Indirect comparison of two real-life studies in immune thrombocytopenia using artificial intelligence

Indirect comparison of two real-life studies in immune thrombocytopenia using artificial intelligence

Manuel Rodríguez-López 1, Raquel Ocampo-Martínez 2, Ramiro J. Núñez-Vázquez 3

1 Servicio de Hematología y Hemoterapia, Hospital Universitario Álvaro Cunqueiro, Vigo, Pontevedra, España; 2 Servicio de Hematología y Hemoterapia, Hospital Universitario Álvaro Cunqueiro, Instituto de Investigación Sanitaria Galicia Sur, EOXI Vigo, Vigo, España; 3 Unidad de Trombosis y Hemostasia, Servicio de Hematologia y Hemoterapia, Hospital Universitario Virgen del Rocío, Sevilla, España

Manuel Rodríguez-López, Raquel Ocampo-Martínez, Ramiro J. Núñez-Vázquez

La información completa de afiliaciones y autor de correspondencia está disponible en la versión original en PDF.

*Correspondence: Manuel Rodríguez-López, Email not available

Summary

Artificial intelligence (AI) is here to stay in medicine. This paper explores the use of generative AI tools (ChatGPT4o and Perplexity AI) to compare the efficacy and safety of two new drugs for chronic primary immune thrombocytopenia (ITP), fostamatinib and avatrombopag, using data from real-life studies. The comparison aims to aid in treatment selection for adult patients with ITP who have not responded to previous treatments. Both treatments are effective and safe, but avatrombopag shows a higher response rate and lower incidence of common adverse events. The choice of treatment may depend on additional clinical factors, patient preferences and economic considerations. ChatGPT4o suggests an initial economic advantage for fostamatinib, while Perplexity AI favours avatrombopag in the long term in the Spanish setting. It is concluded that AI can be a valuable tool for treatment comparison, but validation and human oversight are essential to ensure the quality of the information generated. The analysis suggests that both therapeutic options are valid for the management of patients with chronic ITP, highlighting the importance of a collaborative approach between AI and haematologists to maximise outcomes.

Keywords:  Inmune thrombocytopenia. Artificial intelligence. Real life.

Contents