Clinical features of cerebral cavernous malformations patients with KRIT1 mutations.код для вставкиСкачать
Clinical Features of Cerebral Cavernous Malformations Patients with KRIT1 Mutations Christian Denier, MD,1,2 Pierre Labauge, MD,1,3 Laurent Brunereau, MD,4 Florence Cavé-Riant, MD,2 Florence Marchelli, MD,1 Minh Arnoult,1 Michaelle Cecillon,1 Jacqueline Maciazek,1 Anne Joutel, MD, PhD,1,2 Elisabeth Tournier-Lasserve, MD,1,2 and the Société Française de Neurochirurgie and the Société de Neurochirurgie de Langue Française Cerebral Cavernous Malformations (CCM/OMIM 604214) are vascular malformations causing seizures and cerebral hemorrhages. They occur as a sporadic and autosomal dominant condition, the latter being characterized by the presence of multiple CCM lesions. Stereotyped truncating mutations of KRIT1, the sole CCM gene identified so far, have been identified in CCM1 linked families but the clinical features associated with KRIT1 mutations have not yet been assessed in a large series of patients. We conducted a detailed clinical, neuroradiological and molecular analysis of 64 consecutively recruited CCM families segregating a KRIT1 mutation. Those families included 202 KRIT1 mutation carriers. Among the 202 KRIT1 mutation carriers, 126 individuals were symptomatic and 76 symptom-free. Mean age at clinical onset was 29.7 years (range, 2–72); initial clinical manifestations were seizures in 55% of the cases and cerebral hemorrhages in 32%. Average number of lesions on T2 weighted MRI was 4.9 (ⴞ7.2) and on gradient echo sequences 19.8 (ⴞ33.2). Twenty-six mutation carriers harbored only one lesion on T2-weighted MRI, including 4 mutation carriers, aged from 18 to 55 yr-old, who presented only one CCM lesion both on T2-weighted and on highly sensitive gradient echo MRI sequences. Five symptom free mutation carriers, aged from 27 to 48 yr-old, did not have any detectable lesion both on T2WI and gradient echo MRI sequences. Within KRIT1/CCM1 families, both clinical and radiological penetrance are incomplete and age dependent. Importantly for genetic counseling, nearly half of the KRIT1 mutation carriers aged 50 or more are symptom-free. The presence of only one lesion, even when using gradient echo MRI sequences, can be observed in some patients with an hereditary form of the disease. Incomplete neuroradiological penetrance precludes the use of cerebral MRI to firmly establish a non carrier status, even at an adult age and even when using highly sensitive gradient echo MRI. Altogether these data suggest that the hereditary nature of the disorder may be overlooked in some mutation carriers presenting as sporadic cases with a unique lesion. Ann Neurol 2004;55:213–220 Cerebral cavernous malformations (CCMs/OMIM 116860) are vascular malformations, mostly located in the central nervous system and characterized by abnormally enlarged capillary cavities without intervening brain parenchyma.1 From a large series based on necropsy and/or magnetic resonance imaging (MRI), their prevalence in the general population has been estimated to be close to 0.1 to 0.5%.2,3 Most common symptoms include seizures (38 –51% of cases), cerebral hemorrhages (11–32%), and focal neurological deficits (12– 45%).4 Cavernous angiomas can occur as a sporadic or an From 1INSERM E365, Lariboisière Medical School, Paris; 2Cryogenetics Laboratory, Hôpital Lariboisière, Assistance Publique-Hôpitaux de Paris; 3Neurology Department, CHU Montpellier Nimes; and 4 Radiology Department, CHU de Tours, Tours, France. Received Jul 29, 2003, and in revised form Sep 4. Accepted for publication Sep 4, 2003. autosomal dominant condition. The frequency of the familial form has been estimated as high as 50% in Hispano-American CCM patients5 and close to 10 to 20% in white patients (E. Tournier-Lasserve, unpublished data). Usually, sporadic cases harbor only one CCM lesion, whereas familial cases are characterized by the presence of multiple lesions whose number is strongly correlated to patients’ age.5,6 “Sporadic” cases harboring multiple lesions have been shown to be affected by an hereditary form of the disease because 75% of them have an asymptomatic parent with CCM lesions on brain MRI, due to the incomplete clinical Address correspondence to E. Tournier-Lasserve, INSERM E365, Faculté de médecine Lariboisière, 10, avenue de Verdun, 75010 Paris, France. E-mail: email@example.com The main clinical investigators are listed in the Appendix on page 219. © 2003 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services 213 penetrance of this condition.5,6 A spontaneous de novo mutation of KRIT1/CCM1 gene also has been reported in one apparently sporadic CCM patient with multiple lesions.7 Previous genetic linkage analyses identified three CCM loci on 7q (CCM1), 7p (CCM2), and 3q (CCM3). The proportions of families linked to these loci have been estimated to be close to 40% (CCM1), 20% (CCM2), and 40% (CCM3).8,9 The sole CCM gene identified so far, CCM1, encodes Krit1, a 736 – amino acid protein containing four ankyrin domains and a Four point 1 (4.1) protein/Ezrin/Radixin/Moesin domain.10,11 This protein has been shown to interact in vitro with Rap1A, a small GTPase and icap1␣, a modulator of ␤1 integrin signal transduction pathway.12–14 So far, 74 distinct KRIT1 mutations causing CCMs have been published including 42 from our laboratory.7,10,11,15–21 They are highly stereotyped because all of them lead to premature termination codons, suggesting that these mutations have the same consequence at the protein level and that Krit1 haploinsufficiency, through mRNA decay, is the most likely pathophysiological mechanism involved in CCM1 patients.22 Interestingly, a highly recurrent KRIT1 mutation has been detected in Hispano-American CCM patients in whom a strong founder effect was reported.23 Several clinical and neuroradiological studies conducted before CCM1/KRIT1 gene identification helped to delineate some of the main features of this condition, namely, its autosomal dominant pattern of inheritance, its incomplete clinical penetrance, and the high frequency of multiple lesions. These studies also showed the intrafamilial and interfamilial variability of familial CCM.5,6,24 Despite the high number of KRIT1 mutations reported so far, the phenotypes associated with the presence of a KRIT1 mutation in CCM patients have not yet been documented in a large series of white patients. We consecutively recruited 153 CCM probands who had at least one affected relative with CCM and/or presented multiple cerebral cavernous angiomas. KRIT1 gene mutation screening of these 153 probands led to the identification of 64 mutation carriers. We genotyped herein 313 family members of these 64 KRIT1-positive index patients and report the clinical and neuroradiological features of the 202 KRIT1 mutation carriers identified within these families. Patients and Methods Patients and Families Twenty-eight French University Neurosurgery centers participated in this study which was approved by the local ethics committee. A total of 153 unrelated clinically affected CCM index patients were consecutively enrolled on the basis of one of the two following criteria: to have at least one affected 214 Annals of Neurology Vol 55 No 2 February 2004 relative and/or to have multiple cerebral cavernous angiomas. Detailed clinical and neuroimaging information was collected for all index patients through direct interview and reviewing of medical charts. Clinical assessment included information on cerebral hemorrhages, focal neurological symptoms, seizures, and headaches (when occurring in the absence of any hemorrhage detectable on cerebral imaging). Hemorrhage diagnosis was based on the assessment of clinical symptoms and MRI and was confirmed in most cases by neurosurgery. CCM diagnosis was based on cerebral MRI features and/or pathological analysis. Pedigrees were established with the help of the proband of each family. In addition to clinical information provided by the probands on their relatives, all symptomatic and symptom-free at-risk relatives who provided their written informed consent were directly interviewed by one of us (P.L.) using a standardized procedure. MRI data were retrospectively collected for all symptomatic patients. Symptom-free subjects who gave their informed consent underwent cerebral sagittal T1-weighted and axial T2-weighted MRI. In addition, most of these symptom-free patients underwent MRI gradient-echo sequences investigation. MRI data were independently reviewed by P.L. and L.B. We considered as “affected” all subjects showing cavernous angiomas on cerebral MRI whatever their clinical status. This group included symptomatic and symptom-free subjects. We considered as “healthy” all subjects with a normal MRI. Those who did not undergo MRI screening were classified as “unknown.” Clinical and MRI information for 22 of the 64 families included in this study had been reported in part elsewhere before KRIT1 gene identification.6 Molecular Analysis Genomic DNA from each index patient and all consenting relatives was extracted from peripheral blood using standard procedures. Genomic DNA from 100 unrelated healthy white French subjects was available as a control group. A combination of single-strand conformation polymorphism (SSCP) and sequencing have been used to screen the 16 coding exons and exon–intron boundaries of KRIT1 gene in all 153 probands as described elsewhere.20 Molecular screening data of 121 of these probands were reported previously.20 Thirty-two additional probands were screened herein using the same procedure. This screening led us to identify a deleterious mutation in 64 of these 153 probands, a proportion which is in agreement with previously reported linkage data.9 All these mutations led to a premature stop codon through nonsense, splicing, or frameshift mutations, strongly suggesting that the consequences of these mutations at the protein level would be similar (Fig 1).20 Fig 1. Genealogical trees of families with KRIT1 mutations. The pedigrees are numbered (Cn) and KRIT1 mutations are indicated. Numbering of KRIT1 nucleotides (nt) is according to the full-length KRIT1 cDNA (accession number AF296765), beginning nucleotide numbering at the A of the ATG initiator codon. MRI ⫽ magnetic resonance imaging; CCM ⫽ cerebral cavernous malformation. ‹ Figure 1 Denier et al: Cerebral Cavernous Malformations 215 Table 1. Clinical and Neuroradiological Features in 202 KRIT1 Mutations Carriers All Subjects N ⫽ 202 Feature Symptomatic subjects Age of patients (yr) Mean ⫾ SD Range Age of onset (yr) Mean ⫾ SD Range No. of CCM lesions On T2-weighted MRI ⫾ SD On gradient-echo sequence ⫾ SD Mean age at MRI (yr) Types of first clinical manifestations (%) Cerebral hemorrhages–no/total no Seizures–no. total no. Others–no. total no. Asymptomatic subjects Age of patients (yr) Mean ⫾ SD Range Number of CCM lesions On T2-weighted MRI ⫾ SD On gradient-echo sequence ⫾ SD Mean age at gradient-echo MRI a Males N ⫽ 98 Females N ⫽ 104 N ⫽ 126 N ⫽ 62 N ⫽ 64 41.8 ⫾ 17.9 6–90 40.7 ⫾ 18.4 6–80 42.9 ⫾ 17.4 9–90 29.7 ⫾ 16.6 2–72 27.5 ⫾ 16 5–72 32.1 ⫾ 17 2–67 5.0 ⫾ 6.7 (n ⫽ 79)a 18.2 ⫾ 28 (n ⫽ 37) 36.8 ⫾ 17.2 4.0 ⫾ 3.6 (n ⫽ 39) 15.1 ⫾ 16.2 (n ⫽ 17) 36.4 ⫾ 17.4 6.0 ⫾ 8.7 (n ⫽ 40) 20.8 ⫾ 35.3 (n ⫽ 20) 37.3 ⫾ 17.2 32% (n ⫽ 111) 55% 13% N ⫽ 76 29% (n ⫽ 58) 60% 11% N ⫽ 36 34% (n ⫽ 53) 49% 17% N ⫽ 40 46.8 ⫾ 17.3 10–86 47 ⫾ 18.3 16–86 46.6 ⫾ 16.7 10–86 4.8 ⫾ 8.0 (n ⫽ 53) 21.1 ⫾ 37.4 (n ⫽ 43) 42.8 ⫾ 19.8 6.7 ⫾ 11 (n ⫽ 24) 24.1 ⫾ 40.6 (n ⫽ 17) 41 ⫾ 17.5 3.2 ⫾ 3.7 (n ⫽ 29) 19.2 ⫾ 35.8 (n ⫽ 26) 44.2 ⫾ 14.5 The number between brackets indicates the number of individuals for whom data were available. SD ⫽ standard deviation; CCM ⫽ cerebral cavernous malformations; MRI ⫽ magnetic resonance imaging. Based on these data, SSCP analysis was used herein to genotype all available and consenting members of the 64 KRIT1-positive families including 126 symptomatic CCM patients, 187 at-risk asymptomatic relatives, and 64 spouses. In addition to SSCP screening, genomic DNA sequencing was used for confirmation of “mutation carrier” status in all symptom-free patients with a normal MRI. Statistical Analysis Student t and 2 tests were utilized for comparison of age at inclusion, age of onset, and number of CCM lesions on MRI. A p value less than 0.05 was considered significant. Plus or minus values are means ⫾ SD. Results Clinical and Neuroradiological Features of the 202 KRIT1 Mutation Carriers A total of 202 KRIT1 mutation carriers was identified including the 64 probands and 138 relatives. Among mutation carriers, there were 98 men and 104 women. Mean age at inclusion was similar in men and women and close to 40 years old (Table 1). Among these mutation carriers, 126 were symptomatic. There was no difference when comparing the clinical manifestations present in the probands and their clinically affected relatives, and therefore these two groups were considered as a whole (data not shown). 216 Annals of Neurology Vol 55 No 2 February 2004 Mean age at clinical onset was 29.7 ⫾ 16.6 years (range, 2–72). Mean age at clinical onset was slightly higher in women, but this difference did not reach statistical significance (see Table 1). Grouping of patients according to clinical age of onset in three age groups (0 –24, 25– 44, and ⱖ45 years) showed that half of the patients presented their first neurological symptoms before 25 years of age and 20% after 45 years old (Table 2). Interestingly, comparison of male/female proportions within these three age groups showed a significant deviation toward a higher age of onset in women ( p ⬍ 0.05). This later age of onset in women was not associated with a propensity to a particular type of initial clinical event. MEAN AGE AT CLINICAL ONSET. SYMPTOMS AND FUNCTIONAL OUTCOME. In the whole group of the 126 symptomatic patients, seizures were the initial events in 55% of the patients; 47% of the patients had generalized seizures, 43% had partial seizures, and 10% had both types. Cerebral hemorrhage was the second most common event (32%). At time of inclusion, 15% of the patients had presented more than one hemorrhagic event (two to four symptomatic hemorrhages). For patients who did not present with seizures or hemorrhage, initial symptoms were focal signs (9%) and headache (4%). Mean age of onset of patients in whom seizures were the initial event was Table 2. Comparison of Male/Female Age Range at Clinical Onset in KRIT1 Mutation Carriers Age Groups (yr)a Group 0–24 25–44 ⱖ 45 Symptomatic men (N ⫽ 58)b Symptomatic women (N ⫽ 53) All symptomatic patients (N ⫽ 111) 52% 38% 45% 38% 30% 34% 10% 32% 21% a Comparison of the proportions of males and females within these three age groups showed a significant deviation toward a higher age of onset in females ( p ⬍ 0.05). b The number between brackets indicates the number of individuals for whom data were available. slightly higher than in patients with hemorrhage (28.5 ⫾ 15.2 years old as compared with 24.4 ⫾ 15.2 years old in patients whose first event was hemorrhage), but this difference was not statistically significant. These two groups of patients did not differ for the number of CCM lesions. At last, in addition to cerebral lesions, 10 patients from 5 of these 64 families showed cutaneous hyperkeratotic vascular lesions that have been reported previously to be associated with cerebral cavernous angiomas.24 Clinical disability at time of inclusion was assessed according to neurological handicap: 89% of the whole group of symptomatic patients were fully independent, 10% were handicapped, and 1% were bedridden. Comparison of sex ratio, symptoms at clinical onset, and mean number of lesions did not show any difference between independent and disabled patients. For the 61 patients in whom seizures were the initial event, the grade of epilepsy disability was evaluated according to patient’s response to anticonvulsant drugs: 91% were controlled by anticonvulsants, whereas 9% had intractable epilepsy. We did not detect any difference for age or number of lesions between these two groups of epileptic patients. CEREBRAL MAGNETIC RESONANCE IMAGING FEATURES. T2-weighted standard MRI sequences were available for 132 KRIT1 mutation carriers (including 79 symptomatic subjects) and gradient-echo sequences for 80 (37 symptomatic subjects). Classically, familial CCM patients have multiple CCM lesions, whereas nonfamilial sporadic cases harbor only one CCM lesion. In this panel, 80% of KRIT1 mutation carriers had multiple lesions on T2weighted sequences and 90% multiple lesions on highly sensitive gradient-echo sequence MRI investigations. The average number of lesions per subject was 4.9 ⫾ 7.2 on standard MRI and 19.8 ⫾ 33.2 on gradient-echo sequences ( p ⬍ 0.001). There was no significant difference in the number of lesions in subjects with and without symptom on both T2-weighted MRI sequences (5.0 ⫾ 6.7 vs 4.8 ⫾ 8.0) and gradientecho MRI sequences (18.2 ⫾ 28.0 vs 21.1 ⫾ 37.4; see Table 1). Of 132 mutation carriers for whom MRI investigation was available, 26 subjects (20%) harbored only one lesion on T2-weighted MRI (mean age at MRI, 29.4 ⫾14.8 years; range, 5–50). Among them, 12 subjects underwent gradient-echo sequence MRI, which disclosed multiple lesions in eight cases; the last four subjects, one man and three women, aged, respectively, 18, 34, 34, and 55 years old, were all symptom-free and presented only one CCM lesion both on T2weighted and on gradient-echo sequences MRI (see Fig 1). At last, eight symptom-free mutation carriers had a normal T2-weighted MRI (see next paragraph). There was a strong correlation between age and number of lesions on T2-weighted and gradient-echo sequences. Among subjects younger than 45 years old, the average number of lesions on T2-weighted sequences was 3.3 ⫾ 3.4 (range, 0 –19; n ⫽ 85), whereas it was 8.0 ⫾10.7 in subjects aged 45 years and older (range, 0 – 43; n ⫽ 47; p ⬍ 0.001). On gradient-echo sequences, the mean number of lesions was 10.6 ⫾ 20.2 in subjects younger than 45 years (range, 0 –130; n ⫽ 51), whereas it was 35.9 ⫾ 44.3 in subjects older than 45 years (range, 0 –154; n ⫽ 29; p ⬍ 0.001). INCOMPLETE CLINICAL AND NEUROIMAGING PEN- Among the 249 sampled at-risk relatives of the 64 probands carrying a KRIT1 mutation, 138 were mutation carriers, including 62 symptomatic subjects and 76 symptom-free subjects. Comparison of symptomatic and asymptomatic subjects for mean age at inclusion and sex ratio did not show any significant difference. Among these 76 symptom-free mutation carriers, 53 underwent an MRI investigation; CCM lesions were detected in 45 of them. Among the eight symptom-free subjects who did not show any CCM lesions on T2weighted imaging, two subjects, a 44-year-old man and a 67-year-old woman, showed on gradient-echo sequences MRI, respectively, nine and three hyposignals strongly suggestive of type IV lesions according to Zabramski’s classification25; of the six remaining symptom-free subjects, five subjects underwent gradient-echo sequence MRIs that were strictly normal. In addition to carrying the mutation identified within their affected family members, these five symptom-free KRIT1 mutation carriers had been previously shown to carry the disease CCM1-linked haplotype within their respective families10 (data not shown). These five subjects included a 27-year-old man and four women aged 29, 33, 35, and 48 years old (Figs 1 and 2). ETRANCE. Denier et al: Cerebral Cavernous Malformations 217 Fig 2. Incomplete clinical and neuroradiological penetrance in a KRIT1 mutation carrier. (top) Part of the genealogical tree of Family C005 is shown: men (squares), women (circles), healthy with a normal magnetic resonance image (MRI): (open symbols), clinically affected: (filled symbols), and symptom-free patients with CCM lesions on MRI: (half-filled symbols). The numbering in this part of Family C005 is according to the single-strand conformation polymorphism (SSCP) gel lanes. The proband of this family harbors a C to T nucleotide substitution at nucleotide 1943 within exon 17 which emerged as an abnormal SSCP conformer (lane 3). All members of the family showing CCM lesions (Patients 3, 5, 6, 8) carry the same abnormal conformer that was detected in the proband and that was absent in healthy subjects with the exception of Subject 2, a 48-year-old female mutation carrier whose MRI, including gradient-echo sequences, was strictly normal (bottom panel, showing cerebral gradient-echo sequences MRI of Subjects 2 [normal MRI] and 8 [CCM temporal lesion]). Discussion Before CCM gene mapping and CCM1/KRIT1 gene identification, several studies allowed establishment of 218 Annals of Neurology Vol 55 No 2 February 2004 some of the major clinical and neuroimaging features of familial cavernous angiomas, namely, its autosomal dominant pattern of inheritance, its incomplete clinical penetrance, and the multiplicity of cerebral CCM lesions. Another major feature of this genetically heterogenous condition is its clinical variability both within a given family and from one family to another. The features associated with KRIT1 mutations had not yet been assessed in a large series of patients. Herein, we characterized the clinical and neuroimaging features of this condition in 64 families in which segregates a mutation within the KRIT1/CCM1 gene, all these mutations being highly stereotyped and leading to a premature termination codons and most likely to Krit1 haploinsufficiency. Clinical penetrance in familial CCM was reported to be incomplete. In our panel of KRIT1-positive families, clinical penetrance was also incomplete. The proportion of symptomatic subjects among mutation carriers in this panel was 62%. Importantly for genetic counseling, only half of the sampled mutation carriers subjects aged 50 years or older at inclusion had neurological symptoms that were CCM related (58%). These proportions should be considered as a maximum estimate because not all asymptomatic subjects have been genotyped. Neuroimaging penetrance of CCM was considered previously to be much higher and complete or almost complete. So far, only four obligate transmitters of the disease with a normal T2-weighted MRI scan have been reported6,26,27; sensitive gradient-echo MRI scan was not performed in those patients and, indeed, to our knowledge, there is no published case of an obligate asymptomatic carrier with a normal gradient-echo MRI. Molecular screening allowed us to establish that neuroimaging penetrance is also incomplete, even when using the highly sensitive gradient-echo sequence investigation. Indeed, 5 of our 76 asymptomatic mutation carriers, the oldest one being 48 years old, did not show any lesion on gradient-echo MRI. Previous studies have shown that most patients affected with familial CCM harbor multiple cerebral lesions, and that apparently isolated patients showing multiple lesions are in reality affected with an hereditary form of the disease. In this panel, at least 13% of KRIT1 mutation carriers had only one CCM lesion on T2-weighted sequences and 2% had only one lesion on gradient-echo sequences. These data, combined with the incomplete clinical penetrance of this disorder, strongly suggest that the familial nature of cavernous angiomas may be overlooked in some patients. Additional information was drawn from male/female KRIT1 mutation carriers comparison. The proportions of affected subjects and the nature of the clinical manifestations were similar in these two groups except for a higher proportion of women in the age groups of patients with a late onset (after 45 years old). Although these data need confirmation in a second panel of families, they raise interesting questions. Previous studies including mostly sporadic cases suggested that men have an earlier age of onset.2,28,29 Other studies suggested the role of hormonal factors which may influence the biological behavior of cavernous angiomas in women, in particular, during pregnancy and under hormonal treatment.2,29 –32 So far, to our knowledge, influence of menopause on cerebral cavernous angiomas evolution has not been studied. Studies of larger cohorts of female patients carrying a KRIT1 mutation should allow to better delineate the natural history of the disease regarding main hormonal events. At last, a follow-up of this cohort of genotyped patients and further genetic studies should also help in the future to delineate environmental factors or modifying genes explaining the intrafamilial and interfamilial variability of familial CCM. Appendix The main clinical investigators were Drs Barbieux-Vaquez, Canaple, Le Gars (Amiens), de Bray, Fournier, Guy, PenissonBesnier (Angers), Bergouignan (Bayonne), Beuriat (Beaune), Thibaut (Berck), Bizette, Czorny, Godard, Jacquet (Besançon), Comoy, David, Parker, Said (Bicêtre), Castel, Guerin, Loiseau (Bordeaux), Houtteville, Khoury, Viader (Caen), Attané, Tannier (Carcassonne), Bréchard (Chalons/Saone), Carriere, Chazal, Clavelou (Clermont-Ferrand), Nivelon (Dijon), Benabid, Perret (Grenoble), Latinville (La Rochelle), Born (Liège), Christiaens, Combelles, Lejeune, Louis, Pasquier (Lille), Moreau (Limoges), Fischer, Froment, Guyotat, Mortelese, Sindou (Lyon), Robert, Sauvage, Savet (Macon), Attarian, Gomez, Lena (Marseille), Richard (Montbéliard), Cesari, Coubes, Idee, Frerebeau (Montpellier), Rey (Narbonne), Carriere (Nimes), Hepner, Lescure, Vespignani, Weber (Nancy), Martin, Mussini, Resche (Nantes), Borg, Grellier, Thomas, Lonjon (Nice), Masson, Redondo, Rey (Paris, Beaujon), Delalande (Paris), Georges, Lot, Hagueneau, Sarrazin, Silhouette (Paris, Lariboisière), Zerah, Renier, Pierre-Kahn (Paris, Necker), Baulac, Capelle, Arthuis, Clemenceau, Faillot, Fohano, Philippon (Paris, La Salpétrière), Gil, Lapierre, Neau (Poitiers), Rousseaux, Gaillard (Reims), Carsin, Dufour, Guegan, Seigneuret, Scarabin (Rennes), Drouin-Garraud, Rossi, Frebourg, Freger (Rouen), Brunon, Garnier, Michel (SaintEtienne), Esposito, Metraut (Strasbourg), Aesch, Maheut Lourmière, François, Jan, Velut (Tours), Arrue, Bousquet, Chaix, Clanet, Fabre, Geraud, Lazorthes, Manelfe, Tremoulet (Toulouse), and Joyeux (Valence). This work was supported by an INSERM/AFM/MESR grant (French National Institute for Health and Medical Research/French Association against Myopathies/French Research Ministry, AAE00013HSA, E.T.L.). C.D. is a recipient of a fellowship from Poste Accueil INSERM. We thank all families for their participation. We are also indebted to all members of the Société Française de Neurochirurgie and Société de Neurochirurgie de Langue Française who participated in this study. References 1. Russel DS, Rubinstein LJ. Pathology of tumors of the nervous system. 5th ed. Baltimore: Williams and Wilkins, 1989: 730 –736. 2. Robinson JR, Awad IA, Little JR. Natural history of the cavernous angiomas. J Neurosurg 1991;75:709 –714. 3. Otten P, Pizzolato GP, Rilliet B, Berney J. 131 cases of cavernous angioma (cavernomas) of the CNS, discovered by retrospective analysis of 24,535 autopsies. Neurochirurgie 1989;35: 82– 83, 128 –131. 4. Hsu FPK, Rigamonti D, Huhn SL. Epidemiology of cavernous malformations. In: Awad IA, Barrow DL, eds. Park Ridge, IL: American Association of Surgeons Publications Committee, 1993:13–23. 5. Rigamonti D, Hadley MN, Drayer BP, et al. Cerebral cavernous malformations. Incidence and familial occurrence. N Engl J Med 1988;319:343–347. 6. Labauge P, Laberge S, Brunereau L, et al. Hereditary cerebral cavernous angiomas: clinical and genetic features in 57 French families. Société Française de Neurochirurgie. Lancet 1998;352: 1892–1897. 7. Lucas M, Costa AF, Montori M, et al. Germline mutations in the CCM1 gene, encoding KRIT1, cause cerebral cavernous malformations. Ann Neurol 2001;49:529 –532. 8. Dubovsky J, Zabramski JM, Kurth J, et al. A gene responsible for cavernous malformations of the brain maps to chromosome 7q. Hum Mol Genet 1995;4:453– 458. 9. Craig HD, Gunel M, Cepeda O, et al. Multilocus linkage identifies two new loci for a mendelian form of stroke, cerebral cavernous malformation, at 7p15–13 and 3q25.2-27. Hum Mol Genet 1998;7:1851–1858. 10. Laberge-le Couteulx S, Jung HH, Labauge P, et al. Truncating mutations in CCM1, encoding KRIT1, cause hereditary cavernous angiomas. Nat Genet 1999;23:189 –193. 11. Sahoo T, Johnson EW, Thomas JW, et al. Mutations in the gene encoding KRIT1, a Krev-1/rap1a binding protein, cause cerebral cavernous malformations (CCM1). Hum Mol Genet 1999;8:2325–2333. 12. Serebriiskii I, Estojak J, Sonoda G, et al. Association of Krev1/rap1a with KRIT1, a novel ankyrin repeat-containing protein encoded by a gene mapping to 7q21–22. Oncogene 1997;15: 1043–1049. 13. Zhang J, Clatterbuck RE, Rigamonti D, et al. Interaction between Krit1 and Icap1 infers perturbation of integrin 〉1mediated angiogenesis in the pathogenesis of cerebral cavernous malformation. Hum Mol Genet 2001;10:2953–2960. 14. Zawistowski JS, Serebriiski IG, Lee MF, et al. Krit1 association with the integrin-binding protein icap1: a new direction in the elucidation of cerebral cavernous malformations (CCM1) pathogenesis. Hum Mol Genet 2001;11:389 —396. 15. Davenport WJ, Siegel AM, Dichgans J, et al. CCM1 gene mutations in families segregating cerebral cavernous malformations. Neurology 2001;56:540 –543. 16. Zhang J, Clatterbuck RE, Rigamonti D, Dietz HC. Mutations in KRIT1 in familial cerebral cavernous malformations. Neurosurgery 2000;46:1272–1277. 17. Sahoo T, Goenaga-Diaz E, Serebriiskii IG, et al. Computational and experimental analyses reveal previously undetected coding exons of the KRIT1 (CCM1) gene. Genomics 2001;71: 123–126. 18. Eerola I, Plate KH, Spiegel R, et al. KRIT1 is mutated in hyperkeratotic cutaneous capillary-venous malformation associated with cerebral capillary malformation. Hum Mol Genet 2000; 22:1351–1355. Denier et al: Cerebral Cavernous Malformations 219 19. Chen DH, Lipe HP, Qin Z, Bird TD. Cerebral cavernous malformation: novel mutation in a Chinese family and evidence for heterogeneity. J Neurol Sci 2002;196:91–96. 20. Cavé-Riant F, Denier C, Labauge P, et al. Spectrum and expression analysis of KRIT1 mutations in 121 consecutive and unrelated patients with cerebral cavernous malformations. Eur J Hum Genet 2002;10:733–740. 21. Marini V, Ferrera L, Dorcaratto A, et al. Identification of a novel KRIT1 mutation in an Italian family with cerebral cavernous malformation by the protein truncation test.J Neurol Sci 2003;212:75–78. 22. Frischmeyer PA, Dietz HC. Nonsense-mediated mRNA decay in health and disease. Hum Mol Genet 1999;8:1893–1900. 23. Günel M, Awad IA, Finberg K, et al. A founder mutation as a cause of cerebral cavernous malformations in HispanicAmericans. N Engl J Med 1996;334:946 –951. 24. Labauge P, Enjolras O, Bonerandi JJ, et al. An association between autosomal dominant cerebral cavernomas and a distinctive hyperkeratotic cutaneous vascular malformation in 4 families. Ann Neurol 1999;45:250 –254. 25. Zabramski JM, Wascher TM, Spetzler RF, et al. The natural history of familial cavernous malformations: results of an ongoing study. J Neurosurg 1994;80:422– 432. 220 Annals of Neurology Vol 55 No 2 February 2004 26. Malik S, Cohen BH, Robinson J, et al. Progressive vision loss. A rare manifestation of familial cavernous angiomas. Arch Neurol 1992;49:170 –173. 27. Polymeropoulos MH, Hurko O, Hsu F, et al. Linkage of the locus for cerebral cavernous hemangiomas to human chromosome 7q in four families of Mexican-American descent. Neurology 1997;48:752–757. 28. Robinson JR, Awad IA, Magdinec M, Paranandi L. Factors predisposing to clinical disability in patients with cavernous malformations of the brain. Neurosurgery 1993;32:730 –735. 29. Maraire JN, Awad IA. Intracranial cavernous malformations: lesion behavior and management strategies. Neurosurgery 1995; 37:591– 605. 30. Perez Lopez-Fraile I, Tapiador Sanjuan MJ, Eiras Ajuria J, Gimenez Mas JA. Cerebral cavernous angiomas in pregnancy. Two cases and a review of literature. Neurologia 1995;10: 242–245. 31. Pozzati E, Acciarri N, Tognetti F, et al. Growth, subsequent bleeding, and de novo appearance of cerebral cavernous angiomas. Neurosurgery 1996;38:662– 669. 32. Gazzaz M, Sichez J, Capelle L, Fohanno D. Recurrent bleeding of thalamic cavernous angioma under hormonal treatment. A case report. Neurochirurgie 1999;45:413– 416.