Frequency and Prognostic Impact of NUP98/NSD1 Translocations in Adult AML and MDS Patients

Abstract 1402

Over 20 different fusion partner genes of NUP98 have been reported in hematological malignancies with the majority of NUP98 fusions occurring in acute myeloid leukemia (AML). Recently, a specific fusion of NUP98 with Nuclear receptor-binding SET domain protein 1 (NSD1) has been analyzed in a large cohort of pediatric and adult AML patients. In this report, 16.1% of pediatric AML samples and 2.3% of adult cytogentically normal (CN)-AML cases were found to be NUP98/NSD1-positive. NUP98/NSD1-positive patients had an adverse outcome. The aim of our study was to investigate the frequency, clinical features and the prognostic impact of NUP98/NSD1 in another large, uniformily treated adult AML cohort, and in patients with myelodysplastic syndromes, which frequently precede overt leukemia.

We studied 504 younger AML patients (<60 years) and 193 MDS patients for the presence of NUP98/NSD1 translocations.

Analysis in the AML cohort was also performed for mutations in FLT3-ITD, NPM1, DNMT3A, IDH1, IDH2. Additional mutation analyses were performed in the subgroup of CN-AML for CEBPA, MLL-PTD, WT1 and WT1 SNP rs16754, NRAS. The NUP98/NSD1 fusion was identified by a nested RT-PCR using patient-derived cDNA. cDNA from a patient with proven NUP98-NSD1 fusion was used as a positive control. NUP98/NSD1 fusions were identified in 7 out of 504 younger adult AML patients (1.4%) while the NUP98/NSD1 fusion was not found in any of the 193 MDS patients. In the AML cohort, NUP98/NSD1 positive patients showed a higher number of peripheral blood blasts (P=.002), while other clinical characteristics such as FAB-subtype, type of AML, hemoglobin levels, white cell count or platelet count did not significantly differ between patients with or without the NUP98/NSD1 fusion. The majority of NUP98/NSD1 positive patients (5 out of 7) showed a normal karyotype while one patient was found to have a del(9) and another patient an inv(3)(q21q26) with monosomy 7. We identified an association between FLT3-ITD and NUP98/NSD1 fusion in our cohort (P=.007, 5 patients with NUP98/NSD1 also had a FLT3-ITD). This finding confirms the data by Hollink et al. and suggests a possible functional link between FLT3-ITD and the NUP98/NSD1 fusion in leukemogenesis. NUP98/NSD1 fusions were not associated with other mutations like those suggested in epigenetic regulation (DNMT3A, IDH1 and IDH2). In CN-AML, we also looked for an association between BAALC, ERG, EVI1, MN1, MLL5 and WT1 expression but did not find a significant difference between the expression of these genes and NUP98/NSD1 fusion genes. Due to the low frequency of the aberration outcome analysis was performed for explorative purposes. Considering the whole AML cohort we did not detect a significant difference for OS and for RFS between NUP98/NSD1 positive and negative patients (OS, HR 1.6; 95%CI 0.66–3.88; P=.3; RFS, HR 2.33; 95%CI 0.74–7.30; P=.147). However, NUP98/NSD1 positive patients had significantly lower complete remission (CR) rates (43 vs. 77 percent, P=.037). When considering only patients with CN-AML the NUP98/NSD1 positive patients (n=5) had no significantly different OS and RFS (OS, HR 2.08; 95%CI 0.77–5.64; P=.15; RFS, HR 2.88; 95%CI 0.71–11.71; P=.14). Again, a significantly lower CR rate was observed in the NUP98/NSD1 positive patients compared to NUP98/NSD1 negative patients with CN-AML (40 vs. 80 percent, P=.03). Because of the association between FLT3-ITD and NUP98/NSD1 we also investigated the prognostic impact in the subgroup of CN-AML patients also positive for FLT3-ITD. Again, OS and RFS did not differ between NUP98/NSD1 positive and negative patients (OS, HR 1.31; 95%CI 0.47–3.64; P=.61; RFS, HR 2.04; 95%CI 0.49–8.57; P=.33). For this subgroup a trend towards a lower CR rate was observed for NUP98/NSD1 positive FLT3-ITD positive CN-AML patients (40 vs 75 percent, P=.08).

In summary, our analysis confirms the presence but low incidence of the NUP98/NSD1 fusion gene in adult AML patients and the strong association with FLT3-ITD. The specific association of NUP98/NSD1 with FLT3-ITD mutations warrants clinical investigation with FLT3 inhibitors in these patients.