Acta Endosc (2008) 4: 359-374 © Springer 2008 DOI 10.1007/s101900800018
MISE AU POINT / UPDATE
Maladies Inflammatoires Chroniques Intestinales (MICI) : quelle place pour les traitements conventionnels ? L. Beaugerie Service de gastroentérologie et nutrition, hôpital Saint-Antoine et université Pierre et Marie Curie, 184 rue du Faubourg Saint-Antoine, F-75571 Paris cedex 12, France
Résumé : Les traitements conventionnels des MICI incluent les dérivés 5-amino-salicylés, les corticoïdes et les immunosuppresseurs « conventionnels » (par opposition aux traitements biotechnologiques), appelés « immunomodulateurs » dans la sémantique anglo-saxonne et comprenant essentiellement les thiopurines, le méthotrexate, la ciclosporine et ses analogues. Nous insisterons sur les indications résiduelles des corticoïdes et les développements récents sur les indications et le bon usage des thiopurines et du méthotrexate, en nous appuyant notamment sur les récentes recommandations européennes et américaines [1-5]. La place des traitements conventionnels s’inscrit dans une stratégie thérapeutique d’ensemble des MICI qui tend à être globalement plus ambitieuse et « agressive », du fait de la diversification de l’arsenal thérapeutique et du constat de relatif échec des stratégies adoptées auparavant [6]. L’objectif actuel est d’obtenir chez la majorité des patients une rémission clinique stable et prolongée, l’idéal étant qu’un jour l’expression d’une MICI se limite à celle de sa première poussée. Nos traitements n’étant actuellement que suspensifs, cette évolution n’a de sens que si les risques majeurs liés à l’utilisation des traitements immunosuppresseurs (infections sévères et cancers) sont connus et maîtrisés. Mots clés : Maladies inflammatoires chroniques de l’intestin (MICI) – Azathioprine – 6-mercaptopurine (6-MP)
Inflammatory bowel disease (IBD): what is the role of conventional treatments? Abstract: Conventional treatments for IBD include 5-aminosalicylate derivatives, corticosteroids and ‘conventional’ immunosuppressors (as opposed to biotechnological treatments), called ‘immunomodulators’ which essentially include thiopurines, methotrexate, cyclosporin and cyclosporin analogs. Here, we focus on the residual indications of corticosteroids and recent developments regarding the indications and good use of thiopurines and methotrexate, notably drawing from the recent European and American recommendations [1-5].
Correspondance :
[email protected]
Conventional treatments play a role in therapeutic strategies for IBD, and tend to be globally more ambitious and ‘aggressive,’ due to the diversification of the therapeutic arsenal and the observation of the relative failure of previously adopted strategies [6]. The current objective is to obtain stable, prolonged clinical remission in the majority of patients, with the ideal being that one day the expression of IBD will be limited to the first attack. Since current treatments are only suspensive, this evolution makes sense only if the major risks related to the use of immunosuppressive treatments (severe infections and cancer) are known and mastered. Keywords: Inflammatory bowel diseases (IBD) – Azathioprine – 6-mercaptopurin (6-MP)
Dérivés 5-amino-salicylés (5-ASA) Indépendamment de leur intérêt maintenant admis dans la chémoprévention des lésions néoplasiques coliques [7], les 5-ASA restent au cours de la rectocolite hémorragique (RCH), le traitement de référence des poussées d’intensité faible à moyenne [8] et le traitement de fond de première ligne, avec dans ce dernier contexte, un faible nombre de patients à traiter pour être efficace chez l’un d’entre eux (Number To Treat = NTT). Le NTT est de 3 dans des travaux anciens sur la salazopyrine [9] et de 6 dans la méta-analyse Cochrane, la plus récente portant sur l’ensemble des différents 5-ASA [10]. Lorsque les 5-ASA sont « dépassés » dans le cadre de RCH très évolutives, les experts européens recommandent d’ajouter les thiopurines aux 5-ASA plutôt que de remplacer les 5-ASA par les thiopurines, en visant une addition des effets anti-inflammatoires tout en maintenant l’action de chémoprévention du cancer colorectal [11]. Dans la maladie de Crohn, les 5-ASA restent proposés dans les poussées peu sévères, et le traitement préventif des rechutes de première ligne après résection iléocaecale [5], avec toutefois un NTT élevé (17) [12], et l’obligation, de passer aux thiopurines lorsque la détection endoscopique de la récidive péri-anastomotique à 6 mois - 1 an révèle, malgré les 5-ASA, un score de Rutgeerts supérieur à 1 [5].
360 Chez les patients qui ne développent pas de réactions immuno-allergiques, la toxicité potentielle au long cours des 5-ASA se limite essentiellement à la néphro-toxicité, décelable par le suivi de la fonction rénale : créatininémie semestrielle, Résumé des Caractéristiques du Produit (RCP). Les 5-ASA ne font pas courir de risque d’infection sévère ou de cancer.
Corticoïdes La corticothérapie systémique est un traitement très efficace des poussées de sévérité moyenne à forte de RCH et de maladie de Crohn. C’est une carte maîtresse de l’arsenal thérapeutique des poussées dont on doit savoir se servir, mais le moins souvent possible et en cherchant à ne pas dépasser à chaque fois environ 3 mois d’utilisation. En effet, les corticoïdes font courir les risques : – en traitement aigu, dès la première utilisation, d’effets cosmétiques et de prise de poids parfois psychologiquement désastreux, de diabète parfois définitif, d’accès maniaques ou délirants ; – en traitement chronique, même à faible dose, d’ostéoporose, d’ostéonécrose aseptique, de glaucome et de cataracte postérieure. De plus, indépendamment de la durée du traitement, le fait nouveau est que le registre TREAT a démontré qu’un patient sous corticoïdes a, par rapport aux patients ne recevant ni corticoïdes ni immunosuppresseurs, un risque multiplié par 2 d’infection sévère et de décès, alors que ce risque n’est pas accru pour les patients recevant un immunosuppresseur conventionnel ou un anti-TNF en monothérapie [13]. Il n’ y a pas eu néanmoins, dans cette étude, d’ajustement sur l’activité clinique et l’état nutritionnel et septique des patients. Autrement dit, l’effet délétère propre des corticoïdes est sans doute surestimé par le fait que les patients recevant des corticoïdes sont plus souvent en poussée (donc septiques et dénutris, donc vulnérables aux infections sévères et aux décès) que les patients stabilisés par un traitement immunosuppresseur au long cours. Néanmoins, le fait que les corticoïdes font par eux-mêmes courir un risque d’infection opportuniste, en particulier fungique (candidoses tissulaires) est confirmé dans l’étude cas-témoin de la Mayo Clinic [14]. Les corticoïdes ne doivent donc pas être banalisés. Ils doivent être considérés comme des immunosuppresseurs à part entière, avec, par exemple, comme pour les thiopurines, l’impossibilité d’utiliser les vaccins vivants atténués sous traitement (notamment BCG et vaccin de la fièvre jaune).
Indications Maladie de Crohn Au cours de la maladie de Crohn, la corticothérapie systémique est recommandée pour traiter les poussées de sévérité moyenne à forte, dans toutes les localisations, à l’exception des lésions ano-périnéales suppuratives complexes [1]. Dans la pratique, il faut distinguer la première utilisation de la corticothérapie des suivantes.
La première poussée traitée par corticothérapie systémique est plus d’une fois sur 2 la poussée inaugurale. Dans ce contexte, sauf contre-indication à l’usage des corticoïdes, il n’y a guère d’alternative thérapeutique. En effet, l’utilisation ponctuelle de l’infliximab est déconseillée (pour éviter une immunisation obérant l’efficacité d’une reprise ultérieure), l’alimentation entérale est d’une efficacité un peu moindre et plus lente [15]. Elle sera néanmoins préférée s’il y a une indication nutritionnelle et/ou de croissance. L’alimentation parentérale exclusive est coûteuse et longue, en plus des complications septiques potentielles. Les corticoïdes sont efficaces plus de neuf fois sur dix [16] mais sont incapables d’aboutir au sevrage en temps voulu et au maintien de la rémission dans l’année qui suit plus de huit fois sur dix [16]. Ceci a conduit progressivement à considérer que la première indication d’une corticothérapie systémique constituait par ellemême une indication à débuter en parallèle un traitement par thiopurines, pour tenir compte du délai d’action de ces dernières (notion de « pont » thérapeutique). Ainsi, les experts européens recommandent d’associer d’emblée les thiopurines aux corticoïdes dans le cadre d’une poussée inaugurale [1], attitude ayant démontré une grande efficacité dans le cadre d’un essai thérapeutique pionnier en milieu pédiatrique [17]. Il y a un consensus sur la recommandation de ne pas accepter une corticodépendance [1,2]. Les utilisations ultérieures de la corticothérapie relèvent donc maintenant le plus souvent de l’échec des traitements de fond immunosuppresseurs conventionnels et/ou des anti-TNF et se discutent dans le cadre de l’échec d’une ligne thérapeutique de fond, en transition avec l’attente d’efficacité de la suivante. Deux situations se présentent alors : – la première cure de corticoïdes a été efficace et bien supportée. Chez ces patients cortico-sensibles et « corticotolérants », de nouvelles cures de corticoïdes constituent des « jokers » thérapeutiques appréciables en attendant l’efficacité du méthotrexate ou d’un anti-TNF d’action lente [autorisation de mise sur le marché (AMM) de l’adalimubab] ; – la première cure de corticothérapie a été peu efficace et/ou mal tolérée. Lorsque se présente une nouvelle poussée « cortico-requérante », il convient de discuter à chaque fois que possible un traitement alternatif qui peut être une nutrition entérale à débit continu (possible en relais à domicile après une mise en œuvre hospitalière) [15], une nutrition orale exclusive par Modulen®, et dans les formes plus sévères, un passage à l’infliximab plutôt qu’une nutrition parentérale exclusive (technique longue, coûteuse, et associée à des risques septiques). Le budésonide est considéré comme le traitement de choix des poussées de maladie de Crohn de localisation iléo-caecale d’intensité minime à moyenne [1]. En utilisation prolongée, il est proposé comme un substitut de la corticothérapie systémique (6 mg de budésonide équivalent à peu près à 20-25 mg de prednisolone), globalement moins toxique que la corticothérapie systémique, en attendant de sortir d’une situation de cortico-dépendance à niveau élevé (RCP). En utilisation prolongée, le budésonide peut avoir les mêmes effets cosmétiques, métaboliques
361 et endocriniens (freination surrénalienne) que les corticoïdes systémiques, quoique le plus souvent atténués. La toxicité potentielle du budésonide sur l’os et sur l’œil n’est ni établie ni infirmée. RCH La corticothérapie orale systémique est indiquée dans les formes gauches et étendues de RCH en poussée malgré les 5-ASA [11]. Le traitement corticoïde intraveineux est le traitement de première intention des colites aiguës graves, sauf contre-indication ou intolérance préalable (dans ce cas, la ciclosporine ou l’infliximab sont utilisés) [11]. Les règles d’utilisation répétée ou non des corticoïdes sont les mêmes que pour la maladie de Crohn. Les alternatives aux corticoïdes au cours des poussées sévères de RCH se limitent à la ciclosporine et à l’infliximab.
Modalités et suivi du traitement La dose initiale de corticoïdes systémiques est de 1 mg/ kg/jour d’équivalent prednisolone en France pour les poussées de maladie de Crohn [16], 40 mg/jour per os pour les poussées non graves de RCH [18] et 0,75 mg/kg/jour de méthylprednisolone parentérale pour les colites aiguës graves [11]. Une supplémentation vitamino-calcique est conseillée dans tous les cas [1,3,19]. Une supplémentation en potassium n’est nécessaire qu’en cas de diarrhée abondante, de crampes, d’hypokaliémie pré-existante ou de troubles du rythme cardiaque. Une glycémie à une semaine de traitement est conseillée pour dépister précocement un diabète corticoinduit (guide médecin ALD24 HAS, www.has-sante.fr). Dans le cadre des traitements prolongés, doivent être envisagés une ostéodensitométrie (traitements de plus de 3 mois à plus de 7,5 mg/jour), un examen opthalomologique régulier (traitements de plus de 6 mois), une opothérapie substitutive par 20 à 30 mg d’hydrocortisone (10 à 20 mg au réveil, et 10 mg 6 à 8 heures plus tard) dès que la dose d’équivalent prednisolone est inférieure à 5 mg/jour, pour des corticothérapies de durée supérieure à 3 mois. Avant arrêt de la substitution, lorsque le patient ne reçoit plus que 20 mg d’hydrocortisone par jour, il est conseillé de réaliser un test au Synacthène® après 36 heures d’arrêt de l’hydrocortisone, afin de tester la restauration fonctionnelle de l’axe corticotrope, avant d’arrêter définitivement la substitution par hydrocortisone (RCP).
Thiopurines Indications Elles sont rappelées dans le Tableau 1. Au cours de la maladie de Crohn, elles sont nombreuses dès le diagnostic : poussée inaugurale cortico-requérante, atteinte gastroduodénale, étendue de l’intestin grêle, ano-périnéale complexe [1,5]. D’après les données épidémiologiques [20,21], 2/3 au moins des patients correspondent à ces indications dès le début de la maladie et devraient ainsi
Tableau 1. Indications des thiopurines pour le consensus européen (ECCO) Maladie de Crohn a Corticorésistance – Corticodépendance Poussée inaugurale traitée par corticoïdes Atteinte œso-gastro-duodénale Atteinte étendue (> 100 cm) de l’intestin grêle Lésions ano-périnéales complexes Rechute dans l’année suivant une poussée iléo-caecale sévère ou une poussée colique Rechute dans les 3 mois suivant toute poussée Rectocolite Hémorragique b Corticorésistance – Corticodépendance Poussées fréquentes chez des patients recevant un traitement optimal par les 5-ASA Poussées fréquentes chez des patients ne tolérant pas les 5-ASA Rémission obtenue par la ciclosporine Rémission obtenue par un traitement corticoïde intraveineux intensif (facultatif) a
Gut 2006;55(Suppl I):i16-i58 ; in press
b
Journal of Crohn’s and Colitis,
recevoir des thiopurines dans la première année après le diagnostic. Ce n’est pas encore la cas en France : en 2004-2005, dans la cohorte CESAME, la prévalence de 2/3 de patients recevant ou ayant reçu des thiopurines n’était atteint que 5 à 6 ans après le diagnostic. Ce relatif « retard dans les pratiques » explique sans doute en partie le fait que la probabilité pour un patient de subir une résection intestinale dans les premières années de la maladie est encore inchangée, du fait d’une mise en œuvre trop tardive des thiopurines [6]. Plus tard dans la maladie de Crohn, les thiopurines sont indiquées en cas de cortico-résistance et cortico-dépendance, et en cas de rechute précoce. Au cours de la RCH, les indications des thiopurines sont plus limitées, car une proportion importante des patients a une maladie contrôlée par les 5-ASA. En plus des traditionnelles situations de cortico-résistance et cortico-dépendance, les indications sont l’absence de contrôle de la maladie par les 5-ASA, et le relais de la ciclosporine ou des corticoïdes intraveineux après colite aiguë grave [11].
Bilan avant début du traitement Le patient doit être informé par écrit (fiches du GETAID, www.getaid.org) et par entretien spécifique des enjeux du traitement. Il faut faire un inventaire de départ d’éventuelles verrues et condylomes et de la fréquence d’éventuelles poussées d’herpès pour pouvoir apprécier secondairement l’éventuel facteur aggravant des thiopurines [22]. Le bilan biologique comporte une évaluation des fonctions rénales et un bilan hépatique de référence, et des sérologies du VIH, du virus B et du virus C [19]. Le dosage de l’activité de la TPMT et/ou son génotypage (hors nomenclature, donc non remboursé si prélevé hors de l’hôpital) est recommandé mais facultatif (cf. infra) [2].
362 Pour des raisons pratiques, ce bilan peut être réalisé dès le diagnostic (dans la mesure où 2/3 des patients recevront des thiopurines dans l’année), à défaut dès la première cure de corticoïdes systémiques (c’est le signal d’une très probable mise sous thiopurines prochaine), à défaut encore lorsque la mise sous thiopurines est décidée (source possible de retard au début du traitement (luimême d’action lente), surtout si l’on doit attendre le résultat du dosage de la TPMT).
Contre-indications et précautions d’utilisation Les thiopurines sont contre-indiquées en cas de cancer évolutif et de SIDA avéré. Un déficit complet en TPMT (par mutation homozygote codant pour un gène défaillant) est observé chez 3 patients sur 1 000 et expose à une toxicité médullaire précoce et grave. Certains experts parviennent néanmoins à traiter les patients avec des doses très faibles de thiopurines. En cas de déficit partiel en TMPT (mutation hétérozygote, 10 % de la population environ), les thiopurines doivent être débutées à demi-dose. En cas d’infection chronique virale B ou C, la réplication virale doit être monitorée tant que dure le traitement par thiopurines. Un effet aggravant des thiopurines (réplication virale accrue ou effet rebond), pouvant justifier de modifier l’attitude thérapeutique vis-à-vis de l’hépatite, est possible, mais sans atteindre habituellement l’effet délétère plus marqué des anti-TNF qui justifient eux, un co-traitement anti-viral systématique de l’hépatite B par exemple [23]. En cas d’antécédent personnel de cancer guéri, il faut évaluer individuellement le rapport bénéfice risque ; la prudence est d’attendre si possible 5 ans après guérison du cancer avant de débuter les thiopurines (les transplantés d’organe recevant des thiopurines sont à risque accru de récidive du cancer [24]). Les sujets blancs à peau claire, surtout s’ils ont eu des coups de soleil dans l’enfance, doivent être avertis de leur risque individuel de cancer cutané et de la nécessité absolue de se protéger du soleil sous traitement, car les thiopurines majorent la toxicité, donc l’effet carcinogène des rayons ultraviolets sur la peau [25]. L’expérience des transplantés montre que les sujets de plus de 70 ans mis sous immunosuppresseurs conventionnels révèlent souvent cliniquement un cancer qui existait mais était latent au moment de la mise sous thiopurines. Un scanner thoraco-abdomino-pelvien, une mammographie chez les femmes et un dosage des PSA chez les hommes permettent de réduire ce risque de prescription intempestive de thiopurines [26].
Modalités La dose usuelle efficace d’azathioprine (AZA) est de 2-2,5 mg/kg/j en une prise vespérale ; celle de la 6-mercaptopurine (6-MP) est de 1 à 1,5 mg/kg/j avec un facteur de conversion AZA ⇒ 6-MP de 0,55. Commencer le traitement à dose thérapeutique d’emblée ou à doses progressives est laissé au libre choix des prescripteurs [2]. Le début progressif est susceptible de différer le moment du début d’efficacité des thiopurines, lui-même déjà en
médiane de 3 mois [27]. Il est néanmoins à conseiller en cas de déficit partiel en TPMT, de leuco-neutropénie de base, de traitement par 5-ASA associé (cf. infra) ou d’absence de connaissance du statut TMPT (probabilité de 10 % de déficit partiel). La surveillance biologique des deux premiers mois comporte une NFS-plaquettes hebdomadaire pendant 8 semaines, puis mensuelle jusqu’à stabilisation de la dose, puis trimestrielle indéfiniment. Le bilan hépatique (ALATGGT-Phosphatases alcalines) est prescrit selon le même schéma ou une fois sur deux les 8 premières semaines [19]. La lipasémie ne doit être dosée qu’en cas de douleurs abdominales inhabituelles (HAS).
Effets attendus Environ 40 % des patients vont obtenir une rémission stable sans corticoïdes, avec une délai médian d’action de 3 mois [28]. Les bons répondeurs aux thiopurines ont sous traitement, un faible risque annuel de rechute de la maladie [29], ce qui est un avantage des thiopurines par rapport aux anti-TNF, qui posent plus ou moins le problème de la perte d’efficacité par immunisation [30]. Le taux de rechute après 4 ans d’efficacité des thiopurines peut être estimé à 5 % par an environ tant que le traitement est poursuivi [29].
Intolérance – Complications Les nausées et vomissements sont très fréquents en début de traitement et habituellement transitoires, sans modifier les doses de thiopurines. S’ils sont rebelles, il faut répartir et/ou réduire les doses, et parfois arrêter le traitement si ces dernières mesures sont inefficaces. Les réactions immuno-allergiques (5-10 % des cas) surviennent habituellement entre la deuxième semaine et la huitième semaine de traitement [31]. Elles peuvent comporter rashs, syndrome grippal, fièvre, douleurs articulaires, diarrhée paradoxale, pancréatite (lipasémie > 3N), ictère et diarrhée paradoxale. Tous ces évènements doivent faire interrompre le traitement. Seuls la pancréatite et l’ictère sont une indication définitive à la réintroduction des thiopurines. Dans tous les autres cas, on peut tenter après un intervalle libre, la réintroduction de l’autre thiopurine (le plus souvent en France la 6-MP après l’AZA plutôt que l’AZA après la 6-MP). Dans cette situation, l’absence de récidive des manifestations immuno-allergiques est notée une fois sur deux environ et permet la poursuite du traitement [32]. La toxicité hématologique sévère précoce est prévisible une fois sur deux environ (TMPT). Une toxicité plus tardive peut être expliquée par une interaction médicamenteuse (5-ASA et allopurinol, cf. infra). Elle peut être également secondaire, inexpliquée et progressive, obligeant à une réduction de doses définitive. Elle peut être brutale, inopinée, parfois imputée à une infection virale [33]. Ce risque constant justifie de ne jamais relâcher la surveillance hématologique (au moins trimestrielle) [33]. Des érythroblastopénies précoces sévères sont possibles, obligeant à un arrêt du traitement. En cours de traitement, une
363 lymphopénie est très fréquente, attribuable à l’effet des thiopurines, en plus d’une éventuelle exsudation digestive des lymphocytes dans des atteintes étendues non contrôlées. Nul ne sait exactement en dessous de quel seuil il convient de réduire les doses de thiopurines, voire d’arrêter le traitement, et si la lymphopénie est corrélée avec les risques d’infections sévères et de cancers. En pratique, dès que le taux de lymphocytes est inférieur à 500-600/mm3, il faut doser les lymphocytes CD4 et CD8. Il ne faut pas accepter un taux de lymphocytes CD4 inférieur à 300/mm3, et tant que le taux des lymphocytes n’est pas remonté après réduction de doses, il est suggéré par les infectiologues de prévenir la pneumocystose par Bactrim® ou aérosols de pentamidine, comme au cours de l’infection par le VIH. Une cytolyse et/ou une cholestase minimes précoces peuvent régresser spontanément ou après répartition/ diminution des doses. Sinon, il faut interrompre le traitement et tenter après normalisation du bilan hépatique, un « switch » selon les modalités décrites ci-dessus. La survenue secondaire d’une cytolyse et/ou d’une cholestase [34], même dissociée, doit faire discuter une cause intercurrente (infection virale B, C, CMV, EBV, cholangite), une hépatotoxicité liée aux dérivés méthylés des thiopurines (cf ci-dessous) ou la survenue d’une Hyperplasie Nodulaire Régénérative (HNR). Le risque d’HNR symptomatique semble dans l’expérience française de l’ordre de 1 % à 10 ans [35]. Ce risque, pour des raisons encore totalement mystérieuses, peut atteindre 10 % à 10 ans dans le groupe des hommes ayant un antécédent personnel de résection de plus 50 cm d’intestin (cohorte Saint-Antoine). Les signes d’alarme sont une chute progressive du taux des plaquettes, en particulier lorsqu’elles franchissent le seuil de 150 000/mm3, et l’apparition d’une quelconque anomalie du bilan hépatique (ALAT, GGT, phosphatases alcalines). L’écho-doppler portal peut montrer une réduction ou une inversion des flux, l’IRM hépatique des signes en faveur d’une HNR (mais avec un manque de sensibilité). L’endoscopie digestive haute (VO) et/ou le scanner peuvent montrer des signes d’hypertension portale. En cas de doute persistant, il faut réaliser une ponction biopsie hépatique (PBH). En cas d’HNR avérée ou présumée, il faut arrêter définitivement les thiopurines. La conduite à tenir en cas de survenue d’autres événements divers sous traitement par thiopurines, est développée dans le Tableau 2.
Optimisation pharmacologique L’AZA est la prodrogue de la 6-MP ensuite métabolisée (Fig. 1) selon les 3 voies principales en métabolites inactifs (voie de la xanthine oxydase), en métabolites méthylés hépatotoxiques (6-MMP et 6-MMPR) et en 6-Thio-GuanineNucléotides (6-TGN), métabolites actifs et myélotoxiques. Même si ces notions n’étaient pas encore consolidées au moment du consensus européen de 2005-2006, il est admis maintenant une corrélation significative entre le niveau des 6-TGN et l’efficacité clinique des thiopurines, le seuil étant de 220-230 pmol/8·108 globules rouges [36]. Le dosage des 6-TGN n’est pas remboursé par la Sécurité
Tableau 2. Attitudes pratiques devant diverses situations sous thiopurines Grossesse Pas de tératogénicité avérée (www.lecrat.org, ECCO consensus) Considération individuelle du rapport bénéfice/risque (bénéfice supérieur pendant les 5 premières années d’efficacité des thiopurines) Diminution de dose de 25-50 mg des thiopurines au 9e mois si leucopénie relative de la mère (pour éviter une leucopénie du nouveau-né) Diagnostic d’un cancer sous thiopurines Interruption des thiopurines Attendre si possible 5 ans sans cancer résiduel et sans récidive avant de réintroduire le traitement Vaccins vivants atténués (fièvre jaune++) Arrêt des thiopurines 3 mois, vaccin, réintroduction des thiopurines un mois après (British Society for Rheumatology 2002) Apparition ou aggravation de verrues et de condylomes sous thiopurines Si échec du traitement local, considérer une réduction de dose des thiopurines, voire son arrêt Poussées d’herpès fréquentes Traitement antiviral ou réduction de doses/arrêt des thiopurines
Fig. 1. Métabolisme simplifié des thiopurines 6-MMP : 6-méthylmercaptopurine, 6-TU : acide 6-thiourique, 6-MMPR : 6-méthylmercaptopurine ribonucléotides, 6-TIMP : 6-thioinosine 5’ monophosphate, 6-TGN : 6-thioguanine nucléotides)
Sociale (examen hors nomenclature) et ne peut être recommandé chez tous les patients. Mais en l’absence d’efficacité attendue des thiopurines à 3-6 mois de traitement, il apparaît utile de doser les 6-TGN. Si les 6-TGN sont très élevés (Fig. 2), il n’y a rien à attendre d’une escalade de doses et l’on peut passer à la ligne thérapeutique suivante. Si les 6-TGN sont bas et qu’il y a une marge hématologique (neutrophiles, lymphocytes), on peut augmenter les doses par paliers de 25-50 mg pour tenter d’obtenir l’efficacité thérapeutique. L’association des 5-ASA aux thiopurines majore le taux des 6-TGN (donc l’efficacité potentielle mais aussi la toxicité potentielle des thiopurines) et tend à diminuer le taux des dérivés méthylés [37]. Lorsque les 5-ASA sont ajoutés aux thiopurines, il faut connaître cette interaction et resserrer la surveillance hématologue, voire diminuer un peu la dose de thiopurines en l’absence de marge hématologique.
364 des MICI, avec la logique de ne pas accepter délibérément une progression des lésions vers des dégâts anatomiques irréversibles.
Pas d’efficacité attendue à 3-6 mois Dosage des 6-TGN
Méthotrexate
< 220 > 300 Changement de ligne
Pas de marge hématologique
Marge hématologique
doses/paliers de 25 mg
0 efficacité
TGN > 300
Efficace
6-TGN < 220 Et 0 efficacité Considérer : ASA-allopurinol
Fig. 2. Optimisation du traitement par thiopurines à l’aide du dosage des 6-TGN (une marge hématologique signifie un taux de polynucléaires neutrophiles circulants nettement supérieur à 2 000/mm3, et de lymphocytes supérieur à 500/mm3)
L’allopurinol, en inhibant la xanthine-oxydase, a pour effet de favoriser nettement la production des 6-TGN et de diminuer la production des dérivés méthylés. La prescription d’allopurinol est ainsi une association classiquement contre-indiquée au cours du traitement par thiopurines. L’équipe de Chicago a suggéré plus récemment d’utiliser cette interaction chez les malades n’obtenant pas une efficacité des thiopurines malgré une escalade de doses, avec production préférentielle de dérivés méthylés et fréquente hépatotoxicité [38]. Cette « manipulation » reste expérimentale pour le moment et réservée aux mains expertes. Elle est en effet dangereuse sur le plan hématologique, malgré la réduction de 3/4 de la dose initiale de thiopurines au moment de l’adjonction d’allopurinol.
Changement de ligne thérapeutique Il est évident qu’il faut passer à la ligne thérapeutique suivante (méthotrexate ou anti-TNF) en cas de complication des thiopurines, d’intolérance aux thiopurines ou de survenue d’une poussée sévère malgré des doses déjà ajustées. En cas de poussée non sévère sporadique sous traitement, il est possible d’attendre. En cas de répétition de poussées modérées, il faut chercher à optimiser pharmacologiquement le traitement (Fig. 2) ou, si cela est impossible ou inefficace, changer de ligne thérapeutique. En cas de réponse incomplète des thiopurines et de nécessité permanente de compléments thérapeutiques pour maintenir le contrôle de la maladie (budésonide, antibiotiques…, c’est souvent le cas des atteintes multisténosantes de l’intestin grêle), il est préférable de passer à la ligne thérapeutique suivante si celle-ci laisse espérer une amélioration globale de la situation anatomo-clinique. Lorsque la situation clinique et biologique est contrôlée mais que des lésions anatomiques persistent à l’évidence de façon permanente voire s’aggravent, il n’est pas établi aujourd’hui qu’il faille changer de ligne sur le seul argument d’obtenir une cicatrisation des lésions, mais il est possible que les choses changent rapidement, car de plus en plus d’arguments suggèrent que la cicatrisation des lésions est un pré-requis au bon pronostic à long terme
Indications et efficacité attendue Au cours de la maladie de Crohn, le consensus européen retient comme indications potentielles du méthotrexate, les situations de cortico-résistance et de corticodépendance et les rechutes précoces après corticothérapie, mais dans toutes ces situations, après échec ou complications des thiopurines [1]. Le niveau de preuve de l’efficacité du méthotrexate est élevé, avec en particulier une méta-analyse [39] et deux essais fondateurs publiés dans le New England Journal of Medicine montrant : pour le premier, la supériorité de 25 mg/semaine de méthotrexate par rapport au placebo pour induire la rémission et obtenir un sevrage en corticoïdes, essentiellement dans le sous-groupe partant avec un haut niveau de cortico-dépendance [40] ; pour le deuxième, l’effet significatif de 15 mg/semaine de méthotrexate par rapport au placebo pour maintenir la rémission induite par le méthotrexate [41]. L’efficacité du méthotrexate n’est pas encore validée au cours de la RCH (essai METEOR du GETAID en cours). L’utilisation du méthotrexate au cours de la RCH n’est donc pas recommandée à ce jour.
Contre-indications et précautions d’emploi Le méthotrexate est tératogène de façon suspensive et réversible. Son usage est incompatible avec la conception de l’homme et de la femme, obligeant à une contraception stricte à maintenir trois mois après l’arrêt du médicament chez la femme et cinq mois chez l’homme (RCP). Le méthotrexate ne doit pas être utilisé en cas d’hépatopathie chronique fibrosante sous-jacente avérée. Si l’hépatopathie est possible (obèses, alcooliques), elle doit être infirmée (échographie, fibrotest, fiboscan, voire PBH) avant mise sous méthotrexate. Les insuffisances rénales et respiratoires sévères sont également des contre-indications. Les vaccins vivants ne doivent pas être utilisés comme pour les thiopurines.
Modalités d’utilisation Le bilan biologique pré-requis est le même que pour les thiopurines. Avant le début du traitement, le patient doit être informé en détails (www.getaid.org) en insistant particulièrement sur le caractère impératif de la contraception. Le méthotrexate doit être administré initialement à la dose de 25 mg/semaine par voie sous-cutanée profonde ou intra-musculaire, la biodisponibilité du produit par voie orale étant faible et très aléatoire. Après trois mois, il est possible de passer à la dose de 15 mg/semaine, mais pour garder le patient en rémission prolongée stable, il est souvent nécessaire de revenir temporairement ou définitivement à la dose de 25 mg/semaine. Il faut une supplémentation en folates à distance de l’injection (5 mg d’acide
365 folique un seul jour suffisent). Il est impératif de connaître les interactions médicamenteuses qui peuvent toutes majorer la toxicité hématologique du méthotrexate. L’association du triméthoprime, de l’aspirine et de la phénylbutazone est contre-indiquée ; les AINS et les pénicillines sont déconseillés. La surveillance biologique peut être calquée sur celle des thiopurines, en y ajoutant périodiquement un contrôle de la créatininémie : en effet, d’une part le méthotrexate peut être néphrotoxique ; inversement, une insuffisance rénale d’une autre cause peut majorer la toxicité hématologique du méthotrexate.
Complications Une réaction d’intolérance clinique dans les heures qui suivent l’injection de méthotrexate est fréquente, dès le début du traitement ou retardée. Elle associe diversement une sensation d’ébriété, des céphalées, des nausées, une fatigue. Elles conduisent certains patients à faire leur injection le week-end et à utiliser une prémédication par dompéridone, paracétamol, comme pour l’interféron. Au pire, ces « contrecoups » cliniques de l’injection peuvent conduire à l’arrêt du traitement. Le méthotrexate peut être myélotoxique tout au long de son utilisation. Les pneumonies interstitielles immunoallergiques sont spécifiques du méthotrexate. Leur incidence atteint jusqu’à 3 pour 100 patients-années dans certaines séries. Le signe d’appel est la toux, plus ou moins associée à une fièvre modérée et une dyspnée. Toute toux prolongée sous méthotrexate qui ne s’inscrit pas dans un contexte étiologique intercurrent évident nécessite un avis spécialisé. La toxicité hépatique est habituellement annoncée par une cytolyse. Si le méthotrexate est poursuivi, elle peut aboutir à une fibrose voire une cirrhose. Néanmoins, en l’absence de toute anomalie biologique hépatique et de facteur de risque d’hépatopathie, il n’est plus proposé par les rhumatologues, qui utilisent le méthotrexate beaucoup plus que nous à des doses voisines, de faire une PBH pour une dose cumulée de 3 à 5 g (soit 3 à 5 ans d’utilisation). Le fibrotest apparaît comme un examen de substitution très intéressant pour vérifier l’intégrité hépatique dans ce contexte (JFPD 2008).
Inhibiteurs de la calcineurine La ciclosporine et le tacrolimus sont actuellement indiqués dans les colites aiguës graves de RCH en seconde ligne thérapeutique après échec des corticoïdes intraveineux, en alternative avec l’infliximab [11]. Ils ne peuvent être tout à fait assimilés à un traitement conventionnel et leurs contraintes et modalités d’utilisation ne seront pas développées ici.
Articulation des traitements conventionnels avec les anti-TNF Au cours de la maladie de Crohn, même si l’utilisation des immunosuppresseurs conventionnels devient globalement plus fréquente et plus précoce, le consensus européen ne
remet pas en question l’ordre des lignes thérapeutiques, à savoir les thiopurines en première intention, le méthotrexate ensuite, et les anti-TNF en cas d’échec d’un traitement bien conduit par immunosuppresseurs conventionnels. En pratique, il est à noter qu’en France, le méthotrexate est moins utilisé que les anti-TNF (CESAME). On peut estimer que, sur ces bases d’utilisation des traitements conventionnels, sur 100 patients nouvellement diagnostiqués, 70 recevront des thiopurines dans la première année de la maladie, et 50 iront progressivement aux antiTNF dans les cinq premières années de la maladie pour des raisons diverses (intolérance aux thiopurines, échec primaire, perte d’efficacité), sans toujours passer forcément par l’étape du méthotrexate. Au cours de RCH, la diffusion des thiopurines est relativement récente, du fait de la fréquence des problèmes possibles après coloproctectomie (mauvais résultat fonctionnel, occlusions, sténoses, pochites, hypofertilité), d’où la tendance globale à pousser plus loin le traitement médical avant de conclure à l’indication chirurgicale d’une RCH réfractaire au traitement médical. La même stratégie ascendante est de mise, mais : a) le méthotrexate n’a pour l’instant pas sa place ; b) l’infliximab peut être administré la première fois pour traiter une colite grave chez un patient naïf de thiopurines : dans ce cas, les thiopurines pourront prendre seuls dans certains cas, le relais de l’efficacité de l’infliximab.
Infections et cancers sous TRTS conventionnels et anti-TNF Infections Jusqu’à des temps récents, en dehors de la publication ponctuelle d’infections sévères, parfois opportunistes et fatales, il n’y avait pas d’étude épidémiologique prospective comparant la fréquence et le type d’infections en fonction des traitements anti-inflammatoires et/ou immunosuppresseurs reçus par les patients. Quatre études récentes donnent de premières indications : – les tuberculoses sont plus fréquentes chez les patients traités par anti-TNF [42] ; – chez les patients ambulatoires, l’incidence des verrues vulgaires et des poussées d’herpès est plus fréquente chez les patients recevant une monothérapie par thiopurines que chez ceux ne recevant ni corticoïdes ni immunosuppresseurs [22] ; – les données du registre TREAT suggèrent que les infections sévères et les décès sont environ deux fois plus fréquents chez les patients recevant des corticoïdes que chez ceux n’en recevant pas, après ajustement sur les autres traitements. Ce sur-risque n’est pas constaté pour les thiopurines et les anti-TNF [13] ; – une étude cas-témoin de la Mayo Clinic montre que l’odds-ratio d’infection croît avec le nombre d’immunosuppresseurs (au sens large : corticoïdes, thiopurines, anti-TNF) reçus par les patients (odds-ratio entre 3,1 et 4,4 pour un seul immunosuppresseur, 12,9 pour deux immunosuppresseurs) et que le type d’infections favorisées est différent selon les immunosuppresseurs reçus : candidoses
366 pour les corticoïdes, infections virales pour les thiopurines, agents intra-cellulaires (BK, aspergillose, etc.) pour les anti-TNF [14].
Cancers Les études épidémiologiques menées à partir des registres historiques de MICI (Scandinavie, comté d’Olmstedt) ou de la base de données des médecins généralistes britanniques, n’ont pu à ce jour, globalement préciser les risques de cancers liés à l’utilisation des immunosuppresseurs, du fait de la faible proportion de patients traités jusqu’à des temps récents. Concernant le risque de lymphomes, la méta-analyse de Kandiel et al., portant essentiellement sur des populations hospitalières, suggère un risque multiplié par 4 chez les patients recevant des thiopurines par rapport à ceux n’en recevant pas [43]. Le sur-risque éventuel de lymphomes n’est pas connu pour le méthotrexate (trop peu de patients traités au cours des MICI). Les premières données concernant les anti-TNF sont discordantes (excès de risque dans la méta-analyse des essais cliniques dans la polyarthrite rhumatoïde [44], non retrouvé dans la métaanalyse des essais dans les MICI (JFPD 2008) ni dans le registre TREAT. Concernant les autres types de cancers, il n’y aucune donnée sûre dans ce domaine à ce jour pour les MICI, alors que l’on sait que les transplantés recevant des IS conventionnels ont un excès net de cancers. Pour les thiopurines (lymphomes et autres cancers), l’étude coopérative française CESAME est susceptible de donner dans les mois qui viennent, des informations importantes, sous réserve d’obtenir des informations de fin d’étude pour la majorité des patients.
Conclusion Les pratiques thérapeutiques des MICI ont beaucoup changé ces dix dernières années et sont susceptibles de changer sensiblement et rapidement dans les mois ou années qui viennent. Globalement, les traitements immunosuppresseurs conventionnels et les anti-TNF sont donnés de plus en plus souvent et tôt dans la maladie, sans remettre en question encore à ce jour la séquence traitements conventionnels-anti-TNF. Nous allons vers une utilisation de plus en plus restreinte des corticoïdes et vers le fait de privilégier les monothérapies pour ne pas additionner voire multiplier (pour le risque infectieux) les risques iatrogéniques. L’ordre des lignes thérapeutiques et leur durée dépendront éminemment des données comparées de sécurité d’utilisation, en particulier dans le domaine des infections sévères et des cancers. La gestion de ces risques conditionne déjà beaucoup et conditionnera de plus en plus nos pratiques de soins individuelles au cours des MICI.
Références 1. Travis SP, Stange EF, Lemann M, Oresland T, Chowers Y, Forbes A, D’Haens G, Kitis G, Cortot A, Prantera C, Marteau P, Colombel JF, Gionchetti P, Bouhnik Y, Tiret E, Kroesen J, Starlinger M, Mortensen NJ. European evidence based consensus on the diagnosis and management of Crohn’s disease: current management. Gut 2006;55 Suppl 1:i16-35. 2. Lichtenstein GR, Abreu MT, Cohen R, Tremaine W. American Gastroenterological Association Institute medical position statement on corticosteroids, immunomodulators, and infliximab in inflammatory bowel disease. Gastroenterology 2006;130:935-9. 3. Lichtenstein GR, Abreu MT, Cohen R, Tremaine W. American Gastroenterological Association Institute technical review on corticosteroids, immunomodulators, and infliximab in inflammatory bowel disease. Gastroenterology 2006;130:940-87. 4. Stange EF, Travis SP, Vermeire S, Beglinger C, Kupcinkas L, Geboes K, Barakauskiene A, Villanacci V, Von Herbay A, Warren BF, Gasche C, Tilg H, Schreiber SW, Scholmerich J, Reinisch W. European evidence based consensus on the diagnosis and management of Crohn’s disease: definitions and diagnosis. Gut 2006;55 Suppl 1:i1-15. 5. Caprilli R, Gassull MA, Escher JC, Moser G, Munkholm P, Forbes A, Hommes DW, Lochs H, Angelucci E, Cocco A, Vucelic B, Hildebrand H, Kolacek S, Riis L, Lukas M, de Franchis R, Hamilton M, Jantschek G, Michetti P, O’Morain C, Anwar MM, Freitas JL, Mouzas IA, Baert F, Mitchell R, Hawkey CJ. European evidence based consensus on the diagnosis and management of Crohn’s disease: special situations. Gut 2006; 55 Suppl 1:i36-58. 6. Cosnes J, Nion-Larmurier I, Beaugerie L, Afchain P, Tiret E, Gendre JP. Impact of the increasing use of immunosuppressants in Crohn’s disease on the need for intestinal surgery. Gut 2005;54:237-41. 7. Velayos FS, Terdiman JP, Walsh JM. Effect of 5-aminosalicylate use on colorectal cancer and dysplasia risk: a systematic review and metaanalysis of observational studies. Am J Gastroenterol 2005;100:1345-53. 8. Sutherland L, Macdonald JK. Oral 5-aminosalicylic acid for induction of remission in ulcerative colitis. Cochrane Database Syst Rev 2006: CD000543. 9. Dissanayake AS, Truelove SC. A controlled therapeutic trial of long-term maintenance treatment of ulcerative colitis with sulphazalazine (Salazopyrin). Gut 1973;14:923-6. 10. Sutherland L, Macdonald JK. Oral 5-aminosalicylic acid for maintenance of remission in ulcerative colitis. Cochrane Database Syst Rev 2006: CD000544. 11. Travis SP, Stange EF, Lemann M, Oresland T, Chowers Y, Colombel JF, Gionchetti P, D’Haens G, Ghosh S, Marteau P, Mortensen NJ, Pennincks F, Gassull M. European evidence based consensus: the current management of Ulcerative Colitis. Journal of Crohn’s and Colitis 2008;In press. 12. Camma C, Giunta M, Rosselli M, Cottone M. Mesalamine in the maintenance treatment of Crohn’s disease: a metaanalysis adjusted for confounding variables. Gastroenterology 1997;113:1465-73. 13. Lichtenstein GR, Feagan BG, Cohen RD, Salzberg BA, Diamond RH, Chen DM, Pritchard ML, Sandborn WJ. Serious infections and mortality in association with therapies for Crohn’s disease: TREAT registry. Clin Gastroenterol Hepatol 2006;4:621-30. 14. Toruner M, Loftus EV, Jr., Colombel JF, Orenstein R, Harmsen HJ, Zinsmeister AR, Sandborn WJ, Egan LJ. Risk factors for opportunistic infections in inflammatory bowel diseases: a case-control study. Gastroenterology 2006;130:A-71. 15. Zachos M, Tondeur M, Griffiths AM. Enteral nutritional therapy for induction of remission in Crohn’s disease. Cochrane Database Syst Rev 2007: CD000542. 16. Modigliani R, Mary JY, Simon JF, Cortot A, Soule JC, Gendre JP, Rene E. Clinical, biological, and endoscopic picture of attacks of Crohn’s disease. Evolution on prednisolone.
367
17.
18. 19.
20.
21. 22. 23. 24.
25.
26. 27.
28. 29.
30.
31.
Groupe d’Etude Thérapeutique des Affections Inflammatoires Digestives. Gastroenterology 1990;98:811-8. Markowitz J, Grancher K, Kohn N, Lesser M, Daum F. A multicenter trial of 6-mercaptopurine and prednisone in children with newly diagnosed Crohn’s disease. Gastroenterology 2000;119:895-902. Beaugerie L, Blain A, Brazier F, Gornet J, Parc Y. Traitement de la rectocolite ulcéro-hémorragique dans sa forme étendue (colite grave exclue). Gastroenterol Clin Biol 2004;28:974-83. Lemann M, Beaugerie L, Bouhnik Y, Flourie B, Reimund JM, Seksik P, Marteau P. [Practical forms for the use of the main drugs in the treatment of ulcerative colitis]. Gastroenterol Clin Biol 2004;28:1039-48. Faubion WA, Jr., Loftus EV, Jr., Harmsen WS, Zinsmeister AR, Sandborn WJ. The natural history of corticosteroid therapy for inflammatory bowel disease: a population-based study. Gastroenterology 2001;121:255-60. Beaugerie L, Seksik P, Nion-Larmurier I, Gendre JP, Cosnes J. Predictors of Crohn’s disease. Gastroenterology 2006;130:650-6. Seksik P, Cosnes J, Nion-Larmurier I, Gendre JP, Beaugerie L. Incidence of benign infections in IBD patients treated with azathioprine. Gastroenterology 2006;130:A-72. Cosnes J. Traitements immunosuppresseurs chez les patients à risque. Post-U FMC-HGE 2007:17-30. Vivarelli M, Cucchetti A, Piscaglia F, La Barba G, Bolondi L, Cavallari A, Pinna AD. Analysis of risk factors for tumor recurrence after liver transplantation for hepatocellular carcinoma: key role of immunosuppression. Liver Transpl 2005;11:497-503. O’Donovan P, Perrett CM, Zhang X, Montaner B, Xu YZ, Harwood CA, McGregor JM, Walker SL, Hanaoka F, Karran P. Azathioprine and UVA light generate mutagenic oxidative DNA damage. Science 2005;309:1871-4. Conraads VM, Denollet J, Vorlat A, Moulijn AC, Vrints CJ. Screening for solid organ malignancies prior to heart transplantation. Transplantation 2001;71:1481-3. Present DH, Korelitz BI, Wisch N, Glass JL, Sachar DB, Pasternack BS. Treatment of Crohn’s disease with 6-mercaptopurine. A long-term, randomized, double-blind study. N Engl J Med 1980;302:981-7. Candy S, Wright J, Gerber M, Adams G, Gerig M, Goodman R. A controlled double blind study of azathioprine in the management of Crohn’s disease. Gut 1995;37:674-8. Lemann M, Mary JY, Colombel JF, Duclos B, Soule JC, Lerebours E, Modigliani R, Bouhnik Y. A randomized, doubleblind, controlled withdrawal trial in Crohn’s disease patients in long-term remission on azathioprine. Gastroenterology 2005;128:1812-8. Hanauer SB, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, Rachmilewitz D, Wolf DC, Olson A, Bao W, Rutgeerts P. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. Lancet 2002;359: 1541-9. Beaugerie L, Gendre JP. [Azathioprine, 6-mercaptopurine, and inflammatory diseases of the intestine]. Gastroenterol Clin Biol 1990;14:230-40.
32. Lees CW, Mann A, Arnott SJ. Tolerability and safety of mercaptopurine in azathioprine intolerant patients with IBD. Aliment Pharmacol Ther; in press. 33. Connell WR, Kamm MA, Ritchie JK, Lennard-Jones JE. Bone marrow toxicity caused by azathioprine in inflammatory bowel disease: 27 years of experience. Gut 1993;34:1081-5. 34. Shaye OA, Yadegari M, Abreu MT, Poordad F, Simon K, Martin P, Papadakis KA, Ippoliti A, Vasiliauskas E, Tran TT. Hepatotoxicity of 6-mercaptopurine (6-MP) and Azathioprine (AZA) in adult IBD patients. Am J Gastroenterol 2007; 102:2488-94. 35. Vernier-Massouille G, Cosnes J, Lemann M, Marteau P, Reinisch W, Laharie D, Cadiot G, Bouhnik Y, De Vos M, Boureille A, Duclos B, Seksik P, Mary JY, Colombel JF. Nodular regenerative hyperplasia in patients with inflammatory bowel disease treated with azathioprine. Gut 2007;56:1404-9. 36. Osterman MT, Kundu R, Lichtenstein GR, Lewis JD. Association of 6-thioguanine nucleotide levels and inflammatory bowel disease activity: a meta-analysis. Gastroenterology 2006;130:1047-53. 37. de Boer NK, Wong DR, Jharap B, de Graaf P, Hooymans PM, Mulder CJ, Rijmen F, Engels LG, van Bodegraven AA. Dosedependent influence of 5-aminosalicylates on thiopurine metabolism. Am J Gastroenterol 2007;102:2747-53. 38. Sparrow MP, Hande SA, Friedman S, Cao D, Hanauer SB. Effect of allopurinol on clinical outcomes in inflammatory bowel disease nonresponders to azathioprine or 6-mercaptopurine. Clin Gastroenterol Hepatol 2007;5:209-14. 39. Alfadhli AA, McDonald JW, Feagan BG. Methotrexate for induction of remission in refractory Crohn’s disease. Cochrane Database Syst Rev 2005: CD003459. 40. Feagan BG, Rochon J, Fedorak RN, Irvine EJ, Wild G, Sutherland L, Steinhart AH, Greenberg GR, Gillies R, Hopkins M, et al. Methotrexate for the treatment of Crohn’s disease. The North American Crohn’s Study Group Investigators. N Engl J Med 1995;332:292-7. 41. Feagan BG, Fedorak RN, Irvine EJ, Wild G, Sutherland L, Steinhart AH, Greenberg GR, Koval J, Wong CJ, Hopkins M, Hanauer SB, McDonald JW. A comparison of methotrexate with placebo for the maintenance of remission in Crohn’s disease. North American Crohn’s Study Group Investigators. N Engl J Med 2000;342:1627-32. 42. Keane J, Gershon S, Wise RP, Mirabile-Levens E, Kasznica J, Schwieterman WD, Siegel JN, Braun MM. Tuberculosis associated with infliximab, a tumor necrosis factor alphaneutralizing agent. N Engl J Med 2001;345:1098-104. 43. Kandiel A, Fraser AG, Korelitz BI, Brensinger C, Lewis JD. Increased risk of lymphoma among inflammatory bowel disease patients treated with azathioprine and 6-mercaptopurine. Gut 2005;54:1121-5. 44. Bongartz T, Sutton AJ, Sweeting MJ, Buchan I, Matteson EL, Montori V. Anti-TNF antibody therapy in rheumatoid arthritis and the risk of serious infections and malignancies: systematic review and meta-analysis of rare harmful effects in randomized controlled trials. Jama 2006;295:2275-85.
368 Characteristics 5-aminosilicylate derivatives (5-ASA) Independent of their now accepted use in the chemoprevention of colonic neoplastic lesions [7], 5-ASAs remain the treatment of reference for ulcerative colitis for medium to low intensity attacks [8], and the basic first-line treatment in the latter case, with a low number of patients to treat for efficacy in one patient (number to treat (NTT)). The NTT is three in previous work on salazopyrin [9] and 6 in the Cochrane meta-analysis, the most recent of which comprised the ensemble of the different 5-ASAs [10]. When the 5-ASAs are ‘surpassed’ in the framework of highly evolutive HCR, the European experts recommend adding thiopurines to 5-ASAs rather than replacing the 5-ASAs by thiopurines, aiming for the addition of anti-inflammatory effects while maintaining the effect of chemoprevention of colorectal cancer [11]. For Crohn’s disease, 5-ASAs are still proposed for severe attacks, and as the first-line treatment for prevention of relapse after ileocaecal resection [5], still with a high NTT (17) [12], and the obligation of turning to thiopurines if, despite the 5-ASAs, there is endoscopic detection of a perianastomotic recurrence at six months/one year with a Rutgeerts score > 1 [5]. In patients who do not develop immunoallergic reactions, the potential toxicity during 5-ASA treatment is essentially limited to nephrotoxicity, detectable by monitoring of renal function [bi-yearly creatininemia, Summary of Product Characteristics (SPC)]. The 5-ASAs do not carry a risk of severe infection or cancer.
Corticoids Systemic corticotherapy is a very efficacious treatment for average to high intensity attacks of HCR and Crohn’s disease. It is a master card in the therapeutic arsenal against attacks that one must know how to use; it should be used as little as possible, being careful not to exceed three months of successive use at each interval. In fact, corticosteroids carry the following risks: – for acute treatment, as of the first use, cosmetic effects and weight gain which can sometimes be psychologically disastrous, diabetes which is sometimes definitive, and bouts of mania or delirium; – for chronic treatment, even at low doses, osteoporosis, aseptic osteonecrosis, glaucoma and posterior cataract. In addition, independently of the duration of treatment, the new fact is that the TREAT register has demonstrated that a patient under corticosteroids has, in comparison to patients receiving neither corticosteroids nor immunosuppressors, a two-times greater risk of severe infection and death, while this risk is not accrued for patients receiving a conventional immunosuppressor or anti-TNF in monotherapy [13]. Nevertheless, this study did not comprise an adjustment for clinical activity and the nutritional and septic state of the patients. In other words, the deleterious effect of corticosteroids themselves is without a doubt overestimated due to the fact that patients receiving corticosteroids are more often experiencing an attack (and are thus septic and undernourished, and thus
vulnerable to severe infections and death) than patients stabilised by treatment with a long-term immunosuppressant. Nevertheless, the fact that corticosteroids themselves carry a risk of opportunistic infection, particularly fungal (tissular candidiasis) is confirmed in the Mayo Clinic casecontrol study [14]. Corticosteroids should thus not be trivialized. They must be considered as bonafide immunosuppressors, with, for example, as for the thiopurines, the impossibility of using live attenuated vaccines under treatment (notably BCG and yellow fever vaccine).
Indications Crohn’s disease For Crohn’s disease, systemic corticotherapy is recommended for treating medium and high intensity attacks, in all localisations, with the exception of complex suppurative anoperineal lesions [1]. In practice, the first use of corticotherapy should be distinguished from following uses. In more than one out of two cases, the first attack treated by systemic corticotherapy is the inaugural attack. In this context, with the exception of contra-indication to the use of corticosteroids, there is really no other therapeutic alternative. In fact, the punctual use of infliximab is discouraged (to avoid immunisation interfering with the efficacy of later use), and enteral feeding is slightly less effective and slower [15]. It is nevertheless preferred if there is a nutritional and/or growth indication. Exclusive parenteral alimentation is costly and time-consuming, and carries potential septic complications. Corticosteroids are efficacious more than nine times out of 10 [16] but do not result in weaning within the desired time frame and maintenance of remission in the following year more than eight times out of 10 [16]. This has progressively led to the consideration that the first indication of systemic corticotherapy constitutes in itself an indication for beginning parallel treatment with thiopurines, to take into consideration the delay of action of the latter (the notion of the therapeutic ‘bridge’). Thus, the European experts recommend association, from the beginning, of thiopurines with corticosteroids in the framework of an inaugural attack [1], an attitude which has been shown to be highly efficacious in the framework of a pioneer therapeutic trial in the paediatric milieu [17]. There is a consensus on the recommendation to not accept corticodependance [1,2]. The further uses of corticotherapy thus now more often reveal a failure of basic conventional immunosuppressor and/or anti-TNF treatments and are discussed within the framework of the failure of a basic line of therapy, in transition with the hope that the next therapy will be efficacious. The two following situations are thus presented: – the first treatment with corticosteroids was efficacious and well supported. In corticosensitive and corticosteroid tolerant patients, new corticosteroid treatment constitutes a considerable therapeutic ‘joker’ while waiting for methotrexate or a slow acting anti-TNF (Marketing Authorisation Application of adalimubab) to be effective; – the first corticotherapy treatment was not very efficacious and/or poorly tolerated. When a new ‘corticosteroid-
369 requiring’ attack presents, each time an alternative treatment should be discussed, such as continuous-flow enteral feeding (which is possible to continue at home after being put in place in hospital) [15], exclusive oral nutrition with Modulen®, and for more severe forms, transition to infliximab rather than exclusive parenteral feeding (a long, costly technique associated with high septic risks). Budesonide is considered as the treatment of choice for low to medium intensity attacks of ileocaecal localisations of Crohn’s disease [1]. Its prolonged used is proposed as a substitute to systemic corticotherapy (6 mg of budesonide is equivalent to about 20-25 mg prednisolone), with less overall toxicity than systemic corticotherapy, while trying to get out of a situation of a high level of corticosteroid dependence (SPC). With prolonged use, budesonide can have the same cosmetic, metabolic and endocrine effects (slowing of surrenal function) as systemic corticoids, albeit these effects are often attenuated. The potential toxicity of budesonide on bone and the eye has neither been established nor invalidated.
Table 1. Indications of thiopurines for the European consensus (ECCO)
Ulcerative colitis (UC)
a
Systemic oral corticotherapy is indicated for left and extensive forms of UC attacks despite the 5-ASAs [11]. Intravenous corticoid treatment is the treatment of first intention for serious acute colitis, except for cases where there is a contra-indication or previous intolerance (in this case cyclosporin or infliximab are used [11]. The rules for repeated or single use of corticoids are the same as for Crohn’s disease. The alternatives to corticoids during severe attacks of UC are limited to cyclosporin and infliximab.
Treatment and monitoring The initial dose of systemic corticosteroids is 1 mg/kg/day equivalent of prednisolone in France for attacks of Crohn’s disease [16], 40 mg/day per os for non-serious UC attacks and 0.75 mg/kg/day parenteral methylprednisolone for serious acute colitis [11]. Vitamin-calcium supplementation is recommended in all cases [1,3,19]. Potassium supplementation is only necessary in the case of abundant diarrhoea, cramps, pre-existing hypokalemia or cardiac rhythm difficulties. Glycemia at one week of treatment is recommended for the early detection of corticosteroid-induced diabetes (doctor guide ALD24 HAS, www.has-sante.fr). Osteodensitometry should be considered within the framework of prolonged treatment (treatment over 6 months), and substitutive opotherapy with 20 to 30 mg hydrocortisone (10 to 20 mg upon waking, and 10 mg 6 to 8 hours later) as soon as the equivalent prednisolone dose is lower than 5 mg/day, for corticotherapies over three months. Before cessation of the substitution, when the patient is receiving less than 20 mg of hydrocortisone per day, a Synacthene® test is recommended 36 h after cessation of hydrocortisone, in order to test the functional restoration of the corticotropic axis, before definitively stopping substitution with hydrocortisone (SPC).
Crohn’s disease a Corticoresistance – Corticodependence Onset attacks treated by corticoids Oeso-gastro-duodenal invasion Widespread invasion (> 100 cm) of small intestine Complex anoperineal lesions Recurrence in the year following a severe ileo-caecal or colic attack Recurrence in the 3 months following any attack Ulcerative colitis b Corticoresistance – Corticodependence Frequent attacks in patients receiving an optimal treatment with 5-ASA Frequent attacks in patients intolerant to 5-ASA Remission obtained by ciclosporin Remission obtained by an intensive intravenous corticoid treatment (optional) Gut 2006;55(Suppl I):i16-i58; b Journal of Crohn’s and Colitis, in press
Thiopurines Indications The indications of thiopurines are given in Table 1. There are many indications in Crohn’s disease, as soon as it is diagnosed: corticosteroid requiring inaugural attack, gastro-duodenal attack, and extension to the small bowel, anoperineal complex [1,5]. According to epidemiological data [20,21], at least 2/3 of patients correspond to these indications from the onset of the disease and should thus receive thiopurines within the first year following diagnosis. This is not yet the case in France, in 2004-2005, in the CESAME cohort a prevalence of 2/3 of patients receiving or having received thiopurines was only reached 5/6 years after diagnosis. This relative ‘delay in practice’ in part explains without a doubt the fact that the probability of a patient undergoing intestinal resection in the first few years of the disease remains unchanged, due to starting thiopurines too late [6]. Later in Crohn’s disease, thiopurines are indicated in the case of corticosteroid resistance and dependence, and in the event of precocious relapse. For UC, the indications for thiopurines are more limited, because a large proportion of patients have a disease controlled by 5-ASAs. In addition to traditional situations of corticoid resistance and dependence, the indications are absence of control of the disease by 5-ASAs and the switch to intravenous cyclosporin or corticosteroids following serious acute colitis [11]. Pre-treatment evaluation The patient must be informed of the treatment issues in writing (GETAID files, www.getaid.org) and with a specific interview. A starting inventory of warts and condylomas and of the frequency of eventual herpes attacks must be
370 done in order to appreciate any secondary aggravation due to thiopurines [22]. The biological evaluation comprises a reference evaluation of renal function and a liver panel, and HIV, and hepatitis B and C virus serology [19]. The dosage of TPMT activity and/or genotyping (uncodified in the healthcare system, and therefore covered by national healthcare if the sample is taken outside the hospital) is recommended but optional (cf. infra) [2]. For practical reasons, this evaluation may be performed following diagnosis (in the measure where 2/3 of patients will receive thiopurines within the year), or by default be done upon the first treatment with systemic corticosteroids (this is the sign that thiopurines will likely be used next), or by default again when treatment with thiopurines is decided upon (it is then possible that this will delay treatment (which in itself is slow acting), especially if one must await the results of the TPMT testing).
6-MP conversion factor of 0.55. The prescriber is free to decide whether to begin treatment at the therapeutic dose or with progressive doses [2]. Starting progressively may change the time of beginning of efficacy of thiopurines, which already takes 3 months on average [27]. It is nevertheless recommended in the case of partial TPMT deficiency, of leukoneutropenia, 5-ASA associated treatment (cf. infra) or in the absence of knowledge regarding TPMT status (10% probability of partial deficiency). Biological monitoring of the first two months comprises weekly CBC-platelet testing for eight weeks, followed by monthly testing until stabilisation of the dose, then testing each trimester, indefinitely. The liver panel (ALAT-GGTalkaline phosphatases) is prescribed when the same schedule or once every 2 weeks for the first 8 weeks [19]. Lipasemia should not be tested for in the case of unusual abdominal pain (SAH).
Contra-indications and precautions for use
Expected effects
Thiopurines are contra-indicated in the case of evolutive cancer and known AIDS. A complete deficiency in TPMT (by homozygous mutation coding for a faulty gene) is observed in three out of 1,000 patients and exposes them to precocious and serious medullar toxicity. Certain experts nevertheless manage to treat patients with very low doses of thiopurines. In the case of partial TMPT deficiency (heterozygous mutation, about 10% of the population), one should begin with a half-dose of thiopurines. In the case of chronic hepatitis B or C viral infection, viral replication should be monitored for the duration of thiopurine treatment. An aggravating effect of thiopurines (increased viral replication or rebound response) which may justify a modification of the therapeutic attitude towards the hepatitis is possible, without habitually waiting for the more marked deleterious effect of anti-TNFs which justify systemic anti-viral cotreatment of hepatitis B, for example [23]. In the case of personal antecedent of healed cancer, the risk - benefit ratio must be individually evaluated; it is prudent to wait, if possible, five years after healing of cancer before starting thiopurines (organ transplant patients receiving thiopurines are at a higher risk of cancer relapse [24]). Patients with very fair skin receiving thiopurines, especially if they had sunburn as a child, should be made aware of their individual risk of skin cancer and the absolute necessity of protecting themselves from the sun while under treatment, because thiopurines increase the toxicity, and thus the carcinogenic effect, of ultraviolet rays on the skin [25]. The experience of transplant patients shows that patients over 70 years of age on conventional immunosuppressors often clinically reveal a cancer that existed but was latent at the beginning of thiopurine treatment. A thorax-abdominal-pelvic scan, a mammogram for women and PSA testing in men help reduce the risk of untimely prescription of thiopurines [26].
Therapy The usual efficacious dose of azathioprine (AZA) is from 2 to 2.5 mg/kg/day taken in the evening; that of 6-mercaptopurine (6-MP) is from 1 to 1.5 mg/kg/day with an AZA ⇒
Around 40% of patients achieve stable remission without corticosteroids, with a mean delay of action of three months [28]. Good thiopurine responders have, under treatment, a low annual risk of disease relapse [29], which is an advantage of thiopurines in comparison with anti-TNFs, which more or less pose the problem of lack of efficacy by immunisation [30]. The relapse level after 4 years of efficacy for thiopurines can be estimated at about 5% per year as long as treatment is continued [29].
Intolerance - Complications Nausea and vomiting are very frequent in the beginning of treatment and are usually transitory, without modification of the dose of thiopurines. If they persist, the doses should be divided and/or reduced; if these measures are not effective, sometimes treatment must be stopped. Immunoallergic reactions (5-10% of cases) usually appear between the second and eighth week of treatment [31]. They may include rash, flu syndrome, fever, joint pain, paradoxical diarrhoea, pancreatitis (lipasemia > 3N) and icterus. These events are all cause for interruption of treatment. Only pancreatitis and icterus are definitive indications for the reintroduction of thiopurines. In all the other cases, the reintroduction of another thiopurine can be tried after a free interval (in France, most often 6-MP is used after AZA rather than AZA after 6-MP). In this situation, the absence of recurrence of immunoallergic manifestations is noted in about one out of two cases and treatment can be continued [32]. Early severe haematological toxicity occurs in one out of two cases (TMPT). Later toxicity can be explained by drug interactions (5-ASA and allopurinol, cf. infra). It can also be secondary, unexplained and progressive, and require a definitive dose reduction. It can be brutal, sudden, and sometimes attributed to a viral infection [33]. This constant risk justifies keeping up with haematological monitoring (at least every three months). Severe early erythroblastopenia is possible, requiring cessation of treatment. Lymphopenia during treatment is very frequent, due to the
371 effect of the thiopurines, as well as eventual digestive exsudation of lymphocytes in extensive uncontrolled cases. Nobody knows exactly what the threshold is for reducing the dose of thiopurines, or even for stopping treatment, or if lymphopenia correlates with the risk of severe infection and cancer. In practice, as soon as the level of lymphocytes drops below 500-600/mm3, the patient must have CD4 and CD8 lymphocyte levels tested. A level of lymphocytes below 200/mm3 should not be accepted and, as long as the lymphocyte level has not re-established following dose reduction, infectologists suggest preventing pneumocystosis with Bactrim® or pentamidine aerosols, as done with HIV infection. Minimal and or early cytolysis and/or cholestasis may regress spontaneously or after division/reduction of doses. Otherwise, treatment must be interrupted and, after normalisation of the liver panel, a switch to another treatment, as described above, should be tried. Secondary appearance of cytolysis and/or cholestasis [34], even dissociated, is cause for consideration of a concurrent cause (hepatitis B or C virus infection, CMV, EBV, or cholangitis), hepatotoxicity related to thiopurine methyl derivatives (see below) or the appearance of nodular regenerative hyperplasia (NRH). The risk of symptomatic NRH seems, drawing from French experience, to be on the order of 1% at 10 years [35]. This risk, for reasons entirely unknown, can reach 10% at 10 years in men with a personal antecedent of more than 50 cm resection of the bowel (Saint-Antoine cohort). The alarm signals are a progressive drop in platelet level, particularly when it passes the threshold of 150,000/mm3 and the appearance of any anomaly in the liver panel (ALAT, GGT, alkaline phosphatases). Portal echo-doppler may show a reduction or inversion of flow, and hepatic MRI may show signs in favour of NRH (but with a lack of sensitivity). High digestive endoscopy (oral route) and/or a scan may show signs of portal hypertension. In the case of persistent doubt, an HPB must be performed. In the case of proven or presumed NRH, thiopurines must be definitively stopped. The attitude to take if other events occur while under thiopurine treatment is developed in Table 2.
Pharmacological optimisation AZA is the prodrug of 6-MP which is then metabolised (Fig. 1) according to the three main routes into inactive metabolites (xanthine oxidase route), into methylated hepatotoxic metabolites (6-MMP and 6-MMPR) and into active and myelotoxic 6-thio-guanine-nucleotide (6-TGN) metabolites. Even if these notions were not yet consolidated as of the 2005-2006 European consensus, it is now recognised that there is a significant correlation between the level of 6-TGN and the clinical efficacy of thiopurines, the threshold being 220/230 pmol/8×108 red blood cells [36]. 6-TGN testing is not covered by national healthcare (uncodified exam) and cannot be recommended in all patients. However, in the absence of thiopurine efficacy at 3-6 months of treatment, 6-TGN testing can be useful. If the level of 6-TGN is very high (Fig. 2), there is no reason to try increasing the dose and the next line of therapy can be tried. If the
Table 2. Practical behaviour facing various situations with thiopurines Pregnancy No evidence of teratogenicity (www.lecrat.org, ECCO consensus) Individual consideration of the benefit/risk ratio (higher benefit during the 5 first years when thiopurines are efficacious Diminution dose of 25-50 mg thiopurines at the 9th month if relative leucopenia of the mother (to prevent new-born leucopenia) Diagnosis of cancer with thiopurine treatment Interruption of thiopurines If possible wait for 5 years without any residual cancer and without any recurrence before restarting the treatment Live attenuated vaccines (yellow fever++) Discontinuation of thiopurines for 3 months, vaccine, restarting of thiopurines 1 month later (British Society for Rheumatology 2002) Occurrence or aggravation of warts and condyloma under thiopurine treatment In case of failure of local treatment, consider a reduction of thiopurine dosage, and even discontinuation Frequent herpes attacks Antiviral treatment or reduction of dosage/discontinuation of thiopurines
Fig. 1. Simplified metabolism of thiopurines 6-MMP: 6-methylmercaptopurine, 6-TU: acid 6-thiouric, 6-MMPR: 6-methylmercaptopurine ribonucleotides, 6-TIMP: 6-thioinosine 5’ monophosphate, 6-TGN: 6-thioguanine nucleotides)
level of 6-TGN is low and there is a haematological margin (neutrophils and lymphocytes), the dose can be increased in steps of 25-50 mg to try and obtain therapeutic efficacy. The association of 5-ASA with thiopurines increases the level of 6-TGN (thus the potential efficacy, as well as the potential toxicity of the thiopurines) and tends to reduce the level of methylated derivatives [37]. One must be aware of this interaction when adding 5-ASA to thiopurines, and increase haematological monitoring, or even reduce the dose of thiopurines a little in the absence of a haematological margin. Allopurinol, by inhibiting xanthine-oxidase, has the effect of significantly facilitating the production of 6-TGN and reducing the production of methylated derivatives. Prescription of allopurinol is thus a classically contraindicated association during thiopurine treatment. The Chicago team has recently suggested using this interaction in patients for whom thiopurines are not efficacious despite increases in dose, with the preferential production of
372 No efficacy expected after 3 to 6 months Dosage of 6-TGN < 220 > 300 Change of line
No haematological margin
Haematological margin
doses/steps of 25 mg
0 efficacy
TGN > 300
Efficacious
6-TGN < 220 and 0 efficacy Consider: ASA-allopurinol
Fig. 2. Optimisation of the thiopurine treatment with dosage of 6-TGN (a haematological margin means a circulating neutrophile polynuclear rate neatly higher to 2000/mm3 and lymphocytes higher than 500/mm3)
methylated derivatives and frequent hepatotoxicity [38]. For now, this ‘treatment’ remains experimental and restricted to certain experts. It is in fact haematologically dangerous, despite the reduction of 3/4 of the initial dose of thiopurines upon allopurinol introduction.
Changing the therapeutic strategy It is clear that the next therapeutic strategy (methotrexate or anti-TNF) should be tried in the event of thiopurine complications, intolerance to thiopurines or the appearance of a severe attack despite adjusted doses. In the case of a non-severe sporadic attack under treatment, it is possible to wait. In the case of repeated moderate attacks, one must try to pharmacologically optimise the treatment (Fig. 2) or, if this is impossible or not efficient, change the therapeutic strategy. In the case of incomplete response to thiopurines and the permanent necessity of complementary therapeutics to maintain control of the disease (budesonide, antibiotics) as is often the case of multi-stenotic attacks of the small bowel, it is preferable to move on to the next therapeutic strategy if it provides hope for overall improvement of the anatomical and clinical situation. When the clinical and biological situation is controlled but anatomical lesions persist in an evident manner or even become more aggravated, it is not established today that the therapeutic strategy must be changed based solely on the argument of trying to get the lesions to heal. However, this may rapidly change, because more and more arguments suggest that healing of the lesions is a prerequisite to good long-term prognosis for IBD, with the logic of deliberately not accepting a progression of lesions towards irreversible anatomical damage.
relapse following corticotherapy, but only after failure or complications with thiopurines for all situations [1]. The level of proof of efficacy of methotrexate is high, particularly in one meta-analysis [39] and two founding essays published in the New England Journal of Medicine. The first essay showed the superiority of 25 mg/week of methotrexate compared to placebo for inducing remission and obtaining severage from corticosteroids, essentially in the sub-group of patients with a high level of corticodependance [40]. The second showed the significant effect of 15 mg/weeks of methotrexate compared to placebo for maintaining methotrexate induced remission [41]. The efficacy of methotrexate has not yet been validated for UC (METEOR essay from GETAID, in press). The use of methotrexate for UC is thus not recommended today.
Contra-indications and precautions for use Methotrexate is a suspensive and reversible teratogen. Its use is incompatible with conception for men and women and requires strict contraception to be maintained for three months following cessation of treatment in women and five months in men (SPC). Methotrexate must not be used in the case of proven fibrotic subjacent chronic hepatopathy. If hepatopathy is possible (obese or alcoholic patients) it must be ruled out (ultrasound, FibroTest, FibroScan, or HPB) before beginning methotrexate. Severe renal and respiratory insufficiencies are also contraindications. Live vaccines should not be used as with thiopurines.
Use
Methotrexate
The required biological panel is the same as for the thiopurines. Before commencing treatment, the patient should be informed in detail (www.getaid.org) with particular emphasis placed on the imperative character of contraception. Initially, methotrexate should be administered at 25 mg/week via the deep sub-cutaneous or intra-muscular route, since the bioavailability of the product via the oral route is weak and very uncertain. After three months, it is possible to move to 15 mg/week but, to keep the patient in stable prolonged remission, it is often necessary to temporarily or definitively return to 25 mg/week. Folate supplementation at a distance from the injection (5 mg folic acid one day is enough) is necessary. It is imperative to know all the drug interactions that may increase the haematological toxicity of methotrexate. Association with trimethoprime, aspirin and phenylbutazone is contraindicated; NSAIDS and penicillins are discouraged. Biological monitoring may follow that done for thiopurines, periodically adding creatinaemia testing. In fact, on one hand methotrexate can be nephrotoxic, on the other renal insufficiency from another cause may increase the haematological toxicity of methotrexate.
Indications and expected efficacy
Complications
For Crohn’s disease, according to the European consensus the potential indications for methotrexate are situations of corticosteroid resistance and corticodependance and early
A reaction of clinical intolerance in the hours following the methotrexate injection is frequent, starting from the beginning of treatment or later on. It diversely associates a
373 sensation of inebriation, headaches, nausea, and fatigue. It leads certain patients to have the injection on the weekend and to premedicate with domperidone or paracetamol, like with interferon. In the worst case these clinical ‘drawbacks’ to the injection may lead to stopping treatment. Methotrexate may be myelotoxic throughout its use. Interstitial immunoallergic pneumonia is specific to methotrexate. Its incidence reaches up to three out of 100 patients per year in certain series. The main sign is cough, more or less associated with a moderate fever and dyspnea. Any prolonged cough under methotrexate that is not part of an obvious concurrent etiological context requires the advice of a specialist. Hepatic toxicity is generally announced by cytolysis. If the methotrexate is continued, it may lead to fibrosis or even cirrhosis. Nevertheless, in the absence of any biological hepatic anomaly and hepatopathy risk factor, it is no longer proposed by rheumatologists, who use methotrexate much more than we do in similar doses, to perform a HPB for a cumulative dose of 3-5 g (or 3-5 years of use). The FibroTest appears to be a very interesting substitute test for verifying hepatic integrity in this context (JFPD 2008).
Calcineurin inhibitors Cyclosporin and tacrolimus are currently indicated for serious acute colitis from UC as a second line therapeutic after failure with intravenous corticosteroids, as an alternative to infliximab [11]. They cannot be entirely assimilated to a conventional treatment and their constraints and use are not developed here.
Combination of conventional treatments with anti-TNF therapy For Crohn’s disease, although conventional immunosuppressors are being used more frequently and earlier overall, the European consensus does not cast doubt on the order of therapeutic strategies, which is thiopurines in the first intention, then methotrexate, and anti-TNFs in the event of failure with well conducted treatment with conventional immunosuppressors. In practice, it should be noted that in France, methotrexate is used less than antiTNFs (CESAME). It may be estimated that, based on the use of conventional treatments, out of 100 newly diagnosed patients, 70 will receive thiopurines within the first year of the disease, and 50 will progressively go to anti-TNFs within the first five years of the disease for diverse reasons (intolerance to thiopurines, primary failure, and loss of efficacy), without always transitioning through the methotrexate step. For HCR, the use of thiopurines is relatively recent, due to the frequency of possible problems following coloproctectomy (poor functional result, occlusion, stenosis, pouchitis, and hypofertility), so the overall tendency is to push medical treatment even further before turning to the surgical indication of HCR that is not responsive to medical treatment. The same escalating strategy is mounted, however a) methotrexate currently does not have a role and
b) infliximab may be administered the first time to treat serious colitis in a thiopurine naive patient; in this case, thiopurines alone may, for certain cases, take the efficacy relay of infliximab.
Infections and cancer under conventional treatment therapies and anti-TNF therapy Infections Until recently, with the exception of the periodic publication of severe and sometimes opportunistic and fatal infections, there had been no prospective epidemiological study comparing the frequency and the type of infection as a function of anti-inflammatory and/or immunosuppressive treatment received by patients. Four recent studies give the first indications: – tuberculosis is more frequent in patients treated with anti-TNFs [42]; – in ambulatory patients, the incidence of common warts and herpes is more frequent in patients receiving thiopurine monotherapy than in those receiving neither corticosteroids nor immunosuppressors [22]; – the TREAT register data suggests that severe infections and death are around two times more frequent in patients under corticosteroids than those who are not, after adjusting for other treatments. This additional risk has not been observed for thiopurines and anti-TNFs [13]; – a case control study from the Mayo clinic has shown that the odds ratio of infection increases with the number of immunosuppressors (in the large sense; corticosteroids, thiopurines, and anti-TNFs) received by patients (odds ratio between 3.1 and 4.4 for one immunosuppressor, and 12.9 for two immunosuppressors) and that the type of infection favoured is different according to the immunosuppressor received: candidosis for corticosteroids, viral infections for thiopurines, and intra-cellular agents (BK, aspergillosis, etc.) for anti-TNFs [14].
Cancer To date, epidemiological studies done using historical IBD registers (Scandinavia, Olmstedt County) or the British general physicians’ database have not been able to specify the risk of cancer related to the use of immunosuppressors, due to the low proportion of patients treated until recently. Concerning the risk of lymphoma, the meta-analysis by Kandiel et al. essentially concerning hospitalised populations suggests a four times higher risk for patients receiving thiopurines compared to those who are not [43]. The possible increase in risk of lymphoma is not known for methotrexate (not enough patients treated for IBD). The first data concerning anti-TNFs is discordant, with an excess of risk in the meta-analysis of rheumatoid polyarthritis clinical trials [44], neither found in the metaanalysis on IBD trials (JFPD 2008) nor in the TREAT register. Concerning other kinds of cancer, there is no firm data in this field today for IBD patients, while we know that transplant patients receiving conventional immunosuppressors have a neat excess of cancer.
374 For thiopurines (lymphomas and other cancers), the cooperative French study CESAME may yield, in the coming months, important information, with the reservation that end of study information is obtained from the majority of patients.
Conclusions Therapeutic practices for IBD have changed a lot over the past 10 years and may change significantly and rapidly over the months/years to come. Globally, conventional
immunosuppressor treatments and anti-TNFs are given more and more often and earlier for the disease, without current discussion regarding the sequence of conventional anti-TNF treatments. We move towards increasingly restrained use of corticosteroids and towards preferring monotherapies in order to avoid adding, or even multiplying (for the risk of infection) the iatrogenic risks. The order of lines of therapy and their length will eminently depend on comparative safety data, particularly with regard to severe infections and cancers. The management of these risks already greatly conditions, and will increasingly condition, our practices in the care of patients with IBD.