Abstract

Adult T-cell leukemia/lymphoma (ATL) is a distinct subtype of peripheral T-cell neoplasms associated with human T-cell leukemia virus type-1 retrovirus. ATL includes a heterogeneous group of patients in terms of pathological and clinical features as well as prognosis, suggesting the presence of underlying molecular pathogenesis that could explain such heterogeneity among patients. Recently, we performed an integrated molecular analysis of a large number of ATL cases and delineated a comprehensive registry of gene mutations and other genetic/epigenetic lesions in ATL. In this study, we investigated possible correlations between these genetic/epigenetic lesions and clinical/pathological phenotypes in a large set of ATL patients, with a special focus on the impact of mutations and copy number alterations (CNAs) on clinical outcome.

We analyzed a total of 361 ATL samples, including acute (n = 192), lymphoma (n = 66), chronic (n = 89), and smoldering (n = 14) subtypes, for recurrent mutations and CNAs. Each subtype had characteristic genetic/epigenetic features, suggesting a distinct molecular pathogenesis therein. Aggressive (acute and lymphoma) subtypes were characterized by a higher number of mutations and CNAs including focal amplifications/deletions, hyperploid status, and CIMP phenotype, compared with indolent (chronic and smoldering) tumors. Two mutations (TP53 and IRF4) and eight focal deletions involving 1p13 (CD58), 6p21 (HLA-B), 9p21 (CDKN2A), 10p11 (CCDC7), 13q32 (GPR183), 16q23 (WWOX), 17p13 (TP53), and 19q13 (CEBPA), were more common in aggressive ATL than in indolent ATL. In contrast, showing a similar mutational distribution to those found in large granular lymphocytic leukemia, STAT3 mutations were characteristic of the indolent diseases. Gene set enrichment analysis of RNA-seq data showed a significant enrichment of MYC pathway and genes regulating cell cycle and DNA repair in upregulated genes in aggressive ATL.

Next, we assessed the impact of mutations and CNVs on prognosis among 215 ATL cases, for which survival data were available. In the entire cohort, mutation in CCR4 and IRF4, focal amplification in 9p24 (CD274) and 14q32 (BCL11B), and focal deletion in 9p21 (CDKN2A) were found to be significant predictors of poor overall survival, after adjustment for disease subtype and age. Multivariate analysis revealed that disease subtype (aggressive vs. indolent) was the most significant predictor of clinical outcome in ATL. Subsequent multivariate analysis according to disease subtype showed that within the patients with aggressive ATL, older age (≥ 70 years), CCR4 mutations, and 9p24 amplification were independently associated with an adverse outcome. Based on the number of the risk factors they owned, patients with aggressive ATL were classified into three categories showing marked difference in 3-year overall survival (OS) (P < 0.001): those with no risk factors (OS, 32%), with one risk factor (18%), and with two or more (0%). Among the patients with indolent ATL, we found IRF4 and TP53 mutations, 9p24 amplification, and deletions in 9p21 and 10p11 were independently associated with reduced survival. Interestingly, these alterations, except for 9p24 amplification, were also identified as genes more frequent in aggressive ATL. More importantly, based on these risk factors, the patients with indolent ATL can be classified into two categories showing very different prognostic profiles: patients with no risk factors (OS, 89%) and those with one or more risk factors (21%) (P < 0.001, HR = 16.8, 95% CI:5.4-52.5), suggesting that patients with indolent ATL having a genetic feature of the aggressive subtypes might genetically and biologically represent a distinct subset, which should be better managed as having an aggressive disease. Among these poor prognostic factors, 9p24 amplification and CCR4 mutation are especially interesting, because these lesions might be plausible targets of available agents, including anti-PD1/PD-L1 and anti-CCR4 antibodies.

In conclusion, based on the comprehensive genetic profiling, we demonstrated that the known subtypes of ATL can be further classified into genetically and biologically distinct subsets of tumors characterized by discrete sets of genetic lesions and substantially different prognosis. Our results suggest that molecular profiling can improve the prediction of prognosis in ATL patients and better guide therapy.

Disclosures

Tobinai:Gilead Sciences: Research Funding. Miyazaki:Shin-bio: Honoraria; Chugai: Honoraria, Research Funding; Sumitomo Dainippon: Honoraria; Celgene Japan: Honoraria; Kyowa-Kirin: Honoraria, Research Funding. Watanabe:Daiichi Sankyo Co., Ltd.: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.