From mechanistic maps to medicine: Navigating the translational challenges of m6A modification in autoimmunity and cancer

We read with considerable interest the comprehensive review by Geng et al., (2025) which synthesizes the complex effects on autoimmune diseases and cancer of m6A modification. This modification refers to the methylation, at N6, of adenosine bases in mRNA, a key regulatory mechanism in RNA biology, as it controls the stability, translation and degradation of mRNA. The authors have successfully charted the expansive ‘epitranscriptomic landscape’, illustrating how dysregulation of writers, erasers and readers of such RNA methylation contributes to pathogenesis. While this map is undeniably valuable, the journey from this mechanistic cartography to genuine clinical therapeutics is far more problematic than the review might suggest. We offer a critical appraisal to highlight the substantial gaps and challenges that must be addressed to realize the proposed ‘medicinal value’. The review accurately notes the context-dependent functions of the components (writers, erasers and readers) of m6A modification but underplays the profound therapeutic paradox this creates. For instance, the action of the RNA methyltransferase METTL3 (a writer) leads to starkly opposing outcomes. METTL3 acts as a potent oncogene in acute myeloid leukaemia by promoting the translation of key growth genes (Lee et al., 2024) but functions as a tumour suppressor in glioblastoma, by maintaining the stability of the mRNA for the metalloproteinase ADAM19 which, through its disintegrin activity, suppresses invasiveness (Levental et al., 2024). This duality extends to the immune system, where METTL3 in T cells is crucial for sustaining immune responses but can also enforce immune tolerance. A systemically administered, small-molecule inhibitor of METTL3, developed for leukaemia, could therefore inadvertently promote the progression of other malignancies or disrupt vital immune surveillance. This is not a minor side effect but a fundamental challenge to the entire ‘druggable target’ paradigm for core m6A machinery. The field must pivot towards developing sophisticated delivery systems—such as cell-specific lipid nanoparticles or antibody-drug conjugates targeting lineage-specific surface markers—to achieve the selectivity of effect required. The question is not just if we can inhibit a target, but where, when and in which cell type we can safely do so. Geng et al. rightly highlight the diagnostic potential of m6A modifications. However, the transition from technically complex m6A-seq in bulk tumour tissue to a clinically robust, actionable biomarker is a chasm that has yet to be bridged. The heterogeneity of m6A patterns is immense, not only between patients but also within single tumours and across different immune cell subsets. Bulk sequencing obscures these critical nuances, potentially masking the very signatures with prognostic power. While the authors mention this heterogeneity, they do not critically address the operational and analytical hurdles. The future lies in single-cell m6A mapping technologies (Li et al., 2023), but these are currently far from clinical application. Furthermore, for a biomarker to be clinically useful, it must be measurable in readily accessible biofluids, such as plasma, via a standardized, cost-effective assay (e.g. RT-qPCR or ELISA for m6A-modified transcripts). Currently, no such validated, circulating m6A-based biomarker exists for any autoimmune or oncological disease. The diagnostic promise remains largely confined to the research domain. The therapeutic discussion in the review focuses predominantly on catalytic inhibitors of writers, such as METTL3/14 inhibitors) and of erasers, such as inhibitors of the fat mass and obesity-associated protein (FTO). The FTO protein is a dioxygenase that acts as a demethylase, targeting the m6A in RNA. This narrow focus on writers and erasers overlooks two major strategic opportunities. First, the protein–protein interfaces within the m6A complex, particularly those of the reader proteins, such as YTHDF1/2, represent a vast untapped pharmacological space. Disrupting the interaction between a specific reader and its effector machinery (e.g. the YTHDF1–eIF3 axis) could allow for selective modulation of a subset of m6A-driven pathways, offering a more nuanced therapeutic effect than a global, body-wide blockade of m6A modification. Second, the potential of RNA-targeting therapeutic agents is conspicuously underemphasized. Anti-sense oligonucleotides or small interfering RNAs could be designed to specifically mask or degrade hypermethylated, oncogenic transcripts, such as those for the master transcription factor, c-Myc, and for EGF receptors, or to rescue the expression of hypomethylated tumour suppressors. This approach, targeting the ‘effector’ RNA directly, could bypass the contextual pitfalls of manipulating the machinery of m6A modification itself (Deng et al., 2024). In conclusion, while the review by Geng et al. provides an excellent mechanistic summary, the path to the clinic is obstructed by fundamental biological and translational challenges. The context-dependency of the outcomes of m6A modification necessitates a new class of spatially resolved therapeutic agents. The diagnostic potential requires a leap from bulk omics to single-cell resolution and liquid biopsy validation. Finally, the therapeutic arsenal must expand beyond catalytic inhibition to include protein–protein interaction disruptors and RNA-targeted agents. Addressing these critical issues will determine whether the m6A modifications can transition from compelling mechanistic players to bona fide therapeutic agents in medicine. Du Jiang Yang: Conceptualization; writing—original draft; writing—review and editing; methodology; investigation; methodology. Jiexiang Yang: Conceptualization; writing—original draft; writing—review and editing; methodology; investigation; methodology. Guo You Wang: Investigation; funding acquisition. We have received funding from The Affiliated Traditional Chinese Medicine Hospital of Southwest Medical University, ‘Biomechanical research on cartilage transplantation based on the theory of “equal emphasis on muscle and bone”’ (2023ZYQJ02). All authors have no conflicts of interest or financial ties to disclose.