Keywords acquired coagulation factor deficiency - autoantibody - bleeding disorder - factor
V - inhibitor - nonneutralizing antibody
Coagulation factor V (or FV for the purpose of medical safety) is a cofactor of factor
X (FX) in the amplification process of the clotting reaction in the coagulation cascade.
Severe FV deficiency, either congenital or acquired, causes various bleeding symptoms.[1 ]
[2 ]
[3 ]
Due to increasing incidence of autoimmune coagulation factor deficiency (AiCFD) in
Japan, the Japanese Collaborative Research Group (JCRG) conducted a nationwide survey
on this hemorrhagic disorder during the past decade, with support of the Japanese
Ministry of Health, Labor, and Welfare (MHLW).[4 ]
[5 ]
[6 ] Consequently, autoimmune deficiencies of coagulation factor XIII (or FXIII), coagulation
factor VIII (or FVIII), FV, and the von Willebrand factor (VWF) have been enacted
by the Japanese MHLW as the Designated Intractable Diseases (DIDs) codes 288–1, 288–2,
288–4, and 288–3, respectively. Patients with such DIDs are financially supported
by the public medical expenses subsidy system in Japan. Autoimmune FVIII deficiency
(AiFVIIID or acquired hemophilia A) is the most prevalent AiCFD, followed by autoimmune
FV, FXIII, VWF, and FX deficiency (AiFVD, AiFXIIID, AiVWFD, and AiFXD, respectively);
as of December 2020, the cumulative numbers of AiCFD cases in Japan are as follows:
the estimated number of AiFVIIID cases is 2,160, while the reported numbers of AiFVD,
AiFXIIID, AiVWD, and AiFXD cases are 173, 79, 33, and 3, respectively[4 ]
[5 ]
[ 7 ] (JCRG's achievement report 2020 to the Japanese MHLW).
According to two review articles published in 2009 and 2011, a total number of 76
and 78 patients worldwide had AiFVD, respectively[8 ]
[9 ] (bovine thrombin–related FV inhibitors were excluded by the authors of these articles
as we did in the present study). However, these articles only included 8 and 12 AiFVD
cases, respectively, described in Japan. There were no patient records of physician
consultations for AiFVD at our JCRG headquarters in Yamagata University during the
first 7 years (2009–2015) of our survey.
Thus, we focused our search on AiFVD cases with FV inhibitors (excluding bovine thrombin–associated
cases because they are not considered autoimmune in nature) during the past 5 years
(2016–2020) and identified 24 new cases through our JCRG's nationwide survey (Appendix A ), by employing the Governmental Diagnostic Criterion for AiFVD enacted by the Japanese
MHLW (DID code: 288–4) which was analogous to 2015 Criterion for diagnosing AiFXIIID.[10 ] In addition, we confirmed 177 previously reported cases of AiFVD from the Japanese
medical institutes through periodic extensive literature searches of English and Japanese
reports in PubMed and Igaku Chuo Zasshi (ICHUSHI) databases, respectively. Herein,
we aim to summarize the clinical features (abnormal laboratory findings, presenting
symptoms, diagnoses, treatments, and outcomes) of 201 patients with AiFVD in Japan
to improve the understanding and awareness of this disease. To the best of our knowledge,
this article contains the largest number of Japan's patients with AiFVD reported so
far and provides the most up-to-date overview of its patients.
Number of Autoimmune Factor V Deficiency Cases in Japan
Number of Autoimmune Factor V Deficiency Cases in Japan
In the past 5 years (2016–2020), more than 200 patients with suspected AiCFD participated
in the JCRG survey. Among them, 24 patients were diagnosed with AiFVD due to the presence
of anti-FV autoantibodies detected using immunological assays (e.g., immunoblot, immunochromatography,
and enzyme-linked immunosorbent assay) and FV inhibitors detected using functional
coagulation tests (e.g., Bethesda assay and FV mixing test).[11 ]
[12 ]
[13 ]
[14 ]
[15 ]
[16 ]
[17 ] Our research results do not include cases in which the bovine thrombin preparation
produced an FV inhibitor due to an anti-bovine FV antibody, since the bovine thrombin
preparation was no longer commercially available in Japan at that period. Based on
a population size of 125 million individuals as of August 2021 (
https://www.stat.go.jp/data/jinsui/new.html
), the estimated incidence of AiFVD in Japan is at least approximately 0.04 per million
persons per year.
We also conducted literature search in PubMed (last accessed on March 25, 2021) and
ICHUSHI (last accessed on March 25, 2021), and identified 1,201 and 1,097 relevant
reports, respectively. One suspected AiFVD case was also identified by manually searching
the individual bibliographies of the identified reports.[18 ] By carefully perusing the results of PubMed and ICHUSHI searches, a total of 30
and 263 cases were selected as suspected AiFVD cases in Japan, respectively.
Three identified cases were excluded because of the presence of inhibitors against
multiple coagulation factors and higher levels of their titers against other coagulation
factors than those of titers against FV. Seven patients were excluded from our results
as they had received products containing bovine thrombin–like previously reported
cases[19 ]
[20 ] and thus, had “cross-reacting” antibodies rather than autoantibodies.[21 ] After excluding 98 duplicates and two cases of hereditary FVD, 201 suspected AiFVD
cases were selected and classified into four subgroups according to the diagnostic
criteria of the Japanese MHLW for AiFVD: “definite” (n = 37), “probable-2” (n = 129), “probable-1” (n = 18), and “possible” diagnoses (n = 17; [Table 1 ]).[22 ]
Table 1
Governmental Diagnostic Criterion for AiFVD enacted by the Japanese Ministry of Health,
Labor, and Welfare (the Designated Intractable Disease code: 288–4)
Patients with “definite” or “probable” diagnosis will be provided with governmental
medial/financial support
A. Symptoms and others
1. Recent onset of bleeding symptoms (developed within the previous year mainly in
older adults)
2. No family history of congenital/inherited deficiency of FV/5 (nor other coagulation
factor deficiencies)
3. Lack of previous bleeding symptoms especially in association with previous hemostatic
challenges (e.g., trauma, surgery, invasive tests, tooth extraction, and delivery)
4. Bleeding symptoms are not due to the excessive use of certain medications such
as anticoagulants and antiplatelet drugs
B. Laboratory findings
1. Abnormality of FV parameter(s) on specific laboratory testing (usually FV activity
and/or antigen <50% of reference levels)
(1) FV activity (FV:C): always extremely low
(2) FV antigen (FV:Ag): usually variably low
(3) FV specific activity (FV:C/FV:Ag ratio): usually variably low
2. Laboratory tests for definite diagnosis
(1) PT and/or aPTT cross-mixing test showed an inhibitor pattern in the presence
of isolated severe FV deficiency
(2) The FV inhibitor (circulating anticoagulant) is positive
Results of the 1:1 mixing test of FV activity between patient's plasma and normal
control's plasma (50% each) was not corrected
Measurement of inhibitor titer: The residual FV activity of mixed plasma between patient's
diluted plasma and healthy control's plasma was measured after incubation at 37°C
for 2 hours (Bethesda method)
(3) Detection of anti-FV autoantibodies
Noninhibitory antibodies can be detected by binding methods (immunoblot, enzyme-linked
immunosorbent assay, immunochromatography, etc.). FV inhibitors, that is, neutralizing
anti-FV autoantibodies, are also detected using immunological methods
C. Differential diagnosis
Parahemophilia (hereditary FV deficiency), congenital combined FV and factor VIII/8
deficiency, all secondary FV deficiencies (including disseminated intravascular coagulation,
severe liver disorder, etc.), inherited factor X/10 (FX) deficiency, autoimmune FX
deficiency, all secondary FX deficiencies, inherited prothrombin deficiency, autoimmune
prothrombin deficiency, all secondary prothrombin deficiencies, autoimmune acquired
factor XIII/13 deficiency, anti-phospholipid antibody syndrome, etc., were excluded
Diagnosis category
Definite: All A items plus B1 and B2-(3), but all C items must be excluded
Probable: All A items plus B1 plus B2-(1) or B2-(2), but all C items must be excluded
Probable-1: All A items plus B1 plus B2-(1), but all C items must be excluded
Probable-2: All A items plus B1 plus B2-(2), but all C items must be excluded
Possible: All of A items plus B1
Abbreviations: Ag, antigen; AiFVD, autoimmune factor V deficiency; aPTT, activated
partial thromboplastin time; FV, factor V; PT, prothrombin time.
Since all of these AiFVD patients were reported by Japanese authors from Japanese
medical institutions, they were regarded as Japanese. Based on the Japanese customs,
if the patient is non-Japanese, this will be clearly stated in the article. Nevertheless,
the authors used “Japan's cases” rather than “Japanese cases” in this manuscript to
distinguish nationality from race when needed.
Sex Ratio
The number of male patients was higher than that of female patients, and the sex ratio
(male:female) was 2.8 ([Table 2 ]) which was higher than that reported in 2009 and 2011 (1.8 in 76 cases and 2.0 in
78 cases, respectively[8 ]
[9 ]). The difference was not significant (p = 0.60). The sex ratios in the AiFVIIID and AiFXIIID groups were 0.74 (52/70) and
1.21 (51/42), respectively.[23 ]
[24 ] The reason for the male predominance in the AiCFD group remains unknown.
Table 2
Sex and age information
Subgroup
Total
F
M
n.d.
M/F rate
Male (%)
p -Value
Sex
Total
201
52
146
3
2.81
73.7
Bleeder
138
33
102
3
3.09
73.9
0.60
Nonbleeder
59
17
42
0
2.47
71.2
Subgroup
Total
Average
SD
Maximum
Minimum
Median
p
-Value
Age
Total
200
71.9
11.9
93
4
74
Bleeder
137
71.5
11.1
93
36
73
0.27
Nonbleeder
59
72.6
13.5
92
4
75
Abbreviations: F, female; M, male; n.d., not described; SD, standard deviation.
Age Distribution
The mean age of patients with AiFVD was 71.9 ± 11.9 years (median, 74 years; range,
4–93 years; n = 200; [Table 2 ]) which was similar to that worldwide as reported in 2009 and 2011 (median, 71 years;
range, 3–95 years; n = 76[8 ] and median, 69 years; range, 3–91 years; n = 78,[9 ] respectively) and in patients with AiFVIIID and AiFXIIID (median, 78 years; range,
2–98 years; n = 154[23 ] and median, 70 years; range, 22–89 years; n = 93,[24 ] respectively). The mean age was apparently higher in patients with AiFVD than in
those with AiFXD (56.5 years, n = 26[7 ]) as AiFVD occurred in only one patient aged <30 years. Although the age of the nonbleeder
group tended to be slightly higher than that of the bleeder group ([Fig. 1A ]), the difference was not significant (p = 0.27). Old age per se is not a criterion to suspect AiFVD because some patients
have been diagnosed with congenital FVD at the ages of 80 and 88 years.[25 ]
[26 ]
Fig. 1 Patients' characteristics. (A ) Age distribution. AiFVD was frequently detected in individuals >60 years of age.
The second peak age of bleeders was younger than that of nonbleeders. (B ) Underlying diseases and conditions. More than one-fourth of AiFVD cases are idiopathic,
and approximately one-fifth are associated with infection. AiFVD, autoimmune factor
V deficiency; Hemat. mal., hematopoietic malignancies; n.d., not described; Post-op,
postoperative.
Underlying Diseases/Conditions
Underlying Diseases/Conditions
Fifty-eight patients (28.9%) with AiFVD had no underlying conditions (that is idiopathic
origin; [Table 3 ]; [Fig. 1B ]). Meanwhile, 39 (19.4%) patients with AiFVD had infectious diseases (mainly involving
the respiratory system), and 34 (16.9%) patients were administered antibiotics which
may be related to development of anti-FV autoantibodies, as previously reported in
patients who had anti-FXIII autoantibodies.[24 ] Thirty-four patients (16.9%) underwent surgery. The causality or association in
these cases is difficult to prove because their implicated conditions are commonly
encountered in the normal population,[8 ]
[9 ] and surgery, infection, and antibiotic therapy frequently occur sequentially and/or
simultaneously in patients with AiFVD. However, a relationship between the drug used
and development of FV inhibitors was established in a few cases; in one patient, FV
inhibitors disappeared 3 weeks after treatment with dabigatran was discontinued, but
reappeared after the medication was readministered.[27 ] The presence and disappearance of immunoglobulin (Ig)-G-type anti-FV autoantibodies
was confirmed in this patient by enzyme-linked immunosorbent assay (paper in preparation).
Similar incidents have been reported in two other patients treated with aspirin and
antibiotics.[28 ]
[29 ]
Table 3
Underlying diseases/conditions
Bleeder
Percentage
Nonbleeder
Percentage
p -Value
Cancer
19
13.8
19
32.8
0.10
Hemot. mal.
2
1.4
2
3.4
0.62
Immune
18
13.0
4
6.9
0.0465
Antibiotics
17
12.3
16
27.6
0.18
Infection
24
17.4
18
31.0
0.47
Post-op
9
6.5
25
43.1
<0.0001
Dialysis
11
8.0
4
6.9
0.36
Aneurysm
2
1.4
2
3.4
0.31
Relapse
3
2.2
4
6.9
0.29
Anticoagulants
8
5.8
2
3.4
0.23
None
52
37.7
6
8.6
<0.0001
n.d.
2
1.4
0
0.0
0.27
Abbreviations: Hemat. mal., hematopoietic malignancies; Immune, autoimmune disease;
Post-op, postoperation;
Note: Bold p -values indicate statistical significance.
Twenty-three (11.4%) Japan's patients with AiFVD had autoimmune diseases such as systemic
lupus erythematosus (SLE) and rheumatoid arthritis. Fifteen (7.5%) patients developed
AiFVD during hemodialysis which was detected due to persistent bleeding from puncture
sites, and/or abnormal findings in repeated routine coagulation tests.
Notably, solid cancers (n = 39; 19.4%) and hematopoietic malignancies (n = 4; 2.0%) were found in patients with AiFVD. The destruction of malignant cells
may contribute to the dysregulation of the inflammatory immune response.
Significant differences were found in the frequencies of surgery-associated and idiopathic
cases (both p < 0.0001) between bleeders and nonbleeders (6.5 vs. 43.1 and 37.7 vs. 8.6%, respectively).
Therefore, bleeders were more likely to have an idiopathic origin, whereas nonbleeders
were more likely to be related to surgery.
Bleeding Sites/Symptoms and Bleeding Severity
Bleeding Sites/Symptoms and Bleeding Severity
AiFVD symptoms vary from no hemorrhage to life-threatening bleedings and thrombotic
complications.[8 ]
[9 ]
[21 ] Fifty-nine (29.4%) patients had no bleeding symptoms ([Fig. 2A ]), although patients may manifested bleeding events after 6 years of close monitoring.[30 ] Most patients with AiFVD demonstrated bleeding in soft-tissue regions, such as subcutaneous
tissues (26.4%), urinary tract (23.4%), and intestine (19.4%). Intracranial (6.5%)
and intraperitoneal (4%) bleeding events were reported in AiFVD patients, similar
to those with AiFVIIID and AiFXIIID.[23 ]
[24 ] The AiFVD patients in Japan reported bleeding in the subcutaneous tissues more frequently
than those in all countries worldwide[8 ]
[9 ] (26.4 vs. 9.2%, p = 0.0001). It is possible that Japanese patients and their attending physicians might
have focused more attention to bleeding symptoms, resulting in this difference in
frequency. Postoperation bleeding, post-tooth extraction, and post–needle puncture
are not specific to AiFVD but are common among patients in Japan and all countries
worldwide (5.9–11%[8 ]
[9 ]).
Fig. 2 Bleeding symptoms. (A ) Bleeding sites. Fifty-nine AiFVD patients did not present with bleeding events.
Conversely, many patients had more than one bleeding site and/or symptom. In total,
138 AiFVD patients experienced 275 bleeding events (2.0 on average). (B ) Severity of clinical bleeding. Ratio of bleeding grades. n.a. represents not applicable
because bleeding sites/symptoms were not described. AiFVD, autoimmune factor V deficiency;
GI, gastrointestinal; Post, postoperative; IC, intracranial; IJ, intrajoint; IM, intramuscular;
IT, intrathoracic; IP, intraperitoneal; IJ, intrajoint; n.d., not described; Post-op,
postoperative; SC, subcutaneous.
According to the published categories of clinical bleeding severity,[31 ] 41.3% of patients with AiFVD had grade-III bleeding ([Fig. 2B ]), indicating that this disease may be more severe than congenital FVD[1 ]
[2 ] in which only 14% of patients experienced grade-III bleeding.[31 ]
Nine (4.5%) AiFVD patients in Japan developed thrombosis as a complication.[11 ]
[16 ]
[18 ]
[32 ]
[33 ]
[34 ]
[35 ]
[36 ]
[37 ]
[38 ] FV autoantibodies may inhibit the anticoagulant properties of FV,[21 ]
[39 ] including its function as an activated protein C cofactor in inactivation of activated
factor VIII[40 ]
[41 ] and activated factor V.[42 ]
Abnormal Results on Routine Tests
Abnormal Results on Routine Tests
The mean hemoglobin level in patients with AiFVD was 108 ± 67 g/L (median, 96 g/L;
n = 25) which is consistent with the fact that two-thirds (138/200) of AiFVD patients
presented with bleeding symptoms ([Fig. 2B ]). The mean hemoglobin level in the bleeder group was significantly lower than that
in the nonbleeder group (91.7 ± 31.1 vs. 159 ± 117 g/L, p < 0.042; [Table 4 ] top) and was significantly related with the bleeding severity (p < 0.02; [Fig. 3A ]). The mean platelet count was normal (233 ± 122 × 109 /L, n = 91; [Table 4 ] top).
Fig. 3 Correlation between parameters. (A ) Relationship between bleeding severity and hemoglobin (Hb) levels. (B ) Relationship between bleeding severity and prothrombin time (PT) (s). (C ) Relationship between bleeding severity and activated partial thromboplastin time
(aPTT) (s). (D ) Relationship between PT(s) and aPTT(s). (E ) Relationship between factor V activity (FV:C) and PT(s). (F ) Relationship between FV:C and aPTT(s).
Table 4
Results of laboratory tests
Subgroup
Total
Average
SD
Maximum
Minimum
Median
p -Value
Hemoglobin (g/L)
Bleeder
19
91.7
31.1
171
51
82
0.0417
Nonbleeder
6
159
117
398
101
116
Platelet (×109 /L)
Bleeder
67
228
119
694
1
244
0.61
Nonbleeder
23
251
135
520
70
206
FV:C (%)
Bleeder
116
2.9
2.6
14
0
3.0
0.87
Nonbleeder
50
3.0
3.7
26
0
3.0
FV:Ag (%)
Bleeder
5
17.0
9.5
29
7
14.3
0.05
Nonbleeder
2
55.6
34.8
80.2
31
55.6
PT(s)
Bleeder
60
63.0
42.8
301
17.5
57.8
0.0037
Nonbleeder
31
43.2
17.7
104.9
19.2
38.9
PT (INR)
Bleeder
43
7.4
3.7
20.5
2.34
6.8
0.0005
Nonbleeder
40
4.9
2.0
11.8
2.18
4.6
PT (%)
Bleeder
40
12.8
14.8
86.1
2
9
0.0152
Nonbleeder
15
12.6
2.8
20
10
12
aPTT(s)
Bleeder
98
156.1
61.2
331
35
154.1
0.0045
Nonbleeder
54
126.8
51.0
265.1
46.8
132.1
Lupus anticoagulant[a ]
Bleeder
12
1.35
0.26
1.79
0.9
1.36
0.0349
Nonbleeder
10
1.2
0.1
1.45
1.03
1.2
FV Inhibitor (BU/mL)
Bleeder
90
46.0
170.6
1500
0.69
4.95
0.17
Nonbleeder
35
48.8
165.9
969
1
8
Inhibitory pot. (%)[b ]
Bleeder
3
29.0
12.3
41
16.5
29.5
0.65
Nonbleeder
1
–
28
28
28
28
Hepaplastin test (%)
Bleeder
23
99.4
29.0
166
36.9
96.2
0.44
Nonbleeder
12
92.3
22.5
130
49
89.5
Day to recovery
Bleeder
76
33.1
45.6
365
4
20
0.07
Nonbleeder
36
58.3
76.0
360
5
29
Abbreviations: Ag, antigen; aPTT, activated partial thromboplastin time; FV, factor
V; Inhibitory pot.; inhibitory potential; INR, international normalized ratio; PT,
prothrombin time; SD, standard deviation.
Note: Bold p -values indicate statistical significance.
a Numbers indicate ratio of diluted Russell viper venom times for screen (low phospholipid)
and confirm (high phospholipid) assays.
b Calculated as described in the Appendix A section.66
Concomitant Prolongation of Prothrombin Time and Activated Partial Thromboplastin
Time and Decrease in Factor V Activity (FV:C)
Concomitant Prolongation of Prothrombin Time and Activated Partial Thromboplastin
Time and Decrease in Factor V Activity (FV:C)
Both prothrombin time (PT) and activated partial thromboplastin time (aPTT) were considerably
prolonged in AiFVD patients, indicating abnormalities in factors in the common pathway
of the coagulation cascade. Japan's AiFVD bleeders had statistically longer PT and
aPTT than Japan's nonbleeders (63 vs. 43.2 seconds p < 0.004 and 156.1 vs. 126.8 seconds p < 0.005, respectively; [Table 4 ] middle). PT and aPTT were significantly correlated (p < 0.0001; [Fig. 3D ]). As expected, both PT and aPTT were significantly correlated with bleeding severity
(p ≤ 0.0042 and 0.0029, respectively; [Fig. 3B ] and [C ]). Patients with markedly prolonged PT and aPTT should be monitored closely, as highlighted
in a previous review.[8 ] The hepaplastin test (HPT) was performed in 35 AiFVD patients, and the results were
normal (97 ± 26.8%; [Table 4 ] bottom); this is consistent with findings from our previous study, that HPT can
differentiate FV deficiency from other coagulation factor deficiencies.[43 ]
FV:C was severely reduced in AiFVD bleeders (2.9 ± 2.9%, n = 173; [Table 4 ] top) which may be associated with a relatively severe bleeding tendency. However,
no clear correlation was found between FV:C levels and bleeding severity (p = 0.66); the correlation between the FV level and bleeding phenotype was previously
reported to be lost or limited in the group with low FV range (<5%), wherein patients
with equal FV levels may demonstrate different clinical phenotypes.[1 ] In contrast to the reports of a previous study,[8 ] no significant difference was observed in FV:C between the bleeder and nonbleeder
groups (2.9 vs. 3%, p = 0.87) in this study. This was likely because the Japanese commercial laboratories
uniformly report extremely prolonged one-stage clotting time assays for FV activity
as “ < 3%.” Naturally, the FV:C levels were significantly correlated with both PT
and aPTT (p = 0.007 and <0.0001, respectively; [Fig. 3E ] and [F ]).
Although the number of tested samples was limited, the FV:antigen (Ag) level varied
from normal to extremely low in patients with AiFVD (28.0 ± 24.8%, n = 7). Thus, the specific activity of FV (FV:C/FV:Ag) was low (0.03 and 0.09 in two
patients[44 ]
[45 ]). This finding is consistent with the idea that at least some FV molecules are inhibited
by neutralizing anti-FV antibodies (i.e., FV inhibitor), resulting in the formation
of an antigen–antibody complex and a reduction in the specific activity of FV.
Mixing tests based on PT and/or aPTT were performed in 130 patients, 111 showing an
inhibitor pattern, while 19 demonstrated a deficiency pattern (data not shown). A
deficiency pattern can change to an inhibitor pattern in the same patient, as reported
in at least four patients with AiFXIIID.[46 ]
[47 ]
[48 ] This is likely because when an inhibitor potency is low, the FV inhibitor is overwhelmed
by FV in normal plasma used for the mixing test assay. Conversely, when an inhibitor
potency increases, the FV in normal plasma is overwhelmed by patient's inhibitor.
Accordingly, physicians should either examine the FV inhibitor titer and/or repeat
mixing tests when acquired FVD persists for a prolonged period.
The FV inhibitor titer was determined in 127 AiFVD patients and varied widely (46.3 ± 167.3
BU/mL; [Table 4 ] middle). Two patients showed a titer lower than the cut-off value (<0.5 BU/mL) despite
having anti-FV autoantibodies, suggesting the presence of nonneutralizing autoantibodies
and/or FV-antigen-antibody complex[16 ] (unpublished data). Unexpectedly, FV inhibitor titer was not correlated with bleeding
severity, hemoglobin level, PT, aPTT, and FV:C levels (p = 0.88, 0.65, 0.72, 0.61, and 0.64, respectively; data not shown).
The lupus anticoagulant (LA) test was conducted in 48/201 patients with AiFVD, and
results showed that LAs were positive (= abnormal if >1.3), negative, and unmeasurable
in 11, 28, and 9 patients, respectively (data not shown). Anticardiolipin and anti-β2 -glycoprotein antibodies were positive in 6 and 2 patients but negative in 30 and
38 patients, respectively (data not shown). Both LA and anticardiolipin (or anti-β2 -glycoprotein antibody) were positive in four patients which demonstrated isolated
FV deficiency.[49 ]
[50 ]
[51 ]
[52 ] Their FV inhibitors disappeared shortly after initiation of steroid pulse therapy,
administration of prednisolone, and absence of immunosuppressive therapy, respectively.
Thus, the expression of FV inhibitor in these cases was unlikely caused by presence
of an antiphospholipid antibody which is usually resistant to immunosuppressive therapies.
Anti-Factor V Autoantibodies
Anti-Factor V Autoantibodies
Anti-FV autoantibodies could only be detected in 37 patients with AiFVD (data not
shown), mainly because anti-FV autoantibody detection tests are not routinely performed
in Japan. Among these patients, 34 had IgG class and two had IgA class anti-FV autoantibodies.
One patient had anti-FV autoantibodies, although the sample did not show a positive
result for FV inhibitor,[16 ] suggesting the presence of nonneutralizing autoantibodies against FV and/or FV-antigen-antibody
complexes, as reported by Cortier et al.[53 ] Similarly, another patient showed low FV:C level (5%) in absence of an FV inhibitor
and negative mixing test (paper in preparation). All patients with suspected AiFVD
should be immunologically tested for anti-FV autoantibodies since functional FV inhibitor
assays are unable to detect nonneutralizing autoantibodies against FV.
Therefore, some patients with suspected AiFVD have nonneutralizing autoantibodies
against FV when they display extremely low FV:C levels in the absence of an FV inhibitor
and/or negative mixing test.
Hemostatic Treatment
A total of 110 (54.7%) patients with AiFVD received hemostatic therapy, while 29 (14.4%)
patients did not receive ([Fig. 4A ]). Fresh frozen plasma (FFP) was administered in 42.3% of patients with AiFVD but
was ineffective in arresting bleeding (success rates: 11.8 and 15.4% in Japan and
worldwide, respectively; [Table 5 ] top). This is likely because FFP contains only 1 U/mL of FV and is easily overwhelmed
by FV neutralizing autoantibodies. The repeat administration of FFP may be risky in
patients with ongoing bleeding because of possible circulatory overload,[54 ] aggravation of bleeding symptoms,[55 ] and coagulation abnormalities such as an increase in FV inhibitor titer.
Table 5
Efficacy of therapy
Cases
Cases showing efficacy
Administered
Number (%)
Hemostatic treatment
FFP
85
10 (11.8)
Vitamin K
28
0 (0)
Platelet
30
8 (26.7)
Antifibrinolytics
2
0 (0)
PCC
7
2 (28.6)
rVIIa
2
1 (50)
Antibody eradication/reduction therapy
Steroid alone
101
86 (85.1)
St. pulse alone
10
9 (90)
Combined[a ]
9
9 (100)
Cyclophosphamide
4
3 (75)
Combined[a ]
6
3 (50)
Plasma exchange
10
6 (60)
Combined[a ]
9
5 (55.6)
Abbreviations: FFP, fresh frozen plasma; PCC, prothrombin complex concentrates; rVIIa,
recombinant activated factor VII; St., steroid.
a Employed in association with other immunosuppressants; steroid, mostly prednisolone.
Fig. 4 Treatment. (A ) Hemostatic therapy. Fresh frozen plasma (FFP) was most frequently administered to
AiFVD patients, followed by platelet concentrates (platelet). (B ) Antibody-eradication/reduction therapy. Prednisolone (“Steroid”) was most frequently
used in both bleeders and nonbleeders in high doses or pulse administration. Bleeders
often undergo plasma exchange (PE) and rarely receive high-dose intravenous immunoglobulin
(IVIg) treatment. These regimens were employed much less frequently in nonbleeders.
AiFVD, autoimmune factor V deficiency; antifibrinol., antifibrinolytics; Azathiop.,
azathioprine; Cyclophos., cyclophosphamide; n.d., not described; PCC, prothrombin
complex concentrates; rVIIa, activated recombinant factor VII; Withdraw, withdrawal;
Vit. K, vitamin K.
To date, there are no available commercial FV concentrates. As an alternative, platelet
concentrates were used in 14.9% of patients as platelets contain 20% plasma FV.[56 ] The success rates were 26.7% (8/30) and 68.8% (11/16) in Japan and worldwide, respectively.
The dose and frequency of platelet transfusion may differ from patient to patient
and country to country; thus, its hemostatic effect may also differ between Japan
and the rest of the world. Vitamin K was administered in 13.9% of patients in Japan,
but it had no effect on bleeding, confirming that vitamin-K deficiency was not the
cause of bleeding in patients with AiFVD.
Prothrombin complex concentrates (PCCs) and recombinant-activated factor VII (rFVIIa)
were used in 3.5 and 1% patients with AiFVD, respectively. Antifibrinolytic agents,
such as tranexamic acid, were used in only 1% of patients with AiFVD; this may be
because the patients did not present a hyperfibrinolytic state. Although variable
success rates of PCCs have been reported in Japan and worldwide (28.6 and 80%, respectively),
the number of patients (seven and five cases, respectively) was relatively small to
draw any meaningful conclusion.
Hemostatic medicine was not administered in 14.4% of Japan's AiFVD patients or was
not described (ND) in 30.8% of patients, mainly because these patients did not show
any bleeding symptoms or their bleeding symptoms were not severe.
Antibody Eradication/Reduction Therapy
Antibody Eradication/Reduction Therapy
In total, 73% of Japanese patients with AiFVD initially received prednisolone with
or without other associated treatments (overall prednisolone) for antibody eradication
([Fig. 4B ]), while 13.5% patients received a pulse prednisolone regimen, with success rates
of 74.7 and 70.8%, respectively ([Table 5 ], bottom). The success rate of prednisolone alone was higher (85.4%) than that of
the overall prednisolone. Moreover, pulse prednisolone alone had 90% success rate,
while pulse prednisolone combined with prednisolone had 100% success rate ([Table 5 ], bottom). Therefore, physicians should select either the pulse prednisolone alone
regimen or the combined regimen as a first-line antibody eradication therapy whenever
possible.
Cyclophosphamide and rituximab, an anti-CD20 monoclonal antibody, were used only in
5.1 and 1.5% of patients with AiFVD, respectively, possibly because they are not approved
for treatment of AiFVDs by the Japanese public health insurance system.
Plasma exchange (PE) and high-dose intravenous immunoglobulin (IVIg) therapy were
administered to 17 and 1% of patients with AiFVD in Japan, respectively. Although
PE showed a success rate of 60% in reducing the level of antibodies, the effect of
PE is theoretically transient in nature, as indicated by the temporary success rate
of 50%. None of the Japan's patients with AiFVD underwent antibody adsorption therapy.
Moreover, IVIg was rarely used in AiFVD patients, as reported in AiFXIIID patients.
[24 ] Furthermore, IVIg is not recommended by the Japanese Society of Thrombosis and Hemostasis
for management of AiFVIIID[57 ] due to the variable effects of this treatment.[23 ]
Antibody-directed therapy was not performed in 17.2% of Japan's patients with AiFVD,
while it was not performed in 5 and 11.8% of patients with AiFVD and AiFXIIID, respectively[23 ]
[24 ]). It is likely that some patients with AiFVD did not need this treatment or that
their acquired FVD was transient and self-limiting.
Prognosis
Outcomes were reported in 192 (95.5%) patients with AiFVD ([Fig. 5A ]); of the 147 (73.5%) patients who achieved resolution, 21 (10.5%) showed spontaneous
resolution. Similar overall and spontaneous resolution rates were previously reported
in 69.2% (54/78) and 15.4% (12/54) of patients with FV inhibitor, respectively.[9 ] A total of 10 (5%) patients showed disease deterioration after reinitiating the
antibiotic therapy,[29 ] administering FFP infusion,[55 ] or tapering the dose of prednisolone.[49 ]
[54 ]
[58 ]
[59 ]
Fig. 5 Prognosis. (A ) Outcome of AiFVD. The majority of patients recovered regardless of whether they
are categorized as bleeders or nonbleeders. Spontaneous (Spont.) recovery occurred
more frequently in nonbleeders than in bleeders (p < 0.0002), while all-cause death was more frequent in bleeders than in non-bleeders
(p = 0.027). (B ) Causes of death. Hemorrhagic (Hemorrh.) death occurred only in bleeders, while septic
death occurred more frequently in non-bleeders. AiFVD, autoimmune factor V deficiency;
Exacerb., exacerbation; Fail., failure; MOF, multiple organ failure; n.d., not described.
With regard to the all-cause mortality among the 201 patients with AiFVD, 30 (15%
of 201 cases) patients died ([Fig. 5B ]); this mortality rate (15%) was 50% lower than that reported in a previous worldwide
review article (23/75, 30.7%[8 ]). In addition, 13 (6.5% of 201 cases) Japan's AiFVD patients died due to hemorrhagic
complications such as brain hemorrhage, peritoneal hemorrhage, hemorrhagic shock due
to systemic bleedings or gastric bleeding, etc.; this mortality rate (6.5%) was approximately
50% lower than that reported in a previous worldwide review article (9/75, 12%[8 ]). The major differences between the present and the previous review articles are
as follows: sample size: 201 cases versus 75 cases[8 ]; study population: single country versus worldwide[8 ]; and study period: April 2011 to March 2021 versus 1950 to June 2008.[8 ] A total of 69 Japanese patients were diagnosed with AiFVD before 2010, while 132
patients were diagnosed with AiFVD after 2011. Because the numbers of both reported
and recovered AiFVD cases have approximately two-fold increase over the past 10 years,
there may have been a significant progress in the awareness and management of this
disease, that is, prompt diagnosis was performed and appropriate treatment was provided.
However, the exact reason why Japanese patients with AiFVD have more favorable outcome
remains unclear.
The rate of hemorrhagic death (6.5%) was not higher in the AiFVD patients than in
the AiFVIIID and AiFXIIID patients (9.1% [13/143] vs. 14% [13/93][23 ]
[24 ]).
Relapse was reported in 11.4% (22/201) of AiFVD patients. Two patients experienced
relapse 4 years after remission.[12 ]
[60 ] Thus, long-term observation of patients with AiFVD is necessary. In particular,
relapse tends to occur when patients undergo surgery and/or receive antibiotic therapy.[12 ]
[60 ]
The median time and range to recovery were 25 days and 4 to 365 (40.8 ± 57.6) days
in Japan's AiFVD patients, respectively ([Table 4 ] bottom). In previous worldwide cases, the median time to recovery was 6 weeks (42
days) and the range was 1 week to 29 months (7–890 days).[9 ] This difference can be explained by the recent advances in management of AiFVD.
The mean time to recovery was shorter in bleeders than in nonbleeders (33.1 ± 45.6
vs. 58.3 ± 76 days; p < 0.036), probably because the former group more frequently received antibody eradication
therapies than the latter group (81.2 vs. 50.8%, p < 0.0001 for prednisolone therapy and 16.7 vs. 5.1%, p = 0.013 for the pulse-steroid regimen). In addition, significantly fewer patients
in the bleeder group did not receive any antibody-targeted therapies than in the nonbleeder
group (10.1 vs. 33.9%, p < 0.0001). The prognosis of AiFVD may be related to the underlying diseases, as previously
hypothesized.[8 ]
[9 ]
Limitations
This study has several limitations. First, out literature search only included PubMed
and ICHUSHI databases, not Scopus, Google Scholar, etc. Second, since the observation
periods of patients were generally short, it is difficult to draw any conclusion regarding
their long-term prognosis. Third, reports in the abstract form, yielding limited information,
were also included. Fourth, many AiFVD-suspected patients were excluded due to the
following reasons: they readily responded to FFP administration, their prolonged PT
and aPTT were completely corrected in mixing test studies, they were judged to have
congenital/hereditary FVD by authors of individual reports, and/or authors did not
rule out FV deficiency.[61 ] Fifth, the mechanism of AiFVD was not explored, as it was beyond the scope of this
review.
To overcome some of these limitations, the JCRG started using a database-based registry
system, known as the Japanese DID platform (started on February 1, 2021). This system
will allow long-term follow-up studies to be conducted to determine the prognosis
more precisely and establish optimal second-line therapies for AiFVD.
Conclusion
Because the total number of Japan's AiFVD patients confirmed by JCRG has been steadily
increasing (100 in 2017,[62 ] 164 in 2019,[6 ] and 201 in 2021 [present study]), this review provides the most up-to-date comprehensive
knowledge regarding the Japanese patients with AiFVD. This review had the following
findings: (1) based on a population size of 120 million, the incidence of AiFVD in
Japan is estimated to be at least 0.04 per million persons per year which is similar
to that in Singapore (0.09[8 ]) and less than that in Australia (0.29,[20 ] but this report included bovine thrombin–associated cases); (2) the mean age at
onset of AiFVD is similar to that of other AiCFDs; (3) AiFVD is predominant among
men than women; (4) AiFVD is more severe than previously thought; (5) some patients
with AiFVD have very low FV inhibitor titers, suggesting the presence of nonneutralizing
and hyperclearance-type anti-FV autoantibodies; (6) as nonneutralizing anti-FV autoantibodies
can be overlooked in functional FV inhibitor assays, immunological anti-FV antibody
detection is recommended, especially in the absence of FV inhibitor; (7) the hemorrhagic
death rate was lower than that reported in previous studies; and (8) the prognosis
of AiFVD seemed fairly good.
We hope that the findings of this review will help clinicians face challenges in diagnosing
and treating patients with AiFVD.
Appendix A
The Japanese Collaborative Research Group's Nationwide Survey
The Japanese Collaborative Research Group (JCRG) survey, its integrated/unified laboratory
tests, and our experimental examinations were approved by the institutional review
board of Yamagata University and the individual hospitals of each attending physician.
Written informed consent was obtained from all participants in accordance with the
Declaration of Helsinki.
When a bleeding patient visits a hospital, the physician in charge of the patient
consults a hematologist, who then consults the patient and/or reports the patient
to our JCRG headquarters. Subsequently, blood samples were collected and sent to a
commercial laboratory (SRL Ltd., Hachioji, Japan) and the JCRG headquarters.
Literature Searches
We also performed extensive literature searches at least twice a year in the PubMed
(in English) and ICHUSHI (in Japanese) databases. The following search terms were
used: (immune) OR (antibody) OR (autoimmune) OR (autoantibody) AND (factor V) OR (factor
5) OR (proaccelerin) OR (labile factor) AND (deficiency) OR (inhibitor) for PubMed
and similar Japanese terms for ICHUSHI.
Inclusion and Exclusion
Suspected autoimmune FV deficiency (AiFVD) cases were selected based on the governmental
diagnostic criteria of DID (code: 288–4). These criteria were previously enacted by
the Japanese MHLW ([Table 1 ]) and were analogous to the 2015 criterion for diagnosing AiFXIIID:[10 ] (1) “definite” diagnosis, when anti-factor (F)-V autoantibodies are detected in
blood sample; (2) “probable” diagnosis, when the sample is positive for an FV inhibitor
(termed “probable-2” in this article); or (3) when the 1:1 mixing test and/or cross-mixing
test using patient's and control's plasma samples (based on prothrombin time [PT]
and/or activated partial thromboplastin time [aPTT]) presents an inhibitory but not
deficient pattern in the presence of isolated decreased FV activity (FV:C) (termed
as “probable-1” in this article); and (4) the remaining suspected AiFVD patients were
included in the “possible” group. Patients with congenital and other distinguished
acquired FV deficiencies (acFVDs), such as vitamin-K deficiency, FV deficiency due
to warfarin treatment, severe liver dysfunction, disseminated intravascular coagulopathy,
and other autoimmune coagulation factor deficiencies (AiCFDs), were carefully excluded.
AiFVD is not considered congenital FVD if immunosuppressive therapy resulted in normal
FV levels or if the patient achieved spontaneous resolution.
Data Collection
The following data were collected from the included patients: sex, age at presentation,
bleeding symptoms, underlying diseases, hemoglobin levels, platelet counts, FV:C levels,
FV antigen (FV:Ag) levels, presence of anti-FV autoantibody, FV inhibitor levels,
and mixing test results based on PT, aPTT, or both. In addition, clinical data, such
as data on hemostatic medicines and immunosuppressive agents used, prognosis, and
days to recovery (if recovered), were collected. The FV inhibitory potential (pot.)
was calculated by subtracting the actual residual FV activity of the 1:1 mixture from
the calculated FV activity of the mixture (average FV activity of patient and healthy
control) as previously described for FXIII inhibitory potential.[63 ]
Statistical Analysis
Categorical and continuous variables were compared between the two groups using the
Pearson chi-square test and the Kruskal–Wallis test, respectively. Statistical analyses
were performed using the JMP software (version 12.2.0; SAS Institute, Cary, NC). Statistical
significance was set at p < 0.05.
For statistical purposes, FV:C levels or clotting time (PT or aPTT) lower than or
higher than the indicated values were considered as the indicated values per se, for
example, <1% as 1% or >100 seconds as 100 seconds, respectively.
Erratum: An erratum has been published for this aricle (DOI: 10.1055/s-0042-1759569).
