Review of Potential Barriers to Effective Hemostatic Management of Acquired Hemophilia A by Non-Hemophilia Experts in the United States

Acquired hemophilia A (AHA) is an ultra-rare autoimmune disorder caused by autoantibodies against factor VIII. It often presents with life-threatening bleeding to non-hemophilia experts, who have limited awareness of this condition. This review evaluated hemostatic management and identified barriers to optimal management of AHA by non-hemophilia experts in the United States through a literature review. AHA case reports published by non-hemophilia experts from January 2016 through November 2021 in non-hematology journals were critically reviewed for a chronology of clinical course and management, consultation with a hemophilia expert, referencing of available AHA recommendations, discussion of all hemostatic options, and bleed control outcomes; 24 case reports representing 24 patients were identified. Twelve patients had an apparent delay in diagnosis, 17 cases did not seek expert consultation, and 15 did not reference the 2009 International AHA Recommendations, including six in whom hemostatic treatment was not consistent with the recommendations. Of the 17 articles published after the 2017 AHA Guidance, eight did not reference them. Of the five articles published after the 2020 International Recommendations for AHA, three did not reference them. Overall, 14 articles did not discuss all available hemostatic treatment options. Four patients died. Our findings reveal variability in hemostatic management of AHA by non-hemophilia experts in the United States. Lack of AHA awareness remains a primary barrier for optimal management of AHA among non-hemophilia experts. Increasing education about existing AHA guidelines, including available therapies and access to expert care at hemophilia treatment centers, may help improve the outcomes of patients with AHA.

The treatment of AHA relies on two main principles: (i) control of acute bleeding diathesis and (ii) eradication of the inhibitor [9]. Acute bleeding diathesis can be treated with hemostatic agents, including factor replacement with recombinant porcine FVIII (rpFVIII), bypassing agents such as recombinant factor VIIa (rFVIIa) or activated prothrombin complex concentrate (aPCC) (off-label in the United States), and other agents such as human FVIII or desmopressin [8]. Eradication of the inhibitor requires immunosuppressive therapies such as steroids, cyclophosphamide, and/or rituximab [10].
To help clinicians manage these patients, several expert recommendations for the diagnosis and treatment of AHA have been published [8,9,11]. However, most patients with AHA typically present to physicians who do not have experience in managing patients with hemophilia (e.g., emergency room physicians, geriatricians, obstetricians, oncologists, rheumatologists, surgeons) [5], and their level of awareness of these recommendations is unknown. Therefore, we sought to investigate potential barriers to and suggest solutions for the optimal management of AHA by non-hemophilia experts in the United States via a literature review.
We critically reviewed case reports of AHA published by non-hemophilia experts in non-hematology journals over the last five years. Specifically, we focused on assessing patient outcomes, with particular emphasis on the initial control of bleeding (e.g., chronology and type of hemostatic agents used, when this information was available). We further investigated whether these treaters consulted a hemophilia expert and referenced and/or followed the published recommendations for AHA.  The search strategy included the terms "acquired hemophilia A" and "treatment," and was restricted to articles in English published from January 2016 through November 2021. This range was chosen to reflect the period during which all currently recommended first-line hemostatic options for AHA treatment were available in the United States [8].
AHA case reports and letters have been included. The exclusion criteria were as follows: (i) patients treated outside the United States, (ii) articles not pertaining to AHA, (3) articles not written in the English language, (iv) articles not describing a clinical case, (v) unviable articles, (vi) articles published in hematology journals (to focus on articles targeting a non-hemophilia-expert audience), and (vii) article with authors who appeared to be hemophilia experts (defined as a hematologist with experience managing patients with hemophilia either at a hemophilia treatment center (HTC) or at a hematology clinic), based on author affiliations, internet research, and review of the author's disease expertise). Examples of non-hemophilia experts include emergency room physicians, obstetricians, gynecologists, and surgeons. Hematology/oncology-affiliated authors were included, as they were not necessarily considered hemophilia experts unless there was evidence of bleeding disorder expertise, as evaluated by manual author review. Only articles published in the United States were included because of the potential differences in healthcare delivery systems, including access to HTCs. Moreover, the availability of all treatment options may not have been the same worldwide relative to the start date of our literature search.
The following data were extracted: patient demographics, clinical presentation, medical history, hemostatic regimens used, and author specialties (based on affiliations and internet research). To assess AHA management awareness, the following areas were evaluated: (i) any apparent delay in diagnosis, which was assessed subjectively based on treatment chronology and interventions (e.g., if the patient received another diagnosis before AHA, if the patient was treated with non-hemostatic treatments until AHA was discovered), (ii) chronology of clinical course and type of treatment intervention/s used to control acute bleeding, (iii) consultation with a hemophilia expert, (iv) referencing of available recommendations and guidance for AHA diagnosis and treatment based on the date of publication, specifically those published by Huth-Kühne et al. in 2009 [11], Kruse-Jarres et al. in 2017 [8], and Tiede et al. in 2020 [9], (v) whether the initial hemostatic treatment was consistent with the available recommendations for hemostatic control, (vi) whether all available acute hemostatic treatment options were discussed in the report, and (vii) control of bleeding diathesis and mortality outcomes.

Review
Of the 250 articles initially identified in our literature search, 198 were immediately excluded based on the criteria described earlier (Figure 1). Thirteen of the 52 reports were published in hematology journals and 15 of the remaining 39 were published by hemophilia experts. A total of 24 articles fulfilled the inclusion criteria (data summarized in Table 1).

FIGURE 2:
Overlap between a lack of hemophilia-expert consultation, discussion of treatment options, and apparent delay in diagnosis in 24 AHA case reports published by non-hemophilia-expert authors.
Four of the 24 patients (17%) in our review died: one because of suspected intracranial hemorrhage (without autopsy confirmation, so a thrombotic event was not excluded; treated with aPCC and rFVIIa) [22]; one because of pulmonary embolism possibly attributed to aPCC treatment (dose unspecified) [15]; and one because of cardiac arrest following a stroke (treated with FVIII, rFVIIa, and rpFVIII; Table 1) [20]. The fourth patient died from progressive shock with renal and hepatic failure following cardiac arrest 23 days after admission; bleeding remained uncontrolled despite treatment with 4-factor prothrombin complex concentrate, recombinant FVIII (rFVIII), and rFVIIa, and a pulmonary embolism eventually developed after two days of rFVIIa therapy (30 µg/kg), prompting a switch to plasma exchange [32]. Based on our assessment, all 24 cases either did not discuss all treatment options, did not consult a hemophilia expert, and/or had a diagnostic delay, and all four patients who died had a delay in diagnosis ( Figure 2). Details of case reports published by non-hematology experts that did not discuss all available first-line hemostatic options are available in the Appendices.

Discussion
Our study shows that there is variability in the hemostatic management of AHA among non-hemophilia experts. Our analyses revealed apparent delays in diagnosis in several cases, consistent with the findings of the European Acquired Haemophilia Registry [6]; this registry captured data from 117 centers in 13 European countries and reported a median delay in AHA diagnosis of three days (interquartile range, 0-12 days) after the patient presented with bleeding diathesis. Although these delays might appear relatively short, they could still contribute to poor clinical outcomes, given that most patients presenting with AHA are elderly, have significant comorbidities, and may have life-threatening bleeds [6]. Notably, in our review, one of the patients who died was undiagnosed for two years and had previously presented with life-threatening acute gastrointestinal bleeding [22].
While the identified variability in hemostatic management may have resulted from a lack of awareness of AHA itself, we also noted that most of these clinicians may not be cognizant of HTCs as an important resource for the management of patients with bleeding disorders. HTCs are specialized multidisciplinary healthcare centers that use a team-based shared decision-making process to improve patient outcomes and offer integrated and comprehensive diagnostic and treatment services (including counseling, case management, care coordination, outreach, research, surveillance, and outpatient pharmacy services) to patients with hemophilia and other bleeding disorders [36].
Based on our assessment of potential barriers to the optimal care of people with AHA by non-hemophilia experts, we outlined potential solutions or resources to help physicians in these settings ( Table 2).

Potential barrier
Potential solution or resources to overcome barrier

Diagnosis
Lack of familiarity with AHA symptoms and differential diagnosis; isolated aPTT with normal prothrombin time should prompt differential diagnosis. A hemophilia expert might recommend a mixing assay and an assay to confirm the presence of inhibitors (Bethesda assay, Nijmegen-Bethesda assay, or an anti-FVIII antibody ELISA) Misconception that anti-pFVIII inhibitor assays are required before initiating rpFVIII treatment The rpFVIII prescribing information states to "consider evaluating for presence of anti-rpFVIII antibodies prior to initiation of treatment" [37]. The activity of rpFVIII can be assessed by standard FVIII assays and by observing clinical response [10,38]. An algorithmic approach to dosing rpFVIII developed by Ellsworth et al., can be used which measures FVIII recovery alone to predict rpFVIII treatment efficacy early on, guide dosing and regimen and in turn, increase safety [38]. The most important step for non-hemophilia experts to take if they encounter a patient with bleeding symptoms is to immediately screen with standard coagulation tests, if readily available (including prothrombin time and activated partial thromboplastin time (aPTT), and assessment of FVIII and factor IX levels) [8]. Proper interpretation of coagulation laboratory test results is critical; however, physicians do not always recognize the potential clinical significance of abnormal values, specifically the isolated prolongation of aPTT in patients with AHA. Reding et al. conducted a survey of 302 physicians across several specialties, including internal medicine, emergency medicine, rheumatology, hematology-oncology (17% of the sample), and critical care medicine, to identify potential barriers to the effective recognition and management of AHA [37]. The authors concluded that a large proportion of physicians did not recognize prolongation of aPTT as an indicator of underlying coagulopathy in a patient presenting with bleeding; hence, they were unlikely to investigate the cause of bleeding. Moreover, over half of the physicians would not have chosen to repeat aPTT testing to confirm the initial result. Notably, neither emergency medicine nor critical care physicians considered coagulopathy as the primary explanation for the clinical presentation of the patient in question, and physicians were also reluctant to consult a hemophilia expert. This survey and our own findings suggest that more emphasis should be placed on collaboration between nonhemophilia and hemophilia experts, which may help educate providers about the type of coagulation tests to order and their correct interpretation.
In many of the identified articles, physicians might have been unaware of all available hemostatic treatment options (especially the availability of rpFVIII) or unclear of how they should be dosed [15]. In 2017, Kruse-Jarres et al. provided comprehensive guidance for the management of AHA, including recommended first-line hemostatic treatments and dosing regimens [8]. Briefly, these therapies can be divided into two major categories: (i) FVIII-based therapy (rpFVIII), and (ii) bypassing therapies (aPCC and rFVIIa) [38,39].
rpFVIII is a recombinant analog of porcine FVIII and has enough homology to human FVIII to effectively substitute for endogenous FVIII in the coagulation cascade while remaining unrecognized by FVIII inhibitors because of differences in protein epitopes. rpFVIII has a recommended initial dose of 200 U/kg [40], which is thereafter adjusted in terms of dose and frequency based on FVIII recovery levels and individual clinical response (although real-world experience indicates that an initial dose of 100 U/kg is sufficient) [41]. An algorithmic approach to dosing with rpFVIII, depending on FVIII recovery levels, can be used to guide dosing and does not require rpFVIII inhibitor assay results prior to treatment initiation [41]. Although continued treatment with rpFVIII might cause an anamnestic reaction due to inhibitor cross-reactivity, rpFVIII remains effective in the majority of patients with AHA with pre-existing or de novo inhibitors [40,42]. Most importantly, the activity of rpFVIII can be quantified using the one-stage FVIII clotting assay, which can guide dose titration or therapy switching [40].
Bypassing agents achieve hemostasis by generating thrombin (in the absence of FVIII) at the site of bleeding [10] and are recommended for non-life-threatening or non-limb-threatening bleeding [8]. rFVIIa is a recombinant form of human FVIIa with a recommended dosing frequency of 70-90 µ/kg every two to three hours until hemostasis is achieved. aPCC is a plasma-derived concentrate that contains zymogen forms of procoagulant factors (F) II, FVII, FIX, and FX with trace amounts of their activated forms and anticoagulant protein C in a physiologically balanced ratio (off-label in the United States) [43]. Unlike rpFVIII, the hemostatic activity of both aPCC and rFVIIa cannot be measured by standardized and widely available assays; therefore, it is not possible to predict treatment response or adjust the dose accordingly [8].
The 2017 AHA Guidance recommended rpFVIII as an appropriate first-line replacement therapy for life-or limb-threatening bleeding, where the drug is readily available if no baseline rpFVIII inhibitor is present and if FVIII activity measurement is readily available [8]. Unlike bypassing therapies, rpFVIII may not be readily available to all providers. However, manufacturers often provide options to facilitate swift access to products including rpFVIII ( Table 2). The 2017 AHA guidance recommends bypassing aPCC or rFVIIa therapies for non-life-threatening or non-limb-threatening bleeding [8]. Half of the treaters identified in this review used bypassing agents, whereas only four providers used rpFVIII. Although bypassing agents have an established history of efficacy and safety for the treatment of AHA, rpFVIII effectively controls bleeding. It has the additional advantage of a longer half-life compared with rFVIIa [40] and may have a lower risk of thrombotic complications compared with bypassing agents, as its activity can be measured and doses can be adjusted accordingly during treatment. One of the patients in our review endured a 21-day complicated hospital stay, during which a >500% peak in FVIII levels was reached during rFVIII treatment (despite dose de-escalation), which presented a high risk of thrombosis [34]. The ability to measure rpFVIII levels throughout treatment, not just when the inhibitor was eradicated, may have been beneficial in this case.
Few providers used rpFVIII. Although we were unable to clearly identify barriers to the utilization of rpFVIII, it is possible that physicians were aware of this treatment option but were concerned about the potential for the development of anti-porcine FVIII (anti-pFVIII) antibodies, as acknowledged in the prescribing information and in the 2017 AHA guidance [8,40]. The rpFVIII pivotal phase 2/3 study allowed for the inclusion of patients with anti-pFVIII inhibitor ≤20 BU and 10 such patients were included, with 9/10 and 1/10 achieving an effective or partially effective response at 24 hours, respectively. One of these patients who achieved an effective response at 24 hours was later discovered to have had a baseline anti-pFVIII titer of 29 BU. In addition, five patients developed de novo anti-pFVIII inhibitors in this study, and all of them had effective responses to rpFVIII at 24 hours [42]. Another study has further demonstrated continued hemostatic control with rpFVIII even in the presence of anti-pFVIII inhibitors, which suggests a possible lack of direct correlation between rpFVIII inhibitor presence and treatment efficacy [41]. This may reflect the type 2 kinetics of the acquired inhibitor, which may allow rpFVIII to remain clinically effective despite the presence of anti-pFVIII inhibitors [1,7]. Another barrier to rpFVIII treatment might be the misconception that anti-pFVIII antibody titers must be obtained prior to rpFVIII treatment [40]. Although treaters are encouraged to consider evaluating anti-pFVIII inhibitor titers before using rpFVIII [40], results from these assays are not required before treatment initiation, as FVIII activity can be measured using the standard FVIII activity one-stage clotting assay (OSCA) and the dose adjusted accordingly ( Table 2) [10,41].
Four patients (17%) identified in our literature review died [15,20,22,32]. Although AHA is acknowledged to have a high mortality rate [8], there may be indications of some lack of knowledge and awareness about AHA and its optimal therapies that may have negatively affected the clinical outcome. Clearly, there were delays in diagnosis and three [15,20,22] of four reports did not mention the dosing regimens used for bypassing agents. For example, Roy et al. did not provide doses of both bypassing therapies used or specified if they were administered sequentially or concomitantly [22]; these agents have a known potential for thrombotic adverse events [38,39]. It is particularly important to take caution with the dosing of bypassing agents, as their activity cannot be monitored with readily available assays [8]. Indeed, experts have acknowledged that concern about thrombosis risk associated with bypassing agents may have led to the underdosing of these agents [38,39,44]. This patient was also treated intermittently for two years for gastrointestinal bleeding and had inconclusive endoscopic studies before diagnosis [22], clearly demonstrating the issue of delayed diagnosis of AHA in the United States. Aslam et al. described a patient with AHA who died of pulmonary embolism associated with aPCC treatment [15]. However, the authors did not provide the dose or frequency of aPCC dosing. Moreover, the authors did not appear to seek hemophilia-expert input and did not mention rpFVIII as a treatment option, despite having referenced the 2017 AHA guidance.
Two elderly patients (91 and 87 years old) with a high thrombotic risk (a history of hypertension, atrial fibrillation, mitral stenosis) were treated with emicizumab after achieving hemostatic control with bypassing agents rFVIIa and aPCC, both of which increase the risk of thrombosis [12,17]. rpFVIII might be a better treatment choice for these types of patients because it can be monitored with the OSCA before treatment with emicizumab or with the bovine chromogenic assay after treatment with emicizumab [45]. Moreover, emicizumab is not currently approved for AHA treatment in the United States, and could potentially cause thrombotic complications, especially in elderly patients who have underlying comorbid conditions [46].
Since the beginning of the coronavirus disease 2019 (COVID-19) outbreak, a few cases of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection associated with de novo AHA have been reported in the literature [47][48][49][50]. COVID-19-infected patients have been shown to develop significantly more thrombotic complications and to have increased von Willebrand factor (VWF) activity, VWF antigen, and FVIII activity levels, as well as elevated D-dimer and fibrinogen levels, compared with non-COVID-19infected patients [51]. In one case we reviewed, the physician administered rFVIIa followed by aPCC in a patient with COVID-19 despite the heightened thrombotic risk due to COVID-19 infection [50]. Although one of the authors of this case report is a hemophilia treater and we did not include it in our overall analysis, moving forward, we feel it is important and relevant to carefully consider the appropriate management of AHA in patients with SARS-CoV-2 or similar thrombogenic infections that increase thrombotic risk.
Our critical review has some limitations. We confined our review to the case details reported in previously published papers. Therefore, it is possible that some of our assumptions regarding potential knowledge gaps or choices of treatment and consultation with hemophilia experts may not be accurate. Furthermore, we made assumptions about the availability of treatment options and published recommendations based on the publication date of the articles, which are subject to some uncertainty. We also assumed that articles that did not discuss other treatment options did not consider them at all. While this might not be accurate in every case, the articles in which bleeding was controlled with multiple agents [12,17,18,21,[23][24][25][27][28][29]34] might have achieved better outcomes sooner with a more appropriate choice of first-line agent and, in the cases in which the patient died, not all available recommended agents were tried [15,22,32]. Moreover, we did not compare our findings with a similar analysis of publications by hemophilia experts.
This review lends weight to previous calls to raise the level of awareness of AHA among non-hemophilia experts practicing in the United States and to encourage them to consult with hemophilia experts to assist with diagnosis and treatment [5]. The importance of expert consultation regarding the management of patients with bleeding disorders is recognized by the Medical and Scientific Advisory Council of the National Hemophilia Foundation guideline documents 257 [52] and 265 [53]. These guidelines recommend an ongoing relationship with a bleeding disorder specialist during the perinatal care of women and consultation with a bleeding disorder specialist for the care of patients presenting at the emergency department with bleeding disorders. Moreover, the 2017 AHA guidance states that patients with AHA are best managed or treated in close consultation with physicians experienced in AHA [8].

Conclusions
Our findings indicate variability in hemostatic management by non-hemophilia experts in the United States. They also revealed potential barriers to AHA diagnosis and treatment; we have suggested resources that may help overcome these barriers. Non-hemophilia experts are urged to consult with local hemophilia experts with experience treating AHA in the diagnosis and treatment of patients with suspected or confirmed AHA and to follow the most recent AHA guidelines. Finally, it is apparent that more efforts to help raise the level of awareness of AHA among non-hemophilia experts in the United States are urgently needed to improve clinical outcomes for these patients.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: Ali G. Mokdad declare(s) employment and stock/stock options from Takeda Pharmaceuticals USA. Anjali Sharathkumar declare(s) personal fees and Advisory Board from Takeda. Anjali Sharathkumar received lecture fees from Takeda for presenting at ASH and AIHA. They have also received advisory board payments from Takeda. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.