¿Cómo impacta la inteligencia artificial en la Investigación clínica?: Beneficios, riesgos y desafíos éticos / How does artificial intelligence impact clinical research?: Benefits, Risks and Ethical Challenges

Introducción: La explosión tecnológica de la última década, como la inteligencia artificial (IA) y big data, afectaron positivamente la investigación clínica. Sin embargo, surgen preocupaciones éticas que no se pueden ignorar. Objetivo: Identificar beneficios y riesgos de la IA, destacando desafíos éticos relevantes, para su uso responsable en investigación clínica. Método: Estudio cualitativo y descriptivo, con un enfoque de revisión narrativa de la literatura sobre el tema de interés. Resultado: Beneficios de IA: acelera procesos, optimiza ensayos clínicos, analiza grandes volúmenes de datos, etc. Riegos y desafíos éticos de IA: sesgo algorítmico, heterogeneidad y baja calidad de datos, inseguridad y violación de privacidad, falta de transparencia e interpretabilidad de modelos, etc. Conclusiones: Es esencial contar con políticas para abordar los riesgos y desafíos éticos, garantizando que la IA se utilice de manera equitativa y segura. Estas tecnologías deben actuar como herramientas de apoyo y nunca reemplazar el juicio humano. (AU)

Follow-up of an asymptomatic Chagas disease population of children after treatment with nifurtimox (Lampit) in a sylvatic endemic transmission area of Colombia.

BACKGROUND: Chagas disease is an anthropozoonosis caused by Trypanosoma cruzi. Two drugs are currently used for the etiological treatment of the disease: Nifurtimox (Lampit) and Benznidazole. This study presents a quasi-experimental trial (non-control group) of sixty-two patients who were treated for Chagas disease with Nifurtimox (Lampit), and were then followed for 30 months post-treatment. The safety of Nifurtimox (Lampit) for Chagas disease in this group of children primarily between 4 and 19 years old was also evaluated. MATERIALS AND METHODS: The 62 patients included in the study were selected when resulted seropositive for two out of three fundamentally different serological tests. All children were treated during two months according to protocols established by WHO. Monitoring was performed every twenty days to evaluate treatment safety. In 43 patients, two different serological tests: ELISA and IFAT; and two parasitological tests: blood culture, and real time PCR, (qPCR) were performed to assess therapeutic response, defined as post-treatment serological negativization. PRINCIPAL FINDINGS: All patients completed the treatment successfully, and six patients abandoned the post-treatment follow-up. Adverse effects occurred in 74% of patients, but only 4.8% of cases required temporary suspension to achieve 100% adherence to the 60-day treatment, and all symptoms reverted after treatment completion. Both parasite load (measured through qPCR) and antibodies (ELISA absorbance) evidenced a significant median reduction 6 months after treatment from 6.2 to 0.2 parasite equivalents/mL, and from 0.6 to 0.2 absorbance units respectively (p<0.001). Serological negativization by ELISA was evident since 6 months post-treatment, whereas by IFAT only after 18 months. Serological negativization by the two tests (ELISA and IFAT) was 41.9% (95%CI: 26.5-57.3) after 30 months post-treatment. qPCR was positive in 88.3% of patients pre-treatment and only in 12.1% of patients after 30 months. Survival analysis indicated that only 26.3% (95%CI: 15.5-44.8) persisted with negative qPCR during the whole follow-up period. CONCLUSIONS: Nifurtimox was very well tolerated and successfully reduced parasite load and antibody titers. Re-infection, lysed parasites or a lack of anti-parasitic activity could explain these persistently positive qPCR cases.