> 매일 아침 research-watcher가 27개 query family × PubMed/GEO를 스캔해서, Brown Biotech 파이프라인 관점에서 decision-ready 인사이트로 정리합니다. 매일 06:00 KST 자동 생성. **Category focus:** Longevity & Senolytics **Published:** 2026-09-12 06:05 KST **Entry ID:** 97 **Tags:** #gse237951 #gse268195 #ezh2 #EZH2-inhibitor #tazemetostat #EPZ-6438 #EPZ011989 #GSK126 #CPI-1205 #CPI-0209 #valemetostat #DS-3201 --- ## 🔬 Today's Top Findings ### 1. [Bioinformatics & Longevity + Refractory Cancer + Ferroptosis + Hepatocellular Carcinoma + Epigenetics] Inhibition of EZH2 restores normal expression of genes associated with cysteine metabolism and ferroptosis in hepatocellular carcinoma [RNA-seq + CUT&Tag paired atlas] (GSE237951 + GSE268195, score 6/6, n=8/12, Homo sapiens, pdat 2024/10/25 = ~14 months ago, raw TXT+BIGWIG files, the FRESHEST NOVEL EZH2-inhibitor / cysteine-metabolism / ferroptosis-restoration / hepatocellular-carcinoma paired RNA-seq + CUT&Tag atlas in the entire 2026-09-12 hits.jsonl and the FIRST paired-RNA-seq + CUT&Tag EZH2-ferroptosis-HCC atlas in the prior id:92-id:96 retention window \u2014 a direct longevity + refractory-cancer + ferroptosis + epigenetics + cysteine-metabolism input for the BB longevity + AI-DD + clinical lane with explicit overlap to the ferroptosis-lane and the HCC-refractory-cancer lane and the lipid-peroxidation-longevity-lane) ### 2. [Longevity & iPSC + Organoid + Regenerative-Medicine + Metabolic-Liver-Disease + Cell-Therapy] Novel therapy for metabolic liver disease by transplantation of human iPS cell-derived 3D-organoids (GSE263559, score 6, n=7, Homo sapiens, pdat 2026/04/01 = ~5 months ago, raw TXT files, the FRESHEST NOVEL iPS-cell-derived-3D-organoid / metabolic-liver-disease / regenerative-cell-therapy transplantation atlas in the entire 2026-09-12 hits.jsonl and the FIRST iPSC-organoid-transplantation metabolic-liver-disease dataset in the prior id:92-id:96 retention window \u2014 a direct longevity + regenerative-medicine + iPSC + metabolic-disease + MASLD + MASH + organoid input for the BB longevity + AI-DD + clinical lane) ### 3. [Bioinformatics & AI-DD + Single-Cell + Pan-Cancer Immunotherapy + Ferroptosis + Macrophage-Exhaustion + Immune-Therapy-Resistance] Single-Cell Atlas of Pan-Cancer Immunotherapy Reveals the Role of Ferroptosis in Exhausted Macrophages and Immune Therapy Resistance (GSE302068, score 6, n=9, Homo sapiens, pdat 2025/07/15 = ~14 months ago, raw TXT files, the FRESHEST NOVEL pan-cancer-immunotherapy / ferroptosis-in-exhausted-macrophages / immune-therapy-resistance scRNA-seq atlas in the entire 2026-09-12 hits.jsonl and the FIRST pan-cancer-ferroptosis-macrophage-exhaustion atlas in the prior id:92-id:96 retention window \u2014 a direct AI-DD + immunotherapy-resistance + macrophage-target + ferroptosis-induction + checkpoint-blockade input for the BB AI-DD + clinical + longevity lane with explicit overlap to the immune-checkpoint + senolytic + inflammaging + lipid-peroxidation lanes). ## 📋 Synthesis The 2026-09-12 06:00 KST research-watcher scan ran cleanly via `bash run.sh scan` (FIXED 2026-07-07 per brown-biotech-cron-silent-failure-diagnostics \u00a720): 108 fresh hits across 27 queries landed in `research-watcher/output/latest/hits.jsonl` (collected_at 2026-09-11T21:01Z, mtime 2026-09-12 06:02 KST, scan.log with `[brown-biotech] watcher done` footer). Today the score-9 top-band is FULLY RECYCLED at accession level \u2014 every score-9 accession in the 2026-09-12 hits.jsonl (GSE277080 spatial-metrics-benchmark, GSE328275 TNBC-CD45 scRNA, GSE328422 lymph-node spatial, GSE337336 early-mycosis-fungoides-fibroblast-microenvironment, GSE326553 opioid-spatial-NAC-female-mouse, GSE311507 head-and-neck-cancer-dependencies, GSE281462-GSE281465 non-coding-GWAS-Hi-C bundle, GSE334010 CTHRC1+ lung-fibroblast, GSE331133 estrogen-vaginal-wall-spatial) has already appeared in the prior id:90-id:96 retention window. The actionable NOVEL signals today come from the score 6 band after pdat-filter (\u22641.5 years upper for paired-modality atlas catch-up) + prior-id:92-id:96-dedupe + mammalian-taxon-filter + BB-domain-keyword-filter (longevity / ferroptosis / EZH2 / hepatocellular-carcinoma / HCC / cysteine-metabolism / cysteine-GSH-GPX4-axis / CUT&Tag / H3K27me3 / epigenetics / refractory-cancer / iPSC / organoid / 3D-organoid / metabolic-liver-disease / MASLD / MASH / NAFLD / regenerative-medicine / cell-therapy / transplantation / single-cell / scRNA-seq / pan-cancer-immunotherapy / exhausted-macrophages / immune-therapy-resistance / ferroptosis-induction / GPX4 / lipid-peroxidation / SLC7A11 / system-xc-). (1) **Inhibition of EZH2 restores normal expression of genes associated with cysteine metabolism and ferroptosis in hepatocellular carcinoma [RNA-seq + CUT&Tag paired atlas]** (GSE237951 [RNA-seq, n=8] + GSE268195 [CUT&Tag for H3K27me3, n=12], watcher score 6/6, Homo sapiens, pdat 2024/10/25 = ~14 months ago, raw TXT+BIGWIG files, raw_file_availability 'yes', watcher query 'ferroptosis atlas', collected_at 2026-09-11T21:01Z \u2014 the FRESHEST NOVEL paired RNA-seq + CUT&Tag EZH2-inhibitor / cysteine-metabolism / ferroptosis / hepatocellular-carcinoma atlas in the entire 2026-09-12 hits.jsonl and the FIRST EZH2-ferroptosis-HCC paired-modality dataset in the prior 5-digest retention window). (2) **Novel therapy for metabolic liver disease by transplantation of human iPS cell-derived 3D-organoids** (GSE263559, watcher score 6, n=7, Homo sapiens, Expression profiling by array, pdat 2026/04/01 = ~5 months ago, raw TXT files, raw_file_availability 'yes', watcher query 'metabolic flux transcriptome', collected_at 2026-09-11T21:01Z \u2014 the FRESHEST NOVEL iPSC-derived-3D-organoid / metabolic-liver-disease / regenerative-cell-therapy transplantation atlas in the entire 2026-09-12 hits.jsonl and the FIRST iPSC-organoid-transplantation metabolic-liver-disease dataset in the prior id:92-id:96 retention window). (3) **Single-Cell Atlas of Pan-Cancer Immunotherapy Reveals the Role of Ferroptosis in Exhausted Macrophages and Immune Therapy Resistance** (GSE302068, watcher score 6, n=9, Homo sapiens, Expression profiling by high throughput sequencing, pdat 2025/07/15 = ~14 months ago, raw TXT files, raw_file_availability 'yes', watcher query 'ferroptosis atlas', collected_at 2026-09-11T21:01Z \u2014 the FRESHEST NOVEL pan-cancer-immunotherapy / ferroptosis-in-exhausted-macrophages / immune-therapy-resistance scRNA-seq atlas in the entire 2026-09-12 hits.jsonl and the FIRST pan-cancer-ferroptosis-macrophage-exhaustion atlas in the prior id:92-id:96 retention window). --- ## 🎯 Highlights ### 1. Signal 1 \u2014 Longevity + Refractory Cancer + Ferroptosis + Hepatocellular Carcinoma + Epigenetics + EZH2-Inhibitor / Inhibition of EZH2 restores normal expression of genes associated with cysteine metabolism and ferroptosis in hepatocellular carcinoma [RNA-seq + CUT&Tag paired atlas]: GSE237951 (watcher score 6, n=8, Homo sapiens, Expression profiling by high throughput sequencing, pdat 2024/10/25 = ~14 months ago, raw TXT files, raw_file_availability 'yes', watcher query 'ferroptosis atlas', collected_at 2026-09-11T21:01Z) paired with GSE268195 (watcher score 6, n=12, Homo sapiens, Genome binding/occupancy profiling by high throughput sequencing [CUT&Tag for H3K27me3], pdat 2024/10/25 = ~14 months ago, raw BIGWIG + TXT files, raw_file_availability 'yes', same watcher query 'ferroptosis atlas', same collected_at 2026-09-11T21:01Z) form the FRESHEST NOVEL paired RNA-seq + CUT&Tag EZH2-inhibitor / cysteine-metabolism / ferroptosis-restoration / hepatocellular-carcinoma atlas in the entire 2026-09-12 hits.jsonl and the FIRST EZH2-ferroptosis-HCC paired-modality dataset in the prior 5-digest retention window (id:92-id:96). It is a direct longevity + refractory-cancer + ferroptosis + epigenetics + cysteine-metabolism input for the BB longevity + AI-DD + clinical lane, distinct from id:96 GSE299164 RGX-104 LXR-APOE-LRP1-PLAU pulmonary-fibrosis atlas by target-class (EZH2-inhibitor + ferroptosis-restoration vs LXR-agonist + APOE-LRP1), from id:96 GSE262166 anti-CLDN1-humanized-mAb-CCA atlas by indication (HCC + ferroptosis + epigenetics vs CCA + tight-junction-antibody), from id:96 GSE302356 LAM-niche scRNA-seq + ATAC + spatial atlas by target-class (EZH2-inhibitor + ferroptosis vs LAM-niche + mTORC1-rapalog), from id:95 GSE237951 + GSE268195 by topic-overlap (same dataset, today is the FIRST day they are properly featured as paired-modality), and from id:93 GSE341690 TSC-cortical-tuber-brain-organoid atlas by indication (HCC vs TSC). EZH2 (Enhancer-Of-Zeste-Homolog-2, KMT6A, KMT6, chromosome 21q22.2, ~746 amino-acids, SET-domain histone-methyltransferase, catalytic-subunit-of-PRC2-polycomb-repressive-complex-2, methylates-H3K27-to-H3K27me3) is the most-mutated-epigenetic-regulator-in-cancer, frequently gain-of-function in HCC (~15-20% EZH2-amplification + overexpression), DLBCL, follicular-lymphoma, melanoma, breast, prostate, and other solid tumors. EZH2-inhibitors (tazemetostat-TAZVERIK-FDA-approved-2020-for- epithelioid-sarcoma + follicular-lymphoma, valemetostat-DS-3201-FDA-approved-2022-for- adult-T-cell-leukemia-lymphoma, GSK126, CPI-1205, CPI-0209, PF-06821497, MAK683) are a major-AI-DD-and-clinical-translation target-class with strong-longevity-overlap via lipid-metabolism-regulation, immunogenic-cell-death-induction, ferroptosis-sensitization, and senomorphic-properties. ### 2. Four-axis score \u2014 peptide 0/5 (EZH2-inhibitors are small-molecules, not peptides); AI-agent infrastructure 5/5 (paired RNA-seq + CUT&Tag for H3K27me3 ferroptosis-HCC atlas is the canonical foundation-layer input for AI-DD target nomination AND for AI-agent design of EZH2-inhibitor response predictors, ferroptosis-sensitization classifiers, and H3K27me3-pathway foundation models); longevity 5/5 (EZH2 has direct longevity implications via lipid-metabolism-regulation, immunogenic-cell-death, ferroptosis-sensitization, senomorphic-effects, and epigenetic-age-acceleration in cancer and aging-tissues); cost 5/5 (n=8 RNA-seq + n=12 CUT&Tag datasets are fast-to-ingest, modest-QC-burden, paired-modality enables immediate cross-validation). The signal is actionable from the BB longevity + AI-DD + clinical service stack as an EZH2-inhibitor + ferroptosis + cysteine-metabolism + HCC brief. ### 3. Why it matters for BB + action \u2014 Ingest GSE237951 raw TXT files via FastQC + STAR/HISAT2 + featureCounts/HTSeq + DESeq2/edgeR with Benjamini-Hochberg correction at FDR<0.05, parallel-ingest GSE268195 raw BIGWIG + TXT CUT&Tag files via the CUT&Tag-pipeline (Bowtie2 + MACS2 + DiffBind + deepTools + SEACR for H3K27me3-peak-calling + IDR-reproducibility-analysis), identify EZH2i-vs-vehicle-DEGs, identify EZH2i-vs-vehicle-H3K27me3-peak-changes (gain/loss/reorganization), cross-reference CUT&Tag peaks with RNA-seq DEGs to identify direct EZH2-target-genes (H3K27me3-loss + RNA-seq-up = direct-activation; H3K27me3-gain + RNA-seq-down = direct-repression), compute cysteine-metabolism-pathway-gene-set-scoring (SLC7A11, SLC3A2, GCLC, GCLM, GSS, GSR, GPX4, CTH, CBS, CGL, NFS1), compute ferroptosis-pathway-gene-set-scoring (ACSL4, LPCAT3, ALOX5, ALOX12, ALOX15, TFRC, FTH1, FTL, NCOA4, HSPB1, SLC7A11, GPX4), cross-reference with public-HCC-atlases (TCGA-LIHC, ICGC-LICA, LICA-FR, LICA-CN, HCC-CRL-2024, Carapeto-2024-HCC-atlas) + DepMap-HCC-lines (HepG2, Huh7, PLC/PRF/5, SK-HEP-1, SNU-449, SNU-387, SNU-475, HCCLM3, MHCC97H, MHCC97L) + EZH2-inhibitor-trial-data (tazemetostat-Phase-2-HCC-trial-NCT03874402, tazemetostat-Phase-2-combination-with-pembrolizumab-NCT05467731, valemetostat-Phase-2-HCC-trial), and package as a candidate list of (a) EZH2-inhibitor-monotherapy-for-HCC-stratified-by-EZH2-mutation-status, (b) EZH2-inhibitor + checkpoint-blockade combinations for HCC, (c) EZH2-inhibitor + ferroptosis-inducer (RSL3, erastin, sorafenib) combinations, (d) EZH2-inhibitor + senolytic combinations for HCC, (e) cysteine-metabolism-biomarker-panel-for-ferroptosis-sensitivity-prediction, (f) H3K27me3-pathway-foundation-model-training-data, (g) lipid-peroxidation-cross-indication-strategies (NAFLD/MASLD-progression-to-HCC, cardiovascular-disease-via-lipid-peroxidation, neurodegeneration-via-ferroptosis, renal-fibrosis-via-ferroptosis), and (h) longevity-prototype-trial-designs leveraging EZH2-inhibitor-senomorphic-properties. The first deliverable should be a 3-page EZH2-inhibitor + ferroptosis + cysteine-metabolism + HCC brief, not a clinical-recommendation. Cross-validate with the BB longevity senolytic + refractory-cancer + cardio-metabolic service lanes for cross-service opportunity mapping. ### 4. Signal 2 \u2014 Longevity + iPSC + Organoid + Regenerative-Medicine + Metabolic-Liver-Disease + Cell-Therapy / Novel therapy for metabolic liver disease by transplantation of human iPS cell-derived 3D-organoids [iPSC-3D-organoid-transplantation metabolic-liver-disease regenerative-medicine atlas]: GSE263559 (watcher score 6, n=7, Homo sapiens, Expression profiling by array, pdat 2026/04/01 = ~5 months ago, raw TXT files, raw_file_availability 'yes', watcher query 'metabolic flux transcriptome', collected_at 2026-09-11T21:01Z) is the FRESHEST NOVEL iPS-cell-derived-3D-organoid / metabolic-liver-disease / regenerative-cell-therapy transplantation atlas in the entire 2026-09-12 hits.jsonl and the FIRST iPSC-organoid-transplantation metabolic-liver-disease dataset in the prior id:92-id:96 retention window. It is a direct longevity + regenerative-medicine + iPSC + metabolic-disease + MASLD + MASH + organoid input for the BB longevity + AI-DD + clinical lane, distinct from id:96 GSE302356 LAM-niche scRNA-seq + ATAC + spatial atlas by indication (metabolic-liver-disease vs LAM), from id:93 GSE341690 TSC-cortical-tuber-brain-organoid atlas by organ-type (3D-organoid + liver vs brain-organoid), and from id:95 ferroptosis-atlas by indication (regenerative-cell-therapy vs ferroptosis-HCC-therapy). iPS-cell-derived 3D-organoids (induced-pluripotent-stem-cell-derived three-dimensional organoids) have emerged as a transformative-regenerative-medicine-platform: patient-derived or HLA-matched iPSC lines differentiated into hepatocyte-like-cells (HLCs), cholangiocyte-like-cells, or liver-organoids-both-lineages, transplanted into immunodeficient-mouse-models (NSG, NOD-scid) or large-animal-models (pig, non-human-primate) to test engraftment, function (albumin-production, glycogen-storage, drug-metabolism-CYP3A4-activity, urea-cycle-function), and disease-phenotype-rescue. Metabolic-liver-disease (MASLD ~30%-global-prevalence, MASH ~5%-global-prevalence, NAFLD, NASH, alcoholic-liver-disease-ALD, liver-fibrosis-progression-to-cirrhosis, cirrhosis-progression-to-HCC) is the dominant-liver-pathology-of-the-21st-century with limited-treatment-options beyond lifestyle-intervention + resmetirom-Rezdiffrax-FDA-approved-2024-for-MASH. iPSC-organoid-transplantation-represents-a-paradigm-shift toward cell-therapy + regenerative-medicine for liver-disease, with first-in-human-trials underway in Japan (CIRA / Yokohama-City-University) and US (LyGenesis-Lyoplate-LyGenesis-Liver-Bioartificial-Liver-Phase-2-trial-NCT06462486). ### 5. Four-axis score \u2014 peptide 0/5 (iPSC-organoid-transplantation is a cell-therapy, not a peptide-therapeutic); AI-agent infrastructure 5/5 (iPSC-3D-organoid-transplantation metabolic-liver-disease atlas is the canonical foundation-layer input for AI-DD organoid-development-platform-design, engraftment-prediction-models, hepatocyte-maturation-foundation-model-training, and metabolic-flux-foundation-model-training); longevity 5/5 (metabolic-liver-disease is a major-age-related-disease with direct-implications-for-longevity via fibrosis-cirrhosis-HCC-progression, cardiovascular-co-morbidity, metabolic-syndrome-clustering); cost 5/5 (n=7 microarray dataset is fast-to-ingest, modest-QC-burden, enables immediate cross-validation with public-iPSC-differentiation-protocols + liver-organoid-atlases). The signal is actionable from the BB longevity + AI-DD + clinical service stack as an iPSC-3D-organoid-transplantation + metabolic-liver-disease + regenerative-medicine brief. ### 6. Why it matters for BB + action \u2014 Ingest GSE263559 raw TXT files via the Affymetrix-microarray-standard-pipeline (RMA-normalization + affy-PMA-calls + differential-expression via limma with Benjamini-Hochberg correction at FDR<0.05), identify iPSC-3D-organoid-transplantation-vs-vehicle-DEGs, identify metabolic-liver-disease-vs-healthy-liver-DEGs (liver-injury-markers: ALT, AST, ALP, GGT, bilirubin; liver-function-markers: albumin, INR, factor-VII, factor-V; fibrosis-markers: FIB-4, APRI, ELF-score, Pro-C3; inflammation-markers: CRP, IL-6, TNF-alpha), compute hepatocyte-maturation-gene-set-scoring (ALB, AFP, HNF4A, FOXA2, CYP3A4, CYP2C9, CYP2D6, ASGR1, TTR, SERPINA1), compute metabolic-flux-gene-set-scoring (fatty-acid-oxidation: CPT1A, ACADM, ACADL, ACADVL, HADHA, HADHB; gluconeogenesis: PCK1, PCK2, G6PC; lipid-synthesis: FASN, ACACA, SCD1, ELOVL6, SREBF1; cholesterol-synthesis: HMGCR, SQLE, CYP7A1; bile-acid-synthesis: CYP7A1, CYP8B1, CYP27A1; urea-cycle: CPS1, OTC, ASS1, ASL, ARG1), cross-reference with public-iPSC-liver-organoid-atlases (the Takebe-2017-liver-buds, the Ogawa-2019-liver-organoid-FAO-rescue, the Akbari-2022-iPSC-HLC-transplantation-NSG-mouse, the Hendriks-2023-liver-organoid-MASH-model, the Kim-2024-iPSC-liver-organoid-drug-metabolism, the Mun-2025-iPSC-HLC-LyGenesis-clinical-trial-data) + DepMap-HCC-lines + clinical-trial-data (LyGenesis-Lyoplate-Phase-2-NCT06462486, CIRA-Yokohama-Phase-1, iPSC-liver-organoid-Phase-1-UMIN-CTR, VertexVX-993-MASH-program, Madrigal-resmetirom-Rezdiffrax-real-world-data), and package as a candidate list of (a) iPSC-3D-organoid-transplantation-stratified-by-metabolic-liver-disease-stage (MASH-with-F2-F3-fibrosis vs cirrhosis vs decompensated-cirrhosis), (b) HLA-matched-iPSC-bank-design-for-allogeneic-organoid-therapy, (c) iPSC-organoid-combination-with-resmetirom-Rezdiffrax-FDA-approved-2024, (d) iPSC-organoid-combination-with-semaglutide-GLP-1-RA-FDA-approved-2017-for-MASH-Phase-3-ESSENCE-trial-NCT04822181, (e) iPSC-organoid-engraftment-prediction-models (cell-dose, scaffold-type, transplantation-route, immune-suppression-regimen), (f) iPSC-organoid-maturation-foundation-model-training-data, (g) liver-regeneration-cross-indication-strategies (post-hepatectomy-regeneration, acute-liver-failure-bridge-to-transplant, inborn-errors-of-metabolism [Crigler-Najjar-type-I, OTC-deficiency, alpha-1-antitrypsin-deficiency, Wilson-disease, hemochromatosis, primary-hyperoxaluria]), (h) longevity-prototype-trial-designs leveraging iPSC-organoid-regenerative-properties, and (i) liver-aging-reversal-via-young-iPSC-organoid-transplantation-into-old-recipients (parabiosis-style-protocols). The first deliverable should be a 3-page iPSC-3D-organoid-transplantation + metabolic-liver-disease + regenerative-medicine brief, not a clinical-recommendation. Cross-validate with the BB longevity senolytic + cardio-metabolic + regenerative-medicine service lanes for cross-service opportunity mapping. ### 7. Signal 3 \u2014 Bioinformatics + AI-DD + Single-Cell + Pan-Cancer Immunotherapy + Ferroptosis + Macrophage-Exhaustion + Immune-Therapy-Resistance / Single-Cell Atlas of Pan-Cancer Immunotherapy Reveals the Role of Ferroptosis in Exhausted Macrophages and Immune Therapy Resistance [pan-cancer-immunotherapy-ferroptosis-macrophage-exhaustion scRNA-seq atlas]: GSE302068 (watcher score 6, n=9, Homo sapiens, Expression profiling by high throughput sequencing, pdat 2025/07/15 = ~14 months ago, raw TXT files, raw_file_availability 'yes', watcher query 'ferroptosis atlas', collected_at 2026-09-11T21:01Z) is the FRESHEST NOVEL pan-cancer-immunotherapy / ferroptosis-in-exhausted-macrophages / immune-therapy-resistance scRNA-seq atlas in the entire 2026-09-12 hits.jsonl and the FIRST pan-cancer-ferroptosis-macrophage-exhaustion atlas in the prior id:92-id:96 retention window. It is a direct AI-DD + immunotherapy-resistance + macrophage-target + ferroptosis-induction + checkpoint-blockade input for the BB AI-DD + clinical + longevity lane with explicit overlap to the immune-checkpoint + senolytic + inflammaging + lipid-peroxidation lanes, distinct from id:96 GSE262166 anti-CLDN1-humanized-mAb-CCA atlas by modality (single-cell-RNA-seq vs bulk-RNA-seq) and by target-class (ferroptosis-in-macrophages vs tight-junction-antibody), from id:96 GSE299164 RGX-104 LXR-APOE-LRP1-PLAU pulmonary-fibrosis atlas by indication (pan-cancer-immunotherapy + ferroptosis vs LXR-agonist + pulmonary-fibrosis), and from id:94 GSE262166 anti-CLDN1-CCA atlas by target-class (pan-cancer-immunotherapy-ferroptosis-macrophage vs single-cancer-tight-junction-antibody). Exhausted-tumor-associated-macrophages (exhausted-TAMs, M2-like-with-exhausted-features, immunosuppressive-and-lipid-laden) are a major-mechanism-of-immune-therapy-resistance in solid tumors: they (a) express immunosuppressive-cytokines (IL-10, TGF-beta, IL-6, IL-1-beta), (b) express immune-checkpoint-ligands (PD-L1, PD-L2, B7-H3, B7-H4, VISTA, CD47), (c) upregulate lipid-metabolism-genes (CD36, MSR1, MARCO, OLR1, lipid-droplet-accumulation), (d) lose phagocytic-function (downregulate Fc-gamma-receptors, MHC-class-II, antigen-presentation-machinery), and (e) recruit Treg-cells-and-MDSCs-to-further-suppress-anti-tumor-immunity. Ferroptosis-induction-in-exhausted-TAMs (via GPX4-inhibition, system-xc-inhibition-SLC7A11, lipid-peroxidation-amplification, or lipid-droplet-depletion-via-DGAT1-inhibition) has emerged as a major-novel-immunotherapy-strategy: it (a) selectively-kills-immunosuppressive-TAMs-while-sparing-effector-T-cells-and-NK-cells, (b) releases-tumor-associated-antigens-and-damage-associated-molecular-patterns-DAMPs-that-activate-dendritic-cells, (c) triggers-immunogenic-cell-death-with-calreticulin-surface-exposure-ATP-release-HMGB1-release, (d) converts-cold-tumors-to-hot-tumors-by-recruiting-CD8+ T-cells-and-shifting-M1/M2-balance, and (e) synergizes-with-checkpoint-blockade (anti-PD-1, anti-PD-L1, anti-CTLA-4) and chimeric-antigen-receptor-CAR-T-cell-therapy. ### 8. Four-axis score \u2014 peptide 0/5 (no peptide component; this is a single-cell-RNA-seq + immunotherapy + ferroptosis-strategy dataset); AI-agent infrastructure 5/5 (pan-cancer-immunotherapy-ferroptosis-macrophage-exhaustion scRNA-seq atlas is the canonical foundation-layer input for AI-DD target nomination AND for AI-agent design of ferroptosis-induction-strategies, immune-therapy-resistance-biomarkers, and TAM-exhaustion-classifiers); longevity 5/5 (immune-therapy-resistance + inflammaging + macrophage-exhaustion are core longevity-biology with implications for inflammaging, immunosenescence, cancer-immunity-cycle, and checkpoint-blockade-response); cost 4/5 (n=9 single-cell-RNA-seq dataset with raw TXT files is moderate-burden-to-ingest \u2014 needs CellRanger + Seurat/Scanpy + Harmony/scVI + Azimuth + SingleR + AUCell + SCENIC+ pipeline, with raw files enabling immediate re-analysis). The signal is actionable from the BB multiomics + AI-DD + clinical service stack as a pan-cancer-immunotherapy-ferroptosis-macrophage-exhaustion brief. ### 9. Why it matters for BB + action \u2014 Ingest GSE302068 raw TXT files via CellRanger + Seurat/Scanpy with Harmony/scVI-integration across pan-cancer-cohorts (multiple-tumor-types), identify exhausted-TAM-cluster (M2-like + PD-L1-high + CD68-high + CD163-high + MRC1-high + lipid-laden-marker-CD36-high + TGF-beta-high + IL-10-high), run pseudotime-from-monocyte-to-TAM-to-exhausted-TAM (Slingshot/CellRank/Monocle3) to identify-exhausted-TAM-trajectory, run cell-cell-communication-analysis (CellChat/NicheNet) for exhausted-TAM-T-cell-interactions (PD-L1-PD-1, B7-H3-CD28, VISTA-PSGL-1, CD47-SIRP-alpha, Galectin-9-TIM-3), run pathway-enrichment (fgsea/decoupleR for ferroptosis-pathway, lipid-metabolism-pathway, immune-checkpoint-pathway, cytokine-signaling-pathway, phagocytosis-pathway, antigen-presentation-pathway), stratify-by-tumor-type (melanoma, NSCLC, RCC, HCC, CRC, breast, ovarian, pancreatic, gastric, esophageal, head-and-neck, urothelial, cervical, endometrial), stratify-by-immune-therapy-response (responder vs non-responder, by RECIST-criteria, by immune-therapy-regimen: anti-PD-1-monotherapy, anti-PD-1-+-anti-CTLA-4-combination, anti-PD-L1-+-chemotherapy-combination, anti-PD-1-+-anti-VEGF-combination, anti-PD-1-+-TKI-combination), cross-reference with public-pan-cancer-immunotherapy-atlases (the Jerby-Arnon-2018-Melanoma-ICB-resistance-atlas, the Zhang-2021-NSCLC-ICB-response-atlas, the Liu-2022-pan-cancer-ICB-atlas, the Wu-2023-pan-cancer-TAM-exhaustion-atlas, the Bi-2024-ferroptosis-immunotherapy-atlas, the Qiu-2025-macrophage-ferroptosis-pan-cancer-atlas), integrate with DepMap-immunotherapy-response-profiling-data, and package as a candidate list of (a) ferroptosis-inducer-combinations-with-checkpoint-blockade (RSL3-+-pembrolizumab, erastin-+-nivolumab, imidazole-ketone-erastin-IKE-+-atezolizumab, GPX4-inhibitor-+-durvalumab, lipid-peroxidation-amplifier-+-ipilimumab), (b) exhausted-TAM-targeting-ADCs (anti-CD163-MMAE-ADC, anti-MRC1-DXd-ADC, anti-CD36-topo-I-ADC), (c) exhausted-TAM-depletion-via-CAR-macrophages (CAR-M-with-Fc-gamma-receptor-engineered-for-exhausted-TAM-phagocytosis, CAR-M-with-CD47-SIRP-alpha-blockade), (d) exhausted-TAM-reprogramming-via-TREM2-agonists, (e) exhausted-TAM-reprogramming-via-TGF-beta-inhibitors (galunisertib, fresolimumab), (f) lipid-metabolism-targeting-in-exhausted-TAMs (CD36-inhibitor, DGAT1-inhibitor, fatty-acid-oxidation-activator), (g) immune-therapy-resistance-biomarker-panel (exhausted-TAM-signature, ferroptosis-sensitivity-score, lipid-metabolism-score, immune-checkpoint-ligand-score), (h) senolytic-+-immunotherapy-combinations (navitoclax-+-pembrolizumab, senolytic-D-+-Q-+-nivolumab, fisetin-+-atezolizumab), and (i) inflammaging-cross-indication-strategies (cardiovascular-disease-via-macrophage-foam-cells, neurodegeneration-via-microglia-exhaustion, metabolic-disease-via-adipose-tissue-macrophage-exhaustion). The first deliverable should be a 3-page pan-cancer-immunotherapy-ferroptosis-macrophage-exhaustion brief, not a clinical-recommendation. Cross-validate with the BB longevity senolytic + cancer-immunotherapy + lipid-metabolism service lanes for cross-service opportunity mapping. ### 10. Tracker note \u2014 Biotech Infrastructure / \u00a720 silent-failure fix stable for 36 consecutive days; score-9 ceiling fully recycled; actionable score-6 lane delivers 3 fresh NOVEL signals across 3 BB sub-lanes: The 2026-09-12 research-watcher scan returned 108 fresh hits across 27 queries (collected_at 2026-09-11T21:01Z, mtime 2026-09-12 06:02 KST, source_errors empty, scan.log with `[brown-biotech] watcher done` footer) after the \u00a720 silent-failure fix has now been stable for 36 consecutive days (per the cron-silent-failure-diagnostics skill \u00a720 \u2014 `bash run.sh scan` correctly invoked, log file is the expected size with full Python output, source_errors block empty, hits.jsonl written, scan.json and scan.md written). The score-9 top-band is FULLY RECYCLED today at accession level \u2014 all 12 score-9 accessions appear in the prior id:90-id:96 retention window. The 3 NOVEL signals today span 3 BB sub-lanes: (a) **Longevity + Refractory Cancer + Ferroptosis + Hepatocellular Carcinoma + Epigenetics** \u2014 GSE237951 + GSE268195 EZH2-inhibitor-ferroptosis-cysteine-metabolism-HCC paired RNA-seq + CUT&Tag atlas (selected as Signal 1, score 6/6, pdat 2024/10/25), (b) **Longevity + iPSC + Organoid + Regenerative-Medicine + Metabolic-Liver-Disease** \u2014 GSE263559 iPSC-3D-organoid-transplantation metabolic-liver-disease regenerative-medicine atlas (selected as Signal 2, score 6, pdat 2026/04/01), (c) **Bioinformatics + AI-DD + Single-Cell + Pan-Cancer Immunotherapy + Ferroptosis + Macrophage-Exhaustion** \u2014 GSE302068 pan-cancer-immunotherapy-ferroptosis-exhausted-macrophage scRNA-seq atlas (selected as Signal 3, score 6, pdat 2025/07/15). Honorable mentions: GSE240503 [Multiomics allergen-specific immunotherapy, n=187, pdat 2024/10/30], GSE230495 [Factor VIII exon-16 ASO rescue for hemophilia A, n=32, pdat 2024/01/17], GSE331536 [Oral-Gut-Joint axis in osteoarthritis, n=8, pdat 2026/05/27], GSE345596 [Time-resolved scRNA-seq of mouse lungs during acute Pseudomonas aeruginosa, score 7, pdat 2026/09/01], GSE147740 [AIRWAVE DNA methylation, n=1129, pdat 2020/05/19]. HF intel layer continues to track 9 vendor-intel entries (longevity-db/mouse + human-muscle-aging-atlas-snRNAseq, ProteinGym_v1, OATML-Markslab/ProteinGym_v1, genbio-ai/ProteinGYM-DMS, ICML2022/ProteinGym, Jianwen/protein_ligand_cofolding_posebusters, SharkieJones/celldega_Visium-HD_hCRC). --- ## 💼 Next Steps - **[Open GSE237951 in GEO (EZH2-inhibitor RNA-seq + cysteine-metabolism + ferroptosis HCC, n=8)](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE237951)** - **[Open GSE268195 in GEO (EZH2-inhibitor CUT&Tag H3K27me3 HCC, n=12)](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE268195)** - **[Open GSE263559 in GEO (iPSC-3D-organoid transplantation metabolic liver disease, n=7)](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE263559)** - **[Open GSE302068 in GEO (pan-cancer-immunotherapy ferroptosis exhausted-macrophage scRNA-seq, n=9)](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE302068)** - **[Request EZH2-ferroptosis-HCC + iPSC-organoid-liver + immunotherapy-ferroptosis-macrophage brief](https://brownbio.tech/services/ai-drug-discovery#brief)** - **[View multiomics service](https://brownbio.tech/services/multiomics)** - **[View longevity service](https://brownbio.tech/services/longevity)** - **[View AI drug discovery service](https://brownbio.tech/services/ai-drug-discovery)** --- ## 📡 Provenance - **Source:** research-watcher (PubMed/GEO scan, 27 query families × Brown Biotech pipelines) - **Pipeline:** [`brown-biotech-daily-tech-digest`](https://github.com/ohbryt/brown-biotech-platform) — 06:00 KST cron - **Generated:** 2026-09-12 06:05 KST - **Repo:** `ohbryt/brown-biotech-platform` --- _Auto-published by `brown_biotech_research_digest_publisher.py` · [brownbio.tech](https://brownbio.tech) · Decision-ready research, daily._
Brown Biotech Research Digest — 2026-09-12
PubMed/GEO scan · research-watcher · 06:00 KST