Honing in on the (epi)genetic basis of AITL

In this issue of Blood, Odejide et al report on recurring mutations in a large series of patients with angioimmunoblastic T-cell lymphoma (AITL) clustering around three epigenetic modifiers: TET2, DNMT3, and IDH2.¹ 

Peripheral T-cell lymphomas (PTCLs) are a heterogeneous group of lymphoid malignancies characterized by innate and early chemotherapy resistance and poor overall outcomes. Among the PTCLs, AITL was first described almost 40 years ago as a poorly defined proliferation of T cells in the setting of immune dysregulation, B-cell proliferation, and adenopathy. Although it is rare overall (2% of lymphomas), AITL is among the top 3 most common subtypes of PTCL, with geographic variation showing increased incidence in Western Europe compared with the rest of the world where it accounts for almost one-third of T-cell lymphomas.²  Neither the etiology nor the rationale for regional variation is known.

AITL is morphologically distinguished from other PTCLs by characteristic diffuse proliferation of small to medium-size neoplastic T cells set in a background of reactive polytypic cells, including normal CD4⁺ and CD8⁺ T cells, B lymphocytes, eosinophils, histiocytes, follicular dendritic cells, and immunoblasts. There is marked vascularity and increased high endothelial venules. Recently, the T-follicular helper (T_FH) cell was identified as the normal counterpart of AITL³  with the following immunophenotypic profile: CXCL13⁺, PD1⁺, ICOS⁺, and BCL6⁺, with occasional CD10 and CD30 positivity. The molecular signature of AITL is dominated by cells within the microenvironment,^3,4  and there are several immune suppressive signatures associated with poor outcomes. The clinical outcome is poor overall, with a median survival of approximately 2 years and no specific treatment approach vs other PTCLs. However, up to one-third of patients survive more than 7 years,⁵  and the biologic differences between long-term survivors and others is not understood.

In this issue of Blood, Odejide et al¹  evaluate the coding regions of 219 lymphoma-associated genes in 85 paraffin-embedded tissue samples of AITL in an effort to better characterize its genetic basis. The patients are reflective of most AITL series: advanced median age of 69 years, universally advanced stage, and dismal median survival of 18 months. By using targeted next-generation sequencing, the authors found 80 genes with mutations in the coding region, including 34 genes mutated in more than one patient. In particular, 65 patients (76%) had at least 1 TET2 mutation, and 43 of these patients had more than 1 TET2 mutation. Thirty-three percent of patients had a DNMT3A mutation, and 100% of them also harbored a concomitant TET2 mutation. Seventeen (20%) had IDH2 R172 mutations, and 15 of these patients had concurrent TET2 mutations; this is in contrast to myeloid disorders in which IDH2 and TET2 mutations are thought to be mutually exclusive. In addition to these 3 most common mutations, the authors report a number of gain-of-function and loss-of-function mutations in genes not previously known to be altered in AITL, including TNFSF9, ETV6, CCND3, and STAT3, among others. The significance of these lesser frequency findings is uncertain, particularly whether they are secondary events in subclones or whether they are independently important in the pathogenesis of the disease.

The observation that TET2 and other genes involved in DNA modification are mutated in such a high percentage of patients is noteworthy. TET2 is a member of the ten-eleven translocation (TET) family of genes and plays a crucial role in oxygenation of methylcytosine and DNA demethylation. Loss of TET2 supports its role as a tumor suppressor gene in both myeloid and lymphoid malignancies.⁶  Given the wide spectrum of affected methylated sites, TET2 proteins thus confer significant epigenetic control over transcription. Although it is more frequently associated with myeloid disorders, TET2 was first noted to be mutated in a handful of lymphoid malignancies, particularly T-cell lymphomas.⁷  The report by Quiveron et al found that 11.9% of 177 T-cell lymphoma samples and up to one-third of AITL samples had TET2 deletional or insertional mutations. More recently, another group found that TET2 mutations were present in 47% of AITL and 38% of PTCL-not otherwise specified patients but were distinctly absent in other PTCL histologies and seemed restricted to lymphomas with a T_FH phenotype.⁸  Similarly, IDH2 mutations seem restricted to AITL among lymphoid disorders and can be present in up to 20% of patients in some series.⁹  Others have shown that DNMT3 is mutated in up to 11% of T-cell lymphoma patients and is nearly always associated with a TET2 mutation.¹⁰ 

The contribution of the study by Odejide et al¹  is several-fold. First, it confirms and expands the findings of TET2, DNMT3A, and IDH2 mutations previously identified in AITL in a large international cohort, clearly establishing these mutations as recurrent and frequent abnormalities that distinguish AITL from other PTCLs. The authors remark that the spectrum of mutations is more similar to myeloid disorders and provocatively suggest that treatment should be altered accordingly. Second, the consistent finding of concurrent TET2 and DNMT3A mutations and TET2 and IDH2 mutations suggests that epigenetic regulation is an important motif in AITL pathogenesis. Considering the ever-increasing number of agents targeting epigenetics in hematopoietic malignancies, these findings have important therapeutic potential. Finally, the broad net that was cast to evaluate more than 200 other genes shows that there are a number of genes recurrently mutated in AITL, albeit at a lower frequency, that may be helpful in understanding pathogenesis.

In summary, the report by Odejide et al¹  firmly supports that AITL has a distinct genetic mutational landscape dominated by genes involved in epigenetic programming, and hopefully will pave the way for more targeted approaches for a disease badly in need of better treatment options.