A Cell-based Model of Coagulation and the Role of Factor VIIa - Blood Review 2003

8/13/2019 A Cell-based Model of Coagulation and the Role of Factor VIIa - Blood Review 2003

1/5

A cell-based model ofcoagulation and the role offactor VllaMaureane HoffmanDepartment of Pathology, Duke University Medical Center, Durham. NorthCarol ina. USA

Abstract Our cell-based m odel of haemostasis replaces thetraditi onal cascade hypothesis, and proposes that coagul ationtakes place on different cell surfaces in three ov erlapping steps:initiation, amplification, and propagation. In highlighting the im-portance of cellular co ntrol during coagula tion, the cell-basedmodel allows a more thorough understanding of how haemosta-sis works in vivo, and sheds light on the pathoph ysiologica l mech-anisms behin d certain coagulation disorders. For instance, thismodel proposes that haemophilia involves a failure of platelet-surface FXa generati on, lead ing to a lack of platelet-surfacethrombin production. Our data suggest that high-dose FVlla isable to bind weakly to activated platelets, independently of tissuefactor, in otder to generate sufficient amounts of FXa to supporta burst bf thrombin generation in the absence of FIXa/FVllla.The considerable success of high-dose recombina nt FVll a (rFVlla;NovoSeven@, Novo Nordisk, Copenh agen, Denmark) as a ther-apy for patients with haemophilia and inhibitors has led to its usein a growing number of alternative indications. We believe thateven in the presence of the FIXa/FVllla complex, rFVlla may beable to enhance both FXa and FlXa levels on the surface of acti-vated platelets , thus increasing the production of thrombin .0 2003 Elsevie r Science Ltd. Al l rights reserved.

KEY WORDS: coagulation; cell-based m odel; haemostasis; re-combinant factor Vll a; haemophilia

INTRODUCTION **$Q

he classical model of coagulation descr ibes a cas-cade of reactions involving activation of var ious clot-a ting factors along either an extrinsic or an intrinsic

pathway. According to this model, st imulation of ei ther ofthese two pa thways can resul t in the production of a largeamount of thrombin and subsequent formation of a f ibrinclot (Fig. 1).

Howeve r, al though this cascade paradigm supp orts labo-ratory evaluation of coagulation disorders and demon stratesthe interactions between coagulation factors, i t does notadequately explain the mecha nisms leading to haem ostasisin viva. Furthermore, i t does not provide a great deal ofinformation regarding the pathophysiology of the haemo-static system. In part icular , the model does not explain whycertain categor ies of patients demonstrate a haemorrhagictendency; nor does i t faci l itate accurare prediction of whichpatients wi l l actual ly bleed. For instance, patients with a defi -ciency of factor XII @X II) , high-molecular-weight kininogen,or prekallikrein do not presen t w ith a bleeding tende ncy de-

Intrinsic p8thW8yfactor XIIHMWK4factor Xl- factor XIa

ifactor IX- factor tXi3factor VllIa

Extrinsic pathway

factor VitaTissue facbr

factor X 1 = 4factor xa - facmr xfactor Va

fibrinogen --+ fibrinFig. I. The cascade hypothesis: intr insic and extr insic pathways.

: spi te a prolonged part ial thromboplastin t ime (FIT), which: indicates a disturbance in the functional activ i ty of the in-: tr insic pathw ay.* In contrast, an increased predisposit ion to: haemorrhagic r isk may be present in patients deficient in_ FXI. The degree of prolongation of the FIT in this disor-i der, however, does not necessar i ly predict the extent of the; bleeding tendency, which is typical ly less severe than that: observed in haemophi l ia.

The cascade hypothesis cannot account for the varying: degrees of haemorrhagic tendency and diverse cl inical obser-: vations that resul t from deficiencies of di fferent compone nts: of the two pathwa ys. In an attemp t to explore the process: of haem ostasis from new angles, we developed exper imental

and conceptual models that would al low us to test hy-potheses in a biochemical or Ed r&o system . This, in turn,f would increase understanding of how the normal haemo-

: static sy stem actual ly wo rks in &JO. In addi t ion, we wished: to explore the mechan ism of haem ostasis in haemophi l iac: patients. Such patients have a normal prothrombin t ime: (FT), which measures activi ty o f the extr insic pathway, de-: spi te a prolonged FIT and a pronounced bleeding tendency.: W hy, then, does the extr insic pathway fai l to compen sate forI the dysfunctional intr insic pathway? In other words, why doj haem ophiliacs bleed?

We have developed a cel l-based model of haemo stasis that: wi l l replace the classical model of the coagulation cascade.3: This cel l -based model empha sises the interaction of clott ing: factors with speci f ic cel l surfaces4 and appears to be able to: shed light on many of the unresolved issues highlighted by: the tradit ional cascade theory.

; THE CELL-BASED MODEL OF HAEMOSTASIS

The f i rst step in our investigation was to establ ish an in vitroexper imental syste m incorporating platelets and plasma con-centrations of var ious clott ing factors and coagulation in-hibi tors. A cel lular source of t issue factor (TF ) was consid-ered to be essential , and inclusion of TF-bear ing mon ocytes

0 20 03 Elsevler Sdence Ltd. All r ights reserved. Blood Reviews (2003) 17, 51-55


2/5


3/5

Ila

(4 IXa

b)

TF

+ FreevWF

IIIII--_______ ATFPI = t issue factor pathway inhibitor.

Fig. 3. The ceil-based mode l of haemostasis: (a) init iation, (b) amplif ication , (c) propagation .

coagu lation, as initially postulate d by the original cas cadehypothesis.Al though insuff ic ient to resul t in clot formation by i tsel f,the smal l amount of thrombin generated at the surface ofTF-bearing cells during the initiation phase is essen tial inampl i fy ing the procoagulant signal . At ;he end of the ampl i -f ication p hase, platelets activated by this l imited amount ofthrombin are clad in activated cofactors and FXIa, and theprocess of haemo stasis move s into the propagation phase.

PropagationDuring propagation, FlXa combines with i ts cofactor, FV IIIa,on the surface of activated platelets. Some of the requiredFIXa is produced on the surface of TF-bear ing cel ls byTF/FV IIa, and can di ffuse to the activated platelets as i t isnot inhibi ted by t issue factor pathway inhibi tor (TPPI) , andis only slowly inhibi ted by anti thrombln III (ATIII) . Facto rIXa can also be produced on the platelet surface by FXIa.

Once formed, the FIXa/FV IIIa complex activates FX to

0 200 3 Elsevier Science ltd. All r ights reserved. Blood Reviews (2003) I7, S/-S5


4/5

FXa, which immediately combines with i ts cofactor (Fig. 3~).The FXa/FVa com plex then converts large amoun ts of pro-thrombin to thrombin, resul ting in the cleavage of f ibrinogento f ibrin m onom ers, which polymerise to consol idate the ini-tial platelet plug into a stable fibrin clot.The cel l -based model therefore places an empha sis onthe cellular con trol o f coagu lation, and is subse quen tly ableto explain some cl inical aspects of haemo stasis that theclassical cascade hypothesis cannot.4 It al lows a more thor-ough understanding of how the coagulation process worksin viva, and provides a greater degree of consistency withclinical observation s of various coagulation disorders,

WHY DO HAEMOPHILIACS BLEED?When compared to the tradi tional cascade theory, the cel l -based model faci l i tates a greater understanding of the patho-physiological mecha nisms leading to haemophi l ia. For in-stance, the cascade model does not explain why the extr in-sic pathway appears unable to produce suff ic ient amoun ts ofFX to at least part ially compen sate for a deficiency of FVIIIor FIX. In other words, why does activation of FX by theTP/FV IIa complex fai l to substi tute for the FXa that wouldnormal ly be generated by FIXa/PVIIIa?

The cel l-based model does not sugg est that FXa genera-t ion by the TF/FV IIa complex is insuff ic ient in haemophi l ia,but that i t occurs on the wrong cel l surface. The FIXa/FV IIIacomplex activates FX on the surface of platelets dur ing thepropagation phase, whereas TF/PV IIa can only produce FXaon the surface of the TF-bear ing cel l . The FXa producedon the TF-bearing cel l is unable to move to the activatedplatelet surface, as there exist two very eff ic ient inhibitors ofFXa in the plasma: TFPI and ATIII. At normal plasma levels,both TFPI and ATIII inhibi t FXa so rapidly and effectivelythat the hal f- li fe of FXa is 1 minute or less in the f luidphase.2 Therefore, FXa that remains at the TF-bear ing cel lis relatively protected from inhibit ion, whereas any FXa thatdi ffuses from the surface is rapidly inhibited.

Accordingly, the cel l -based model proposes thathaem ophilia is spec ifically a failure of platelet-surfac e FXageneration, which resul ts in a lack of platelet-surface throm-bin production. Haemophi l iac patients demonstrate rela-tively normal initiation and amp lification phase s of coagula-tion, and so are able to form an initial platelet plug a t thebleeding si te, but they cannot generate the burst of throm-bin at the platelet surface that is necessary to stabi l ise theinitial plug into a fibrin clot.

HOW DOES HIGH-DOSE FVBa ENHANCEHAEMOSTASIS IN HAEMOPHJLIA?As discussed above, the cel l -based model of coagulationsuggests that the total amou nt of FXa produced is lessimportant than the location in which i t is generated.* Webel ieve that FXa mu st be formed on the platelet surfaceby FIXa/PV IIIa, in close proximity to w, in order to beincorporated into prothrombinase complexe s. This meansthat the TF/FV IIa complex cannot compensa te for a lack of

FIX/F VIII, as i t make s FXa in the wrong place. If this is thecase, then eff ic ient haemophi l ia treatment mus t involve therestoration of FXa generation on the platelet surface.

Our data imply that high-dose FVIIa is able to do just that- i t can enhance haem ostasis in haemophi l iacs by activatingsuff ic ient PX on the surface of activated platelets to supporta burst o f thrombin generation.

Or iginaIly, our group favoured a TFdependent mecha nismin which high doses of FVIIa could dr ive the TF pathwayin haemophi l iacs, enhancing the performance of the extr in-sic pathway and therefore producing haem ostasis. I t is wel lrecognised that PVIIa exhibi ts very l i tt le proteolyt ic activ i tyin the absence of TE Howeve r, the doses of FVIIa required toachieve coagulation in haemophi l iacs produced plasma lev-els that were several orders of magnitude greater than the &for binding of PVIIa to TF, leading some researchers to sug-gest that FVIIa is unl ikely to work through a TFdependentmechanism.

We used our exper imental model to determine howhigh-dose FVIIa supports haem ostasis in patients withhaemophi l ia. I t was found that FVIIa binds weak ly to ac-t ivated platelets, even though platelets do not carry TEOnce bound to the platelet, PVIIa generates a smal l amountof PXa, leading to the production of a l imited amount ofthrombin on the platelet surface. These f indings are alsoconsistent wi th our conceptual model of coagulation, whichpostulates that platelet-surface FXa generation is requiredfor the assemb ly of the prothrombinase complex and subse-quent thrombin generation. Furthermore, the concentrationof FVIIa required to produce detectable thrombin generationcorrelates with the lowest concentration of PVIIa necessaryfor clinical eff icacy in haem ophilia patien ts. loWhen compared to the amount of FXa that would usu-al ly be produced by the FIXa/FV IIIa complex, the quanti tygenerated by platelet-bound FVIIa is low. However, i t is sig-ni f icantly higher than the level of FXa normal ly producedon platelets of haemophi l iacs, and is certainly suff ic ient toenhance thrombin generation in exper imental models of FIXand FVIII deficiency.

We bel ieve that haemophi l ia is character ised pr imar ilyby a faihrre o f platelet-surface thrombin generation. If thisis the c ase, then resul ts from our studies in exper imentalmodels suggest that high levels of FVIIa may part ially restoreFXa generation on the platelet surface, leading to enhancedthrombin production in the absence of FIX or FVIII (Pig. 4) .

We have tentatively made two extrapolations of our invi tro data to the in vivo effects of high-dose FVIIa therapyin haemophi l ia. First, OUT data suggest that a high dose isneeded because PVIIa binds to platelets w ith a low aff ini tyo(d of 50-100 nM, rather than 5 1 nM or less for FVIIabinding to TF).2 As a resul t, a high concentration of FVIIais required to achieve even a mode st degree of plateletbinding. At the concentrations of PVIIa attained in z&o,binding to platelets is not saturated. This observation led usto predict that an escalation of FVIIa dose should thereforeincrease platelet-surface thrombin generation. Several groUpshave confi rmed this theory by demonstrating that cl inicaleff icacy may be attained by increasing the dose of FVIIa inthose haemophikac patients who fai l to respond to ini tialdose recommendations.

m Blood Reviews (2003) 17, S -55 0 2003 Elsevier Science Ltd. All r ights reserved.


5/5

Fig. 4. High-dose FVlla partially restores platelet-surface thrombin gen-eration in haem ophil ia . (Reproduced from Hoffman M., Monroe D.M.3rd, Roberts H.R. Activated factor V II activates factors IX and X on thesurface of activated platelets: thoughts on the mechanism of action ofhigh-dose activated factor VII. Blood Co agul Fibrinolysis 1998; 9 (Suppl.I): S6 IS65, with permission.)

The second extrapolation from studies of FVIIa in theexper imental model is that the action o f high-dose FVIIain vivo is not directly dependent on n, but is insteadplateletdepe ndent. Earlier theories postulating a TFdep en-dent mode of action for FVIIa explain the local isation ofFVIIa a$ti$i ty to the injury si te, which may account for therelative lack of thrombotic compl ications observed dur inghighdose FVIIa therapy, but do not adequately justi fy therequirement for high do ses. lo Howeve r, a platelet-dependentmecha nism in which FVIIa binds to platelets w ith low aflin-i ty al lows not only foe the local isation of FVIIa activi ty, butalso explains why high dos es are required to attain clinicallyeffective levels of thrombin generation. Whi le this theoryof platelet dependence does not preclude other actions andeffects of FVIIa, i t is consistent wi th empir ical ly determineddosing requirements. Howev er, i t is important to be awarethat this mechanism is not truly TF- independent, as TF is st i l lrequired for the initiation of coagu lation. The theory of aplateletdependent mechan ism of action simply impl ies thatthe pr imary effect of FVIIa occurs on the platelet surface.

i

0 2003 Elsevier Science Ltd. All r ights mserved. Blood Reviews (2003) 17, 5 l-55 m

CONCLUSIONS

When compared to the tradit ional cascade hypothesis, webel ieve that the cel l -based conceptual model of haemosta-sis al lows a more fundam ental understanding of the cl inicalproblems observed in some coagulation disorders by focus-ing on the cen tral role of spec ific cell surfac es in controll ingand directing the haemo static process.Our cel l -based model bui lds upon the foundations laid bythe tradi tional cascade theory, but places greater empha sison the roles of speci f ic receptors present on the surfaces ofthe cel ls involved. Importantly, the cel l -based model sug geststhat understanding the structure and function of coagula-t ion proteins is necessary, but not suff ic ient, to understandhaemo stasis in vivo. Accordingly, this more recent modelprovides a potential ly more accurate representation of the

haemo static process in vivo, and facil i tates a greater un-derstanding of the pathophysiological mecha nisms behindcoagulation disorders such as haemophi l ia.

Haemophi l ia may be chamcter ised by a fai lure of platelet-surface thrombin generation in the f inal propagation stageof the haem ostatic process. High-dose recombinant FVIIa(rFVIIa; NovoSeven@ , Novo Nordisk, Copenhagen, Denma rk)has shown considerable succe ss as a therapy for haemophi l-iacs and inhibi tor patients, and this success may be due toa mechan ism of action involving platelet-surface FXa gener-ation. This resul ts in enhanced thrombin production, andmay part ial ly compensa te for the deficiency of FIX or FVIII.The eff icacy of high-dose rFVIIa in haemophi l ia and Inhibi torpatients has led to i ts use in a growing number of al ternativeindications, and data regarding i ts mechan ism of action insuch circumstance s are scarce. However, we bel ieve tha teven in the presence of the FIXa/FV IIIa complex, FVIIa maybe able to enhance both FXa and FIXa levels on the plateletsurface, thus augmenting the production of vi tal thrombin.

References

I. Davie EW, Ratnoff OD. Waterfall sequence for intr insic blood c lot-t ing. Science 1964; 145: I 3 IO- I 3 12.

2. Hoffman M. Mechanism of action of NovoSeven@using a cell-basedmode l. B loodli ne Reviews 2002; I : 5-6.

3. Veldman A. Hoffman M, Ehrenforth S. New insights into the coag-ulation system and implications for new therapeutic options withrecombinant factor Vlla. Curr Med Chem 2003; IO: 797-81 I.

4. Hoffman M, Monroe DM 3rd. A cell-based mode l of hemostasis.Thromb Haemost 2001; 85: 958-965.

5. Monroe DM. Roberts H R. Hoffman M. Platele t procoagulant com-plex assembly in a t issue factor- init iated system. Br J Haem atol 1994;88: 364-37 I.

6. Mann KG. Poten tial analytes for the diagnosis of thrombosis. Anoverv iew. Ann Ep idemiol 1992; 2: 365-370.

7. Bag lia FA, Bad ell ino KO. Li CQ, Lopez JA, Walsh PN. Factor Xl bind-ing to the platelet glycoprotein lb-IX-V complex promotes factorXl activation by thrombin. J Biol Chem 2002; 277: I662- 16 68.

8. Oliver ]A, Monroe DM, Roberts HR. Hoffman MR. Feedback acti-vation of factor XI on platelets in the absence of factor XII. Arte-r ioscler Thromb Vast Biol 1999; 19: 170-l 77.

9. Hoffma n M, Monroe DM 3rd. Roberts HR. Activated factor VII acti-vates factors IX and X on the surface of activated platelets: thoughtson the mechanism of action of high-dose activated factor VII. Bloo dCoagu l Fibrinolysis 1998; 9 (Suppl I): S6 I-S65.

IO. Monroe DM, Hoffman M. Oliver ]A, Roberts HR. Platele t activ ity ofhigh-dose factor Vlla is indepen dent of t issue factor. Br J Haema tol1997; 99: 542-547.

Documents

A Cell-based Model of Coagulation and the Role of Factor VIIa - Blood Review 2003