Activated FVII levels in factor VII Padua (Arg304Gln) coagulation disorder and in true factor VII deficiency: a study in homozygotes and heterozygotes

Abstract

Congenital FVII deficiency is usually subdivided into two forms: type I and type II. Type I is characterized by a concomitant deficiency of FVII activity and FVII antigen (true deficiency). Type II is characterized by a discrepancy between FVII activity which is always low and FVII antigen which may be normal, near normal, or reduced. Thromboplastins of different origins may show a discrepant behaviour towards type II FVII deficiencies. The abnormal factor VII present in these forms may, in fact show, different levels of activity, according to the thromboplastin used in the assay system. Typical of these variants is the Arg304Gln mutation (know as FVII Padua). In this variant, FVII level is low when rabbit brain thromboplastin is used, whereas the level is perfectly normal when ox-brain thromboplastin is employed. Intermediate levels are obtained if human placenta or human recombinant is used. Since ox-brain thromboplastin is very sensitive to activated FVII, the normal FVII levels obtained in FVII Padua could be due to abnormally high circulating levels of activated FVII. The purpose of the present paper was to investigate the level of activated FVII present in homozygotes and heterozygotes with FVII Padua. For comparison, a group of patients with type I or ‘true’ deficiency was also investigated. A group of 21 normal patients served as controls. The activated FVII level found in FVII Padua was 8·4 and 41·0 mU/ml for homozygotes and heterozygotes, respectively. The level found in homozygous true deficiency was unassayable, whereas that found in heterozygotes was 36·2 mU/ml. The level found in the control population was 64·9 mU/ml in agreement with other reports. The low levels of activated FVIIa found in homozygotes with FVII Padua indicate that the normal FVII activity found with ox-brain thromboplastin cannot be attributed to higher than normal circulating levels of FVIIa.

Keywords:

• FVII
• Activated FVII
• Tissue factor
• Thromboplastin

Introduction

The reaction between tissue factor (TF) and FVII has received considerably attention during the last two decades. This is mainly due to the importance attributed to the TF–FVII complex in the initiation of blood coagulation via a direct activation of FIX as well as FX.1,2

The relation between total FVII level and activated FVII has not been completely clarified. It is known that only about 1% of total FVII circulates in an activated form in balance with TF pathway inhibitor levels.3,4 The usual levels in the normal population are in the range 50–170 mU/ml.5 Information about the levels of activated factor VII found in congenital FVII deficiency is very scanty.5^–9

This is a particularly true for type 2 defects, namely, for those cases of FVII deficiency in which there is a clear discrepancy between FVII activity and FVII antigen.9

Factor VII Padua is a type II defect due to an Arg304Gln mutation in exon8 and characterized by a sharp discrepancy in FVII activity levels according to the tissue thromboplastins used.10 This is illustrated when a rabbit brain thromboplastin is used in the assay system when FVII results are low (4–10% of normal), whereas they appear normal when an ox-brain preparation is used.10 Thromboplastins of human origin (placenta) or recombinant human thromboplastin yield levels of about 30–40% of normal.11 Bleeding manifestations are mild and, occasionally, absent. After our original description, in the late 1970s and early 1980s, about 25 additional homozygous cases have been reported from several countries, together with many heterozygotes and compound heterozygotes.12^–14

The purpose of the present study is to investigate the level of activated FVII found in both homozygous and heterozygous patients with this disorder and, for comparison in a few patients with ‘true’ FVII deficiency.

Patients, Materials and Methods

Six homozygous and 14 heterozygous patients with the Arg304Gln mutation (FVII Padua) together, with three homozygous and 11 heterozygous patients with ‘true’ severe FVII deficiency, were the object of the present investigation. All these patients were diagnosed and studied in Padua during the years 1978–2002. Twenty-one normal subjects (10 male and 11 female) aged 20–60 years were used as controls.

The main clinical and laboratory features of five of the homozygous patients with FVII Padua have been previously reported by us in detail.10,12,13

The sixth patient is a 64-year-old female first studied by us in 1995. She was a native of the same valley where all other patients came from (Piave River Valley in north-eastern Italy). The patient came to the attention of a local gynaecologist because of vaginal bleeding from the uterine stump. She had undergone hysterectomy about 10 years earlier because of multiple fibromas. At time of surgery, the patient was reported to have shown only mild bleeding that required the administration of 1 unit of fresh frozen plasma. The vaginal bleeding for the uterine stump was also mild and consisted mainly of a continuous oozing. A local cauterization was sufficient in stopping the bleeding, and no substitution therapy was required.

Since histology was negative, a tentative diagnosis of inflammation of the stump was made. Previously, she had no bleeding manifestations, except for a mild post-partum hemorrhage (second delivery) which, however, did not require blood transfusion. Her first delivery was uneventful and she had no history of thrombosis.

The coagulation pattern was typical for factor VII Padua, namely, prolonged prothrombin time (PT), normal partial thromboplastin time (PTT), and normal Thrombotest; the FVII level was low (5% of normal) using rabbit brain thromboplastin but was normal (105% of normal) using ox-brain thomboplastin. Genetic analysis revealed the typical Arg304Gln mutation in exon 8. Both children were found to be heterozygous for the abnormality with FVII activity levels around 55% of normal but normal antigen.

The three homozygous patients with true factor VII deficiency have also been studied by us in the past. The heterozygotes for the Arg304Gln FVII abnormality belong to the families of the homozygotes. The same is true for the heterozygotes for ‘true’ FVII deficiency.

Routine basis tests such as PT, PTT, and Thrombotest (Nygaard Laboratories) and total factor VII assays were carried out as previously reported.10

The thromboplastins used were rabbit brain (Ortho or Dade-Behring Laboratories), human placenta (Dade-Behring), human recombinant (Baxter, Dade-Behring), and ox-brain thromboplastin (kindly prepared by us by Stago Laboratoies and Nygaard Laboratories).

Immunological FVII evaluation was carried out by electro-immunoassay using an antigen kindly supplied by Behring Laboratories and by an Elisa method (Asserachrom Stago Laboratories).

In the electroimmunoassay, the concentration of antiserum in agarose was 4% with an overnight migration in a refrigerated electrophoresis chamber under 4 V/cm electrical power.

Activated FVII was assayed using the kit (Statclot) produced by Stago Laboratories, Asnieres, France. The calibration curve was obtained by using freeze-dried preparations containing known amounts of human recombinant FVIIa. The standard is equivalent to 89/688 World Heath Organization standard. The shorter clotting time, the higher FVII levels, and the results obtained with the two control reagents were within the range established by the manufacturers. All female control subjects and patients, at time of study, were not pregnant and were not taking oral contraceptives. None of the normal subject had ever had either bleeding or thrombotic episodes.

All tests were carried out on fresh plasma or on deep frozen plasma kept for less than 60 days at −40°C. In all instances, blood sample were collected without any difficulty in plastic tubes.

Results

The mean activated FVII level in our normal population was 64·8 mU/ml (range: 36–135 mU/ml). Factor VIIa level in homozygous patients with the FVII Padua abnormality was 8·4 mU/ml (range: 5–12 mU/ml), while that found in heterozygotes was 41 mU/ml (range: 25–90 mU/ml) (Table 1).

Table 1. Main coagulation parameters in six homozygotes with FVII Padua and in three cases homozygous for ‘true’ factor VII deficiency

		Factor VII activity (%)
	PT (second)	Rabbit brain	Human placenta	Recombinant human	Ox-brain	Factor VII antigen	Activated FVII levels (mU/ml)	Comments
FVII Padua
Case 1	30·2	8	36	30	105	95	5
Case 2	30·0	9	32	20	110	110	5·5
Case 3	32·5	10	36	32	100	105	12
Case 4	34·0	11	34	29	100	110	12
Case 5	31·0	8	33	30	90	120	10
Case 6	31·6	7	29	25	95	100	6
True FVII deficiency
Case 1	48	1	1·2	1	1	Traces	Traces	Unassailable
Case 2	42	1·5	1·5	2	1·5	Traces	Traces	Unassailable
Case 3	50	1	1	1	1	Traces	Traces	Unassailable
Normal values	13–15	80–120	80–120	80–120	80–120	80–120	36–135

Table 2. Comparison of levels of activated FVII observed in the different groups of patients investigated

Defect	No. of cases (M/F)	Mean value	Range	Comments
Homozygous FVII Padua	6 (1/5)	8·4	5–12	P>0·01 with all other groups for which calculation could be carried out
Heterozygous FVII Padua	14 (8/6)	41·0	25–90
True FVII deficiency	3 (1/2)	Traces	…	Unassailable traces
Heterozygous for True deficiency	11 (6/5)	36·2	20–75
Normal controls	21 (10/11)	64·8	36–135

The levels were unassayable in the three cases with type I FVII deficiency.

Mean FVII activated level in the heterozygous patients with true deficiency was 36·2 mU/ml (range: 20–75 mU/ml). Mean FVII activities in the homozygous population using rabbit brain, human placenta, or ox-brain thromboplastin were 8·8%, 33·3%, and 100·2%, respectively.

Mean FV antigen was 106·6% of normal in the homozygous and 98·5% in the heterozygous. No appreciable FVII antigen was found in the homozygous for true deficiency. FVII activities in the heterozygous population using the same three thromboplastins were 59·2%, 57·5%, and 97·5%, respectively, in the case of FVII Padua, and 46·2%, 49·5%, and 48·6%, respectively, in the case of true deficiency. Due to the homogeneity of patients who came all from the same valley in north-eastern Italy, consistent results were obtained in every assay, regardless of the time of assay.

Discussion

The diagnosis of congenital factor VII defect is based on the presence of a prolonged PT and PT-derived clotting tests, mainly PP test, Normotest, and Thrombotest. Specific FVII levels in type I congenital deficiency are low and comparable to antigen levels. By contrast, in type 2 forms, antigen levels are normal or near normal, while FVII activity is low. Variably increased activated FVII levels have been reported in disseminated intravascular coagulopathy, during oral contraceptive (OC) therapy and, occasionally, in arterial diseases.4,6,7,15^–17 However, the clinical significance of these changes in activated FVII levels is still to be clarified and the use of the assay is limited to research use and in the follow-up of haemophilic patients with inhibitors being treated with FVIIa concentrates.18^–22 It is surprising that there are no data reporting the levels of activated factor VII present in patients with congenital FVII deficiencies and even for acquired conditions, such as coumarin therapy, data are scanty and still equivocal being described as normal7 or decreased.6 We suspect that it is likely that different levels of anticoagulation of the tested patients could explain the discrepancies.

The levels obtained by us in the control group was similar to that observed by others and underscore the reliability of the method used.5,6 The small variations observed in the normal ranges are acceptable.

Ox-brain thromboplastin is particularly sensitive to the presence of activated factor VII.23,24 Since FVII Padua shows normal levels in clotting assays which employed ox-brain thromboplastin (specific factor VII assay or a PT-like global test such as Thrombotest), it could have been speculated that the normal levels seen in this variant could be due to increased levels of circulating activated FVII. Although this explanation is implausible, it was felt important to confirm or deny it.

The potential correlation between activated FVII in these patients could also have been suggested by the observation that one of these six homozygous patients presented a diffuse great saphenous vein thrombosis. Furthermore, two additional patients with this defect and a deep vein thrombosis have been reported by others.25,26

Activated FVII levels have been postulated to be associated with an increased risk for thrombosis. However, the low FVIIa levels found in all patients have excluded this hypothesis as the basis for the occurrence of the thrombotic manifestations, which was probably due, as in other cases of FVII deficiency, to the presence of concomitant risk factors. The proposita, for example, had varices in the affected leg.

The low FVIIa levels found both in homozygous and heterozygous patients with the Arg304Glu mutation clearly indicate that the normal clotting results obtained with ox-brain thromboplastin are not due to the presence of elevated levels of activated FVII. The levels obtained in patients with the Arg304Gln were low but still assayable. On the contrary, no significant levels of activated FVII were present in cases of true deficiency. This is not surprising, if one takes into account that the FVIIc activity present in FVII Padua is about 8% of normal, whereas in true deficiency, the activity is only around one or less than 1% of normal. There seems to be a fair correlation between FVIIc and activated FVII. It is commonly accepted that only about 1% of FVIIc activity circulates in an activated form.3 The data that have been presented seem to support this since even individuals who have FVIIc levels around 50% or 60% of normal, such as the heterozygotes presented here, have proportionally reduced levels of activated FVII (Table 2).

The Arg304Glu mutation is located on exon 8 in an area that appears to be crucial for binding and activation by tissue factor. These are at least two other FVII type II abnormalities in a nearby area, namely, the Cys310Phe and Gly331Ser mutations.27,28 Unfortunately, these two abnormalities which are also cross-reacting material-positive, have not been tested with an ox-brain thromboplastin. It would be interesting to find out whether these abnormalities behave as FVII Padua towards tissue thromboplastins.

Ox-brain thromboplastin has been tested by other authors in additional cases of Arg304Gln mutation and the results obtained by us have been repeatedly confirmed.25,26 To our knowledge, in none of these additional cases, the level of activated FVII has been evaluated. However, there is no reason to suspect that different negative results might be obtained.

A final comment concerns the observation that at least four homozygotes for factor VII Padua have presented with thrombotic events.10,25,26,29 Furthermore, another patient with the same clotting pattern, but without the confirmation of molecular confirmation of the abnormality, has been presented by Triplett et al.30 Unfortunately, the activated FVII level for these patients is available only in one case (proband no. 1 of this report). The level observed was 5 mU/ml, which was the lowest value observed in our group of patients. Finally, it has been recently demonstrated that associated prothrombotic conditions were present in most of these patients with FVII deficiency and thrombosis.31

In conclusion, it appears that activated FVII assays may sometimes be useful in cases with potential bleeding conditions and as an additional measure of clotting activity in patients receiving activated FVII concentrates or in the evaluation of prothombotic settings.

This study was supported in part by the Associazione Emofilia ed altre Coagulopatie delle Tre Venezie.