- Top of page
- Case reports
Summary. Background: Inhibitors are rare in boys with mild hemophilia A (MHA; factor (F)VIII:C > 5%) but may arise following intense FVIII exposure, e.g. continuous infusion (CI). Objectives: To determine the impact of intense FVIII exposure in inhibitor formation in MHA at our institution and to compare this with previous reports. Patients and methods: We reviewed FVIII exposure and inhibitor development in boys (ages 0–18 years) with MHA followed at our institution from 1996 to 2001 and conducted a Medline search (1966–2002) on the experience of inhibitor development following intensive/CI exposure to FVIII. Results: We identified 54 boys with MHA. Twenty-nine (54%) had been exposed to FVIII. Seven had received FVIII by CI. Four developed inhibitors; three high titer (at ages 10 years, 16 years and 17 years) and one low titer (at 1 month old). All four had received a CI of recombinant (r) FVIII of at least 6 days within 6 weeks of developing inhibitors. Baseline FVIII levels fell to < 1% in all cases and the three with high-titer inhibitors developed severe bleeding. Immune tolerance therapy (ITT) was attempted in two boys and was successful in one. Our literature search identified 35 cases (only four children) with MHA developing inhibitors following intense FVIII exposure often in the context of surgery. Conclusions: The incidence of inhibitors in our MHA population was 7.4%. If expressed according to exposure the incidence was significantly higher: 14% (4/29) for any exposure to FVIII and 57% (4/7) for exposure by CI. A prospective study to address whether CI is associated with an increased incidence of inhibitor development in MHA is warranted.
- Top of page
- Case reports
Bleeding in hemophiliacs may be prevented or arrested by the augmentation of endogenous factor (F)VIII or administration of exogenous FVIII. The development of allo-antibodies against exogenous ‘wild’ type FVIII, known as ‘inhibitors’, is currently the most serious complication in the management of hemophilia patients . Inhibitors neutralize FVIII leading to treatment failure and are therefore usually detected when bleeding episodes fail to respond to appropriate FVIII replacement [2,3].
Inhibitor formation is reflective of an immune response against a ‘foreign’ FVIII molecule. The pathogenesis of inhibitor development is better understood in severe hemophilia A (SHA), where hemophilia is associated with mutations (large deletions, inversion mutations and premature stop codon mutations) in the FVIII gene resulting in a complete absence (<1%) of circulating FVIII:C [4,5]. In contrast, other patients with hemophilia have measurable levels of endogenous FVIII; 1–5% for moderate hemophilia and >5% for mild hemophilia A (MHA). Current opinion is that in many such cases this FVIII is conformationally altered with subtle changes rendering it antigenically distinct from exogenous ‘wild’ type FVIII [1,4–9]. Consequently, in this setting, wild type FVIII may still be potentially immunogenic.
In MHA the development of inhibitors is a serious but infrequently reported complication (3–13%), occurring more commonly later in life, often following intensive FVIII replacement for surgery or trauma [1,5,10–14]. Due to cross-reactivity with endogenous FVIII, the development of inhibitors in most patients with MHA results in a fall in the endogenous level of FVIII:C, converting patients to a severe phenotype (FVIII:C <1%) [5–8,15,16].
In contrast to SHA [4,8,9,17–22], little is known regarding risk factors for inhibitor development in MHA and specifically regarding factor exposure (amount and intensity of FVIII exposure, type of FVIII used, and context in which exposure occurs). In 1970 Crowell commented that the less frequent need for transfusions in mild hemophiliacs might be partly responsible for their low incidence of inhibitors . In contrast, Strauss did not find a higher cumulative factor exposure in severe hemophiliacs who developed inhibitors than in those who did not . Neither paper examined the relationship between inhibitor development and intensity of exposure.
Traditionally, FVIII has been administered by episodic bolus injections (BI). Disadvantages of BIs are the concentration peaks and troughs that occur resulting in a risk of bleeding during troughs and inefficient and costly usage associated with peaks. Continuous infusion (CI) of FVIII eliminates peaks and troughs and, since it has been shown to be a safe and effective method of administering FVIII, has become extensively advocated [24,25].
Recent reports associating inhibitor development in MHA following exposure to FVIII by CI  prompted our group to review our institutional experience in MHA (54 boys) to determine the impact of FVIII exposure in the pathogenesis of inhibitor formation. These 54 boys were prospectively followed with yearly inhibitor screening. For boys developing inhibitors we present detailed information regarding their clinical courses and management. Additionally, we conducted a search of all available literature (1966–2002) on the experience of inhibitor development in MHA cases following intensive or continuous exposure to FVIII.
- Top of page
- Case reports
Retrospective chart reviews were conducted to evaluate demographic data relating to hemophilia (FVIII:C level, desmopressin; 1-deamino-8-d-arginine vasopressin (DDAVP) responsiveness, and family history of hemophilia), lifetime FVIII usage (type, amount and intensity of factor used, and age at first exposure) and inhibitor development. Total amount of exposure was expressed as cumulative exposure days (CED), an exposure day (ED) being defined as a day in which a patient receives at least one dose of FVIII regardless of source [plasma derived (pd) or recombinant (r)], amount or method of administration (BI vs. CI). Intensity of exposure was expressed as the maximum number of consecutive ED at any one time regardless of method of administration.
For patients who developed inhibitors, chart reviews were conducted to evaluate the context of inhibitor development, management of bleeds postinhibitor development, specific management of the inhibitor and patients' clinical course.
For all patients, determination of FVIII:C levels was performed using a one-stage clotting assay. Inhibitor testing was performed using the Bethesda assay. For patients developing inhibitors FVIII gene mutation analysis was performed in the Canadian National Hemophilia Genotyping Laboratory at Queen's University, Kingston, Ontario. In each case, genomic DNA was obtained by a salt extraction method. Initial mutation screening was performed on polymerase chain reaction-amplified FVIII exons by conformation-sensitive gel electrophoresis. Fragments demonstrating heteroduplex formation were subjected to automated DNA sequencing .
A Medline search (1966–2002) was conducted using combinations of key words: mild hemophilia A, inhibitor, continuous infusion, and supplemented by additional references located in the bibliographies of listed articles. Cases were accepted only if it was clearly stated that the patient had received a CI or daily exposure to FVIII for at least 4 days. Four days was chosen as bleeds may be treated for 2 or 3 days but not usually for 4 days or more. Articles that used qualitative terms describing intense exposure were also included.
- Top of page
- Case reports
The 54 boys were diagnosed with MHA at a mean age of 2.5 years (median 1.5 years; range 4 days to 16 years). Patients' mean baseline FVIII:C level was 17.5% (median 15.0%; range 6–38%). Forty-five of 51 boys in whom DDAVP response testing was performed had either a complete (n = 35; 69%) or partial response (n = 10; 20%) to DDAVP. Complete response was defined as a response satisfying the following two criteria: at least a doubling of FVIII:C level over baseline and a rise in FVIII:C level to at least 30% 1 h post-DDAVP. Partial response was defined as a response satisfying only one of the above two criteria.
Twenty-nine (54%) of the 54 boys have been exposed to FVIII some time during their childhood: four to pdFVIII, 15 to rFVIII, and 10 to both (Table 1). Mean age at initial exposure to any FVIII was 5.4 years (range 5 days to 16.3 years); 2.3 years (range 21 days to 6.6 years) for pdFVIII and 8.0 years (range 5 days to 16.3 years) for rFVIII. Boys were much more likely to be exposed if they were non-responders (n = 6; 86% exposed) vs. partial/complete responders (n = 45; 48% exposed) to DDAVP (P < 0.01). Boys exposed to factor had lower endogenous baseline FVIII:C levels (mean 14.9%; range 6–29%) than non-exposed boys (mean 20.5%; range 6–38%), but among exposed boys cumulative exposure (CED) or intensity of exposure (consecutive ED) was not correlated with baseline FVIII:C levels.
Table 1. Characteristics of factor (F)VIII exposure in boys with mild hemophilia A (MHA) pre-inhibitor development (ranked according to intensity of exposure)
|Patient||Baseline FVIII %||pdFVIII||rFVIII||Infusion strategy BI/CI||Inhibitor, yes/no|
|Age (years) at first exposure||CED (days)||Max. intensity of exposure (days)||Age (years) at first exposure||CED (days)||Max. intensity of exposure (days)|
For the 29 boys exposed to any FVIII, the mean number of CEDs was 16.3 days (1–46 days). Sixteen boys received FVIII daily for at least 6 consecutive days, seven by CI and nine by BI. Four of the seven exposed to CI developed inhibitors (high-titre in three); all within 6 weeks of exposure. In contrast, none of the nine boys exposed to daily BI (minimum of 6 consecutive ED) alone developed inhibitors (P = 0.02 by Fisher's exact test). Details of the boys developing inhibitors following CI are presented below and in Table 2. Three boys (patients 5, 12 and 16 of Table 1) also exposed to a CI of rFVIII (10 days, 8 days and 6 days) did not develop inhibitors. These three boys were exposed to CI at the ages of 2.2 years, 2 months and 7.7 years for a depressed skull fracture, an incarcerated hernia and a tonsillectomy, respectively.
Table 2. Characteristics of patients with inhibitors
| ||Case 1||Case 2||Case 3||Case 4|
|Baseline FVIII (%)||16||13||29||4–10|
|Post-DDAVP FVIII (%)||36||55||129||14|
|Age at INH detection||16.3 years||10.1 years||16.2 years||33 days|
|Event leading to INH||Hemarthrosis||Ankle fracture||Knee hemarthrosis with arthrocentesis||Neonatal ICH|
|Exposure details||CI × 14 days, then BI q 12 h × 13 days||CI × 6 days, then BI q 2 days × 2 weeks||CI × 11 days, then 4 BI over 10 days||CI × 27 days|
|FVIII ED (pre INH)||27||13||16||27|
|Presentation of INH||Failure to respond to FVIII||Failure to respond to FVIII||Failure to respond to FVIII||Routine surveillance|
|Time: exposure to INH detection (weeks)||6||6||6||4|
|FVIII (%) at INH detection||< 1||< 1||< 1||< 1|
|Max. INH to H and P FVIII (BU)||12 and 16 BU||51 and 122 BU||166 and 56 BU||2 and 0 BU|
|Bleeds post-INH||(Over 3 months)||(Over 1 year)||(Over 2 years)||None|
| Joint bleeds||2||2||9|| |
| ST/muscle||3||4||13|| |
| Hematuria||1||3||5|| |
| Other|| || ||Epistaxis-1|| |
|Treatment of bleeds post-INH||PFVIII, FEIBA, rFVIIa||rFVIII, PFVIII, FEIBA, rFVIIa||rFVIII, FEIBA, rFVIIa||rFVIII|
|Management of INH||At 3 months: plasmapheresis + corticosteroids and ITT × 1 month||Therapy refused||At 19 months: plasmapheresis + ITT × 1 year||Prophylaxis (3×/ week) for 1 year|
|Course of INH||Disappeared 3.5 months||Disappeared 14 months||Persistent||Disappeared 12 months|
|Gene mutation analysis||Missense mutation: Val2016Ala (exon 19: A3 domain): documented 10 times in HAMSTeRS. No prior report of inhibitors||Missense mutation: Pro1854Leu (exon 17: A3 domain): not previously documented in HAMSTeRS||Mutation not identified: CSGE and MDE screens negative Exons 12, 15, 18 and 26 normal by sequencing HAMSTeRS||Missense mutation: Asn2286Lys (exon 26; C2 domain): not previously documented in|
The incidence of inhibitors in our patients, expressed according to overall number of patients with MHA, is 7.4%. However, the incidence is 14% (4/29) if expressed according to MHA patients exposed to any exogenous FVIII. The incidence is 25% if expressed as a percentage of patients exposed to at least 6 consecutive ED (reflecting intensity of exposure). The incidence is 57% (4/7) if expressed as a percentage of patients exposed to FVIII by CI.
- Top of page
- Case reports
MHA is genotypically heterogeneous and mutations have been reported throughout all domains, other than the B domain, of the FVIII gene. In MHA only particular missense mutations appear to be associated with inhibitor formation [4,5,9,27]. These tend to be clustered in or around certain immunogenic ‘hot spots’; amino acids 482–501 of the A2 domain and 2248–2312 of the C2 domain near the C1–C2 junction [1,5,19,28,29]. Various investigators have suggested that the risk of developing an inhibitor in MHA may be largely confined to a few kindreds possessing high-risk genotypes. Hay et al. reported a 41% inhibitor incidence among siblings of boys with MHA and inhibitors , and Knobe et al. described two families with MHA each with multiple members with inhibitors .
Our four patients were unrelated and had different mutations. Two of the three mutations in our patients were missense mutations in the A3 domain; one (case 1) Val2016Ala (exon 19) substitution represents a relatively minor change to a small hydrophobic amino acid. A high prevalence of this mutation has been found in a small region in the Canadian province of Newfoundland where 44 affected males have been identified with this mutation . None of these 44 males has developed inhibitors. There are a number of other reports of this mutation but as yet no reports of inhibitors associated with this mutation. The second mutation (case 2), Pro1854Leu (exon 17), is a novel missense mutation and probably involves the interruption of an α helix which may induce a conformational change in the mutant protein. Case 4 had a missense mutation (Asn2286Lys) in the C2 domain which alters the charge of the molecule and probably its protein conformation. These boys emphasize the mutational heterogeneity associated with inhibitor development in MHA. Two of our four boys with inhibitors had younger hemophilic brothers—three in total. None of these brothers has developed inhibitors despite two of them having been exposed to FVIII, albeit receiving considerably less total, and less intense exposure. A lingering concern is what is the risk of inhibitor development in these younger siblings should they experience a similar intense exposure to FVIII as their older siblings, and should this concern influence management of bleeds in these younger siblings.
Our observations as well as studies involving monozygotic hemophilic twins discordant for inhibitors suggest that factors other than FVIII genotype predispose patients with MHA to the development of inhibitors . Some of these factors may be genetic and might include mutations and polymorphisms of proteins involved in immune response and cytokine production . Other factors may be related to the intensity of factor exposure [9,22,32,33].
Although the overall incidence of inhibitors in our MHA population (7.4%) was low and consistent with literature reports (3–13%) , all of our patients who developed inhibitors did so after CI. This suggests that the actual CI method of administering FVIII may be associated with an increased risk of inhibitor development. There are no prospective studies reporting the incidence of inhibitors in MHA patients exposed to a CI. Available literature reports primarily consist of single case reports or small case series, and these papers often omit details necessary to link intense exposure to inhibitor development, e.g. duration and type of exposure and time to inhibitor development post-exposure. Furthermore, many publications fail to describe adequately the exposure but instead use qualitative terms (‘extensive perioperative’, ‘frequent high doses’, ‘several times’ and ‘intense use’) to suggest intense exposure.
Our literature search identified 21 reports describing 35 patients with MHA developing an inhibitor following intensive exposure to FVIII (Table 3). Most reports involved one to five cases and FVIII exposure occurred usually in the context of surgery. In 12 patients CI was specifically described as the method of exposure. Most patients developing inhibitors were exposed to at least 7–10 days of either daily BI or CI. All types of factor concentrate were involved in these cases and no association could be made between inhibitor development and type of factor used. Most reports involved adults and only four children (ages 6 months, 7 years, 8 years, and 11 years) with MHA have been previously reported to have developed inhibitors following intensive FVIII exposure. In keeping with the observation that MHA patients developing inhibitors tend to be older, we found that our three patients developing inhibitors following CI were all 10 years of age or older. In contrast, the three boys in our population who did not develop an inhibitor following CI were all under 10 years of age.
Table 3. Medline search of inhibitor development in mild hemophilia A patients receiving intensive factor VIII exposure (minimum, daily BI × 4 days or CI)
|Reports||No. of patients||Baseline FVIII (%)||Age at INH (years)||Type of FVIII||Exposure||Days (post exp) to INH development||Max. INH titer (BU)||Reason for treatment|
|Beck 1969 ||2||8||21||C||BI × 12 days||9||1||Surgery|
|5||52||C||BI × 13 days||15||15||Surgery|
|Crowell 1970 ||1||4–7||21||C||BI × 20 days||23||12.5||Surgery|
|Lechner 1972 ||1||6||20||C||BI × 36 days||34||0.75||Trauma|
|Shapiro 1975 ||1||10||50||C||BI > 7 days||≈ 30||40||Surgery|
|Kesteven 1984 ||1||7–14||26||pd||BI × 16 days||90||173 Oxford U||Trauma|
|Bovill 1985 ||1||14||68||pd + C||BI × 10 days||11||4.0||Surgery|
|Suzuki 1995 ||1||17||60||r||CI × 15 days||106||30||Surgery|
|Fijnvandraat 1997 ||1||20||19||pd + C||‘Extensive perioperative’||NR||22||Surgery|
|Thompson 1997 ||1||8–11||41||r||BI × 35 days||28||128||Surgery|
|Baglin 1998 ||2||11||49||pd||CI (NR)||<20||67||Surgery|
|Hay 1998 ||≈ 2/3 of 26||>5 (in 22 of 26 patients)||Median = 33 (7–71)||NR||‘Intensive replacement therapy’||Immediately preceding INH development||Mean: 22.5||NR|
|Van den Brink 1999 ||1||29||63||pd||CI × 5 days||42||250||Surgery|
|Peerlinck 1999 ||1||23||39||pd||NR||‘Shortly thereafter’||305||Surgery|
|Tagariello, 1999 ||1||NR||41||pd||CI × 12 days||12||NR||Surgery|
|Koestenberger 2000 ||1||NR||0.5||r||CI × 9 days||9||170||Surgery|
|White 2000 ||2||5||19||r||CI × 5 days + BI × 5 days||30||70||Surgery|
|7||40||r||CI × 5 days + BI × 5 days||21||87||Surgery|
|Robbins 2001 ||1||16||16||r||CI × 2 days + BI × 5 days||28||50||Surgery|
|Puetz 2001 ||1||10||7||r||CI × 4 days||13||421||Trauma|
|Knobe 2001 ||5||10||8||r||‘Frequent high doses’||120||8.4||Burn|
|20||67||NR||‘Short period’||NR||11||GI bleed|
|Vlot 2002 ||1||30||68||pd||BI × 24 days||150||36.1||Surgery|
|Liu 2002 ||3||NR||NR||NR||‘Intense use’||NR||91||Surgery|
The incidence of inhibitor development in MHA following intense FVIII exposure is not known. Certainly not all MHA patients exposed to a CI develop an inhibitor; in our series, three of seven boys exposed to a CI of FVIII did not develop an inhibitor. Nonetheless, some hemophilia treatment centers have recommended against the use of CI in MHA patients . In the absence of adequate data to support an evidence-based recommendation, we believe that the role of intensive FVIII exposure as a risk factor for inhibitor formation, both in the form of bolus and CI, merits further study. Such a study will need to be both prospective and multicenter.
The mechanism(s) by which CI of FVIII might result in a higher incidence of inhibitor formation are unknown. One possibility is that FVIII given as a CI is modified into a more antigenic form during storage ex vivo by dilution (if performed) or by prolonged contact with plastic infusion materials or with inflamed veins. We speculate that perhaps when FVIII is given as a CI into peripheral veins of patients with MHA some leakage subcutaneously could result in a more immunogenic type of exposure. Intensive FVIII exposure in patients with MHA always occurs in the context of significant bleeds or major surgery, as in all four of our patients. In these situations, it is possible that extensive tissue damage and associated inflammation facilitates an antibody response against exogenous FVIII due to the presence of immunological ‘danger signals’. In keeping with this is the observation of Kaufman et al. that animals exposed to FIX during periods of inflammation can develop inhibitors to FIX .
All four of our patients who developed inhibitors did so after exposure to rFVIII. This includes one patient who had previously been exposed to pdFVIII (two non-consecutive BI) without inhibitor formation. At this time there is no definitive evidence to indicate a higher risk of inhibitor development with recombinant vs. plasma-derived FVIII products [11,28,38,39]. Although a higher incidence of inhibitors has been documented with the increased use of rFVIII, confounding variables (increased inhibitor surveillance, increasing use of prophylaxis, and greater use of CI) may explain the apparent increase in inhibitor development .
The clinical impact of the inhibitors in our four patients varied from minimal (case 4; low titer) to severe bleeding (cases 1, 2, and 3; all high titer). Consistent with literature reports, our three cases with high-titer inhibitors developed a severe and unusual bleeding pattern reminiscent of that seen in acquired hemophilia [5,19,29,41,42]. The morbidity and mortality for such patients is high . Reports suggest that these MHA patients with inhibitors respond differently to management. Unlike in cases of SHA, spontaneous disappearance of inhibitors in MHA occurs (as in case 2), and is thought to result from auto-tolerance from ongoing production of endogenous FVIII [5,43]. ITT appears to be less effective in patients with MHA [5,44]. Immune suppression (corticosteroids, cyclophosphamide, etc.) is perceived to be more important than immune tolerance in patients with MHA . In our series of four patients, ITT was attempted in two cases, and was successful in one (with plasmapharesis and corticosteroids). Presently, there are insufficient data to recommend any specific ITT regimen for MHA patients with inhibitors. Prospective studies are needed to determine the subset of MHA patients with high-titer inhibitors who would merit this demanding and expensive treatment strategy.
Based on our observations and previous reports of inhibitor development in MHA patients following intensive exposure to FVIII, we recommend that physicians who care for such patients closely monitor them for inhibitor development in the event that they are intensely exposed to FVIII.
In conclusion, inhibitor formation is an increasingly recognized problem in patients with MHA and prospective studies are needed to determine which patients are at risk for the development of inhibitors. Our observations suggest that intense exposure to FVIII, possibly in the form of CI, may be associated with an incidence of inhibitor formation similar to that observed in SHA cases. Boys with MHA who are intensely exposed to FVIII by CI should be carefully monitored for inhibitor development. A prospective multicenter study to address the issue of inhibitors in MHA post-CI and following other intense exposures to FVIII is warranted.