Why doctors have a moral imperative to prescribe and support medical cannabis—an essay by David Nutt

Dear Editor,

In his argument that doctors have a moral imperative to “support medical cannabis,” David Nutt encourages cannabis prescription for various indications, generally without the justification provided by conventional evidence, such as randomized controlled trials. To make such sweeping advocacy seems radical, particularly since available evidence paints a rather incomplete and sometimes disappointing picture of cannabinoid efficacy (1).

Equally concerning is Nutt’s tendency to downplay risks of harm from cannabis, which range from motor vehicle accidents (2) and emergency presentations (3) to severe psychiatric illness. Links between high dose THC and psychosis (4) are particularly discounted, at variance with the experience of inpatient psychiatrists around the world. Cannabis use is associated with increased contact with mental health services and poorer prognosis of anxiety and depressive disorders as well (5) (6).

The ‘moral imperative’ to prescribe cannabis should be trumped by our ethical obligation to act in patients' best interest, based on available evidence and tailored to the individual. While low rates of cannabis prescription in the UK may be due, in part, to stigma and organisational inertia, they may also reflect doctors’ wariness of adverse effects and a cautious wait for better evidence of efficacy.

1. Menkes DB. The double edged sword of medicalising cannabis. BMJ. 2019;365:l4066.
2. Asbridge M, Hayden JA, Cartwright JL. Acute cannabis consumption and motor vehicle collision risk: systematic review of observational studies and meta-analysis. BMJ. 2012;344:e536.
3. Zhu H, Wu LT. Trends and Correlates of Cannabis-involved Emergency Department Visits: 2004 to 2011. J Addict Med. 2016;10(6):429-36.
4. Murray RM, Quigley H, Quattrone D, Englund A, Di Forti M. Traditional marijuana, high-potency cannabis and synthetic cannabinoids: increasing risk for psychosis. World Psychiatry. 2016;15(3):195-204.
5. Bahorik AL, Leibowitz A, Sterling SA, Travis A, Weisner C, Satre DD. Patterns of marijuana use among psychiatry patients with depression and its impact on recovery. J Affect Disord. 2017;213:168-71.
6. Duperrouzel J, Hawes SW, Lopez-Quintero C, Pacheco-Colón I, Comer J, Gonzalez R. The association between adolescent cannabis use and anxiety: A parallel process analysis. Addict Behav. 2018;78:107-13.

Dear Editor

The causal connection of cannabis with mental illness is as impressive and wide ranging as it is legendary. Depression, anxiety, bipolar disorder, schizophrenia, various personality disorders, an amotivational state and suicide are all well substantiated as being causally linked with cannabis exposure [1]. Cannabis has also been mentioned with heritable syndromes of mental illness include ADHD-like, and autism-like features [2]. Indeed one detailed study linked the incidence of cannabis exposure at substance level with any mental illness, serious mental illness, major depressive illness and suicidal ideation both across space time and in a causal inferential modelling analysis [1]. Indeed in USA the incidence of autistic spectrum disorder has been shown to be growing at an exponential rate [2] and Colorado lawmakers have moved to have it declared an epidemic. These findings from pediatric populations clearly indicate the transgenerational heritability of cannabis neurotoxicity – which is a very serious multigenerational impact presently not factored in to public health discussions.

In general theoretical terms one might suppose that either that cannabinoids may be inhibiting fundamental processes in the cells of the brain (mainly neurons, glia, microglia and vascular cells) or affecting many diverse processes all at once – or perhaps some combination of the two. The challenge in explaining the underlying mechanisms for this host of neurotoxic manifestations is to know which of the many demonstrated mechanistic links are most responsible.

Described pathways include inhibition of mitochondria activity limiting energy supply to neurons, inhibition of mitochondrial activity limiting energy and substrate supply to the epigenomic and genomic machinery, direct interference with the neurexin- and neuroligin- scaffolding which forms the framework of synapses, interference with the actin cytoskeleton of synaptic boutons and the machinery that allows them to grow with use and forms one of the structural bases of memory and long term synaptic potentiation, 88% reduction in white matter connections across the brain including splenium to precuneus connection [3], altered cerebellar connectivity with amygdala and anterior temporal pole, epigenomically altered D2R2 dopamine receptor signalling which is a gene linked with autism and schizophrenia, direct interference with the lysine demethylases of the epigenomic machinery, potentiation of low grade brain inflammation including suppression of neuronogenesis, and inhibition of the slit-robo ligand receptor complex which allows and facilitates the massive outgrowth of the human forebrain.

A fascinating paper just appeared describing control of the massive outgrowth of the human prefrontal cortex (PFC), the locus of most higher computations and learning consolidation in the brain and one of whose primary organic substrates is the unusually high rate of synapses and connectivity in this region [4]. PFC synaptogenesis and long range connections are controlled by a retinoic acid gradient which is elevated in the frontal pole and depressed at the premotor cortex [4]. High local gradients are established by ALDH1A1, transduced by retinoic acid receptors RXRG and RARB, and catabolised by CYP26B1 posteriorly. Moreover a companion paper showed that PFC spinogenesis was controlled by cerebellin 2.

Fascinatingly every one of these key synaptogenesis controllers has recently been shown to be disrupted epigenomically by cannabis dependence and withdrawal [5]. Searching the term “neuro” in the supplementary material reveals 555 hits; for brain there were 15 hits, for neurological disease 67 hits, 4 hits for cerebral disorders and 32 references to neurons. Indeed for cerebellin 4, which is distributed more in the midbrain and hindbrain there were 71 hits!

When confronted with the appalling epidemic of cannabis related serious mental illness we simply cannot afford to be mechanistically naïve regarding the pleiotropic neurotoxicity of numerous cannabinoids of which their exponential effects in the higher dose range and multigenerational and epigenomic inheritability are its most concerning features. Without properly factoring in this multifaceted, profound and multigenerational neurotoxicity formulation of viable public health policy is impossible.

References

1. Reece AS, Hulse GK: Co-occurrence across time and space of drug- and cannabinoid- exposure and adverse mental health outcomes in the National Survey of Drug Use and Health: combined geotemporospatial and causal inference analysis. BMC Public Health 2020, 20(1):1655.
2. Reece A.S., Hulse G.K.: Effect of Cannabis Legalization on US Autism Incidence and Medium Term Projections. Clinical Pediatrics: Open Access 2019, 4(2):1-17.
3. Zalesky A, Solowij N, Yucel M, Lubman DI, Takagi M, Harding IH, Lorenzetti V, Wang R, Searle K, Pantelis C et al: Effect of long-term cannabis use on axonal fibre connectivity. Brain 2012, 135(Pt 7):2245-2255.
4. Shibata M, Pattabiraman K, Lorente-Galdos B, Andrijevic D, Kim SK, Kaur N, Muchnik SK, Xing X, Santpere G, Sousa AMM et al: Regulation of prefrontal patterning and connectivity by retinoic acid. Nature 2021, 598(7881):483-488.
5. Schrott R, Murphy SK, Modliszewski JL, King DE, Hill B, Itchon-Ramos N, Raburn D, Price T, Levin ED, Vandrey R et al: Refraining from use diminishes cannabis-associated epigenetic changes in human sperm. Environmental Epigenetics 2021, 7(1):1-10.

Dear Editor

One of the abiding mysteries of cannabis genotoxicity has for some time been “Why is cannabis so destructive to chromosomes?”

The signal for chromosomal congenital anomalies is amongst the strongest of all the congenital anomaly relationships of cannabis exposures and is uniformly observed in large series from Canada, Australia, Hawaii, Colorado, USA and Europe [1, 2]. Chromosomal mis-segregation syndromes are a prominent feature of cannabis teratogenicity. Hence trisomies (trisomy 13, 18 and 21; syndromes described by Patau, Edwards and Down), disomies (Klinefelter male XXY; male disomy X), and monosomies (Turner syndrome female XO), have been observed wherever data is available [1, 2].

Deletions and microdeletions are observed in both USA and Europe [1, 2] making double stranded DNA breaks also a notable clinical feature of cannabis-related chromosomal anomalies.

Testicular cancer is widely recognized as being cannabis-related and has been shown in metanalysis to develop 2.6-fold faster than background. This disorder is characterized by an isochromosome 12 usually involving a centromeric break and a reduplication and joining of the short arms of this chromosome, along with chromosomal anomalies of ten other chromosomes [3]. Non-seminoma has been described as losing up to 70 chromosomal arms trimmed after whole genome reduplication events [3]. Remarkably, given a median age of cannabis exposure of around 20 years of age, and a median age of testicular cancer incidence of 33 years, this implies an acceleration of the usual oncogenic incubation period by 2.5-fold. 2.5 x 2.6 = 6.5-fold acceleration in the frequency-oncogenic induction rate index.

Both acute lymphoid and myeloid leukaemias of childhood with which cannabis is associated are also linked with stable oncogenic chromosomal translocations [4].

So why is cannabis so genetically destructive?

One fascinating possibility was provided by a recent profoundly insightful whole epigenome screen of human sperm in cannabis addiction and cannabis withdrawal [5]. It is well known that the sister chromosomes line up on the metaphase plate prior to their separation during mitotic anaphase. The chromosomes are bound to the ends of the microtubules of the two mitotic half-spindles by a complex multiprotein network of 90 proteins called the kinetochore and each sister chromatid is pulled polarwards by cellular kinesin motors. The kinetochore binds to a highly repetitive central segment of the chromosome called the centrosome which is marked by specialized chromatin and nucleoproteins especially those of the CENP (centromere proteins) family, initially CENPA which is a modified form of histone 3. Clearly anything which disturbs centromeric or kinetochore function will disrupt this delicate machinery.

US researchers showed that 16 CENP proteins were inhibited by cannabis exposure including 105 hits overall and 66 hits for CENPN which is the second protein in the sequence to bind after CENPA [5]. Moreover they showed dramatic inhibition of the high fidelity pathway for DNA damage repair (DDR) known as homologous repair (HR) with nine hits for inhibition of the main HR enzyme RAD51. RAD52, which heads the low fidelity highly mutagenic microhomology end joining pathway therefore becomes the main workhorse for centromeric DDR repair and the cycle becomes self-perpetuating through centromeric breakage-fusion-bridge cycles [4].

Hence these epigenomic insights may possibly explain prominent chromosomal mis-segregation, double stranded DNA breaks, and aggressive acceleration of repeated chromosomal breakage-fusion-bridge cycles observed epidemiologically in cannabinoid carcinogenesis, cannabinoid teratogenesis and to some extent cannabis-accelerated ageing.

References

1. Reece A.S., Hulse G.K.: Cannabinoid- and Substance- Relationships of European Congenital Anomaly Patterns: A Space-Time Panel Regression and Causal Inferential Study. Environmental Epigenetics 2022, 8(1):1-40.
2. Reece A.S., Hulse G.K.: Geotemporospatial and Causal Inferential Epidemiological Overview and Survey of USA Cannabis, Cannabidiol and Cannabinoid Genotoxicity Expressed in Cancer Incidence 2003–2017: Part 1 – Continuous Bivariate Analysis. Archives of Public Health 2022, In Press.
3. Shen H, Shih J, Hollern DP, Wang L, Bowlby R, Tickoo SK, Thorsson V, Mungall AJ, Newton Y, Hegde AM et al: Integrated Molecular Characterization of Testicular Germ Cell Tumors. Cell Rep 2018, 23(11):3392-3406.
4. Yilmaz D, Furst A, Meaburn K, Lezaja A, Wen Y, Altmeyer M, Reina-San-Martin B, Soutoglou E: Activation of homologous recombination in G1 preserves centromeric integrity. Nature 2021, 600(7890):748-753.
5. Schrott R, Murphy SK, Modliszewski JL, King DE, Hill B, Itchon-Ramos N, Raburn D, Price T, Levin ED, Vandrey R et al: Refraining from use diminishes cannabis-associated epigenetic changes in human sperm. Environmental Epigenetics 2021, 7(1):1-10.

Dear Editor

I read David Nutt’s essay “Why doctors have a moral imperative to prescribe and support medical cannabis” with interest. I am a member of the British Paediatric Neurology Association (BPNA), but these are entirely my personal views.

In November 2018 cannabis was re-assigned to schedule 2 legal status in the UK, allowing cannabis-based medicinal products (CBMP) to be legally prescribed. Professor Nutt argues that “stigma, fear and entrenched resistance in the medical profession” explain why, more than three years later, only a handful of prescriptions have been made. I wonder if it is more likely that the primary reason for the low number of prescriptions is a lack of any substantial evidence-base. We must be careful not to interpret a change in legal status as a carte blanche to prescribe CBMP for any indication, regardless of supportive evidence. Clinicians need to know that a medication is safe and efficacious before they can confidently prescribe it. This goes for all medications, whether they are plant-derived, or have historically been used for recreational purposes, or not.

In his essay, I wonder if Professor Nutt has made a “straw man” out of the BPNA in accusing the association of a “refusal to recommend the prescription of medical cannabis to children with severe treatment refractory epilepsy.” As I understand it, producing guidance and “recommending medications” is not one of the roles of the BPNA, nor is it something the association has specifically done for any one of the 30+ anti-seizure medications that are prescribed to children by its members. For epilepsy treatment, the BPNA website provides a link to the National Institute of Health and Care Excellence (NICE) guidelines on management of epilepsies in children. NICE, in turn, regularly appraises the evidence and cost-effectiveness of new medicinal products.

Professor Nutt states that CBMP have “shown unprecedented efficacy.” In support of this bold statement, he references just one paper (on which he is an author) – a case series of 10 children, published in BMJ Paediatrics Open. It is well worth reading the authors’ own section on the limitations of their data in that paper (page 5), but I will summarise it here. This was a trial with no control group. Seizure frequency changes were retrospective and based on parent recall. There was likely bias towards selection of parents who were motivated to demonstrate efficacy of CBMP since they were recruited from pro-cannabis support charities. This section for his paper demonstrates that Professor Nutt and his co-authors have insight into what methodology could be used to generate a more robust evidence base.

I would add to the limitations of the identified by the authors of the BMJ Paediatrics Open paper two more. Firstly, there is no detail at all on seizure types. This is a detail that is important in epilepsy trials. Some seizure types can be more accurately counted, and are more disruptive for patients, than others. Secondly, analysis was not done on intention to treat. Five patients with “incomplete data” were simply removed from analysis without any description of exactly what data were missing. The authors recognise that “further [and I would add more robust] research is required” if we are to understand whether these products are likely to benefit our patients. I agree.

Prior to the UK law change in 2018, proponents of CBMP argued that a change in legal status would facilitate more robust investigation of these products and allow proper trials to take place without legal barriers. I am open to the idea that these products may benefit my patients, but if we are to make progress, we need to have a constructive dialogue between proponents of CBMP and the clinicians who may one day be prescribing them. If Professor Nutt wants to achieve better treatment for children with epilepsies, I struggle to understand what he hopes to achieve by demonising the BPNA as he has in this essay, since most of the clinicians initiating new medications in severe therapy-resistant childhood epilepsies are members of this association. I would like to think that Professor Nutt has no interest in allowing this debate to become polarised. The most obvious way forward for the CBMP movement, which enjoys significant financial backing from the wider pro-cannabis lobby, is to focus resources on investigating these products robustly, rather than (for example) discrediting professionals who devote their careers to looking after some of the most vulnerable children.