Background The use of social media platforms for sharing health-related information is on the rise. Sickle cell disease (SCD) affects millions of people worldwide. However, discussions by SCD stakeholders on social media remain unexplored. This study aimed to gain understanding on patient and other stakeholders’ awareness about SCD and their perception of the patient journey by social media listening. Methods Data was retrospectively collected from April 2019 to April 2021 on SCD specific terms in 14 European countries from blogs, forums, and social networking sites (Twitter, public Facebook, YouTube, and Instagram). Results Of 317.9K conversations on SCD (93% Twitter), 945 posts on relevant patient-centric conversation were analyzed. Most patients were females (73%) and ≤ 30 years old (75%). Patient journey stages were addressed in 52% of conversations. Patient journey conversations were mainly regarding symptoms (56%) (mainly pain episodes, pain in general, and vaso-occlusive crises) and treatment (44%). Conversations on hospital visits or hospitalization mostly revolved around crisis faced due to symptoms. Impact on quality of life, especially emotional impact (56%), was also extensively discussed. Unmet needs were derived from 24% of the conversations, lack of awareness of SCD (42%) and lack of empathy and support from HCPs (24%) being the most frequent topics. Patients reported having their symptoms questioned or dismissed by healthcare professionals, which they attributed to racial bias. Conclusions Social media listening is a useful tool to explore the real experiences, concerns and needs of SCD patients and other stakeholders, providing valuable insights for improving the management of these patients.

e18774 Background: Malignant melanoma (MM) is one of the most aggressive skin cancers and its incidence has been increasing worldwide. Deep understanding of patient characteristics and the course of the disease, specially through the evaluation of real-world evidence, is extremely relevant for an adequate treatment approach and better outcomes. This study aims to comprehensively evaluate demographic and clinical characteristics and also treatment outcomes of patients with stage III and IV MM, treated at a Portuguese institution. Methods: Retrospective cohort study of patients with de novo MM stage III/IV or that evolved from earlier MM stages, between 2015 and 2017 (considered the index date). Patients were followed until 12/31/2019. Demographic, clinical and treatment characteristics were evaluated. Survival was assessed, from the index date, using the Kaplan Meier method and log-rank test to compare groups. Results: We included 215 patients with a median age of 66 years (20-96) and 50.2% (n = 108) were male. At index date, 63.7% (n = 137) were stage III. From those, 41.6% (n = 57) progressed to stage IV during follow-up. At diagnosis, the majority of patients had ulceration (53.3%; n = 119), normal LDH ( < 248 U/L; 56.3%; n = 121) and from 110 patients tested for BRAF, 45.4% (n = 50) had a mutation. In earlier stages, 41.8% (n = 81) performed sentinel LN only and from those 61.7% (n = 50) had latter metastatic disease. Complete LND was performed in 49% (n = 95) and 58.9% (n = 56) had a distant relapse. Brain metastasis were diagnosed in 28.4% (n = 61) of the patients, and 50.8% (n = 31) were not eligible for any treatment due to poor clinical status. Systemic treatment was performed in 70 patients with advanced disease. In 1st line, 34 (48.6%) patients underwent anti-PD-1, 28 (40.0%) BRAF/MEKi, 5 (7.1%) BRAFi and 3 (4.3%) chemotherapy. A 2nd line treatment was performed in 21 (30.0%) patients and 2 (9.5%) underwent 3rd line treatment. With a median follow-up of 29 months OS for all patients at 24 months was 54.9% (95% CI; 48.6-62.0): 69.3% (95% CI; 62.0-77.5) for stage III patients and 29.5% (95% CI; 20.9-41.6) for stage IV patients. OS was worst for known risk factors (ulceration, mitotic rate and LDH). OS at 24 months for patients under systemic treatment was 37.4% (95% CI; 26.9-52.0), with no differences between immunotherapy and targeted therapy. Finally, 22 patients were submitted to limb perfusion with an OS of 58.1% (95% CI; 41.2-81.9) at 24 months and a median PFS of 7.4 months (95% CI; 3.9-11.3). Conclusions: Analysis of real-world data is a solid tool in the evaluation, development and improvement of treatment strategies. Demographic and clinical characteristics are comparable to those of other studied cohorts. Longer follow-up of this population and the inclusion of new patients submitted to contemporary approaches will allow improving knowledge and care for melanoma patients in Portugal.