SARS-CoV-2 continues to produce new subvariants, and the landscape shifts every few months. As of the most recent surveillance data, the dominant circulating forms in many countries descend from the JN.1 lineage of Omicron, with names like KP.3.1.1, XEC, and LP.8.1 appearing in hospital sequencing reports. These are not entirely “new strains” in the strict virological sense, but they carry enough mutations to partially sidestep existing immunity, which is why waves of infection keep rolling through even highly vaccinated populations.
What Is Circulating Right Now
The variants making headlines in 2024 and into 2025 are all descendants of JN.1, itself a sublineage of Omicron BA.2.86. The naming has become alphabet soup, but the key players include KP.2, KP.3, KP.3.1.1, XEC, and LP.8.1. These share a common ancestor but differ in a handful of spike protein mutations that affect how well they dodge antibodies and latch onto human cells. A study estimating the effectiveness of the 2024–2025 updated vaccine found that protection against hospitalization varied by subvariant: roughly 49% against KP.3.1.1, about 34% against XEC, and a lower, statistically uncertain estimate against LP.8.1.1JAMA Network Open. Estimated Effectiveness of 2024-2025 COVID-19 Vaccination Against Severe COVID-19 The fact that vaccine effectiveness differs across these subvariants tells you they are meaningfully distinct from one another even though they share a recent common lineage.
XEC is a recombinant, meaning it formed when two different Omicron subvariants infected the same person at the same time and swapped genetic material. LP.8.1, meanwhile, carries additional mutations that researchers are still characterizing. These names change often enough that by the time you read this, a newer subvariant may have overtaken whatever is dominant today. That turnover is itself a defining feature of where SARS-CoV-2 is in its evolution.
Why Scientists Say “Variant,” Not “Strain”
The words “variant” and “strain” get used interchangeably in everyday conversation, but virologists draw a meaningful line between them. Every time SARS-CoV-2 copies itself inside your cells, small errors creep into its genetic code. Viruses carrying those errors are called variants. A variant only earns the label “strain” if it behaves in a physically distinct way: binding to a different receptor, replicating faster, or causing notably different disease. Essentially, all strains are variants, but not all variants are strains.2Journal of Medical Virology. The variants question: What is the problem?
By that definition, most of the subvariants circulating right now are not truly new strains. They belong to the same broad Omicron family and use the same cell-entry receptor. What makes them matter clinically is not a dramatic change in how the virus works, but a steady accumulation of mutations that let them slip past your antibodies more easily.
How the Virus Keeps Reinventing Itself
Three main engines drive the constant appearance of new subvariants: point mutations, convergent evolution, and recombination.
SARS-CoV-2 mutates more slowly than many other RNA viruses because it has a built-in proofreading mechanism that corrects some copying errors. Its mutation rate is roughly one to two errors per million nucleotides per replication cycle, lower than viruses like HIV or hepatitis C that lack such proofreading.3Nature Reviews Microbiology. The evolution of SARS-CoV-2 But because billions of people have been infected and each infection involves trillions of viral copies, even a relatively low error rate generates enormous diversity.
Convergent evolution is a particularly striking pattern. Unrelated lineages around the world independently land on the same mutations, suggesting those changes provide a genuine survival advantage. During the first phase of the pandemic, before mass vaccination, researchers identified 31 spike mutations that evolved independently at least three times across separate lineages, and those mutations included the key changes found in every variant of concern at the time.4PubMed Central. Perspectives: SARS-CoV-2 Spike Convergent Evolution as a Guide to Explore Adaptive Advantage Since Omicron took over, this convergence has intensified at a new set of residues, driven by the selective pressure of widespread immunity from both infection and vaccination.5PubMed Central. Convergent Evolution in SARS-CoV-2 Spike Creates a Variant Soup from Which New COVID-19 Waves Emerge
Recombination adds another layer. When a single person is simultaneously infected with two different SARS-CoV-2 lineages, the viruses can swap chunks of their genomes, producing a hybrid that inherits features of both parents. One study found co-infections with different SARS-CoV-2 clades in about 0.25% of sequenced samples, and roughly a quarter of those co-infected patients harbored recombinant viral populations.6PubMed Central. SARS-CoV-2 Co-Infections and Recombinations Identified by Long-Read Single-Molecule Real-Time Sequencing The XEC subvariant circulating now is a product of exactly this process.
What Makes the Latest Subvariants Different
The mutations getting the most attention in KP.2, KP.3, and related subvariants cluster in the part of the spike protein that physically grabs onto your cells. Two key changes, called L455S and F456L, reduce the spike’s natural stickiness to the ACE2 receptor. That sounds like it should be bad for the virus, but a third mutation at position R346T structurally compensates by reinforcing the shape of the receptor-binding region, restoring the virus’s ability to infect cells.7PubMed Central. Characteristics of JN.1-derived SARS-CoV-2 subvariants SLip, FLiRT, and KP.2 in neutralization escape, infectivity and membrane fusion The net result is a virus that holds onto its infectiousness while gaining a meaningful escape from existing antibodies.
Computational modeling has confirmed that these mutations work together through epistatic interactions, meaning the effect of one mutation depends on which other mutations are present. The L455S and F456L changes impair the ability of certain classes of antibodies to neutralize the virus, while a subsequent change at position Q493 can rescue ACE2-binding affinity that would otherwise be lost.8PubMed Central. AlphaFold2 Modeling and Molecular Dynamics Simulations of the Conformational Ensembles for the SARS-CoV-2 Spike Omicron JN.1, KP.2 and KP.3 Variants This push-and-pull between immune escape and maintaining the ability to infect cells is the central tension driving SARS-CoV-2 evolution right now.
It is also worth noting that mutations outside the spike matter too. Changes in non-spike proteins like NSP3, NSP6, and several accessory proteins can affect how fast the virus replicates, how well it evades innate immune responses like natural killer cells, and how it disrupts the cell’s own protein-making machinery.9PubMed Central. Unique mutations in SARS-CoV-2 Omicron subvariants’ non-spike proteins: Potential impacts on viral pathogenesis and host immune evasion The spike gets the headlines because it is the primary vaccine target, but the virus is evolving across its entire genome.
How Well Do Current Vaccines Work Against These Subvariants
Vaccines do not stop every infection, but they remain the most effective tool against severe disease. The 2024–2025 updated vaccine showed roughly 49% effectiveness against hospitalization from KP.3.1.1 and about 48% against XEC when measured within the first three months after vaccination.1JAMA Network Open. Estimated Effectiveness of 2024-2025 COVID-19 Vaccination Against Severe COVID-19 For context, the previous season’s XBB.1.5-targeted vaccine showed about 54% effectiveness against symptomatic infection at a median of 52 days post-vaccination.10PubMed Central. Early Estimates of Updated 2023-2024 (Monovalent XBB.1.5) COVID-19 Vaccine Effectiveness Against Symptomatic SARS-CoV-2 Infection Attributable to Co-Circulating Omicron Variants Among Immunocompetent Adults
Protection wanes over time, as it does with most respiratory virus vaccines. One study found that the protective benefit of a variant-targeted booster against hospitalization from COVID pneumonia was significant for the first six months but no longer statistically significant between six and twelve months, regardless of whether a person had vaccine-derived immunity alone or hybrid immunity from both vaccination and prior infection.11PubMed Central. Annual Variant-Targeted Vaccination to Prevent Severe COVID-19 in Cohorts With Vaccine-Derived and Hybrid Immunity This is the core rationale behind the annual update cycle: the virus evolves enough and protection fades enough that a yearly reformulated dose provides meaningful benefit.
People with hybrid immunity, meaning they have been both vaccinated and previously infected, generate memory B cells at five- to ten-fold higher levels compared to either infection or vaccination alone, and their protection against symptomatic disease tends to last about six to eight months.12PubMed Central. Hybrid Immunity to SARS-CoV-2 from Infection and Vaccination-Evidence Synthesis and Implications for New COVID-19 Vaccines By this point in the pandemic, most adults in high-vaccination countries have some degree of hybrid immunity, which is part of why severe outcomes are less common now than during the first waves.
Are Current Variants Milder
The general trend with Omicron subvariants has been toward less severe disease compared to the original virus and earlier variants like Delta and Gamma. A study of hospitalized patients in Brazil found that those admitted during Omicron-dominant waves had lower rates of loss of smell, loss of taste, fever, and respiratory distress compared to patients admitted during earlier waves.13PubMed Central. Clinical symptom profile of hospitalized COVID-19 Brazilian patients according to SARS-CoV-2 variants Genomic surveillance data from 2022–2023 in Europe also showed that after adjusting for age, sex, and timing, most later Omicron lineages were associated with equal or decreased odds of hospitalization compared to BA.2.14PubMed Central. Monitoring viral evolution and epidemiological characteristics of SARS-CoV-2 during 2022–2023 using Integrated Genomic Surveillance
But disentangling intrinsic viral severity from population immunity is nearly impossible at this stage. The virus looks “milder” in part because almost everyone now has some immune memory from prior infection, vaccination, or both. For people with weakened immune systems, the elderly, and young children, COVID-19 still lands plenty of people in hospitals. The same European surveillance study found that the highest odds of hospitalization fell in children under five and adults over 60.14PubMed Central. Monitoring viral evolution and epidemiological characteristics of SARS-CoV-2 during 2022–2023 using Integrated Genomic Surveillance
Long COVID Across Different Variants
A common hope has been that milder acute illness means less long COVID, but the evidence is more complicated. A meta-analysis covering over 33,000 patients found that the overall picture of long COVID symptoms did not differ dramatically between variants, though certain specific symptoms did. The Alpha variant was associated with the highest rates of fatigue and general persistent symptoms, while the Omicron variant showed the highest rate of muscle pain compared to the original virus.15PubMed Central. Comparison of Long COVID-19 Caused by Different SARS-CoV-2 Strains: A Systematic Review and Meta-Analysis
Perhaps more sobering, a study that followed patients for a year and a half after infection found that those infected with Omicron and Delta actually reported more severe long COVID symptoms and worse health scores than those infected with the original virus. At the 18-month mark, over half of all patients had failed to improve, with no difference between variants.16PubMed. Long-term prognosis at 1.5 years after infection with wild-type strain of SARS-CoV-2 and Alpha, Delta, as well as Omicron variants The assumption that Omicron equals mild long-term outcomes does not hold up well under scrutiny.
Do Home Rapid Tests Still Detect New Variants
Most rapid antigen tests target the nucleocapsid protein rather than the spike, which is why they have generally continued working as the spike mutated heavily. But “generally” is not “always.” A comprehensive evaluation of 34 commercially available rapid tests against five major variants found that several tests showed reduced performance with certain variants.17PubMed Central. Comparison of the analytical and clinical sensitivities of 34 rapid antigen tests with prevalent SARS-CoV-2 variants of concern during the COVID-19 pandemic in the UK Specific nucleocapsid mutations can impair test sensitivity; one study demonstrated that a single amino acid substitution in the Delta variant completely abolished antigen detection in some tests even at high viral loads.18Frontiers in Virology. Mutations in SARS-CoV-2 nucleocapsid in variants of concern impair the sensitivity of SARS-CoV-2 detection by rapid antigen tests
In practical terms, a negative rapid test when you have symptoms is not as definitive as it once was. Testing twice, 48 hours apart, remains a reasonable approach if you suspect infection and the first test comes back negative. The tests still work well enough for population-level screening, but any individual negative result should be taken with a grain of salt, especially early in an illness when viral load may be low.
How Scientists Track What Is Circulating
Clinical testing has dropped dramatically since the early pandemic, so surveillance increasingly relies on wastewater monitoring. By sequencing viral RNA from sewage, researchers can detect which variants are circulating in a community without needing anyone to take a test. Wastewater surveillance has proven sensitive enough to detect low-abundance variants before they show up in clinical sequencing.19PubMed Central. Detecting SARS-CoV-2 variants in wastewater and their correlation with circulating variants in the communities In Australia, this approach identified the first incursion of Omicron BA.1 into the state of Victoria through sewage testing before clinical cases were widely confirmed.20PubMed Central. Highly sensitive wastewater surveillance of SARS-CoV-2 variants by targeted next-generation amplicon sequencing provides early warning of incursion in Victoria, Australia
More recently, machine-learning approaches applied to wastewater data have shown the ability to detect emerging variants even earlier than conventional computational methods, and can uncover unusual co-varying mutation patterns that do not match any known variant, flagging potentially novel lineages before they receive official names.21PubMed Central. Early detection of emerging SARS-CoV-2 Variants from wastewater through genome sequencing and machine learning This is increasingly how public health authorities know what is out there, even as individual testing becomes rarer.
Where New Variants Come From
Two underappreciated sources of new variants deserve attention: immunocompromised patients and animal reservoirs.
People with weakened immune systems, whether from cancer treatment, organ transplants, or other conditions, can carry active SARS-CoV-2 infections for months. During these prolonged infections, the virus replicates extensively in an environment where immune pressure is present but insufficient to clear it, creating ideal conditions for rapid mutation. Studies have shown that persistent infections in immunocompromised patients trigger the accumulation of an unusually high number of mutations, some of which match the defining changes in variants of concern.22PubMed Central. Long-Term Evolution of SARS-CoV-2 in an Immunocompromised Patient with Non-Hodgkin Lymphoma It is plausible that some major variant jumps, including the original emergence of Omicron, originated in chronically infected individuals.23Clinical Infection in Practice. Persistent SARS-CoV-2 infection in immunocompromised patients facilitates rapid viral evolution: Retrospective cohort study and literature review
Animal populations represent another concern. SARS-CoV-2 has jumped from humans into white-tailed deer, mink, hamsters, and other species. Once circulating in an animal host, the virus can mutate along different evolutionary pathways and potentially spill back into humans carrying novel combinations of changes. The possibility of an animal reservoir producing a variant with unexpected properties remains a genuine concern among virologists studying pandemic preparedness.24PubMed Central. Role of Spillover and Spillback in SARS-CoV-2 Transmission and the Importance of One Health in Understanding the Dynamics of the COVID-19 Pandemic
Mucosal Vaccines and the Next Frontier
All widely used COVID vaccines are injected into muscle, which is good at generating antibodies in the blood but less effective at building immunity right at the surfaces where infection begins: the nose and throat. Mucosal vaccines, delivered as nasal sprays or oral drops, are designed to trigger immune responses directly at these entry points. In theory, a strong mucosal immune response could prevent infection altogether rather than just reducing severity, which would also limit transmission and slow the emergence of new variants.25Vaccine. A review of currently licensed mucosal COVID-19 vaccines
A handful of mucosal COVID vaccines have been approved in countries including China and India, though none have yet gained widespread use in North America or Europe. The scientific rationale is compelling, but the practical challenges are real: delivering a consistent dose to mucosal surfaces is harder to standardize than an intramuscular injection, and measuring mucosal immunity in clinical trials requires different tools than measuring blood antibody levels. Still, if a mucosal vaccine could meaningfully reduce transmission, it would change the calculus on how quickly new variants emerge by shrinking the total pool of viral replication in the population. This is an area where the next few years of research could substantially shift how we manage SARS-CoV-2 going forward.
How Omicron Changed the Way the Virus Enters Cells
One nuance that got a lot of early attention was the idea that Omicron had fundamentally changed which cellular doorway it uses to enter cells, switching from a surface enzyme pathway to an internal one mediated by cathepsins. That finding held up in monkey kidney cell lines commonly used in labs, but when researchers tested Omicron’s cell-entry mechanism in human airway and intestinal tissue models, the picture reversed: Omicron still relied on the same surface enzyme (TMPRSS2) that earlier variants used, not on cathepsins.26PubMed Central. SARS-CoV-2 Omicron entry is type II transmembrane serine protease-mediated in human airway and intestinal organoid models This is a useful reminder that laboratory cell lines do not always reflect what happens in actual human tissue, and that drugs targeting surface serine proteases may still have relevance against current variants despite the earlier reports suggesting otherwise.