Genomics of lymphoblastic lymphoma

NOTCH1 and FBXW7 variants (which also induce NOTCH pathway signaling) are common in T-ALL.[50] In T-cell lymphoblastic lymphoma, NOTCH1 variants are observed in approximately 60% to 65% of cases, and FBXW7 variants are observed in approximately 15% to 25% of cases.[51-54] T-cell lymphoblastic lymphomas with NOTCH1 gene fusions, which have gene expression signatures that are different from cases with NOTCH1 gene variants, are discussed below.

CDKN2A at chromosome 9p21 is commonly altered in both T-ALL and in T-cell lymphoblastic lymphoma, with approximately three-fourths of each showing deletions of this gene locus.[50,54]

Loss of heterozygosity at chromosome 6q is observed in approximately 15% of T-ALL cases.[54]

PTEN variants are observed in approximately 15% of T-ALL cases and in a comparable percentage of T-cell lymphoblastic lymphoma cases.[50,53,54]

KMT2D variants are observed in approximately 10% of T-cell lymphoblastic lymphoma cases.[54] Other genes associated with epigenetics that are altered in T-ALL include PHF6 and KMT2C.

All 12 patients with TRB::NOTCH1 fusions were older than 10 years.

Patients with TRB::NOTCH1 gene fusions rarely had additional variants in NOTCH1. However, patients without this fusion commonly had NOTCH1 variants (about 60%).

The cumulative incidence of relapse was 67% in patients with TRB::NOTCH1 fusions, compared with less than 20% in patients with T-cell lymphoblastic lymphoma who did not have the fusion.

Only one of six patients with a fusion was younger than 10 years. The ages of patients ranged from 8 to 17 years.

Five of six patients with NOTCH1 fusions experienced an event. Four patients had disease relapse during therapy, and one patient developed a therapy-related AML.

CCL17 (TARC) levels, which are commonly increased at diagnosis for patients with Hodgkin lymphoma, were markedly elevated in all patients with T-cell lymphoblastic lymphoma with NOTCH1 gene fusions, but they were not elevated in patients without NOTCH1 gene fusions. CCL17 (TARC) levels decreased when remission was achieved and then increased again at disease relapse.

Clonal immunoglobulin and T-cell receptor gene rearrangements were detected in 89% and 79%, respectively, of the B-cell lymphoblastic lymphoma cases. Most clonal rearrangements were unproductive or nonfunctional, reflecting an early stage in B-cell development, which is consistent with the model that B-cell lymphoblastic lymphoma and B-ALL share the same cell of origin.

The variant landscape and focal deletions of B-cell lymphoblastic lymphoma show great overlap with those of B-ALL. The most common variants and deletions involved in B-cell lymphoblastic lymphoma were CDKN2A or CDKN2B (21%), NRAS (13%), IKZF1 (12%), and KMT2D (12%). RAS pathway variants were equally represented between B-cell lymphoblastic lymphoma and B-ALL, while variants in genes controlling B-cell development and cell cycle control were more common in B-ALL. Genes encoding epigenetic regulators (e.g., KMT2D, EP300, ARID1A, and ATF7IP) were more frequently altered in B-cell lymphoblastic lymphoma.

High hypodiploidy was seen in 29% of B-cell lymphoblastic lymphoma cases (similar to B-ALL), while the ETV6::RUNX1 gene fusion was detected in 13% of B-cell lymphoblastic lymphoma cases, a frequency somewhat lower than that reported for B-ALL (25%).

B-ALL high-risk groups (intrachromosomal amplification of the RUNX1 gene [iAMP21], ABL-class fusions, Philadelphia chromosome-like, KMT2A-rearranged/like, near haploid, and low haploid) were detected in 24% of B-cell lymphoblastic lymphoma cases. There was no association between stage and risk group. While the cumulative incidence of relapse was greater for patients in the high-risk group than for those in the non-high–risk group, the difference did not reach statistical significance.