NRF2 pathway activation and SPP1⁺TREM2⁺ macrophages drive chemoradiotherapy resistance in esophageal squamous cell carcinoma

This study reveals that NFE2L2/KEAP1 mutations drive chemoradiotherapy resistance in esophageal squamous cell carcinoma by activating the NRF2 pathway, which synergizes with SPP1⁺TREM2⁺ macrophages to promote tumor relapse, thereby identifying these mechanisms as key targets for overcoming treatment failure.

The Gist

The Big Picture: The "Unbeatable" Escaped Prisoner

Imagine Esophageal Squamous Cell Carcinoma (ESCC) as a very aggressive prisoner. The standard way to catch this prisoner is Chemoradiotherapy (a combination of chemotherapy and radiation). Think of this treatment as a high-tech laser grid and a flood of water designed to wash the prisoner away.

For many patients, this works. But for others, the prisoner doesn't just survive; they come back stronger, hiding in the shadows and eventually taking over the whole city (the body) again. This is called relapse.

This paper asks a simple but crucial question: "Why do some prisoners escape the laser grid, and how do they come back?"

The researchers found that the escape isn't random. It's a coordinated team effort between two specific groups: a super-powered cell inside the tumor and a corrupt security guard outside it.

---

The Two Villains in the Story

1. The "Super-Cell" (The NRF2 Mutant)

Inside the tumor, there are cells that have a specific genetic glitch. In about 40% of patients who relapse, a gene called NFE2L2 (or its partner KEAP1) is broken.

• The Analogy: Imagine a normal cell has a "self-destruct button" that radiation pushes to kill it. The NRF2 pathway is the cell's internal alarm system. Usually, when the alarm goes off, the cell panics and dies.
• The Glitch: In these resistant cells, the alarm system is jammed in the "ON" position. It's like the cell has installed a forcefield. When radiation hits, the cell doesn't panic; instead, it instantly activates a "cleanup crew" (antioxidants) that neutralizes the damage before it can kill the cell.
• The Result: These cells are like super-soldiers that are immune to the standard laser grid. They survive the treatment and wait for the right moment to multiply.

2. The "Corrupt Security Guard" (The SPP1+TREM2+ Macrophage)

Tumors aren't just made of cancer cells; they are surrounded by a neighborhood of normal cells, immune cells, and fluids. This is the Tumor Microenvironment (TME).

• The Analogy: Think of the immune system (specifically T-cells) as the police force trying to arrest the cancer.
• The Betrayal: The "Super-Cells" described above have a secret weapon. They send out chemical signals (like a distress beacon) that recruit a specific type of immune cell called a Macrophage.
• The Corruption: Most macrophages are good guys (they eat bad cells). But these specific ones, labeled SPP1+TREM2+, are like corrupt security guards. Once recruited by the Super-Cells, they stop fighting the cancer. Instead, they:
  1. Build a physical wall around the cancer.
  2. Turn off the real police (the T-cells), making them tired and exhausted.
  3. Create a safe house where the cancer can hide and grow back.

The "Handshake" That Seals the Deal

The most exciting discovery in this paper is how these two villains work together.

• The Discovery: The researchers used high-tech "GPS" (spatial transcriptomics) to look at where these cells are located in the tumor.
• The Finding: In patients who relapsed, the Super-Cells and the Corrupt Guards were found standing right next to each other, holding hands. They were physically touching.
• The Mechanism: The Super-Cells send out specific chemical "keys" (ligands like CXCL14 and Laminins) that fit perfectly into the "locks" (receptors) on the Corrupt Guards. This handshake tells the guards: "Don't fight us. Protect us."

In patients who didn't relapse, this handshake never happened. The Super-Cells were either absent, or the guards didn't listen to them.

Why This Matters: A New Game Plan

For years, doctors have been trying to kill the cancer with stronger lasers (more radiation). But this paper suggests that's like trying to shoot a tank with a water pistol when the tank has a forcefield.

The paper proposes a new strategy:

1. Predict the Escape: Before starting treatment, doctors can take a small sample of the tumor and check for the "Super-Cell" glitch (NFE2L2/KEAP1 mutations). If the glitch is there, the patient is at high risk of the standard treatment failing.
2. Break the Alliance: Instead of just attacking the cancer, we need to break the team.
   • Step A: Use a drug to turn off the "forcefield" (NRF2 inhibitor) so the radiation can actually kill the Super-Cells.
   • Step B: Use a drug to fire the "Corrupt Guards" (antibodies against SPP1/TREM2) so the real police (T-cells) can do their job.

The Bottom Line

This study is like finding the blueprint for a prison break. It tells us that the reason some patients fail treatment isn't just bad luck; it's because their tumors have evolved a two-part defense system: a cell that can't be killed by radiation, and a bodyguard that stops the immune system from helping.

By identifying these players early, doctors can stop treating everyone the same way and instead use a customized "lockpick" to break the alliance, giving patients a much better chance of staying cancer-free.

Technical Summary

1. Problem Statement

Esophageal squamous cell carcinoma (ESCC) is a highly aggressive malignancy with a high rate of recurrence and poor response to standard chemoradiotherapy (CRT). While CRT is the primary treatment for locally advanced ESCC, a significant proportion of patients (51–64%) fail to achieve a pathological complete response, and many experience local or metastatic relapse within two years. The molecular mechanisms driving this resistance, particularly the interplay between tumor-intrinsic genetic alterations and the tumor microenvironment (TME), remain poorly understood. There is an urgent need to identify predictive biomarkers for patient stratification and novel therapeutic targets to overcome CRT resistance.

2. Methodology

The study employed a comprehensive, multi-omics approach integrated with a prospective clinical trial (NCT04694391) involving 114 ESCC patients. The research was structured into three cohorts:

• Discovery Cohort (n=90): Patients were stratified into Relapsed (Rel) (relapse within 2 years) and Non-Relapsed (NR) (no relapse for 3+ years).
  • Techniques: Whole-exome sequencing (WES) on 106 tumor samples and Whole-genome sequencing (WGS) on 49 samples to identify somatic mutations and copy number alterations (SCNAs).
  • Analysis: Assessment of tumor mutation burden (TMB), microsatellite instability (MSI), intra-tumoral heterogeneity (MATH score), mutational signatures, and clonal evolution.
• Mechanism Investigation Cohort (n=14): Focused on dissecting the TME and signaling pathways.
  • Techniques: Bulk RNA-seq (8 samples), Single-cell RNA sequencing (scRNA-seq) on 5 relapsed samples (integrated with a public dataset of 60 treatment-naïve samples), and 10× Visium Spatial Transcriptomics (ST) on matched pre- and post-relapse biopsies.
  • Analysis: Cell-type deconvolution, cell-cell communication inference (NicheNet, CellChat), and spatial co-localization analysis.
• Validation Cohort (n=10): Used to confirm findings.
  • Techniques: Multiplex Immunohistochemistry (mIHC) on pre- and post-treatment samples to validate protein expression and cellular proximity.
• Functional Validation:
  • In vitro: Luciferase reporter assays in ESCC cell lines (ECA109) to measure Antioxidant Response Element (ARE) activity.
  • Data Integration: Correlation of NRF2 pathway scores with radiation sensitivity data from 533 cancer cell lines and Reverse Phase Protein Array (RPPA) data from 19 ESCC cell lines.

3. Key Contributions

• Identification of a Primary Driver: Established NFE2L2 (NRF2) and KEAP1 somatic alterations as the dominant molecular drivers of CRT resistance in ESCC, occurring in nearly 40% of relapsed patients.
• TME Remodeling Mechanism: Uncovered a novel mechanism where NRF2-activated tumor cells physically recruit and interact with a specific immunosuppressive macrophage subset (SPP1⁺TREM2⁺).
• Spatial Resolution: Demonstrated that the co-localization of NRF2-activated epithelial cells and SPP1⁺TREM2⁺ macrophages is a unique feature of relapsed tumors, not observed in treatment-naïve or non-relapsed tissues.
• Clinical Translation: Proposed a dual-strategy therapeutic approach: targeting the NRF2 pathway directly and inhibiting the recruitment/function of SPP1⁺TREM2⁺ macrophages.

4. Key Results

A. Genetic Drivers of Resistance

• Heterogeneity: Pre-relapse (Pre-Rel) tumors exhibited significantly higher intra-tumoral heterogeneity (MATH score) and transversion mutation rates compared to non-relapsed (Pre-NR) tumors, though TMB and MSI were not significantly different.
• NFE2L2/KEAP1 Alterations:
  • SCNAs: Pre-Rel samples showed significant amplification of the 2q31.2 region containing NFE2L2.
  • Mutations: NFE2L2 mutations were found in 7/33 Pre-Rel samples but 0/29 Pre-NR samples. KEAP1 mutations were also enriched. Combined, NFE2L2/KEAP1 alterations were present in 40% of Pre-Rel patients vs. 3% of Pre-NR patients ($p = 7 \times 10^{-4}$).
  • Evolution: These mutations were often early clonal events that expanded under CRT pressure, leading to higher Variant Allele Frequencies (VAF) in post-relapse samples.
• Functional Impact: NFE2L2 mutations (specifically in the DEG/ETGE domains) led to constitutive NRF2 activation, evidenced by elevated ARE reporter activity and upregulation of antioxidant genes.

B. NRF2 Pathway Activation and Resistance

• Transcriptomic Signature: Bulk RNA-seq confirmed significant upregulation of the NRF2 pathway and its downstream targets (detoxification enzymes, antioxidant genes, metabolic regulators, and transporters) in Pre-Rel samples.
• Correlation with Survival: High NRF2 pathway scores in cancer cell lines correlated significantly with increased survival under radiation ($p = 5.7 \times 10^{-4}$). RPPA analysis confirmed higher protein levels of NRF2 targets in radioresistant cell lines.

C. TME Remodeling: The Macrophage Connection

• Cellular Composition: scRNA-seq revealed an enrichment of NRF2-activated epithelial cells (marked by AKR1B10) in relapsed tumors.
• Macrophage Subsets: Relapsed tumors showed a significant increase in SPP1⁺TREM2⁺ macrophages and SPP1⁺ macrophages, which exhibited the highest immunosuppressive scores among all myeloid subsets.
• Spatial Co-localization: Spatial transcriptomics and mIHC demonstrated that NRF2-activated epithelial cells exclusively co-localized with SPP1⁺TREM2⁺ macrophages in relapsed tissues. This interaction was mediated by ligand-receptor pairs (e.g., CXCL14/CCL26 from tumor cells binding to CXCR4/CCR1/CCR5 on macrophages; Laminins binding to Integrins).
• Immune Exhaustion: These SPP1⁺TREM2⁺ macrophages were identified as the primary signal senders driving the exhaustion of CD8⁺ T cells (specifically precursor exhausted and terminally exhausted states) via the ADGRG signaling pathway.

5. Significance and Clinical Implications

• Predictive Biomarker: The study proposes that pre-treatment sequencing for NFE2L2/KEAP1 mutations can serve as a robust biomarker to stratify patients likely to fail CRT. This is superior to gene expression signatures as it relies on established DNA sequencing workflows.
• Therapeutic Strategy: The findings suggest a rational combination therapy:
  1. NRF2 Inhibitors: To block the intrinsic antioxidant defense of tumor cells.
  2. Macrophage Targeting: Using antibodies against SPP1 or TREM2, or agents that disrupt the tumor-macrophage crosstalk, to reverse immunosuppression.
• Mechanistic Insight: The paper provides a unified model where tumor-intrinsic mutations (NRF2 activation) drive extrinsic TME remodeling (macrophage recruitment), creating a protective niche that fosters resistance and relapse.

In conclusion, this study bridges the gap between genomic alterations and the immune microenvironment in ESCC, offering a clear path toward personalized treatment strategies for chemoradiotherapy-resistant patients.