Resumen
Despite the emerging success of immunotherapy in non-small-cell lung cancer (NSCLC), it remains clinically important to better identify patients who are likely to respond to treatment, especially considering the existence of immune-related adverse events (irAEs). In recent years, the gut microbiome has been correlated with treatment response, but no predictive models relating the two have been developed. In this study, we used random forest and neural networks to predict the progression-free survival of NSCLC patients treated with immunotherapy. Our results showed that a functional profile of the human gut microbiome outperformed the taxonomical profile across different studies, which can be utilized to establish a model with good predictive value in lung cancer immunotherapy.