Genetic diagnosis, treatment, monitoring, and surveillance

Table of Contents

Genetic diagnosis, treatment, monitoring, and surveillance
Initial notes
Annotations on common readings
Annotated additions by students
Idea: Genetic analysis has begun to identify genetic risk factors. We need to consider the social infrastructure needed to keep track of the genetic and environmental exposures with a view to useful epidemiological analysis and subsequent healthcare measures. Even in cases where the condition has a clear-cut link to a single changed gene and treatment is possible, there is complexity in sustaining that treatment.

Initial notes

For a few years this last decade, genome-wide association studies seemed to hold promise for detecting genes related to diseases and and the invention of drug-based treatments.

But, in November 2009, , A Genetics Company Fails, Its Research Too Complex. See also 2010 commentaries on low yield from GWAs, e.g., Couzin-Frankel , J.: 2010, Major Heart Disease Genes Prove Elusive. Science 328(5983),1220-1221.
Khera et al. (2016) claims success, but also integrates knowledge about many genes of small effect with environmental/lifestyle factors that ameliorate or exacerbate the effects.

In 2009, Khoury et al. were concerned that the promises were not over-stated. Look at the table giving their quality control proposal.

In 2005, Frank cautions that epidemiology needs as much data about environmental factors as genes, but observes that the playing field is not level. (Give credit if you ever cite this powerpoint.)

Even for (rare) diseases governed by single genes, the path from genetic diagnosis to therapy is complicated as the poster-child case of PKU shows. From Taylor (2009):

Genetic analysis
This week’s readings deal with analysis of actual genes, whereas most of the previous week’s readings about heritability looked at variation in some observed trait. With the expanding role of genetic analysis comes the challenge of how to fit this into our healthcare system. Genetic risk factors associated with common diseases, individually may only have a weak risk-ratio, but combinations of these factors may, researchers hope, have a larger impact of the population.
Khoury et al try to make a case for establishing standards for “presenting and interpreting cumulative evidence on gene-disease associations.” They describe problems such as publication and selection biases, differences in collection and analysis of samples and the presence of undetected gene-environment interactions among studies of genome-wide analysis. This has lead to a high incidence of type 1 errors in GWA studies (false positives). Networks are attempting to establish consensus guidelines for reporting and publishing gene-disease associations to reduce this risk. (Do a web search to see how things have developed in the two years since.)
Bowcock describes how a consortium of 50 British groups examined genetic variance in a genome-wide association study. They examined the genetic issues for 7 common diseases including RA, CAD, bipolar disorders, diabetes, hypertension and Crohn’s disease. To identify the genetic risk factors for these common diseases, they examined 500,000 genetic markers(or SNPs-single nucleotide polymorphisms) from the genomes of 17,000 individuals. They found very little difference between controls and cases, but they did find some SNPs that can be considered genetic risk factors for a particular disease, some confirming previous studies, but others identifying unique genes that affect susceptibility to a disease.
The more advanced the genetic analysis becomes, the issue of how this information is going to be utilized for the treatment or monitoring of a person’s risk for disease and what part prevention and screening plays in the individual’s health status presents itself. Bowcock cautions that translating someone’s risk into “medical practice” should not be done without “larger patient populations, well-annotated clinical databases and sophisticated environmental assessment.”
Frank's powerpoints remind us that knowledge about environmental factors is needed as well, but because it costs more to collect and store, it tends not to be collected. This will make some epi. research questions impossible to address and shape the kinds of knowledge that can be put into biomedical practice and social policy.

Social application of knowledge about genes
The Paul article gives us an example of how a rare genetic disorder, Phenylketonuria (PKU) has been managed. Even though the incidence is between 1 in 11,000 and 1 in 15,000 births, all newborns are tested for it in the US, Canada, Australia, New Zealand, Japan and most other Western and Eastern European countries. The article chronicles the history of instituting the screening procedures for PKU. PKU was described as a “treatable genetic disease.” If left untreated, it results in severe mental retardation and behavioral abnormalities. PKU can be treated by a special diet which eliminates phenylalanine toxicity in the blood of those with PKU. There were policy issues involved in the PKU screening process that warrant examination. What were the societal factors that contributed to the federal initiative in the US in 1961? Not everyone was a proponent of the testing of every newborn for such a rare genetic disease. Problems of treatment efficacy and the question of the “cost” of the program are also addressed.
As we become more advanced in genetic analysis, many similar issues may be encountered for other conditions. One current related topic is the role of BRCA1 and BRAC2 inherited breast cancer gene abnormalities. Although they only account for about 10% of all breast cancers, there is much discussion about the Pros and Cons of seeking your genetic profile for breast cancer. Issues of prophylactic breast removal surgery, discrimination by health insurers and stress and anxiety associated with knowing your genetic profile are all ones that can be related to other genetic testing.
Taylor (2009) looks in broad brush at the overall project of application of genetic information.

Notes and annotations from 2007 course, 2009
Common readings: Khoury 2007 (Many genes as small risk factors), Paul 1998 (Complexities of social support after PKU diagnosis)
Supplementary Reading: Bowcock 2007, Frank 2005, Taylor 2009

Annotations on common readings

Muin J Khoury, Julian Little, Marta Gwinn and John PA Ioannidis. On the synthesis and interpretation of consistent but weak gene–disease associations in the era of genome-wide association studies. International Journal of Epidemiology 2007; 36:439–445

The completion of the Human Genome Project and subsequent advances in genome-wide scanning technologies has enabled researchers to study hundreds of thousands of genetic variants simultaneously, yielding susceptibility findings for common diseases. Khoury and colleagues contend that the “slow and tedious” progress in unraveling genetic risk factors for common chronic diseases (such as diabetes etc.) can largely be attributed to their complex pathogenesis (which involves numerous genetic and environmental risk factors along with their interactions). It is proposed that ‘major’ gene effects, manifesting as Mendelian or single-gene disorder, account for only a small fraction of cases for a given disease. Furthermore, it is noted that larger genetic effects may only be observed in certain population subgroups, prefigured by their genetic background, environmental exposure or disease subtype. However, Khoury and colleagues purport that a combination of even a few small effects (e.g. <20 common genetic variants) could account for a sizeable proportion of the genetic predisposition linked to common complex diseases.

In this commentary, the authors examine the implications of putative associations between single genetic variants and various complex diseases generated by Genome-Wide Association (GWA) studies. Khoury and colleagues’ fundamental assertion is that weak associations (risk ratios (RR) typically<1.5) will be the norm rather than the exception in GWA studies. However, differentiating true weak associations from spurious effects, resulting from a litany of oversights (such as significance-chasing, genotyping errors etc), is emphasized as being a challenge. The authors argue that standards are urgently needed for qualifying evidence of gene–disease associations, particularly for consistent yet weak associations. Khoury and colleagues present a schema for grading the credibility of evidence synthesized through GWA studies; postulating the important criteria domains to be effect size, amount of evidence/replication, protection from bias, biological plausibility and relevance.

The authors indicate that although weak risk factors have little validity (or utility) as predictive or diagnostic tests, they are poised to offer insights to disease pathogenesis, aetiology and environmental risk factors through ‘Mendelian randomization’. Khoury and colleagues conclude that “more sophisticated approaches are needed to integrate information on multiple genetic variants with other biomarkers, as well as physiological and clinical data, for use in population medicine (Khoury et al.).” The authors insist that the capacity of the Human Genome Project to procure medical advances is mediated by the scientific community’s ability to accurately characterize the relationship between genetic variation and human disease. (SY)

Annotated additions by students

(In alphabetical order by author's name with contributor's initials and date at the end.)