SAMHD1 expression contributes to doxorubicin resistance and predicts survival outcomes in diffuse large B-cell lymphoma patients

Abstract

Diffuse large B-cell lymphoma (DLBCL) is a commonly diagnosed, aggressive non-Hodgkin's lymphoma. While R-CHOP chemoimmunotherapy is potentially curative, about 40% of DLBCL patients will fail, highlighting the need to identify biomarkers to optimize management. SAMHD1 has a dNTPase-independent role in promoting resection to facilitate DNA double-strand break (DSB) repair by homologous recombination. We evaluated the relationship of SAMHD1 levels with sensitivity to DSB-sensitizing agents in DLBCL cells and the association of SAMHD1 expression with clinical outcomes in 79 DLBCL patients treated with definitive therapy and an independent cohort dataset of 234 DLBCL patients. Low SAMHD1 expression, Vpx-mediated, or siRNA-mediated degradation/depletion in DLBCL cells was associated with greater sensitivity to doxorubicin and PARP inhibitors. On Kaplan-Meier log-rank survival analysis, low SAMHD1 expression was associated with improved overall survival (OS), which on subset analysis remained significant only in patients with advanced stage (III-IV) and moderate to high risk (2-5 International Prognostic Index (IPI)). The association of low SAMHD1 expression with improved OS remained significant on multivariate analysis independent of other adverse factors, including IPI, and was validated in an independent cohort. Our findings suggest that SAMHD1 expression mediates doxorubicin resistance and may be an important prognostic biomarker in advanced, higher-risk DLBCL patients.

Graphical Abstract

Figure 1.

Low SAMHD1 expression is associated with greater sensitivity to doxorubicin and Veliparib in DLBCL. (A) DLBCL cell lines RCK-8, RIVA, SUDHL-4 and SUDHL-6 were transfected with siSAMHD1, lysed, run on SDS-PAGE, and analyzed for SAMHD1 expression by western blot analysis. (B, C) Indicated cells were seeded and treated with varying doses of Veliparib, and viability was analyzed via PI/Annexin V assay (B) or AlamarBlue assay (C). (D, E) Indicated cells were seeded and incubated with varying concentrations of doxorubicin, and viability was analyzed via PI/Annexin V assay (D) or AlamarBlue assay (E). (F) RCK8 and SUDHL-6 were fixed, PI stained, and cell cycle was analyzed by flow cytometry. (G) RCK-8 and SUDHL-6 were lysed, dNTP extracted, and RT-based primer extension assay was analyzed for dNTP pool size. For B, C, D, E and G, the mean and standard deviation from three replicas are shown. ** P < 0.01, *** P < 0.001.

Figure 2.

SAMHD1 degradation with VLPs with Vpx causes doxorubicin sensitization. (A) Schematic of VLP preparation and viability assay. (B) Western blot analysis showing Vpx-mediated degradation of SAMHD1 in RCK-8 and SUDHL-6 cells. (C) RCK-8 and SUDHL-6 were cultured in fresh media at sub-confluency and transduced with VLPs with or without Vpx. 48 h post-transduction, cells were fixed, stained with PI, and analyzed for cell cycle profile by flow cytometry. (D) 48 h post-transduction with VLPs with Vpx (+Vpx) or without Vpx (-Vpx), RCK-8 and SUDHL-6 cells were seeded in 96-well plates, incubated with indicated concentrations of doxorubicin for 48 hr, and viability was determined by assessing metabolization of AlamarBlue. (E) Western blot analysis showing siRNA-mediated SAMHD1 depletion in RCK-8 and SUDHL-6 cells. (F) Evaluation of cell viability using PI/Annexin V staining following siRNA-mediated SAMHD1 knockdown following treatment with varying concentrations of Doxorubicin for 48 h. (G) SUDHL-4 cells were transduced with VLPs without Vpx (-Vpx) or with Vpx (+Vpx) and incubated for 24 h and treated with 250 nM doxorubicin for 48 h. Then, cells were harvested, lysed, resolved on SDS gel, and probed for indicated proteins. For D and E, mean and standard deviation obtained from three replicas is shown. ** P < 0.01 and * P < 0.05

Figure 3.

Low SAMHD1 expression is associated with improved overall survival in patients with DLBCL. (A) Representative immunohistochemical analysis for SAMHD1 expression in tissue from patients with DLBCL as used for scoring. (B) Kaplan–Meier plot showing patient outcome association with SAMHD1 expression level. Overall survival (OS) was defined as the time from diagnosis to death or the last contact when living patients were censored at the last contact. OS was estimated using the Kaplan–Meier method and was compared using log-rank tests. (C–F) The benefit of SAMHD1 expression level on OS of patients was assessed in a subgroup of patients with different IPI scores (C, D) and stages (E, F). (G) Analysis of an independent cohort of DLBCL patients from GDC indicates that low SAMHD1 expression is associated with improved OS.