Gene expression signatures delineate biological and prognostic subgroups in peripheral T-cell lymphoma

Abstract

Peripheral T-cell lymphoma (PTCL) encompasses a heterogeneous group of neoplasms with generally poor clinical outcome. Currently 50% of PTCL cases are not classifiable: PTCL-not otherwise specified (NOS). Gene-expression profiles on 372 PTCL cases were analyzed and robust molecular classifiers and oncogenic pathways that reflect the pathobiology of tumor cells and their microenvironment were identified for major PTCL-entities, including 114 angioimmunoblastic T-cell lymphoma (AITL), 31 anaplastic lymphoma kinase (ALK)-positive and 48 ALK-negative anaplastic large cell lymphoma, 14 adult T-cell leukemia/lymphoma and 44 extranodal NK/T-cell lymphoma that were further separated into NK-cell and gdT-cell lymphomas. Thirty-seven percent of morphologically diagnosed PTCL-NOS cases were reclassified into other specific subtypes by molecular signatures. Reexamination, immunohistochemistry, and IDH2 mutation analysis in reclassified cases supported the validity of the reclassification. Two major molecular subgroups can be identified in the remaining PTCL-NOS cases characterized by high expression of either GATA3 (33%; 40/121) or TBX21 (49%; 59/121). The GATA3 subgroup was significantly associated with poor overall survival (P = .01). High expression of cytotoxic gene-signature within the TBX21 subgroup also showed poor clinical outcome (P = .05). In AITL, high expression of several signatures associated with the tumor microenvironment was significantly associated with outcome. A combined prognostic score was predictive of survival in an independent cohort (P = .004).

Figures

Figure 1

Molecular diagnostic signatures of PTCL subgroups. (A) Unique gene expression signatures were identified for major PTCL entities using compound covariate prediction model (see Materials and methods for details), and the predictor score from top ranking genes for each subtype was used to classify a PTCL patient. ALCL and ENKTL groups are further differentiated into ALK(+)ALCL and ALK(–)ALCL, and NK and γδ T-cell subgroups, respectively. Each column represents a PTCL patient and each row represents a unique gene of the classifier. The relative gene expression scale is indicated below. (B) Pathological vs molecular diagnosis comparison. Substantial number of cases from PTCL-NOS were molecularly classified into WHO recognized PTCL subgroups: (i) AITL (n = 21, 14%); (ii) ALK(–)ALCL (n = 17, 11%); (iii) ATLL (n = 4, 3%); (iv) γδ-PTCL (n = 13, 9%). However, 26 AITL cases (22%) were not molecularly classifiable and changed to PTCL-NOS.

Figure 2

ALK(–)ALCL is molecularly distinct from PTCL-NOS and ALK(+)ALCL. (A) A unique gene expression signature for ALCL was identified and then ALK(+)ALCL cases were separated from ALK(–)ALCL (reclassified ALCL cases are indicated at the top). (B) Representative gene signatures/pathway enriched in ALK(+)ALCL compared with ALK(–)ALCL. (C) A representative re-classified ALCL case (hematoxylin and eosin staining [H&E]) with a pathological diagnosis of PTCL-NOS showing expression of CD30, GZMB, and TIA, but no expression of ALK by immunostaining.

Figure 3

Two major molecular subgroups within PTCL-NOS with biological and overall survival differences. (A) Bayesian predictor for the GATA3 and TBX21 subgroups was derived using cases from hierarchical cluster 1 and cluster 2. LOOCV was used for classification precision. (B) Overall survival (OS) analysis of molecularly-defined GATA3 and TBX21 subgroups showed significant difference in clinical outcome (P = .01). A subset of cases mentioned in the text lacked clinical outcome data and are not included in OS analysis. (C) Representative cases in the TBX21 subgroup (H&E) with TBX21 immunostain showed positivity ranging from 40% to 80% of cells, whereas positivity in the GATA3 subgroup was <10%. (D) Representative cases in GATA3 subgroup immunostained with GATA3 showing positivity ranging from 50% to 80% of cells, and were negative for TBX21.

Figure 3

Two major molecular subgroups within PTCL-NOS with biological and overall survival differences. (A) Bayesian predictor for the GATA3 and TBX21 subgroups was derived using cases from hierarchical cluster 1 and cluster 2. LOOCV was used for classification precision. (B) Overall survival (OS) analysis of molecularly-defined GATA3 and TBX21 subgroups showed significant difference in clinical outcome (P = .01). A subset of cases mentioned in the text lacked clinical outcome data and are not included in OS analysis. (C) Representative cases in the TBX21 subgroup (H&E) with TBX21 immunostain showed positivity ranging from 40% to 80% of cells, whereas positivity in the GATA3 subgroup was <10%. (D) Representative cases in GATA3 subgroup immunostained with GATA3 showing positivity ranging from 50% to 80% of cells, and were negative for TBX21.

Figure 4

Survival prediction in AITL. (A) Prognostic model in a training set and validated on an independent validation set. The optimal model represents that cases with high B-cell signature and low monocytic signature show favorable outcome compared with other cases (P = .004). (B) AITL cases are arranged according to model score, which represents differences between averages expression of B-cell signature and monocytic/P53 signature (mean). (C) Representative case from the upper (4th) quartile of the model score showing fewer CD20+ cells, whereas in (D) cases in lower(2nd) quartile showed a higher number of CD20+ B cells.