Background. Acquired hemophilia A (AHA) is a rare autoimmune bleeding disorder caused by autoantibodies against factor VIII (FVIII), predominantly affecting elderly patients and associated with substantial bleeding risk and mortality. The inhibitor titer, measured by the Nijmegen-modified Bethesda assay and expressed in Bethesda Units per milliliter (BU/mL), is a key diagnostic parameter. The 5 BU/mL cut-off, derived from congenital hemophilia, is conventionally used to distinguish low- from high-titer inhibitors, but it has never been validated as a predictor of bleeding risk in AHA: owing to the non-linear inactivation kinetics of AHA autoantibodies, neither the titer nor residual FVIII activity consistently reflects bleeding severity. Objective. To assess the prognostic value of the inhibitor titer at diagnosis by comparing patients stratified according to the 5 BU/mL cut-off with respect to clinical presentation, treatment, and outcomes. Methods. Observational, retrospective, single-center cohort study of 39 patients with AHA followed at Padua University Hospital between February 2019 and July 2026, divided into a low-titer group (<5 BU/mL, n=16) and a high-titer group (≥5 BU/mL, n=23). Variables were compared using the Mann-Whitney and Fisher's exact tests. Time to first bleeding event was analyzed in 35 patients using the Kaplan-Meier method and a Cox model, unadjusted and adjusted for FVIII activity at diagnosis, with incidence rate calculation and a 90-day sensitivity analysis. Results. The groups were comparable for age, sex, bleeding presentation, and etiology. FVIII activity was higher in the low-titer group (median 7.9% vs 3.2%; p=0.010). High-titer patients received more intensive treatment, with greater use of susoctocog alfa (34.8% vs 6.3%; p=0.056) and transfusion support (82.6% vs 31.3%; p=0.002). Median bleeding-free time was 35 days in the low-titer group and was not reached in the high-titer group (log-rank p=0.009). In the Cox model, high titer was associated with a lower bleeding risk (HR 0.26; p=0.016), also after adjustment for FVIII activity (HR 0.24; p=0.016); the finding was confirmed by incidence rates (0.81 vs 0.16 events per 100 patient-days; IRR 0.20; p=0.003) and by the 90-day analysis. Complete remission (87.5% vs 73.9%; p=0.432) and median time to remission (63 vs 72 days; p=0.825) did not differ. Relapses were more frequent in the low-titer group (64.3% vs 29.4%; p=0.076) and infections in the high-titer group (56.5% vs 25.0%; p=0.051), where five deaths occurred before inhibitor clearance versus none in the low-titer group (p=0.066). Conclusions. A low inhibitor titer at diagnosis was associated with a higher incidence of subsequent bleeding. This may reflect differences in clinical phenotype and treatment intensity; the observational design does not allow causal inference. Inhibitor titer should not be used as the sole determinant of bleeding-risk assessment, which should also incorporate clinical presentation and bleeding evolution. Prospective multicenter studies are needed to confirm these findings.
Introduzione. L'emofilia A acquisita (AHA) è una rara malattia emorragica autoimmune, causata da autoanticorpi diretti contro FVIII, che colpisce prevalentemente soggetti anziani ed è gravata da un rischio significativo di sanguinamento e di mortalità. Il titolo dell'inibitore, misurato con il metodo Bethesda-Nijmegen ed espresso in Unità Bethesda per millilitro (UB/mL), è un parametro diagnostico centrale. Il cut-off di 5 UB/mL, derivato dall'emofilia congenita, distingue per convenzione le forme a basso e ad alto titolo, ma nell'AHA non è mai stato validato come predittore del rischio emorragico: la cinetica di inattivazione non lineare degli autoanticorpi fa sì che né il titolo né l'attività residua del FVIII riflettano in modo affidabile la gravità del fenotipo emorragico. Obiettivi. Valutare il valore prognostico del titolo dell'inibitore alla diagnosi, confrontando i pazienti stratificati secondo il cut-off di 5 UB/mL per presentazione clinica, approccio terapeutico ed esiti. Materiali e metodi. Studio osservazionale retrospettivo monocentrico su 39 pazienti con AHA seguiti presso l'Azienda Ospedale-Università di Padova tra febbraio 2019 e luglio 2026, suddivisi in basso titolo (<5 UB/mL, n=16) e alto titolo (≥5 UB/mL, n=23). Le variabili sono state confrontate con i test di Mann-Whitney e di Fisher. Il tempo al primo evento emorragico è stato analizzato in 35 pazienti con il metodo di Kaplan-Meier e un modello di Cox, non aggiustato e aggiustato per l'attività del FVIII, con calcolo dei tassi di incidenza e analisi di sensibilità a 90 giorni. Risultati. I gruppi erano sovrapponibili per età, sesso, presentazione emorragica ed eziologia. L'attività del FVIII era più elevata nel basso titolo (mediana 7,9% vs 3,2%; p=0,010). I pazienti ad alto titolo hanno ricevuto trattamenti più intensivi, con maggior ricorso a susoctocog alfa (34,8% vs 6,3%; p=0,056) e a supporto trasfusionale (82,6% vs 31,3%; p=0,002). Il tempo mediano libero da sanguinamento è stato di 35 giorni nel basso titolo, non raggiunto nell'alto titolo (log-rank p=0,009). Nel modello di Cox l'alto titolo si è associato a un minor rischio di sanguinamento (HR 0,26; p=0,016), anche dopo aggiustamento per il FVIII (HR 0,24; p=0,016); il dato è confermato dai tassi di incidenza (0,81 vs 0,16 eventi per 100 giorni-paziente; IRR 0,20; p=0,003) e dall'analisi a 90 giorni. Remissione completa (87,5% vs 73,9%; p=0,432) e tempo per raggiungerla (63 vs 72 giorni; p=0,825) non differivano. Le recidive sono state più frequenti nel basso titolo (64,3% vs 29,4%; p=0,076), le infezioni nell'alto titolo (56,5% vs 25,0%; p=0,051), dove si sono verificati cinque decessi prima della negativizzazione dell'inibitore contro nessuno nel basso titolo (p=0,066). Conclusione. Un basso titolo dell'inibitore alla diagnosi si è associato a una maggiore incidenza di sanguinamenti successivi. L'associazione potrebbe riflettere differenze nel fenotipo clinico e nell'intensità dei trattamenti; la natura osservazionale dello studio non consente di stabilire un rapporto causale. Il titolo non dovrebbe essere usato come unico parametro di valutazione del rischio emorragico, che deve integrare la presentazione clinica e l'evoluzione del sanguinamento. Sono necessari studi prospettici multicentrici per confermare questi dati.
Titolo dell’inibitore alla diagnosi e decorso clinico nell’emofilia A acquisita: uno studio retrospettivo di coorte.
RIGHETTO, FRANCESCO
2025/2026
Abstract
Background. Acquired hemophilia A (AHA) is a rare autoimmune bleeding disorder caused by autoantibodies against factor VIII (FVIII), predominantly affecting elderly patients and associated with substantial bleeding risk and mortality. The inhibitor titer, measured by the Nijmegen-modified Bethesda assay and expressed in Bethesda Units per milliliter (BU/mL), is a key diagnostic parameter. The 5 BU/mL cut-off, derived from congenital hemophilia, is conventionally used to distinguish low- from high-titer inhibitors, but it has never been validated as a predictor of bleeding risk in AHA: owing to the non-linear inactivation kinetics of AHA autoantibodies, neither the titer nor residual FVIII activity consistently reflects bleeding severity. Objective. To assess the prognostic value of the inhibitor titer at diagnosis by comparing patients stratified according to the 5 BU/mL cut-off with respect to clinical presentation, treatment, and outcomes. Methods. Observational, retrospective, single-center cohort study of 39 patients with AHA followed at Padua University Hospital between February 2019 and July 2026, divided into a low-titer group (<5 BU/mL, n=16) and a high-titer group (≥5 BU/mL, n=23). Variables were compared using the Mann-Whitney and Fisher's exact tests. Time to first bleeding event was analyzed in 35 patients using the Kaplan-Meier method and a Cox model, unadjusted and adjusted for FVIII activity at diagnosis, with incidence rate calculation and a 90-day sensitivity analysis. Results. The groups were comparable for age, sex, bleeding presentation, and etiology. FVIII activity was higher in the low-titer group (median 7.9% vs 3.2%; p=0.010). High-titer patients received more intensive treatment, with greater use of susoctocog alfa (34.8% vs 6.3%; p=0.056) and transfusion support (82.6% vs 31.3%; p=0.002). Median bleeding-free time was 35 days in the low-titer group and was not reached in the high-titer group (log-rank p=0.009). In the Cox model, high titer was associated with a lower bleeding risk (HR 0.26; p=0.016), also after adjustment for FVIII activity (HR 0.24; p=0.016); the finding was confirmed by incidence rates (0.81 vs 0.16 events per 100 patient-days; IRR 0.20; p=0.003) and by the 90-day analysis. Complete remission (87.5% vs 73.9%; p=0.432) and median time to remission (63 vs 72 days; p=0.825) did not differ. Relapses were more frequent in the low-titer group (64.3% vs 29.4%; p=0.076) and infections in the high-titer group (56.5% vs 25.0%; p=0.051), where five deaths occurred before inhibitor clearance versus none in the low-titer group (p=0.066). Conclusions. A low inhibitor titer at diagnosis was associated with a higher incidence of subsequent bleeding. This may reflect differences in clinical phenotype and treatment intensity; the observational design does not allow causal inference. Inhibitor titer should not be used as the sole determinant of bleeding-risk assessment, which should also incorporate clinical presentation and bleeding evolution. Prospective multicenter studies are needed to confirm these findings.| File | Dimensione | Formato | |
|---|---|---|---|
|
Righetto_Francesco.pdf
Accesso riservato
Dimensione
1.11 MB
Formato
Adobe PDF
|
1.11 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/116078