Vivace Therapeutics co-founder, president and CEO Sofie Qiao, Ph.D., acknowledges that immunotherapy has dominated headlines in discussions of new cancer treatments, but she believes small molecules are still a valuable treatment modality, particularly with new, advanced research tools available for discovering new compounds.

“In the past 20-to-30 years, the trend has been to incorporate more kinds of information, earlier, on the properties of compounds relevant to development,” Qiao says. “For example, structural biology tools have improved significantly and more information is available for potential compounds and their biological targets. Automation also has contributed to faster and more accurate drug discovery.”

Over the next decade, Qiao says, the major treatment modalities, such as small molecules, biologics, and cell and gene therapies, will continue to play a role not only as single agents, but also in various combinations.

“In the future,” she says, “I think combinations will become much more rational, based on mechanism of action or really sophisticated hypotheses.”

Vivace’s small molecule drug development targets the Hippo-Yap pathway, which is involved in the regulation of cell proliferation, programmed cell death and cell migration. Mutations in the pathway are implicated in gastric, colon, cervical, ovarian and breast cancers as well as uveal melanoma, mesothelioma, hepatocellular carcinoma and esophageal squamous cell carcinoma. The company, which was founded in 2015, expects to begin clinical trials in late 2019.

Qiao previously served as managing director of WuXi Ventures, the predecessor to WuXi Healthcare Ventures, which merged with Frontline BioVentures in 2017 to form 6 Dimensions Capital, a China and US-focused investment group.

In 2006, Qiao cofounded LEAD Therapeutics, acquired in early 2010 by BioMarin, mostly for its PARP inhibitor talazoparib, then still in preclinical development stage. On October 16, 2018, FDA approved this drug for gBRCAm HER2-negative locally advanced or metastatic breast cancer. She began her industry career at Genzyme as a medicinal chemist after earning a Ph.D. in organic chemistry from Massachusetts Institute of Technology and an A.B. in chemistry from Harvard University.

WuXi AppTec Communications’ interview with Qiao is part of an exclusive series spotlighting the inside perspectives of thought leaders on topics shaping the future of new medicines.

WuXi: How diverse are small molecule drugs today? How would you define small molecules?

Sofie Qiao: Small molecules refer to things that are not proteins, nucleic acids or carbohydrates. But there are some gray areas like small peptides or oligonucleotides, which some people call small molecules.

The old definition is that small molecules are synthetically accessible and producible by chemical synthesis, but now some large peptides and oligonucleotides are also made synthetically. I think the idea is that they are not proteins, nuclei acids or carbohydrates and peptides and oligonucleotides are sort of the gray zone.

In our company, we work on the classically defined small molecules, meaning synthetically designed small molecules. They are not peptides or nucleic acids or oligonucleotides.

WuXi: How has small molecule drug discovery changed?

Sofie Qiao: In the past 20-to-30 years the trend has been to incorporate more kinds of information, earlier, on the properties of compounds relevant to development; meaning you do ADME (absorption, distribution metabolism and excretion) and PK (pharmacokinetics) earlier. You do off target screening earlier and get a better idea about the various pharmaceutical properties earlier in the drug discovery process. And structural biology tools that used to be very highly specialized are now routinely used in small molecule drug discovery.

There is also a lot more biological information available both about the target and the compound, so more versatile, cellular in vivo read-outs are available for drug activities because of the tools and technologies available for fancy and sophisticated assays.

Also, automation has made a big difference and that means assays can be conducted on a scale and speed that couldn’t have been done before when most work was more manual. So that really increases the data available to make better choices during lead discovery and optimization.

Essentially the technology that has been developed from academic labs and from industry really has contributed to making drug discovery more efficient because of the vast amount of data available to guide the whole drug discovery process.

WuXi: Do small molecule drugs have advantages over biologics and other treatment modalities, such as cell and gene therapies? If so, what are they?

Sofie Qiao: One advantage that immediately comes to mind is that small molecules have much less CMC (chemistry, manufacturing and controls) complexity than biologics.

Small molecules are much easier to make and also much cheaper because of the costs of goods. They can be a tiny fraction of say, antibodies or proteins. Not only are they easier to make by synthetic means, but also characterization is straightforward, so there is no issue like heterogeneous post-translation modification of proteins or conformational differences. There’s no need for some of the assays that biologics have to encounter.

Another big advantage of small molecules is the potential for convenience of oral dosing, especially for chronic therapies because antibodies and proteins have to be administered intravenously. Small molecules, as long as they have the right properties, can be in a pill. Very few biologics have that potential for oral dosing even though people have been working on different methods of delivering proteins and antibodies.

Yet, another advantage of small molecules is that they can access intracellular targets. Small molecules are the modality of choice for intracellular targets because it is very difficult to deliver proteins beyond the endosome and lysosome.

Gene therapy, or RNA therapies, may be able to deliver a protein to most cellular compartments, but small molecules are the best to access intracellular targets.

WuXi: Are there certain diseases where small molecule drugs would be preferable to other treatment modalities?

Sofie Qiao: In general, if targets are amenable to small molecules and you can discover those molecules, a small molecule is preferable in the sense that it is a lot easier to make and it’s much cheaper.

Biologics usually are applied to things that for whatever reason cannot be done with small molecules. For example, a lot of companies work on antibodies; immuno-oncology is really the rage right now, such as anti- PD-1, PDL-1, and CTLA-4 antibodies. These had to be antibodies in order to address target specificity and/or cell type specificity that cannot be accomplished with small molecules. In that case, you have to resort to antibodies.

I should mention that for genetic diseases which have the potential of being cured by gene therapy in a single dose, gene therapy would be much more preferable to taking a small molecule or a protein because small molecules and proteins are just modulating some functions, whereas gene therapy has the potential to cure genetic diseases.

Small molecules are definitely not going away, but yes indeed, cell therapy, gene therapy and antibodies are currently taking up a lot of the attention in our industry.

WuXi: Are small molecules being used to improve our understanding of biology? If so, how?

Sofie Qiao: Yes, definitely. Biology is very complicated and often a very powerful tool in deciphering biology is to have a good, specific small molecule to study a target. For example, the MEK and BRAF inhibitors were instrumental in understanding the pathways and their role in cancer

We believe that the kinds of compounds Vivace Therapeutics has discovered so far to inhibit activation of YAP driven gene expression are providing us unique insight into the hippo YAP pathway, even though our academic founders and plenty of other researchers in the field have studied the pathway from the biology perspective using knockouts and other models for many years before. Having those small molecules as chemical tool compounds, we have gained a much better understanding of the pathway’s role in certain cancer indications and how to modulate the pathway to address those cancer indications.

WuXi: What are the differences in market pricing between small molecule drugs and biologics? How will these differences change over the next five-to-10 years?

Sofie Qiao: Prices commanded by drugs, especially in the US, are generally proportional to the value they bring to patients.

One example that comes to mind is Gilead Sciences’ HCV therapeutic, which has generated some controversy. However, let’s not forget that it is a fundamental cure and how phenomenal that is for the patients. Hence value-based pricing will likely continue to govern small molecules and biologics.

At the same time, we know that some biologics are extremely expensive to manufacture and cell therapies are even more so. Therefore, in those two cases I would say cost of goods more often play a role in pricing, more often than it does in small molecules.

WuXi: What kinds of small molecule drugs is Vivace Therapeutics developing?

Sofie Qiao: At Vivace we’re working on the Hippo-YAP pathway, which has been demonstrated to be very important for various cancer indications, where the pathway has been activated.

Our founders, Kun-Liang Guan, Ph.D., of the University of California, San Diego, and Sheng Ding, Ph.D., of University of California, San Francisco, were collaborating on this pathway for a while before we raised money to work on the pathway. In terms of the specific cancer indications, there are certain cancers such as gastric cancer, mesothelioma, liver cancers and uveal melanoma, where the pathway has been activated. We’re working on addressing some of those cancer indications with our small molecule approach.

WuXi: Do you see your small molecules as a single therapy or in combination with biologics?

Sofie Qiao: That’s a great question. These days when we talk about oncology, it would seem that everybody is combining everything with PD-1 or PD-L1. Our preference is to find a molecule that works as a single agent, based on really solid mechanisms. The rationale is as follows – if we have a small molecule that works as a single agent in specific cancer indications, we can then look into combination possibilities – right now we have some in vivo and in vitro studies combining our molecule with some known therapies to see whether they lead to synergistic effect.

I think for a small company, the most prudent approach is to discover a molecule that can work as a single agent first, and then look into combination possibilities afterward, which can expand the potential. That’s basically our approach.

WuXi: What technologies is Vivace using for drug discovery?

Sofie Qiao: We have strong medicinal chemistry and very sophisticated assays. In structural biology, we use x-ray crystallography and computational modeling. In terms of technology, there is nothing we’re using that’s unique or proprietary to our company. We are using a suite of sophisticated drug discovery technologies, and we have a very experienced team of drug hunters.

WuXi: What role do you see for artificial intelligence (AI) and other computational tools in drug discovery?

Sofie Qiao: As I mentioned, computational modelling, or molecular modelling, is very useful to us and as we move first-in-class molecules into development, bioinformatics will play a very major role.

I’m not one of the most IT savvy people in drug discovery, so when people talk about AI in drug discovery, I always pause a little. I feel that AI and other computational tools can be very valuable, if whatever information that feeds it is good and solid. I’m more concerned about the possibility of garbage in and garbage out, meaning a lot of wrong data and wrong interpretations of data going in, resulting in the wrong predictions or judgment made by AI.

We want to make sure most of the inputs to the computational tools are the right or good inputs so we can have the right outputs.

If you think about where AI has demonstrated its power, for example, in chess, all the inputs are definitely the right inputs because there are so many moves you can anticipate – millions or billions of moves – they are all possible moves and all the games won or lost by people were truly won or lost, unequivocally.

There is no bad data going into AI for chess, whereas in drug discovery, that’s where I do struggle a little bit. When we look in the literature, since we are working on a novel pathway, people are still trying to figure out the biology, sometimes with conflicting results and conclusions. Some of the reported results later turn out to be irreproducible and some of the conclusions turn out to be wrong. If machine learning is based on 50% wrong data or wrong interpretation – how does that affect AI? That is why I have this concern.

WuXi: Are CROs critical to Vivace’s drug development?

Sofie Qiao: One thing that is unique with our company is that even though we are working on innovative drug discovery, we are entirely virtual in the sense we don’t have any internal labs or any internal R&D center. We work with WuXi AppTec intensely, to whom we outsource chemistry, biology, DMPK (drug metabolism and pharmacokinetics) animal models, and scale-up.

As a virtual company, we rely heavily on CROs (contract research organizations) to do our work. In addition to WuXi AppTec, we also work with other specialty CROs, depending on the expertise needed.

The fact that we work with WuXi AppTec so heavily is that, first, they are the world’s leading drug discovery and development CRO and we get great services and competitive pricing. Second, I did come from WuXi Ventures and that helps a little since I know the organization well. With the virtual business model, we need to work with CROs, and we also have an experienced team that is very good at working virtually. In fact, half of our team came from the old LEAD Therapeutics team, who invented the recently approved Talazoparib by working entirely virtually.

WuXi: Vivace also has been working on bispecific antibodies. What are your plans for this platform?

Sofie Qiao: We are spinning off our bispecific antibody platform into a separate company. That’s why this interview is interesting in its timing. While big companies can often accommodate both small molecules and antibodies, small companies usually take on one. Each has its place and one of the reasons we are spinning off (the antibody platform) is that we do see a more compelling value proposition for the (two platforms) to become separate companies rather than under one roof; and that’s why we are separating them.

WuXi: How do you see the drug modality mix – our toolkit for disease treatment – evolving over the next five to 10 years?

Sofie Qiao: There is a place for each modality and all modalities will continue to get more versatile and efficient. So, there will be an increase in rational design into discovery of all classes of agents and also an increase in rational combinations.

Right now, it seems that people are doing a lot of combinations in oncology, and sometimes we struggle to see good rationale. In the future I think combinations will become much more rational based on mechanism of action and very sophisticated hypotheses.

Biology is complex enough that on the 10-year scale, the experimental process will still dominate discovery of all classes of agents, meaning the experiments themselves are not going to be replaced by machine learning or computational tools.

The biology is just too complex and then the mix of modalities probably will include a greater proportion of things like gene therapy, oligonucleotides and cell therapies, whereas in the past we have been working mostly on small molecules, antibodies and therapeutic proteins.

In the future, there will be a greater proportion of new things, like gene therapy and cell therapy and oligonucleotides; and that’s only because there are so few of them now. Future successes will increase their percentage in the mix and there will be areas where they can better address the unmet medical needs than proteins, antibodies or small molecules. So, we will see more combinations and we see all these modalities continue to grow and mature.