Cancer is the leading cause of death in Singapore. But early detection can make a huge difference to patients – both for their survival, and saving money on hospital bills.

If clinicians could detect cancers at stage one or zero, they can get the treatment they need earlier, and “we don’t have to give them chemo”, says Dr Joanne Ngeow, Head of the Cancer Genetics Service at National Cancer Centre Singapore.

If the challenge is to diagnose cancers earlier, our genes could be the key, Dr Ngeow believes. GovInsider caught up with her to learn more about how genetic data can help with early detection and prevention of cancer.

Finding mutations hiding in DNA

Dr Ngeow’s goal at the Cancer Genetics Service is to identify patients with a high risk of cancer from a young age. Genomics has become more widely used in Singapore over the last few years, and today, patients from across Singapore’s hospitals are referred to the cancer centre if they have a strong family history of the disease, a rare form of it, or multiple forms, Dr Ngeow says. “We use either blood or saliva to detect what genetic mutations they might have that might increase the risk.”

Those with the highest risk of certain cancers will benefit most, and these are the groups the centre is concentrating on. “It’s been well shown that if you were to treat them like all patients, you would miss things, so you need to be more intensive at different time points,” she argues.

Genomics can broadly be used in two ways – to understand the biology of tumours, and to understand the cancer risks of individual patients on a genetic level. The former is more focused on the actual treatment of cancer, according to Dr Ngeow, while the latter is a preventive strategy against late-stage cancer. And from a healthcare system perspective, the latter is “less costly” in the long run, she points out.

Early detection

The value that genomics brings to the table is that patients’ families can identify their own individual risks of getting the same disease. “We offer the testing to their respective family members, because they may have a certain probability of actually sharing the same mutation.” On the flipside, family members of patients who test negative will not need to go for early screenings, she points out: “In that way we avoid a lot of unnecessary testing for people who don’t need it.”

“In that way we avoid a lot of unnecessary testing for people who don’t need it.”

“One issue with treating patients only after they have been diagnosed with cancer is that very often, it’s actually really difficult to achieve a cure,” she notes. By harnessing genomics, “we can become more precise in terms of offering patients better treatment”.

Meanwhile, genetic counsellors at the cancer centre are supporting doctors by freeing up their time to see more patients – and in this way, helps the centre to maximise on resources. These counsellors hold master’s degrees and cannot physically examine patients, but instead, help patients understand what genomics is, and support patient education in other ways.

Just one genetic counsellor helped Dr Ngeow’s group to grow its capacity by 350%, where the number of patients seen rose from two per week in April 2014 to seven per week in November 2014, according to a paper published by the group. This did not affect patient satisfaction: “Genetic counsellors are equivalent to physicians in leading counselling sessions across outcomes such as patient anxiety, satisfaction, and knowledge”, Dr Ngeow and her colleagues wrote.

The costs of genomics has come down significantly, too: right now, genetic testing for hereditary cancers, or to identify cancer risks in those with a strong family history, costs less than S$1,000 (US$730), Dr Ngeow continues.

Where AI factors in

To make a difference in treatment, cancer needs to be diagnosed and tackled much earlier on. Artificial intelligence could potentially be used in “predictive analytics” to support the advance of genomics, and provide much more accurate predictions in the future, Dr Ngeow believes.

AI can be linked to an electronic medical record to send alerts to patients with increased chances of cancer, for example, Dr Ngeow says. Take a hypothetical patient diagnosed with breast cancer at 30: “An alert should probably come up to say you need to consider whether this is genetic, because you are very young,” she explains. “For a patient it might pop up and say, ‘you really need to consider this condition and that condition’, anytime anything new is added to a particular patient’s history.”

AI can also be used to process larger amounts of data on patients, pulling out patterns and classifying risk groups. These patterns could identify patients that are predisposed to a certain cancer, but are not already aware of it, Dr Ngeow continues.

And AI can help understand and predict patient behaviour outside of the hospital as well, where environmental exposure and other factors can influence their recovery, explains Dr Ngeow. “Let’s say somebody needs to go for surveillance – colonoscopies every year. We can understand first of all if our patients are doing that; if they are not, what are some of the factors that are influencing why they are not going?” she says.

Genomics is particularly exciting for disease prevention in other ways – it is making a difference in stemming the spread of epidemics. In 2015, the Genome Institute of Singapore stopped a Group B streptococcus (GBS) outbreak from spreading by identifying exactly what was making patients sick. “We were able to show that the GBS causing infections in human patients was the exact same as the GBS we isolated from raw fish samples from the stalls they ate at,” Dr Swaine Chen, Senior Research Scientist at the institute, told GovInsider.

Singapore is prioritising a shift to preventive healthcare, and at the same time, emphasising the need for cost-effective treatments. When it comes to cancer, genomics could be a key piece in this new shift.