Highlights
FLT3 inhibitor resistance in FLT3-ITD acute myeloid leukemia frequently emerges through acquired RAS-pathway mutations, particularly NRAS, creating a major barrier to durable disease control.
In this preclinical study, the clinically available sphingosine-1-phosphate receptor modulators fingolimod (FTY720) and mocravimod (KRP-203) restored sensitivity to FLT3 inhibition in multiple NRAS-mutated FLT3-ITD AML models, including primary patient blasts.
Resensitization was associated with ERK inactivation, transcriptional downregulation of SPHK1, and suppression of downstream AKT, p70 S6K, and BAD signaling, supporting a mechanistically coherent combination strategy.
The combination also showed in vivo activity in an orthotopic mouse model, strengthening translational interest for clinical testing in molecularly defined relapsed or resistant AML.
Background and Clinical Context
Acute myeloid leukemia (AML) with internal tandem duplication of FLT3, commonly termed FLT3-ITD AML, represents a biologically aggressive subtype associated with high relapse risk and historically poor outcomes. The development of FLT3 inhibitors has materially changed the therapeutic landscape. Agents such as midostaurin, gilteritinib, quizartinib, and sorafenib have demonstrated important activity in selected settings, and FLT3-directed therapy is now embedded in modern AML management. Even so, responses are often not durable, especially in relapsed or refractory disease, where adaptive and acquired resistance remains the rule rather than the exception.
Among the best recognized mechanisms of resistance are secondary mutations that reactivate signaling downstream of FLT3. RAS-pathway alterations, including NRAS mutations, are particularly relevant because they can bypass the need for upstream FLT3 signaling and sustain leukemia cell survival despite potent receptor inhibition. Clinically, this is highly consequential: patients who initially respond to a FLT3 inhibitor may subsequently relapse with emergent NRAS-mutant subclones, limiting the effectiveness of re-treatment or continued FLT3-targeted therapy.
The present study by Chatterjee and colleagues addresses that problem through a different axis of signaling biology. Sphingosine kinase 1 (SPHK1) phosphorylates sphingosine to generate sphingosine-1-phosphate (S1P), a bioactive lipid that supports cell survival, proliferation, trafficking, and inflammatory signaling. Prior observations suggest that SPHK1 is upregulated and redistributed to the plasma membrane in RAS-mutated cells, raising the possibility that S1P signaling may be a nonredundant collaborator in NRAS-driven FLT3 inhibitor resistance. The therapeutic question is therefore clinically attractive: can modulation of S1P receptor signaling restore drug sensitivity in FLT3-ITD AML with NRAS-mediated resistance?
Study Design
This was a mechanistic and translational preclinical investigation using AML cell lines, patient-derived blasts, and an orthotopic mouse model. The study focused on FLT3-ITD-expressing human AML models with engineered or selected NRAS mutations known to confer resistance to FLT3 inhibitors.
Experimental Models
The authors examined FLT3-ITD AML cell lines MOLM-14 and MV4-11 carrying NRAS variants G12D, G12S, Q61K, or Q61H, with comparison to G12C. They also studied patient AML blasts harboring NRAS mutations including G13D, G13V, and G12D. In vivo, they used G12D NRAS-mutated M14(R)701 cells in an orthotopic mouse model to test whether pharmacologic S1P receptor modulation could restore gilteritinib sensitivity.
Interventions and Comparators
Cells were treated with FLT3 inhibitors alone or in combination with the S1P receptor modulators fingolimod (FTY720) or mocravimod (KRP-203). The core comparison was whether adding an S1P receptor modulator could overcome resistance observed with FLT3 inhibition alone in NRAS-mutated FLT3-ITD AML cells.
Endpoints
The investigators evaluated signaling and antileukemic effects using immunoblotting, cytotoxicity assays, apoptosis assays, and colony formation studies. To explore mechanism, they measured downstream RAS and SPHK1 effectors by immunoblotting and qRT-PCR. The in vivo endpoint was restoration of antileukemic efficacy of gilteritinib in the orthotopic model.
The abstract does not report detailed quantitative effect sizes, hazard ratios, confidence intervals, or formal toxicity datasets. Accordingly, interpretation should remain centered on directionality, biological plausibility, and translational potential rather than on estimates of clinical magnitude.
Key Findings
1. S1P receptor modulators reversed FLT3 inhibitor resistance in several NRAS-mutated FLT3-ITD AML models
The principal finding is that fingolimod and mocravimod resensitized NRAS-mutated FLT3-ITD AML cells to FLT3 inhibitors. This effect was observed in human MOLM-14 and MV4-11 cells expressing NRAS G12D, G12S, Q61K, or Q61H mutations. Importantly, the benefit was not universal across all NRAS substitutions: the abstract specifically notes lack of resensitization in G12C-mutated cells. That genotype specificity is clinically notable because it suggests the resistance biology may differ by NRAS codon and amino acid substitution rather than simply by the presence or absence of any NRAS mutation.
The investigators also demonstrated resensitization in primary patient blasts carrying NRAS G13D, G13V, or G12D mutations. Inclusion of patient-derived samples materially strengthens the translational relevance of the work because many resistance mechanisms that look compelling in cell lines fail to reproduce in primary AML material.
2. The combination showed in vivo efficacy in an orthotopic AML model
In vivo confirmation is an important bridge between signaling biology and therapeutic plausibility. In this study, co-treatment with fingolimod resensitized G12D NRAS-mutated M14(R)701 cells to gilteritinib in an orthotopic mouse model. Although the abstract does not specify survival curves, leukemia burden measurements, or pharmacodynamic kinetics, the finding indicates that the combination retained activity in a more physiologic setting than in vitro culture alone.
That result is particularly relevant because fingolimod is already clinically used in other fields, which may lower some of the development barriers typically associated with entirely novel compounds. However, its immune effects and safety profile in heavily pretreated AML populations would still require careful re-evaluation.
3. Mechanistically, the combination suppressed ERK and downregulated SPHK1
The mechanistic data are central to the paper’s contribution. Co-treatment inactivated ERK, a canonical downstream mediator of RAS signaling. At the same time, the combination transcriptionally downregulated SPHK1, suggesting that S1P pathway modulation is not merely a parallel phenomenon but part of a functionally linked signaling circuit in resistant cells.
This is biologically important because NRAS-driven resistance often sustains MAPK pathway activity despite FLT3 blockade. If S1P receptor modulation helps collapse that compensatory signaling, it creates a rational means of re-establishing dependence on FLT3-targeted therapy.
4. Downstream survival signaling through AKT, p70 S6K, and BAD was also inhibited
Beyond ERK, the combination also inactivated AKT, p70 S6K, and BAD, indicating broader suppression of prosurvival signaling. This matters because resistant AML clones are rarely sustained by a single pathway node. Successful resensitization typically requires coordinated attenuation of multiple survival circuits, including MAPK and PI3K/AKT/mTOR-associated pathways.
The authors further report that constitutive SPHK1 expression abrogated this inactivation. This is a strong mechanistic observation because it supports a causal role for SPHK1 rather than a simple association. In other words, restoring SPHK1 signaling appears sufficient to blunt the downstream effects of S1P receptor modulator co-treatment, reinforcing the importance of the SPHK1-S1P axis in this resistance state.
5. The response appears mutation-selective rather than universal
The lack of benefit in G12C-mutated cells is more than a footnote. It suggests that NRAS-mutated FLT3-ITD AML may need further molecular stratification before this strategy can be optimally applied. Such mutation-selective behavior could reflect differences in membrane localization, GTPase cycling, effector preference, or interactions with SPHK1 and receptor signaling. For clinicians and trialists, this means a future study should not simply enroll “RAS-mutant” AML broadly without attention to the exact NRAS allele.
Expert Commentary
This study addresses a very practical therapeutic problem in FLT3-ITD AML: how to target relapse biology after emergence of NRAS-mediated resistance. The work is compelling for several reasons. First, it uses clinically relevant agents rather than exclusively experimental compounds. Second, it spans cell lines, patient blasts, and an orthotopic in vivo model. Third, the authors go beyond phenotypic resensitization and provide a coherent mechanism involving ERK suppression, SPHK1 downregulation, and collapse of AKT-p70 S6K-BAD signaling.
The translational appeal of fingolimod and mocravimod deserves particular attention. Fingolimod is a well-known S1P receptor modulator with established clinical use in multiple sclerosis, and mocravimod has also been studied in immune-mediated settings. Drug repurposing can accelerate clinical development, especially when there is already human pharmacology experience. In AML, however, prior experience in neurology or transplantation does not automatically translate to safety in patients receiving myelosuppressive therapy. These agents can affect lymphocyte trafficking, heart rate, infection risk, and other systems that matter greatly in leukemia care.
Several limitations should temper interpretation. The report available here is an abstract, not the full article, so exact numerical results, replicate structure, dose-response details, and safety observations are not available for close scrutiny. Without those data, it is not possible to judge the magnitude of synergy, the therapeutic window, or whether the combination is equally effective across different FLT3 inhibitors and NRAS alleles. In addition, preclinical resensitization does not guarantee clinical efficacy, particularly in AML where marrow microenvironment effects, clonal heterogeneity, and co-occurring mutations can markedly alter drug response.
Another unresolved issue is whether S1P receptor modulation would be best deployed at overt resistance, at molecular relapse, or upfront in patients at high risk for RAS-pathway escape. Combination therapy earlier in treatment could, in theory, delay clonal selection, but it could also increase toxicity and immunologic complications. Biomarker development will therefore be essential. At minimum, future trials should incorporate serial molecular profiling for FLT3, NRAS, KRAS, PTPN11, and related signaling alterations, alongside SPHK1 expression or pathway readouts if technically feasible.
Clinically, the most immediate niche for this strategy may be relapsed or refractory FLT3-ITD AML after or during gilteritinib exposure with documented NRAS-mutant emergence. This is a setting with limited options and high unmet need. If the signal proves reproducible, a molecularly selected early-phase study could be justified.
Context Within Current FLT3-Targeted Therapy
FLT3 inhibition is already a standard component of AML treatment in several scenarios. Midostaurin added to induction and consolidation improved overall survival in newly diagnosed FLT3-mutated AML in the RATIFY trial. Gilteritinib improved survival over salvage chemotherapy in relapsed or refractory FLT3-mutated AML in the ADMIRAL trial. Quizartinib has also shown benefit in newly diagnosed FLT3-ITD AML when added to intensive chemotherapy. Yet across these advances, on-target and bypass resistance remains common.
RAS/MAPK pathway activation is among the best documented bypass routes. The current study fits well with that broader literature by proposing that the S1P-SPHK1 axis is not just another parallel pathway, but a therapeutically actionable dependency in at least a subset of NRAS-mutated resistant leukemias. If validated, this would move the field one step closer to mechanism-based salvage combinations rather than empiric layering of drugs.
Implications for Research and Practice
For practicing hematologist-oncologists, this study is not practice changing today, but it is hypothesis generating in an important and clinically recognizable resistance state. Molecular relapse profiling in FLT3-ITD AML is already increasingly relevant, and these findings reinforce the value of identifying emergent NRAS mutations rather than treating post-FLT3 inhibitor relapse as biologically uniform.
For researchers, several next steps are clear. Prospective work should define which NRAS alleles predict response, clarify whether the effect extends to other RAS-pathway lesions, and test combination schedules that balance efficacy with tolerability. Correlative studies should determine whether SPHK1 expression, localization, or transcriptional state can serve as a predictive biomarker. Given the known immunologic effects of S1P receptor modulators, studies in immunocompetent or more clinically representative models would also be informative.
Ultimately, the most relevant translational milestone would be an early-phase clinical trial enrolling patients with FLT3-ITD AML and molecular evidence of NRAS-mediated resistance, ideally with mandatory serial sequencing and pharmacodynamic sampling.
Conclusion
Chatterjee and colleagues provide persuasive preclinical evidence that S1P receptor modulation can restore FLT3 inhibitor sensitivity in NRAS-mutated FLT3-ITD AML. Fingolimod and mocravimod resensitized resistant AML cell lines and patient blasts, and fingolimod enhanced gilteritinib activity in an orthotopic mouse model. Mechanistically, the strategy appears to work through ERK inactivation, SPHK1 downregulation, and suppression of AKT-centered survival signaling.
The findings are notable because they address a common resistance mechanism with clinically recognizable molecular biomarkers and with agents that already have human use experience outside AML. At the same time, the current evidence remains preclinical, and key questions about allele specificity, toxicity, and clinical magnitude of benefit remain open. The study should be viewed as a strong translational rationale for biomarker-driven early-phase trials rather than as immediate support for off-label routine use.
Funding and ClinicalTrials.gov
The abstract provided does not report funding sources or a ClinicalTrials.gov registration number. Because this is a preclinical study rather than a clinical trial, a registration number may not apply. Readers should consult the full publication for complete funding and disclosures.
Citation and References
Primary article: Chatterjee A, Mustafa Ali MK, Bailey CM, Liu Y, Small D, Smith CC, Traer E, Wang Y, Silvestri G, Baer MR. Sphingosine-1-phosphate receptor modulators resensitize FLT3-ITD acute myeloid leukemia cells with NRAS mutations to FLT3 inhibitors. Leukemia. 2026-06-02. PMID: 42230959. Available at: https://pubmed.ncbi.nlm.nih.gov/42230959/
Relevant background references:
Stone RM, Mandrekar SJ, Sanford BL, et al. Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 mutation. N Engl J Med. 2017;377(5):454-464.
Perl AE, Martinelli G, Cortes JE, et al. Gilteritinib or chemotherapy for relapsed or refractory FLT3-mutated AML. N Engl J Med. 2019;381(18):1728-1740.
Erba HP, Montesinos P, Kim HJ, et al. Quizartinib plus chemotherapy in newly diagnosed FLT3-ITD acute myeloid leukemia. N Engl J Med. 2023;389(23):2133-2146.
Daver N, Schlenk RF, Russell NH, Levis MJ. Targeting FLT3 mutations in AML: review of current knowledge and evidence. Leukemia. 2019;33(2):299-312.
Pyne NJ, Pyne S. Sphingosine 1-phosphate and cancer. Nat Rev Cancer. 2010;10(7):489-503.