Background: The Quebec Cannabis Registry (QCR) was launched in 2015 to allow physicians to prescribe medical cannabis (MC) in the province of Quebec, Canada. This study aimed to investigate the safety and effectiveness of MC in cancer patients using pharmacovigilance data prospectively collected for up to 24 months. Methods: Patients were enrolled in the QCR between May 2015 and October 2018 and followed every 3 months. Study outcomes included adverse events (AE), pain severity and interference (Brief-Pain Inventory), wellbeing (Revised-Edmonton Symptom Assessment Scale) and overall health scale (EQ5D5L) at baseline and at each follow-up (F-UP). Significance of changes over time were assessed using repeated-measures ANOVA. Results: Out of the 2991 patients enrolled in the QCR, 358 (12.8%) were cancer patients (mean age 57.7 (± 14.6); 171 (47.8%) males). The main cancer types were breast (16.2%), lung (11.7%), leukemia (11.5%) and colorectal (11.2%). MC was prescribed primarily for pain (72.1%), anxiety (4.7%), nausea (4.5%), anorexia (3.9%), and insomnia (3.1%). A total of 13 patients (3.6%) reported AE with only three being serious (one unrelated to MC: stroke; and two possibly related: diarrhea, from CBD oil overdose and pneumonia from smoking MC). Mean scores significantly (p < 0.05) improved between baseline and 3 months F-UP for pain severity (4.8 ± 1.5 vs 4.1 ± 1.8), pain interference (4.6 ± 1.8 vs 3.8 ± 1.7), and the overall health scale (60 ± 21 vs 71 ± 18). Well-being scores also significantly improved between baseline and 6 months F-UP (4.4 ± 2.1 vs 3.5 ± 2.8). Conclusions: Population-based data shows that cancer patients can benefit safely and effectively from MC as a complementary treatment, when prescribed and monitored under medical-nursing supervision.