What’s the risk of different human populations to develop a disease?
To find out, a team led by Université de Montréal professor Guillaume Lettre created an international consortium to study the blood of hundreds of thousands of people worldwide.
In one of the largest studies of its kind, published today in Cell, close to 750,000 participants from five major populations – European, African, Hispanic, East Asian and South Asian – were tested to see the effect of genetic mutations on characteristics in their blood.
These characteristics include such things as hemoglobin concentration and platelet counts.
“Each human population is subject to different environments,” said Lettre, a researcher at the Montreal Heart Institute.
“Over thousands of years,” he said, “these environmental pressures have resulted in the progressive appearance of variations in DNA, called genetic mutations, which can influence our physical characteristics, such as skin size or color, but also our risk of getting certain diseases.”
He added: “This observation (of how the environment affects how people’s appearance and health vary in different parts of the world) represents the cornerstone of the theory of evolution by natural selection proposed by Charles Darwin in 1859.”
The consortium founded by Lettre and his colleagues chose to study 15 characteristics of blood cells because previous studies had already uncovered mutations whose consequences were limited to certain populations.
45 million genetic mutations
By testing more than 45 million genetic variations in each participant, Lettre and his collaborators have found more than 5,000 mutations in human DNA that affect the blood characteristics of populations around the world.
Done in conjunction with another study focusing exclusively on individuals of European origin, the new study shows that the vast majority of mutations associated with blood cells were common to all five major population groups.
But aside from these, the researchers also found about 100 mutations whose effect was restricted to certain populations and which, it turns out, are not found in people of European descent.
For example, in individuals of South Asian origin, the researchers identified a mutation in the interleukin-7 gene that stimulates the secretion of this molecule and thus increases the levels of lymphocytes (a type of white blood cell in the immune system) circulating in their blood.
“Of course, this kind of mutation can affect the health of people of South Asian origin,” Lettre noted. “It’s thought that this mutation could influence their capacity to resist certain infections or develop diseases like blood cancer.”
However, he cautioned, “these are, at present, only hypotheses, as researchers do not have the capacity to test them, given the immense costs and the difficulty of finding participants for this type of study.”
Improving ways of predicting
By comparing the genetic results obtained in each population, the researchers were able to prioritize certain genes that appear to have an overall effect on blood cell production.
This will make it possible, over the long term, to improve ways of predicting the risk of suffering from certain diseases and to develop new, more effective treatments.
Here again, however, major investments in research will be required to analyze the consequences of these mutations on the health of these population groups.
Another major obstacle will be to convince researchers how important it is for all population groups globally to be included in these types of genetic studies.
“Despite the size of our study, the vast majority of participants – about 560,000 out of 740,000 individuals – were of European origin,” Lettre noted. “This necessarily introduces a bias into the study.”
In the future, he said, “we hope to work with populations that have been little studied so far – for example, East African populations or indigenous peoples – in order to shed light on new genes that regulate blood cells.”
One thing is clear, he concluded: in order to better understand human diseases and to ensure that everyone, regardless of ethnic origin, is able to benefit from advances in genetics and precision medicine, diseases will have to be studied in all populations worldwide.
The article “Trans-ethnic and ancestry-specific blood-cell genetics in 746,667 individuals from 5 global populations”, by Ming-Huei Chen, Laura M. Raffield, Abdou Mousas et al, was published Sept. 3, 2020 in Cell.
COVID-19, the disease caused by the SARS-COV2 virus, has led to a global pandemic [1, 2]. The SARS-COV2 virus has had varying effects on the global population; those who are older and with comorbidities such as cardiovascular disease, diabetes, and pulmonary diseases have proven more vulnerable to severe disease [3–7].
Given the significant morbidity and mortality associated with COVID-19, there has been scientific interest in eliciting data that details characteristics that may render individuals more susceptible to COVID-19 infection and determining what risk factors may be associated with progression and severity of disease from the virus [8–11].
There have been numerous molecular level hypotheses raised for the variable susceptibility to disease and vulnerability to severe disease, such as the variable expression of ACE-2 expression in the airway epithelia [12]. Landsteiner’s ABO carbohydrate moieties are genetically inherited and previous reports have suggested a correlation between ABO blood type, cardiovascular disease, and cancers, as well as typing and susceptibility to certain infections, including SARS coronavirus [13–18].
In currently pre-printed data, Zhao et al. reported a possible association between blood type A and a higher risk for COVID-19 infection and mortality while blood group O was associated with a lower risk of infection and mortality [13]. Zietz and Tatonetti found that blood type A was correlated with a higher odds of testing positive for disease [19].
There is a paucity of data regarding the relationship between ABO blood typing and severity of COVID-19 disease. Using a large multi-institutional cohort of patients, this study aimed to determine if there is an association between ABO blood type and severity of COVID-19 disease as well as ascertain if there is variability in testing positive for COVID-19 between blood types.
Results
During the study period, there were 7648 symptomatic patients who received a COVID-19 test throughout the five institutions included. Of these, 1289 tests were positive and had their blood group documented hence included in the analysis; the demographics, comorbidities, and medications of this population are outlined in Table Table1.1. Of these, 484 (37.5%) were admitted to hospital, 123 (9.5%) were admitted to the ICU, 108 (8.4%) were intubated, 3 (0.2%) required ECMO, and 89 (6.9%) died. Of the 1289 patients who tested positive, 440 (34.2%) were blood type A, 201 (15.6%) were blood type B, 61 (4.7%) were blood type AB, and 587 (45.5%) were blood type O. Six hundred and four COVID-19 positive patients had their WBC recorded (blood type A: 204, blood type B: 104, blood type AB: 35, blood type O: 261); 511 had their LDH evaluated (blood type A: 169, blood type B: 91, blood type AB: 29, blood type O: 222); 487 had their CRP evaluated (blood type A: 85, blood type B: 28, blood type AB: 212, blood type O: 487); and 393 had their ESR evaluated (blood type A: 130, blood type B: 71, blood type AB: 28, blood type O: 164). There was no association between blood type and any of the peak inflammatory markers (peak WBC, p = 0.25; peak LDH, p = 0.40; peak ESR, p = 0.16; peak CRP, p = 0.14). Moreover, there was no association between any of the clinical outcomes (admission p = 0.20, ICU admission p = 0.94, intubation p = 0.93, required proning while intubated p = 0.58, requiring ECMO p = 0.09, and death p = 0.49, Table 2).
Table 1
Demographics and patient comorbidities. N total = 1289. BMI: body mass index, COPD: chronic obstructive pulmonary disease, MI: myocardial infarction, ESRD: end-stage renal disease, CAD: coronary artery disease, CKD: chronic kidney disease, DVT: deep venous thrombosis, PE: pulmonary embolism, DOAC: direct oral anticoagulant. White refers to non-white, non-Hispanic
| Factor | Level | Blood type A | Blood type B | Blood type AB | Blood type O | P value |
|---|---|---|---|---|---|---|
| N | 440 | 201 | 61 | 587 | ||
| Age, mean (SD) | 56.9 (18.6) | 57.6 (18.1) | 57.1 (19.9) | 54.8 (18.1) | 0.14 | |
| BMI, mean (SD) | 30.8 (6.5) | 30.6 (6.7) | 29.4 (5.4) | 32.0 (14.9) | 0.32 | |
| Rhesus positive | 392 (89.1%) | 183 (91.0%) | 53 (86.9%) | 533 (90.8%) | 0.63 | |
| Female sex | 299 (68.0%) | 136 (67.7%) | 33 (54.1%) | 404 (68.8%) | 0.14 | |
| Language (primary) | English | 328 (74.5%) | 149 (74.1%) | 54 (88.5%) | 382 (65.1%) | <0.001 |
| Spanish | 88 (20.0%) | 36 (17.9%) | 4 (6.6%) | 180 (30.7%) | ||
| Other | 24 (5.5%) | 16 (8.0%) | 3 (4.9%) | 25 (4.3%) | ||
| Race | White | 221 (50.2%) | 80 (39.8%) | 28 (45.9%) | 224 (38.2%) | 0.008 |
| Black | 84 (19.1%) | 49 (24.4%) | 15 (24.6%) | 114 (19.4%) | ||
| Hispanic | 52 (11.8%) | 25 (12.4%) | 4 (6.6%) | 103 (17.5%) | ||
| Other | 77 (17.5%) | 41 (20.4%) | 13 (21.3%) | 128 (21.8%) | ||
| Not Reported | 6 (1.4%) | 6 (3.0%) | 1 (1.6%) | 18 (3.1%) | ||
| Hypertension | 256 (58.2%) | 124 (61.7%) | 42 (68.9%) | 341 (58.1%) | 0.34 | |
| Smoker | 97 (22.0%) | 39 (19.4%) | 15 (24.6%) | 117 (19.9%) | 0.69 | |
| Hyperlipidemia | 251 (57.0%) | 105 (52.2%) | 35 (57.4%) | 323 (55.0%) | 0.70 | |
| COPD | 55 (12.5%) | 26 (12.9%) | 9 (14.8%) | 72 (12.3%) | 0.95 | |
| Diabetes mellitus | 150 (34.1%) | 66 (32.8%) | 25 (41.0%) | 197 (33.6%) | 0.68 | |
| Cancer diagnosis | 131 (29.8%) | 57 (28.4%) | 19 (31.1%) | 161 (27.4%) | 0.83 | |
| Cirrhosis | 14 (3.2%) | 9 (4.5%) | 1 (1.6%) | 21 (3.6%) | 0.72 | |
| Asthma | 124 (28.2%) | 46 (22.9%) | 14 (23.0%) | 163 (27.8%) | 0.44 | |
| History of stroke | 50 (11.4%) | 30 (14.9%) | 10 (16.4%) | 71 (12.1%) | 0.47 | |
| ESRD | 17 (3.9%) | 13 (6.5%) | 5 (8.2%) | 30 (5.1%) | 0.33 | |
| CAD | 171 (38.9%) | 71 (35.3%) | 30 (49.2%) | 231 (39.4%) | 0.28 | |
| CKD | 74 (16.8%) | 41 (20.4%) | 12 (19.7%) | 111 (18.9%) | 0.70 | |
| Dysrhythmia | 199 (45.2%) | 95 (47.3%) | 34 (55.7%) | 274 (46.7%) | 0.49 | |
| Congestive heart failure | 71 (16.1%) | 41 (20.4%) | 13 (21.3%) | 107 (18.2%) | 0.51 | |
| History DVT | 39 (8.9%) | 14 (7.0%) | 6 (9.8%) | 43 (7.3%) | 0.71 | |
| History PE | 26 (5.9%) | 10 (5.0%) | 4 (6.6%) | 24 (4.1%) | 0.55 | |
| Aspirin | 66 (15.0%) | 45 (22.4%) | 16 (26.2%) | 96 (16.4%) | 0.029 | |
| Warfarin | 14 (3.2%) | 7 (3.5%) | 1 (1.6%) | 20 (3.4%) | 0.90 | |
| Statin use | 132 (30.0%) | 64 (31.8%) | 25 (41.0%) | 141 (24.0%) | 0.007 | |
| Calcium channel blocker | 56 (12.7%) | 29 (14.4%) | 5 (8.2%) | 76 (12.9%) | 0.65 | |
| Thiazide diuretic | 36 (8.2%) | 14 (7.0%) | 4 (6.6%) | 46 (7.8%) | 0.94 | |
| ACE inhibitor | 59 (13.4%) | 32 (15.9%) | 7 (11.5%) | 81 (13.8%) | 0.78 | |
| ARB | 32 (7.3%) | 19 (9.5%) | 6 (9.8%) | 50 (8.5%) | 0.76 | |
| Beta blocker | 97 (22.0%) | 51 (25.4%) | 14 (23.0%) | 124 (21.1%) | 0.66 | |
| DOAC | 31 (7.0%) | 12 (6.0%) | 6 (9.8%) | 23 (3.9%) | 0.071 | |
| p2y12 inhibitor | 11 (2.5%) | 11 (5.5%) | 4 (6.6%) | 9 (1.5%) | 0.006 |
Table 2 – Univariate analysis
| Peak creatinine | 1.8 (2.3) | 1.9 (2.4) | 1.5 (1.7) | 1.7 (2.2) | 0.64 |
| Peak WBC, mean (SD) | 9.9 (5.9) | 10.2 (12.1) | 10.9 (7.9) | 8.9 (5.9) | 0.25 |
| Peak LDH, mean (SD) | 484.8 (1180.4) | 414.1 (198.0) | 324.3 (141.3) | 375.5 (165.0) | 0.40 |
| Peak ESR, mean (SD) | 64.5 (37.9) | 63.3 (36.4) | 63.7 (37.9) | 55.7 (33.4) | 0.16 |
| Peak CRP, mean (SD) | 139.3 (110.3) | 140.7 (97.8) | 139.0 (116.5) | 118.0 (95.3) | 0.14 |
| Admitted | 158 (35.9%) | 85 (42.3%) | 28 (45.9%) | 213 (36.3%) | 0.20 |
| ICU admission | 41 (9.3%) | 18 (9.0%) | 7 (11.5%) | 57 (9.7%) | 0.94 |
| Intubated | 38 (8.6%) | 15 (7.5%) | 6 (9.8%) | 49 (8.3%) | 0.93 |
| Required proning | 18 (4.1%) | 4 (2.0%) | 2 (3.3%) | 23 (3.9%) | 0.58 |
| ECMO | 1 (0.2%) | 2 (1.0%) | 0 (0.0%) | 0 (0.0%) | 0.088 |
| Dead | 36 (8.2%) | 14 (7.0%) | 5 (8.2%) | 34 (5.8%) | 0.49 |
| Intubation/death (ID) | 63 (14.3) | 23 (11.4) | 8 (13.1) | 68 (11.6) | 0.57 |
WBC white blood count, LDH lactate dehydrogenase, ESR erythrocyte sediment rate, CRP C-reactive protein, ICU Intensive Care Unit, ECMO extracorporeal membrane oxygenation
In the multivariable analysis, blood type was not determined to be independently associated with COVID-19 disease severity (blood type A: ref., blood type B: AOR: 0.72, 95% CI: 0.42–1.26, blood type AB: AOR: 0.78, 95% CI: 0.33–1.87, blood type O: AOR: 0.77, 95% CI: 0.51–1.16), Rh+: AOR: 1.03, 95% CI: 0.93–1.86) (Table 3).
Table 3
Multivariable analysis: blood type versus intubation/death. Referent is blood type A. Also adjusted for sex, primary language, aspirin use, calcium channel blocker use, diagnoses of chronic kidney disease, coronary artery disease, prior stroke and diabetes mellitus, race not reported (referent: white), sex and presence of rhesus factor. Rh +: Rhesus factor positive. Hosmer and Lemeshow goodness of fit p = 0.98
| Blood type | AOR | 95% CI | P value |
|---|---|---|---|
| A | Ref | ||
| B | 0.72 | 0.42–1.26 | 0.25 |
| AB | 0.78 | 0.33–1.87 | 0.58 |
| O | 0.77 | 0.51–1.16 | 0.21 |
| Rh+ | 1.03 | 0.93–1.86 | 0.10 |
For the analysis evaluating for correlation of blood type with a positive test, blood type A had 440 (16.6%) positive tests, blood type B had 201 (19.4%) positive tests, blood type AB had 61 (19.8%) positive tests, and blood type O had 587 (16.1%) positive tests (p = 0.036). After multivariable analysis, blood type A had no correlation with positive testing (AOR: 1.00, 95% CI: 0.88–1.13), blood type B was associated with higher odds of testing positive for disease (AOR: 1.28, 95% CI: 1.08–1.52), AB was also associated with higher odds of testing positive (AOR: 1.37, 95% CI: 1.02–1.83), O was associated with lower odds of testing positive (AOR: 0.84, 95% CI: 0.75–0.95), and Rh+ blood was associated with a higher odds of testing positive (AOR: 1.22 (1.003–1.50) (Table 4).
Table 4
Rate of positive test by blood type. Overall P value 0.036 derived by Chi-squared testing. Adjusted odds ratio (AOR) with adjustment for sex, primary language, age, and rhesus factor with each blood type compared to all others. Rh+ model included blood type as a categorical covariate. Rh+: Rhesus factor positive. Hosmer and Lemeshow goodness of fit p > .5 for all models
| Blood type | Total N | N positive (%) | AOR (95% CI) | P value |
|---|---|---|---|---|
| A | 2649 | 440 (16.6) | 1.00 (0.88–1.13) | 0.96 |
| B | 1035 | 201 (19.4) | 1.28 (1.08–1.52) | 0.004 |
| AB | 308 | 61 (19.8) | 1.37 (1.02–1.83) | 0.035 |
| O | 3656 | 587 (16.1) | 0.84 (0.75–0.95) | 0.007 |
| Rh+ | 6707 | 1161 (17.3) | 1.22 (1.003–1.50) | 0.047 |
| All | 7648 | 1289 (16.9) |
resounce link : https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7354354/
More information: Ming-Huei Chen et al, Trans-ethnic and Ancestry-Specific Blood-Cell Genetics in 746,667 Individuals from 5 Global Populations, Cell (2020). DOI: 10.1016/j.cell.2020.06.045
