Sphingosine Rheostat Reemerges as an Immune Suppression Switch in Metastatic Colorectal Cancer

Three decades ago, cancer biologists embraced an elegant idea known as the sphingosine rheostat: a molecular dial inside tumor cells that weighed the pro-survival lipid sphingosine-1-phosphate against its pro-death counterpart, ceramide. When the dial tipped toward sphingosine-1-phosphate, cells survived; when it tipped toward ceramide, they died. The model shaped a generation of lipid-focused cancer research. Now a large multi-omics study published in the Journal of Experimental & Clinical Cancer Research argues that the field has been looking in the wrong place. The rheostat, the researchers report, is not primarily a tumor-cell survival switch at all. Instead, it operates most powerfully in the stromal and immune cells that surround the tumor, where it appears to orchestrate an immunosuppressive microenvironment that correlates with advanced disease, poor survival and weak predicted responses to checkpoint inhibitor immunotherapy.

The new work, led by Michelle M. Maurin and Siddabasave Gowda B. Gowda with senior author Timothy J. Yeatman, began with a straightforward hypothesis: that the sphingosine-1-phosphate-to-ceramide ratio matters more in the tumor microenvironment than in the malignant epithelial cells that have dominated the literature since the 1990s. To test it, the team performed quantitative liquid chromatography–tandem mass spectrometry, a gold-standard technique for measuring lipid species with high precision, on 480 resected primary colorectal tumors and 94 metastatic colorectal cancer specimens. The comparison revealed a striking biochemical shift. Metastatic lesions were enriched in sphingosine-1-phosphate and its close relative sphinganine-1-phosphate, while a specific ceramide species, ceramide 24:1, was depleted. In other words, the metastatic lipid landscape had tipped decisively toward the pro-survival, pro-signaling arm of the pathway.

Lipid measurements alone could not say which cells were responsible, so the investigators layered on gene-expression analysis at remarkable scale. Quantitative real-time PCR confirmed that enzymes that synthesize sphingosine-1-phosphate and sphinganine-1-phosphate, particularly sphingosine kinase 1, or SPHK1, were elevated in metastatic colorectal cancer compared with primary disease, along with the lipid’s receptors S1PR3, S1PR4 and S1PR5. The team then mined a large gene-expression dataset of 2,373 colorectal cancer samples, combining clinical outcomes with molecular profiles, and applied CIBERSORT, a computational method that estimates the relative abundance of different cell types from bulk RNA transcripts. Public single-cell RNA-sequencing data provided an independent check at single-cell resolution.

The cellular attribution was unambiguous. SPHK1 and its receptor S1PR3 were expressed predominantly in stromal cells and immune cells within the tumor microenvironment, whereas the related kinase SPHK2 predominated in epithelial tumor cells. High-definition spatial transcriptomics added a geographic dimension, showing where in the tumor architecture these programs were active, and immunofluorescence imaging of patient sections revealed SPHK1 staining co-localizing with CD163, a marker of M2 macrophages, in the stroma surrounding tumor nests. The kinase signal sat outside the keratin-positive cancer cells, in the very neighborhoods where immune suppression is thought to be manufactured.

From these data the researchers distilled a three-gene sphingosine-1-phosphate signature and asked what it predicted. The answer was grim but coherent. High signature scores were associated with stromal enrichment, with infiltration by myeloid-derived suppressor cells and M2 macrophages, with advanced stage, with nodal involvement and with poor survival. Perhaps most consequentially, the signature showed weak to negative correlations with validated gene-expression scores that predict response to checkpoint inhibitor therapy, the immunotherapies that have transformed treatment for many cancers but delivered only modest benefits in colorectal cancer, particularly microsatellite-stable disease.

That last finding carries the study’s translational punch. Checkpoint inhibitors work by releasing the brakes on T cells, but they depend on a tumor microenvironment that still permits T cells to function. A lipid program that enriches the stroma with suppressive myeloid cells and M2 macrophages is precisely the kind of barrier that renders such therapies ineffective. The sphingosine-1-phosphate axis, the authors conclude, is not merely a bystander correlate of aggressive disease; it is plausibly a mechanistic contributor to the immunosuppressive state that shields metastatic colorectal cancer from immune attack.

The laboratory experiments in the study reinforce that interpretation. In vitro, the team cultured human macrophages under M1, classically inflammatory, or M2, alternatively activated, conditions and examined the effects of their conditioned media on primary T cells. Flow cytometry measured T cell phenotypes, proliferation, activation markers such as interferon-gamma and CD107a granule mobilization, and cytotoxic killing of tumor target cells. Pharmacological inhibition of sphingosine kinase activity with the SPHK1 inhibitor PF-543 and the S1PR3 antagonist CAY10444 was used to probe whether blocking the axis could rescue T cell function in these co-culture systems, alongside assays of T cell migration toward macrophage-conditioned media.

Conceptually, the study reframes a thirty-year-old model. The original rheostat placed the sphingosine-1-phosphate/ceramide balance inside the tumor cell, where rising sphingosine-1-phosphate was read as a cell-autonomous survival advantage. The new data reposition the dial in the tumor’s supporting ecosystem: stromal cells and myeloid immune cells that generate sphingosine-1-phosphate, signal through receptors such as S1PR3, and thereby help construct a microenvironment hostile to cytotoxic immunity. This stromal-immune program links lipid metabolism to inflammation and immunosuppression simultaneously, explaining why the same pathway that once looked like a survival factor for cancer cells now looks like a communication network between the tumor and its immune entourage.

The clinical implications are significant for a disease with a large unmet need. Colorectal cancer remains a leading cause of cancer death worldwide, and metastatic disease in particular responds poorly to immunotherapy outside the small microsatellite-instability-high subset. If the sphingosine-1-phosphate/ceramide imbalance helps maintain that resistance, then therapeutic strategies aimed at rebalancing the ratio, whether by inhibiting SPHK1 in stromal and myeloid cells, blocking S1PR3 signaling, or otherwise reprogramming the suppressive microenvironment, could sensitize tumors to checkpoint blockade. The three-gene signature itself may also serve as a biomarker, identifying patients whose tumors harbor the suppressive lipid program and who might be candidates for combination approaches.

Cautions remain, as they must in any study moving from association toward mechanism. The human data are largely correlative, drawn from resected specimens and retrospective datasets, and the authors note that the published version was shared early as accepted peer-reviewed research. Sphingolipid biology is intricate, with dozens of ceramide species, multiple kinases and five receptors whose roles differ by context, and the field has previously seen promising lipid-targeted strategies falter in translation. Still, the scale of the analysis, spanning targeted lipidomics in hundreds of tumors, a 2,373-sample expression cohort, spatial transcriptomics, single-cell data and functional co-culture assays, gives the reframing unusual weight. The sphingosine rheostat, first sketched as a switch inside the cancer cell, now appears to be a dial held jointly by the stroma and the immune system, and turning it back may be a route to making metastatic colorectal cancer visible to immunotherapy.

Subject of Research: The sphingosine-1-phosphate/ceramide rheostat as a stromal-immune program driving immune suppression in metastatic colorectal cancer

Article Title: Re-defining the sphingosine rheostat as a stromal-immune program associated with immune suppression in metastatic colorectal cancer

Article References: Maurin, M. M., Gowda, S. G. B., Eksioglu, E. A., Nebozhyn, M. V., Razabdouski, T., Gowda, D., Adapa, S. R., Jiang, R. H. Y., Wang, H., Soundararajan, R., Carr, M., Sundaraswamy, P. M., Alden, A. J., Martinez, C., Bennett, R. D., Chudzinski, A. P., Karachristos, A., Nywening, T. M., Cavallaro, P. M., … Yeatman, T. J. (2026). Re-defining the sphingosine rheostat as a stromal-immune program associated with immune suppression in metastatic colorectal cancer. Journal of Experimental & Clinical Cancer Research. https://doi.org/10.1186/s13046-026-03832-1

Image Credits: AI Generated

DOI: 10.1186/s13046-026-03832-1

Keywords: sphingosine-1-phosphate, ceramide, sphingosine rheostat, metastatic colorectal cancer, tumor microenvironment, SPHK1, S1PR3, M2 macrophages, myeloid-derived suppressor cells, immunosuppression, checkpoint inhibitor resistance, lipidomics