Why does ancestry matter for some medical decisions?

Tags

Ancestry, Ancestry-informative marker - Wikipedia (AIMs), Single-nucleotide polymorphism - Wikipedia (SNPs)

Short answer: Most often ancestry and environment together mould disease risk. Ancestry alone directly confers disease risk less often, specifically so in cases where single gene mutations have outsize effects. Tay–Sachs disease – Wikipedia, Sickle-cell disease – Wikipedia and Cystic fibrosis – Wikipedia are some well-known examples of the latter.

Longer answer if interested

Why It’s So Difficult to Scientifically Tease Apart The Role Of Ancestry In Disease Causes

Human ancestry is the outcome of choice that liberally pockmarks both past and present with gratuitous violence and calamities since it’s a choice contrived to mediate and enforce differential access to resources. This choice manifests itself as caste, class, ethnicity, linguistic group, race, sect, tribe, etc. They’re social, i.e., explicitly political and cultural, not biological categories (1) but they end up influencing biology anyway. Here’s how.

• Human societies tend to practice some form of social stratification or another. Over time, differential access to critical healthful life-sustaining resources such as quality education, health care and nutrition impact health, especially since social stratification-engendered privations tend to be experienced across generations.
• Thus, health disparity is the outcome of historical inequities a particular social category experiences at the hands of what usually tends to be a long-prevailing hegemony.
• Persistence of social stratifications across generations thus end up influencing biology by differentially influencing disease risk through the human-created construct of health disparity.
• The ethically unambiguous and appropriate place for ancestry in biomedicine lies in helping to try to tease apart the relative contributions of health disparity versus genetic predisposition to disease.
• However, it’s well nigh impossible to disentangle genetic predisposition from the many environmental confounders health disparities impose on disease risk. This stymies the effort to accurately parse and pinpoint the role of ancestry in many disease causes.

Problem is by tending to examine it devoid of its inherent sociological context, biomedicine artificially insulates ancestry. That leaves its role vulnerable to exploitation by the scientific flavor of the day, which these days is genomics.

The genomic era is like the proverbial hammer primed to seek and find nails everywhere so it’s become quite the fashion to privilege or attempt to privilege genetic predilection even in the case of multi-factorial diseases. This problem is also grounded in the fact that biomedicine has evolved an inherently siloed approach such that super-specialists, be they molecular biologists, geneticists, epidemiologists, public health researchers, etc., examine a given health issue through the lens of their training often without simultaneously attempting to look beyond, especially at the sociological context of disease. Thus, ancestry in the form of race/ethnicity has been and continues to be used as a proxy for genotype, albeit devoid of sociological context.

Official guidelines add to the problem, being inadequate and/or inaccurate and/or too riddled with ambiguity. For example, in the US, when classifying their research subjects, NIH-funded scientists are required to adhere to the racial and ethnic categories specified by the Office of Management and Budget (OMB)’s Directive #15, the so-called NIH Inclusion Policy and Guidelines (see below from 2),

‘American Indian or Alaska Native, Asian, Black or African American, Hawaiian or Pacific Islander, and White, Hispanic or Latino and Not Hispanic or Latino’

• At least one survey of 18 NIH-funded scientists (11 men, 7 women) (3) reports these guidelines
  • Are applied too unquestioningly regardless of their utility or accuracy.
  • Are bureaucratic, one size fits all, catch all, inflexible, an example cited being Barack Obama. Categorize as Black or Caucasian?
  • Are difficult to comply with in geographic areas with few minority residents.
  • Force minority inclusion numbers that yield data lacking sufficient statistical power to provide meaningful results, i.e., difficult to generate representative and therefore generalizable data sets.
• A study from the UK (4) reports similar flawed approach to study of racial/ethnic contribution to health and disease.

Genomic Approaches To Assess Ancestry Remain Inherently Flawed

Ancestry-informative marker – Wikipedia (AIMs) are inherently misleading and fraught with flawed assumptions (5, 6, 7, 8, 9). So what are AIMs? They’re population-specific markers, typically Single-nucleotide polymorphism – Wikipedia (SNPs) that occur at different frequencies in different populations. Since they’re shared by all humans, rather than presence or absence, analysis focuses on their frequency.

These days TV ads about learning one’s ancestry are plentiful. Just send a saliva sample in and get genetic genealogical results back. Too deceptively simple but also flawed to boot.

Obviously a test sample is compared to reference samples of Africans, Asians, Europeans, Native Americans, etc.

• Who are the ‘reference populations’ for each race or ethnicity? To assign ‘purity’, ideally they should be groups that have remained immobile, isolated and endogamous for millennia till date since reference samples should represent ‘pure’ examples of different ethnic/racial categories, the standard against which the test sample would be compared.
• Obviously such reference populations are an impossibility for much of the world’s population.
• Instead ‘small groups of contemporary people‘ (9) are chosen as representative samples for a particular continent or region or ethnic, linguistic or tribal group.
• Who is chosen? What are the criteria used to choose one individual and not another to supply the reference sample?
• How many people from a given population (caste/ethnicity/linguistic group/race/sect/tribe) are sampled to develop a representative reference sample base?
• How many are necessary? How many sufficient? 100, 1000, 10000, 100000, …?
• How are thresholds set to determine how a given result is interpreted to include or exclude a particular population?

The answers to these scientifically crucial questions lie hidden behind the legal iron curtain of proprietary information preventing even the bare minimum in terms of rigorous science, namely to independently replicate and thereby verify (9). Thus, the widely advertised genomic categorization of ancestry using AIM touted by various commercial entities is more or less the outcome of a technological Sleight of hand – Wikipedia or two or three or more.

Diseases Found To Be More Prevalent in A Particular Race/Ethnicity Are Typically Monogenic

Tay–Sachs disease – Wikipedia & BRCA1 – Wikipedia mutations

At least 2 disease-causing/associated genes are more prevalent among Ashkenazi Jews – Wikipedia, infantile form of Tay–Sachs disease – Wikipedia and BRCA1 – Wikipedia mutations associated with higher risk for breast cancer.

Tay-Sachs is an autosomal recessive (Dominance (genetics) – Wikipedia) Genetic disorder – Wikipedia caused by a Mutation – Wikipedia in HEXA – Wikipedia gene on Chromosome 15 (human) – Wikipedia where children usually die by the age of 4. Disease is caused by the impaired function of lysosomal enzyme, Hex A.

Higher prevalence in these two instances is presumed to owe to the fact that the Jewish population descended from a small number of founders and remained largely endogamous (10).

Genetic counseling and prenatal screening are also advised for Cajuns – Wikipedia in Louisiana and French Canadians – Wikipedia since similar mutations have been identified among them.

Sickle-cell disease – Wikipedia

An adaptation to thwart malaria, sickle cell is more common among those with West African ancestry, specifically those with the globin S (betas) mutation (11).

Cystic fibrosis – Wikipedia

An autosomal recessive genetic disorder caused by mutations in Cystic fibrosis transmembrane conductance regulator – Wikipedia (CFTR) gene, it’s found to be more prevalent among people of European descent (12, 13).

More Accurate To Envision Ancestry As A Continuum Rather Than Groups Of Independently Evolving Discrete Units

An additional challenge is the fact that ancestry as a social construct is fast becoming less categorical as populations meet and meld as perhaps never before, even while they may have remained geographically isolated for varying lengths of time here and there in previous millennia.

Consider USA for example, a country that assesses race in its census.

• The category ‘Other’ was first listed in the US 1910 census. Now listed as ‘Some Other Race’, in the 2010 census it had become the 3rd largest category after ‘White’ and ‘Black’ (14, 15).
• In recent years, 15% of US marriages are between people of different ethnicities and races.
• One in seven US infants is today born into a ethnically and/or racially mixed family.
• A particular genomics example perfectly hints at the potentially vast complexity hidden underneath the surface of the race/ethnicity categories commonly used in our times. Complete genomic sequences of two famous European origin American scientists, James Watson – Wikipedia, Craig Venter – Wikipedia, and Seong-Jin Kim, an Asian-origin scientist, showed the former shared fewer (461000) SNPs with each other than they each shared with the Asian (569000 and 481000, respectively) (16, 17), something utterly unlikely to be discerned from physical appearance alone.
• A genomic analysis of self-identified European Americans (n = 326), African Americans (n = 324) and Hispanics (n = 327) in Manhattan, New York, revealed such substantial ancestral mix in both African Americans and Hispanics, the authors concluded (18, emphasis mine; see figure below from 19).

‘A pooled analysis of the African Americans and Hispanics from NY demonstrated a broad continuum of ancestral origin making classification by race/ethnicity uninformative’

• Needless to say, such melding happened or is happening faster in some countries and especially faster in large cosmopolitan cities. Largely the mix of Native American (Amerindian), European (mainly Portugal) and African, Brazil is a country that famously embodies more than anything else racial ambiguity (20).

Finally, data also suggests higher genetic diversity within races (85%) rather than between races (15%) (21), which only further undermines the practical value of race/ethnicity in dissecting disease risks and causes at group level. This is especially the case for Africa, the continent with the greatest degree of genetic variation (see below from 22 quoted in 9).

‘For many regions of the human genome, there are more variants found among people of Africa than found among people in the rest of the world. This is probably because humans have resided in Africa for much longer than we have resided any place else in the world, so our species had time to accumulate genetic changes within the people in Africa.’

In other words, race/ethnic categories such as African, Asian, Caucasian, Hispanic, Latino, White poorly predict human biological similarity and diversity. As Cuban geneticist Dr. Beatriz Marcheco put it (23),

‘The classic mirror reflects skin color; but the DNA mirror reflects our common ancestors’

Race/ethnicity are thus becoming less and less relevant as proxies for genotype or rather the discernible truth about ancestry lies more and more between rather than within these commonly accepted social categories.

Some Examples Where Misapplication Of Ancestry Obfuscates Rather Than Clarifies Cause For Disease Predisposition

Hypertension – Wikipedia

Hypertension and its clinical outcomes such as heart disease, stroke and renal failure are so much more prevalent among African Americans that a racial predisposition ascribed back in the 20th century still erroneously prevails as a dogma (19). Erroneous because large studies comparing West African, Caribbean and American Blacks show high prevalence of hypertension among African Americans is an outlier, being lower among other Blacks (24, also see figures below from 19).

• Low blood pressures in rural West Africa that change little with age.
• Similar average blood pressures to White North Americans among West Indian Blacks.
• Higher blood pressure among urban African Americans from Maywood in Chicago.
• Obesity, high sodium and low potassium intake, the lifestyle factors known to increase blood pressure matched blood pressure averages among these three groups of Blacks. In other words, abrupt diet and lifestyle changes better explain hypertension rates among African Americans.
• The specific example of hypertension reveals how difficult it is to assert which is more consequential, nature or environment, simply because it’s practically impossible to observe the obverse, people from Africa leading a US lifestyle without experiencing either racial or class inequities.

An even larger study of 85000 subjects including Whites from 8 surveys in the US, Canada and Europe, and 3 surveys among Blacks in Africa, the Caribbean and the US, showed preventable causes of hypertension overlap across races and ethnicities (25). Meantime a much smaller (n = 1056) US study (26) on African Americans served as the basis for the US FDA’s approval of a hypertension drug, BiDil (Isosorbide dinitrate/hydralazine – Wikipedia), supposedly designed for African Americans (27, 28).

Moral? Far from vaunted impartiality, an example of how economics (patents) and politics (tokenism) trump science.

The issue of causality is further complicated by the fact that blood pressure regulation is extremely complex and thus unlikely to be explained by genes alone. One of the largest blood pressure GWAS (Genome-wide association study – Wikipedia) examined 200000 subjects and found the 29 genetic markers most strongly associated with blood pressure could only account for 23% of risk for hypertension (29). Since lifetime hypertension risk in the US is ~85%, this means genomics has so far provided little by way of predictive value. Cherry on the cake is blood pressure susceptibility variants are similar among subjects with African, Asian, European and South Asian ancestry (29).

Diabetes

Judenkrankheit or Jews’ disease, as recently as 1904, this is how physicians in the US and Europe tended to perceive diabetes (see below from 30).

‘THERE IS NO RACE, WHICH is so subject to diabetes as the Jews,” wrote W. H. Thomas in 1904 in the eugenically obsessed language of his day. Thomas, a New York physician, was voicing an almost universally held belief in the United States that of all the “races,” Jews had the greatest likelihood of developing diabetes. At the same time, most members of the medical community considered the prevalence of diabetes among Blacks to be unusually low. In the words of a Johns Hopkins physician in 1898, “Diabetes is a rare disease in the colored race”.’

Fast forward 100 or so years and in the US,

• Diabetes rates have sky-rocketed among African Americans to 2X those in Whites while they’ve declined among Jews.
• Today, Pima people – Wikipedia have the highest rate of Diabetes mellitus type 2 – Wikipedia in the world and of course, since they form a homogenous group, unsurprisingly, mapping their genetics has become an intense focus of research interest. Laughable weren’t it so soul-crushingly tragic for the following reasons.
  • Before the advent of European American encroachers on their land after the American Civil War, the Pima had a reputation for excellent farming and lived independent, autonomous lives along the Gila river on lands presently known as Arizona (31). They even called themselves Akimel O’Tham or the River People.
  • White settlers directly competing for irrigation rights, the 1877 Desert Land Act which ‘required bona fide application of water to the land to obtain title‘, new dams that re-directed water away from traditional Pima farms (9), all these human-made interventions forced Pima to abandon their age-old ways of life in a matter of a few decades.
  • A health survey in 1902 found a single case of diabetes among the Pima. By the 1930s, this number had increased to >500.
  • With the completed Coolidge Dam not sending enough water their way, their traditional farming essentially going bust, the Pima quickly sank into abject poverty and started dying early.
  • Like a benighted god the US federal government rode to the rescue, providing Pima free government surplus food and what food it is! Refined white flour, processed cheese, lard, candy, chips, refined sugar, grape juice, macaroni when the Pima’s original diet consisted of (32 quoted in 9).

‘…seeds, buds, fruits and joints of various cacti; seeds of the mesquite, ironwood, palo verde, amaranth, salt bush, lambsquarter, horsebean and squash; acorns and other wild nuts; . . . roots and bulbs of the sandroot (wild potato) . . . deer, antelope, ..rabbits, quail, dove, wild ducks, wild turkey.’

• • By the mid-20th century, this ancient diet had been entirely supplanted by boxes and boxes of macaroni and cheese. Where Pima dietary fat intake was 15% in the 1890s, it had increased to an incredible 40% by the 1990s (33).
  • And yet it apparently sounds eminently reasonable and soundly scientific to probe and probe Pima genetics to sincerely try to understand their sky-high rates of diabetes these days (34). An exercise in callousness, ignorance, stupidity or all three.

At this point it becomes necessary to ask whether it is really reasonable to highlight ancestry as a mechanistic contributing factor to diabetes when rates can be evidently higher than the norm and drop to average in just 100 years in one group while they increase and increase in two other groups over the same period?

A simpler explanation is how abrupt diet and lifestyle changes impact life trajectories and chronic disease risk in the short-term. Plausible proof? Traditional rural dwelling societies practicing ‘traditional culture’ have vanishingly low rates of diabetes compared to their counterparts newly adapted to ‘westernized’ diets and lifestyle (see below from 9).

Attitudes ranging from the cavalier to sheer ineptitude suggest the prevailing culture of biomedical science is ill-equipped to deal with divisive political topics such as ancestry. Science exists within society, not outside of it and the prevalent untenable allegiance to the implausible notion of striving to be perceived as ahistorical and apolitical ill-serves biomedical science and society alike.

And so we’re back where we started, namely unable to parse environmental and genetic factors in assigning causes to many, especially multi-factorial diseases. Impasse largely owing to biomedical scientists ignoring sociology when probing the role of ancestry, specifically race/ethnicity, in diseases.

Bibliography

1. Schwartz, Robert S. “Racial profiling in medical research.” New England Journal of Medicine 344.18 (2001): 1392-1393.

2. Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity

3. Knerr, Sarah, Dawn Wayman, and Vence L. Bonham. “Inclusion of racial and ethnic minorities in genetic research: advance the spirit by changing the rules?.” The Journal of Law, Medicine & Ethics 39.3 (2011): 502-512. https://www.ncbi.nlm.nih.gov/pmc…

4. Smart, Andrew, et al. “Social inclusivity vs analytical acuity? A qualitative study of UK researchers regarding the inclusion of minority ethnic groups in biobanks.” Medical Law International 9.2 (2008): 169-190.

5. Fullwiley, Duana. “The molecularization of race: institutionalizing human difference in pharmacogenetics practice.” Science as Culture 16.1 (2007): 1-30.

6. Fullwiley, Duana. “The Biologistical Construction Of RaceAdmixture’Technology And The New Genetic Medicine.” Social studies of science 38.5 (2008): 695-735. http://beck2.med.harvard.edu/wee…

7. TallBear, Kimberly. Native American DNA: Tribal belonging and the false promise of genetic science. 2013.

8. Fujimura, Joan H., and Ramya Rajagopalan. “Different differences: The use of’genetic ancestry’versus race in biomedical human genetic research.” Social Studies of Science (2010): 0306312710379170. http://www.ssc.wisc.edu/soc/facu…

9. Duster, Troy. “A post‐genomic surprise. The molecular reinscription of race in science, law and medicine.” The British journal of sociology 66.1 (2015): 1-27. http://geneticsandsociety.org/do…

10. Burchard, Esteban González, et al. “The importance of race and ethnic background in biomedical research and clinical practice.” New England Journal of Medicine 348.12 (2003): 1170-1175. https://www.researchgate.net/pro…

11. Grosse, Scott D., et al. “Sickle cell disease in Africa: a neglected cause of early childhood mortality.” American journal of preventive medicine 41.6 (2011): S398-S405. https://www.researchgate.net/pro…

12. Cutting, Garry R., et al. “Analysis of four diverse population groups indicates that a subset of cystic fibrosis mutations occur in common among Caucasians.” American journal of human genetics 50.6 (1992): 1185. https://www.ncbi.nlm.nih.gov/pmc…

13. Zvereff, Val V., et al. “Cystic fibrosis carrier screening in a North American population.” Genetics in Medicine 16.7 (2013): 539-546.

14. Black? White? Asian? More Young Americans Choose All of the Above. The New York Times, Susan Saulny, January 29, 2011. More Young Americans Identify as Mixed Race

15. The Rise of the American ‘Others’. The Atlantic, Sowmiya Ashok, August 27, 2016. More Americans Are Selecting “Some Other Race” on U.S. Census Forms

16. Levy, Samuel, et al. “The diploid genome sequence of an individual human.” PLoS Biol 5.10 (2007): e254. http://journals.plos.org/plosbio…

17. Ahn, Sung-Min, et al. “The first Korean genome sequence and analysis: full genome sequencing for a socio-ethnic group.” Genome research 19.9 (2009): 1622-1629. Full genome sequencing for a socio-ethnic group

18. Tayo, Bamidele O., et al. “Genetic background of patients from a university medical center in Manhattan: implications for personalized medicine.” PLoS One 6.5 (2011): e19166. http://journals.plos.org/plosone…

19. Cooper, Richard S. “Race in biological and biomedical research.” Cold Spring Harbor perspectives in medicine 3.11 (2013): a008573. Race in Biological and Biomedical Research

20. Santos, Hadassa C., et al. “A minimum set of ancestry informative markers for determining admixture proportions in a mixed American population: the Brazilian set.” European Journal of Human Genetics (2015). http://www.nature.com/ejhg/journ…

21. Mersha, Tesfaye B., and Tilahun Abebe. “Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities.” Human genomics 9.1 (2015): 1. Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities

22. Ossorio, Pilar N. “Myth and mystification: The science and race of IQ.” Race and the Genetic Revolution: Science, Myth, and Culture (2009).

23. Genes Prove Mixed Ancestry of All Cubans: Interview Director, National Medical Genetics Center, Havana

24. Cooper, R., et al. “Hypertension prevalence in seven populations of African origin.” Am J Public Health 87 (1997): 160-168. http://ajph.aphapublications.org…

25. Cooper, Richard S., et al. “An international comparative study of blood pressure in populations of European vs. African descent.” BMC medicine 3.1 (2005): 1. An international comparative study of blood pressure in populations of European vs. African descent

26. Taylor, Anne L., et al. “Combination of isosorbide dinitrate and hydralazine in blacks with heart failure.” New England Journal of Medicine 351.20 (2004): 2049-2057. http://www.nejm.org/doi/pdf/10.1…

27. Roberts, Dorothy. Fatal invention: How science, politics, and big business re-create race in the twenty-first century. The New Press, 2013.

28. Kahn, Jonathan. Race in a bottle: The story of BiDil and racialized medicine in a post-genomic age. Columbia University Press, 2013.

29. International Consortium for Blood Pressure Genome-Wide Association Studies. “Genetic variants in novel pathways influence blood pressure and cardiovascular disease risk.” Nature 478.7367 (2011): 103-109. https://csg.sph.umich.edu/boehnk…

30. Tuchman, Arleen Marcia. “Diabetes and race a historical perspective.” American journal of public health 101.1 (2011): 24-33. https://www.researchgate.net/pro…

31. Dejong, David H. “Abandoned Little by Little:” The 1914 Pima Adjudication Survey, Water Deprivation, and Farming on the Pima Reservation.” Agricultural History (2007): 36-69.

32. Mark, Albyn K. “Ecological Change in the History of the Papago Indian Population.” Master of Arts thesis, University of Arizona (1960).

33. Demouy, J., et al. “The Pima Indians: Pathfinders of Health. Bethesda, MD: Nat. Inst.” Diabetes Digestive Kidney Diseases (1995).

34. Pearson, Ewan R. “Dissecting the etiology of type 2 diabetes in the Pima Indian population.” Diabetes 64.12 (2015): 3993-3995. http://diabetes.diabetesjournals…

https://www.quora.com/Why-does-ancestry-matter-for-some-medical-decisions/answer/Tirumalai-Kamala

Advertisements

Does analysing and organising statistics count as research?

Tags

Cochrane Library, Meta-analysis, Reproducibility crisis

Not only does analyzing statistics count as research, proper application of statistics is the cornerstone of credible research. I’ll use the example of biomedical research to elucidate the decisive role of statistics in research.

Home – ClinicalTrials.gov is a human clinical trial database run by the US National Library of Medicine (ClinicalTrials.gov). Currently, it lists 196,459 studies with locations in all 50 [US] States and in 190 countries. That’s a lot of trials.

• How could we assess the value and validity of the data human clinical trials generate for a given disease?
• If a particular trial concludes that a treatment approach is 90% effective, can we/should we take it at face value?

This is where statistics come in. In particular, a statistical methodology called Meta-analysis. In meta-analysis, data from numerous clinical trials are statistically analyzed in order to rigorously query the methodology they used and the data they generated, and to identify as-yet-unseen patterns among the data. Combining results from many studies in this manner yields more robust data. How?

• For one, when we aggregate data from different studies, we end up with larger data sets. For example, larger number of study patients in the case of human clinical trials. Obviously, as an example of a hypothetical meta-analysis, aggregate data from 10000 individuals is more applicable to the population at large compared to data from 1000 individuals from a single trial.
• For another, biases inherent to the study researchers in one trial get minimized (normalized) when multiple studies are combined together for analysis. And bias is inherent in all of us, even when we set up a study with the best of intentions to minimize bias in its design.
• Further, subjecting multiple studies to such rigorous, impartial scrutiny reveals flaws, weaknesses, deficiencies that serve as signposts to the larger research community, enabling them to get a better grasp of their own research field by helping them sift out and separate relevant studies from the more questionable ones.

Of course, meta-analyses also have flaws such as their unavoidable reliance on published studies. Since negative results hardly ever get published, this can result in skewing. Nevertheless, in the aggregate, benefits of meta-analyses outweigh their drawbacks. In fact, over the years a particular kind of clinical trial meta-analysis called the Cochrane Library has become the gold standard for rigorous statistical assessment of clinical trial data, helping guide the development of Evidence-based medicine.

I will even go so far as to say that lack of rigorous statistics in basic biomedical research is a key reason for its current reproducibility crisis (1, 2, 3). My personal journey in science enabled me to understand that this is a major problem.

• My Ph.D. project on mycobacteria addressed a question of relevance to the largest vaccine trial for tuberculosis (TB), the South Indian BCG vaccine trial.
• Conducted in South India on ~360,000 people spread across 209 villages and 1 town, it showed that BCG didn’t protect against adult TB. Why?
• While that question still hasn’t been answered conclusively, prior exposure to environmental mycobacteria emerged as a plausible factor.
• What kinds of environmental mycobacteria and in which environments? In a nutshell, my Ph.D project had to answer these questions. But how could I, just one person, cover such a vast area?
• My road map was an experiment design worked out with the help of a trained statistician.
• Every step of my Ph.D project was rigorously vetted and co-ordinated by this trained statistician using rigorous statistical tools.
• And it didn’t stop with just the experiment design but worked its way through every step of the research process.
• In fact, I still remember how I’d come back from the study area with my bottles of soil, water and dust samples, and hand over my sample list to the statistician. He’d come back with a list of randomly generated numbers. I’d turn my back, and he’d rub out all my labels and re-label my sample bottles with those random numbers. Samples would be decoded only after I’d generated all the data from each sample. That’s how rigorous the study was.

Then I came to the US and found, instead of equivalent rigor, a process of basic biomedical research that was notable for being totally arbitrary and capricious.

• In basic biomedical research, scientists design experiments how they like, can and do drop animals from the final data set on a whim, and then use statistics rather loosely and arbitrarily at the back end to try to make sense of the data they generate.
• In fact, misuse statistics is a more accurate way of putting it since statistics isn’t used to draw up the study design at the front end, rather only for analyzing the data at the back end.
• Even more shockingly, consulting trained statisticians anywhere in this process is a rarity, certainly not the norm.
• Thus, increasingly over the last few decades, even as the mouse model became the de facto experimental model in both basic and applied biomedical research, an anything-goes, Wild West type of science culture accompanied its experiment design and data analysis.
• Ironically, at the same time, the statistical science behind human clinical trials only became more, not less, rigorous.

• We may finally be coming to the proverbial light at the end of tunnel with the recent publication of the first randomized clinical trial in mice (4).
• Applying the rigorous statistically undergirded clinical trial model to basic biomedical research, i.e., preclinical research, makes the data generated from such models more rigorous and therefore more credible.
• And it also increases opportunity for those trained in statistics because their expertise is now needed and valued across a larger spectrum of biomedical research rather than remaining confined to the purview of human clinical trials alone.

This is why even though I’ve worked in basic research since I came to the US, I know that none of this work approaches the rigor of my Ph.D. study. After all, it was the only one where statistics were properly applied from beginning to end by a trained statistician, starting with the study design itself. So an emphatic yes, in my book, analyzing and organizing statistics counts as research of the highest order, as long as it’s done with guidance and input from trained statisticians rather than at the dictate of statistical novices, be they colleagues or bosses.

Bibliography

1. Prinz, Florian, Thomas Schlange, and Khusru Asadullah. “Believe it or not: how much can we rely on published data on potential drug targets?.” Nature reviews Drug discovery 10.9 (2011): 712-712.  Page on wustl.edu
2. Begley, C. Glenn, and Lee M. Ellis. “Drug development: Raise standards for preclinical cancer research.” Nature 483.7391 (2012): 531-533. Page on mckeonreview.org.au
3. Landis, Story C., et al. “A call for transparent reporting to optimize the predictive value of preclinical research.” Nature 490.7419 (2012): 187-191. Page on nih.gov
4. Llovera, Gemma, et al. “Results of a preclinical randomized controlled multicenter trial (pRCT): Anti-CD49d treatment for acute brain ischemia.” Science Translational Medicine 7.299 (2015): 299ra121-299ra121. Page on sciencemag.org

https://www.quora.com/Does-analysing-and-organising-statistics-count-as-research/answer/Tirumalai-Kamala

With medical research, isn’t there a risk of missing the individuals that have atypical response/reactions?

Tags

N-of-1 Trials, Nicholas J. Schork, Precision Medicine

Assuming the question pertains to atypical responses/reactions to drugs as an example of the outcome of medical research, a brief overview of the FDA drug approval process, a prototype of the prevailing system, helps highlight its major shortcoming w.r.t. rapidly and economically identifying those who might be harmed or not helped by a drug (see figures below from 1).

Problem: Current Drug Approval Process Prioritizes Average, not Individual, Response

Clinical trials typically assess a few responses from a few thousand people. As well, the system in place isn’t optimized for uncovering atypical/adverse reactions early in the process. It also takes inadequate account of genetic (age, gender, ethnic, etc.) and lifestyle (activity, diet, metabolic profile, etc.) differences. This drug approval process evolved post-WW-II. Over the same period, several crucial bodies of knowledge increased, some, such as medical knowledge, considerably, others, such as medical tools and technologies, even exponentially. As a result, drug responses can be assessed more comprehensively than previously possible. Scientific understanding of underlying mechanisms of action have also substantially improved in the interim so much so that it’s now becoming ever-apparent that we’re living through a transitional period with a rather imprecise 20th century medicine in place even as our medical and scientific knowledge has brought us to the cusp of personalized medicine.

As Nicholas J. Schork highlighted in a recent opinion piece in Nature (2), every day millions of people take medications that won’t help them. As he highlights in a compelling figure, the top ten highest-grossing drugs in the US help between 1:25 to 1:5, i.e., only 4 to 25%  of those who take them (see figure below from 2).

Schork also points out that drugs such as statins, routinely used to lower cholesterol, may benefit as few as 1:50, i.e., 2% (3). Why? Not because of deliberate, intentional oversight or laxity but simply because the current drug approval process prioritizes average, not individual, efficacy. Logical step then is to move from average to individual responses, i.e., N-of-1 trials. Schork suggests that aggregates of N-of-1 trials could better assess true efficacy and adverse outcomes of drugs.

Possible Solution: N-of-1 Trials are More Likely to Deliver Precision Medicine
N-of-1 trials would collect a lot of data from one person, sometimes every day, often for months or years. Schork mentions one example from Australia (4).

• 132 osteoarthritis or chronic pain patients took different drugs over three years.
• Reported pain levels, swelling and other associated symptoms were measured every 2 weeks for 12-week periods, when patient was off/on a particular drug.
• Pre- and post-Rx data comparisons enabled more effective Rx.
• Costs of such an approach were, unsurprisingly, much higher.

Cost is definitely a bottleneck in personalized medicine (6). Making patient re-imbursements easier would be an obvious measure. Molecular tests that stratify patients according to likelihood of response to or safety of a drug, i.e., Companion diagnostic, are key building blocks of personalized medicine. Cohen and Felix’ analysis (6) of the US healthcare market suggests recent measures by insurers to cover them will help build an evidence base for companion diagnostics. This is the first step necessary to accelerate the personalized medicine process.

Apart from costs, the other consideration is whether personalized medicine is actually a reality at present. While we can measure ever more responses, for many diseases we’re still far from understanding what’s relevant, i.e., Biomarker that are reliable stand-in/surrogates for disease onset or progression or for drug response.

Obviously critical knowledge gaps remain. However, Schork says physicians are already doing N-of-1 trials in an ad hoc manner. For e.g., prescribe one medication, monitor its effect then try another. Anyone who’s ever been to a doctor would be familiar with this scenario. However, this process hasn’t yet been formalized into a rigorous, clinical trial approach. The crux is to transform this ‘everyday clinical care‘ into N-of-1 trials (2).

The key pillars of such an approach would include (2),

• Genetic data, so-called ‘omics‘: Getting ever-easier to assess, this includes not just DNA, RNA, but also blood metabolites (metabolome) and microbiota (microbiome).
• Personalized health monitoring: Becoming ever-accessible with devices such as wearable electronic monitors, continuous glucose monitors, portable electroencephalogram (EEG) monitors, etc.
• Institutional support for precision medicine: Governments, regulatory agencies (7), funding agencies are increasing their support. For e.g., in the US, President Obama announced a $215 million national Precision Medicine Initiative (8). The FDA ‘s even stated that the era of one-size-fits-all medicine may even be over (7).

Another key step: Atypical/adverse outcomes could be uncovered earlier by assessing standard outcomes
Obviously, current clinical trial design doesn’t prioritize uniform outcome measurements. Different trials measure different outcomes, even for the same disease. This is another area where considerable standardization is necessary. One approach to standardization is the Core Outcome Measures in Effectiveness Trials (COMET) Initiative to measure uniform clinical trial outcomes, the brainchild of Paula Williamson, a statistician at the University of Liverpool, UK (see figure below from 5). In the US, the NIH’s NLM (National Library of Medicine) launched a similar initiative, a database of ‘core data elements‘ that NIH Institutes recommend or require in trials they fund (5).

Today’s drug approval process certainly carries greater risk of missing atypical/adverse drug or intervention reactions, especially early in the process. Over the next few decades, certainly 20th century medicine will morph into personalized, precision medicine, and as this process gathers pace, this risk will automatically decline or least that’s the hope.

Bibliography

1. http://www.fda.gov/downloads/Dru…
2. Schork, Nicholas J. “Personalized medicine: Time for one-person trials.” Nature 520.7549 (2015): 609-611. Page on researchgate.net
3. Mukherjee, Debabrata, and Eric J. Topol. “Pharmacogenomics in cardiovascular diseases.” Progress in cardiovascular diseases 44.6 (2002): 479-498.
4. Scuffham, Paul A., et al. “Using N-of-1 trials to improve patient management and save costs.” Journal of general internal medicine 25.9 (2010): 906-913. Page on nih.gov
5. Keener, Amanda B. “Group seeks standardization for what clinical trials must measure.” Nature medicine 20.8 (2014): 798-799.
6. Cohen, Joshua P., and Abigail E. Felix. “Personalized medicine’s bottleneck: diagnostic test evidence and reimbursement.” Journal of personalized medicine4.2 (2014): 163-175. Page on mdpi.com
7. Simoncelli, T. “Paving the way for personalized medicine: FDA’s role in a new era of medical product development.” Silver Spring, MD: US Food and Drug Administration. Published October (2013). Page on fda.gov
8. Precision Medicine Initiative

https://www.quora.com/With-medical-research-isnt-there-a-risk-of-missing-the-individuals-that-have-atypical-response-reactions/answer/Tirumalai-Kamala