Page 114 - 2021_09-Haematologica-web

E. Oppliger Leibundgut et al.
Figure 4. Best molecular response of patients with additional mutations. Mutant allele burdens of each mutation before imetelstat treatment and at best response. Driver mutations are depicted in blue. Limit of detection at 2%, dashed line. BT: before treatment; BR: at best response.
cy is higher than mutation rates reported from other cohorts of ET patients,7,8,12 although still lower than the 86% and 98% overall rates of additional mutations detect- ed in patients with myelofibrosis.17,18 The high frequency of additional mutations in our ET cohort might reflect the concept of genetic instability in MPN and subsequent clonal evolution in a subset of these highly pretreated and partially resistant patients who had been diagnosed a median of 7.2 years previously. This concept is further supported by the finding that more than half of patients carried more than one additional mutation, as has also been reported by others.27 In addition, only patients with additional mutations at study entry acquired even more somatic mutations late during treatment (n=3).
The most frequently mutated gene was DNMT3A fol- lowed by TET2. DNMT3A mutations co-occurred with other somatic mutations in three of four patients, in line with published data.17 In ET, DNMT3A and TET2 muta- tions are often early events involved in disease initiation, which may precede the JAK2 V617F mutation and influ- ence the phenotype.9,28,29 Of the other genes mutated at
study entry, TP53, SF3B1, U2AF1 and EZH2 are part of a group of “adverse risk mutations” for ET, based on their significantly poor impact on overall, leukemia-free and myelofibrosis-free survival, and ASXL1 mutations are known as molecular risk factors for transformation to myelofibrosis.12,14
Patients with or without additional mutations had sim- ilar molecular responses to imetelstat with 63% of MMR in both groups; however, the presence of additional muta- tions had a negative effect on the depth of response, as mutant allele burden reductions were significantly deeper in patients without additional mutations. Of interest, all patients with additional mutations who gained a deep response (MMR) had JAK2 V617F driver mutations while patients with CALR or MPL driver mutations had poorer responses. In contrast, response depth in patients without additional mutations was not assigned to a specific driver mutation type. Further evidence from larger cohorts of patients is needed to support this observation.
The majority of cells with additional mutations were suppressed by imetelstat, and additional mutations
2402
haematologica | 2021; 106(9)