Tumoral and stromal hMENA isoforms impact tertiary lymphoid structure localization in lung cancer and predict immune checkpoint blockade response in patients with cancer.

BACKGROUND: Tertiary Lymphoid Structures (TLS) correlate with positive outcomes in patients with NSCLC and the efficacy of immune checkpoint blockade (ICB) in cancer. The actin regulatory protein hMENA undergoes tissue-specific splicing, producing the epithelial hMENA11a linked to favorable prognosis in early NSCLC, and the mesenchymal hMENAΔv6 found in invasive cancer cells and pro-tumoral cancer-associated fibroblasts (CAFs). This study investigates how hMENA isoforms in tumor cells and CAFs relate to TLS presence, localization and impact on patient outcomes and ICB response. METHODS: Methods involved RNA-SEQ on NSCLC cells with depleted hMENA isoforms. A retrospective observational study assessed tissues from surgically treated N0 patients with NSCLC, using immunohistochemistry for tumoral and stromal hMENA isoforms, fibronectin, and TLS presence. ICB-treated patient tumors were analyzed using Nanostring nCounter and GeoMx spatial transcriptomics. Multiparametric flow cytometry characterized B cells and tissue-resident memory T cells (TRM). Survival and ICB response were estimated in the cohort and validated using bioinformatics pipelines in different datasets. FINDINGS: Findings indicate that hMENA11a in NSCLC cells upregulates the TLS regulator LTßR, decreases fibronectin, and favors CXCL13 production by TRM. Conversely, hMENAΔv6 in CAFs inhibits LTßR-related NF-kB pathway, reduces CXCL13 secretion, and promotes fibronectin production. These patterns are validated in N0 NSCLC tumors, where hMENA11ahigh expression, CAF hMENAΔv6low, and stromal fibronectinlow are associated with intratumoral TLS, linked to memory B cells and predictive of longer survival. The hMENA isoform pattern, fibronectin, and LTßR expression broadly predict ICB response in tumors where TLS indicates an anti-tumor immune response. INTERPRETATION: This study uncovers hMENA alternative splicing as an unexplored contributor to TLS-related Tumor Immune Microenvironment (TIME) and a promising biomarker for clinical outcomes and likely ICB responsiveness in N0 patients with NSCLC. FUNDING: This work is supported by AIRC (IG 19822), ACC (RCR-2019-23669120), CAL.HUB.RIA Ministero Salute PNRR-POS T4, "Ricerca Corrente" granted by the Italian Ministry of Health.

Through the looking glass: Distinguishing neural correlates of relational and non-relational self-reference and person representation.

BACKGROUND: Neuroimaging studies have found a substantial overlap between self-related and social cognitive processes. This study examines three different ways of conceptualizing a person - one that requires considering how they are embedded in their social environment (roles), one that requires considering their generalized qualities (traits), and one that identifies their relevant group memberships. To the extent that relational aspects of identity require considering how a person is embedded in their social environment we should find greater activation for role judgements in regions associated with social cognitive processes. METHODS: During fMRI scanning, 38 participants made stimulus judgments about themselves and a close other regarding the target's traits, social roles, and group memberships in a 2 (target of judgment) x 3 (stimulus category) within-participant design. RESULTS: Relatively greater activation in areas broadly associated with theory of mind and mentalizing (e.g., PCC, TPJ) was found for social role, compared to trait judgments. By contrast, trait judgments, compared to role judgments, activated regions associated with semantic memory (e.g., IFG). Conjunction analyses showed that activations associated with roles overlapped with regions associated with a meta-analytic map of mentalization, judgments made about others, and stimuli reflecting higher social specialization, indicating that roles may require considering how a person is socially embedded. Judgments about group memberships were associated with brain regions found for both trait and role judgments. CONCLUSION: Our findings provide evidence for a distinction between two modes of social reference - one that is relatively more associated with social relational processing (roles) and that is relatively more dependent on semantic memory processes (traits). Given the substantial overlap between the pattern of results for roles (relative to traits) and other (relative to self), it may be the case that at least part of our representation of ourselves and others may fundamentally require representing people as embedded in social networks.