NPRL2 Expression Associated with Cytotoxic Immune Function Independent of Estimated Immune Cell Abundance in TCGA-KIRC

Document Type

Conference Proceeding

Publication Date

9-2026

Publication Title

Oncologist

Abstract

Background: NPRL2, a GATOR1 complex tumor suppressor on chromosome 3p21.3, regulates mTOR signaling and is frequently lost in ccRCC, yet its association with antitumor cytotoxic immunity in ccRCC remains unexplored. We evaluated whether tumor NPRL2 expression in TCGA-KIRC associates with cytotoxic immune function and whether these associations persist after adjustment for estimated immune-cell abundance. Methods: Bulk RNA-seq data (log2(TPM + 1)) from 537 TCGAKIRC primary tumors were analyzed using transcriptomic immune module scores. A CD8 abundance score was defined from CD8A and CD8B. Cytotoxic effector program activity was quantified using a 5-gene signature (GZMB, PRF1, GNLY, NKG7, KLRD1) representing a shared granule-mediated cytotoxic effector program. Cytotoxic efficiency was defined as the difference between the z-scored cytotoxic effector score and the z-scored CD8 abundance score. Associations were tested using Spearman correlation. To evaluate whether findings extended beyond estimated immune-cell abundance, abundance-adjusted cytotoxic efficiency was examined after further adjustment for xCell and MCP-counter CD8 and NK abundance covariates. Sensitivity analyses were performed using the published Rooney cytolytic signature (GZMA, PRF1) and an expanded 6-gene cytotoxic signature including GZMA. Exploratory analyses examined CD8- anchored (CD8A, CD8B, GZMB, PRF1) and NK-anchored (NKG7, GNLY, KLRD1, FCGR3A) cytotoxic modules separately. P-values are unadjusted given the exploratory nature of the analyses. Results: Higher NPRL2 expression was associated with greater cytotoxic effector program activity (rho = 0.251, p < 0.001) and greater cytotoxic efficiency (rho = 0.232, p < 0.001). NPRL2 showed little or no association with estimated CD8 or NK abundance by xCell or MCP-counter, except for a modest positive association with MCP-counter NK cells (rho = 0.180, p < 0.001). After adjustment for abundance-related covariates, NPRL2 remained positively associated with abundance-adjusted cytotoxic efficiency (rho = 0.238, p < 0.001). Findings were directionally consistent in sensitivity analyses using the Rooney cytolytic signature (GZMA, PRF1; abundance-adjusted efficiency rho = 0.125, p = 0.004) and an expanded 6-gene cytotoxic signature including GZMA (abundance-adjusted efficiency rho = 0.228, p < 0.001). Exploratory lineage-anchored analyses suggested contributions from both CD8-anchored and NK-anchored cytotoxic programs, with a numerically stronger association for NK-anchored activity (rho = 0.193, p < 0.001) than CD8- anchored activity (rho = 0.097, p = 0.024). Conclusions: In TCGA-KIRC, higher NPRL2 expression is associated with stronger cytotoxic immune function after adjustment for estimated immune-cell abundance. Sensitivity analyses support the association across alternative cytotoxic signatures and abundance-adjusted formulations, while exploratory lineage-anchored analyses suggest a mixed cytotoxic effector program with both CD8-anchored and NK-anchored components. These findings are hypothesis-generating and warrant mechanistic study in systems with cell-type resolution.

Volume

31

Issue

Suppl 3

First Page

S34

Comments

2026 Kidney Cancer Research Summit, July 23-24, 2026, Boston, MA

Last Page

S35

DOI

10.1093/oncolo/oyag312.040

Share

COinS