Matched adjusted indirect comparisons and artificial intelligence: complementary or competitors? About a case

We proceeded to compare the efficacy and safety of two prophylactic treatments for severe hemophilia A in patients ≥ 12 years, efanesoctocog alfa and emicizumab, using both traditional adjusted indirect comparison methods (MAIC, matched adjusted indirect comparison) and the artificial intelligence tool ChatGPT4o, highlighting the importance of these studies due to the lack of direct comparisons between treatments in patients with hemophilia, a rare disease. Data from Phase III clinical trials are used: the XTEND-1 study for the efanesoctocog alfa and the HAVEN-3 study for emicizumab. The MAIC analysis concluded that efanesoctocog alfa was more effective than emicizumab in reducing bleeding episodes, with an improvement in joint health score. ChatGPT4o also highlighted the greater efficacy of efanesoctocog alfa, but noted the influence of sample size on the statistical significance and robustness of the results. In terms of cost, we proceeded to estimate what the unit cost of efanesoctocog alfa should be as or more efficient than emicizumab. Both treatments are shown to be safe and effective. In conclusion, MAIC and ChatGPT4o may be complementary assessment strategies and may help in the choice of treatment and can help in the choice of treatment. However, the therapeutic selection process is more complex and must consider additional factors, such as the additional factors, such as treatment convenience, patient preferences and clinical judgment.