We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.


The Road to a Sustainable Precision Health Ecosystem in Taiwan

The Road to a Sustainable Precision Health Ecosystem in Taiwan content piece image
Credit: iStock
Listen with
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 6 minutes

Numerous databases used in healthcare are based on data from Caucasian patients and research participants, but predictions derived from these data aren’t always as accurate when applied to patients from other ethnicities. The Taiwan Precision Medicine Initiative aims to use Han Chinese-specific data to optimize patient care in Taiwan. The global Han Chinese population is around 1.4 billion people, with 95% of the Taiwanese population having Han Chinese ancestry.

Technology Networks spoke to Dr. Pui-Yan Kwok, the leader of the Taiwan Precision Medicine Initiative, about what the initiative is doing to implement a sustainable precision health ecosystem in Taiwan, the challenges facing precision medicine and what the initiative hopes to achieve in the future.

Katie Brighton (KB): What impact could precision medicine have on healthcare?

Pui-Yan Kwok (PK): Precision medicine means different things to different people. Some people think of it as finding mutations that drive a patient’s cancer and then matching that patient with the best therapy. Another area of precision medicine is sequencing genes and identifying mutations that cause rare genetic disorders. This is a research exercise for many rare genetic disorders. For some patients, it can be life-changing if a treatment strategy can be identified that alleviates their symptoms or improves their prognosis. Both areas of precision medicine (cancer and rare genetic disease) are a work in progress.

A third understudied area of precision medicine, and what I’m most interested in, is the prediction of common disease risks so we can match patients with management plans to prevent or slow down disease, or help guide optimal treatments. This area of precision medicine is not new; when your doctor asks if you have any diseases that run in your family, that’s shorthand for genetic risk. As we are better able to define genetic profiles to predict disease risk more accurately and precisely it will help enable clinicians to practice better evidence-based medicine.

KB: Can you tell us a bit more about what the Taiwan Precision Medicine Initiative is and what its aims are?

PK: We launched the Taiwan Precision Medicine Initiative, or TPMI, in 2019 as a joint effort across 16 hospital systems to establish a sustainable precision health ecosystem in Taiwan. In the project’s first phase, we partnered with Thermo Fisher Scientific to design a custom genotyping research array for identifying genetic factors associated with the risk of diseases relevant to the Han Chinese population. We have now enrolled more than 500,000 people and genotyped 400,000 of them across Taiwan, so we are well on our way toward the goal of building a cohort of 1 million people.

TPMI is different from many other genetic studies in that our participants are working with our researchers interactively – it’s not a one-way street. That interactive participation enables us to collect longitudinal data from participants on an ongoing basis. We also provide them feedback on their risk and guidance on how they can manage their risk. If we can encourage them to adopt those guidelines to manage their health, we’ll be able to show that this approach is helpful and see, over time, the impact of precision medicine both at an individual and cohort level.

We are starting to return genetic test results to participants now through our website. Over time, as we analyze genetic and clinical data, we’ll generate robust polygenic risk score algorithms for each disease and can then share common disease risk results with participants. We are also working on an app so we can relay results and help guide how participants use that information. Clinicians may be reluctant to use research results to guide patient care, understandably. Our goal is to take the burden off clinicians by providing resources such as medical societies and educational sites participants can turn to for information. We have also talked about establishing clinician-moderated chat groups for people at high risk for certain diseases to ask questions and get credible answers.

KB: What are the challenges in implementing precision medicine and how does the TPMI aim to address these?

PK: The biggest challenge in implementing precision medicine is the need to establish strong scientific validation for health management guidelines. When I sit across from a patient, I need to be able to say, “based on your risk, this is what we should do.” But having that kind of clinical evidence will take time. Right now, we have studies based on existing data. These provide statistics associating genetics with disease risk, but we don’t yet have clinical evidence showing how individuals can improve their life expectancy using this information.

In a few years, we can figure out if our predictions are true for common diseases such as diabetes and hypertension. With ongoing longitudinal studies, we can ask patients if they want to follow the standard of care or the disease risk-based guidelines for screening, intervention, etc. We can then look for health outcome differences between the two groups and determine if the disease risk-based guidelines are effective.

KB: How did you select the technology to power TPMI?

PK: We chose a custom Axiom genotyping research array from Thermo Fisher based on three things. First, the flexibility of design. Being able to tailor the single-nucleotide polymorphism content to populations of different ancestries is very important since every ethnic group has an optimal set of markers. With a big project there are always challenges, and Thermo Fisher has a very responsive applications team that has helped us through technical challenges to achieve our goals.

Second, we worked closely with the analytical team at Thermo Fisher to make sure the genetic testing was robust. With clinical testing, when you get back an inconclusive result you can run the test again. However, with big studies when you’re genotyping tens of thousands of people with several hundred thousand markers, you can’t run tests again and again; you need the results to be credible. We have two sets of markers on our array – genetic testing markers and genetic profiling markers, and those are very different. With genetic profiling markers, it does not affect the results too much even if 5% fail. But for genetic testing, every failed marker is a failed test and useless data. There are tens of thousands of those markers you don’t want to miss and you can’t afford wrong results.

And third, the cost and ease of implementation. If you’re doing research on millions of people the technology needs to be affordable. The cost of using the Axiom array platform is actually very manageable; it is no more expensive than a cholesterol test.

KB: Can you explain why it is so important for genetic studies to be diverse?

PK: All genetic experts worry about the lack of diversity in genetic research to some extent. Disease susceptibility is different based on genetics – a lot of physiological diseases across backgrounds are based on genes. For historical and economic reasons, most big studies have been done with Caucasians. When you use an algorithm developed from the European data to analyze Chinese populations the results are not as strong as when you use that algorithm to look at a European population. With early data from TPMI, we have already shown we can bring predictor values up for the Han Chinese population by using algorithms developed from Han Chinese data.

With custom technology we’re now able to capture similar disease risk variants for other populations but very little has been done in non-European populations. Countries have limited resources to get studies done, so it’s important that testing is accessible and affordable.

KB: Are there any genetic conditions that are particularly prevalent in the Han Chinese population that you will be screening for? What other diseases will be screened for?

PK: It’s no secret that some diseases affect Asians differently. For example, in the U.S. and western countries type 2 diabetes is associated with high BMI. In Asia, people without high BMI still get type 2 diabetes; it’s actually quite prevalent. There’s also a high rate of lung cancer among non-smoking women in Asia. We plan to develop risk prediction algorithms for at least 20 common diseases relevant to Han Chinese people and we’d like to be able to drill down into the risk of more serious disease subtypes for people with these conditions.

As one example, when you have uncontrolled hypertension you’re at risk of stroke, but are there certain genetic factors that will make you more prone to that outcome? If we could determine the people at the highest risk for serious diseases, we can focus our resources on them. As we need large numbers of cases for most diseases to extract the maximal genetic signal for them, our model predicts that by the time we reach 1 million people, we’ll get to maximal genetic risk prediction for the common diseases in Han Chinese people.

KB: How long might it take to translate the research by TPMI into the clinic?

PK: If we strictly follow current public health practice, it will take some time to amass enough scientific evidence for clinicians to start guiding patient care based on TPMI results. However, if participants choose to trust these predictions, then they can use that information to make lifestyle decisions much sooner.

While our participants may be able to see some benefits from early TPMI results, my goal is to develop a more comprehensive healthcare management plan for each person. By designing an integrated risk calculator combining genetic and non-genetic information (environmental exposure, lifestyle, etc.), we can establish healthcare management guidelines including nutrition, lifestyle, disease screening and treatment to help people stay healthy longer.

KB: The TPMI reached the 500,000 participants milestone earlier this year. What’s next?

PK: The bigger the number the better – the more participants, the more indications, subtypes and categories we can create. But research funding is not unlimited and our goal is to reach 1 million participants to cover major diseases.

From there, the next step is to offer the genotyping test to others so they can leverage the data for their own disease risk assessment and then have the option of joining a community of people who share their clinical data. We really want everyone of Han Chinese ancestry to take this test and benefit. In return, we’ll have an ever-expanding dataset. We want this to be patient-focused and to find ways to accurately predict who is going to be at high risk for disease. For most of us, we’ll be at high risk for something. 

Dr. Pui-Yan Kwok was speaking to Katie Brighton, Scientific Copywriter for Technology Networks.