 months (range, 3-62), Pettit said that it is difficult to characterize treatment response due to the short follow-up and the small cohort. Four patients have been treated with supportive care alone and have not received treatment for MF.
“Three of the patients have been treated with JAK inhibitors to date, and all 3 have had great clinical responses, as defined by significant reduction in spleen size or improvement in symptom burden,” Pettit said. “Those responses have been maintained for 15 months and counting in 1 patient, 10 months and counting in another, and 4 months up until the time of transplant in the last.”
Pettit said that investigators will next compare the NGS results and the expression data to other patients with overt myelofibrosis, as well as a separate cohort with prefibrotic disease and those with essential thrombocythemia and polycythemia vera.
“The pathophysiology and genomic drivers of triple-negative MF remain incompletely understood and warrant further study,” she added.
Pettit KM, Kandarpa M, Robinson D, et al. Genomic landscape and clinical features of triple-negative myelofibrosis.Clin Lymphoma Myeloma Leuk.2018;18(suppl; abstr MPN-285). doi: 10.1016/j.clml.2018.07.187.