Project Optimus Series | Clinical Pharmacology and Project Optimus

We continue our Project Optimus series with a focus on clinical pharmacology—a foundational element in redefining dose optimization in oncology drug development. 

Dr. Wael Harb is joined by Pierre-Olivier Tremblay, Vice President of Clinical Pharmacology at Syneos Health, to discuss how model-informed strategies, exposure-response analysis and biomarkers are transforming early phase trials. They examine how tools like PK/PD modeling and tumor dynamics simulations enable a shift from traditional maximum tolerated dose approaches to more patient-centric dosing strategies, and explore how AI, data sharing and regulatory innovation are shaping the future of precision oncology.

The views expressed in this podcast belong solely to the speakers and do not represent those of their organization.

If you want access to more future-focused, actionable insights to help biopharmaceutical companies better execute and succeed in a constantly evolving environment, visit the Syneos Health Insights Hub. The perspectives you’ll find there are driven by dynamic research and crafted by subject matter experts focused on real answers to help guide decision-making and investment. You can find it all at https://www.syneoshealth.com/insights-hub.

Like what you’re hearing? Be sure to rate and review us!

We want to hear from you! If there’s a topic you’d like us to cover on a future episode, contact us at podcast@syneoshealth.com. 

Transcript

SPEAKER_00 0:00

Welcome and thank you for joining us. I am Dr. Yell Hart, Vice President of Medical Management at Sineas Health. This episode continues our special series on the FDA's Project Optimus, a transformative initiative that's redefining how we approach dose optimization in oncology. In our previous episodes, we explored the regulatory, operational, and biostatistical perspectives. Today we shift to a cornerstone of dose optimization clinical pharmacology. Joining me is Pierre Olivier Tremblay, Vice President of Clinical Pharmacology at Sinus Health. He brings a wealth of experience in applying PKPD modeling, exposure response analysis, and simulation tools to optimize dose selection and accelerate oncology drug development. Clinical Pharmacology and Project Optimus. Next on the Sinus Health podcast. Pierre Olivier, welcome to the podcast. Hey, hello. I'm really happy to be here with you today. Pierre Alvier, let's start at the top. From your perspective, how is clinical pharmacology central to the implementation of Project Optimus?

SPEAKER_01 1:22

Well, it forms a large part of the backbone of Project Optimus with the shift from toxicity and maximum tolerated doses to assessing clinical efficacy and safety at an appropriate dose. All of the techniques we use in clinical pharmacology to support optimal dose selection is then applied with Project Optimus, meaning that really early in the clinical trials in oncology, we need to assess exposure as well as response and uncover tendencies that the maximum tolerated dose paradigm can't do right now.

SPEAKER_00 2:02

So the FDA is asking developers to rethink how they structure their early phase studies. Why is the traditional MTD or maximum tolerated dose-based approach no longer enough, especially in oncology?

SPEAKER_01 2:18

Yes. So the NTD type of output or endpoint was created for chemotoxic agents with a high, as the name implies, a high toxicity level. Nowadays, with individualized therapy or immunogy, these drugs come with a larger safety window, meaning that we may not necessarily find obvious toxicities or severe toxicities early on in the drug administration process. That means that we incur the risk of giving doses that are really high without additional efficacy benefit. Project Optimus aims at circumventing this by making online during a study assessments of exposure and response and also uncover longitudinal adverse events of low grade intensity. Some adverse events with immunocology or individualized therapy may be low grade, grade one, grade two. They may occur later than in the very first few cycles of treatment, and they may be long-lasting but under threshold for an MTD. So using a totality of evidence approach, Project Optimus aims at taking a totality of evidence approach where we will optimize, as the name implies, the dosage for maximum efficacy and limit unnecessary toxicity as best as we can.

SPEAKER_00 4:01

So new biological requires a new paradigm in how we find the optimal dose. What kind of exposure response data does the FDA expect under Project Optimus?

SPEAKER_01 4:15

Well, first we'll start with the traditional exposure response analyses where we'll try and see both dose response and exposure response, trying to see how exposure relates to potentially adverse events, even if they're low grade, try to see whether they are appearing later and if they last for a long time. We'll try as well to link exposure to early response biomarkers, so circulating tumor DNA, for example, PSA and prostate cancer, for example, as well. The idea is to make all of these measurements and take all of these data points and tell a story of what we're seeing happening in the treatment to help support selecting a dose and a dosage as well. So those and dose regimen and often combinations that will lead to maximum response and limiting adverse events. And so the idea is taking longitudinal information along the trial and incorporating all of that into often will use model-based approaches to integrate exposure and response data, whether they be biomarkers, safety, or even observable endpoints such as tumor growth inhibition, for example.

SPEAKER_00 5:46

It's interesting you mentioned the CT DNA as a BD marker. Do you see that being used often in oncology trials? It is used more and more.

SPEAKER_01 5:56

I know that there's often a fear that these biomarkers need to be validated. And it is the case if we were intend on using CTDNA in a confirmatory phase three trial, but certainly in the very early trials, so first in human studies, early phase two studies, where treatment duration is short, where sample size is also short, biomarkers like CTDNA can provide insights on early responses that are not possible to see with these types of early phase trials. And if it's there to help support dosage optimization, the requirement for fully validated assays is not necessary in these very early trials. What we want is gather knowledge, gather evidence of therapeutic action. And we don't necessarily want to use that for residual disease assessment, for example. It's more for an exploratory endpoint. So there's a lot of potential for biomarkers and PD markers in early drug development to guide dose selection. That's great to know.

SPEAKER_00 7:10

You mentioned modeling and modeling forum drug development. What tools or modeling strategies are most useful in this context?

SPEAKER_01 7:20

A lot of the tools that we have today in clinical pharmacology can be used in these early phase oncology trials for understanding exposure population decay modeling, so nonlinear mixed effect modeling. Similar models can also be used for tumor size or tumor inhibition dynamics along the way. So dynamic models taking into account changes in time. These models can also account for baseline intrinsic factor from the patient perspective and extrinsic factors, as previous treatment lines, for example, and things like that. So these models are very powerful in uncovering characteristics that will provide responses in a specific population. Other types of models, I'm sure XQ has talked about those as well, but all of the models involving Bayesian approaches for determining whether to go to the next dose or not. So these can also be applied. But a lot of the clinical pharmacology models are dynamic models that have continuous responses. So again, tumor growth dynamics, CT DNA changes, and these can be linked to other markers or other measurements, hazards ratios, and other forms of odds ratios for modeling survival, for example. So these can all be linked together. So the great advantage of these tools is that they allow quantification of multimodal outputs and bring them together. So that also allows for affirming or delineating clear assumptions on what we're observing. These assumptions are part of the modeling exercise, and so we call them out specifically, and so that also helps drive the decisions we're making along the way.

SPEAKER_00 9:49

Alternatively, do we need to do both? What's your take on how the PKPD modeling can utilize data from backfields and dose escalation and the necessity to do randomization?

SPEAKER_01 10:01

I'd say that backfills may be interesting when we observe in our modeling exercise we may be lacking data information that wouldn't allow us necessarily to do a randomized dose optimization. And so it's interesting as an add-on in to data that is incomplete to some extent, because we can't change the protocol necessarily, but we can make some amendments for further biomarkers, for example. But I guess backfields is interesting for supplementing data prior to making a decision, supplementing the models to then go to a randomized dose optimization cohort.

SPEAKER_00 10:48

Let's talk about patient impact. How does dose optimization supported by CONCOPharmacology benefit patients?

SPEAKER_01 10:55

Well, one of the important aspects of these modeling activities, as I said earlier, is the capacity to include in these models intrinsic factors. So patient covariates, for example. A very simple example could be age of the patient, number of prior therapies, what type of prior therapies. These can all be part of the modeling exercise, allowing to uncover patient characteristics that make them more or less amenable to treatment. That's a powerful tool in allowing us to tailor treatment and dosage to specific patient populations. Another covariant that could be used is also specific mutations or not, presence or absence of certain immune cells, for example. But the modeling exercise allows us to take all of these into consideration, given enough information and data, and see what would be a patient that responds better to some dosage, and what other types of patients would respond better to other types of dosage or combinations, for example.

SPEAKER_00 12:10

Sounds great. Looking to the future, how do you see the role of clinical pharmacology evolving within the context of Project Optimus and oncology interrupt development overall?

SPEAKER_01 12:22

We'll have to do a lot with technical advancements. A lot of these models right now are based on data from clinical trials that belong to a specific sponsor, pharmaceutical company, university, or whatnot. These models will evolve. The models we build today will evolve through machine learning and artificial intelligence, but we can only do that if we have access to sufficient data. So I think one of the ways clinical pharmacology will be impactful in oncology and other indications. And when I say clinical pharmacology, I also incorporate quantitative biology, so computer biology and other methods, other quantitative aspects, but getting access to larger data sets, so sharing data repositories, validating some of these models as fit for purpose into certain types of clinical study designs in certain form of indications. So certainly the future looks bright for quantitative methods and clinical pharmacology through the use of large data sets, real-world evidence, and linking all of that information into models that can then be validated to automate some of the analyses that we're currently doing on a case-by-case basis for some of the drugs we're working on. What we're going to see in the future are more validated tools, quantitative tools that will accelerate and support drug development, in particular in oncology. And to that effect, the FDA has just released a roadmap for reducing animal testing for biopharmaceuticals. That means that there's a wish for regulatory bodies to take advantage of these quantitative methods and new approach methodologies for supporting drug development and trial design, optimize them as much as possible. And so this roadmap for reducing animal testing in biopharmaceutics and biologics, specifically for antibodies, means a significant change in how we do immune oncology for the future, for example, with the use of more specific human-related nonclinical models, but also using in silical computer simulations for optimizing initial dose selection and then dosage recommendations and optimal doses for future trials. A lot of things happening right now in that field.

SPEAKER_00 15:25

Very exciting. Do you see AI will play a role in drug development?

SPEAKER_01 15:30

Absolutely. The challenge with drug development and AI is the access to large data sets. And so we do have that existing in specific databases, either by government bodies or universities, but sharing anonymized data and making it accessible to groups and organizations to develop and validate AI tools is still a challenge nowadays. But as soon as we go through that hurdle and solve that, AI will become increasingly used, both for selecting targets and combinations, but also for optimizing dosages of different therapies, including combination therapy and rationalizing drug development process, making sure that for smaller clinical trials, but more targeted clinical trials, for which you could have trials targeting a specific population, go to your proof of concept, phase three confirmatory trial for that specific population, and then run another small phase three study for an additional indication based on AI research. So more nimble clinical development is probably coming along the way. We're still some way from that. I don't think we'll ever see the disappearance of actual clinical trials with patients. But as I've said, they'll be more nimble, smaller, quicker with the help of AI and machine language and models will have developed for supporting drug development. And those models would be validated as per DFDA requirements. And once they are validated, I think they'll be used very efficiently by organizations to develop smaller, more targeted trials that will go quicker to approval.

SPEAKER_00 17:38

Less use of animal studies and smaller, nimbler studies in clinical trials would be very helpful in reducing costs and accelerating timelines.

SPEAKER_01 17:48

Definitely. Yeah, that's the aim in the longer term. Less animal use and targeted clinical trials where the likelihood of seeing immediate patient benefits exist and then expanding from these success to larger indications or modified indications, but always through nimbler, smaller trials that'll go more rapidly to approval to the regulatory agencies.

SPEAKER_00 18:16

Pierre Olivier, this has been a truly insightful discussion. Thank you for helping us understand how clinical pharmacology is enabling the success of Project Optimus and bringing us closer to precision oncology. It was my pleasure. Thank you so much.

SPEAKER_02 18:33

That's all for today's episode of the Senior's Health Podcast. I'm your host, Nick Kenny. If you have other topical issues you would like to hear on this podcast, please send us a message at podcast at SeniorsHealth.com. And for access to more future-focused, actionable life science insights, please visit the Senior's Health Insights Hub at insightshub.health. Sonyous Health, shortening the distance from lab to life.