Sustained Improvement of Survival in Patients Receiving Emicizumab Prophylaxis Instead of Immunosuppression during Early Management of Acquired Hemophilia a (AHA)

Acquired hemophilia A (AHA) is a serious bleeding disorder characterized by neutralizing autoantibodies against coagulation factor VIII (FVIII), also called inhibitors. Standard of care is immunosuppressive therapy (IST) to eradicate inhibitors. IST achieves remission in 60-80% of patients but is often related to adverse events, in particular infections, which constitute the leading cause of death in AHA patients in Western countries. The GTH-AHA-EMI study, conducted by our group between March 2021 and June 2022, used emicizumab to protect AHA patients from bleeding during the first 12 weeks of management and allowed IST only thereafter. The choice of IST regimen was at the investigator’s discretion. Promisingly low rates of severe infection and mortality were reported at end of study after 24 weeks. The objective of the current analysis was to report long-term outcomes of this treatment strategy and to compare them with patients receiving traditional IST immediately after diagnosis. We collected 2-year individual patient data from the GTH-AHA-EMI study (n=47, emicizumab weeks 1 to 12, delayed IST after week 12, database closure in June 2024) and the historic GTH-AH 01/2010 study (n=101, immediate IST week 1 to 10, no emicizumab). Information was included on the use of IST, adverse events, partial remission (PR, defined as FVIII recovered to >50% and no bleeding), and complete remission (CR, defined as PR plus negative inhibitor test, prednisolone tapered to <15 mg/day, and any other IST stopped). Data were compared using descriptive statistics and propensity score (PS)-matched Kaplan-Meier analysis and Cox proportional hazards regression. The two study cohorts were very similar with regard to age, underlying and concomitant disorders, baseline FVIII activity, inhibitor titer, and physical performance status. Of the 47 GTH-AHA-EMI study patients, 44 were observed beyond week 12, and 19 of them (43%) continued emicizumab prophylaxis after week 12 at the investigator’s discretion for a further median of 21 weeks. Eight GTH-AHA-EMI patients never started IST. Two of them died early after week 12, while all of the remaining 6 patients achieved spontaneous PR (median week 40, range 7-96). 35 (80%) of 44 GTH-AHA-EMI patients started IST, 29 of them between week 12 and week 24, and 6 thereafter. IST included glucocorticoids (n=25), rituximab (n=24), mycophenolate mofetil (n=8), and cyclophosphamide (n=6), alone or in combination. 29 (83%) of 35 patients achieved PR (median week 30, range 18-97), and 28 (80%) achieved CR (median week 38, range 18-97). For comparison, GTH-AH 01/2010 patients receiving immediate IST after diagnosis achieved PR in 84% (median week 4, range 1-52) and CR in 61% (median 10 weeks, range 4-84). 11 (31%) out of the 35 GTH-AHA-EMI patients receiving delayed IST had a total of 12 infections recorded median 7 weeks (range 1-12) after IST start. They were of grade 1 (7 events), grade 2 (3 events), or grade 3 (2 events), but never life-threatening or fatal. 36 (36%) of 101 GTH-AH 01/2010 patients had a total of 55 infections, reported as fatal in 17 patients. PS-matched overall survival rates were higher for GTH-AHA-EMI patients (emicizumab and delayed IST; year 1: 86%, year 2: 82%) as compared to GTH-AH 01/2010 patients (immediate IST; year 1: 69%, year 2: 63%). The hazard ratio for overall survival was 0.39 (95% confidence interval 0.19 to 0.80, p=0.011) in favor of emicizumab and delayed IST. In conclusion, emicizumab prophylaxis allowed to postpone IST during the early management of AHA. The 2-year overall survival rate improved by an absolute 19% comparing PS-matched individual patient data of emicizumab and delayed IST (GTH-AHA-EMI, 82%) vs. immediate IST (GTH-AH 01/2010, 63%). Spontaneous remission was observed in 6 patients never receiving IST. 80% of GTH-AHA-EMI patients receiving delayed IST achieved CR, although time to remission was longer compared to historic controls.