Informed consent constitutes a fundamental ethical principle in medical practice. With the in-depth integration of generative artificial intelligence （AI） represented by Chat Generative Pre-trained Transformer （ChatGPT） with medicine， it has brought revolutionary development to traditional informed consent while also introducing new ethical challenges. ChatGPT offers features such as improving the readability of informed consent content， enhancing its comprehensiveness and accuracy， and increasing the convenience of obtaining informed consent. However， as the application of ChatGPT in informed consent is still in the exploratory stage， it is imperative to proactively and fully consider the accompanying ethical issues， such as information security， liability determination， transparency， and fairness. This paper conducted an ethical analysis on the challenges faced by generative AI， represented by ChatGPT， in the application of informed consent and proposed countermeasures， such as upholding free and fully informed consent， strengthening the balance of rights and obligations in informed consent， and establishing a transparent and fair supervision mechanism. The aim was to promote the ethically compliant， orderly， and controllable development of generative AI in the field of medical informed consent.

BACKGROUNDTo evaluate three new chemotherapeutic regimens for non-small cell lung cancer (NSCLC) by pharmacoeconomic analysis in guiding rational use of drugs.

METHODSOne hundred and one cases of NSCLC in clinic stage III or IV were treated by one of the three chemotherapeutic schemes-PC: paclitaxel (135mg/m²,d1)+DDP; TC: docetetaxel (75mg/m²,d1)+DDP; VC: vinorelbine (25mg/m²,d1 and d8)+DDP, DDP were given at 80mg/m² in 3 groups. Pharmacoeconomic cost-effectiveness analysis was used to compare the efficacy of the three regimens.

RESULTSThe response rate was 46.9%, 48.6% and 47.1% and median survival duration was 7.8, 7.5 and 7.6 months for PC, TC and VC regimen respectively, with 1-year survival rate of 37.5%, 37.1% and 38.2% respectively. There was remarkable difference in the response rate and median survival duration between PC and TC, but no statistical difference was observed between PC and VC. There was no statistical difference in 1-year survival rate among the three regimens. The average cost of one patient for one therapeutic cycle was RMB 15840.5, 15831.1 and 9401.8 Yuan respectively. Escalation of 1% of response rate costed RMB 337.75, 325.74 and 199.61 Yuan respectively. Prolongation of 1 month of median survival duration costed RMB 2030.83, 2110.97 and 1237.08 Yuan respectively. Escalation of 1% of one year survival rate costed RMB 422.41 , 426.71 and 246.12 Yuan respectively.

CONCLUSIONSAmong these three new chemotherapeutic regimens for the advanced patients with NSCLC, the expenditure of VC is much cheaper than PC and TC. The cost effectiveness of VC is the lowest among the three regimens.