Tuesday, July 23, 2024
Homecna_insider singaporeA lab for bespoke drug treatments — where AI, researchers find ways...

A lab for bespoke drug treatments — where AI, researchers find ways to help cancer patients

SINGAPORE: He had just completed the “great experience” of full-time National Service and was looking forward to the future after having studied gaming at the Institute of Technical Education.

Four months later, in January 2020, Donavan Koh’s voice began to grow hoarse.

What he initially thought was a sore throat worsened to the point where his parents could not hear what he was saying. His father had a hunch that it was abnormal and got him to see a doctor.

The result of the biopsy shocked Koh: It was Stage 4 natural killer (NK) cell lymphoma, a “very rare” variant of lymphoma, diagnosed his doctor, Daryl Tan, a haematologist in private practice.

Lymphoma is a cancer that starts in infection-fighting white blood cells called lymphocytes, which are part of the body’s immune system, according to the National University Cancer Institute, Singapore (NCIS).

On his doctor’s advice, Koh immediately started chemotherapy, the first-line treatment for his condition.

The side effect made him cry. “After two months, my hair started to fall,” said Koh, who had not told his friends about his condition. “I was devastated because I care a lot about my appearance.”

Worse still, the treatment did not work.

It was time to put him on standard second-line drugs. Only about 40 per cent of patients, however, respond to this treatment. Koh was not one of them, and his situation was dire at this point.

“There isn’t any proven way to treat the disease after a patient’s relapsed or progressed beyond the first one or two lines of treatment,” Tan said.

Driven by robots and artificial intelligence, QPOP accomplished the task in the span of about six days — which would be impossible with manual tests, said Edward Chow, an associate professor at the N2CR and principal investigator at the Cancer Science Institute of Singapore (CSI Singapore), NUS.

His team started working on QPOP in 2012 and have applied it to medical settings since around 2017. A collaboration among molecular biologists, cancer researchers and engineers, QPOP “marries the wet lab with computer science and mathematical analysis”, said Chow.

Simply put, the platform is “a bartender lab for cocktails of drugs”, said fellow N2CR assistant professor and CSI Singapore principal investigator Anand Jeyasekharan.

“Edward’s lab (team) are like expert bartenders. They come up with the best cocktails of these anti-cancer drugs, which are … bespoke for our particular patients,” said Jeyasekharan, who is also a consultant in the department of haematology-oncology at the NCIS.

He added: “There are plenty of new drugs being discovered every day, but we’re often still uncertain as to how best to use these new drugs, which patients should be receiving a particular new drug over another, (and) in what combination.”

These are the “unknowns” that Chow’s team and QPOP seek to solve.

WATCH: Personalised cancer treatment (6:51)


Chow’s motivation for doing cancer research is personal: When he was in graduate school, his father was diagnosed with renal cancer.

Chow witnessed the progression of the disease as his father went through “the different lines of standard therapy until, ultimately, he ran out of options”.

“We did feel helpless, especially because his treatment was very regimented, so I wanted to see if I could do anything to make sure that people have as many options as possible,” he said.

He decided to switch his area of research — which was on the effects of antiviral immune response on drug metabolism — to cancer, for more clinical impact and “to give cancer patients and their families as much time together as possible”.

And that was what he did for Koh.

QPOP ranked two drugs highly for the patient, but both were not approved for use in Singapore at the time. The medical team successfully sought to bring the drugs here with the Health Sciences Authority’s approval.

To the doctors’ surprise, Koh, who had been unable to talk intelligibly for about six months, regained his voice after the first treatment cycle, said Tan.


The scientists will keep working to make individualised treatment a reality for more patients. “There’s still more to be done. We still need to figure out how to keep (cancer) away once it’s gone away,” said Jeyasekharan.

The researchers believe QPOP can ultimately apply to all types of cancer, but this will depend on advances in culturing cancer cells and tumour tissues in the laboratory, Chow said.

He and Jeyasekharan want to find a way to incorporate more drugs into QPOP’s drug set — not only drugs that work directly on cancer cells, but also those that help patients’ immune cells to target and attack cancer cells.

- Advertisment -

Most Popular