[Analysis of clinicopathological and molecular abnormalities of angioimmunoblastic T-cell lymphoma].

OBJECTIVE: To analyze the clinicopathological features, molecular changes and prognostic factors in angioimmunoblastic T-cell lymphoma (AITL). METHODS: Sixty-one cases AITL diagnosed by Department of Pathology of Peking University Cancer Hospital were collected with their clinical data. Morphologically, they were classified as typeâ [lymphoid tissue reactive hyperplasia (LRH) like]; typeâ ¡[marginal zone lymphoma(MZL)like] and type â ¢ [peripheral T-cell lymphoma, not specified (PTCL-NOS) like]. Immunohistochemical staining was used to evaluate the presence of follicular helper T-cell (TFH) phenotype, proliferation of extra germinal center (GC) follicular dendritic cells (FDCs), presence of Hodgkin and Reed-Sternberg (HRS)-like cells and large B transformation. The density of Epstein-Barr virus (EBV) + cells was counted with slides stained by Epstein-Barr virus encoded RNA (EBER) in situ hybridization on high power field (HPF). T-cell receptor / immunoglobulin gene (TCR/IG) clonality and targeted exome sequencing (TES) test were performed when necessary. SPSS 22.0 software was used for statistical analysis. RESULTS: Morphological subtype (%): 11.4% (7/61) cases were classified as type â ; 50.8% (31/61) as type â ¡; 37.8% (23/61) as type â ¢. 83.6% (51/61) cases showed classical TFH immunophenotype. With variable extra-GC FDC meshwork proliferation (median 20.0%); 23.0% (14/61) had HRS-like cells; 11.5% (7/61) with large B transformation. 42.6% (26/61) of cases with high counts of EBV. 57.9% (11/19) TCR(+)/IG(-), 26.3% (5/19) TCR(+)/IG(+), 10.5% (2/19) were TCR(－)/IG(－), and 5.3% (1/19) TCR(－)/IG(+). Mutation frequencies by TES were 66.7% (20/30) for RHOA, 23.3% (7/30) for IDH2 mutation, 80.0% (24/30) for TET2 mutation, and 33.3% (10/30) DNMT3A mutation. Integrated analysis divided into four groups: (1) IDH2 and RHOA co-mutation group (7 cases): 6 cases were type â ¡, 1 case was type â ¢; all with typical TFH phenotype; HRS-like cells and large B transformation were not found; (2) RHOA single mutation group (13 cases): 1 case was type â , 6 cases were type â ¡, 6 cases were type â ¢; 5 cases without typical TFH phenotype; 6 cases had HRS-like cells, and 2 cases with large B transformation. Atypically, 1 case showed TCR(－)/IG(－), 1 case with TCR(－)/IG(+), and 1 case with TCR(+)/IG(+); (3) TET2 and/or DNMT3A mutation alone group (7 cases): 3 cases were type â ¡, 4 cases were type â ¢, all cases were found with typical TFH phenotype; 2 cases had HRS-like cells, 2 cases with large B transformation, and atypically; (4) non-mutation group (3 cases), all were type â ¡, with typical TFH phenotype, with significant extra-GC FDC proliferation, without HRS-like cells and large B transformation. Atypically, 1 case was TCR(－)/IG(－). Univariate analysis confirmed that higher density of EBV positive cell was independent adverse prognostic factors for both overall survival (OS) and progression free survival(PFS), (P=0.017 and P=0.046). CONCLUSION: Pathological diagnoses of ALTL cases with HRS-like cells, large B transformation or type â are difficult. Although TCR/IG gene rearrangement test is helpful but still with limitation. TES involving RHOA, IDH2, TET2, DNMT3A can robustly assist in the differential diagnosis of those difficult cases. Higher density of EBV positive cells counts in tumor tissue might be an indicator for poor survival.