Analysis of Mutational Landscape in Systemic Anaplastic Large Cell Lymphoma Identifies Novel Prognostic Markers

BACKGROUND: Systemic Anaplastic Large Cell Lymphoma (sALCL) is comprised of two distinct T-cell non-Hodgkin lymphoma entities: ALK-positive (ALK+) ALCL and ALK-negative (ALK-) ALCL. These entities are characterized by either the presence of absence of an ALK-translocation. It has been reported that ALK+ ALCL has a better prognosis compared to ALK- ALCL, with a 5-year overall survival (OS) of 70-80% versus 15-45%, respectively. Furthermore, more than 25% of ALK+ ALCL patients undergo relapse. sALCL is a genetically heterogeneous disease whose genomic characterization has been improved through the implementation of high-throughput technologies. Despite this, the prognostic value of somatic mutations has been poorly described. Here we present the analysis of genetic aberrations of sALCL, shedding light on disease pathogenesis, novel diagnostic biomarkers and prognostic markers for the detection of refractory and/or relapsed patients which could hold clinical implications.

METHODS: Formalin-fixed paraffin-embedded (FFPE) and/or fresh frozen tissue and related clinical information for 82 sALCL patients (47 ALK+ and 35 ALK-) were obtained after written informed consent from 5 centres across Europe. Using deep targeted DNA sequencing, the entire coding region of 275 selected genes (QIAseq Targeted DNA - Human Comprehensive Cancer Panel, Qiagen, Germany) was investigated in our retrospective cohort of patient' samples as well as 6 ALCL cell lines (Karpas-299, SU-DHL-1, DEL, SR786, FEPD and MAC2a). The average depth achieved across all the sequenced samples was approximately 2000x allowing the detection of aberrations with low allele frequencies (<10%) not usually detectable by other methods (e.g. Sanger sequencing or Whole Exome Sequencing). Moreover, we explored the clonal expansion in sALCL relapsed patients comparing the mutational status in matched samples at both diagnosis and relapse (n=4).

RESULTS: ALK- patients in our cohort have a worse prognosis than ALK+, whereby the 7y-OS was 45.1% and 73.2%, respectively, although the 7 year progression free survival (PFS) was comparable at 57.1% for ALK+ and 42.4% for ALK- patients. As previously reported, ALK+ patients were significantly younger in our series than ALK- patients with an average age of 23.5 and 55.2 years, respectively. Among the 275 genes sequenced, 148 (53.8%) genes harbor at least one mutation throughout the entire patient and cell line cohort; almost 1/3 of mutated genes were shared between ALK- and ALK+ ALCL patients. On average 4.2 mutations per patient were detected in ALK- ALCL, a higher level than observed in ALK+ samples (2.7), most likely due to the oncogenic-driving role of the ALK-translocation in this group. Seventy-two out of 82 (88%) patients carried at least one single mutation within the genes analysed in our panel. The most recurrent genes mutated were TP53 in both sALCL categories (16% of sALCL patients), followed by LRP1B being prevalently mutated in ALK+ disease (15% of sALCL patients). As expected, STAT3 and JAK1 were mutated solely in the ALK- ALCL cohort and represent the most commonly mutated genes in this group. Prognostically, the most recurrent genes mutated in patients with a dismal outcome (dead and/or relapsed patients) were TP53, STAT3, EPHA5, JAK1, LRP1B, KMT2D, PRDM1 and SOCS1. Comparing samples at diagnosis versus relapse highlighted two key findings: firstly, the clonal expansion of TP53 mutated clones in relapsed ALCL patients and secondly, the acquired mutations in EPHA5 which were only detected at relapse. These data suggest their possible use as biomarkers associated with clonal evolution.

CONCLUSION: This study provides information regarding the genetic landscape in sALCL across 275 select genes. We have confirmed the importance of known frequently mutated genes (STAT3 and JAK1) and describe novel recurrently mutated genes (LRP1B, EPHA5 and KMT2D). Mutations in STAT3 were more recurrent in ALK- ALCL patients with shorter OS, and clones harboring mutated TP53 and EPHA5 were detectable more often in relapsed sALCL thereby suggesting a possible driving role.