People with accelerated biological age at increased risk of COVID-19 symptom severity and mortality
By Dr. Liji Thomas, MDJul 13 2020
Even after six months, there are no successful definitive therapies for COVID-19, nor an effective vaccine, other the recently approved, ‘Emergency Use Authorization’ of the antiviral remdesivir in COVID-19 disease. The greatest threat of mortality remains among the elderly and those with coexisting illnesses. Now, a new study published on the preprint server medRxiv* in July 2020 discusses the explanation in terms of the biological aging markers in patients with severe disease, rather than the chronological age.
Novel Coronavirus SARS-CoV-2 Colorized scanning electron micrograph of an apoptotic cell (green) heavily infected with SARS-CoV-2 virus particles (orange), isolated from a patient sample. Image at the NIAID Integrated Research Facility (IRF) in Fort Detrick, Maryland. Credit: NIAID
Age and Disease Boosts COVID-19 Mortality
The mortality rate in COVID-19 is projected to be between 10% and 27% in people aged 85 years and over, compared to 3-11% for the age group 65-84 years, 1-3% for those aged 55-64 years, and <1% for those 20-54 years of age. Overall, the mortality risk is a hundred times higher for people aged 85-plus relative to the under-50s. And people above 65 years account for 80% of deaths in the USA at present.
Another high-risk factor for mortality is the presence of diseases like cardiovascular disease, chronic kidney disease, liver disease, and chronic obstructive pulmonary disease.
The Study: Predictive Ability of Phenotypic Age
To pull these together, the current authors used their previously validated biomarkers of aging, which have been found to predict both sickness and death risk in the short and long term for non-COVID-19 individuals. These markers have led them to consider the theory that biological aging rather than chronological aging is a more powerful influence on the severity profile of COVID-19.
This new study by an international team of researchers aimed to measure the risk and the predictive ability of advanced aging biomarkers, using a risk profile called Phenotypic Age or PhenoAge, for short. This is a supervised machine learning model that was trained using 42 biomarker inputs to predict all-cause mortality.
The researchers used this measure on biomarker data from 2006 to 2010, from over 440,000 participants, to arrive at the PhenoAge. This was then combined with information on diseases developed over the long-term, up to 2017. The researchers then examined the presence of an association between PhenoAge and COVID-19 severity.
Of the participants, about 25,000 died before the pandemic, using a cut-off of February 1, 2020. Of the usable data, more than 55% came from females, and 94% said they were White, 2% as Black, and 4% all others.
Of these, 6,100 were tested from March 16 to March 31, 2020. About 21% were positive and recovered. 3% were positive and died after being infected, between March 5 and April 26, 2020.
The mean chronological age was 68 years on April 26, 2020, with 64% being at least 65 years. The baseline PhenoAge was available for over 347,000 samples. The researchers found that the mean chronological age, at 56 years, was 2.5 years more than the mean PhenoAge at 54 years.
Does Accelerated Aging Increase Risk of COVID-19 Death?
The PhenoAge was used to estimate accelerated phenotypic age (PhenoAgeAccel) at baseline, which represents the difference between an individual’s phenotypic age compared to that expected at the current age.
The researchers also looked at two outcomes: whether the COVID-19 tests were positive, and whether the patient with a positive test died. The researchers used the two outcomes above in four different models, namely:
- Current chronological age
- Current age plus baseline PhenoAgeAccel
- Current age plus pre-existing disease conditions
- Current age plus pre-existing disease states plus baseline PhenoAgeAccel
They then considered the possibility that a higher PhenoAgeAccel, meaning more rapid biological aging, would positively associate with severe COVID-19. They found that men and Black patients were more likely to have a positive test and to die of confirmed disease, and to test positive, respectively.
In the first model, age at COVID-19 was not associated significantly with positive tests when adjusted for demographic factors. Model 2, when adjusted for demographic and accelerated phenotypic age evaluated 10-14 years before the current date also showed minimal correlation.
Current age was associated with lower risk when adjusted for pre-existing disease states – the third model. The fourth model was also protective.
Accelerating Aging Linked to Severe COVID-19
The researchers found that in the second model, the odds that the individual would test positive for COVID-19 went up by 1.3 with each 5-year increase in PhenoAgeAccel, and the odds of death from any cause went up by 1.55. The third and fourth models included diseases that were individually linked to higher test positivity and deaths from all causes, even when adjusted for current age and demographics.
These models showed that accelerated phenotypic age was linked to more severe COVID-19 symptoms as measured by either outcome. This continued to be significant even when various chronic disease conditions were adjusted for.
Secondly, once they adjusted for older biological age, COVID-19 severity was less heavily linked with some disease conditions like a history of pneumonia, diabetes, and chronic lung disease. The researchers suggest that this association could be due to the underlying pathways of immune activation in both disease conditions.
Underlying Chronic Inflammation Holds the Key?
- Dutch trial of convalescent plasma in COVID-19 discontinued
- Research suggests protective effect of influenza vaccine against COVID-19 severity and mortality
- Study links fermented vegetable consumption to low COVID-19 mortality
Prior research by the same team showed that PhenoAgeAccel was associated with biological processes linked to immune activation, cell cycle functions, and carbohydrate digestion and metabolism. These previous studies also trained a biological clock based on DNA methylation, called DNAmPhenoAge, where PhenoAge represented the biological age rather than the chronological age.
They found that this clock was related to the activation of various chemicals that marked the presence of inflammation, repair of DNA damage, active transcription and translation, and immune senescence. These indicate inflammaging, or an acceleration of chronic inflammation, which underlies both accelerated biological aging and severe COVID-19 infection.
Such inflammaging processes include loss of genomic stability, senility of individual cells, impaired function of mitochondria, changes in the microbial populations of the body, inflammasome activation, and an abnormal.loss of regulation of immune cell responses without any apparent underlying factor, as well as chronic infections. This is important because these could present targets to prevent severe COVID-19.
Current Age Less Important than PhenoAge
Why was there no obvious impact of current chronological age on these outcomes? The researchers suggest this could be because of the time lag between the baseline blood draw and the current pandemic. Instead of this parameter, when the length of follow-up is used to predict risk, there is a positive correlation with increasing age and COVID-19 severity.
The researchers feel that simultaneous assessment of PhenoAgeAccel for all participants would have shown that chronological age has a much less significant or even insignificant effect on these outcomes. In support of this, they point to the reversal of the association of chronological age with test positivity when they adjusted for the presence of other diseases.
It could be that among people with comparable illnesses at the time of the earlier assessment, younger people were aging faster, and this accounts for the higher mortality. On the other hand, the adjusted age-related risk remains higher, which shows that disease alone is not the sole cause of the increased risk of death.
Implications of the Study
The researchers stress that the most important finding from the study is that the PhenoAge measure had reliable predictive power, even though it was computed ten years or more before the current pandemic appeared, and though it was not connected to either the chronological age or the presence of other disease states. This could be because the biomarkers are based on their predictive power for age-linked mortality when used in combination.
They say, “The biomarker-specific associations in this study may shed light on the key aging signals that may be driving the association between biological aging and COVID-19 severity.”
The most robust links were between albumin levels and a positive COVID-19 test, and between death due to any cause in an individual with confirmed COVID-19 with high glucose levels. Glucose may promote viral replication and antiviral immunosuppression. High glucose levels are already known to increase the risk of death in flu patients.
Low albumin was also linked to severe COVID-19, as found in many prior studies. This correlation with albumin levels measured more than ten years earlier confirms that the low levels of albumin are due to chronic inflammation, as also shown by a high C-reactive protein (CRP) and leukocyte count, rather than because of reduced albumin production.
Despite the current relevance of an aging study that occurred long ago, it could also be that some individuals who had accelerated aging later improved their health, but others showed reverse movement. This would mean that a more recent PhenoAge could improve the predictive value of this measure.
Secondly, the study points to two different target pathways for treating COVID-19. One is the underlying inflammatory mechanisms that link accelerated aging and severe COVID-19. The other is the aging process. In the latter, drugs like metformin and rapamycin, that slow aging, could be used to treat this condition. Rapamycin boosts the efficacy of the flu vaccine, and metformin lowers blood glucose levels. Moreover, it is being suggested that the latter could suppress the inflammatory activation of immune cells.
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.