Artificial intelligence (AI) comparison of social media-based patient-reported outcomes of PD-1, BRAF, and CTLA-4 inhibitors for melanoma treatment.

Abstract

e21572 Background: Due to side effects and adverse psychosocial factors, there can often be a disconnect between clinical impression and the patient perspective of treatment. Melanoma patients frequently use social media to discuss their disease sentiments and outcomes, providing a window into patient insight, otherwise not readily attainable. We analyzed publicly available full text posts on multiple social media platforms concerning melanoma treatments PD-1 inhibitors (pembrolizumab/Keytruda, nivolumab), BRAF inhibitors (dabrafenib, vemurafenib), and CTLA-4 inhibitor (ipilimumab/Yervoy) associated with Patient Global Impression of Change (PGIC) terms to compare and identify patient burden. Methods: Millions (12,599,313) of publicly available online social media text data were extracted and run through Brandwatch Artificial Intelligence-powered database and machine learning algorithm to categorize treatment-specific posts with PGIC terms associated with sentiment. Results: Out of 52,962 posts related to a select list of melanoma treatments, we identified Keytruda (6,080), nivolumab (1,614), dabrafenib (529), vemurafenib (329), and ipilimumab (1715) posts. The top ten types of posts by volume for each treatment were predominantly positive for PGIC terms (improving, well, progress) while being associated with negative emotions (fear and sadness). Patient-perceived better treatments were associated with decrease fear. PD-1 inhibitor Keytruda displayed markedly less fear (56.9%) with a very high positive PGIC (92.3%) in comparison with PD-1 inhibitor nivolumab with decreased positive PGIC (78.0%) and increased fear (81.6%). Likewise, this patten was discovered for the BRAF inhibitors and CTLA-4 inhibitor, albeit less distinct: dabrafenib (positive PGIC 86.4%, fear 71.9%), vemurafenib (positive PGIC 78.0%, 81.6%), ipilimumab (positive PGIC 75.0%, fear 79.3%). Conclusions: Our initial results indicate that despite patients reporting good outcomes from melanoma treatment, they remain fearful. Nonetheless, a trend is identified between patient-perceived better treatment and decreased patient fear. This analysis can help bridge the gap of understanding patient perspective and provides an indication for translation into more effective clinical and pharmaceutical response.

Artificial Intelligence Sponge City Wisdom Management Model

“…is model is a single-layer multidirectional arrangement, which is divided into three levels. e first is the writing layer, then the implicit layer, and the final is the output layer, but there is a hidden layer, which can be one layer or multiple layers [14].…”