The research focuses on two lipids with opposing biological effects: sphingosine-1-phosphate, commonly abbreviated S1P, and ceramide. Under the classical rheostat model proposed roughly 30 years ago, elevated S1P within a cancer cell promotes growth and survival, while higher concentrations of ceramide push the cell toward death. The analogy to a dimmer switch was apt in its original formulation: cancer cells were thought to adjust the relative levels of these two signaling molecules internally, tipping the balance toward uncontrolled division or, conversely, toward self-destruction. Generations of studies have been designed, and many experimental therapies conceived, on the assumption that this regulatory circuit was an intracellular affair.

The new findings upend that assumption. Using advanced spatial mapping technology applied to 574 human colon cancer tissue samples, the team discovered that as colorectal cancer progresses and spreads, levels of the pro-growth lipid S1P rise sharply while levels of the cell-killing lipid ceramide fall significantly. Crucially, however, the molecules responsible for producing S1P and for receiving its survival signals were not concentrated within the malignant cells. Instead, they clustered in the surrounding stromal tissue and in specialized immune cells, most notably macrophages, the scavenger cells that normally serve as first responders of the innate immune system. In other words, the chemical balancing act that determines whether tumors thrive appears to be orchestrated, in substantial part, by the neighborhood around the tumor rather than by the tumor itself.

This spatial redistribution has profound consequences for how metastatic colorectal cancer evades immunotherapy. According to the study, the accumulation of S1P signaling in the tumor microenvironment creates a state of chronic, unhealthy inflammation that effectively walls off cytotoxic T cells, the immune system’s primary tumor-killing weapons. The killer cells are prevented from getting close enough to the malignant cells to recognize and destroy them. Senior author Timothy J. Yeatman, MD, Professor of Surgery at the USF Health Morsani College of Medicine and associate center director for Translational Research and Innovation at the TGH Cancer Institute, framed the shift in thinking directly. For decades, he noted, researchers have thought of this pathway largely in terms of what happens inside the cancer cell and whether the cell survives or dies. What the new data suggest, he said, is that much of the important activity is actually happening in the neighborhood around the tumor, where it creates an environment that prevents the immune system from effectively attacking the cancer.

The technical details of the pathway add an important layer of specificity. The researchers identified the S1P receptor S1PR3 as a key mediator of the immunosuppressive effect. In the language of the paper, elevated S1P and its receptor S1PR3 act as a localized biochemical shield, driving out cytotoxic T cells and creating an immunosuppressive microenvironment. That characterization came from Siddabasave Gowda B. Gowda, Associate Professor at Hokkaido University’s Graduate School of Global Food Resources and Faculty of Health Sciences, who emphasized that targeting this pathway offers a therapeutic strategy to break these barriers, restore immune cell infiltration and potentially make immunotherapies effective for patients with advanced disease. The formulation is significant because it reframes immune evasion not as a passive absence of immune cells but as an actively constructed barrier, built from lipid signaling molecules secreted and sensed within the stromal compartment.

Beyond the lipid measurements, the team uncovered a three-gene signature that functions as a molecular warning sign of aggressive disease. High activity of this gene signature correlated strongly with late-stage cancer, with tumors that had spread into lymph nodes, with lower patient survival rates and with poor responses to standard checkpoint inhibitor immunotherapy. In practical terms, the signature could eventually help clinicians identify which metastatic colorectal cancer patients are least likely to benefit from existing immune therapies and might therefore be candidates for combination approaches that include pathway inhibition. The multiomics approach, integrating lipidomics, spatial transcriptomics and clinical outcome data across hundreds of patient samples, was essential to detecting these associations, since bulk tissue analysis would have averaged away the spatial information that proved decisive.

Laboratory experiments provided the functional proof of concept. When the researchers blocked two key components of the S1P signaling pathway, the effect on the immune landscape was striking: healthy killer T cells returned to the vicinity of the cancer cells and successfully attacked them again. This result demonstrates that the immunosuppressive state is not irreversible. It is maintained by ongoing signaling that can, at least in experimental settings, be interrupted. The authors caution that these are preclinical findings, but they support further investigation into whether pharmacological targeting of the pathway could remodel the immune environment surrounding a tumor and restore the effectiveness of immune-based treatments.

The clinical stakes are considerable. Colorectal cancer is the second-leading cause of cancer-related mortality worldwide, and the prognosis deteriorates dramatically once the disease metastasizes. Five-year survival rates drop to approximately 14 percent when colorectal cancer spreads to other parts of the body. Immune checkpoint inhibitors have transformed outcomes in several tumor types, including melanoma and certain lung cancers, but most metastatic colorectal tumors remain notoriously resistant to these drugs. Explaining that resistance has been a major clinical challenge, and the current study offers a mechanistic answer rooted in the tumor microenvironment rather than in the cancer cells’ own biology. If the S1P/S1PR3 axis can be therapeutically disabled, checkpoint inhibitors might gain the access they currently lack.

The study also carries conceptual weight for the broader field of cancer biology. The rheostat model has influenced how researchers interpret sphingolipid signaling across many cancer types, and the demonstration that a major component of that signaling is stromal and immune-mediated invites a re-examination of assumptions in other tumors as well. The authors describe their work as re-defining the sphingosine rheostat as a stromal-immune program, a phrase that signals a shift from a cell-autonomous view to an ecosystem view of tumor lipid biology. Such reframing matters because drugs designed to hit targets inside cancer cells may fail if the decisive biology resides in adjacent support tissue, a possibility that could help explain past therapeutic disappointments in this pathway.

Funding for the research came from the Tampa General Hospital Foundation, the National Cancer Institute through grants U01CA293474, R35CA197570 and P50CA236733, and the Japan Society for the Promotion of Science through KAKENHI grant 25K00258. The collaboration itself, spanning surgical oncology expertise in Tampa and lipid biochemistry in Hokkaido, reflects the increasingly international and multidisciplinary character of cancer microenvironment research. Whether S1P pathway inhibition can be translated into a safe and effective combination regimen for patients with metastatic colorectal cancer remains to be tested in clinical trials, but the study provides a clear molecular target, a biomarker signature for patient stratification and a mechanistic rationale for why the immune system, in so many of these patients, never gets the chance to fight.

Subject of Research: Sphingosine-1-phosphate signaling in the tumor microenvironment and immune evasion in metastatic colorectal cancer

Article Title: New study finds metastatic colorectal cancer’s immune defenses may begin outside the cancer cell, challenging view held for decades

Article References: New study finds metastatic colorectal cancer’s immune defenses may begin outside the cancer cell, challenging view held for decades. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: metastatic colorectal cancer, sphingosine-1-phosphate, S1PR3, ceramide, tumor microenvironment, macrophages, cytotoxic T cells, immunotherapy resistance, checkpoint inhibitors, spatial mapping, three-gene signature, sphingolipid signaling

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