Dark DNA’s Long RNAs Emerge as Master Regulators of Colorectal Cancer

For decades, the vast stretches of the human genome that do not encode proteins were dismissed as evolutionary baggage, a biological wasteland dismissed with the shorthand of “junk DNA.” A comprehensive new review published in Nature Reviews Gastroenterology & Hepatology makes a compelling case that this so-called dark matter of the genome is not only functional but may hold the keys to understanding and ultimately treating colorectal cancer, one of the world’s most common and lethal malignancies. The review, led by researchers at the University of Southern California’s Norris Comprehensive Cancer Center and Humanitas University in Milan, synthesizes an enormous body of evidence showing that long non-coding RNAs, or lncRNAs, act as master architects of the tumour microenvironment, the complex cellular ecosystem that surrounds and shapes a growing tumour.

The scale of the non-coding genome is staggering. Only a small fraction of human DNA encodes proteins; the rest is transcribed into tens of thousands of RNA molecules that never become proteins but instead perform regulatory duties of remarkable sophistication. Long non-coding RNAs, defined as transcripts longer than 200 nucleotides, have emerged as pivotal players in this regulatory landscape. They can guide proteins to specific genomic locations, scaffold multi-molecular complexes, sponge microRNAs away from their targets, modulate splicing, and stabilize messenger RNAs. In cancer, these functions go awry: individual lncRNAs can behave as oncogenes that drive malignant growth or as tumour suppressors that restrain it, and the same molecule can play opposite roles depending on cellular context.

What distinguishes the new review is its focus on the tumour microenvironment rather than the cancer cell in isolation. Colorectal tumours are not simply masses of dividing epithelial cells; they are embedded in a dynamic neighbourhood of immune cells, cancer-associated fibroblasts, endothelial cells, extracellular matrix and signalling molecules. This ecosystem determines whether a tumour grows aggressively, metastasizes to the liver, evades immune surveillance or succumbs to therapy. The authors argue that lncRNAs exert profound influence over every one of these components, making them central to cancer progression, metastasis and drug resistance.

Consider the immune compartment. The review documents numerous lncRNAs that help colorectal tumours escape destruction by cytotoxic T cells and natural killer cells. The lncRNA SNHG16 drives PD-L1-mediated immune escape by regulating the miR-324-3p/ELK4 signalling axis, while a feedback loop involving LINC00460, miR-186-3p and MYC enhances the expression of both CD47 and PD-L1, two molecular shields that tumours raise against immune attack. KCNQ1OT1, carried in tumour-derived exosomes, mediates CD8-positive T cell exhaustion by regulating CD155 expression. VPS9D1-AS1 amplifies intratumoural TGF-beta signalling, promoting tumour cell escape from T cell killing, whereas MIR22HG acts in the opposite direction, functioning as a tumour suppressor through TGF-beta/SMAD signalling and facilitating immunotherapy responses. These findings suggest that lncRNA profiles could predict which patients benefit from immune checkpoint inhibitors, a question of enormous clinical importance in colorectal cancer, where immunotherapy currently helps only a subset of patients.

Macrophages, the immune cells that can either attack tumours or be co-opted to support them, are another major target of lncRNA regulation. Tumour-associated macrophages exist along a spectrum from pro-inflammatory, anti-tumour M1 states to immunosuppressive, pro-tumour M2 states, and multiple lncRNAs push this balance toward the M2 pole. The lncRNA RPPH1 promotes metastasis by driving exosome-mediated M2 polarization, while RP11-417E7.1 activates Wnt/beta-catenin signalling and facilitates exosome-driven macrophage reprogramming. Conversely, NBR2 suppresses colorectal cancer progression by regulating macrophage polarization, illustrating that lncRNAs can be harnessed as anti-tumour forces. NEAT1, which activates inflammasomes in macrophages, promotes cancer-associated inflammation, linking chronic inflammatory conditions such as ulcerative colitis to malignant transformation through non-coding RNA networks.

Cancer-associated fibroblasts, the most abundant stromal cells in many tumours, are equally subject to lncRNA control. Normal fibroblasts can be converted into tumour-promoting CAFs through lncRNA cargo delivered by extracellular vesicles: LINC01915 facilitates this conversion through the miR-92a-3p/KLF4/CH25H axis. Once converted, CAFs return fire, secreting exosomes loaded with lncRNAs such as H19, which transfers stemness and chemoresistance to cancer cells, FAL1, which promotes oxaliplatin resistance through autophagy regulation, and LINC00355, which drives epithelial-mesenchymal transition through the miR-34b-5p/CRKL axis. Exosomal PWAR6 from myofibroblastic CAFs accelerates liver metastasis by altering glutamine availability and impairing natural killer cell function. The lncRNA FENDRR, by contrast, suppresses CAF activity and serves as a favourable prognostic indicator, underscoring the therapeutic potential of tipping these stromal conversations in the patient’s favour.

Angiogenesis, the formation of new blood vessels that feed tumour growth, is a third pillar of microenvironmental control. The review catalogues lncRNAs that regulate vascular remodelling through well-defined pathways. PVT1 promotes tumorigenesis by stabilizing miR-16-5p and engaging the VEGFA/VEGFR1/AKT axis, while ZFAS1, induced by the transcription factor SP1, drives progression through the miR-150-5p/VEGFA axis. NORAD induction under hypoxia accounts for chemoresistance and vasculogenic mimicry by sponging miR-495-3p and elevating HIF-1alpha. On the suppressive side, the lncRNA HITT forms a regulatory loop with HIF-1alpha to restrain angiogenesis and tumour growth, and GAS5 inhibits both angiogenesis and metastasis through Wnt/beta-catenin signalling. Because anti-angiogenic drugs are a mainstay of colorectal cancer treatment, lncRNAs that mediate resistance to these agents represent attractive predictive biomarkers and combination-therapy targets.

Perhaps the most clinically translatable theme is the role of extracellular vesicles, particularly exosomes, as carriers of lncRNA cargo between cells. Tumour cells package specific lncRNAs into vesicles that travel through blood and other bodily fluids, reprogramming recipient cells at distant sites. CRNDE-h delivered by tumour exosomes promotes Th17 cell differentiation; HLA-F-AS1 in cancer-derived vesicles induces macrophage polarization and metastasis; MIR181A1HG in vesicles from highly metastatic cells remodels the extracellular matrix and recruits myeloid-derived suppressor cells to the liver. Crucially, because these vesicle-associated lncRNAs circulate in accessible biofluids, they are emerging as promising non-invasive biomarkers. Plasma exosomal lncRNAs have already shown utility for early detection of colorectal cancer, and circulating lncRNAs such as CRNDE-h, SNHG11, DANCR, HOTAIR and MALAT1 have been proposed as diagnostic and prognostic markers that could complement or even outperform conventional markers like carcinoembryonic antigen.

The review also confronts the challenge of therapeutic resistance, an area where lncRNAs play a decisive role. HOTAIR contributes to 5-fluorouracil resistance by suppressing miR-218 and activating NF-kappaB/thymidylate synthase signalling. MALAT1 is associated with poor response to oxaliplatin-based chemotherapy and promotes resistance through the epigenetic regulator EZH2. MIR100HG-derived microRNAs mediate cetuximab resistance through Wnt/beta-catenin signalling, while exosomal UCA1 predicts cetuximab-resistant disease. The lncRNA CCAL, transferred from fibroblasts via exosomes, promotes chemoresistance in cancer cells. These mechanisms suggest that measuring lncRNA expression before and during treatment could guide therapy selection, and that targeting resistance-driving lncRNAs could resensitize tumours to existing drugs.

Translating these discoveries into the clinic will require overcoming substantial hurdles, including the safe and efficient delivery of RNA-targeted therapeutics, the tissue specificity of lncRNA function and the need for rigorous validation in large patient cohorts. Yet the tools are advancing rapidly: antisense oligonucleotides and small interfering RNAs can silence oncogenic lncRNAs, spatial transcriptomics is revealing where lncRNAs act within intact tumours, and machine-learning approaches are distilling multi-centre data into lncRNA signatures capable of stratifying patients with stage II and III colorectal cancer. The authors conclude that a deeper understanding of the intricate interplay between lncRNAs and the tumour microenvironment paves the way for innovations in precision oncology, transforming molecules once dismissed as genomic noise into the next generation of biomarkers and drug targets for a disease that claims hundreds of thousands of lives each year.

Subject of Research: The role of long non-coding RNAs in regulating the colorectal cancer tumour microenvironment and enabling precision oncology

Article Title: Long non-coding RNAs in colorectal cancer: shaping the tumour microenvironment and advancing precision oncology

Article References: Bartolini, M., Battaglin, F., Soni, S., Algaze, S., Ashouri, K., Torres-Gonzalez, L., Mittal, P., Shah, U., Zhang, W., Millstein, J., Puccini, A., & Lenz, H.-J. (2026). Long non-coding RNAs in colorectal cancer: shaping the tumour microenvironment and advancing precision oncology. Nature Reviews Gastroenterology & Hepatology. https://doi.org/10.1038/s41575-026-01243-3

Image Credits: AI Generated

DOI: 10.1038/s41575-026-01243-3

Keywords: long non-coding RNAs, colorectal cancer, tumour microenvironment, dark DNA, exosomes, biomarkers, immune evasion, cancer-associated fibroblasts, angiogenesis, drug resistance, precision oncology, liquid biopsy