SARS-CoV-2 と COVID-19 に関する備忘録 Vol.51

投稿日投稿者しとらす

SARS-CoV-2 と COVID-19 に関するメモ・備忘録

◆Virological characteristics of the SARS-CoV-2 NB.1.8.1 variant【THE LANCET : Infectious Diseases　2025年6月6日】

Abstract

After the spread of SARS-CoV-2 JN.1, its subvariants, such as KP.3 (JN.1.11.1.3) and KP.3.1.1 (JN.1.11.1.3.1.1), and XEC (a recombinant lineage of two JN.1 subvariants), emerged and rapidly spread globally. Subsequently, LP.8.1 (JN.1.11.1.1.1.3.8.1), a descendant lineage of KP.1.1.3 (JN.1.11.1.1.1.3), accounts for approximately 30% of all global infections as of April, 2025, as per data from Nextstrain. Thereafter, NB.1.8.1 (XDV.1.5.1.1.8.1), a descendant lineage of XDV, has started to spread worldwide. XDV is a recombinant lineage of XDE (a recombinant lineage of GW.5.1 [XBB.1.19.1.5.1] and FL.13.4 [XBB.1.9.1.13.4]) and JN.1. NB.1.8.1 has acquired seven spike substitutions and 23 non-spike substitutions compared with JN.1. Compared with the XEC spike protein, the NB.1.8.1 spike bears four substitutions: G184S, K478I, A435S, and L1104V. We estimated the relative effective reproduction number (Re) of NB.1.8.1 using a Bayesian multinomial logistic model1–4 based on genome surveillance data from Singapore, Hong Kong, Australia, and the USA, where this variant has spread as of April, 2025. The Re of NB.1.8.1 was 1·17-fold higher than that of LP.8.1 in Singapore, suggesting its potential to outcompete the other major SARS-CoV-2 lineages.
We then assessed the properties of NB.1.8.1 by virological experiments. A lentivirus-based pseudovirus assay showed that the infectivity of NB.1.8.1 was significantly higher (2·5-fold; p<0·0001) than that of LP.8.1. Among the substitutions detected in the NB.1.8.1 spike, K478I and L1104V significantly decreased the infectivity of the XEC pseudovirus. A neutralisation assay using convalescent sera after XEC infection and vaccine sera with the JN.1 mRNA vaccine showed that the 50% neutralisation titre is similar among XEC, LP.8.1, and NB.1.8.1 variants. Overall, although NB.1.8.1 does not show enhanced humoral immune evasion ability compared with LP.8.1, NB.1.8.1 showed higher pseudovirus infectivity than LP.8.1. In line with our data, a previous study showed that the receptor binding domain of NB.1.8.1 binds more strongly to human ACE2 receptor than that of LP.8.1. Therefore, it is plausible to assume that the increased fitness of NB.1.8.1 is due to its augmented infectivity rather than its increased immune evasion. However, in a Chinese cohort, NB.1.8.1 evaded the humoral immunity induced by BA.5 and JN.1 infections more efficiently than LB.8.1. Given that NB.1.8.1 has more than 20 substitutions in non-spike regions compared with LP.8.1 and XEC, it would be warranted to further investigate the virological features of NB.1.8.1 using live isolated virus.

◆Mobius: Mixture-Of-Experts Transformer Model in Epigenetics of ME/CFS and Long COVID【bioRxiv　2025年5月27日】

Abstract

Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) and Long COVID are chronic debilitating post-infectious illnesses that collectively affect up to 470 million individuals. Unlike illnesses of comparable scale, there are no validated blood or imaging tests for the clinical diagnosis of these conditions. Currently, these conditions are diagnosed through clinical exclusion, resulting in approximately 90% of ME/CFS patients being incorrectly diagnosed as Long COVID patients. This misdiagnosis contributes to delayed care and millions of dollars in healthcare burdens. We present Mobius, a transformer-based model that uses autoencoder-derived features from blood DNA methylation to distinguish ME/CFS, Long COVID, and healthy controls. Using 852 samples from 14 distinct datasets, our method employs three innovations: (i) self-supervised masked pretraining to learn epigenetic patterns, (ii) a sparsely-gated mixture-of-experts architecture to handle heterogeneous data, and (iii) an adaptive computation time mechanism for dynamic inference. Mobius achieved 97.06% accuracy (macro-F1 0.95, AUROC 0.96), outperforming current symptom-based diagnostics (58%) and baseline models such as XGBoost (82%). Ablation experiments showed that pretraining added 6% accuracy and that the gating and adaptive depth contributed an additional 7%. Our open-source pipeline could enable a much-needed objective blood test for these conditions and guide targeted precision medicine therapies.

◆Post-Covid-19 Symptoms in Children: A Cross-Sectional Survey【WILEY Online Library　2025年6月9日】

Abstract

Aim

To assess children’s self or parental-rated health following Aotearoa New Zealand’s (NZ) first widespread community transmission of SARS-CoV-2 in February 2022.

Methods

This cross-sectional study recruited participants aged 3–20 years who had consented to be contacted after taking part in the NZ Health Survey. Participants over 15 years or guardians of younger children were surveyed by telephone between November 2022 and April 2023. Ordinal logistic regression was used to estimate the effect of Covid-19 infection on self-reported health.

Results

The study included 4264 children and young adults, with 70.6% reporting having tested positive for Covid-19 at least once (via PCR or RAT test). Almost one-quarter (24.5%) reported more frequent coughs, colds and stomach bugs after Covid-19 infection. One-fifth reported headaches (21.7%), fatigue (20.6%) or stomach aches (14.6%) and 13.1% reported anxiety that was new since having Covid-19. At baseline, there were no significant differences between the self-rated health status of those children who later had Covid-19 and those who did not (p = 0.5274). Following widespread SARS-CoV-2 transmission, those who reported having had Covid-19 were significantly more likely to report a poorer health status than those who did not (p < 0.0001). Conclusions

Greater than one-fifth of NZ children reported persisting symptoms after Covid-19 infection. Post-Covid-19 symptoms impacted the quality of children’s day-to-day lives. Preventing infection is key to preventing post-Covid-19 symptoms.

◆Still Not Feeling the Same After COVID-19? You’re Not Alone【TIME　2025年6月12日】

Most people have put the COVID-19 pandemic behind them. Infections, vaccinations, or a combination of both have bolstered people’s immunity, and while new variants continue to pop up, getting sick does not induce the same panic it once did.

But a new study shows that recovery from COVID-19 might not be as quick or straightforward as most of us now expect. The study, published in Open Forum Infectious Diseases, found that on average, it takes many people up to three months to return to good physical health after a COVID-19 infection, and nine months to recover good mental well-being. For up to 20% of infected people who were analyzed in the study, this mental-health recovery took even longer: up to a year or more.

Lauren Wisk, assistant professor in the division of general internal medicine and health services research at the David Geffen School of Medicine at University of California, Los Angeles, and her team looked at data from people who had COVID-19 at eight health facilities across the U.S. from Dec. 2020 to Aug. 2022. People were asked to fill out surveys every three months for one year about their recovery, recording physical and mental symptoms like anxiety, depression, fatigue, social participation, sleep disturbances, and pain.

It took people far longer to regain their mental well-being than it did their physical health. “To be totally honest, we didn’t necessarily expect to see different recovery trajectories as big as the ones we are seeing,” says Wisk. “While it makes sense that some people recover faster physically, and other people recover faster mentally, on average the difference that we saw was surprising.”

Wisk and her team also asked people to self-report if they experienced Long COVID, meaning symptoms stemming from their infection that lingered for at least three months. Nearly half of people who reported both poor physical and mental qualify of life following their infections also believed they had Long COVID. While the assessment was subjective, it tracked with the data Wisk’s team collected; among people who reported just poor physical health, poor mental health, or neither, there were fewer reports of Long COVID.

The findings point to the need for a deeper understanding of how COVID-19 infections affect the body, physically and mentally, in the short and long term, says Wisk. “We need to be thinking about a longer road to recovery for people, because even if someone recovers physically from their symptoms, it might not end there for them.”

Appreciating these longer lasting effects could help people seek treatment for their symptoms, which may condense their recovery period. Wisk says that short courses of anxiety medications and sleep therapies, for example, could address some of the lingering effects of COVID-19.

“We know how to treat the initial infection and how to keep people alive, but we don’t have a great treatment protocol for the after effects and the lingering symptoms,” says Wisk. “These data should help to guide development of protocols in which we think of recovery over a potentially long time horizon before people get back to normal.”

◆Life insurers grapple with post-Covid mortality spike, see 25% decline in insurance coverage【Moneycontro : Malvika Sundaresan　2025年6月11日】

Mortality claims for top life insurance companies have remained persistently high between FY20 and FY25 – hovering near one million annually.

This marks an almost 75 percent increase in death claims over five years, a sharp rise for an industry that fundamentally relies on stable mortality patterns.

Despite the spike in mortality claims, insurance companies saw a 25 percent decline in the number of lives covered by top private life insurers. This decline is primarily due to major insurers reducing their participation in the group insurance segment, industry sources said.

“This is a crisis,” said Sridhar Sivaram, President, Enam Holdings.

He pointed out that in FY24, the mortality rate for HDFC Life, ICICI Prudential and SBI Life is actually higher than the COVID-19 peak, Sivaram pointed out.

According to data from the Insurance Regulatory and Development Authority of India (IRDAI), mortality claims rose to 1.085 million in FY25, up from 1.009 million in FY24 and significantly higher than the 623,000 reported in FY20.

This marks an almost 75 percent increase in death claims over five years, a sharp rise for an industry that fundamentally relies on stable mortality patterns.

“Life insurance as a sector is built on actuarial science,” said a life insurer who chose to stay anonymous, “where pricing and risk assessments depend on the assumption that death rates change gradually and predictably over time.”

When that assumption is disrupted, as it has been in recent years, the entire pricing model for protection products like term life is thrown off balance, the insurer said.

The surge began in FY21, with claims rising by nearly 50 percent from the previous year to 930,000, largely due to the initial COVID-19 wave. FY22 saw the highest volume, with 1.17 million claims during the devastating Delta variant outbreak.

While claims receded slightly in the following years, they remained well above pre-pandemic levels, with FY25 still recording over a million deaths covered under life insurance.

This suggests that life insurers are now covering fewer lives, but are paying out more death claims, indicating a concentrated risk pool.

“Unfortunately, there is no comprehensive data available on mortality at the national level,” Sivaram said, adding that the Health Ministry needs to take steps to make such information accessible.

FY21, which coincides with the initial pandemic wave, had the highest ratio of mortality claims to total deaths at 1.35 percent.

“There could be multiple reasons that can be attributed to this,” said a health economist.

“Those who continue to remain insured may belong to older age brackets or have underlying health conditions that make them more vulnerable. Additionally, many of these policyholders could be holding legacy insurance policies taken during or just after the pandemic years, when health risks were already high but not fully accounted for in the underwriting process.”

He explained that COVID-19’s long-term effects, such as cardiovascular complications, respiratory issues, and undiagnosed illnesses due to deferred medical care, could still be playing out, leading to higher mortality claims even years after the peak of the pandemic.

“The virus may have triggered or worsened chronic illnesses that are now resulting in fatalities, and insurance data is just beginning to reflect that delayed impact,” he added.

Moreover, this persistent pressure from rising mortality claims has led Indian life insurers to revise and recalibrate pricing for term life insurance.

Major companies like HDFC Life, ICICI Prudential Life, Axis Max Life and Bajaj Allianz Life are adjusting term life pricing, to cope with this persistent pressure from rising mortality claims.

In order to compensate for this contraction in group coverage and rising term life claims, these insurers have pivoted sharply toward retail-oriented, investment-linked policies, most notably Unit Linked Insurance Plans (ULIPs).

These products bundle insurance with market-linked returns and appeal to a growing segment of financially literate, urban consumers seeking both protection and wealth creation.

SBI Life derived approximately 67 percent of its Annual Premium Equivalent (APE) from ULIPs. ICICI Prudential Life reported a ULIP share of 50.8 percent of its APE. HDFC Life indicated a ULIP share ranging from 24 percent to 37 percent, depending on the quarter in FY25.

◆Deoxygenation Trends and Their Multivariate Association with Self-Reported Fatigue in Post-COVID Syndrome【MDPI　2025年6月3日】

Abstract

Background/Objectives: A relevant subgroup of post-COVID-19 syndrome (PCS) patients suffers from post-exertional malaise (PEM) and cardiovascular or neurological symptoms, impairing daily functioning up to becoming even house- or bedbound. Recent data suggest that PCS summarizes different subgroups, one of them being characterized by an impaired microcirculation. Thus, the aim of the present study was to investigate local deoxygenation, measured with non-invasive near-infrared regional spectroscopy (NIRS), and its association with self-reported fatigue in patients with PCS compared to controls in light exercise.
Methods: 150 participants (100 PCS patients and 50 controls) were recruited. PEM was assessed using FACIT, Chalder, and Bell scoring and Canadian Criteria. NIRS was used to measure local oxygenation while kneading a stress ball and during recovery.
Results: PCS patients showed fatigue scores of 30 (Bell score), 20.6 (FACIT fatigue score), and 9.914 (Chalder fatigue score). Decreased deoxygenation peaks at the start of exercise were observed in patients with PCS, compared to controls (p = 0.0002). Multivariate analysis identified a subgroup, showing an association between strong fatigue and restricted oxygenation dynamics.
Conclusions: NIRS could be a potential tool to assess deoxygenation deficits even in moderate to severely impaired PCS patients using light exercise protocols.

◆Distinct Transcriptomic Biomarkers and Pathways Associated with Cardiovascular and Neurovascular Dysregulation in Long COVID-19 Brain Fog【bioRxiv　2025年6月11日】

Abstract

Background Long COVID-19 often manifests as persistent cognitive impairments, commonly referred to as brain fog, with poorly understood biological mechanisms. Previous studies highlighted neuroinflammation and blood-brain barrier disruption. Here, we identify distinct biomarkers and dysregulated pathways, with a focus on cardiovascular and neurovascular involvement.

Methods We analyzed public RNA-sequencing data (GEO: GSE251849) comparing long COVID-19 patients with brain fog, long COVID-19 patients without cognitive symptoms, convalescent patients, and healthy controls. Differential gene expression, principal component analysis, and pathway enrichment analyses were performed using DESeq2, KEGG, GO, and Enrichr tools.

Results Our analysis revealed a distinct gene expression profile in patients with brain fog, identifying exclusive biomarkers such as NRCAM, GRIN2C, NOS2, and EDNRB. Pathway analysis revealed significant dysregulation in cardiovascular, calcium signaling, and relaxin signaling pathways, suggesting novel biological mechanisms that contribute to cognitive impairment. These findings expand beyond neuroinflammation and highlight potential cardiovascular involvement in long COVID-related cognitive dysfunction.

Conclusion Our study reveals novel transcriptomic signatures that highlight cardiovascular and neurovascular dysregulation, providing promising diagnostic biomarkers and therapeutic targets. Future research should validate these findings and investigate therapeutic interventions addressing these unique pathways.

◆Distinct Transcriptomic Biomarkers and Pathways Associated with Cardiovascular and Neurovascular Dysregulation in Long COVID-19 Brain Fog【cell : Trends in Neurosciences　2025年6月11日】

Highlights

CD8 T cells have emerged as potential cellular mediators of neuropathology in neurological diseases as diverse as Alzheimer’s disease, long COVID, glioblastoma, Susac syndrome, multiple sclerosis and cerebral malaria.
Immune-mediated blood–brain barrier (BBB) disruption can be initiated in the brain or through peripheral immune cell infiltration through innate and adaptive immune system processes.
Among adaptive immune cells, CD8 T lymphocytes primed by professional antigen-presenting cells that express MHC class I induce BBB disruption.
Cellular components of the neurovascular unit (NVU) express MHC class I molecules which can alter brain-infiltrating CD8 T cell responses and cause BBB disruption during neuroinflammatory conditions.
The human perforin gene is diverse and could influence CD8 T cell-mediated BBB disruption in infectious diseases.

Abstract

Blood–brain barrier (BBB) disruption is a hallmark of many neurological diseases. It is known that proinflammatory cytokines can disrupt tight junctions between endothelial cells of the BBB, allowing larger molecules to penetrate the brain. Along with BBB disruption, these diseases also feature increased numbers of CD8 T cells as components of the immune cell population in the brain. In this review, we discuss the mechanisms of CD8 T cell-mediated BBB disruption, including the roles of antigen presentation through MHC class I molecules, which emerge as an important determining feature of BBB disruption and immune cell infiltration into the brain.

◆Distinct Transcriptomic Biomarkers and Pathways Associated with Cardiovascular and Neurovascular Dysregulation in Long COVID-19 Brain Fog【ACS Publications　2025年6月9日】

Abstract

SARS-CoV-2 infections in children lead to symptoms from mild respiratory illness to severe postacute sequelae of COVID-19, including multisystem inflammatory syndrome in Children (MIS-C). We conducted a metabolic profiling of 147 children’s serum samples, including acute COVID-19 patients, MIS-C patients, and healthy controls. Using nuclear magnetic resonance spectroscopy and liquid chromatography–mass spectrometry, we measured 1101 metabolites. The results revealed distinct metabolic profiles in acute COVID-19 and MIS-C patients, with significant alterations in lipid classes. Both conditions exhibited an elevated Apo-B100/Apo-A1 ratio and increased serum inflammatory markers. MIS-C patients showed unique disruptions, including increased triglycerides and altered lipoprotein composition. Despite milder clinical respiratory symptoms, children’s metabolic disturbances mirrored those seen in severe adult COVID-19 patients, indicating a shared inflammatory response to SARS-CoV-2. This suggests potential long-term health impacts, underscoring the need for continued research into the metabolic consequences of COVID-19 in children.