What Bella taught us — ten years of cannabinoid discovery

A note on the image below: this photograph shows Bella during her illness, with visible skin lesions. It is included with her family's wish that her story is told honestly.

Bella, in early 2016. Six years old, 40 kilograms, and our entire world.

Ten years ago this month, Sandra and I started treating our six-year-old American Staffordshire Terrier, Bella, for an aggressive skin cancer that her vet — and our family friend of more than twenty years, Dr Baron Jonsson — had told us was going to take her in a matter of weeks. Cutaneous epitheliotropic T-cell lymphoma. Prognosis: a few weeks to three months at most.

What followed in the nineteen months between that diagnosis and the day Bella left us did two things that have shaped every choice Sandra and I have made since.

First, it opened our eyes to cannabinoids and the Endocannabinoid System. Until 2016, neither of us had any reason to know that mammals carry a regulatory system specifically tuned to cannabinoid molecules — endogenous ones we make ourselves, and plant-derived ones the cannabis genus has spent its evolutionary history producing. We learned about it because Bella's response, when we ran out of options, taught us it was real.

Second, it taught us — the hard way — that THC and CBD are not interchangeable in dogs. The two oils Bella was given over those nineteen months produced two completely different clinical pictures. We didn't have the language for it at the time. Ten years on, we do.

This is written for vets and for thoughtful pet parents. It assumes you already know the Endocannabinoid System exists. It doesn't promote a product. It's the post I wish I'd been able to read in May 2016.

We tried to do this the right way

Baron did his homework. In the days following Bella's diagnosis, he pulled together a treatment plan that drew on two distinct bodies of evidence.

The first was linoleic acid. A small but real peer-reviewed literature documents oral high-dose linoleic acid — typically delivered as safflower oil — as an adjunctive treatment for the form of cutaneous T-cell lymphoma Bella had. Iwamoto et al. (1992, Cancer Letters) documented remission in six of eight dogs.¹ The protocol is referenced in the standard veterinary oncology reviews at approximately 3 ml/kg orally.² For Bella, at 40 kg, that worked out to 240 ml of oil a day. Roughly a cup, morning and night, every day. To Baron, it was the evidence-based front line of what could be offered.

The second was cannabinoids. By 2016, the international literature suggested cannabinoids might play an adjunct role in some lymphomas. Baron and I discussed it. We agreed the evidence was preliminary, but given Bella's prognosis, it was worth investigating through legitimate channels.

Australia in 2016 had no cannabis-containing veterinary chemical product registered by the APVMA. It still doesn't.³ But the APVMA does administer a minor use permit scheme, with a companion consent-to-import pathway. Baron approached them. We got nowhere — not because anyone was obstructive, but because no veterinary cannabis product had ever been registered or permitted in Australia. The pathway was closed. Compounding wasn't an option either; medicinal cannabis was still three months from being legalised for humans.

We did not bypass the regulatory system. We tried every lawful channel available. None had an answer for us in Bella's timeframe.

What happened next did not come from a clinical decision. It came from a friendship.

Sandra's best friend Magda was, at that same moment, in the final stages of breast cancer. Her husband Les was caring for her at home. He had tried everything a husband can try when he is watching his wife die — and somewhere along the way, he had sourced a bottle of cannabis oil. We don't know where it came from. He didn't either, in any precise sense. He had it because he had been looking for anything that might help Magda.

Sandra and I had been spending much of that period supporting Les and Magda. Then Bella's diagnosis landed on top of everything else. One evening, sitting with us in the heaviness of it all, Les said five words that changed the next ten years of our lives:

"Why don't you try this with Bella?"

We discussed it with Baron, weighed it against Bella's timeline of weeks-to-three-months, and in early May 2016 we started Bella on the oil alongside the linoleic acid protocol.

Magda lost her battle a month later. Bella lived another nineteen months.

What happened next — and what we now recognise

Within three weeks, Bella's condition changed. The tumours on her thorax and nose stopped progressing. The nose lesion reduced. She ate, she ran, she was herself. We bought her fourteen months beyond the three-month outer edge of her original prognosis.

I want to be precise about what we can and can't attribute. We were running two interventions in parallel — high-dose linoleic acid, with its published veterinary evidence base, and an unknown-composition cannabis oil. It is genuinely not possible, looking back, to disentangle their contributions. We don't know which drove Bella's response, or whether it was the combination. For that alone, I remain grateful to both Baron and to Les — to Baron for his clinical thinking, and to Les for putting that bottle in my hand at the worst time of his own life.

But those fourteen months were also marked by signs we didn't have the vocabulary for at the time. Bella was periodically ataxic. She had episodes of urinary dribbling. She became hyperaesthetic. Her pupils would dilate. With ten years of retrospect and a much-matured veterinary literature, we now recognise these as textbook canine THC toxicity signs. The oil contained Δ9-THC at unknown concentration, with no batch testing and no way to titrate.

What happened when we found CBD

In April 2017, the original oil ran out. It took us five months to source a replacement. By September 2017, we'd found a locally-sourced CBD-predominant oil.

What we observed is the most important clinical data point in Bella's whole story:

No effect on the tumours. They had returned during the five-month treatment gap and the CBD oil did not arrest them the way the original oil had.

But Bella was noticeably more comfortable. Less pain behaviour, more settled, no ataxia, no urinary incontinence, no hyperaesthesia. Where the first oil had bought tumour regression at the clinical cost of THC toxicity, the CBD oil provided genuine wellness support without any of the same behavioural signs.

This is the moment HEMP PET began, intellectually. The contrast between those two oils — same plant family, radically different clinical pictures — forced Baron, Sandra and me to ask the question almost no Australian vet in 2017 could answer: what is actually in this bottle, and what is the functional difference between THC and CBD in dogs?

The answer to the first question was "nobody knows without a laboratory." The answer to the second has been transformed by peer-reviewed literature since.

What we now know — ten years of literature

The clinical evidence on canine cannabinoid biology has matured substantially since 2016, and three findings stand out for any vet or pet parent reading this:

The canine ECS is real, mapped, and clinically relevant. Freundt-Revilla and colleagues (2017) mapped CB1 receptor distribution across the canine central and peripheral nervous system, documenting expression across cerebral cortex, hippocampus, cerebellum, medulla, and spinal cord.⁴ The popular "10× more CB1 than humans" claim does not appear in the primary literature and should be treated as a hypothesis. What the paper does establish is that the canine ECS is architecturally primed in motor-coordination and autonomic regions — which is why dogs respond to cannabinoids the way they do.

THC toxicity is now well-characterised in Australian veterinary practice. Lauinger and Peacock's 2022 paper in the Australian Veterinary Journal documented fifteen Melbourne cases — ataxia (13/15), mydriasis (6/15), hyperaesthesia (5/15), urinary incontinence (4/15), stupor (3/15).⁵The signs Bella showed in 2016 are the signs that turn up in Melbourne casualty today. The largest modern series — Binagia et al., JAVMA 2024, 223 cases — reinforces the pattern with ataxia in 88.3% of cases.⁶

CBD has a fundamentally different safety profile. Gamble et al. (2018)⁷, Vaughn et al. (2020)⁸, and McGrath et al. (2019)⁹ have collectively established that CBD-predominant formulations are well-tolerated in dogs at standard doses, with the recurring tolerability signal being mild alkaline phosphatase elevations rather than CNS or behavioural effects. Bella's clinical course on the CBD oil in late 2017 was a preview of exactly this safety profile.

Disclosure: the Wakshlag/Gamble group has consulting ties to ElleVet Sciences; the Vaughn/Kulpa study was industry-funded by Canopy Growth. These should be weighed when reading the primary sources, but neither disclosure invalidates the safety signal.

And the product-quality problem is severe. Bonn-Miller et al. (JAMA 2017) found only 30% of online CBD products tested within ±10% of label CBD content.¹⁰ Wakshlag et al. (2020) — analysing 29 over-the-counter veterinary hemp supplements — found only 10 of 29 within 10% of label, four with heavy-metal contamination, and two with no detectable cannabinoids.¹¹ The TGA has separately warned that medicinal cannabis sourced outside the regulated supply chain may not be accurately labelled.¹²

The first oil we gave Bella in 2016 would almost certainly have been among those products. That's why batch-level testing matters, and why HEMP PET built its product around it.

What Bella taught us

Three lessons, plainly stated:

Two anniversaries this month

Bella's story happened inside a specific moment in Australian medical history. In May 2016, when we started Bella on treatment, medicinal cannabis was three months from being legalised for human use. The off-label veterinary prescribing pathway, as it operates today, did not yet exist in anything like its current form. The peer-reviewed canine literature on CBD was all still to come.

This month, alongside the ten years since we began treating Bella, marks another anniversary worth noting. On 1 May 2025, the APVMA quietly removed an incorrect statement from its public guidance — a statement that had wrongly suggested veterinarians could only prescribe imported cannabis-derived products. Australian vets have always retained, under state and territory veterinary practice legislation, the discretion to prescribe off-label what they consider clinically fit for purpose. The May 2025 correction did not "open up" a new prescribing right. It removed a misleading statement. The distinction matters.

In 2026, the broader landscape is almost unrecognisable from where we started. Medicinal cannabis is a standard therapeutic option for humans across the developed world. Full-spectrum cannabinoid preparations are now the subject of rigorous peer-reviewed research for diverse animal wellness applications — mobility, anxiety, sleep, appetite, settled behaviour. None of this was true when we lost Bella.

What has not changed — and this is the part I think about most — is how little of the underlying ECS science has reached the training of the medical professionals caring for our pets, our families, and ourselves. That gap is what HEMP PET is still working on.

A note on HEMP PET — and where ten years of expertise have led

Sandra and I founded HEMP PET because Bella's illness showed us, up close, what cannabinoids and the Endocannabinoid System could do — and how little of that was understood by the medical professionals around us. The brand exists for a simple reason: healthier pets and happier pet parents.

HEMP DROPS launched in December 2019. Six years on, we have become genuine experts in full-spectrum CBD for animal wellness — not because we set out to be, but because the questions Bella raised in 2016 and 2017 demanded answers we had to find ourselves. We grow Australian hemp, extract the cannabinoids, formulate the product, and test every batch — end-to-end in Australia, because that is the only way to know what's in the bottle. Every HEMP DROPS bottle ships with a current Certificate of Analysis from ACS Laboratories, an Australian NATA-accredited testing facility (ISO/IEC 17025).

Six years and many thousands of treated animals later, the wellness applications we hear about most often from the vets we work with are everyday quality-of-life ones: senior dogs moving more comfortably, anxious dogs settling more easily, cats and dogs eating better through stress or recovery. That is what full-spectrum CBD does well, and what HEMP DROPS is designed for.

Baron has continued that work with us. After Bella passed, Dr Jonsson formalised what had effectively been a twenty-year relationship by joining HEMP PET as Veterinary Advisor. His involvement is why I am confident writing a post like this one for Australian vets and pet parents: it has been read, questioned, and sharpened by the clinician who has been in the room from the start, and who knew Bella since she was a puppy.

Vets who want to know more about HEMP DROPS — composition, current batch CoAs, or the off-label prescribing pathway — can register at the vet portal at HempPet.au. Pet parents can ask their own vet about HEMP DROPS, or visit our consumer site for the broader nutritional range.

Medically reviewed by {{REVIEWER_NAME}}, {{REVIEWER_CREDENTIALS}}. Review date: {{REVIEW_DATE}}.