Primary mediastinal B-cell lymphoma (PMBL) is a rare B-cell non-Hodgkin lymphoma (NHL) that presents with a large mediastinal mass, often in young women in their 20s and 30s. It has a distinct gene expression profile from diffuse large B-cell lymphoma (DLBCL) involving the mediastinum, and it is more genetically similar to nodular sclerosis Hodgkin lymphoma (HL). Similarities include variable CD30 and PDL1/PDL2 expression, the latter through the same amplification of chromosome 9p24.1 that has been described in HL. Since its recognition as a distinct entity with distinct histopathologic findings, prognosis following anthracycline-containing chemotherapy in PMBL is appreciably better than previous estimates, with five-year progression-free and overall survival rates of approximately 80 percent and 90 percent, respectively. Phase II data of dose-adjusted R-EPOCH (etoposide, doxorubicin, and cyclophosphamide with vincristine, prednisone, and rituximab) from the National Cancer Institute yielded even better results, with five-year event-free and overall survival rates of 93 percent and 97 percent, respectively. The role of radiotherapy for patients with bulky disease and a negative positron emission tomography scan at the end of treatment is uncertain. Despite these very good outcomes in a vast majority of patients, prognosis for patients with refractory or relapsed disease is quite poor. Response rate to salvage chemotherapy is only 20 to 30 percent, and two-year overall survival is only 15 percent. The anti-CD30 antibody drug conjugate brentuximab vedotin and the anti-PD1 antibody pembrolizumab have overall response rates of 15 percent (15% CR rate) and 33 percent (11% CR rate), respectively. These patients represent an unmet medical need.
Dr. Larry Mansouri and colleagues previously identified a recurrent 4–base pair truncating deletion in the NFKBIE gene that codes for IκBε in clinically aggressive chronic lymphocytic leukemia (CLL). IκBε is a negative feedback regulator of the NFκB pathway, and mutations are associated with decreased IκBε levels, decreased p65 inhibition, and increased phosphorylation and nuclear translocation of p65. They went on to screen nearly 1,500 lymphoid malignancies for mutations in NFKBIE and found them to be highly enriched in HL (in four of 11) and PMBL (in 46 of 203). In contrast, mutation frequency was less than 5 percent in follicular lymphoma, splenic marginal zone lymphoma, T-acute lymphoblastic leukemia, DLBCL, mantle cell lymphoma, primary central nervous system lymphoma, and small lymphocytic lymphoma. Among the 203 patients with PMBL, mutations were associated with chemoresistance (25% vs. 6%; p=0.022) and poor prognosis (5-year overall survival, 59% vs. 78%; p=0.034) compared with wild-type tumors. Whole-exome sequencing of 14 PMBL tumors (seven of which harbored the NFKBIE deletion) demonstrated overlapping mutations in other NFκB factors like TNFAIP3 in up to 50 percent of cases; mutations in STAT6, however, which occur in up to 36 percent of PMBL cases, were mutually exclusive with alterations in NFKBIE.
Finally, gene expression profiling was performed with NanoString PanCancer Pathways to evaluate both canonical cancer pathways, as well as 30-gene transcripts known to be important in PMBL and/or the NFκB pathway. Eight NFKBIE-deleted and 21 NFKBIE–wild-type cases of PMBL were analyzed. The investigators found differential expression of 79 genes, with upregulation of several NFκB target genes and upstream mediators including CD79B, EGFR, CD40, BCL2L1, and NFKBIB among the NFKBIE-deleted cases.
In Brief
These findings highlight the role of disregulated NFκB signaling in a subset of chemoresistant PMBL and offer potential new therapeutic targets for these patients with known poor outcomes. Although this study offers important insight into the pathobiology of a disease with a poor prognosis, there are several limitations to this series. First, patients were treated with variable chemotherapy regimens, with or without rituximab, and it is therefore not known if these results would apply to the most current regimens used clinically, such as R-EPOCH. However, up to one quarter of patients received dose-intensive regimens, and that did not seem to improve outcomes in the NFKBIE-deleted patients. Similarly, deletions in NFKBIE were not predictive of a poor response to radiotherapy; therefore, these mutations may not identify the most vulnerable PMBL patients refractory to both chemotherapy and radiation therapy. They may, however, help to identify a group of patients for whom adjuvant radiation therapy may be beneficial. Finally, the overlap of NFKBIE mutations and other mutations in the NFκB pathway raises questions as to whether such molecular abnormalities are markers of chemoresistance, and therefore a good therapeutic target. Nevertheless, the abundance of mutations in genes involved in the NFκB pathway suggests that drugs like the proteosome inhibitors or drugs that promote apoptosis, such as venetoclax, alone or in combination with chemotherapy or immunotherapy, may result in improved outcomes and survival. Certainly, further identification and development of novel NFκB inhibitors is a strategic approach worth pursuing in these patients.
Competing Interests
Dr. Jacobson indicated no relevant conflicts of interest.
