This research aimed at gaining knowledge on the usage of cannabinoid substances among cancer patients receiving systemic anti-cancer treatment in the Netherlands. The current study revealed that almost a quarter of the cancer patients have used cannabinoid substances for medical purposes, with a prevalence of active users of 15%. Potential future medical usage by current non-users was reported to be 23%. CBD-oil was the most frequently reported way of consumption. Users consisted mainly of patients undergoing treatment with palliative intent. More than half of the patients reporting use of cannabinoid substances, received immunotherapy treatment. Intention of treatment, as well as type of therapy, turned out to be predictive factors for cannabinoid consumption. Although motivations for usage varied, about half of the users, reported the supposed anti-cancer properties of cannabinoids as their motivation to engage in consumption of these substances.

The high prevalence of self-reported use of cannabinoid substances among cancer patients in the Netherlands, is in line with the findings from several other studies, suggesting that cannabinoid usage among this patient group should not be underestimated. A Canadian study, performed prior to the Canadian Cannabis Act, reported a similar prevalence of active users of 18% (Martell et al. 2018). We expected the cannabinoid usage in our study to be higher than reported in the Canadian study, for several reasons. First of all, the current study is the first study selecting patients receiving systemic therapy, which was a predictor for cannabinoid consumption in the Canadian study (Martell et al. 2018). Furthermore, in Canada consumption of cannabis for medical purposes was an exemption under the law prior to Canadian Cannabis Act in 2018 (Martell et al. 2018), while in the Netherlands both recreational as well as medical use of cannabis are tolerated though not legalized (Gielen and de Vrey 2021). This would justify higher prevalence of users in the current study compared to the Canadian study. Pergam et al. (2017), who performed a similar study in Washington, where cannabis has been fully legalized, revealed a prevalence of active users of 24% (Pergam et al. 2017). Although cultural differences between the USA and the Netherlands could affect the prevalence of consumption, a more likely explanation is the methodology of both studies. In the Pergam study a considerable percentage of users reported exclusive recreational use, whereas our study allowed for specifying consumption based on medical intent. The innovative aspect of the current study, including the selection of patients receiving systemic therapy combined with the potential exclusive focus on consumption of cannabinoid substances with medical intent, contributes to the differences in prevalence found between the studies and allows for a conscientious appraisal of medical cannabinoid consumption among cancer patients in the Netherlands receiving systemic therapy, which turned out to be clinically significant.

The prevalence of cannabinoid users we found in the current study, is considerably higher than the reported prevalence rate of prescribed cannabis in the general population (Stichting Farmaceutische Kengetallen 2021). This difference is mainly due to the fact that most patients rely on a different resource for their cannabinoids, as also reported in the current study. Additionally, the prevalence found in the current study, is higher than the most recently reported yearly prevalence of cannabis in the general population (7.5%) of the Netherlands (van Laar and van Gestel 2019), especially when specified for the age group of the current study (1.8%) (Trimbos Instituut 2017). These findings imply an increased susceptibility to engagement in consumption of cannabinoids among the oncology patients within this age group.

The consumption of cannabinoids by cancer patients is significant, particularly when compared to the general population. The observation that over half of the patients that reported active use of cannabinoids, concurrently received immunotherapy, may be of concern. The immunomodulatory role of cannabinoid substances in cancer is not yet clear and their safety during immunotherapy is not guaranteed. Recent data from a prospective observational study by Bar-Sela et al., reported decreased time to progression and shorter overall survival for cannabis-users receiving immunotherapy treatment (Bar-Sela et al. 2020). Additionally, in an earlier retrospective observational study of Taha et al. (2019), consumption of cannabis was associated with reduced response rates to Nivolumab in patients with advanced cancer (Taha et al. 2019). Underlying mechanisms concern the number and functioning of available lymphocytes, which might be altered through exposure to cannabinoids (Bar-Sela et al. 2020). These findings suggest that adjunctive treatment with cannabinoids should be approached with caution.

A commonly reported symptom for consumption of cannabinoids, was the treatment of pain, which could imply inadequate pain management or pain refractory to commonly used pain medication, including opioids. Although in total only 4 consumers reported an NRS score equal or higher than 5, information about additional pain medication was incomplete. Despite the limited evidence, the Dutch Guideline for Policy and Treatment of Pain in Cancer, advises to consider consumption of cannabinoids in pain refractory to other pain medication (Federatie Medisch Specialisten 2019). The consumption of cannabinoids to manage pain, is in line with previous studies, where also mainly the expected analgesic effect was reported as the motivation for usage (Donovan et al. 2019; Mousa et al. 2019; Pergam et al. 2017). However, whereas in other studies particularly THC was consumed (Martell et al. 2018; Mousa et al. 2019; Pergam et al. 2017), the current study reported CBD-oil as the most common way of consumption. However, THC is proposed to be the analgesic component in cannabinoid substances (Good et al. 2019; Hardy et al. 2020; MacDonald and Farrah 2019) (Whiting et al. 2015), rather than the CBD-component. Strikingly, of the patients who used the cannabinoids to treat pain, only 50% used a THC compound.

Another great concern is the observation of the assumed anti-cancer effect of cannabinoids as driving motivation for use. With 46%, the percentage of patients in our study reporting this potential therapeutic effect as a reason for consumption, is higher than found in some previous studies, where values around 25–30% have been reported (Mousa et al. 2019; Pergam et al. 2017). Not only current or previous users adhere to this therapeutic believe of cannabinoids, but also a remarkable 20% of non-users reported considering using cannabinoids because of its proposed anti-cancerous effect. To date no results of large clinical trials have been published supporting the use of cannabinoids for anti-cancer purposes. Rather, potential evidence regarding effectivity has been limited to in vitro and in vivo studies (Abrams and Guzman 2015; Bouquié et al. 2018; Daris et al. 2019; Turgeman and Bar-Sela 2017). The role of social media seems to be a compelling factor in creating this believe, with news stories claiming cannabis as an alternative treatment to cure cancer being widely spread (Shi et al. 2019).

The persisting profound believe in the assumed anti-cancerous effect in absence of clinical evidence, could be related to presumed unawareness by the clinicians regarding consumption of cannabinoids by their patients. This is in line with the low rates of reported usage in the patient files found in our study, underlining the lack of active involvement of the clinician towards this topic. This suggests that even in a country where cannabis consumption is relatively tolerated, cannabinoid consumption as alternative or adjunctive treatment is not yet a topic which is easily addressed by either the clinician or the patient, which is in agreement with previous research (Braun et al. 2020; Kleckner et al. 2019; Pergam et al. 2017). Absence of clinical guidance, however, leaves the patients relying on non-medical sources of information, with presumed lower degrees of clinical evidence (Zolotov et al. 2021), thereby potentially posing patients at medical risks. This does not only concern the anti-cancer believe, but also the potential risks related to concurrent immunotherapy. Furthermore, self-medicating might lead to consumption of products for purposes which are not actually served by the substance. Additionally, it increases the risk of overdosing products, leads to a higher risk of dependency (Fitzcharles and Eisenberg 2017; Hazekamp and Pappas 2014) and increases the risk of using polluted products. The startling amount of users among cancer patients in absence of clinical guidance, combined with patients’ presumed misperception regarding its purposes and utility, as shown in the current study, stresses the need for a change. It necessitates a diligent role of the clinician and addresses the urgency of adequate patient education on both the potential therapeutic benefits, as well as the eventual risks, of cannabinoid consumption in cancer (Donovan et al. 2019). This is especially important in light of the high number of users receiving immunotherapy, as revealed in the current study.

Considering the observational cross-sectional design, this study has its limitations. Even though different types of tumors are represented in the study, the study might not completely reflect the diversity of tumors as presented in the general population, as a result of using a convenience sample, thereby increasing the risk of selection bias. Furthermore, even though the response rate might have been increased through the support of the researcher while filling in the questionnaire, it could have negatively affected the report of usage due to social desirability, thereby underreporting the actual prevalence of cannabinoid usage. Finally, statistical analysis was applied on a small sample size. Even though the study should be interpreted in light of these intrinsic limitations of the study, this study greatly emphasizes the uncertainties and difficulties regarding the current issue of cannabinoid usage among cancer patients.