Fn-Associated ESCC Resistance Tied to SPP1+ Macrophages

Key Takeaways

• In esophageal squamous cell carcinoma treated with neoadjuvant chemotherapy plus PD-L1 blockade, Fn-positive disease was associated with poorer long-term survival, although some Fn-positive tumors still appeared treatment-sensitive.
• Nonresponder tumors showed immune exclusion rather than immune scarcity, with CD8-positive T cells concentrated in macrophage- and Fn-rich stromal regions despite similar overall immune-cell density.
• Macrophages from nonresponder tumors showed marked SPP1 overexpression, supporting a resistance program linked to cancer-associated fibroblast activation and intratumoral immune disorganization.
• The mannose-targeted NC-TNPM platform suppressed macrophage SPP1, reduced SPP1-positive macrophages, and was associated with deeper T-cell penetration and stronger tumor control when added to chemo-immunotherapy in preclinical models.
• Reported short-term safety testing found no evident tissue damage or liver function impairment, while longer-term translation questions remained unresolved.

Esophageal squamous cell carcinoma (ESCC) remains a setting in which neoadjuvant chemotherapy plus programmed death-ligand 1 (PD-L1) blockade can produce very different outcomes among tumors that otherwise look similar clinically. Fusobacterium nucleatum (Fn)-positive disease sharpens the question of whether resistance reflects where immune cells can reach inside the tumor rather than how many are present overall.

In the Theranostics study of NC-TNPM SPP1 editing in Fn-associated ESCC resistance, investigators integrated human ESCC specimens, in vitro macrophage infection and co-culture experiments, a mannose-modified non-cationic thiourea lipid nanoparticle carrying Cas9 mRNA plus secreted phosphoprotein 1 (SPP1)-targeting sgRNA, subcutaneous mouse work, and humanized patient-derived xenografts (PDX). Samples were stratified by Fn status and by response versus nonresponse to neoadjuvant chemo-immunotherapy, with attention to tumor-associated macrophages and cancer-associated fibroblasts (CAFs). Responder and nonresponder PDX mice received cisplatin, paclitaxel, and anti-PD-L1 on days 0, 7, and 14, and the humanized model added autologous CD4-positive and CD8-positive T cells after human CD45-positive chimerism exceeded 30%.

The clinical and mechanistic Fn-associated ESCC resistance findings linked Fn positivity during combination therapy to about 70% lower 5-year overall survival than Fn-negative disease, while still showing that some Fn-positive tumors remained sensitive and carried lower macrophage SPP1 expression. Responders and nonresponders had similar overall immune-cell density, but nonresponder tumors kept CD8-positive T cells in macrophage- and Fn-enriched stromal regions instead of tumor nests. Macrophages from nonresponder tumors also carried much higher Fn burden and markedly higher SPP1 expression, including 20-fold higher transcriptomic signal and 30.86-fold higher reverse-transcription quantitative polymerase chain reaction expression.

When investigators targeted macrophage SPP1 with NC-TNPM, flow cytometry showed 92% knockdown in F4/80-positive macrophages, and in vivo SPP1-positive macrophages fell by 85% to 93.5% without depleting total macrophages. T cells then redistributed deeper into tumor parenchyma, and tumor control improved when NC-TNPM was added to cisplatin, paclitaxel, and anti-PD-L1 in both subcutaneous and humanized PDX models, with reported T-cell activation effects varying by model and comparator. Biodistribution showed tumor accumulation but was not tumor-specific, with substantial liver and spleen uptake.

The spatial barrier model rested on colocalization and redistribution rather than direct proof of a continuous physical CAF wall, so the authors framed the data as evidence of a functional stromal barrier. Humanized PDX systems also capture only part of human antitumor immunity. Within those limits, the work positions SPP1-positive macrophage abundance together with spatial immune organization as candidate mechanistic markers beyond microbial status alone.

Investigators linked an Fn-SPP1-macrophage-CAF axis to immune exclusion in ESCC and reported that macrophage-targeted SPP1 editing reversed that phenotype and strengthened chemo-immunotherapy activity in preclinical models. The reported short-term safety readout showed no evident tissue damage or liver function impairment, but long-term safety, biodistribution specificity, off-target editing, and immunogenicity still need evaluation before clinical translation.

Clinician Questions

Which Fn-positive ESCC tumors did the reported resistance mechanism appear to apply to?

The mechanism did not fit all Fn-positive ESCC tumors. Some Fn-positive tumors remained sensitive to neoadjuvant chemo-immunotherapy and showed lower macrophage SPP1 expression, so microbial status alone did not explain response.

How did the study distinguish immune exclusion from simply having fewer immune cells in ESCC tumors?

The distinction was spatial rather than numeric. Responders and nonresponders had similar overall immune-cell density, but nonresponder tumors confined CD8-positive T cells to macrophage- and Fn-enriched stromal regions instead of allowing deeper entry into tumor nests.

What unanswered translational questions remain for NC-TNPM in Fn-associated ESCC?

Short-term histology and liver-function testing were reassuring, but long-term safety, biodistribution specificity, off-target editing, immunogenicity, and fuller validation in systems that better reflect human immunity remain unresolved before clinical translation.

Recommended Reading

• For more on immunochemotherapy in ESCC: Survival Improvements in Recurrent and Metastatic ESCC Cohort