When we talk about cannabis in the context of health, we most often talk about a handful of compounds that most people have heard of — CBD, THC, perhaps CBG. These are the cannabinoids that dominate commercial conversations, regulatory discussions, and the wellness industry. But the cannabis flower is chemically far more complex than any of those conversations suggest.

A study published in April 2025 in the journal Pharmaceuticals, authored by researchers at Wonkwang University in South Korea, took a systematic approach to that complexity. They extracted and isolated eleven separate compounds from the flowers of Cannabis sativa. Two of those compounds had never been described in the scientific literature before. Three others had never previously been found in this plant at all. And then the researchers did something critical: they tested every compound against SK-N-SH neuroblastoma cells — the type used to study a cancer that primarily strikes children — and measured what happened.

The findings add an important dimension to how we understand the therapeutic potential locked inside the cannabis plant, and they raise questions that will matter both for cancer research and for the broader scientific understanding of what cannabis actually contains.

What Is Neuroblastoma — And Why Does It Matter Here

Before getting into what the researchers found, it is worth understanding what neuroblastoma is and why this line of research is clinically significant.

Neuroblastoma is a solid tumour that develops in nerve tissue, most commonly in the adrenal glands situated above the kidneys. It is the most common solid cancer in children and the most frequent malignancy in the first year of life. Approximately 65% of primary tumours are found in the abdomen. Standard treatment involves chemotherapy, surgery, and radiation — but for children with advanced disease, outcomes remain poor, and many patients do not respond adequately to existing therapies. New treatment strategies are urgently needed, and identifying new biologically active compounds that interfere with neuroblastoma cell growth is a meaningful step in that direction.

Previous research has shown that cannabis plant extracts with rich cannabinoid content can exhibit antitumour effects in human neuroblastoma cells. This study set out to go further — to isolate individual compounds, identify precisely what they are, and test each one separately so the science can distinguish which specific molecules are responsible for any effect observed.

Two Classes of Compounds — One Expected, One a Genuine Surprise

The eleven compounds isolated in this study fall into two chemically distinct groups: cannabinoids, which anyone familiar with cannabis will recognise, and chlorin-type compounds, which are something else entirely.

The chlorin compounds are the real surprise. Cannabis sativa has been studied for decades. Its chemical constituents — cannabinoids, flavonoids, terpenes, phenolics, and alkaloids — are well catalogued, with more than 560 potentially bioactive compounds identified across the literature. But chlorin-type compounds had never been reported in this plant before. Their discovery here expands the known chemistry of cannabis in a direction nobody was looking.

What Chlorin-Type Compounds Are — And Why They Are Interesting

Chlorins are a class of compounds structurally related to porphyrins — the same family as haemoglobin and chlorophyll. They are found in most higher plants and have attracted serious scientific interest as potential drug candidates, primarily in oncology.

Chlorin derivatives have been shown to exhibit strong toxicity against cancer cell lines, including human lung carcinoma, breast adenocarcinoma, malignant melanoma, and ovarian carcinoma. They are also structurally suited for photodynamic therapy — a cancer treatment in which a compound is activated by light to generate reactive oxygen species that destroy tumour cells. The discovery that cannabis flowers contain compounds from this class, including two new ones, raises a question the field has not previously had reason to ask: what role, if any, do these molecules play in the biological effects associated with cannabis use and the plant's therapeutic potential?

The Bioactivity Results — What Actually Happened in the Cancer Cells

Using the MTT assay — a standard method for measuring cell viability — the researchers exposed SK-N-SH neuroblastoma cells to each of the eleven isolated compounds at concentrations of 2.5, 5, and 10 µM and measured survival.

Three compounds were inactive at all concentrations tested: cannabielsoxa (the new cannabinoid, compound 1), pyropheophorbide A (compound 3), and dihydroxy-cannabidiol (compound 11). The remaining seven compounds — compounds 2, 4, 5, 6, 7, 8, 9, and 10 — all showed significant inhibitory activity against neuroblastoma cell proliferation.

The Molecular Docking Analysis — How Do These Compounds Interact With the Target?

The researchers did not stop at measuring cell death. They used molecular docking simulations to investigate the mechanism by which the most active compounds interact with a known neuroblastoma target protein.

The target used was the crystal structure of the SK-N-SH neuroblastoma protein with the Protein Data Bank identifier 4KUM. The three compounds with the highest bioactivity — CBDA (6), CBDA methyl ester (7), and CBD (8) — were docked into its active site and their binding scores calculated.

The in silico results aligned with the in vitro experiments — the compounds that showed the strongest cell-killing activity in the MTT assay also showed the most favourable binding to the target protein computationally. This coherence between experimental methods strengthens the credibility of the findings.

The Drug-Likeness Question — Could These Compounds Actually Become Medicines?

Identifying a compound that kills cancer cells in a lab dish is an early step. Whether that compound could ever become a medicine depends on a different set of properties entirely — how it is absorbed, distributed, metabolised, and excreted in the body, and what its toxicity profile looks like. This is what ADMET analysis assesses.

On the positive side, all three leading compounds showed high gastrointestinal absorption, acceptable solubility, and no P-glycoprotein substrate activity — meaning they would not be easily pumped out of cells by the body's own drug-resistance mechanisms. CBD (compound 8) was also predicted to cross the blood-brain barrier, which is particularly relevant for a neurological cancer.

None of the compounds violated Lipinski's Rule of Five — the standard pharmaceutical benchmark for oral bioavailability. In other words, from a basic drug-likeness perspective, these molecules look like viable candidates for further development, with appropriate caution around drug interaction profiles.

What This Means for How We Think About the Cannabis Plant

There is a tendency in conversations about cannabis — both inside the industry and in mainstream media — to reduce the plant to two or three key molecules. THC for psychoactivity. CBD for everything else. That reduction is practically understandable but scientifically misleading.

This study is a reminder of what the cannabis flower actually is: a complex chemical factory producing over 560 potentially bioactive compounds across multiple chemical classes. The cannabinoids are one class. The terpenes are another. The flavonoids, the phenolics, the alkaloids — all distinct. And now, apparently, chlorin-type compounds as well, including two that nobody knew existed until this year.

For growers and cultivators, the implications are also worth sitting with. The research was conducted on the flowers of Cannabis sativa — the same part of the plant that is harvested commercially. The chemical profile of those flowers is shaped by genetics, growing conditions, harvest timing, and post-harvest handling. The compounds that showed anti-neuroblastoma activity in this study are not exotic laboratory creations — they are molecules that exist, in varying quantities, in the flower material that the industry already produces.

The Honest Limitations

Scientific integrity requires stating clearly what this study is and what it is not. It is an in vitro study — meaning the compounds were tested on cancer cells in a laboratory dish, not in a living organism. Cell-based studies like this are how drug discovery begins, but they are many steps removed from a clinical treatment.

The molecular docking analysis is a computational prediction, not a direct observation of how these molecules behave in a human body. ADMET properties computed by SwissADME are estimates based on algorithms, not clinical measurements. This research is promising early-stage science. It is not a clinical recommendation, and it should not be interpreted as evidence that cannabis treats childhood cancer.

The researchers themselves are clear on this point, noting that in vivo and clinical studies are necessary to confirm the therapeutic potential observed here. They also acknowledge that strict regulations in South Korea around cannabis limited their ability to conduct in vivo work as part of this study — an honest reflection of the regulatory reality that shapes what cannabis research can and cannot do in different jurisdictions.

What the study does establish is a scientifically credible basis for further investigation, a set of specific compounds worth pursuing, and a structural understanding of why some molecules in this plant appear more active than others. That is meaningful scientific progress, even if it is progress measured in early steps rather than finished answers.

The practical takeaway for anyone engaged with this plant — whether as a grower, a patient, a clinician, or a researcher — is the same one that emerges from the broader science: cannabis is more chemically complex than the industry's current vocabulary gives it credit for. The compounds that matter therapeutically may not all be the ones most prominently on the label. And the research that will eventually clarify the picture is still, in meaningful ways, just beginning.