Abstract LB-062: XPO1 is a rational target for double and triple-hit aggressive B-cell lymphomas

• Published in: Cancer research (Chicago, Ill.) Ill.), 2015-08, Vol.75 (15_Supplement), p.LB-062-LB-062
• Main Authors: Marullo, Rossella, Yang, ShaoNing, Rashal, Tami, Landesman, Yosef, Carlson, Robert, Shacham, Sharon, Cerchietti, Leandro C.
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2015-LB-062

Mutation and constitutive expression of MYC and BCL2 and/or BCL6 (a.k.a. double and triple-hit lymphomas) defines subsets of diffuse large B-cell lymphoma (DLBCL) pts with particularly poor outcome. Almost 60% of pts with BCL2 and MYC translocations die within six months of diagnosis due to chemorefractory disease, a prognosis that cannot be overcome with intense chemotherapy. A further hindrance to patient survival is that these double and triple-hit lymphomas are frequently found in the elderly who have limited tolerability to aggressive chemotherapeutic regimens. To identify chemoresistant double and triple-hit DLBCL, we screened 38 DLBCL cell lines for mutations in BCL2, MYC and BCL6 by FISH and response to chemotherapy administration in vitro. We found 3 triple-hit and 2 double-hit DLBCL cell lines. Previous work in our laboratory showed that the nuclear export protein, XPO1, is a critical regulator of MYC, BCL2 and BCL6 mRNA transport in DLBCL. We therefore analyzed XPO1 amplification (by PCR) and expression (by immunoblot) in these cell lines and found that all 5 cells expressed XPO1 at higher levels than centroblasts and that, in at least 2 of them, this was result of gene amplification. Inhibition of XPO1 with selinexor, a Selective Inhibitor or Nuclear Export (SINE) compound, increased nuclear localization of MYC, BCL2 and BCL6 transcripts and consequently decreased their protein expression. Moreover, selinexor administration induced chemosensitization in doxorubicin-resistant DLBCL cells through decreasing DNA damage repair mechanisms as demonstrated by comet assays. To determine the potential extent of DLBCL patients that could benefit from such a treatment, we first analyzed 110 DLBCL patient samples by immunohistochemistry and found that XPO1 expression was increased in 85 cases compared to normal tonsils. Within the cohort analyzed, 6 patients had double or triple hit lymphoma (by FISH) and all overexpressed XPO1. Since this population has higher incidence of chemorefractory disease, we decided to develop a pre-clinical model of this disease using primary patient samples. We developed two patient-derived xenograft (PDX) models representing a double and a triple-hit lymphoma. In both cases XPO1 was also amplified. We compared the molecular and pathological characteristics of 12 generations of PDXs with the original pt sample, and found that protein expression and mutations of these genes were stable. We therefore administered selinexor alone,