Genetic Variations Associated with Recurrent Venous Thrombosis Astrid van Hylckama Vlieg, Linda E. Flinterman, Lance A. Bare, Suzanne C. Cannegieter, Pieter H. Reitsma, Andre R. Arellano, Carmen H. Tong, James J. Devlin and Frits R. Rosendaal Circ Cardiovasc Genet. published online September 10, 2014; Circulation: Cardiovascular Genetics is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2014 American Heart Association, Inc. All rights reserved. Print ISSN: 1942-325X. Online ISSN: 1942-3268
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circgenetics.ahajournals.org/content/early/2014/09/10/CIRCGENETICS.114.000682
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation: Cardiovascular Genetics can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation: Cardiovascular Genetics is online at: http://circgenetics.ahajournals.org//subscriptions/
Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
Genetic Variations Associated with Recurrent Venous Thrombosis
Running title: van Hylckama Vlieg et al.; Genetic variations and recurrent thrombosis
Astrid van Hylckama Vlieg, PhD1,3; Linda E. Flinterman, PhD1; Lance A. Bare, PhD4; Suzanne C. Cannegieter, MD, PhD1,3; Pieter H. Reitsma, PhD2,3; Andre R. Arellano, BSc4; Carmen H. 11,2,3 2,,3 MD,, Ph PhD D1, Tong, MSc4; James J. Devlin, PhD4; Frits R. Rosendaal, MD
1 3
Department a tme art m nt off C Clinical linical Epidemiology, Epi p demiologgy, y 2Depar Department rtm ment of Thr Thrombosis h ombosis and Haemostasis hr Haemostasis, s
Einthovenn Laboratory Laaboratory for forr Experimental Exp xpperim mentall Vascular Vascuulaar Medicine, Va Medic iccin inee, Le Leiden eid den U University niiversitty Me Medicall C Center, Leiden, Le eid iden en n, th the he Ne Neth Netherlands; t errla th l nd ndss; 4Ce Celera, Cele lera le r , Al ra A Alameda, amed am ed da, a, C CA A
Correspondence: A. van Hylckama Vlieg Leiden University Medical Center Department of Clinical Epidemiology, C7-97 PO Box 9600 2300 RC Leiden The Netherlands Tel: +31 (0)71 526 1562 Fax: +31 (0)71 526 6994 Email: [email protected]
Journal Subject Codes: [173] Deep vein thrombosis 1 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
Abstract: Background - The prediction of recurrent venous thrombosis using individual genetic risk predictors has proven to be challenging. The aim of this study was to assess whether multiple genetic SNP analysis would predict recurrent venous thrombosis. Methods and Results - Patients with a first venous thrombosis were followed for a recurrent venous thrombosis up to 2009 (MEGA follow-up study), which occurred in 608 out of 4100 patients (2.7%/yr). 31 Common thrombosis-associated single nucleotide polymorphisms (SNPs) were associated with the risk of recurrence. A genetic risk score (GRS) for each individual was calculated by summing the number of risk-increasing alleles for each of the 31 SNPs and for a simplified model consisting of 5 SNPs: rs6025, rs1799963, rs8176719,, rs2066865, rs2 s206 0668 06 6865 68 65,, and 65 and rs2036914. The risk of recurrence associated with the GRS was calculated continuously ated co cont ntin nt inuo in uous uo usly us ly and a after an stratification on inn a low low and a d high an h gh score. All individual hi al SNPs SNPs were at most mo ostt mildly associated with w recurrence ri Regarding the recurrence highest patients with rrisk. iskk. Regardin ing th in he 31 331-SNP -SN SNP P GR GRS,, re eccu urren ence ri risk skk wa wass hi high g est in gh n pat ati at tient en nts w ithh 31 it and lowest inn pa T) and nd rs2036914 (F11 (F11). 1). ) 3876 patients had a valid measurement for the five SNPs included in this model. odel. Thee ris risk five-SNP sk off rrecurrent ecuurre ec rennt venous thrombosis associated re asssociated with the fiv iv ve-SNP GRS was calculated co continuously, ontinuously, i.e., on i.ee., per per addition add d itio dd on of of onee risk riskk allele, alleele le,, and a d af an after st stratification trattificati tioon inn a low ti l w and lo high risk score c core (i.e., (i.ee., d 1 risk riiskk alleles all lleles l ((P10) P110) andd t 5 ri risk iskk alle alleles l les ((P90)) P90) P9 P90) 0)) )) by b calcula calculating l ting HRs, HR ad adjusted d for age and se recurrence was after four, sex. x. The The cumulative cumu cu mula mu lati tive ve iincidence ncid nc iden ence en ce ooff re recu curr cu rren rr ence en ce w as ccalculated alcu al cula cu late tedd af te afte terr fo te four ur,, an ur andd si sixx yyears of follow-up ffor iindividuals di id l with i h a low l GRS and d iindividuals di id l with i h a hi highh GRS. GRS Since predictors of recurrence may be different from those of a first event, we also constructed a prediction model using the five genetic variants that were, in this study, most strongly associated with the risk of recurrence. Analyses were performed in the overall patient group and after stratification into patients with a first unprovoked or provoked event and after stratification by sex.4, 5 The risk of recurrent venous thrombosis associated with the five-SNP GRS was also assessed after exclusion of patients with malignant disease at the time of the first venous thrombosis and in patients of Caucasian descent only (defined as both parents born in a North or West European country). Furthermore, we performed stratified analyses on type of the first venous thrombosis (deep
7 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
venous thrombosis of the leg and pulmonary embolism) and family history of venous thrombosis (father, mother, or sibling with venous thrombosis). Unprovoked thrombosis was defined as having none of the following provoking factors: malignant disease in the five years prior to the first thrombosis, surgery, trauma, hospitalisation, immobilisation, plaster cast, hormone use (oral contraceptives and hormone therapy), or pregnancy within three months before the first event, within 4 weeks postpartum, or long-haul fight (> 4 hours) in the two months prior to the first thrombosis.
Results The mean age of the the 4100 4 000 patients at the time of the first 41 fi event was 488 years (range: 18-70 18 70 years) and 2233 (54.5%) women. with a 54.55%) were wo 54 ome men. n. Th Thee ttotal otal ot a vo vvolume olume me off ffollow-up olllow o -u -up was was 22040 wa 2 0400 person-years 22 perrso sonn-ye yeearrs wi w mean follow-up (range months-10.7 w up of w-up of 5.4 5 4 years 5. yearrs (r ye (ran angee 11.1 an .11 m onnth hss-10 10.7 0.7 7 years). y ar ye a s)). Off these theese ppatients atieent at ntss 608 60 had had a recurrent recc re event resulting overall CI t ting in an overa all incidence inc n id iden ence en cee of o rrecurrence e urrre ec renc ncce of 27 227.6 .66 pper er 11000 0000 pe pperson-years rson rs on--years (95% C on 25.4-29.8). 1 (28.3%) Patients 1133 1333 (28.3 13 (28. (2 8.33%) %) P atie at ient ntss ha hhad add an uunprovoked npro np provo voke kedd fi ffirst irs rstt event even ev entt an andd 28 2870 70 (71.7%) (71 71.7 .7%) 7%)) hhad ad a provoked first event (information missing in 97 patients). The incidence of recurrent thrombosis was higher in patients with a first unprovoked event than in patients with a provoked first event (provoked: 21.8 per 1000 person-years, 95% CI: 19.5-24.1; unprovoked: 42.9 per 1000 personyears, 95% CI: 37.6-48.1). As expected, the risk of recurrence was higher in men than in women. Table 1 shows the risk of recurrent venous thrombosis associated with the 31 SNPs. The SNPs were ordered according to the risk of recurrent thrombosis in an additive model. All individual SNPs were at most mildly associated with the risk of recurrence, with the strongest association for the factor V Leiden mutation (HR additive model: 1.60, 95% CI: 1.34-1.89). We tested whether the number of risk-increasing alleles was associated with the risk of recurrent venous thrombosis. The number of risk-increasing alleles ranged from 13 to 40; there 8 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
was a trend towards more risk alleles in patients who had a recurrence compared with patients who did not (figure 1). The addition of one risk allele to the 31-SNP genetic risk score increased the risk of a recurrence on average by 4% (HR: 1.04; 95% CI: 1.02-1.07). Using the 31-SNP genetic risk score we were able to stratify patients according to their risk of recurrent thrombosis (figure 2, panel A). The discriminative power became apparent mainly after four years of follow-up. Table 2 shows the cumulative incidence of recurrence for groups with varying numbers of risk alleles. The risk of recurrence was highest in patients with 31 risk alleles (6-year CI: 19.2%, 95% CI 15.3-23.3) and lowest in patients risk atients with A, rs8176719 (ABO), rs2066865 (FGG 10034 C>T) and rs2036914 (F11), could stratify patients into high and low risk of recurrence, with an over two-fold difference. The predictive power was even stronger after stratification into provoked and unprovoked first events. The measurement of five SNPs was sufficient.
12 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
The high frequency of recurrent venous thrombosis (12-20% in five years) begs the question of long-term anticoagulant prophylaxis after a first thrombosis. However, since this therapy is associated with a risk of major haemorrhage of 2% per year23, indiscriminate use in all patients is not warranted. Identification of individuals at low risk of recurrence in whom anticoagulant therapy can safely be withheld, and in those with high risk who should receive long-term treatment, will lead to a minimum of untoward events, i.e. thrombotic and bleeding events. Groups with over 25% six-year risk of recurrence were men with a high risk score, and all individuals with a high risk score and an unprovoked first event. These hese constituted 6.0% 6..0% of all patients, and in them long-term anticoagulation could be considered. Vice versa, groups pss w with i e rrisk ear i k of is o recurrence rec e ur urrrence were women with a lo low risk score, an nd all individuals withh a low 7% six-year and risk score an and nd a provokedd ffirst irst event. even nt. These The h see cconstituted he onnsttituuteed almost alm lmos o t 10% os 10 of of alll ppatients, atieent n s, aand nd inn th them a ation may be be safely saf afel felly discon ddiscontinued. di i ntin tiinued. anticoagulation may hough houg gh the factor V Leiden Leiiden andd th he pr pprothrombin othhrombi b n 20210A bi 2002110A mutation mutatio i n were the stron Although the strongest genetic markers rkers rk k among th the h 31 meas measured red d SN SNP SNPs SNPs, Ps neither ithhe were ere strongly strongl tr gll associated ciat i ed d with ith ithh the th h risk rii of recurrence when considered individually. Furthermore, when the SNPs were ranked according to the risk of recurrence, the top five SNPs performed less well than the five SNPs previously identified as performing well in predicting first thrombosis, which have repeatedly proven their ability to predict the risk of a first thrombosis in multiple studies. Compared with studies on the risk of a first venous thrombosis, little information is available on the risk of recurrence regarding genetic variation. The 5-SNP genetic risk score using the five genetic variants that were most strongly associated (i.e. by univariate HR) with the risk of recurrent venous thrombosis was only based on information from the current study so the risk estimates associated with the selected variants contain more uncertainty than those of the well-established five SNPs
13 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
in the genetic risk score. This most likely led to the somewhat worse predictive value of the former 5-SNP genetic risk score. Limitations of our study are that we excluded children and individuals aged over 70 at the time of the first event. Therefore, our findings cannot be extrapolated to these patients. Per analysis, we also excluded patients with missing measurements. This resulted in the exclusion of 660 patients for the 31-SNP model and 223 for the 5-SNP model. Arguably, this exclusion may have led to a loss in power to detect potential effects on the risk of recurrence. Of the total study population, 3569 (89.5%) were of Caucasian descent. Therefore, we can apply an only app pply these se findings to patients with a similar ethnic background. Further studies are needed needded ttoo assess asses esss the th predictive value ethnic Strengths v ue of of th tthee genetic geenetic model in populations ns with with different ethn hnnic backgrounds. Stre e of our studyy are events homogeneous arre the large number numb mb ber of of recurrent recu cu urren ent ev ventss oobserved bsserve vedd in n a hom om moggeneo ous u we wellcharacterised cohort patients, which enabled assess risk e co ed oho hort ort of pa pati tien ti ientts, ts wh w ichh enab ic ble ledd us u tto o as asse esss tthe he ri iskk ooff ve venous us tthrombosis us hrom ombo b si bo sis is associated with risk w different ge ggenetic netiic ri iskk scores in iincluding clluddingg subgroup subggroupp analyses, anally lyses,, andd the prolonged p olonge pr g follow-up. In conclusion, we showed that, using genetic markers for venous thrombosis, it is possible to stratify patients who have had a first venous thrombosis, into subgroups with a high and low risk of recurrence. Risk stratification became more pronounced when the genetic risk score was combined with information regarding provoking factors at the time of the first event and with sex. We believe that the results indicate that for some high-risk groups long-term anticoagulant treatment is indicated. With the increasing use of exome and whole genome sequencing, novel genetic markers for venous thrombosis will be discovered. These will most likely be rare genetic variants with a high thrombosis risk. These genetic markers will be of great importance in further unravelling causes of venous thrombosis but due to their low prevalence in
14 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
the general population, these genetic markers may be of little clinical relevance when considered individually. However, the prediction of recurrence risk using multiple genetic and environmental markers may be further optimized.
Funding Sources The Multiple Environmental and Genetic Assessment of Risk Factors for Venous Thrombosis was supported by The Netherlands Heart Foundation (grant NHS 98.113), the Dutch Cancer Foundation (RUL99/1992), and The Netherlands Organization for Scientific Research (grant 912-03-033l 2003). The MEGA follow-up study was supported by The Netherlands Heart Foundation (grant NHS 2008B86). The funding organizations played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or the preparation, review, or approval of the manuscript. Conflict-of-Interest Disclosure: Lance A. Bare, James J. Devlin, Pieter ter .H H Re R Reitsma, itsma, and it nd F Frits R. Rosendaal hold or have applied for patents related to SNPs in this manuscript nuscrip iptt (notably ip (not (n otab ot ably ab ly rrs6025, s60 s6 rs2066865, and nd rs2036914). rs203 s2203 0 69 6 144). Andre R. Arellano, Carmen Caarm rmen H. Tong, Tongg, James Jame mes J. Devlin, and Lance me Lan A. Bare are employees mployees mp plo oyees off Celera Ceele Cele l raa C Corporation, orrpo p rati raati tion onn, a wholly whol wh o ly y oowned wnned d ssubsidiary ubbsi ubsi s di diar aryy of Q ar Quest uest ue st Dia Diagnostics. Diagnostic iagn ia gnos gn ostic The os remaining au aauthors uth hors declaree no no co competing ompe p tingg ffinancial pe in nanciiall in interests. nteeressts.. References: s s: Incidence nci cide denc de ncee of nc o vvenous enou en ouss th ou thro thromboembolism: romb ro mboe mb oemb oe mbol mb olis ol ism: is m: a community-based com ommu m ni mu nity ty y-bbas a ed study stu tudy dyy in in Western West We ster st ernn France. er Fr 1. Oger E. In EPI-GETBP P Study Stu tudy dy Group. Gro roup up.. Groupe Groupe Grou pe dd'Etude 'E Etude tude ddee la la Thrombose Throm hrombo bose se de de Bretagne Bret Br etag tagne ne Occidentale. Occ ccid id den enta tale le.. Thromb Thro Th roo Haemost. 2000;83:657-660. 2. Naess IA, Christiansen SC, Romundstad P, Cannegieter SC, Rosendaal FR, Hammerstrom J. Incidence and mortality of venous thrombosis: a population-based study. J Thromb Haemost. 2007;5:692-699. 3. Young L, Ockelford P, Milne D, Rolfe-Vyson V, Mckelvie S, Harper P. Post-treatment residual thrombus increases the risk of recurrent deep vein thrombosis and mortality. J Thromb Haemost. 2006;4:1919-1924. 4. Christiansen SC, Cannegieter SC, Koster T, Vandenbroucke JP, Rosendaal FR. Thrombophilia, clinical factors, and recurrent venous thrombotic events. JAMA. 2005;293:23522361. 5. Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet. 2003;362:523-526. 6. Hansson PO, Sorbo J, Eriksson H. Recurrent venous thromboembolism after deep vein thrombosis: incidence and risk factors. Arch Intern Med. 2000;160:769-774. 15 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
7. Prandoni P, Tormene D, Spiezia L, Pesavento R, Simioni P. Duration of anticoagulation and risk of recurrent thromboembolism in carriers of factor V Leiden or prothrombin mutation. J Thromb Haemost. 2008;6:2223-2224. 8. Heit JA, Phelps MA, Ward SA, Slusser JP, Petterson TM, De Andrade M. Familial segregation of venous thromboembolism. J Thromb Haemost. 2004;2:731-736. 9. Larsen TB, Sorensen HT, Skytthe A, Johnsen SP, Vaupel JW, Christensen K. Major genetic susceptibility for venous thromboembolism in men: a study of Danish twins. Epidemiology. 2003;14:328-332. 10. Souto JC, Almasy L, Borrell M, Blanco-Vaca F, Mateo J, Soria JM, et al. Genetic susceptibility to thrombosis and its relationship to physiological risk factors: the GAIT study. Genetic Analysis of Idiopathic Thrombophilia. Am J Hum Genet. 2000;67:1452-1459. 0;67:1452-1459. 11. Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen ggen CJ CJ. The Th value valu ue ooff family history as a risk indicator for venous thrombosis. Arch Intern Med. 2009;169:610-615. Med 20 2009 09;1 09 ;169 ;1 69:6 69 :6 610 61 12. de Haan HG, Bezemer ID, Doggen CJ, Cessie S,, R Reitsma AR,, et Multiple n HG G, Beze eme m r ID D, Do Dogg ggen gg en nC J, le le C essiee S es eits ei tsma ts maa PH, PH, Arellano Are r llan anoo AR an A e aal. l. Mu u SNP testingg improves prediction venous thrombosis. Blood. 2012;120:656-663. im mproves risk k pred ed diction o off ffirst on irsst veno nous th hrom ombo bossis.. B bo loood. 20 2012;1 120 20:6 656-6663 13. van Hylckama CA, Rosendaal Baglin of lckama maa V Vlieg lieg g A, A, Ba B Baglin glin gl l C A, Baree LA LA, Ro R send se ndaa nd aall FR aa FR, Ba Bagl gllin in TP. P Proof Proof off of of principle pri pr princ in inc potential clinical thrombosis. J inical utility off multiple mu ult ltip iple ip le SNP SNP analysis ana naly ly ysi siss for for prediction pred pr e iccti ed t onn of of recurrent recu re cu urr rren en nt vvenous enous thromb Thromb Haemost. 2008;6:751-754. a aemost. 2008;6 ; :7 751-754 7 4. 14. Blom JW, CJ, Osanto S, Rosendaal FR. Malignancies, and W Doggen D CJ Os to S Ro ndd l FR M Malignancies ali li cii prothrombotic othhr botii mutations, m ta tation tii the risk of venous thrombosis. JAMA. 2005;293:715-722. 15. Flinterman LE, van Hylckama Vlieg A, Cannegieter SC, Rosendaal FR. Recurrent venous thrombosis: incidence and characteristics in a large cohort of VT patients (MEGA study). J.Thromb.Haemost. 9 [Supplement s2], 42. 2011. Ref Type: Abstract 16. Bezemer ID, Bare LA, Doggen CJ, Arellano AR, Tong C, Rowland CM, et al. Gene variants associated with deep vein thrombosis. JAMA. 2008;299:1306-1314. 17. Arellano AR, Bezemer ID, Tong CH, Catanese JJ, Devlin JJ, Reitsma PH, et al. Gene variants associated with venous thrombosis: confirmation in the MEGA study. J Thromb Haemost. 2010;8:1132-1134. 18. Germer S, Holland MJ, Higuchi R. High-throughput SNP allele-frequency determination in pooled DNA samples by kinetic PCR. Genome Res. 2000;10:258-266. 19. Shiffman D, O'Meara ES, Bare LA, Rowland CM, Louie JZ, Arellano AR, et al. Association of gene variants with incident myocardial infarction in the Cardiovascular Health Study. Arterioscler Thromb Vasc Biol. 2008;28:173-179.
16 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
20. Shiffman D, Ellis SG, Rowland CM, Malloy MJ, Luke MM, Iakoubova OA, et al. Identification of four gene variants associated with myocardial infarction. Am J Hum Genet. 2005;77:596-605. 21. Iakoubova OA, Tong CH, Chokkalingam AP, Rowland CM, Kirchgessner TG, Louie JZ, et al. Asp92Asn polymorphism in the myeloid IgA Fc receptor is associated with myocardial infarction in two disparate populations: CARE and WOSCOPS. Arterioscler Thromb Vasc Biol. 2006;26:2763-2768. 22. Shiffman D, Rowland CM, Louie JZ, Luke MM, Bare LA, Bolonick JI, et al. Gene variants of VAMP8 and HNRPUL1 are associated with early-onset myocardial infarction. Arterioscler Thromb Vasc Biol. 2006;26:1613-1618. 23. Kearon C, Ginsberg JS, Kovacs MJ, Anderson DR, Wells P, Juliann JA, et al. Comparison Comp par aris i o of low-intensity warfarin therapy with conventional-intensity warfarin therapy long-term erap er apyy for ap for lo long ng--term ng te m prevention of recurrent venous thromboembolism. N Engl J Med. 2003;349:631-639. 3;349: 9 63 6311-63 639. 9
17 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
Table 1: The risk of recurrent venous thrombosis associated with 31 individual SNPs previously associated with the risk of a first venous thrombotic event (HR adjusted for age and sex) Gene
SNP
Chrom
Position
Risk allele freq, %
N
HR*
95% CI
HR†
95% CI
F5
rs6025
1
167.785.673
8.3
4098
1.34-1.89
rs1799963
11
46.717.631
2.7
4099
1.23
0.88-1.71
F11
rs3822057
4
187.425.146
54.7
3934
1.22
1.09-1.37
F2
rs3136520
11
46.699.808
2.8
3910
1.20
0.87-1.64
ABO
rs8176719
9
1366.1332.90 13 908 136.132.908
45.4
39077
1.18
1.05-1.34
F13A1
rs5985
6
6.263.79 6. 7944 79 6.263.794
755.55 75.5
4094 94 4094
1.17
1.02-1.34
F11
rs2036914
4
187.429.4775 187.429.475
58.4 58.4
39943 3943
1.15
1.02-1.30
F5
rs4524
1
1677 778 16 778 379 379 167.778.379
77 9 77.9
3942 3942
1.13
0.98-1.30
VWF
rs1063856
12
6.153.534
36.6
3961
1.13
1.00-1.27
F11
rs2289252
4
187.444.375
47.7
3912
1.12
0.99-1.26
SERPINC1
rs2227589
1
172.152.839
11.5
4030
1.11
0.94-1.32
FGG
rs2066865
4
155.744.726
33.4
4099
1.11
0.99-1.25
RGS7
rs670659
1
239.228.398
66.1
4046
1.08
0.96-1.23
PROCR
rs2069952
20
33.227.612
63.4
3914
1.08
0.96-1.22
F9
rs6048
X
138.460.946
72.8
4035
1.37-2.00 0.93-4.13 0.85-1.69 0.36-18.1 1.04-1.67 1.19 1. 19 1.19-1.96 0.81 0. 0.81-1.61 0 60 6 0.60-9.65 1.033 1.03-1.54 1 088 1. 1.08-1.78 0.888 0. 0.88-1.96 1.000 1.00-2.19 1.000 1.00-1.63 1.07 07 7 1.07-1.78 0 63 0.63-1.37 0.76-1.62 0.91-1.30 1.01-1.65 0.84-1.26 0.99-1.57 0.90-1.33 0.77-2.41 0.81-1.14 1.07-1.73 0.89-1.59 0.93-1.66 0.91-1.54 0.94-1.60 0.85-1.62 0.93-1.43
1.60
F2
1.66 1.96 1.20 2.54 1.32 1.53 1.53 1.14 14 1.14 2 40 2.40 1.26 1.39 39 1.39 31 1.31 1.48 4 1.48 1.28 1.28 1.38 38 1.38 0 93 93 0.93 1.11 1.09 1.29 1.03 1.25 1.09 1.36 0.96 1.36 1.19 1.24 1.19 1.22 1.17 1.15
1.06
0.96-1.18
18 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
PROCR
rs867186
20
33.228.215
14.1
4030
NAT8B
rs2001490
2
73.781.606
38.8
4057
PROC
rs1799809
2
127.892.345
47.6
4058
F3
1208 indel
1
94.780.000
46.9
3959
F13B
rs6003
1
195.297.644
10.2
3840
F8
rs1800291
X
153.811.479
85.2
4100
TFPI
rs8176592
2
188.040.937
68.9
3915
GP6
rs1613662
19
60.2288.407 60 60.228.407
83.8
52 4052
F2
rs3136516
11
46.7177.3 46 .332 3 32 46.717.332
52 6 52.6
40085 4085
HIVEP1
rs169713
6
11.920.5177
21 9 21.9
38887 3887
NR1I2
rs1523127
3
1200 983 12 983 729 729 120.983.729
40 4 40.4
4061 4061
STXBP5
rs1039084
6
147.635.413
42.2
3989
F9
rs4149755
X
138.451.778
6.7
3913
CPB2
rs3742264
13
45.546.095
68.6
3893
PROCR
rs2069951
20
33.277.425
6.7
3914
F5
rs1800595
1
167.776.972
5.8
3959
1.08 1.00 1.15 1.04 1.04 1.09 1.02 1.06 0.96 1.45 1.09 1 08 1. 1.08 0.97 0. 97 0.97 1 03 1.03 0.98 98 0.98 88 0.88 0.97 0.90 0.90 1 04 1. 1.04 0 86 0.86 0.90 0.86 0.87 0.89 0.82 0.92 0.76 0.86 0.62 0.81
0.90-1.30 0.57-1.73 0.97-1.37 0.81-1.33 0.87-1.27 0.87-1.37 0.84-1.24 0.84-1.34 0.77-1.20 0.82-2.57 0.73 7 0.73-1.64 0.82 0. 82 0.82-1.44 0.7 73 0.73-1.31 0 0.77 0.77-1.38 0.600 0.60-1.62 0 611 0. 0.61-1.60 0.73 0. 3 0.73-1.08 0.788 0.78-1.20 0.75 5 0.75-1.07 0.711 0.71-1.52 0 72 0.72-1.03 0.71-1.14 0.72-1.03 0.69-1.10 0.58-1.34 0.52-1.28 0.70-1.20 0.58-1.00 0.66-1.11 0.15-2.47 0.61-1.07
0.76
0.11-5.44
SNP: Single nucleotide polymorphism, Chrom: Chromosome, HR: Hazard Ratio * Hazard ratio for heterozygous and homozygous carriers of risk allele versus homozygous carriers of non-risk allele † Hazard ratio for the risk allele using an additive model
19 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
1.05
0.90-1.23
1.05
0.93-1.17
1.05
0.94-1.17
1.03
0.92-1.16
1.03
0.86-1.24
1.03
0.91-1.18
1.03
0.91-1.17
1.00
0.86-1.16
0.99
0.88-1.10
0.95
0.82-1.09
0.93
0.83-1.04
0.92
0.82-1.03
0.90
0.74-1.10
0.86
0.76-0.97
0.85
0.66-1.08
0.82
0.62-1.07
DOI: 10.1161/CIRCGENETICS.114.000682
Table 2. 4-year and 6-year cumulative risk of recurrence using different genetic risk scores # risk alleles
N total (%)
N rec
4-year cumulative incidence (95% CI)
6-year cumulative incidence (95% CI)
Incidence rate (95% CI)
31-SNP GRS
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circgenetics.ahajournals.org/content/early/2014/09/10/CIRCGENETICS.114.000682
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation: Cardiovascular Genetics can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation: Cardiovascular Genetics is online at: http://circgenetics.ahajournals.org//subscriptions/
Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
Genetic Variations Associated with Recurrent Venous Thrombosis
Running title: van Hylckama Vlieg et al.; Genetic variations and recurrent thrombosis
Astrid van Hylckama Vlieg, PhD1,3; Linda E. Flinterman, PhD1; Lance A. Bare, PhD4; Suzanne C. Cannegieter, MD, PhD1,3; Pieter H. Reitsma, PhD2,3; Andre R. Arellano, BSc4; Carmen H. 11,2,3 2,,3 MD,, Ph PhD D1, Tong, MSc4; James J. Devlin, PhD4; Frits R. Rosendaal, MD
1 3
Department a tme art m nt off C Clinical linical Epidemiology, Epi p demiologgy, y 2Depar Department rtm ment of Thr Thrombosis h ombosis and Haemostasis hr Haemostasis, s
Einthovenn Laboratory Laaboratory for forr Experimental Exp xpperim mentall Vascular Vascuulaar Medicine, Va Medic iccin inee, Le Leiden eid den U University niiversitty Me Medicall C Center, Leiden, Le eid iden en n, th the he Ne Neth Netherlands; t errla th l nd ndss; 4Ce Celera, Cele lera le r , Al ra A Alameda, amed am ed da, a, C CA A
Correspondence: A. van Hylckama Vlieg Leiden University Medical Center Department of Clinical Epidemiology, C7-97 PO Box 9600 2300 RC Leiden The Netherlands Tel: +31 (0)71 526 1562 Fax: +31 (0)71 526 6994 Email: [email protected]
Journal Subject Codes: [173] Deep vein thrombosis 1 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
Abstract: Background - The prediction of recurrent venous thrombosis using individual genetic risk predictors has proven to be challenging. The aim of this study was to assess whether multiple genetic SNP analysis would predict recurrent venous thrombosis. Methods and Results - Patients with a first venous thrombosis were followed for a recurrent venous thrombosis up to 2009 (MEGA follow-up study), which occurred in 608 out of 4100 patients (2.7%/yr). 31 Common thrombosis-associated single nucleotide polymorphisms (SNPs) were associated with the risk of recurrence. A genetic risk score (GRS) for each individual was calculated by summing the number of risk-increasing alleles for each of the 31 SNPs and for a simplified model consisting of 5 SNPs: rs6025, rs1799963, rs8176719,, rs2066865, rs2 s206 0668 06 6865 68 65,, and 65 and rs2036914. The risk of recurrence associated with the GRS was calculated continuously ated co cont ntin nt inuo in uous uo usly us ly and a after an stratification on inn a low low and a d high an h gh score. All individual hi al SNPs SNPs were at most mo ostt mildly associated with w recurrence ri Regarding the recurrence highest patients with rrisk. iskk. Regardin ing th in he 31 331-SNP -SN SNP P GR GRS,, re eccu urren ence ri risk skk wa wass hi high g est in gh n pat ati at tient en nts w ithh 31 it and lowest inn pa T) and nd rs2036914 (F11 (F11). 1). ) 3876 patients had a valid measurement for the five SNPs included in this model. odel. Thee ris risk five-SNP sk off rrecurrent ecuurre ec rennt venous thrombosis associated re asssociated with the fiv iv ve-SNP GRS was calculated co continuously, ontinuously, i.e., on i.ee., per per addition add d itio dd on of of onee risk riskk allele, alleele le,, and a d af an after st stratification trattificati tioon inn a low ti l w and lo high risk score c core (i.e., (i.ee., d 1 risk riiskk alleles all lleles l ((P10) P110) andd t 5 ri risk iskk alle alleles l les ((P90)) P90) P9 P90) 0)) )) by b calcula calculating l ting HRs, HR ad adjusted d for age and se recurrence was after four, sex. x. The The cumulative cumu cu mula mu lati tive ve iincidence ncid nc iden ence en ce ooff re recu curr cu rren rr ence en ce w as ccalculated alcu al cula cu late tedd af te afte terr fo te four ur,, an ur andd si sixx yyears of follow-up ffor iindividuals di id l with i h a low l GRS and d iindividuals di id l with i h a hi highh GRS. GRS Since predictors of recurrence may be different from those of a first event, we also constructed a prediction model using the five genetic variants that were, in this study, most strongly associated with the risk of recurrence. Analyses were performed in the overall patient group and after stratification into patients with a first unprovoked or provoked event and after stratification by sex.4, 5 The risk of recurrent venous thrombosis associated with the five-SNP GRS was also assessed after exclusion of patients with malignant disease at the time of the first venous thrombosis and in patients of Caucasian descent only (defined as both parents born in a North or West European country). Furthermore, we performed stratified analyses on type of the first venous thrombosis (deep
7 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
venous thrombosis of the leg and pulmonary embolism) and family history of venous thrombosis (father, mother, or sibling with venous thrombosis). Unprovoked thrombosis was defined as having none of the following provoking factors: malignant disease in the five years prior to the first thrombosis, surgery, trauma, hospitalisation, immobilisation, plaster cast, hormone use (oral contraceptives and hormone therapy), or pregnancy within three months before the first event, within 4 weeks postpartum, or long-haul fight (> 4 hours) in the two months prior to the first thrombosis.
Results The mean age of the the 4100 4 000 patients at the time of the first 41 fi event was 488 years (range: 18-70 18 70 years) and 2233 (54.5%) women. with a 54.55%) were wo 54 ome men. n. Th Thee ttotal otal ot a vo vvolume olume me off ffollow-up olllow o -u -up was was 22040 wa 2 0400 person-years 22 perrso sonn-ye yeearrs wi w mean follow-up (range months-10.7 w up of w-up of 5.4 5 4 years 5. yearrs (r ye (ran angee 11.1 an .11 m onnth hss-10 10.7 0.7 7 years). y ar ye a s)). Off these theese ppatients atieent at ntss 608 60 had had a recurrent recc re event resulting overall CI t ting in an overa all incidence inc n id iden ence en cee of o rrecurrence e urrre ec renc ncce of 27 227.6 .66 pper er 11000 0000 pe pperson-years rson rs on--years (95% C on 25.4-29.8). 1 (28.3%) Patients 1133 1333 (28.3 13 (28. (2 8.33%) %) P atie at ient ntss ha hhad add an uunprovoked npro np provo voke kedd fi ffirst irs rstt event even ev entt an andd 28 2870 70 (71.7%) (71 71.7 .7%) 7%)) hhad ad a provoked first event (information missing in 97 patients). The incidence of recurrent thrombosis was higher in patients with a first unprovoked event than in patients with a provoked first event (provoked: 21.8 per 1000 person-years, 95% CI: 19.5-24.1; unprovoked: 42.9 per 1000 personyears, 95% CI: 37.6-48.1). As expected, the risk of recurrence was higher in men than in women. Table 1 shows the risk of recurrent venous thrombosis associated with the 31 SNPs. The SNPs were ordered according to the risk of recurrent thrombosis in an additive model. All individual SNPs were at most mildly associated with the risk of recurrence, with the strongest association for the factor V Leiden mutation (HR additive model: 1.60, 95% CI: 1.34-1.89). We tested whether the number of risk-increasing alleles was associated with the risk of recurrent venous thrombosis. The number of risk-increasing alleles ranged from 13 to 40; there 8 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
was a trend towards more risk alleles in patients who had a recurrence compared with patients who did not (figure 1). The addition of one risk allele to the 31-SNP genetic risk score increased the risk of a recurrence on average by 4% (HR: 1.04; 95% CI: 1.02-1.07). Using the 31-SNP genetic risk score we were able to stratify patients according to their risk of recurrent thrombosis (figure 2, panel A). The discriminative power became apparent mainly after four years of follow-up. Table 2 shows the cumulative incidence of recurrence for groups with varying numbers of risk alleles. The risk of recurrence was highest in patients with 31 risk alleles (6-year CI: 19.2%, 95% CI 15.3-23.3) and lowest in patients risk atients with A, rs8176719 (ABO), rs2066865 (FGG 10034 C>T) and rs2036914 (F11), could stratify patients into high and low risk of recurrence, with an over two-fold difference. The predictive power was even stronger after stratification into provoked and unprovoked first events. The measurement of five SNPs was sufficient.
12 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
The high frequency of recurrent venous thrombosis (12-20% in five years) begs the question of long-term anticoagulant prophylaxis after a first thrombosis. However, since this therapy is associated with a risk of major haemorrhage of 2% per year23, indiscriminate use in all patients is not warranted. Identification of individuals at low risk of recurrence in whom anticoagulant therapy can safely be withheld, and in those with high risk who should receive long-term treatment, will lead to a minimum of untoward events, i.e. thrombotic and bleeding events. Groups with over 25% six-year risk of recurrence were men with a high risk score, and all individuals with a high risk score and an unprovoked first event. These hese constituted 6.0% 6..0% of all patients, and in them long-term anticoagulation could be considered. Vice versa, groups pss w with i e rrisk ear i k of is o recurrence rec e ur urrrence were women with a lo low risk score, an nd all individuals withh a low 7% six-year and risk score an and nd a provokedd ffirst irst event. even nt. These The h see cconstituted he onnsttituuteed almost alm lmos o t 10% os 10 of of alll ppatients, atieent n s, aand nd inn th them a ation may be be safely saf afel felly discon ddiscontinued. di i ntin tiinued. anticoagulation may hough houg gh the factor V Leiden Leiiden andd th he pr pprothrombin othhrombi b n 20210A bi 2002110A mutation mutatio i n were the stron Although the strongest genetic markers rkers rk k among th the h 31 meas measured red d SN SNP SNPs SNPs, Ps neither ithhe were ere strongly strongl tr gll associated ciat i ed d with ith ithh the th h risk rii of recurrence when considered individually. Furthermore, when the SNPs were ranked according to the risk of recurrence, the top five SNPs performed less well than the five SNPs previously identified as performing well in predicting first thrombosis, which have repeatedly proven their ability to predict the risk of a first thrombosis in multiple studies. Compared with studies on the risk of a first venous thrombosis, little information is available on the risk of recurrence regarding genetic variation. The 5-SNP genetic risk score using the five genetic variants that were most strongly associated (i.e. by univariate HR) with the risk of recurrent venous thrombosis was only based on information from the current study so the risk estimates associated with the selected variants contain more uncertainty than those of the well-established five SNPs
13 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
in the genetic risk score. This most likely led to the somewhat worse predictive value of the former 5-SNP genetic risk score. Limitations of our study are that we excluded children and individuals aged over 70 at the time of the first event. Therefore, our findings cannot be extrapolated to these patients. Per analysis, we also excluded patients with missing measurements. This resulted in the exclusion of 660 patients for the 31-SNP model and 223 for the 5-SNP model. Arguably, this exclusion may have led to a loss in power to detect potential effects on the risk of recurrence. Of the total study population, 3569 (89.5%) were of Caucasian descent. Therefore, we can apply an only app pply these se findings to patients with a similar ethnic background. Further studies are needed needded ttoo assess asses esss the th predictive value ethnic Strengths v ue of of th tthee genetic geenetic model in populations ns with with different ethn hnnic backgrounds. Stre e of our studyy are events homogeneous arre the large number numb mb ber of of recurrent recu cu urren ent ev ventss oobserved bsserve vedd in n a hom om moggeneo ous u we wellcharacterised cohort patients, which enabled assess risk e co ed oho hort ort of pa pati tien ti ientts, ts wh w ichh enab ic ble ledd us u tto o as asse esss tthe he ri iskk ooff ve venous us tthrombosis us hrom ombo b si bo sis is associated with risk w different ge ggenetic netiic ri iskk scores in iincluding clluddingg subgroup subggroupp analyses, anally lyses,, andd the prolonged p olonge pr g follow-up. In conclusion, we showed that, using genetic markers for venous thrombosis, it is possible to stratify patients who have had a first venous thrombosis, into subgroups with a high and low risk of recurrence. Risk stratification became more pronounced when the genetic risk score was combined with information regarding provoking factors at the time of the first event and with sex. We believe that the results indicate that for some high-risk groups long-term anticoagulant treatment is indicated. With the increasing use of exome and whole genome sequencing, novel genetic markers for venous thrombosis will be discovered. These will most likely be rare genetic variants with a high thrombosis risk. These genetic markers will be of great importance in further unravelling causes of venous thrombosis but due to their low prevalence in
14 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
the general population, these genetic markers may be of little clinical relevance when considered individually. However, the prediction of recurrence risk using multiple genetic and environmental markers may be further optimized.
Funding Sources The Multiple Environmental and Genetic Assessment of Risk Factors for Venous Thrombosis was supported by The Netherlands Heart Foundation (grant NHS 98.113), the Dutch Cancer Foundation (RUL99/1992), and The Netherlands Organization for Scientific Research (grant 912-03-033l 2003). The MEGA follow-up study was supported by The Netherlands Heart Foundation (grant NHS 2008B86). The funding organizations played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or the preparation, review, or approval of the manuscript. Conflict-of-Interest Disclosure: Lance A. Bare, James J. Devlin, Pieter ter .H H Re R Reitsma, itsma, and it nd F Frits R. Rosendaal hold or have applied for patents related to SNPs in this manuscript nuscrip iptt (notably ip (not (n otab ot ably ab ly rrs6025, s60 s6 rs2066865, and nd rs2036914). rs203 s2203 0 69 6 144). Andre R. Arellano, Carmen Caarm rmen H. Tong, Tongg, James Jame mes J. Devlin, and Lance me Lan A. Bare are employees mployees mp plo oyees off Celera Ceele Cele l raa C Corporation, orrpo p rati raati tion onn, a wholly whol wh o ly y oowned wnned d ssubsidiary ubbsi ubsi s di diar aryy of Q ar Quest uest ue st Dia Diagnostics. Diagnostic iagn ia gnos gn ostic The os remaining au aauthors uth hors declaree no no co competing ompe p tingg ffinancial pe in nanciiall in interests. nteeressts.. References: s s: Incidence nci cide denc de ncee of nc o vvenous enou en ouss th ou thro thromboembolism: romb ro mboe mb oemb oe mbol mb olis ol ism: is m: a community-based com ommu m ni mu nity ty y-bbas a ed study stu tudy dyy in in Western West We ster st ernn France. er Fr 1. Oger E. In EPI-GETBP P Study Stu tudy dy Group. Gro roup up.. Groupe Groupe Grou pe dd'Etude 'E Etude tude ddee la la Thrombose Throm hrombo bose se de de Bretagne Bret Br etag tagne ne Occidentale. Occ ccid id den enta tale le.. Thromb Thro Th roo Haemost. 2000;83:657-660. 2. Naess IA, Christiansen SC, Romundstad P, Cannegieter SC, Rosendaal FR, Hammerstrom J. Incidence and mortality of venous thrombosis: a population-based study. J Thromb Haemost. 2007;5:692-699. 3. Young L, Ockelford P, Milne D, Rolfe-Vyson V, Mckelvie S, Harper P. Post-treatment residual thrombus increases the risk of recurrent deep vein thrombosis and mortality. J Thromb Haemost. 2006;4:1919-1924. 4. Christiansen SC, Cannegieter SC, Koster T, Vandenbroucke JP, Rosendaal FR. Thrombophilia, clinical factors, and recurrent venous thrombotic events. JAMA. 2005;293:23522361. 5. Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet. 2003;362:523-526. 6. Hansson PO, Sorbo J, Eriksson H. Recurrent venous thromboembolism after deep vein thrombosis: incidence and risk factors. Arch Intern Med. 2000;160:769-774. 15 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
7. Prandoni P, Tormene D, Spiezia L, Pesavento R, Simioni P. Duration of anticoagulation and risk of recurrent thromboembolism in carriers of factor V Leiden or prothrombin mutation. J Thromb Haemost. 2008;6:2223-2224. 8. Heit JA, Phelps MA, Ward SA, Slusser JP, Petterson TM, De Andrade M. Familial segregation of venous thromboembolism. J Thromb Haemost. 2004;2:731-736. 9. Larsen TB, Sorensen HT, Skytthe A, Johnsen SP, Vaupel JW, Christensen K. Major genetic susceptibility for venous thromboembolism in men: a study of Danish twins. Epidemiology. 2003;14:328-332. 10. Souto JC, Almasy L, Borrell M, Blanco-Vaca F, Mateo J, Soria JM, et al. Genetic susceptibility to thrombosis and its relationship to physiological risk factors: the GAIT study. Genetic Analysis of Idiopathic Thrombophilia. Am J Hum Genet. 2000;67:1452-1459. 0;67:1452-1459. 11. Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen ggen CJ CJ. The Th value valu ue ooff family history as a risk indicator for venous thrombosis. Arch Intern Med. 2009;169:610-615. Med 20 2009 09;1 09 ;169 ;1 69:6 69 :6 610 61 12. de Haan HG, Bezemer ID, Doggen CJ, Cessie S,, R Reitsma AR,, et Multiple n HG G, Beze eme m r ID D, Do Dogg ggen gg en nC J, le le C essiee S es eits ei tsma ts maa PH, PH, Arellano Are r llan anoo AR an A e aal. l. Mu u SNP testingg improves prediction venous thrombosis. Blood. 2012;120:656-663. im mproves risk k pred ed diction o off ffirst on irsst veno nous th hrom ombo bossis.. B bo loood. 20 2012;1 120 20:6 656-6663 13. van Hylckama CA, Rosendaal Baglin of lckama maa V Vlieg lieg g A, A, Ba B Baglin glin gl l C A, Baree LA LA, Ro R send se ndaa nd aall FR aa FR, Ba Bagl gllin in TP. P Proof Proof off of of principle pri pr princ in inc potential clinical thrombosis. J inical utility off multiple mu ult ltip iple ip le SNP SNP analysis ana naly ly ysi siss for for prediction pred pr e iccti ed t onn of of recurrent recu re cu urr rren en nt vvenous enous thromb Thromb Haemost. 2008;6:751-754. a aemost. 2008;6 ; :7 751-754 7 4. 14. Blom JW, CJ, Osanto S, Rosendaal FR. Malignancies, and W Doggen D CJ Os to S Ro ndd l FR M Malignancies ali li cii prothrombotic othhr botii mutations, m ta tation tii the risk of venous thrombosis. JAMA. 2005;293:715-722. 15. Flinterman LE, van Hylckama Vlieg A, Cannegieter SC, Rosendaal FR. Recurrent venous thrombosis: incidence and characteristics in a large cohort of VT patients (MEGA study). J.Thromb.Haemost. 9 [Supplement s2], 42. 2011. Ref Type: Abstract 16. Bezemer ID, Bare LA, Doggen CJ, Arellano AR, Tong C, Rowland CM, et al. Gene variants associated with deep vein thrombosis. JAMA. 2008;299:1306-1314. 17. Arellano AR, Bezemer ID, Tong CH, Catanese JJ, Devlin JJ, Reitsma PH, et al. Gene variants associated with venous thrombosis: confirmation in the MEGA study. J Thromb Haemost. 2010;8:1132-1134. 18. Germer S, Holland MJ, Higuchi R. High-throughput SNP allele-frequency determination in pooled DNA samples by kinetic PCR. Genome Res. 2000;10:258-266. 19. Shiffman D, O'Meara ES, Bare LA, Rowland CM, Louie JZ, Arellano AR, et al. Association of gene variants with incident myocardial infarction in the Cardiovascular Health Study. Arterioscler Thromb Vasc Biol. 2008;28:173-179.
16 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
20. Shiffman D, Ellis SG, Rowland CM, Malloy MJ, Luke MM, Iakoubova OA, et al. Identification of four gene variants associated with myocardial infarction. Am J Hum Genet. 2005;77:596-605. 21. Iakoubova OA, Tong CH, Chokkalingam AP, Rowland CM, Kirchgessner TG, Louie JZ, et al. Asp92Asn polymorphism in the myeloid IgA Fc receptor is associated with myocardial infarction in two disparate populations: CARE and WOSCOPS. Arterioscler Thromb Vasc Biol. 2006;26:2763-2768. 22. Shiffman D, Rowland CM, Louie JZ, Luke MM, Bare LA, Bolonick JI, et al. Gene variants of VAMP8 and HNRPUL1 are associated with early-onset myocardial infarction. Arterioscler Thromb Vasc Biol. 2006;26:1613-1618. 23. Kearon C, Ginsberg JS, Kovacs MJ, Anderson DR, Wells P, Juliann JA, et al. Comparison Comp par aris i o of low-intensity warfarin therapy with conventional-intensity warfarin therapy long-term erap er apyy for ap for lo long ng--term ng te m prevention of recurrent venous thromboembolism. N Engl J Med. 2003;349:631-639. 3;349: 9 63 6311-63 639. 9
17 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
Table 1: The risk of recurrent venous thrombosis associated with 31 individual SNPs previously associated with the risk of a first venous thrombotic event (HR adjusted for age and sex) Gene
SNP
Chrom
Position
Risk allele freq, %
N
HR*
95% CI
HR†
95% CI
F5
rs6025
1
167.785.673
8.3
4098
1.34-1.89
rs1799963
11
46.717.631
2.7
4099
1.23
0.88-1.71
F11
rs3822057
4
187.425.146
54.7
3934
1.22
1.09-1.37
F2
rs3136520
11
46.699.808
2.8
3910
1.20
0.87-1.64
ABO
rs8176719
9
1366.1332.90 13 908 136.132.908
45.4
39077
1.18
1.05-1.34
F13A1
rs5985
6
6.263.79 6. 7944 79 6.263.794
755.55 75.5
4094 94 4094
1.17
1.02-1.34
F11
rs2036914
4
187.429.4775 187.429.475
58.4 58.4
39943 3943
1.15
1.02-1.30
F5
rs4524
1
1677 778 16 778 379 379 167.778.379
77 9 77.9
3942 3942
1.13
0.98-1.30
VWF
rs1063856
12
6.153.534
36.6
3961
1.13
1.00-1.27
F11
rs2289252
4
187.444.375
47.7
3912
1.12
0.99-1.26
SERPINC1
rs2227589
1
172.152.839
11.5
4030
1.11
0.94-1.32
FGG
rs2066865
4
155.744.726
33.4
4099
1.11
0.99-1.25
RGS7
rs670659
1
239.228.398
66.1
4046
1.08
0.96-1.23
PROCR
rs2069952
20
33.227.612
63.4
3914
1.08
0.96-1.22
F9
rs6048
X
138.460.946
72.8
4035
1.37-2.00 0.93-4.13 0.85-1.69 0.36-18.1 1.04-1.67 1.19 1. 19 1.19-1.96 0.81 0. 0.81-1.61 0 60 6 0.60-9.65 1.033 1.03-1.54 1 088 1. 1.08-1.78 0.888 0. 0.88-1.96 1.000 1.00-2.19 1.000 1.00-1.63 1.07 07 7 1.07-1.78 0 63 0.63-1.37 0.76-1.62 0.91-1.30 1.01-1.65 0.84-1.26 0.99-1.57 0.90-1.33 0.77-2.41 0.81-1.14 1.07-1.73 0.89-1.59 0.93-1.66 0.91-1.54 0.94-1.60 0.85-1.62 0.93-1.43
1.60
F2
1.66 1.96 1.20 2.54 1.32 1.53 1.53 1.14 14 1.14 2 40 2.40 1.26 1.39 39 1.39 31 1.31 1.48 4 1.48 1.28 1.28 1.38 38 1.38 0 93 93 0.93 1.11 1.09 1.29 1.03 1.25 1.09 1.36 0.96 1.36 1.19 1.24 1.19 1.22 1.17 1.15
1.06
0.96-1.18
18 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
DOI: 10.1161/CIRCGENETICS.114.000682
PROCR
rs867186
20
33.228.215
14.1
4030
NAT8B
rs2001490
2
73.781.606
38.8
4057
PROC
rs1799809
2
127.892.345
47.6
4058
F3
1208 indel
1
94.780.000
46.9
3959
F13B
rs6003
1
195.297.644
10.2
3840
F8
rs1800291
X
153.811.479
85.2
4100
TFPI
rs8176592
2
188.040.937
68.9
3915
GP6
rs1613662
19
60.2288.407 60 60.228.407
83.8
52 4052
F2
rs3136516
11
46.7177.3 46 .332 3 32 46.717.332
52 6 52.6
40085 4085
HIVEP1
rs169713
6
11.920.5177
21 9 21.9
38887 3887
NR1I2
rs1523127
3
1200 983 12 983 729 729 120.983.729
40 4 40.4
4061 4061
STXBP5
rs1039084
6
147.635.413
42.2
3989
F9
rs4149755
X
138.451.778
6.7
3913
CPB2
rs3742264
13
45.546.095
68.6
3893
PROCR
rs2069951
20
33.277.425
6.7
3914
F5
rs1800595
1
167.776.972
5.8
3959
1.08 1.00 1.15 1.04 1.04 1.09 1.02 1.06 0.96 1.45 1.09 1 08 1. 1.08 0.97 0. 97 0.97 1 03 1.03 0.98 98 0.98 88 0.88 0.97 0.90 0.90 1 04 1. 1.04 0 86 0.86 0.90 0.86 0.87 0.89 0.82 0.92 0.76 0.86 0.62 0.81
0.90-1.30 0.57-1.73 0.97-1.37 0.81-1.33 0.87-1.27 0.87-1.37 0.84-1.24 0.84-1.34 0.77-1.20 0.82-2.57 0.73 7 0.73-1.64 0.82 0. 82 0.82-1.44 0.7 73 0.73-1.31 0 0.77 0.77-1.38 0.600 0.60-1.62 0 611 0. 0.61-1.60 0.73 0. 3 0.73-1.08 0.788 0.78-1.20 0.75 5 0.75-1.07 0.711 0.71-1.52 0 72 0.72-1.03 0.71-1.14 0.72-1.03 0.69-1.10 0.58-1.34 0.52-1.28 0.70-1.20 0.58-1.00 0.66-1.11 0.15-2.47 0.61-1.07
0.76
0.11-5.44
SNP: Single nucleotide polymorphism, Chrom: Chromosome, HR: Hazard Ratio * Hazard ratio for heterozygous and homozygous carriers of risk allele versus homozygous carriers of non-risk allele † Hazard ratio for the risk allele using an additive model
19 Downloaded from http://circgenetics.ahajournals.org/ at New York University/ Medical Center--New York on October 17, 2014
1.05
0.90-1.23
1.05
0.93-1.17
1.05
0.94-1.17
1.03
0.92-1.16
1.03
0.86-1.24
1.03
0.91-1.18
1.03
0.91-1.17
1.00
0.86-1.16
0.99
0.88-1.10
0.95
0.82-1.09
0.93
0.83-1.04
0.92
0.82-1.03
0.90
0.74-1.10
0.86
0.76-0.97
0.85
0.66-1.08
0.82
0.62-1.07
DOI: 10.1161/CIRCGENETICS.114.000682
Table 2. 4-year and 6-year cumulative risk of recurrence using different genetic risk scores # risk alleles
N total (%)
N rec
4-year cumulative incidence (95% CI)
6-year cumulative incidence (95% CI)
Incidence rate (95% CI)
31-SNP GRS
