What are the key takeaways from this GEN Podcasts episode?

Key insights include: **NOAEL/MABEL Integration:** Mechanistic modeling enhances traditional dose-setting frameworks by replacing concentration-based thresholds like IC50 with mechanism-driven metrics such as target occupancy percentages (e.g., 80–90%) or protein synthesis perturbation rates. This allows species differences to be explained beyond allometric scaling, enabling teams to justify higher, safer starting doses with quantified uncertainty.; **Backwards and Forwards Translation:** Because mechanistic models are built on biological parameters — cell counts, binding affinities, synthesis and degradation rates, feedback loop strengths — they enable bidirectional translation between preclinical and clinical settings. Teams can update model parameters as Phase 1 data arrives, progressively reducing prediction uncertainty across SAD, MAD, and recommended Phase 2 dose decisions.; **Virtual Patient Analysis in Phase 1:** Rather than waiting for Phase 3 statistical power, mechanistic models can simulate patient subpopulations during Phase 1 by sampling distributions of biological parameters. Teams can define dose targets as: achieve greater than X% response in greater than Y% of patients, enabling earlier, data-driven dose escalation decisions with fewer enrolled subjects.

What did John Burke discuss on GEN Podcasts?

John Burke, president and CEO of Applied BioMath, explains how mechanistic modeling integrates with NOAEL and MABEL frameworks to improve first-in-human dose predictions, manage patient variability in clinical trials, and accelerate therapeutic development for gene and cell therapies through quantitative systems pharmacology approaches. Key topics include: **NOAEL/MABEL Integration:** Mechanistic modeling enhances traditional dose-setting frameworks by replacing concentration-based thresholds like IC50 with mechanism-driven metrics such as target occupancy percentages (e.g., 80–90%) or protein synthesis perturbation rates. This allows species differences to be explained beyond allometric scaling, enabling teams to justify higher, safer starting doses with quantified uncertainty.; **Backwards and Forwards Translation:** Because mechanistic models are built on biological parameters — cell counts, binding affinities, synthesis and degradation rates, feedback loop strengths — they enable bidirectional translation between preclinical and clinical settings. Teams can update model parameters as Phase 1 data arrives, progressively reducing prediction uncertainty across SAD, MAD, and recommended Phase 2 dose decisions..

How long is this episode of GEN Podcasts?

This episode is 23 minutes long. SignalCast provides an AI-generated summary so you can get the key insights in about 3 minutes.

GEN Podcasts

Mechanistic Modeling Improves Drug Discovery Workflows and Speeds Therapeutic Development

December 1, 2022

23 min episode · 2 min read

John Burke

Episode

23 min

Read time

2 min

Topics

Productivity, Startups, Fundraising & VC

AI-Generated Summary

Published Mar 15, 2026

Key Takeaways

✓NOAEL/MABEL Integration: Mechanistic modeling enhances traditional dose-setting frameworks by replacing concentration-based thresholds like IC50 with mechanism-driven metrics such as target occupancy percentages (e.g., 80–90%) or protein synthesis perturbation rates. This allows species differences to be explained beyond allometric scaling, enabling teams to justify higher, safer starting doses with quantified uncertainty.
✓Backwards and Forwards Translation: Because mechanistic models are built on biological parameters — cell counts, binding affinities, synthesis and degradation rates, feedback loop strengths — they enable bidirectional translation between preclinical and clinical settings. Teams can update model parameters as Phase 1 data arrives, progressively reducing prediction uncertainty across SAD, MAD, and recommended Phase 2 dose decisions.
✓Virtual Patient Analysis in Phase 1: Rather than waiting for Phase 3 statistical power, mechanistic models can simulate patient subpopulations during Phase 1 by sampling distributions of biological parameters. Teams can define dose targets as: achieve greater than X% response in greater than Y% of patients, enabling earlier, data-driven dose escalation decisions with fewer enrolled subjects.
✓Project Optimus Alignment: The FDA Oncology Center of Excellence's Project Optimus initiative explicitly targets dose optimization reform in oncology. Mechanistic and systems modeling directly supports this regulatory direction under PDUFA's Model-Informed Drug Development framework, giving teams a regulatory pathway to justify higher starting doses in end-stage oncology patients where homeopathic starting doses create ethical concerns.
✓PopPK Remains Relevant Late-Stage: Mechanistic modeling does not replace population PK/PD approaches; both run in parallel. PopPK becomes the standard tool at Phase 2–3 and pre-BLA/NDA stages when sufficient patient numbers allow statistical parameter distributions. For novel gene and cell therapies, however, no established PopPK standards yet exist, making mechanistic QSP models the current primary framework.

What It Covers

John Burke, president and CEO of Applied BioMath, explains how mechanistic modeling integrates with NOAEL and MABEL frameworks to improve first-in-human dose predictions, manage patient variability in clinical trials, and accelerate therapeutic development for gene and cell therapies through quantitative systems pharmacology approaches.

Key Questions Answered

•NOAEL/MABEL Integration: Mechanistic modeling enhances traditional dose-setting frameworks by replacing concentration-based thresholds like IC50 with mechanism-driven metrics such as target occupancy percentages (e.g., 80–90%) or protein synthesis perturbation rates. This allows species differences to be explained beyond allometric scaling, enabling teams to justify higher, safer starting doses with quantified uncertainty.
•Backwards and Forwards Translation: Because mechanistic models are built on biological parameters — cell counts, binding affinities, synthesis and degradation rates, feedback loop strengths — they enable bidirectional translation between preclinical and clinical settings. Teams can update model parameters as Phase 1 data arrives, progressively reducing prediction uncertainty across SAD, MAD, and recommended Phase 2 dose decisions.
•Virtual Patient Analysis in Phase 1: Rather than waiting for Phase 3 statistical power, mechanistic models can simulate patient subpopulations during Phase 1 by sampling distributions of biological parameters. Teams can define dose targets as: achieve greater than X% response in greater than Y% of patients, enabling earlier, data-driven dose escalation decisions with fewer enrolled subjects.
•Project Optimus Alignment: The FDA Oncology Center of Excellence's Project Optimus initiative explicitly targets dose optimization reform in oncology. Mechanistic and systems modeling directly supports this regulatory direction under PDUFA's Model-Informed Drug Development framework, giving teams a regulatory pathway to justify higher starting doses in end-stage oncology patients where homeopathic starting doses create ethical concerns.
•PopPK Remains Relevant Late-Stage: Mechanistic modeling does not replace population PK/PD approaches; both run in parallel. PopPK becomes the standard tool at Phase 2–3 and pre-BLA/NDA stages when sufficient patient numbers allow statistical parameter distributions. For novel gene and cell therapies, however, no established PopPK standards yet exist, making mechanistic QSP models the current primary framework.

Notable Moment

Burke points out that standard dose-setting methods sometimes produce homeopathic starting doses even for terminal oncology patients who urgently need higher exposures — a counterintuitive ethical problem that mechanistic modeling, combined with FDA's Project Optimus, is specifically designed to solve.

Know someone who'd find this useful?

Episode Transcript

Welcome to Gencast, a sponsored podcast series brought to you by Genetic Engineering and Biotechnology News. I'm your host, Jeff Pugaliskis. In my last conversation with doctor John Burke, president, CEO, and cofounder of Applied Biomath, we discussed how the company's applying advanced mathematical and systems biology techniques at critical decision points in the drug discovery process. In particular, we touched upon what the process is for deciding what will be the first human dose and what's involved in filing the investigational new drug application or IND. Now in this final episode of our three part podcast series, John and I dive even deeper into the mechanistic modeling approach and how organizations can and are specifically applying it to their drug discovery and development workflows. Let's jump in and join the conversation. Hello, everyone, and welcome to this new episode of Gencast. As always, I'm your host, Jeff Bogoliskas, the technical editor for genetic engineering and biotechnology news, which has been bringing you the latest news and insights on new tools and technologies within the life science industry for over forty years. Once again, it's my great pleasure to be joined by the cofounder, president, and CEO of Applied Biomath, doctor John Burke. Welcome back, John. Thank you very much. I'm excited to be back. So, John, in the previous podcast, we discussed determining if developing a therapeutic is feasible and whether or not it could be best in class. And you also gave us a glimpse into what goes into predicting first in human dose and filing an IND. But if you don't mind, could we dig a little bit deeper into how modeling approaches can help? So, really, the first question I have is there are lots of approaches and guidances to picking a first in human starting dose, such as NOAEL and Mabel. You know, how does the mechanistic modeling fit into these approaches? Starting with a a very good question. I'll do my best to overview, you know, my viewpoint on it, but it's gonna be, you know, really just the surface. So when when you're making your your human dose predictions, right, there's clearly, this is highly regulated by governments. Right? Or or, like, the European Union. Right? It's it's, we wanna make sure that we're putting in safe and efficacious therapies into human, but it's also a function of what is tolerable. So, for example, if if you have cancer and you're on your fifth therapy, you know, you can be a little risky. You can take maybe a therapeutic that might give you a terrible rash or other issues, right, to save your life. Right? Whereas if you are in I and I, let's say, an autoimmune disease. Right? Let's say you have, psoriasis or MS or, you know, RA, these these these can be tough. These can be nasty, but people aren't gonna tolerate very bad side effects. Right? Because it's it's it's just a different risk profile. So I I say …

Get the full transcript (3,479 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all GEN Podcasts transcripts →