PB2061 THE PROGNOSTIC IMPACT OF NUCLEOLIN OVEREXPRESSION IN HEMATOLOGIC NEOPLASMS

Background: Nucleolin (NCL) is a multifunctional nucleolar DNA‐ and RNA‐binding protein involved in multiple intracellular processes ‐ including chromatin remodeling, DNA recombination and replication, rRNA synthesis and processing, and mRNA transcription and metabolism ‐, several of which are dysregulated in hematologic malignancies. It is overexpressed on the cytoplasm and cell surface of highly proliferative cells, including several solid tumors, such as gastric, colorectal, breast, lung and hepatocellular carcinomas, and gliomas and ependymomas. In these neoplasms, overexpression increases with histologic grade, metastatic potential and clinical stage, and is a biomolecular marker of event‐free survival (EFS), relapse‐free survival and overall survival (OS). NCL is also overexpressed in acute myeloid leukemia (AML) blast cells, and upregulates oncogenes and downregulates tumor suppressors in both acute and chronic leukemias. In a small heterogeneous set of AML patients (pts), higher levels of NCL were found to associate with decreased OS, a finding that was echoed in a small series of diffuse large B cell lymphoma (DLBCL) pts. The prognostic (Px) value of NCL overexpression in other hematologic neoplasms is unclear. Aims: To analyze the Px value of NCL overexpression in hematologic malignancies. Methods We analyzed NCL mRNA expression and survival data from publicly available datasets archived in the NIH Gene Expression Omnibus repository, in 1579 treatment‐naïve pts with lymphoma ‐ DLBCL (Series 1: 414 pts; Series 2: 498), mantle cell (MCL, 123 pts), and peripheral T cell (PTCL, 193 pts) ‐ and multiple myeloma (MM, 351 pts). Survival was compared at the third quartile (p75), the median (p50), and the first quartile (p25) of NCL expression; the ideal prognostic cut‐off (IPCO) was identified using the #R2 Genomics Analysis and Visualization Platform tool. Results: In Series 1 of DLBCL, the increased expression of NCL was associated with decreased OS, at p75 (39 vs 83 months, p = 0.033), p50 (64 vs 132 m, p = 0.02) and p25 (73 m vs median not reached (NR), p = 0.0066). The IPCO, at the 93 rd percentile (p93), was able to differentiate two cohorts with OS of 11 vs 90 months, p < 0.001. In Series 2, there were no differences in OS at the quartile cut‐offs up to p75; however, the IPCO at the p98 segregated two cohorts, with a decrease in OS (24 vs 82 m, p = 0.018) and EFS (17 m vs NR, p = 0.036) for NCL overexpressors. In MCL, the increased expression of NCL decreased OS at p75 (15 vs 39 m, p = 0.0033), p50 (19 vs 53 m, p = 0.0035) and p25 (25 vs 66 m, p = 0.042), with an IPCO at p90 (9 vs 39 m, p < 0.001). In PTCL, the IPCO at p8 identified a subgroup of pts (low‐expressors) with a particularly favorable Px, marginally approaching the level of significance (OS: 21 vs 177 m, p = 0.054); at higher expressions, the Px was indifferently unfavorable (25 vs 27 at p75; 21 vs 25 at p50; 18 vs 29 at p25; p = NS). In MM, the IPCO at p95 identified a subgroup of pts (high‐expressors) with the worst Px (38 m vs NR, p < 0.001); the median OS was NR for all cohorts at lower expression cut‐offs. Summary/Conclusion: In the four lymphoid‐lineage hematologic neoplasms considered in this study, the increased expression of NCL associated with worsening Px, with a decrease in OS and EFS. NCL was also able to identify subgroups of over‐expressors with markedly reduced OS in “good” Px malignancies, and of under‐expressors with a particularly favorable Px in “bad” Px malignancies. These findings are concordant with previously described data for non‐hematologic solid tumors, suggesting a comprehensive Px role for NCL in oncology.