Page 137 - Haematologica-April 2018

Histone modifier gene mutations in PTCL-NOS
mutations, such as mutations of histone acetylation genes, were mutually exclusive of each other, suggesting that his- tone modifying genes might be involved in distinct biolog- ical processes (Figure 1D).
Alterations of histone modifier genes were primarily located at well-conserved amino acid positions across dis- tinct species (Online Supplementary Figure S1). Typically, KMT2D/KMT2A mutations affected the PHD domain (e.g. residues 134-320, 1378-1556, 5032-5138, and 1433- 1624, 1871-1983), HMG domain (residues 2021-2072), undetermined domain (e.g. residues 2487-4658) and SET domain (e.g. residues 5397-5519 and 3825-3969). EP300/CREBBP mutations affected the HAT domain (e.g. residues 1306-1612 and 1342-1649).
Histone modifier gene mutations were associated with disease progression in peripheral T-cell lymphoma not otherwise specified
One hundred and forty patients were treated with CHOP-based chemotherapy in a historical cohort of Shanghai Ruijin Hospital from 1997 to 2011, and referred to as the training cohort. The validation cohort consisted of 99 patients enrolled in two prospective studies (NCT 01746992 and NCT 02533700, randomized trials to com- pare CHOP-based chemotherapy with sequential chemotherapy with CEOP/IVE/GDP or CTOP/ITE/MTX). Since 2012, 49 and 50 patients have been randomized to CHOP-based or sequential chemotherapy, respectively. No obvious differences in clinical and pathological characteristics or treatment
A
response were observed either between the training and the validation cohort, or between the two arms within the validation cohort (Online Supplementary Table S2). Gene mutation data were available for 73 and 52 patients of the training and validation cohorts with available tissue sam- ples, respectively (Figure 1A).
In the training cohort, the median follow-up time was 29.1 months (range, 0.5-162.0 months). The 2-year pro- gression-free and overall survival rates of the patients were 36.7% and 47.1%, respectively. In the univariate analysis, the International Prognostic Index was a signifi- cant prognostic factor for both progression-free survival and overall survival (both P<0.001), but histone modifier mutations were only prognostic for progression-free sur- vival and not overall survival (P=0.012 and P=0.095, respectively) (Figure 2A,B). In the multivariate analysis, when the International Prognostic Index was controlled for, the presence of a histone modifier gene mutation was an independent prognostic factor for progression-free sur- vival (P<0.001) (Table 1). The 2-year progression-free and overall survival rates were 26.4% and 56.6% for patients with histone modifier gene mutations and 49.6% and 63.3% for patients without mutations (Figure 2A,B). In the validation cohort, the median follow-up time was 19.5 months (range, 2.1-43.0 months). Histone modifier gene mutations were associated with shorter progression-free survival in multivariate analysis (P=0.049) (Table 1). The 2- year progression-free and overall survival rates were 22.2% and 24.2% for patients with histone modifier gene mutations and 41.1% and 57.5% for patients without
BCD
Figure 1. Histone modifier gene mutations in peripheral T-cell lymphoma, not otherwise specified. (A) Gene mutations identified by targeted sequencing in 125 patients with peripheral T-cell lymphomas. The number of patients (N) with mutations is listed on the right. The mutations are classified into the categories indicated on the left: I, histone methylation; II, histone acetylation; III, DNA methylation; IV, chromatin remodeler. (B) Number and type of non-silent somatic mutations. (C) Number and percentage of non-silent somatic single nucleotide variants. (D) Circos diagram according to mutation categories.
haematologica | 2018; 103(4)
681