Cannabis-Derived Compounds Target Colorectal Cancer

Cannabis research continues to uncover new possibilities that extend far beyond the plant’s well-known cannabinoids. While much of the public conversation has focused on compounds such as THC and CBD, scientists are increasingly exploring cannabis as a source of entirely new drug candidates that could one day be engineered to treat a wide range of diseases.

A 2026 computational study¹ offers an intriguing example of this approach. Rather than testing cannabis extracts directly against cancer, researchers used advanced computer modeling to design and evaluate new cannabis-derived molecules that may interfere with one of colorectal cancer’s key drivers. After screening dozens of compounds through multiple layers of virtual testing, two candidates emerged as especially promising.

Although these findings remain at the earliest stage of drug discovery and have not yet been validated in laboratory or human studies, they highlight how cannabis chemistry could serve as a valuable starting point for developing more targeted cancer therapies. Let’s dive into the research and what it means for the future of cannabinoid medicine.

Why Colorectal Cancer Needs New Treatments

Colorectal cancer is one of the most commonly diagnosed cancers worldwide and remains a leading cause of cancer-related deaths. While surgery, chemotherapy, radiation, targeted therapies, and immunotherapy have improved patient outcomes, treatment can still be challenging, particularly when cancers become resistant to existing drugs.

One strategy researchers are pursuing involves identifying highly specific molecular targets that cancer cells rely on for growth while minimizing damage to healthy tissue. This precision medicine approach aims to improve treatment effectiveness while reducing unwanted side effects. This study focused on cyclin-dependent kinase 2, better known as CDK2.

Understanding CDK2

CDK2 is an enzyme that helps regulate the cell cycle, which is the process where cells grow and divide. Healthy cells rely on this system to replace damaged tissue and maintain normal function. Cancer cells, however, often exploit these pathways to divide uncontrollably.

When CDK2 activity becomes dysregulated, it can contribute to rapid tumor growth. Because of its important role in cell proliferation, CDK2 has become an attractive target for developing new anti-cancer drugs. If scientists can safely inhibit CDK2 in cancer cells, they may be able to slow or stop tumor growth.

Using Cannabis Chemistry for Drug Discovery

Instead of searching randomly for potential drugs, the research team combined several advanced computational techniques that allow scientists to predict which molecules are most likely to succeed before entering expensive laboratory testing.

The researchers began with a collection of 33 cannabis-derived compounds that had previously demonstrated activity against colorectal cancer cells. Using quantitative structure-activity relationship (QSAR) modeling, specifically CoMFA and CoMSIA, they analyzed how subtle differences in molecular structure affected anti-cancer activity. These models allowed the researchers to identify structural features associated with stronger biological effects and use that information to design 20 entirely new molecules, labeled V1 through V20.

This process represents a modern form of rational drug design. Rather than relying solely on trial and error, scientists use computational tools to predict which molecular changes may improve a compound’s performance before synthesizing it in the laboratory.

Narrowing the Best Candidates

Designing new molecules is only one part of the challenge. Potential drugs must also possess characteristics that allow them to function inside the human body. The research team therefore evaluated each new compound for drug-likeness and predicted ADMET properties, which estimate how a compound may be absorbed, distributed, metabolized, excreted, and whether it could present toxicity concerns. Among the twenty newly designed molecules, four stood out as they demonstrated favorable predicted pharmacokinetic properties while maintaining strong anti-cancer potential.

This filtering process helps researchers avoid investing time and resources in compounds that may appear biologically active but are unlikely to become viable medicines.

Molecular Docking Revealed Strong Binding

After narrowing the list, the researchers used molecular docking simulations to predict how tightly each compound bound to the enzyme and identified the interactions responsible for that binding. Among all candidates, V7 and V8 produced the strongest binding to CDK2, outperforming the reference compound used for comparison.

Strong binding is important because it increases the likelihood that a compound can effectively interfere with the protein’s normal function. However, docking represents only a static snapshot. Researchers also wanted to determine whether these interactions remained stable over time.

Simulating Real-World Protein Behavior

To answer that question, the team conducted molecular dynamics simulations lasting 100 nanoseconds. Unlike molecular docking, which captures a single predicted interaction, molecular dynamics simulates how proteins and drug molecules move naturally over time. This creates a more realistic picture of whether a potential drug remains securely attached under changing conditions.

Throughout the simulations, both V7 and V8 maintained stable interactions with CDK2. The researchers evaluated several indicators of stability, including:

• Root mean square deviation (RMSD), which measures overall structural stability.
• Root mean square fluctuation (RMSF), which examines movement of individual protein regions.
• Radius of gyration (Rg), which reflects protein compactness.
• Solvent-accessible surface area (SASA), which estimates how much of the protein remains exposed to surrounding water.
• Molecular surface area (MolSA), another measure of structural behavior.

Together, these analyses suggested that both compounds formed stable protein-ligand complexes throughout the simulation period. However, among the two leading candidates, V8 consistently demonstrated the most favorable dynamic behavior.

What Makes This Study Different?

One of the study’s biggest strengths is its comprehensive computational workflow. Rather than relying on a single prediction method, the researchers combined multiple independent techniques, including QSAR modeling, drug-likeness screening, ADMET prediction, molecular docking, and molecular dynamics simulations. Each stage served as a checkpoint, gradually narrowing the list from dozens of possible molecules to two particularly promising candidates.

This layered approach increases confidence that selected compounds deserve further investigation while reducing the number of molecules requiring costly laboratory experiments.

Important Limitations

Despite the encouraging findings, it is important to recognize what this study does, and does not, demonstrate. Everything reported was generated through computational modeling. None of the newly designed compounds have yet been tested in living cells, animal models, or human clinical trials. While computer simulations are valuable because they help researchers prioritize the most promising candidates, they cannot prove that a molecule will be safe or effective as a medicine.

Future research will need to:

• Synthesize the newly designed compounds.
• Test them in colorectal cancer cell cultures.
• Evaluate their safety in animal studies.
• Determine whether they perform as predicted in laboratory experiments.
• Eventually assess safety and effectiveness through clinical trials.

Many promising computational drug candidates never advance to approved therapies, making experimental validation an essential next step.

Cannabis May Offer a New Drug Discovery Platform

Perhaps the most significant takeaway from this research is the remarkable chemical diversity found within cannabis that illustrates how that diversity can inspire the design of entirely new pharmaceutical compounds. Rather than using cannabis itself as the treatment, researchers are leveraging its molecular framework as a blueprint for creating more selective, drug-like molecules capable of targeting specific disease pathways.

While these findings remain firmly in the preclinical discovery phase, they highlight the growing role of computational drug design in accelerating pharmaceutical research. As artificial intelligence, molecular modeling, and medicinal chemistry continue to evolve, cannabis-derived compounds may increasingly contribute not only to cannabinoid medicine but also to the broader search for next-generation cancer therapies.

References:

^{1. Khadija Khaddam Allah, Khaoula Mkhayar, Rachid Haloui, Amine Ballari, Ossama Daoui, Kaouakeb El Khattabi, Abdelmoula El Abbouchi, Samir Chtita, Souad El Khattabi, Cannabis as a source of novel therapeutics: A computational study combining LBDD and structure-based docking/dynamics to identify novel drug-like compounds for colorectal cancer, Journal of Genetic Engineering and Biotechnology, Volume 24, Issue 3, 2026, 100763, ISSN 1687-157X, https://doi.org/10.1016/j.jgeb.2026.100763}