What Is an SRSF2 Mutation and Its Role in Disease?

SRSF2 is a protein that helps cells correctly process their genetic instructions, and mutations in the gene that encodes it are found in a substantial fraction of blood cancers, particularly myelodysplastic syndromes (MDS) and chronic myelomonocytic leukemia (CMML). These mutations don’t simply switch the protein off. Instead, they change the way it reads RNA, causing it to mishandle specific genetic messages in blood-forming stem cells. The consequences ripple outward: genes that should be active get silenced, DNA damage accumulates, and the affected blood cells gain a competitive edge that can eventually tip toward leukemia.

What SRSF2 Normally Does

SRSF2 belongs to a family of proteins called serine/arginine-rich splicing factors. Its day job is binding to stretches of RNA and helping the cell decide which segments of a gene’s message to keep and which to cut out, a process called splicing. Almost every human gene produces an RNA transcript that needs editing before it can be translated into a working protein, and SRSF2 is one of the key editors. It recognizes short sequence motifs in the RNA and recruits the cellular machinery that performs the cuts and joins.

Beyond splicing, SRSF2 has roles in transcription itself. It helps RNA polymerase, the enzyme that reads DNA into RNA, move smoothly along the gene. It also plays a part in maintaining the integrity of the genome during DNA replication. These additional functions matter because they explain why a mutation in a “splicing factor” can cause problems that go well beyond misassembled RNA messages.

The P95 Hotspot and How It Changes RNA Recognition

Nearly all disease-linked SRSF2 mutations cluster at a single amino acid: proline at position 95. The most common change swaps proline for histidine (P95H), though substitutions to leucine, arginine, alanine, and threonine also occur at the same spot. In one study of 275 CMML patients, 120 of the 129 who carried an SRSF2 mutation had a missense change at Pro95.¹ A small number of cases carried insertions or deletions in the same region, all predicted to damage the protein’s structure.

The normal SRSF2 protein recognizes certain short RNA sequences and binds to them to mark exons for inclusion in the final transcript. The P95 mutation doesn’t destroy this binding ability. Instead, it shifts the protein’s preferences, making it latch onto RNA motifs it would ordinarily ignore while loosening its grip on motifs it usually favors.² The result is that hundreds of exons across the genome get spliced differently than they should. Some exons that belong in the finished message get skipped; others that should be excluded get left in.

How Missplicing Drives Disease

One of the most important downstream targets of SRSF2-mutant missplicing is a gene called EZH2, which encodes a protein involved in silencing other genes during blood cell development. When SRSF2 is mutated, EZH2’s RNA transcript gets spliced incorrectly, producing a defective message that the cell destroys through a quality-control system called nonsense-mediated decay. The net effect is a loss of EZH2 protein, which impairs the normal maturation of blood cells.³ EZH2 is itself a known tumor suppressor in blood cancers, so knocking it down through missplicing is one concrete path from an SRSF2 mutation to disease.

EZH2 is not the only victim. The altered splicing preferences of mutant SRSF2 affect many genes involved in blood-cell regulation. The missplicing is not random; it follows predictable patterns dictated by the changed RNA-binding preferences of the mutant protein.⁴ That predictability is what distinguishes SRSF2 mutations from a generic loss of splicing accuracy: the mutation creates a specific, reproducible pattern of gene misregulation.

R-Loops, DNA Damage, and Genome Instability

Splicing is only part of the story. Research has shown that mutant SRSF2 triggers the accumulation of structures called R-loops, three-stranded tangles of DNA and RNA that form when freshly made RNA threads back onto the DNA template instead of peeling away. A normal copy of SRSF2 actually suppresses R-loops, but the mutant form does the opposite, amplifying them across the genome.⁵

R-loops are a problem because they stall the machinery that copies DNA before cell division, creating what researchers call replication stress. That stress can lead to DNA breaks. Cells with mutant SRSF2 activate emergency DNA-damage-response pathways in an attempt to cope, but the chronic activation of these pathways can itself become destabilizing over time.⁶ The R-loop problem links SRSF2 mutations mechanistically to the chromosomal instability often seen in MDS and related cancers.

Separate work has found that deleting SRSF2 entirely or making both copies of the gene carry the P95H mutation causes extensive DNA damage and cell-cycle arrest. The protein appears to help regulate the transcription of genes needed for DNA replication and repair, a function that operates independently of its splicing role. When only one copy of the gene is mutated (the situation in most patients), DNA damage still accumulates but is not severe enough to halt the cell cycle, allowing the damaged cells to keep dividing.⁷ That partial impairment may be part of what lets mutant clones persist and gradually accumulate additional mutations.

SRSF2 Mutations in Myelodysplastic Syndromes

MDS is the disease most thoroughly studied in connection with SRSF2 mutations. In large patient cohorts, roughly 15 to 20 percent of MDS patients carry an SRSF2 mutation. One study of 233 patients found the mutation in about 15 percent of cases; in another cohort of 111 patients with detectable mutations, SRSF2 was the second most commonly mutated gene at 19 percent, trailing only SF3B1.⁸

The mutation tends to show up alongside others. In one cohort, more than 85 percent of SRSF2-mutated MDS patients carried at least one additional gene mutation, with RUNX1, IDH2, and ASXL1 being especially common co-mutations.⁹ SRSF2 mutations were also more common in men and in older patients in this study.

A meta-analysis pooling data from ten cohort studies and nearly 1,900 MDS patients found that SRSF2 mutations were associated with shorter overall survival and a higher likelihood of transformation to acute myeloid leukemia. That adverse impact held even among patients in the lower-risk categories of MDS, where the expectation is for relatively indolent disease.¹⁰ A cross-entity study confirmed the survival gap: median overall survival was roughly 2.8 years in SRSF2-mutated MDS patients versus 5.6 years in those without the mutation.¹¹

Chronic Myelomonocytic Leukemia

If SRSF2 mutations are common in MDS, they are a hallmark of CMML. Studies consistently report SRSF2 mutation rates between 25 and 47 percent in CMML cohorts, making it the most frequently mutated splicing factor gene in this disease.¹ A more recent analysis of national and international cohorts found SRSF2 mutation rates of about 37 to 39 percent.¹²

The prognostic picture in CMML is murkier than in MDS. One meta-analysis found that while SRSF2 mutations clearly predicted worse outcomes in MDS, they did not significantly affect overall survival in CMML.¹³ A smaller study did report SRSF2 as an independent negative predictor for both overall and progression-free survival in CMML, though the confidence intervals were wide.¹⁴ The inconsistency may reflect the fact that CMML is a heterogeneous disease with many competing genetic drivers, making it harder to isolate the impact of any single mutation.

Acute Myeloid Leukemia and Myeloproliferative Neoplasms

SRSF2 mutations are also found in AML, particularly in cases that evolve from a prior MDS or myeloproliferative neoplasm (MPN) rather than arising on their own. In AML arising from MPNs, SRSF2 mutations were detected in about 19 percent of patients, a rate significantly higher than in AML arising de novo (roughly 6 percent) or after MDS (about 5 percent). Patients with MPN-derived AML who carried SRSF2 mutations had worse survival, with a hazard ratio of about 2.1 in multivariate analysis.¹⁵

In newly diagnosed AML patients treated with the combination of venetoclax and azacitidine, a regimen that has become standard for older patients who cannot tolerate intensive chemotherapy, SRSF2 mutations were associated with poorer overall and leukemia-free survival.¹⁶ This is a finding with real clinical relevance, since venetoclax-azacitidine is now one of the most widely used AML regimens.

When SRSF2 mutations co-occur with ASXL1 mutations, the resulting AML tends to display features that overlap with CMML, including a monocytic appearance and a tendency to carry additional mutations in TET2, STAG2, and IDH genes. TET2 was mutated in 63 percent of these co-mutated cases versus 19 percent of AML cases without the co-mutation.¹⁷ The overlap suggests these diseases share underlying biology driven partly by SRSF2-related splicing disruption.

Clonal Hematopoiesis and Early Detection

Not every SRSF2 mutation is found in someone who already has cancer. SRSF2 is one of several genes whose mutations can be detected in the blood of apparently healthy people, a phenomenon called clonal hematopoiesis of indeterminate potential (CHIP). In CHIP, a single blood-forming stem cell acquires a mutation that gives it a growth advantage, and its descendants gradually come to make up a measurable fraction of a person’s blood cells.¹⁸

CHIP is increasingly common with age and is associated with a modestly elevated risk of eventually developing a blood cancer, as well as higher rates of cardiovascular disease and other age-related conditions. Among splicing factor mutations found in CHIP, SRSF2 is one of the more worrying ones because it is classified as a high-risk mutation for progression to an overt myeloid neoplasm.¹⁹ The intermediate stage between CHIP and full-blown MDS, sometimes called clonal cytopenia of undetermined significance (CCUS), is where SRSF2 mutations often first become clinically significant and where close monitoring is most warranted.

Beyond Blood Cancers

SRSF2’s best-studied disease connections are in blood cancers, but it has roles elsewhere. In advanced systemic mastocytosis, a rare disease driven by mast-cell overgrowth, SRSF2 P95 hotspot mutations are found almost exclusively in patients who also have an associated blood cancer, linking the mutation to the non-mast-cell component of the disease.²⁰ Comprehensive profiling of advanced mastocytosis patients confirmed that SRSF2, along with TET2 and ASXL1, was among the most frequently mutated genes in this setting.²¹

In solid tumors, the picture is different. Rather than point mutations at position 95, some lung cancers show overexpression of the normal SRSF2 protein. Research in lung adenocarcinoma has found that too much SRSF2 increases global transcription and replicative stress, leading to DNA double-strand breaks. The overexpressed protein also rewires DNA repair, favoring one repair pathway over another in ways that may promote tumor progression.²² This is a reminder that the same protein can contribute to cancer through entirely different mechanisms in different tissues: loss-of-function missplicing in blood cells versus gain-of-function overexpression in lung cells.

How SRSF2 Mutations Are Detected

Because SRSF2 mutations cluster at a single hotspot, they are relatively straightforward to detect with modern sequencing. Most hematology centers now use next-generation sequencing panels that cover a set of genes commonly mutated in myeloid cancers, and SRSF2 is a standard inclusion on these panels.²³ The mutation’s stability over the course of disease is another practical advantage: once detected, it tends to persist rather than disappearing and reappearing, which makes it a reliable marker for tracking a patient’s clone over time.

For patients with unexplained blood count abnormalities, the detection of an SRSF2 mutation can sometimes tip the diagnostic balance. A borderline MDS case that might otherwise be chalked up to age-related changes becomes more convincing when molecular testing reveals a known pathogenic mutation. The same applies to distinguishing CMML from reactive monocytosis: finding an SRSF2 mutation in the right clinical context adds strong evidence for a clonal, neoplastic process.

Therapeutic Implications and Emerging Strategies

There is no approved drug that specifically targets mutant SRSF2, but the mutation’s biology has opened up several therapeutic angles. One of the most explored ideas takes advantage of the fact that SRSF2 mutations always occur in a heterozygous state, meaning cells keep one normal copy of the gene alongside the mutant copy. Research has shown that the normal copy is essential for survival of the mutant cell, and that further disrupting splicing in these cells is disproportionately lethal compared to cells with two normal copies. In mouse models of SRSF2-mutant leukemia, treatment with the splicing inhibitor E7107 preferentially killed the mutant cells.²⁴ This kind of “synthetic lethal” strategy, where you exploit a vulnerability that exists only because of the mutation, is a promising concept, though clinical development of splicing modulators has been slow.

Another line of research has focused on enzymes called protein arginine methyltransferases, which modify splicing factors and are needed for their function. PRMT5 inhibitors, now in clinical trials for various cancers, have shown activity against spliceosomal-mutant leukemias. Work with SRSF2-mutant cells identified PRMT1 as a particularly important target within the broader family of type I arginine methyltransferases.²⁵

In the meantime, the standard treatment backbone for many SRSF2-mutant patients remains hypomethylating agents like azacitidine or decitabine, often combined with venetoclax in AML. There is an interesting wrinkle here: one study of high-risk MDS and secondary AML patients treated with azacitidine found that those carrying SRSF2 mutations actually had better overall and progression-free survival compared to SRSF2-wild-type patients.²⁶ Whether that reflects something about how azacitidine interacts with the splicing defect, or whether it reflects the broader mutation profile of those patients, remains unclear. It does underscore that the prognostic impact of an SRSF2 mutation can shift depending on the treatment context, a reality that makes blanket statements about the mutation being “good” or “bad” unreliable.

Why SRSF2 Mutations Are Always Heterozygous

One detail that researchers have puzzled over is why SRSF2 mutations in cancer are never found in a homozygous state. Patients always retain one normal copy of the gene. The reason appears to be straightforward: losing SRSF2 function entirely is lethal to the cell. Laboratory experiments have confirmed that deleting both copies or making both carry the P95H mutation causes so much DNA damage that cells stop dividing. The single mutant copy is enough to create a competitive advantage through altered splicing and other mechanisms, but a double dose overwhelms the cell’s ability to cope. This built-in ceiling on how far the mutation can go is exactly what makes it a potential therapeutic target: the mutant cell is balanced on a knife’s edge, and pushing it further in the same direction could be enough to kill it.