2024 Karnofsky Award winner Lillian Siu talks about her career in phase I studies, ctDNA, & her mentor’s evil red pencil

Lillian L. Siu discovered her passion while perusing employment ads in the Journal of Clinical Oncology.

“I literally opened JCO, the Journal of Clinical Oncology, one day, and in the back there was an advertisement for a drug development fellowship in San Antonio, Texas. I read the description and felt, ‘Wow, this is really what I want to do. I want to develop drugs and bring new drugs to patients and bring them hope,’” Siu, director of the Phase I Clinical Trials Program and clinical lead of the Tumor Immunotherapy Program at University Health Network’s Princess Margaret Cancer Centre, said on the Cancer History Project Podcast. “It opened my eyes that this is definitely a field that I would love to be involved in—to take a drug from bench to bedside and then translate back to understand why it works or it doesn’t work.

“To me, that was, ‘Wow, I would love to spend my whole career doing this.’ And I did.”

Siu is also a professor of medicine at the University of Toronto, and a senior scientist, a medical oncologist, and the BMO Chair in Precision Cancer Genomics at Princess Margaret Cancer Centre.

I’m a phase I trialist, so I only have to figure out the dose and go to the phase II trialists and say, ‘Here’s the dose, test it.’ But life is not like this anymore, because many of the drugs now start with phase I and perhaps end in phase I, and then they get registered and approved. So, you have to sort of think about more like a phase III oncologist now in the phase I arena.

The insight led to Siu’s nearly 30-year-long career in novel drug development and phase I clinical trials.

Through her work in early-phase clinical trials, Siu was among the first scientists to read out the signals of safety, pharmacology, and preliminary efficacy of therapeutic agents that ushered in a new era of cancer therapy.

And it’s quite a list: Epidermal growth factor inhibitors, monoclonal antibodies, antiangiogenic multi-kinase tyrosine kinase inhibitors, epigenetic agents such as histone deacetylase inhibitors and hypomethylating agents, PI3K-AKT-mTOR inhibitors, and immunotherapeutic agents like costimulatory molecules and checkpoint inhibitors.

She has been involved in the early development of over 50 drugs and has focused on ways to improve efficiency and scientific relevance of clinical trials.

Ultimately, her career landed her the 2024 David A. Karnofsky Memorial Award, the American Society of Clinical Oncology’s highest scientific honor.

Over the course of her career as a phase I clinical trialist, Siu watched as the field moved away from using “maximum tolerated dose,” the growing need for early-phase efficacy data, and the emergence of intermediate biomarkers.

Consistent throughout was the field’s desire to make clinical trials better for the patients, Siu said.

“There are many like-minded individuals around the world who are interested in developing clinical trials in a better way—more agile, more dynamic, and more efficient,” Siu said. “I’m very glad I went to that area, because when you’re dealing with drugs day-to-day on the hardcore, translational research, having some moments that you back off and think, ‘Are we doing all this right? Could we have done this better and more efficiently? And how can we get at the right dose with the fewest resources and not put patients at low or non-subtherapeutic doses, and at the same time avoid toxic doses?’ These are all really, really interesting thoughts and ideas.”

It is important to think long-term, even in the phase I setting, Siu said. Particularly now, when some drugs—larotrectinib and entrectinib, for example—get  approved before phase III trials are conducted.

“You need to look far enough to think what happens to the end of the experiment, not just tomorrow, and be able to project from the end looking backwards and then work your way back. That makes more sense than to just think, ‘If I get through this experiment or this trial, I’m happy I can publish this.’ I don’t think that’s enough,” Siu said.

“I’m a phase I trialist, so I only have to figure out the dose and go to the phase II trialists and say, ‘Here’s the dose, test it,’” Siu said. “But life is not like this anymore, because many of the drugs now start with phase I and perhaps end in phase I, and then they get registered and approved. So, you have to sort of think about more like a phase III oncologist now in the phase I arena.”

Keeping with this thinking, Siu’s recent interests have gravitated towards developing surrogate endpoints for preliminary efficacy.

“Because drugs are now more active… there’s more interest to make go/no-go decisions early on, so we don’t subject patients to drugs that are not going to be ever helpful to them. With that, I think there is a need to have early readout,” Siu said. “For example, I’m very interested in the area of circulating biomarkers such as circulating tumor DNA, whether we can use changes in ctDNA early on, even in a phase I trial, to understand whether there is efficacy to move forward or not.

“In a phase I trial where patients are very heterogeneous, typically heavily pretreated with many lines of treatment, to see bonafide tumor shrinkage might be challenging. But I think having a molecular assay that allows us to see whether the tumor burden changes on a microscopic level such as with ctDNA is very attractive.”

Siu credits her success to a network of renowned mentors, but more importantly, to quality mentorship.

“I think it is important to have mentors such as Ian Tannock and Malcolm Moore, etc., not just because they’re famous, but because they care about their mentees’ careers,” Siu said. “They met with me. They would suggest, for example, if they had to give a talk but they couldn’t, or they thought that I was a more appropriate speaker, they would put me up for that kind of opportunity. Writing papers together, getting feedback from them.”

Ian Tannock, now emeritus professor of medicine and medical biophysics at Princess Margaret Cancer Centre and University of Toronto, had a teaching method that stuck with Siu.

“Ian may not like me telling the public this, but he had this ‘evil red pencil,’ we called it,” Siu said. “We would dread seeing back a manuscript that is from black turned to red—because it’s all red. You knew you didn’t do a good job when you got back your manuscript and it’s red and not black anymore, because of the evil red pencil. But that’s the opportunity for you to learn.”

Mentorship is more than about imparting knowledge, Siu said. It’s about passing on a way of thinking.

“You think about how giants like Malcolm Moore, etc., think, and after you do this a few times you start to say, ‘Hey, I think this is how they would think, but I’m going to even think more beyond that. I’m going to think one step further and challenge their thinking,’” Siu said. “I think that kind of iteration really helped me grow, in terms of being able to challenge my mentees. I think the most important thing when mentoring someone is not tell them what you think. It’s really to teach them how to think, and there comes a point where they are able to come up with their own ideas or challenge what you say.

“Then, I think, you’ve done something really successful, because you have now stimulated somebody to have their own original innovative thinking.”

Siu is continuing the legacy, albeit with modern tools.

“I don’t have an evil red pencil. It’s all electronic now,” she said.

Siu’s mentees may dread a red-lined document filling their screens, but they have a mentor that encourages them to push past such setbacks.

“Learn from every mistake or every challenge and rise above it and be tenacious,” Siu said. “Be persistent, because there’s never an end that is a bad ending. It is always a good ending if you put enough effort in it. Maybe not entirely the way you want it, but at least if you put in the effort, something will return to you that is worth your effort. I truly believe in that, and certainly I see that in my career.

“I don’t only have positive results, I have very often negative results, but it’s fun. Learning from your mistakes is half of the fun, and cherish that kind of moment to learn from it.”

Siu spoke with Jacquelyn Cobb, a reporter with The Cancer Letter.

Transcript

Lillian Siu: Thank you very much, Jacquelyn.

LS: Like many people in this field, there’s a family history that started the whole interest of studying about cancer. I think it affects so many of us in terms of our family and our friends, and I really wanted to tackle this disease.

Even though it might be a very small contribution, I felt that this was a calling from very early on in my career.

LS: When I was in high school, I was very fond of art class. I had probably had about 200, 300 drawings. I was very fond of pencil drawings. In fact, the first money I ever made was from this lady who wanted me to draw her horse for her, so she gave me a picture of her horse, and I made $60, which, when you’re a high school student, $60 is a lot of money.

So, I thought for sure art was my calling.

Well, medicine came along, and medicine took more interest out of me than art.

I still try to draw at times, but not as much as I would like.

But medicine is not just science, as we all learn now. The art of it is probably as much, if not more, important than the science—to understand how our patients feel, to put ourselves in their shoes and listen.

Many of us don’t do very well in listening to our patients, who are giving us all their answers. I think the art of medicine is really what ties the two parts of my brain together.

LS: I was born in Hong Kong, and then I came over to Canada in my teens, and I studied medicine here in Toronto, and I did much of my undergraduate training also in Toronto. I wanted to see the world.

My mom was sick with cancer—hence the whole background of oncology. But she was well enough for me to go away for a bit of time to get some more training.

I literally opened JCO, the Journal of Clinical Oncology, one day, and in the back there was an advertisement for a drug development fellowship in San Antonio, Texas. I read the description and felt, “Wow, this is really what I want to do. I want to develop drugs and bring new drugs to patients and bring them hope.”

I applied. Unfortunately, I was a bit late for that year, so I stayed behind for a fellowship training here in Toronto under Dr. Malcolm Moore to learn a bit about drug development and the landscape here in Canada. And then I went to San Antonio, and there I trained with all the giants of drug development, Dr. Daniel D. Von Hoff, Dr. Eric Rowinsky, and Dr. Gail Eckhardt.

It opened my eyes that this is definitely a field that I would love to be involved in—to take a drug from bench to bedside and then translate back to understand why it works or it doesn’t work. To me, that was, “Wow, I would love to spend my whole career doing this.” And I did.

LS: The INSPIRE study was a very interesting story.

Many years ago, when every scientific group was interested in the immune system and trying to “drug” the immune system to attack cancer, our institution at Princess Margaret was no different.

We have very well renowned basic scientists, Pamela Ohashi, Tak Mak, Naoto Hirano, and we have clinicians like myself who are really interested in doing clinical trials involving drugs that attack the cancer cells through the immune system.

But there was not enough crosstalk between the basic scientists and the clinicians. So, I was tasked to try, with Dr. Ohashi together, to build a truly translational program.

Because drugs are now more active… there’s more interest to make go/no-go decisions early on, so we don’t subject patients to drugs that are not going to be ever helpful to them. With that, I think there is a need to have early readout.

So, we thought, how are we going to do this? The scientists don’t want to leave the labs, and the clinicians are too busy in the clinic. So, we said, why don’t we create a flagship study that everybody wants to be a part of?

And we created the INSPIRE study—the name of the study was really meant to inspire people to be a part of it. We got pembrolizumab from the pharmaceutical company, Merck, and we wanted drug-only support because we wanted to apply for grants to do a lot of the correlative studies.

And lo and behold, that was in 2016, 2017. Fast forward 2024, the study obviously has long completed accrual, but we still have papers coming out of the INSPIRE study.

Now, the clinicians are meeting with the basic scientists, translational scientists, pretty much regularly, not just for INSPIRE but for other projects related to immunotherapy and other scientific discoveries.

That particular study [INSPIRE] has brought on many high-impact papers, including in the Nature family, Cancer Discovery, etc. While I think papers and manuscripts are important, I think the team science is much more critical—that it really has been able to bring the group to work together such that the whole is much greater than the sum of the parts.

So, it’s a good story, and I’m very proud of it, because it all started with trying to brainstorm how do we make them play in the same sandbox—thinking they’ll never want to work together.

And now—they are.

LS: I trained, as I mentioned, in Toronto before I went to San Antonio.

In addition to Malcolm Moore, who taught me a lot of my experimental therapeutic learnings, I also learned from Ian Tannock, who is a world-renowned clinical trial methodologist. At the beginning, I dabbled in this area. I thought, “Oh, this is a good way to get some papers out.”

But it became a passion, and it’s also a very good subject area to network because there are many like-minded individuals around the world who are interested in developing clinical trials in a better way—more agile, more dynamic, and more efficient. So, it became not as much a hobby now as really a true interest.

And it has also attracted a lot of trainees who have come to work with us and they have developed projects in clinical trial methodology, whether it is chart review or white papers and other ideas and thoughts.

I started from early-phase drug development, phase I trial design, dose escalation on how to get to the MTD (maximum tolerated dose) properly and accurately. Fast forward to 2023-24, where dose optimization is a very hot topic, because we have not been giving the right doses to many patients based on the drug designs that we have, and because drugs have changed.

They’re not all cytotoxics now. In the era of immunotherapy and targeted therapy, the same principles don’t apply anymore as much to these drugs.

So, a large extent of my interest in clinical trial methodology has passed on to these various topics.

I’m very glad I went to that area, because when you’re dealing with drugs day-to-day on the hardcore, translational research, having some moments that you back off and think, “Are we doing all this right? Could we have done this better and more efficiently? And how can we get at the right dose with the fewest resources and not put patients at low or non-subtherapeutic doses, and at the same time avoid toxic doses?”

These are all really, really interesting thoughts and ideas.

LS: There are many individuals, FDA, regulatory agencies around the world, as well as academic oncologists and pharmaceutical representatives who are really interested in this, because, obviously, it now is considered a necessity and not just a “may do” kind of project. It’s a “must do” project, and there’s a lot of interest on how to do that right—really looking at the data in totality, whether it’s pharmacology, pharmacokinetic or pharmacodynamic data, early efficacy data, even preclinical data can add to that model to really understand how best to create a dose range, not just a dose.

That dose range can be further studied towards the end of dose escalation to choose the correct dose to move forward.

It sounds like it’s a long effort, but in the end, I think we all feel that that’s the right thing to do early on, rather than when you have a phase III study and then you find out you’re testing patients on the wrong dose. That probably would lead to a really detrimental effect on the path of the drug.

LS: Because drugs are now more active—of course they still have toxicities, but by and large, I think drugs are more active, especially combinations with active drug plus new drug and hopefully the new drug has activity—I think there’s more interest to make go/no-go decisions early on, so we don’t subject patients to drugs that are not going to be ever helpful to them.

With that, I think there is a need to have early readout.

For example, I’m very interested in the area of circulating biomarkers such as circulating tumor DNA, whether we can use changes in ctDNA early on, even in a phase I trial, to understand whether there is efficacy to move forward or not.

In a phase I trial where patients are very heterogeneous, typically heavily pretreated with many lines of treatment, to see bonafide tumor shrinkage might be challenging. But I think having a molecular assay that allow us to see whether the tumor burden changes on a microscopic level such as with ctDNA is very attractive.

Now, obviously, we need to follow to see whether these kinds of surrogate markers do prove to be reliable, but I think overall it’s a really exciting time when technology catches up to our knowledge—maybe even exceeding our knowledge at this point.

LS: I think the biggest challenge is I don’t think we share enough of our knowledge. So, if we are to look at one particular disease, for example, microsatellite-stable colon cancer, a very common disease—the vast majority of colon cancer is MSS—and yet we don’t have great drugs beyond chemotherapy for this particular disease.

We know that very few of these cancers, if any, respond to immunotherapy. And targeted therapy has a limited role.

So, clearly, there’s a lot of interest in tackling this challenge—from the basic scientists to the clinicians to even clinical trials that have been completed and not been successful—I think we need to learn from all that and put it all together to really move the field to the next chapter.

I think a lot of the data get hidden, either because they’re negative and not published, or they’re in the competitive space where sharing is challenging. It really stalls our progress, because there’s less interest in sharing when the data are not yet open to the public. I think it stalls the field.

While I think trying to put a lot of multimodal data together is challenging, not having the ability to access that kind of data to analyze them is even more difficult.

LS: It’s one of my disease sites of interest.

When I was training, we had to pick disease sites to focus on. Phase I trials cover every solid tumor, so I have to be a jack of all trades and know a little bit about every cancer type, but we are also specialized. I had chosen head-and-neck cancer and gastrointestinal cancers.

Over the years, I have had to give up some, because time is limited. So, I have stopped seeing gastrointestinal cancers, colorectal cancers, and other GI cancers, but I still maintain my head-and-neck cancer interests. So, I still have a clinical practice beyond the phase I program in head-and-neck cancer. And, of course, I was mentored by Ian Tannock who is also a head-and-neck oncologist. So, there’s some roots to it.

LS: I think it is important to have mentors such as Ian Tannock and Malcolm Moore, etc., not just because they’re famous, but because they care about their mentees’ careers.

They met with me. They would suggest, for example, if they had to give a talk but they couldn’t, or they thought that I was a more appropriate speaker, they would put me up for that kind of opportunity. Writing papers together, getting feedback from them.

Ian may not like me telling the public this, but he had this “evil red pencil” we called it. We would dread seeing back a manuscript that is from black turned to red because it’s all red.

You knew you didn’t do a good job when you got back your manuscript and it’s red and not black anymore, because of the evil red pencil. But that’s the opportunity for you to learn.

You think about how giants like Ian Tannock, Malcolm Moore, etc., think, and after you do this a few times, you start to say, “Hey, I think this is how they would think, but I’m going to even think more beyond that. I’m going to think one step further and challenge their thinking.”

I think that kind of iteration really helped me grow, in terms of being able to challenge my mentees. I think the most important thing when mentoring someone is not tell them what you think. It’s really to teach them how to think, and there comes a point where they are able to come up with their own ideas or challenge what you say. Then, I think, you’ve done something really successful, because you have now stimulated somebody to have their own original innovative thinking.

I don’t have an evil red pencil. It’s all electronic now.

LS: I think, obviously, evidence-based thinking. But beyond evidence-based medicine is to sometimes read beyond your own field. Sometimes the answer is not only in oncology.

There are other fields that have done similar or even more advanced testing, and learning from other fields such as rheumatology, cardiology, infectious disease—I mean I draw to those subspecialties because they were the subspecialties that if I weren’t an oncologist I would have been an infectious disease or nephrology expert.

I really like those fields that make you think quite a lot, where every patient case is different and there’s a lot of deep thinking involved.

I would say I draw from other subspecialties in my scientific thinking, and really think about a little bit further than immediate results. I think you need to look far enough to think what happens to the end of the experiment, not just tomorrow, and be able to project from the end looking backwards and then work your way back. That makes more sense than to just think, “If I get through this experiment or this trial, I’m happy I can publish this.”

I don’t think that’s enough. What do you do if this study is positive? What do you do if this study is negative, and what is that line of sight? And of course, I’m a phase I trialist, so I only have to figure out the dose and go to the phase II trialists and say, “Here’s the dose, test it.”

But life is not like this anymore, because many of the drugs now start with phase I and perhaps end in phase I, and then they get registered and approved.

So, you have to sort of think about more like a phase III oncologist now in the phase I arena. So, I think this whole thinking about the end before you start is very important.

LS: Both of these studies are in head-and-neck cancer. I was the co-global principal investigator for both studies. Keynote-122 is a nasopharyngeal cancer trial. I was co-PI with Professor Anthony Chan in Hong Kong. Keynote-412 is in locally advanced head-and-neck squamous cell cancer. I was co-global PI with Professor Jean-Pascal Machiels.

Just to take one for example, in the KEYNOTE-412 trial, even though it was a negative study at the end, we learned a lot from it and it just got published in Lancet Oncology this week.

Be persistent, because there’s never an end that is a bad ending. It is always a good ending if you put enough effort in it. Maybe not entirely the way you want it, but at least if you put in the effort, something will return to you that is worth your effort.

We met at ASCO, I would say probably five, six years ago.

Interestingly, Jean-Pascal and I came up with a very similar idea about adding immunotherapy to concurrent chemoradiation, which is the standard of care for patients receiving definitive treatment with locally advanced head-and-neck squamous cancers. We decided to join forces and we were planning on how to do this under the cooperative group setting, but in the end we felt it was easier that we join forces.

Merck, the company that is obviously leading that study and [manufacturing] pembrolizumab, was interested in our proposals and decided that we would all work together, the company and the two investigators from different parts of the world to design the study together and run it through the last few years.

The accrual went really well and the study read out. Obviously looking back there are things that we probably would’ve changed. Hindsight is always 20/20, but the point is that we actually both came up with a similar idea and worked together.

We are very good friends. In fact, we continue to work on several other head-and-neck studies together now as co-global investigators. So, I think in many ways having the opportunity to work with colleagues that have very similar passion about a disease, very similar thoughts and ideas is really rewarding.

That’s why I think networking is so important, and I tell my mentees, “Going to meetings is not just presenting your poster, presenting your talk. You need to talk to others, network, understand how you can make friends and build bridges across the world.”

It sounds very cliche, but it is so important in the career of an oncologist.

LS: It’s an interesting question. I think the geographical barrier is much less than others think. Honestly, I think it’s much easier to go to Canada from the U.S. and vice versa than many, many parts of the world to the U.S., I think it’s so global that it doesn’t matter where you are.

Of course, our healthcare system is a bit different. We have universal healthcare, so basically everybody gets very similar, if not the same, medications and treatments and tests, and some drugs are not immediately available, compared to the U.S. for quite a while, because the approval process and the reimbursement process are not the same.

Even if a drug is approved by Health Canada, it may take a few months or a year before you can get reimbursed or maybe never get reimbursed, which is always challenging when there’s an effective drug, but you cannot use it for your patients.

So, I think that part is a bit challenging for me in Canada to handle, because you know the science and the evidence, but you may not have the ability to prescribe the drug because it’s not available, because it’s not funded.

Other than that, I love Canada.

I have been here the vast majority of my life, and I have no intention of leaving it anytime soon. I love my travels to the U.S., because during that time I feel like I am completely Americanized.

I can eat at an American restaurant and know all the favorite American dishes, just like my colleagues next to me—but yet I have poutine, and you don’t have poutine.

LS: I’m hesitating, because I don’t think I have a lot of challenges. I’m very blessed, but every challenge I see as an opportunity. I do.

When I hit a roadblock, I don’t get depressed too much. I try to say, “Hey, that’s an opportunity. I haven’t dealt with this before. I’m going to try and see how I can rise up and face the challenge.”

I’ve mentioned some challenges. Obviously, one is that sometimes, as an oncologist here in Canada, I cannot prescribe a lot of the treatments that I think should be given.

When you don’t have a healthcare system that allows you to prescribe everything based on evidence because of funding, that’s a challenge.

I don’t always get my proposals, letters of intent, or grants funded, because that’s a fact of life. If we get every grant funded, that would be just a perfect world. I see that as an opportunity because if my grants are rejected, there must be a reason, and the reviews will teach me something. So, I don’t get down when it comes to that too much.

I teach my mentees, if they get a paper rejected. I say, “Get used to it. This is not going to be your last time, and every time you get a rejection, you should take advantage of it and improve your paper, because it will only make it better.”

So, as I said, I don’t see every challenge as a bad thing. I see every challenge as an opportunity to make it a better thing.

One of the biggest challenges is that I don’t have enough time, and I don’t have great time management. I like to do a lot of things myself, because I have that kind of urge to see to it myself. I think that’s not a good trait. It’s better to be able to delegate and let go.

I’m trying to do that a lot more and have a better work-life balance. I think that’s a challenge for sure. I’m here on the weekends, I’m here on holidays, like today, and probably shouldn’t be, but hey, I love my work. What can I do?

LS: I do. I have trained many, many over the years.

LS: I think the proudest moment really has been seeing my mentees win career development awards or have a paper or become head of a department, or see them really high up in the drug development ladder—that they have been able to succeed.

None of us can be in the limelight, in the prime of our career, forever. I am very realistic, and seeing your next generation that you train—and I’m very proud that I’m friends with probably 99.9% of the people I have trained.

We get together every ASCO and we try to touch base every year to catch up—I think seeing those individuals that I’ve touched becoming successful leaders, to me that is probably much more than just myself having a big award.

I mean, I’m grateful to ASCO for the Karnofsky, don’t get me wrong. It’s a big honor.

I think those individuals who are very successful—and really not just in academia, some are very successful in the pharmaceutical business and very high up in the pharmaceutical companies developing new drugs that affect many lives.

To me, this is really a success of my career to be able to touch on their life, even for a few years during their training.

LS: I think what worries me about oncology is there’s a lot of redundancy. We do a lot of “me-too” drug development. When there’s an interesting target, everybody is there, and people tend not to think out of the box as much.

I also think for cancer, for example, we tend to be very cross-sectional. We test one tumor at one point in time. We’re not as dynamic as cancers are. So, I am worried that we don’t have that kind of insight to really try and predict where the cancer is going, rather than try and follow it and beat it based on following and see what it’s doing rather than predicting what it will do.

Obviously, we have realized that there is an interest in getting rid of disparities—or reducing disparities—and access. We’re in the part of the world where it’s not difficult to access a lot of good medicines and good health, but there are many, many parts of the world where getting next-generation sequencing is not the most important thing.

Getting surgery done is the most important thing. That kind of disparity obviously is disheartening, when you should be able to make treatments fairly equal around the world.

So, I think there’s a lot of work cut out for all of us in oncology and other subspecialties.

LS: I was one before. I would say don’t give up. I would say learn from every mistake or every challenge and rise above it and be tenacious.

Be persistent, because there’s never an end that is a bad ending. It is always a good ending if you put enough effort in it. Maybe not entirely the way you want it, but at least if you put in the effort, something will return to you that is worth your effort.

I truly believe in that, and certainly I see that in my career. I don’t only have positive results, I have very often negative results, but it’s fun.

Learning from your mistakes is half of the fun, and cherish that kind of moment to learn from it.

LS: I think you asked me about my whole life.