Coordinate Regulation of NF-κB Subunit Expression In Pediatric Hodgkin Lymphoma Patients with Rapid Early Response to Therapy, but Not Slow Early Response to Therapy

Abstract 2680

The nuclear factor-κB (NF-κB) pathway plays an integral role in Hodgkin's lymphoma (HL), and enhances survival of Hodgkin Reed Sternberg (H/RS) cells. Previous studies have shown that both the classical and alternative NF-κB pathways are constitutively active in adult HL. However, it is not known which proteins in the NF-κB pathway are constitutively active in pediatric HL, or if NF-κB pathway activation and regulation are similar between patients who respond rapidly to treatment (RER) and patients who are slow early responders (SER).

We used a tissue microarray (TMA) to examine specimens from 102 pediatric HL patients (56 male, 78 Caucasian, median age 15y (range 1–21y), 85 nodular sclerosing subtype, 99 evaluable, 80 RER, 19 SER), and 10 lymphoid controls for NF-κB pathway subunit expression. All HL samples were intermediate-risk HL treated on the Children's Oncology Group protocol AHOD0031. RER was defined as a >60% shrinkage in tumor volume after two therapy cycles as assessed by CT; SER had <60% tumor shrinkage. After CD30 quality control analysis, the 308-spot microarray was examined by IHC for both classical and alternative NF-κB pathway proteins (Rel-A, phospho-RelA, Rel-B, c-Rel, IκB-a, IKK-α, IKK-β, IKK-γ (NEMO), NIK and the negative NF-κB regulator A20). Staining intensity (3-point scale), localization (nuclear vs. cytoplasmic), and pathway regulation were compared between RER and SER patients.

Cytoplasmic NF-κB-inducing kinase (NIK) was overexpressed in SER patients compared to RER patients (p=0.005, Wilcoxon rank sum test). As expected, increased cyto-NIK was associated with increased cyto-IKK-α (r= 0.87). However, increased cyto-NIK was not associated with increases in either cyto-NF-κB2 (r = 0.12), cyto-Rel-B (r = 0.138) or nuclear Rel-B (r= 0.084). Data suggests a deregulation in the alternative NF-κB pathway, specifically IKK-mediated proteasomal cleavage of NF-κB2 in SER patients.

In RER patients, there were several correlations between NF-κB pathway proteins (nuclear phospho-Rel-A with nuclear c-Rel [CC-0.51, p<0.01], cyto-NIK with cyto-Rel-B [CC-0.36, p<0.01], and cyto-Rel-A with cyto-IkB-α [CC-0.32, p<0.05]). In total, RER patients had 19 significant (p <.05 to <.01) nuclear NF-κB associations (CC 0.32 to 0.62) and 30 significant cytoplasmic NF-κB associations (CC 0.28–0.68). In contrast, SER patients had only 4 significant nuclear associations (CC 0.65–0.80) and 4 significant cytoplasmic associations (CC 0.67–0.87). Although there were a larger number of significant nuclear and cytoplasmic associations in RER patients, this data should be interpreted with caution due to the difference in sample size between the SER and RER patients. However, these data suggest that the degree NF-κB pathway coordination may correlate with rapid treatment response in pediatric HL.