November 17, 2025
by Bara Townsend, Marketing Communications Specialist
The Key Takeaway Messages From This Year’s PhysChem Forum
The annual PhysChem Forum meeting brings together scientists from industry and academia to discuss topics around physical chemistry and its role in directing absorption, distribution, metabolism, excretion (ADME), and toxicity properties of chemical compounds.
The 22nd PhysChem Forum, hosted by Roche, focused on ‘Advances and Challenges in Physicochemical Profiling for Emerging Modalities in Drug Research’. Andrius Sazonovas, the Director of Percepta Solutions, shares his thoughts on the current hot topics in the world of physical chemistry and physicochemical properties.
1. Beyond Rule of 5 (bRo5)
Lipinski’s ‘Rule of 5’ was originally published to help scientists identify orally bioavailable drug candidates and initially the ‘Beyond Rule of 5’ (bRo5) compounds were largely dismissed. However, over time, it became clear that many targets and novel mechanisms (e.g., protein degradation) cannot be effectively “drugged” by the traditional small molecules, plus the evidence on successfully marketed orally bioavailable molecules that do not comply with the Rule of 5 also accumulated (e.g., Cyclosporine A, Atorvastatin, etc.). This realization, along with advances in drug delivery and permeability understanding, opened the door to the exploration of bRo5 chemical space, broadening the range of druggable targets and reshaping modern drug discovery.
The 2025 PhysChem Forum confirmed that interest and research into bRo5 compounds is not slowing down. PROTACs, macrocycles, oligopeptides, and other large, flexible molecules continue to challenge our understanding of permeability and bioavailability. More than half of the talks this year focused on how these molecules cross biological membranes and reach systemic circulation, with two major themes emerging:
Intrinsic Permeability Matters More Than Ever
Researchers are delving deeper into the properties that influence a molecule’s intrinsic permeability. The presentations discussed conformational flexibility, or “chameleonicity,” which describes a molecule’s ability to adopt different shapes under varying conditions. Both experimental and computational approaches are being used to characterize these properties, helping scientists identify trends and design bRo5 molecules that are intrinsically better at crossing membranes.
Improving Bioavailability Beyond Intrinsic Properties
The second theme explored strategies to enhance bioavailability beyond what a molecule’s inherent properties allow. Presentations ranged from traditional formulation strategies, such as physical state modifications, excipients affecting drug microenvironment and release, or even co-administering with other drugs adjusting gastrointestinal pH, to cutting-edge methods that modulate intestinal membrane structure and transporter behavior. A particularly memorable presentation from Randy Mrsny illustrated what could be described as true “bio-hacking” of the human gut to improve absorption. Collectively, these studies showed just how creative the field has become in improving the therapeutic potential of complex molecules.
2. Rethinking P-gp Efflux
Efflux by P-glycoprotein (P-gp) continues to be a hot topic, but this year’s discussions highlighted just how nuanced it really is. Simply classifying a compound as a P-gp substrate or non-substrate is not enough to explain bioavailability. What truly matters is the balance between passive permeability and active efflux. For instance, a compound with low passive diffusion may exhibit poor absorption even if it is not a P-gp substrate, while one with higher passive permeability may remain well absorbed despite being actively transported out of membrane cells.
This is where quantitative P-gp models would offer a major advantage, but developing them is a challenge of its own due to the high prevalence of semi-quantitative (censored) values among the experimental P-gp efflux data. Recently ACD/Labs has developed a modelling methodology that allows these data to be included in quantitative regression analysis.
Due to the a priori mechanistic considerations in descriptor selection, these machine learning models can predict the impact of efflux across a wide range of molecules, including challenging bRo5 compounds such as PROTACs.
This is evidenced by the validation study addressing this very topic, which we presented at this year’s PhysChem Forum, demonstrating that these models can successfully capture the effects of P-gp on the bioavailability of publicly available PROTAC datasets. This work highlights the value of moving beyond simple qualitative classifications and applying predictive, quantitative tools to guide the design and evaluation of emerging modalities.
3. Trustworthy Computational Models Are Still Key
The only truly computational presentation this year came from Raquel Rodriguez-Perez of Novartis. What stood out in the discussion that followed was a simple but crucial point: no matter how advanced the machine learning methods, evaluating a model’s applicability domain cannot be skipped. The most reliable approach remains the classic one—checking similarity to the training set, coherence across predictions in an ensemble, and consistency between predicted and experimental values for the most similar structures. These are the same principles that ACD/Labs has relied on and promoted in its GALAS models for over a decade.
Software for Physicochemical Property Prediction
ACD/Labs offers predictive software to empower scientists with information for data-driven decision-making without the need for measurement of physicochemical properties. PhysChem Suite provides tools to help you investigate structure modifications that meet the target property profile in addition to calculators for lipophilicity (logP, logD), ionization (pKa), aqueous solubility, rule of 5 compliance, and more.
About the Author
