Patients with treatment-resistant epilepsy can develop a tolerance to cannabis-based therapy, a finding that may have implications for long-term management, new research suggests.

“Physicians should be aware that tolerance may occur,” study investigator Shimrit Uliel-Sibony, MD, pediatric epileptologist, Dana-Dwek Children’s Hospital, Tel Aviv University, Israel, told Medscape Medical News.

“There’s this notion that cannabis is a great medicine, that it’s much better in terms of response rates than other interventions for patients with treatment-resistant epilepsy, but that may only be at the beginning,” she added.

Long-term exposure may produce a different picture. With time, efficacy may decrease, resulting in a need to increase the dose. This suggests the development of tolerance.

“We know marijuana may have cognitive consequences with long-term, chronic recreational use. I think we need to be a little more cautious on our expectations of this therapy,” said Uliel-Sibony.

The study was presented here at the American Epilepsy Society (AES) 72nd Annual Meeting 2018.

Few Long-term Data

In the recent past, many studies have suggested that medical cannabis is superior to traditional antiepileptic drugs (AEDs). Studies have reported dramatic improvement in seizure control, as well as a favorable safety profile.

However, said Uliel-Sibony, there is a need for data on long-term safety and efficacy. The long duration of treatment, the investigators note, “raises the possibility of withdrawal and tolerance.” Some animal studies suggest that prolonged exposure to tetrahydrocannabinol (THC), the psychoactive ingredient in cannabis, but not cannabidiol (CBD) leads to tolerance.

To assess the long-term efficacy of cannabinoids as well as the development of tolerance in the treatment of refractory epilepsy, the investigators conducted an observational, longitudinal study of pediatric and adult patients.

The analysis included 92 patients (59 males) ranging in age from 1 year to 37 years (mean age, 11.8 years). All had treatment-resistant epilepsy; two had Dravet syndrome; and three had Lennox-Gastaut syndrome (LGS). The remainder had other types of seizures and epilepsy syndromes.

All patients had experienced treatment failure with numerous drug therapies, and for some, use of the ketogenic diet or vagal nerve stimulation therapy was also ineffective. None had been previously treated with medical cannabis.

Participants were initially given a cannabis oil extract in which the ratio of CBD to THC was 20:1. That amount of THC has negligible psychoactive activity, noted Uliel-Sibony.

The patients had been receiving stable doses of AEDs for at least 4 weeks before study enrollment. All participants had at least 3 months of follow-up (mean follow-up, 20 months).

No “Honeymoon” Tolerance

Tolerance was defined as a 30% or greater reduction in response rate that continued for more than 3 months.

Patients in whom efficacy decreased during the first 3 months of treatments were not included. The intention was to exclude patients who showed short-lasting improvement that was possibly the result of the “honeymoon” effect.

The investigators found that about 29% of the study population experienced a reduction in seizures by 50% to 75%.

Of 84 patients included in the tolerance analysis, 25% developed tolerance, which was reported with an average dosage of 12.6 mg/kg/day.

This dosage is lower than in some studies that used more purified cannabis. There was some suggestion that combining CBD with a very small amount of TCH “enables use of a lower dose of CBD,” a phenomenon referred to as the entourage effect, said Uliel-Sibony.

None of the patients with Dravet syndrome or LGS developed tolerance, she added.

In an attempt to counteract tolerance, the CBD dose was increased. Using this approach, about 20% of the patients were able to achieve the same response rate as before the development of tolerance, but this was not the case for almost half of the cohort.

In some patients, tolerance was reversed by gradually reducing the CBD dose, waiting 2 to 3 weeks, and then restarting the medication at the same dose that achieved response.

The tolerance may be due to “receptor desensitization,” said Uliel-Sibony. “If we take patients off the medicine, the receptors become sensitive again.”

This new study used two similar CBD products that are available in Israel. There was no difference between the two products in terms of tolerance.

It is not clear whether the tolerance seen in the study was due to the presence of THC. The literature on the subject suggests that there is no tolerance to CBD, only to THC, she said.

But it may be a question of exposure. In this study, the mean time for tolerance to occur was 7.3 months.

“Tolerance is something that develops over time,” said Uliel-Sibony.

Having a small amount of THC in the product seems to have some positive side effects, she said. Parents reported that children showed improvements in sleep and appetite and that they were more alert.

Great Expectations

She noted that overall, the side effect profile for the formulation is superior to that reported in other studies that used a more purified formulation of CBD. Of the 87 patients included in the safety analysis, 51% experienced adverse reactions.

Uliel-Sibony hopes to continue to follow the patients in the study for a longer period “to get a better picture” of tolerance, she said.

But she recognizes the challenges of following this patient population. Moreover, even if the number of seizures is reduced by half, for those patients who were having 10 seizures at the start of treatment, the condition is still not well controlled.

She believes expectations are high for cannabis-based therapy. She cited one study that showed that parents who relocated to Colorado just to access one of the early CBD treatments for epilepsy reported better response to the drug.

Families may be misled by some of the hype surrounding medical marijuana, said Uliel-Sibony. “Parents believe they’re giving their child something natural, something that isn’t really a chemical, but CBD is a medicine like any other.”

She noted that her study has limitations, including the fact that it is not randomized controlled trial.