Background: Clinical trial evidence showed that anti-PD1 ± anti-CTLA4 as well as BRAF ± MEK-inhibitors for BRAF-mutated tumours dramatically improved outcomes for patients with advanced melanoma. Large prospective data sets provide real-world insight into the management of patients with advanced melanoma in routine practice. Methods: Patients ≥ 18 years with unresectable or metastatic melanoma receiving therapy with first- (1L) or second-line (2L) anti-PD1 alone (PD1), with anti-CTLA4 (C-IO) or combination BRAF- and MEK-inhibitors (C-TT) were enrolled in a multi-centre prospective observational study across Canada. Data was collected from May 1, 2016 to November 30, 2021 and entered into the Canadian Melanoma Research Network Registry to include demographics and clinical details. Each patient was followed until death, up to 3 years, or date of data extraction, whichever occurred first. Results: Data for 401 (1L) and 128 (2L) patients was analyzed. There was a significant difference in the age at diagnosis for PD1 (69.2y), C-IO (57.8y), and C-TT (58.8y) in the 1L cohort (p < 0.0001) and 2L cohort (59.5y, 50.8y, and 50.9y respectively; p = 0.036). Patients treated with either C-IO or C-TT had a significantly higher baseline LDH (p = 0.0003) and proportion with brain metastases (p = 0.021) compared to PD1. The probability of survival for PD1, C-IO, and C-TT at 1 year was 0.85, 0.78, and 0.66; and at 3 years was 0.63, 0.54, and 0.36 respectively. The probability of survival for 2L PD1, C-IO, and C-TT at 1 year was 0.81, 0.53, and 0.56; and at 3 years was 0.55, 0.42, and 0.20 respectively. When comparing treatment regimens, the overall survival (OS) in 1L and 2L showed a superior survival for PD1 when compared with C-TT using a pairwise log-rank comparison (p < 0.0001 and p = 0.0009). There was no difference between C-IO and C-TT, or C-IO and PD1 in 1L or 2L. Using a Cox proportional hazard model for OS, the presence of brain metastases was significant only in 1L (HR 1.663, p = 0.01). There was no difference in OS based on age in either 1L or 2L for all treatments. When comparing regimens, the progression-free survival (PFS) in 1L showed no difference using a pairwise log-rank comparison; however, there was a significant improvement in PFS in 2L for PD1 compared with C-TT (p = 0.0014). There was no difference in PFS between C-IO and C-TT, or C-IO and PD1 in 2L. Conclusions: This real-world data suggests that patient selection is key when deciding on the most appropriate treatment in 1L or 2L as PD1 therapy appeared to have superior OS and PFS across comparisons. Patients with high-risk features such as high LDH and the presence of brain metastases received C-IO or C-TT more often than PD1. Age was not associated with OS in 1L or 2L for each treatment.