Cancer-induced nerve injury promotes resistance to anti-PD-1 therapy [RNA-seq]

Peri-neural invasion (PNI) is a well-established poor prognostic factor in multiple cancer types. However, the mechanisms driving the PNI's detrimental clinical effect remain elusive. Here, we provide clinical and mechanistic insights into PNI and cancer-induced injury of tumor-associated nerves (TANs) and their role in resistance to anti-PD-1 therapy. Our work demonstrates that poor response to anti-PD-1 therapy in cutaneous squamous cell carcinoma (cSCC), melanoma, and gastric cancer is associated with PNI and TANs injury. Ultrastructural electron microscopy analysis reveals that direct contact between cancer cells and nerve fibers leads to cancer-induced nerve injury (CINI) via myelin degradation. Injured neurons respond by autonomously initiating an interleukin (IL)-6 and interferon (IFN) type I inflammatory response. This inflammatory response alters the immune activity in the peri-neural niche in melanoma, cSCC, and pancreatic adenocarcinoma, leading to an immuno-suppressive activity aimed at nerve healing and regeneration. As the tumor grows, the CINI burden increases, the inflammatory signal within the niche becomes chronic, and eventually skews the general immune tone within the tumor microenvironment to a suppressive and exhaustive state. The CINI-driven anti-PD-1 resistance can be reversed by targeting multiple steps in the CINI signaling process: denervating the tumor, conditional knockout of the transcription factor mediating the injury signal within neurons (cKO-Atf3), knockout of the IFN-a receptor signaling (Ifnar1-/-), or by combining anti-PD-1 and anti-IL-6-receptor blockade. Our findings demonstrate the direct immuno-regulatory roles of TANs and their therapeutic potential. Bulk RNA sequencing of lung metastasis–innervating neurons was performed following intravenous injection of 5 × 105 B16F10-eGFP melanoma cells or vehicle into nociceptor neuron reporter mice (Trpv1Cre::Td-tomatofl/wt). Two weeks later, the mice were euthanized, and lung metastases were visually confirmed in the B16F10-eGFP-inoculated group. Jugular nodose ganglia (JNC) were dissected into ice-cold HEPES-buffered DMEM (Thermofisher, #12430062). For each biological replicate, JNC from Three mice were pooled and transferred into HEPES-buffered DMEM containing 1 mg/mL collagenase IV (Sigma, #C5138) and 2.4 U/mL dispase II (Sigma, #04942078001), then incubated at 37 °C for 70 minutes. After washing with DMEM, ganglia were gently triturated using glass Pasteur pipettes of decreasing diameter. Cells were centrifuged at 200 g over a 15% BSA gradient in PBS to remove debris, stained with SYTO 40 (10 μM, Thermofisher, #S11351) for 5 minutes at room temperature to distinguish cells from axonal debris, washed with PBS, and resuspended in sterile flow cytometry buffer (PBS with 2% FBS and 1 mM EDTA). The cell suspension was filtered through a 70 µm mesh (VWR, #10204-924), and nociceptor neurons (td-tomato+) were FACS-enriched on a BD FACSAria cell sorter24 , collected directly into 500 µL TRIzol reagent (Invitrogen, #15596026), and stored at −80 °C until RNA extraction, following established protocols33. Library preparation was conducted at the Institut de Recherche en Cancérologie et en Immunologie (IRIC), Université de Montréal. RNA quality was confirmed on an Agilent Bioanalyzer, with all samples achieving an RNA Integrity Number (RIN) ≥ 7.5. Libraries were prepared using a poly(A)-enrichment, single-stranded RNA-seq method (KapaBiosystems, KAPA RNA Hyperprep Kit, #KR1352) and sequenced on an Illumina NextSeq500 platform with 75-cycle single-end reads. Basecalling was performed using Illumina RTA 2.4.11, and demultiplexing was done with bcl2fastq 2.20 (allowing one mismatch in the index). Trimmomatic was used to remove adapter sequences and low-quality bases from the 3′ end of each read, and the resulting high-quality reads were aligned to the GRCm38 mouse genome using STAR v2.5.11, which also generated gene-level read counts. Differential expression analysis was performed in DESeq2 using normalized read counts, and genes were considered differentially expressed if they had an adjusted p-value (false discovery rate, FDR) < 0.05. Log2 fold changes and −log10 p-values were calculated from the normalized data, and additional data analysis and visualization were carried out in RStudio.