Survival rate of patients with combined hepatocellular cholangiocarcinoma receiving medical cannabis treatment: A retrospective, cohort comparative study

Introduction

Combined hepatocellular-cholangiocarcinoma (cHCC-CCA) is an uncommon and aggressive form of primary liver cancer that exhibits both hepatocytic and cholangiocytic differentiation within the same tumor.¹ The global incidence rate is approximately 0.59 per 1,000,000 people,² whereas Thailand reports significantly higher rates.³ Notably, Northeastern Thailand—particularly Khon Kaen Province—has the highest reported incidence of CCA worldwide at 118.5 per 100,000 population, exceeding the global rate by over 100-fold.³ Due to its asymptomatic nature in early stages, CCA is typically diagnosed at an advanced stage when metastasis has already occurred, resulting in limited therapeutic options, aggressive disease progression,⁴ and poor prognosis.⁵

Previous studies have shown the median post-diagnosis survival of CCA patients to be about 9 months (95% CI 7–11), with 1-, 3-, and 5-year survival rates at 43.4%, 21.5%, and 17.1%, respectively.⁶ Mean overall survival rate at 1-, 3-, and 5-year was 66.6, 41.5, and 32.7% for patients with transitional HCC-CC,⁷ with median survival time from diagnosis 4.3 months (95% CI: 3.3–5.1),⁸ and after supportive treatment was 4 months.⁹ Survival time was increased among CCA patients receiving surgery, an average of 29.38 months, best supportive treatment, 5.12 months, and 13.38 months for chemotherapy patients.¹⁰

Cannabis-based medicinal products are now legally available in several countries¹¹ and are commonly used in palliative care to alleviate pain, reduce nausea, stimulate appetite, and improve quality of life in adults with cancer.¹² Observational evidence also suggests that supervised medical cannabis is generally well-tolerated and associated with quality-of-life improvements over about six months of use.^13,14 Thailand legalized medical cannabis in February 2019, the first country in Southeast Asia to do so^15,16 and has since integrated cannabis-based options into palliative care services as an adjunct or alternative to standard care.¹⁷ In a recent retrospective cohort of advanced cholangiocarcinoma treated in northeastern Thailand, patients receiving cannabis-based treatment had a median overall survival of 5.66 months and an approximate one-year (12-month) survival of 29.98%, based on Kaplan–Meier estimates,¹⁸ highlighting a potential role alongside symptom management.

Survival data for cannabis-treated cholangiocarcinoma (CCA) remain limited, but some evidence suggests potential benefits. A U.S. inpatient study found that cannabis users with CCA had significantly lower in-hospital mortality than matched non-users (OR = 0.40; 95% CI: 0.16–0.97; p < 0.04).¹⁹ In Thailand, a prospective cohort reported improved functional status and quality of life at two and four months among CCA patients receiving cannabis-based treatment compared to standard care.¹³ Preclinical studies also indicate anticancer effects: cannabidiol (CBD) and cannabigerol (CBG) inhibited CCA cell proliferation and migration, inducing apoptosis and autophagy,²⁰ while delta-9-tetrahydrocannabinol (THC) suppressed proliferation and promoted apoptosis via MAPK/MEK and Akt pathway inhibition.²¹ These findings, though preliminary, highlight the need for further clinical research in advanced CCA, where current survival outcomes remain poor.²²

This study aims to analyze survival outcomes (SA) and identify factors associated with survival rates among patients with combined hepatocellular–cholangiocarcinoma (CHCC/CCA) diagnosed by their physicians as requiring palliative care, who then choose either cannabis treatment (CT) or standard treatment (ST; conventional medical care according to national clinical practice guidelines) after receiving information on available options. Retrospective cohort data were obtained from hospital electronic medical records (EMR) and cancer clinic databases across four tertiary and two secondary hospitals in five Northeastern provinces of Thailand. The results could provide preliminary evidence to support policy discussions and the development of evidence-based palliative care strategies under the national medicinal cannabis framework.

Methods

Results

Survival outcomes

Survival outcomes are presented in Table 2 and illustrated in Figure 1. For ST patients, the total follow-up time was 790 person-months, with a mortality rate of 48.35 per 100 person-months. For CT patients, the total follow-up time was 476 person-months, with a mortality rate of 10.9 per 100 person-months.

Table 3. Multivariable Cox proportional hazards regression analysis for overall survival among patients with advanced cholangiocarcinoma (n = 491).

Variable	Adjusted HR	95% CI	p-value
Treatment group
Standard treatment (ST)	Reference	—	—
Cannabis treatment (CT)	0.28	0.20–0.37	<0.001
Age group
<60 years	Reference	—	—
60–69 years	0.85	0.66–1.09	0.212
≥70 years	0.87	0.68–1.09	0.236
Sex
Male	Reference	—	—
Female	0.89	0.73–1.08	0.236
Type of prior cancer treatment
Surgery	1.00	—	0.106
Chemotherapy	1.43	0.93–2.20	0.093
Combined treatment	1.27	0.82–1.93	0.272
Palliative care only	1.23	0.79–1.92	0.355
Time from diagnosis to registration
<3 months	Reference	—	—
3–6 months	1.31	1.01–1.71	0.044
6–9 months	1.21	0.89–1.65	0.222
>9 months	1.16	0.82–1.63	0.398

Figure 1. Kaplan–Meier survival curves for advanced cholangiocarcinoma (CCA) and hepatocellular carcinoma (HCC) patients receiving standard palliative care treatment (ST) and medicinal cannabis treatment (CT).

The median survival time was calculated from the date of registration at the palliative care or cannabis clinic to the date of death or the study endpoint (30 June 2021). The median survival time was 0.83 months (95% CI: 0.71–0.95) for the ST group and 5.66 months (95% CI: 1.94–9.38) for the CT group. Kaplan–Meier survival curves ( Figure 1) demonstrated significantly longer survival for the CT group compared with the ST group (log-rank test, p < 0.001).

Discussion and Conclusion

In the present study, we examined the survival outcomes of patients with advanced cholangiocarcinoma (CCA) or hepatocellular carcinoma (HCC) receiving either standard palliative care treatment (ST) or medicinal cannabis treatment (CT) across tertiary and secondary hospitals in five provinces. Survival time was calculated from the date of registration at the palliative or cannabis clinic until death or censoring. The CT group demonstrated a markedly longer median survival time (5.66 months) compared with the ST group (0.83 months).

These findings are generally consistent with previous research in Northeastern Thailand, where patients receiving only supportive treatment had a median survival of 4.3 months after diagnosis.¹⁸ Moreover, patients diagnosed at an advanced stage were almost twice as likely to die (HR: 1.8; 95% CI: 1.1–2.9).^23–25 However, our results contrast with Bar-Sela et al.,¹⁴ who reported shorter overall survival among advanced cancer patients using cannabis compared with non-users. This discrepancy may reflect differences in baseline characteristics, timing of cannabis initiation, disease stage, and access to other systemic treatments.

In the adjusted Cox regression model ( Table 2), receiving medicinal cannabis treatment was significantly associated with prolonged survival (adjusted HR = 0.28; 95% CI: 0.20–0.37; p < 0.001). However, the set of adjustment variables was limited to age, sex, type of cancer treatment, and time from diagnosis to registration due to the constraints of using secondary data.²⁶ Important clinical parameters such as performance status, tumor burden, comorbidities, and detailed treatment history could not be included, and residual confounding is therefore possible.²⁷

Baseline differences in care pathways may partly explain the observed survival advantage in the CT group. Many ST patients were referred to palliative clinics late in their disease trajectory, often after exhausting surgery, chemotherapy, or combination regimens.²⁸ In contrast, CT patients—often older and treatment-naïve—were frequently registered directly at cannabis clinics in community hospitals following imaging or biopsy confirmation of advanced metastases.²⁹ Some of these patients also received chemotherapy concurrently with cannabis, which may have contributed to extended survival.³⁰

From a biological perspective, cannabinoids may improve survival indirectly by alleviating symptoms (e.g., pain, anorexia, nausea), enhancing nutritional intake, improving sleep, and enabling greater tolerance to systemic therapy.^11,13,31 Preclinical studies also suggest potential anti-tumor effects via apoptosis induction, inhibition of angiogenesis, and suppression of tumor proliferation,²⁸ although these effects remain to be validated in large-scale clinical trials. Taken together, the observed survival benefit in the CT group may be partly attributable to both patient- and treatment-related covariates, as well as the potential symptom-modulating and anti-tumor mechanisms of cannabinoids demonstrated in preclinical studies.^11,13,15,18 This integrated interpretation underscores the multifactorial nature of survival outcomes in advanced CCA/HCC.

To our knowledge, this is the first multicenter Thai study to compare survival outcomes between ST and CT in advanced CCA/HCC patients treated with standardized, FDA-approved medicinal cannabis products under physician supervision.^17–33 Our findings support the integration of medicinal cannabis into palliative care,³³ especially in community settings with limited oncology resources.

Strengths of this study include the relatively large sample size, multi-level hospital participation, and the use of survival analysis adjusted for available covariates. Limitations include the retrospective design, reliance on secondary data,²⁵ incomplete information on cannabis dosage/formulation/adherence, and the inability to adjust for important prognostic factors. Consequently, while the association between cannabis treatment and improved survival is compelling, causality cannot be established without prospective randomized controlled trials.²⁶

Author contributions

N.P. contributed to the research design, data collection, and manuscript writing. P.P., A.W., contributed to data collection and revised the manuscript, and R. W contributed to research administrator, research design, data collection, review and revised the manuscript.

Statements