استكشاف العلاقة المحتملة بين ناهضات مستقبلات GLP-1 والسلوكيات الانتحارية أو الذاتية الجرح: دراسة مراقبة دوائية مستندة إلى قاعدة بيانات نظام الإبلاغ عن الأحداث الضارة التابع لإدارة الغذاء والدواء الأمريكية Exploration of the potential association between GLP-1 receptor agonists and suicidal or self-injurious behaviors: a pharmacovigilance study based on the FDA Adverse Event Reporting System database

المجلة: BMC Medicine، المجلد: 22، العدد: 1
DOI: https://doi.org/10.1186/s12916-024-03274-6
PMID: https://pubmed.ncbi.nlm.nih.gov/38355513
تاريخ النشر: 2024-02-14

استكشاف العلاقة المحتملة بين ناهضات مستقبلات GLP-1 والسلوكيات الانتحارية أو الذاتية الجرح: دراسة مراقبة دوائية مستندة إلى قاعدة بيانات نظام الإبلاغ عن الأحداث الضارة التابع لإدارة الغذاء والدواء الأمريكية

جيانشينغ زو يو زينغ باوهوا شيو سونغجونغ لونغ لي إيه زو يونهوي ليو تشنغليانغ لي ييفان زانغ ماوباي ليو و شيو مي وو (د)

الملخص

الخلفية إن تحديد ما إذا كان هناك علاقة محتملة بين ناهضات مستقبلات الببتيد الشبيه بالجلوكاجون 1 (GLP-1RAs) والسلوكيات الانتحارية أو السلوكيات الذاتية الجرح (SSIBs) أمر بالغ الأهمية لسلامة الجمهور. وقد بحثت هذه الدراسة في العلاقة المحتملة بين GLP-1RAs وSSIBs من خلال استكشاف قاعدة بيانات نظام الإبلاغ عن الأحداث الضارة التابع لإدارة الغذاء والدواء (FAERS).

الطرق تم إجراء تحليل عدم التناسب باستخدام بيانات ما بعد التسويق من مستودع FAERS (الربع الأول من 2018 إلى الربع الرابع من 2022). تم تحديد حالات SSIB المرتبطة بمثبطات GLP-1RAs وتحليلها من خلال تحليل عدم التناسب باستخدام مكون المعلومات. تم استخدام التوزيع المعلمي مع اختبار ملاءمة لت分析 وقت بدء الأعراض، و تم استخدام الانكماش لتقييم التأثير المحتمل للعلاج المشترك على حدوث سلوكيات الانتحار الذاتي.

النتائج: تم تحديد 204 حالة من حالات السلوكيات الانتحارية المرتبطة بمستقبلات GLP-1RA، بما في ذلك السيماغلوتيد، والليراجلوتيد، والدولاجلوتيد، والإكسيناتيد، والألبiglوتيد، في قاعدة بيانات FAERS. أظهر تحليل وقت الظهور عدم وجود آلية متسقة لفترة تأخر السلوكيات الانتحارية في المرضى الذين يتلقون GLP-1RAs. لم يشير تحليل عدم التناسب إلى وجود ارتباط بين GLP-1RAs والسلوكيات الانتحارية. كشف تحليل الأدوية المشتركة عن 81 حالة مع مضادات الاكتئاب، ومضادات الذهان، والبنزوديازيبينات، والتي قد تكون مؤشرات على الاضطرابات النفسية المصاحبة. الاستنتاجات: لم نجد أي إشارة إلى تقارير غير متناسبة عن ارتباط استخدام GLP-1RA والسلوكيات الانتحارية. يحتاج الأطباء إلى الحفاظ على يقظة متزايدة تجاه المرضى الذين تم علاجهم مسبقًا بأدوية نفسية عصبية. وهذا يساهم في قبول أكبر لمستقبلات GLP-1RA في المرضى الذين يعانون من داء السكري من النوع الثاني أو السمنة.

أهم النقاط

  • لتحديد ما إذا كان هناك علاقة محتملة بين ناهضات مستقبلات الببتيد الشبيه بالجلوكاجون 1 (GLP1RAs) والسلوكيات الانتحارية أو السلوكيات الذاتية الجرح (SSIBs).
  • هل هناك ارتباط مباشر بين GLP-1RAs و SSIBs؟ لا توجد أدلة تشير بشكل معقول إلى وجود ارتباط بين GLP-1RAs و SSIBs بناءً على الخصائص السريرية، ومدة البداية، وعدم التناسب، وتحليل الأدوية المساعدة.
  • يجب على الأطباء إيلاء المزيد من الاهتمام للحالة النفسية للمرضى الذين لديهم تاريخ من استخدام الأدوية النفسية العصبية، ويحتاج الأمر إلى مراقبة أكثر شمولاً للنظر بعناية في قابليتهم للإصابة بالآثار الجانبية النفسية.
    الكلمات الرئيسية: GLP-1RAs، السلوكيات الانتحارية أو الذاتية الجرح، اليقظة الدوائية، قاعدة بيانات FAERS، داء السكري من النوع 2، السمنة

الملخص الرسومي

النقطة البارزة: لم تجد الدراسة الحالية أي إشارة إلى تقارير غير متناسبة عن وجود ارتباط بين استخدام GLP-1RA و SSIBs. يحتاج الأطباء إلى الحفاظ على يقظة متزايدة تجاه المرضى الذين تم تحضيرهم مسبقًا بأدوية نفسية عصبية.
دراسة مراقبة سلامة الأدوية غير المتناسبة لبيانات ما بعد التسويق في قاعدة بيانات نظام الإبلاغ عن الأحداث الضارة التابع لإدارة الغذاء والدواء الأمريكية
أسفر استخدام أدوية GLP-1RAs عن تسجيل إجمالي 91,431 حالة من عام 2018 إلى عام 2022.
تشمل الأدوية المعنية السيماغلوتيد، الليراجلوتيد، الدولاكوتيد، الإكسيناتيد، الليكسيسيناتيد، والألباجلوتيد
204 حالة من السلوكيات الانتحارية أو السلوكيات المؤذية للذات
استكشاف العلاقة المحتملة بين ناهضات مستقبلات GLP-1 والسلوكيات الانتحارية أو السلوكية الذاتية المؤذية: دراسة مراقبة سلامة الأدوية استنادًا إلى قاعدة بيانات نظام الإبلاغ عن الأحداث الضارة التابعة لإدارة الغذاء والدواء الأمريكية
جيانشينغ زو، يو زينغ، باوهوا شيو، سونغجون لونغ، لي-إي زو، يونهوي ليو، تشينغليانغ لي، ييفان زانغ، ماوباي ليو، شيويمي وو*

تحليل وقت بدء الأعراض

لا يمكن لآلية واحدة أن تفسر الارتباط الزمني بين بداية استخدام GLP-1RAs و SSIBs.
استكشاف العلاقة المحتملة بين ناهضات مستقبلات GLP-1 والسلوكيات الانتحارية أو السلوكية الذاتية المؤذية: دراسة مراقبة دوائية مستندة إلى قاعدة بيانات نظام الإبلاغ عن الأحداث الضارة التابعة لإدارة الغذاء والدواء الأمريكية
جيانشينغ زو، يو زينغ، باوهوا شيو، سونغجون لونغ، لي-إي زو، يونهوي ليو، تشينغليانغ لي، ييفان زانغ، ماوباي ليو، شيويمي وو*

التحليل الوصفي

جميع الدوائر المتكاملة لا توجد علاقة ذات دلالة إحصائية بين GLP-1RAs و SSIBs
استكشاف العلاقة المحتملة بين ناهضات مستقبلات GLP-1 والسلوكيات الانتحارية أو السلوكية الذاتية المؤذية: دراسة مراقبة سلامة الأدوية استنادًا إلى قاعدة بيانات نظام الإبلاغ عن الأحداث الضارة التابعة لإدارة الغذاء والدواء الأمريكية
جيانشينغ زو، يو زينغ، باوهوا شيو، سونغجون لونغ، لي-إي زو، يونهوي ليو، تشينغليانغ لي، ييفان زانغ، ماوباي ليو، شيوماي وو*

تحليلات التزامن الدوائي

كان من المرجح أن تُنسب SSIBs إلى الحالة النفسية للمريض بدلاً من أن تكون ناتجة عن GLP-1RAs مباشرة.
استكشاف العلاقة المحتملة بين ناهضات مستقبلات GLP-1 والسلوكيات الانتحارية أو السلوكية الذاتية المؤذية: دراسة مراقبة دوائية مستندة إلى قاعدة بيانات نظام الإبلاغ عن الأحداث الضارة التابعة لإدارة الغذاء والدواء الأمريكية
جيانشينغ زو، يو زينغ، باوهوا شيو، سونغجون لونغ، لي-إي زو، يونهوي ليو، تشينغليانغ لي، ييفان زانغ، ماوباي ليو، شيويمي وو*
تقييم الرابط السببي العالمي
على مستوى العالم، لم يتم الوفاء بمعايير برادفورد هيل

الخلفية

محفزات مستقبلات الببتيد الشبيه بالجلوكاجون 1 (GLP-1RAs) هي مشتقات معدلة من الببتيد الشبيه بالجلوكاجون 1 (GLP-1) التي تحاكي التأثير التنظيمي لـ GLP-1 على مستويات الجلوكوز في الدم. تكمن المزايا الرئيسية لـ GLP-1RAs في تقليل الجلوكوز، والفوائد القلبية الكلوية، وتحسين متلازمة الأيض، وكل ذلك مدعوم بوفرة من الدراسات المستندة إلى الأدلة [1-4]. لذلك، ظهرت GLP-1RAs كواحدة من التدخلات الدوائية الموصى بها لمرض السكري من النوع 2، خاصة للمرضى الذين لديهم عوامل خطر قلبية وعائية وأمراض مصاحبة موجودة (مثل،
مرض الكلى المزمن) [5]. علاوة على ذلك، تظهر أدوية GLP-1RAs تأثيرات كبيرة في تقليل الوزن من خلال تسهيل امتصاص الجلوكوز وكبح الشهية [6]. وفقًا لمراكز السيطرة على الأمراض والوقاية منها، يحتل مرض السكري والسمنة المرتبتين بين أعلى 10 أمراض مزمنة من حيث العبء الاقتصادي في الولايات المتحدة الأمريكية. نظرًا للازدياد المتسارع في انتشار هذه الأمراض على مستوى العالم، هناك فرصة غير مسبوقة لتوسيع السوق العالمية لأدوية GLP-1RAs.
منذ موافقة إدارة الغذاء والدواء (FDA) على الإكسيناتيد لعلاج مرض السكري من النوع 2 في عام 2005، أصبحت الأبحاث حول مثبطات GLP-1 موضوعًا ساخنًا
الموضوع والأنالوجات الجديدة يتم تقديمها باستمرار إلى السوق. في ديسمبر 2014، وافقت إدارة الغذاء والدواء الأمريكية على الليراجلوتيد كدواء لفقدان الوزن. بعد ذلك مباشرة، أصبح السيماغلوتيد متاحًا مع دلالة مرض السكري في عام 2017، ثم في عام 2020، تم أيضًا الترخيص لعلاجه للسمنة. الطلب العالمي على السيماغلوتيد والليراجلوتيد يتجاوز العرض الحالي، والذي قد يكون مرتبطًا باستخداماتهما غير المصرح بها وسوء الاستخدام. على الرغم من الإمكانات الكبيرة والمزايا لمثبطات GLP-1RAs، إلا أنها مصحوبة حتمًا بمجموعة من ردود الفعل السلبية المحتملة (ADRs)، كما هو الحال مع أي دواء. تشمل أكثر ردود الفعل السلبية شيوعًا المرتبطة بعلاج GLP1RA أحداثًا معوية مثل الغثيان والقيء والإسهال. ومع ذلك، فإن هذه الردود غير الخطيرة مقبولة لدى الأفراد المصابين بالسكري والسمنة. ما يثير القلق هو بعض التقارير الأخيرة عن ردود الفعل السلبية الخطيرة المرتبطة بمثبطات GLP1RAs: (i) إفصاح شركة فايزر عن ارتفاع إنزيمات الكبد في مستخدمي اللوتيغليبرون (وهو مثبط GLP-1RA جديد عن طريق الفم) يدل على إصابة خلايا الكبد؛ (ii) كشفت الدراسات عن ارتفاع خطر الإصابة بسرطان الغدة الدرقية المرتبط والتهاب المرارة مع استخدام مثبطات GLP-1RAs؛ (iii) قدمت وكالة الأدوية الأيسلندية 150 تقريرًا عن سلوكيات انتحارية أو ذاتية الإصابة مرتبطة بمثبطات GLP-1RAs إلى وكالة الأدوية الأوروبية. بعد الحوادث المذكورة، تم إيقاف تطوير اللوتيغليبرون، وصنفت وكالات تنظيم الأدوية في عدة دول مثبطات GLP-1RAs كعامل خطر محتمل لسرطان الغدة الدرقية وأمراض المرارة والقنوات الصفراوية. ومع ذلك، لم تقم أي دراسات بالتحقيق في العلاقة بين مثبطات GLP-1RAs وسلوكيات الانتحار أو الإصابة الذاتية. تعتبر سلوكيات الانتحار أو الإصابة الذاتية، المعترف بها كردود فعل سلبية هامة، قد حظيت باعتراف واسع كعوامل حاسمة قد تعرض سلامة المرضى للخطر. أظهرت الدراسات تصاعدًا مستمرًا عامًا بعد عام في حالات انتحار المرضى المرتبطة باستخدام الأدوية، مما زاد من القلق بين الجمهور العام والمهنيين السريريين بشأن هذه الظاهرة. لذلك، في سياق الاستخدام العالمي الواسع لمثبطات GLP-1RAs، من الضروري التحقيق فيما إذا كانت هناك علاقة محتملة بين مثبطات GLP-1RAs وسلوكيات الانتحار أو الإصابة الذاتية، وهو أمر ذو أهمية قصوى لسلامة الأدوية العامة والسريرية.
نظام الإبلاغ عن الأحداث السلبية التابعة لإدارة الغذاء والدواء (FAERS) هو قاعدة بيانات شاملة للأحداث السلبية الناتجة عن الأدوية، تم تصميمها خصيصًا من قبل إدارة الغذاء والدواء لتسهيل المراقبة بعد التسويق للأدوية والبيولوجيات العلاجية. تشمل قاعدة البيانات جميع الأحداث السلبية وأخطاء الأدوية الموثقة من قبل إدارة الغذاء والدواء. تسهل هذه القاعدة تحديد وتحليل كمي للإشارات التي تشير إلى الإبلاغ غير المتناسب عن الأحداث السلبية، مما يساعد في التعرف على الروابط بين أدوية معينة وأحداث سلبية معينة.
هدفت هذه الدراسة إلى استكشاف خصائص المرض المرتبطة بـ SSIBs المرتبطة بـ GLP-1RAs من خلال تحليل
قاعدة بيانات FAERS وقياس العلاقة بين GLP-1RAs و SSIBs من خلال مقاييس مثل وقت الظهور، وتحليل عدم التناسب، وتحليل الأدوية المشتركة. ستؤثر نتائج هذه الدراسة بشكل إيجابي على سلامة الأدوية للمرضى الذين يعانون من داء السكري من النوع الثاني أو السمنة، مما يوفر دعمًا كبيرًا لخيارات الأدوية السريرية.

طرق

مصادر البيانات وتصميم الدراسة

قاعدة بيانات FAERS، واحدة من أكبر قواعد البيانات المتاحة للجمهور لتقارير ردود الفعل السلبية، توفر للباحثين بيانات خام من موقع إدارة الغذاء والدواء الأمريكية (FDA).https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html). لمعالجة البيانات الخام المستخرجة من قاعدة بيانات FAERS، استخدمنا Microsoft SQL Server (الإصدار 2019؛ شركة مايكروسوفت، ريدموند، واشنطن، الولايات المتحدة الأمريكية). نظرًا لأن التقارير المكررة أو غير المعقولة يمكن أن تؤدي إلى أخطاء كبيرة [22]، تم إجراء تنظيف شامل وتطبيع لضمان بيانات قابلة للتفسير (تم تقديم العملية بالتفصيل في الملف الإضافي 1: الشكل S1 [23-25]).
تم اختيار سنة الدراسة بناءً على توفر الأدوية والبيانات. عند بدء هذه الدراسة، كانت إدارة الغذاء والدواء الأمريكية (FDA) قد نشرت فقط بيانات الربع الأول من عام 2023. تم استبعاد تيرزيباتيد (الذي تم تسويقه في سبتمبر 2022) بسبب نقص البيانات، وكان سيماجلوتيد هو أحدث دواء تم تسويقه (ديسمبر 2017). لذلك، قمنا بإجراء تحليل غير متناسب لبيانات ما بعد التسويق من مستودع FAERS للتحقيق في العلاقة المحتملة بين SSIBs وGLP-1RAs. تم اختيار البيانات لكل سنة كاملة منذ تسويق أحدث دواء من الربع الأول من 2018 إلى الربع الرابع من 2022. لم يكن هناك حاجة للحصول على موافقة أخلاقية حيث تم إجراء الدراسة باستخدام بيانات عامة متاحة بدون تعريف.

تعريف الحالات والأدوية ذات الاهتمام

في قاعدة بيانات FAERS، يتم تصنيف ردود الفعل الضارة (ADRs) وفقًا لمصطلحات القاموس الطبي للأنشطة التنظيمية (MedDRA) من حيث العلامات والأعراض، والتي تُسمى المصطلحات المفضلة (PTs). تشمل ردود الفعل الضارة المعنية أفكار الانتحار، وأفكار إيذاء النفس، وسلوكيات الانتحار، وسلوكيات إيذاء النفس (انظر الملف الإضافي 1: الجدول S1 للحصول على قائمة مفصلة)، والتي تُنسب بشكل متجانس إلى مصطلح مجموعة عالية المستوى (HLGT) يسمى سلوكيات الانتحار وإيذاء النفس حسب MedDRA (الإصدار 26.0). تشمل الأدوية المعنية أدوية GLP-1RAs المعتمدة من إدارة الغذاء والدواء [26]: سيماجلوتيد، ليراجلوتيد، دولاجلوتيد، إكسيناتيد، ليكسيسيناتيد، وألبiglutide. في هذه الدراسة، تم استخدام فينلافاكسين كعنصر تحكم إيجابي لأنه مرتبط بشكل كبير بسلوكيات إيذاء النفس والانتحار [27]. تم اختيار أورليستات وإمباغليفلوزين كعناصر تحكم سلبية بناءً على تشابههما في السكان المشار إليهم لأدوية GLP-1RAs، دون أي دليل مُبلغ عنه عن إيذاء النفس.
أو إمكانية الانتحار في المواصفات أو الدراسات ذات الصلة. شملت الدراسة الحالية جميع الحالات التي تم فيها إعطاء GLP1RAs وتم إدراجها كأدوية مشتبه بها رئيسية، أدوية مشتبه بها ثانوية، أو أدوية مصاحبة لـ SSIBs.

التحليل الوصفي

تم توثيق الخصائص السكانية للمرضى (العمر، الجنس، ونوع المراسل) والجوانب السريرية (فترة الكمون، النتيجة، والدلالة) في تقارير حالات الآثار الجانبية الخطيرة المرتبطة بمستحضرات السيماغلوتيد، والليغلايتيد، والدولاجلوتيد، والإكسيناتيد، والليكسيسيناتيد، والألباجلوتيد.

تحليل الوقت المبلغ عنه لبدء الأعراض

لوصف زمن الاستجابة لمثبطات SSIBs مع GLP-1RAs، قمنا بإجراء تحليل زمن البداية (TTO) من خلال اختيار نموذج التوزيع المعلمي الذي أظهر أفضل نتيجة في اختبار جودة الملاءمة بين توزيعات ويبول، اللوغاريتمي الطبيعي، غاما، والأسية. يتميز التوزيع المعلمي بمعامل المقياس ومعامل الشكل [28]. معامل المقياس يمثل حجم الأحداث الضائرة التي تحدث في الكمية داخل دالة التوزيع. اعتمادًا على قيمة معامل الشكل الحد الأقصى لـ فترة الثقة (CI) لـ يشير إلى زيادة أولية، تليها انخفاض في معدل المخاطر (نوع الفشل المبكر). على العكس، الحد الأدنى من CI لـ يظهر خطرًا متزايدًا مع مرور الوقت (نوع فشل التآكل)، بينما CI يشمل 1 لـ ، مما يشير إلى معدل خطر ثابت طوال فترة التعرض (نوع الفشل العشوائي). مزيد من التفاصيل حول تحليل TTO وتصنيف يمكن العثور عليه في الملف الإضافي 1، الأقسام 4-5.

تحليل عدم التناسب

تم اعتماد نهج تسلسلي لمعالجة العوامل المربكة الرئيسية بشكل منهجي:
(1) نهج استكشافي غير متناسب يقارن بين GLP-1RAs وجميع الأدوية الأخرى المبلغ عنها في قاعدة بيانات FAERS، باستخدام فينلافاكسين كعنصر تحكم إيجابي وإمباغليفلوزين/أورليستات كعنصر تحكم سلبي. يمكن العثور على تعريف نهج عدم التناسب في الملف الإضافي 1: القسم 4 [29].
(2) استخدمنا مكون المعلومات البايزية (IC) لحساب الحد الأدنى من سي آي ( ). هذه الطريقة أكثر دقة من الاعتماد على نسبة الأرجحية في التقرير، خاصة في الحالات التي تحتوي على عدد محدود من الحالات [30]. يمكن العثور على إجراء حساب IC في الملف الإضافي 1: القسم 5.
(3) تم إجراء تحليل للنتائج السلبية الكاذبة من خلال زيادة معدل حدوث GLP-1RA المبلغ عنه بشكل مصطنع.
SSIBs المرتبطة بـ تهدف هذه الدراسة إلى التحقق مما إذا كان يمكن أن يُعزى غياب النتائج الإيجابية إلى استبعادنا لبعض الحالات التي تتضمن الاستخدام المتزامن لعدة أدوية GLP-1RAs. هذه الفحص الدقيق يدعم بشكل فعال مصداقية النتائج السلبية.
(4) نظرًا للاختلافات الجوهرية والانحياز المحتمل في التقارير في قاعدة بيانات FAERS، قمنا بإجراء سلسلة من تحليلات الحساسية. أولاً، استبعدنا تقارير الحالات وغير الحالات التي تتعلق بالأحداث المعوية (جميع PTs تنتمي إلى اضطرابات الجهاز الهضمي في فئة الأعضاء النظامية) من مجموعة البيانات. كانت هذه الخطوة تهدف إلى تقليل تأثير التعتيم الناتج عن الأحداث المعوية لتجنب الانحياز التنافسي المحتمل. علاوة على ذلك، كانت المؤشرات الرئيسية في الحالات المشمولة هي داء السكري من النوع 2 والسمنة، وهما من عوامل الخطر المحتملة لـ SSIBs. لذلك، قمنا بتضييق مجموعة بيانات التحليل لتشمل الأفراد الذين يعانون من داء السكري من النوع 2 أو فقدان الوزن من خلال حقل indi_pt (الذي يمثل المؤشر). أظهرت الدراسات السابقة أن تقييد السكان الذين تم تحليلهم يمكن أن يخفف من انحياز المؤشر [21]. IC تم حسابه باستخدام مجموعة البيانات المعالجة لإجراء تحليلات الحساسية المذكورة أعلاه.
(5) إن اعتبار عدد منخفض من الحالات المتوقعة قد يؤدي إلى حساسية غير كافية لاكتشاف عدم التناسب بالقوة ذات الصلة [31]، قمنا بالإبلاغ عن حساسية عينة تمثيلية لقياس موثوقية النتائج السلبية. تم اعتبار النتيجة موثوقة عندما كانت الحساسية لاكتشاف العينة التمثيلية يمكن العثور على إجراء الحساب في الملف الإضافي 1: الشكل S2.

تحليل التداخل الدوائي

تم إجراء تحليل لحالة بحالة من قبل صيدليين مستقلين لتحديد الأدوية المجمعة من خلال البحث في ملفات DRUG وTHER للحالات الـ 204 المشمولة. عندما كانت هناك تناقضات، تم إدخال صيدلي ثالث لاتخاذ القرار النهائي. قمنا بإجراء تحليل للتداوي المشترك لمستقبلات GLP-1 والأدوية العشرين الأكثر استخدامًا في حالات SSIBs المرتبطة بمستقبلات GLP-1. حساب “ لهذه الأدوية يتماشى مع الإجراء الموصوف في القسم 2.5. استخدمنا الانكماش لقياس تفاعلات الأدوية مع الأدوية الأخرى لأن دراسة سابقة أظهرت أنه الطريقة الأكثر تحفظًا بين عدة خوارزميات [32]. معيار الكشف هو الحد الأدنى من CI من عملية الحساب لـ وتم وصف قائمة الأدوية في الملف الإضافي 1: الجدول S2. عندما تم تسجيل دواء نفسي عصبي واحد على الأقل (مثل مضادات الذهان، ومضادات الاكتئاب، ومهدئات القلق) في التقرير، تم تعريف المريض بأنه لديه تاريخ من الأدوية.

تقييم عالمي للأدلة

تم استخدام معايير برادفورد هيل المعدلة لتقييم العلاقات المحتملة بين الأدلة المتاحة المختلفة [33-35]. يمكن أن تقيم الاحتمالية البيولوجية ما إذا كان هناك آلية لدعم العلاقة المحتملة بين ردود الفعل السلبية للأدوية (ADRs) والدواء المدروس. كما أخذنا في الاعتبار قوة الأدلة (حجم التأثيرات الملحوظة) وتناسق النتائج عبر دراسات أو مصادر مختلفة. بالإضافة إلى ذلك، تم استخدام الخصوصية لتقييم ما إذا كانت بعض الأحداث مرتبطة بعوامل محددة، وكان التماسك لفحص التسلسل الزمني للأحداث الذي يتماشى مع العلاقة المحتملة. وأخيرًا، تم استخدام أدوية مشابهة أخرى (إمباغليفلوزين وأورليستات) كتحكم سلبي لدعم الاستنتاج النهائي.

التحليل الإحصائي

تم تعريف عتبة عدم التناسب لإشارة التقرير على أنها ، وتم تعريف التفاعل الإيجابي بين الأدوية على أنه تم استخدام R (الإصدار 4.3.1؛ مؤسسة R للحوسبة الإحصائية، فيينا، النمسا) للرسم، وتحليل TTO، وحساب IC و النسبة المتضخمة المستخدمة في تحليل النتائج السلبية الكاذبة كانت محدد بناءً على استراتيجية خطوة بخطوة (الملف الإضافي 1: الشكل S3، والجدول S3).

النتائج

التحليل الوصفي

من الربع الأول من عام 2018 إلى الربع الرابع من عام 2022، تم تحديد 57 و74 و51 و20 و2 تقرير عن أحداث سلبية خطيرة مرتبطة بالسماغلوتيد والليراجلوتيد والدولاجلوتيد والإكسيناتيد والألبiglوتيد، على التوالي، في قاعدة بيانات FAERS. تم استبعاد ثلاثة مرضى بسبب الاستخدام المتزامن لعدة أدوية من فئة GLP-1RAs، ولم يتم الكشف عن أي تقارير عن أحداث سلبية خطيرة لليكسيسيناتيد. أظهرت توزيع السنوات المبلغ عنها نمطًا متساويًا بشكل عام، مع وجود اتجاه تصاعدي ملحوظ في عام 2022. كانت الغالبية (54.90%) من التقارير عن أحداث سلبية خطيرة باستخدام GLP-1RAs قد قدمها محترفو الرعاية الصحية. على الرغم من الكمية الصغيرة من البيانات المفقودة أو غير المعروفة ( )، كانت نسبة التقارير في الإناث أعلى بشكل ملحوظ من تلك في الذكور ( 64.22 مقابل ، على التوالي). من بين التقارير التي تتوفر فيها بيانات، تم ملاحظة نصف الحالات في البالغين (18-65 سنة). داء السكري من النوع 2 ( ) وفقدان الوزن ( ) ظهرت كأكثر المؤشرات سطوعًا، بينما شكلت مؤشرات أخرى والمجهول أو المفقود يشكل (الجدول 1).

تحليل الوقت المبلغ عنه لبدء الأعراض

تم استخدام إجمالي 52 حالة صالحة لتحليل TTO (الشكل 1). من بينها، أبلغت ثلاث حالات عن SSIBs بعد يوم واحد، ويوم واحد، وثلاثة أيام من توقف GLP-1RAs (حالة واحدة لـ
liraglutide وحالتين لـ semaglutide، والتي قد تكون مرتبطة بتأخر ردود الفعل الضارة الناتجة عن الأدوية المعنية.
في تحليل TTO استنادًا إلى توزيعات المعلمات والحالات الصالحة، كانت الوسيطات TTO (نطاق الربيع الربعي) لـ SSIBs المرتبطة بالسيماجلوتيد، والليراجلوتيد، والدولاجلوتيد، والإكسيناتيد هي ، “، و 2.5 (1.5-169.5، ) أيام، على التوالي (لم يتم تضمين الألبغليتيد في تحليل TTO بسبب عدم كفاية البيانات). أظهرت نتائج اختبار جودة الملاءمة (الملف الإضافي 1: الجدول S4) أن نموذج اللوغاريتم الطبيعي كان الأفضل لوصف فترة تأخير SSIBs المرتبطة بالليراجلوتيد والإكسيناتيد. كان نموذج غاما مناسبًا لـ SSIBs المرتبطة بالسيماغلوتيد، ونموذج ويبول وصف فترة تأخير SSIBs المرتبطة بالدولاجلوتيد بشكل جيد. تم تصنيف السيماغلوتيد والدولاجلوتيد كأنواع فشل مبكر، والإكسيناتيد كنوع فشل عشوائي، والليراجلوتيد كنوع فشل نتيجة للاستخدام (الملف الإضافي 1: الجدول S5 للبيانات الخام).

تحليل عدم التناسب

أولاً، تم إجراء تحليل أولي لمقارنة GLP1RAs مع أدوية أخرى في قاعدة بيانات FAERS. كانت GLP-1RAs التي تم تقييمها هي السيماغلوتيد ( liraglutide )، دولاجلوتيد ( )، إكسيناتيد ( ) ، وألبجليتيد ( 3.91)، ولم يظهر أي منها إشارة للإبلاغ غير المتناسب. سلبية (إمباجليفلوزين، ; أورليستات، ) وإيجابي (فينلافاكسين، أكدت الضوابط صحة البيانات الداخلية لقاعدة البيانات. ثم قمنا بإجراء تحليل للنتائج السلبية الكاذبة حيث لم تحقق نتائج التحليل غير المتناسب لجميع أدوية GLP-1 مع SSIBs. بعد زيادة عدد التقارير بـ ، وظلت نتائج التحكم السلبية والإيجابية صالحة (الشكل 2). أخيرًا، تمت ملاحظته أيضًا في تحليلات الحساسية (الشكل 3). الحساسية لاكتشاف هذا كان لكل مثبطات GLP-1 في التحليلات الأولية والتحليلات السلبية الكاذبة. في بعض تحليلات المجموعات الفرعية، هناك بعض الحساسية لاكتشاف ذلك. كان < 0.8 للألبجليتيد والإكسيناتيد. يمكن العثور على البيانات الخام للتحليل أعلاه في الملف الإضافي 1: الجدول S8.

تحليل التداخل الدوائي

قمنا بإجراء تحليل للتداخل الدوائي لأعلى 20 دواءً تم الإبلاغ عن تكرار استخدامه لأعلى نسبة من SSIBs المرتبطة بـ GLP-1RA (مع الأخذ في الاعتبار نفس الحالات المبلغ عنها، تم تضمين 23 دواءً). تم تحديد عدة أدوية نفسية عصبية، بما في ذلك بوبروبيون وكويتيابين وأريبيبرازول، وأظهر تحليل عدم التناسب مع SSIBs إشارة للإبلاغ. وكشفت الفحوصات الإضافية أن 81 مريضًا كان لديهم تاريخ من
الجدول 1 خصائص الحالات المرتبطة بمثبطات GLP-1RA في قاعدة بيانات FAERS
فئات سيماجلوتيد (%) ليراجلوتيد (%) دولاجلوتيد (%) إكسيناتيد (%) ألبجليتيد (%) الإجمالي (%)
تقارير SSIBs ٥٧ 74 51 20 2 204
سنة التقرير
2018 2 (3.51) 16 (21.62) 12 (23.53) 4 (20.00) 1 (50.00) ٣٥ (١٧.١٦)
2019 4 (7.02) 11 (14.86) 7 (13.73) 7 (35.00) 0 ٢٩ (١٤.٢٢)
٢٠٢٠ 8 (14.04) 19 (25.68) 10 (19.61) 4 (20.00) 1 (50.00) 42 (20.59)
2021 16 (28.07) 12 (16.22) 8 (15.69) 2 (10.00) 0 ٣٨ (١٨.٦٣)
2022 ٢٧ (٤٧.٣٧) 16 (21.62) 14 (27.45) 3 (15.00) 0 60 (29.41)
مراسل
محترف الرعاية الصحية 30 (52.63) 53 (71.62) 19 (37.25) 9 (45.00) 1 (50.00) ١١٢ (٥٤.٩٠)
غير متخصص في الرعاية الصحية ٢٦ (٤٥.٦١) 21 (28.38) 31 (60.78) 6 (30.00) 1 (50.00) 85 (41.67)
غير معروف أو مفقود 1 (1.75) 0 1 (1.96) 5 (25.00) 0 7 (3.43)
جنس
ذكر 18 (31.58) 20 (27.03) 19 (37.25) 9 (45.00) 1 (50.00) 67 (32.84)
أنثى ٣٨ (٦٦.٦٧) 52 (70.27) ٢٩ (٥٦.٨٦) 11 (55.00) 1 (50.00) 131 (64.22)
غير معروف أو مفقود 1 (1.75) 2 (2.70) 3 (5.88) 0 0 6 (2.94)
فئة العمر، سنوات
قاصر (< 18) 0 0 0 3 (15.00) 0 3 (1.47)
بالغ (18-65) ٢٨ (٤٩.١٢) ٤٣ (٥٨.١١) ٢٤ (٤٧.٠٦) 7 (35.00) 0 ١٠٢ (٥٠.٠٠)
كبار السن (>65) 1 (1.75) 7 (9.46) 10 (19.61) 2 (10.00) 0 20 (9.80)
غير معروف أو مفقود ٢٨ (٤٩.١٢) ٢٤ (٣٢.٤٣) 17 (33.33) 8 (40.00) 2 (100.00) 79 (38.73)
الوسيط (المدى interquartile) ٤٣ (٣٥-٥٢) ٤٧ (٣٦-٥٧) ٥٥ (٤٦-٦٧) 40 (25-54) / /
إشارة
داء السكري من النوع الثاني 13 (22.81) 14 (18.92) 30 (58.82) 13 (65.00) 1 (50.00) 71 (34.80)
فقدان الوزن 15 (26.32) 15 (20.27) 0 0 0 30 (14.71)
الآخرون 1 (1.75) ٤ (٥.٤١) 0 3 (15.00) 0 8 (3.92)
غير معروف أو مفقود ٢٨ (٤٩.١٢) 41 (55.41) 21 (41.18) ٤ (٢٠.٠٠) 1 (50.00) 95 (46.57)
النتائج
الموت 1 (1.75) 17 (22.97) 3 (5.88) 1 (5.00) 0 22 (10.78)
يهدد الحياة 6 (10.53) 6 (8.11) 3 (5.88) 5 (25.00) 0 20 (9.80)
الاستشفاء 8 (14.04) 20 (27.03) 19 (37.25) 7 (35.00) 0 54 (26.47)
إعاقة 2 (3.51) 1 (1.35) 2 (3.92) 1 (5.00) 0 6 (2.94)
أمراض خطيرة أخرى 52 (91.23) 55 (74.32) ٣٧ (٧٢.٥٥) 12 (60.00) 2 (100) 158 (77.45)
نطاق الربيع بين الربيعيين (IQR)، سلوكيات الانتحار والإيذاء الذاتي (SSIBs)، ناهضات مستقبلات GLP-1 (GLP-1RAs)
المهنيون في الرعاية الصحية بما في ذلك المراسلون مثل الأطباء والصيادلة؛ غير المهنيين في الرعاية الصحية بما في ذلك المراسلون مثل المستهلكين والمحامين
نظرًا لأن الحالة قد تشهد نتائج سريرية مختلفة خلال العلاج الدوائي، فمن المعقول أن تتجاوز النسبة المئوية الإجمالية للنتيجة تحت هذا البند 100%
الأدوية النفسية العصبية، واستخدم 63 مريضًا أكثر من دواء نفسي عصبي واحد (الملف الإضافي 1: الشكل S4). بعد ذلك أظهرت نتائج الانكماش عدم وجود احتمال لتفاعلات دوائية بين جميع أدوية GLP-1RAs والأدوية النفسية العصبية (جميع) ; راجع الملف الإضافي 1: الجدول S6 للحصول على قائمة بأفضل 20 دواءً ، و ).

تقييم العلاقة السببية العالمية

عالمياً، لم يتم استيفاء معايير برادفورد هيل، كما يتضح من قوة عدم التناسب وتماسكها طوال التحليل، الزمني
العلاقات، والاحتمالية البيولوجية، وبالتالي عدم دعم ارتباط سببي محتمل بين مثبطات GLP-1 (سيماجلوتيد، ليراجلوتيد، دولاجلوتيد، إكسيناتيد، ليكسيزيناتيد، وألبiglوتيد) و SSIBs (الجدول 2).

نقاش

أجرت الدراسة الحالية تحليلًا شاملاً يتضمن أربعة أهداف رئيسية: أولاً، قمنا بتوصيف الميزات السريرية لحالات SSIB المتعلقة بـ GLP-1RA من خلال تحليل بيانات ما بعد التسويق من قاعدة بيانات FAERS؛ ثانيًا، قمنا بتطوير نموذج للكمون لهذه الحالات؛ ثالثًا، توصلنا إلى عدم وجود تقارير غير متناسبة.
الشكل 1 تحليل TTO لـ SSIBs المبلغ عنها لكل GLP-1 RA. تحليل TTO المبلغ عنه ومدة العلاج لـ SSIB المرتبطة بـ ليراجلوتيد، ب سيماغلوتيد، ج دولاجلوتيد، و د إكسيناتيد. تمثل الماس المظلمة TTO لـ SSIBs بعد إعطاء GLP-1RAs، بينما تشير القضبان الرمادية إلى مدة علاج GLP-1 لكل حالة. بسبب قيود توفر البيانات، يمكن رسم مدة العلاج فقط لمجموعة فرعية من الحالات. اختبار جودة الملاءمة لـ SSIB المرتبطة بـ ليراجلوتيد سيماجلوتيد دولاجلوتيد، و إكسيناتيد. يقع فوق الرسم البياني مخطط الصندوق الكمي. تتوافق القضبان الخضراء مع عدد الحالات التي حدثت ضمن التوزيع، بينما تمثل الخط الأخضر المنحنى الملائم للنموذج الذي يظهر أفضل النتائج. اعتمدنا توزيع لوغاريتمي طبيعي لوصف فترة الانتظار لليراجلوتيد وإكسيناتيد، وتوزيع غاما لسيماجلوتيد، وتوزيع ويبول لدولاجلوتيد. يمكن العثور على البيانات الخام لتحليل TTO في 52 حالة صالحة في الملف الإضافي 1: الجدول S7. TTO، الوقت حتى الظهور؛ SSIBs، السلوكيات الانتحارية والذاتية الإصابات؛ GLP-1 RA، منبه مستقبلات GLP-1.
الإشارات المتعلقة بمستقبلات GLP-1 ومؤشرات SSIB؛ وأخيرًا، قمنا بدراسة أنماط التزامن في استخدام مستقبلات GLP-1 والتفاعلات المحتملة بين الأدوية مع أدوية أخرى لاستكشاف تأثيرها على تكرار الإبلاغ عن مؤشرات SSIB.
في البداية، قمنا بمقارنة الخصائص السريرية للمرضى الذين تم علاجهم بمثبطات GLP-1RAs و SSIBs. أظهرت التقارير من 2018 إلى 2021 توزيعًا متوسطًا عامًا. ومع ذلك، كان هناك اتجاه تصاعدي في كل من ADR مع الحالة و
لم يتم ملاحظة أي حالة في عام 2022، مما قد يعكس الزيادة في وصف أدوية GLP-1RAs في السنوات الأخيرة. من بين البيانات المتاحة حول الأحداث السلبية، كانت النسبة أعلى بشكل ملحوظ من الإناث (64.22%) مقارنة بالذكور. تم الإبلاغ عن SSIBs. شكل البالغون الفئة الرئيسية المبلغة، حيث مثلوا الغالبية العظمى من جميع التقارير (الأحداث: البالغين: ، وكبار السن: ). الفجوات الجنسية المذكورة أعلاه هي
شكل. من SSIBs المرتبطة بـ GLP-1 RA في التحليل الأساسي وتحليل النتائج السلبية الكاذبة. عندما تم الكشف عن إشارة تقرير غير متناسبة. FAERS، نظام تقارير الأحداث السلبية التابع لإدارة الغذاء والدواء الأمريكية؛ IC، مكون المعلومات؛ SSIBs، السلوكيات الانتحارية والذاتية الإصابات؛ GLP-1 RA، منبه مستقبلات GLP-1
شكل. من SSIBs المرتبطة بـ GLP-1 RA في تحليل الحساسية. عندما تم الكشف عن إشارة تقرير غير متناسبة. FAERS، نظام تقارير الأحداث السلبية التابع لإدارة الغذاء والدواء الأمريكية؛ IC، مكون المعلومات؛ SSIBs، السلوكيات الانتحارية والذاتية الأذى؛ GLP-1 RA، منبه مستقبلات GLP-1
كما يتضح في التحقيقات الوبائية لمحاولات الانتحار والأفكار الانتحارية، التي أظهرت انتشارًا أعلى لسلوكيات إيذاء النفس بين
بين الإناث أكثر من الذكور. ومن الجدير بالذكر أن هذه الفجوة تكون ملحوظة بشكل خاص لدى الأفراد البدينين [40، 41]. فيما يتعلق بالعمر، يجب على الأطباء توخي الحذر بشكل أكبر
الجدول 2 التقييم العالمي من خلال معايير برادفورد هيل المعدلة
معايير وصف المصدر/الطريقة
قوة الارتباط على الرغم من أن مؤشرات التناسب ليست مقاييس للمخاطر، فإن قوة عدم التناسب سواء في التحليل الأساسي (مقارنة بجميع الأدوية الأخرى) أو في تحليل النتائج السلبية الكاذبة تشير إلى إشارة سلبية. تحليل عدم التناسب
تشبيه تم إثبات عدم الصلة أيضًا بالنسبة لأدوية مضادة للسكري/مضادة للسمنة الأخرى (إمباغليفلوزين وأورليستات)، التي تم استخدامها كتحكم سلبي في هذه الدراسة. تحليل عدم التناسب والتسميات
الاحتمالية البيولوجية / الأدلة التجريبية يُقترح أن يكون لمستقبلات GLP-1 تأثيرات إيجابية على الإدراك. خاصة من حيث الآليات العلاجية المزدوجة التي قد تحسن كل من العجز في الجهاز العصبي المركزي والعبء الأيضي. لا توجد أدلة تدعم أن مستقبلات GLP-1 ستسبب سلوكيات تناول الطعام غير الطبيعية. تحليل عدم التناسب والأدبيات
الاتساق كانت نتائج أساليب عدم التناسب متسقة في النتائج السلبية الكاذبة تحليل عدم التناسب
التماسك تقرير تجربة عشوائية محكومة يذكر ثلاث حالات مراهقين مرتبطة بأفكار/سلوكيات انتحارية باستخدام الليراجلوتيد. ومع ذلك، فإن المشارك الذي انتحر، والذي كان في مجموعة الليراجلوتيد، كان لديه تاريخ من اضطراب نقص الانتباه مع فرط النشاط وكان هناك محاولة انتحار واحدة في مجموعتي الليراجلوتيد والدواء الوهمي، على التوالي [37]. أبلغ تحليل استكشافي مجمع عن 34 من أصل 3291 أفكار انتحارية مع الليراجلوتيد. لكن لم يتم ملاحظة أي اختلالات بين العلاجات في أفكار/سلوكيات انتحارية أو اكتئاب من خلال تقييمات استبيانات مستقبلية [38]. يصف تقرير حالة حالتين من الاكتئاب المرتبط بالسيماجلوتيد [39]. بحث أدبي
خصوصية أظهرت نتائج التحليل الأولي وتحليل النتائج السلبية الكاذبة عدم وجود ارتباط بين أدوية GLP-1RAs وSSIBs. وأشارت تحليل الأدوية المشتركة إلى أن حدوث SSIBs كان أكثر احتمالاً أن يرتبط بالحالة النفسية للمريض نفسه. تحليل عدم التناسب والتداوي المشترك
علاقة زمنية تشير البيانات المتاحة إلى وجود حالات من SSIBs بعد التوقف عن استخدام GLP-1RA، ولكن لم يكن بالإمكان إجراء دراسات إضافية بسبب نقص البيانات. تحليل وقت البدء
الرجعية هذا المعيار ذو قيمة محدودة هنا حيث لا توجد بيانات كافية عن إعادة التحدي وإلغاء التحدي في قاعدة بيانات FAERS. وصفي
عند النظر في استخدام مثل هذه الأدوية في الفئات السكانية من الأطفال وكبار السن، نظرًا لملفات الأمان الخاصة بها. وقد أشارت الدراسات السابقة أيضًا إلى أن البالغين قد يظهرون قابلية أعلى للميل نحو الانتحار وإيذاء النفس مقارنة بالفئات العمرية الأخرى، مما قد يسهم في زيادة عدد الحالات المبلغ عنها بين البالغين. علاوة على ذلك، بين عامي 2018 و2022، تم الإبلاغ عن 91,431 حالة تتعلق بمثبطات GLP-1RAs، وكانت النسب المئوية للإناث والذكور وغير المعروف 53.77 و38.79 و على التوالي. كانت نسبة الأطفال والبالغين وكبار السن والأشخاص ذوي العمر غير المعروف ، و نعتقد أن الزيادة في التقارير من قبل الإناث والبالغين قد تكون مرتبطة أيضًا بتوزيع التقارير. من خلال تحليل هذه الأنماط والاتجاهات في البيانات، حصلنا على رؤى أولية حول العلاقة المحتملة بين GLP-1RAs و SSIBs.
بعد ذلك، استكشفنا العلاقة المحتملة بين GLP-1RAs وTTO لـ SSIBs. كان هدف تحليلنا هو تحديد ما إذا كانت هذه الأدوية قد تؤثر على تطور SSIBs من خلال بعض الآليات المشتركة. GLP-1RAs هي فئة من الأدوية المستخدمة لعلاج داء السكري من النوع 2 والتي يمكن أن تحاكي التأثير التنظيمي لـ GLP-1 على
مستويات الجلوكوز في الدم. على الرغم من أن هذه الأدوية من فئة GLP-1RAs تختلف في التركيب الجزيئي ومدة الفعالية، إلا أن آلياتها الدوائية متشابهة. لذلك، افترضنا أنه إذا كانت هناك ارتباطات ميكانيكية محتملة بين GLP-1RAs وSSIBs، فإن أنماطها المسببة للأمراض ستكون مشابهة. ومع ذلك، كشفت نتائجنا عن اكتشافات غير متوقعة. لاحظنا ثلاثة نماذج مسببة للأمراض مختلفة للأدوية في تحليل TTO. على سبيل المثال، تم تصنيف السيماغلوتيد والدولاجلوتيد كأنواع فشل مبكر، بينما تم تحديد الإكسيناتيد كنوع فشل عشوائي، وتم تصنيف الليراجلوتيد كنوع فشل نتيجة التآكل. وهذا يشير إلى أنه لا يوجد آلية واحدة (تأثير التحمل/تأثير التراكم) تفسر الارتباط الزمني لظهور GLP-1RAs وSSIBs. وبالتالي، يبدو أن تنوع أنماط الظهور التي لوحظت لـ GLP-1RAs يتحدى الافتراضات المذكورة أعلاه، مما يجعل من الصعب استنتاج علاقة واضحة بينها.
ثم قمنا بإجراء تحليل استكشافي لمثبطات GLP1RAs مقابل جميع الأدوية الأخرى في قاعدة بيانات FAERS باستخدام نفس المنهجية المستخدمة في دراسات اليقظة الدوائية السابقة [20،21]. تم استخدام ضوابط سلبية وإيجابية لتحديد الصلاحية الداخلية لـ
قاعدة البيانات. كانت الأورليستات والإمباغليفلوزين بمثابة ضوابط سلبية، بينما كانت الفينلافاكسين بمثابة ضابط إيجابي. كانت النتائج لهذه الأدوية الضابطة متسقة مع المعلومات المتاحة في الأدبيات والملصقات، مما يؤكد موثوقية دراستنا من حيث المنهجية وتحليل البيانات. أظهرت نتائج التحليل غير المتناسب أنه لم يظهر أي من GLP-1RAs إشارات مع SSIBs. تشير هذه النتائج السلبية إلى عدم وجود ارتباط مباشر بين GLP-1RAs وSSIBs. بالنظر إلى أننا استبعدنا ثلاث حالات تتعلق بالاستخدام المتزامن لعدة GLP-1RAs، قمنا أيضًا بإجراء تحليل للنتائج السلبية الكاذبة لتجنب النتائج السلبية الكاذبة. ومع ذلك، حتى مع زيادة في عدد التقارير، لم تحقق نتائج التحليل غير المتناسب بين جميع GLP-1RAs و SSIBs معيار IC. . بالنظر إلى تأثير التمويه للأحداث المعوية وحدوث SSIBs الناتجة عن الأمراض الأساسية، تم إجراء تحليلات حساسية. بعد تقييد الحالات وغير الحالات وفقًا لشروط الفحص، ظلت النتائج سلبية. بالإضافة إلى ذلك، أسفرت تحليلات المجموعات الفرعية المختلفة للأعمار و ADR أيضًا عن نتائج سلبية (الملف الإضافي 1: الشكل S5-6). الحساسية لاكتشاف هذا IC كانت قيمة GLP-1 RAs أكبر من 0.8، مما يشير إلى أن غياب الإشارة له دلالة. ومع ذلك، في بعض تحليلات المجموعات الفرعية، قد يُعزى غياب الإشارة (حساسية < 0.8) بالنسبة للألباجلوتيد والإكسيناتيد إلى العدد المحدود من الأدوية والأعراض الجانبية التي تم استخدامها في ظروف الفحص. وقد أكدت عملية التحقق من التحليل أعلاه عدم وجود دليل على وجود علاقة بين GLP-1RAs و SSIBs.
قمنا بتحليل إضافي لأعلى 20 دواء مصاحب تم الإبلاغ عنه بتكرار عالٍ في حالات السلوكيات الانتحارية المرتبطة بمستقبلات GLP-1RA لتحليل الأدوية المشتركة. تم تضمين العديد من الأدوية النفسية العصبية، بما في ذلك بوبروبيون وكواتيبين وأريبيبرازول، ضمن هذه الأدوية. عند التعمق أكثر في استخدام الأدوية النفسية العصبية في الحالات المشمولة في هذه الدراسة، وجدنا أن 81 فردًا يعانون من سلوكيات انتحارية مرتبطة بـ GLP-1RA كان لديهم تاريخ من استخدام الأدوية النفسية العصبية. أظهرت نتائج تحليل عدم التناسب الذي تم إجراؤه لهذه الأدوية والسلوكيات الانتحارية إشارة للإبلاغ، وقد اقترحت الدراسات أن الأفراد الذين يستخدمون هذه الأدوية قد يكونون في خطر أعلى من السلوكيات الانتحارية. تشير هذه النتائج إلى وجود ارتباط بين الأدوية النفسية العصبية والسلوكيات الانتحارية المرتبطة بـ GLP-1RA. الآثار الجانبية الأيضية السلبية المتزايدة (مثل السكري، وزيادة الوزن، واضطراب الدهون، وزيادة خطر الإصابة بأمراض القلب والأوعية الدموية) شائعة خلال علاج الأدوية النفسية العصبية. لذلك، فإن احتمال استخدام الأدوية النفسية العصبية مع GLP-1RAs مرتفع. قمنا أيضًا باستكشاف التفاعلات المحتملة بين GLP-1RAs والأدوية النفسية العصبية من خلال إجراء تحليل الانكماش. أظهرت النتائج أن أيًا من أدوية GLP-1RA لم تظهر تفاعلات محتملة مع الأدوية.
مع الأدوية النفسية العصبية. عند الجمع مع الأدوية النفسية العصبية، لم تزد GLP-1RAs من خطر تطوير SSIBs. علاوة على ذلك، فإن GLP-1RAs لها تأثيرات تعزز الإدراك وت activates مسار GLP-1R/cAMP/PKA، مما يوفر تدخلاً جديداً لعلاج الاكتئاب وتقليل SSIBs. توفر هذه الدراسات منظوراً بيولوجياً شاملاً ودقيقاً لفهم أعمق للعلاقة بين GLP-1RAs وSSIBs. وجهة نظرنا بدأت تميل نحو فكرة أن حدوث SSIBs من المرجح أن يكون مرتبطاً بالحالة العقلية للمريض بدلاً من GLP-1RAs.
استنادًا إلى النتائج والتقييمات المذكورة أعلاه للمعلومات البحثية المتاحة، اعتمدنا معايير برادفورد هيل المعدلة من المنشورات السابقة وقيمنا الرابط نقطة بنقطة بين أدوية GLP-1RA وSSIBs. بشكل عام، لا تلبي الأدلة الحالية تمامًا معايير برادفورد هيل المعدلة. وقد أفادت الدراسات بوجود حالات انتحار مرتبطة بالليراجلوتيد والاكتئاب المرتبط بالسيماغلوتيد في معيار التماسك [37-39]. ومع ذلك، خلص الباحثون إلى أن الليراجلوتيد لم يكن مرتبطًا بحدوث SSIBs. في التجربة السريرية العشوائية (RCT) التي أجراها كيلي وآخرون [37]، تم الإبلاغ عن حالة انتحار في مجموعة الليراجلوتيد كان لديها تاريخ من اضطراب نقص الانتباه مع فرط النشاط، وهو عامل محتمل من عوامل SSIBs. في الوقت نفسه، حاول مراهق الانتحار في كل مجموعة (مجموعة الدواء الوهمي ومجموعة الليراجلوتيد)، مما يبرز أن المراهقين هم فئة ضعيفة يجب مراقبتها بعناية. في تقرير آخر عن الاكتئاب المرتبط بالسيماغلوتيد [39]، كان لدى الفرد تاريخ طويل من الاكتئاب، مما يجعل من الصعب استبعاد تأثير تكرار المرض. علاوة على ذلك، في برنامج تأثير علاج السيماغلوتيد في الأشخاص الذين يعانون من السمنة [49-52]، لم يتم الإبلاغ عن أي SSIBs باستثناء بعض الاضطرابات النفسية الخفيفة أو المتوسطة (مثل الأرق، والقلق). من خلال تحليل شامل بأثر رجعي لقوة الارتباط، والتشابه، ومعايير التناسق بين أدوية GLP-1RA وSSIBs، خلصنا إلى أنه لا توجد أدلة تشير بشكل معقول إلى وجود ارتباط بين أدوية GLP-1RA وSSIBs. ومع ذلك، أكدت دراسات RCT، وتقارير الحالات، وقاعدة بيانات FAERS على أنه يجب أن نكون على دراية بالأدوية النفسية العصبية المساعدة التي قد تكون مرتبطة بـ SSIBs. يجب على الأطباء أن يدركوا أن المرضى الذين يتلقون أدوية متعددة يحتاجون إلى يقظة متزايدة، ومراقبة شاملة، واستشارة متخصصة.
لدراستنا بعض القيود. أولاً، كنظام للإبلاغ العفوي، يحتوي FAERS على عيوب جوهرية، مثل السجلات المكررة ذات جودة معلومات متغيرة. على الرغم من التصحيحات والحذف اليدوي، قد توجد فقط بعض الحالات المكررة [53،54]. ثانياً، بسبب التوافر المحدود لبيانات TTO، فإن عدد الحالات الصالحة
قد تكون غير كافية. وبالتالي، يجب اعتبار الاستنتاجات المستخلصة من تحليل TTO كأدلة ذات جودة منخفضة. ثالثًا، تحليل النتائج السلبية الكاذبة هو اختبار نقترحه بناءً على الوضع المحدد لهذه الدراسة. ومع ذلك، لم يتم استخدامه على نطاق واسع بعد في مجال اليقظة الدوائية. لذلك، فإن تطبيقه يتطلب اعتبارات دقيقة. رابعًا، كانت SSIBs المحددة في دراستنا هي PTs المضمنة للسلوكيات الانتحارية والسلوكيات الذاتية الجرح (على مستوى HLGT). نظرًا للاختلافات في قواعد تصنيف MedDRA، قد توجد بعض PTs المحتملة التي لم يتم تضمينها، مثل الاكتئاب الانتحاري (الذي ينتمي إلى مستوى HLGT، اضطرابات المزاج الاكتئابية والاضطرابات) والجرعة الزائدة المتعمدة (التي تنتمي إلى مستوى HLGT، الجرعات الزائدة والجرعات الناقصة NEC).
لقد أصبحت قاعدة بيانات FAERS الأكثر استخدامًا على نطاق واسع في السنوات الأخيرة. من خلال استخراج وتحليل قاعدة بيانات FAERS، تناولت هذه الدراسة المخاوف بشأن العلاقة المحتملة بين أدوية GLP-1RAs وحالات السلوك الانتحاري وساعدت في تبديد المخاوف العامة بشأن سلامة الأدوية. الغالبية العظمى من حالات السلوك الانتحاري المبلغ عنها في التجربة السريرية العشوائية الحالية والدراسة بعد التسويق لا تعتبر مرتبطة بأدوية GLP-1 RAs. التحليلات غير المتناسبة التي أجريت بالتزامن مع دراستنا، استنادًا إلى قاعدة بيانات FAERS مع فترات زمنية متفاوتة، تشير أيضًا إلى عدم وجود علاقة مباشرة بين أدوية GLP-1RAs وحالات السلوك الانتحاري. في تحليل استعادي باستخدام السجلات الصحية الإلكترونية، لاحظ الباحثون أن السيماغلوتيد يمكن أن يقلل حتى من خطر الانتحار مقارنة بأدوية مكافحة السمنة غير GLP-1 RAs. ومع ذلك، نظرًا لندرة حدوث حالات السلوك الانتحاري، وحجم التجربة السريرية العشوائية، وقيود الدراسات الاستعادية، هناك حاجة إلى تجارب مستقبلية أكبر وأعلى جودة لتحديد مصداقية الاستنتاجات الحالية.

الاستنتاجات

استكشفت هذه الدراسة العلاقة بين GLP-1RAs و SSIBs من خلال تحليل عدم التناسب. لم تجد هذه الدراسة أي دليل يشير بشكل معقول إلى وجود ارتباط بين GLP-1RA و SSIBs بناءً على الخصائص السريرية، و TTO، وعدم التناسب، وتحليل الأدوية المساعدة. بالمقارنة، يجب على الأطباء إيلاء المزيد من الاهتمام للحالة النفسية للمرضى الذين لديهم تاريخ من الأدوية النفسية، ويجب أن يكون هناك مراقبة أكثر شمولاً للنظر بعناية في قابليتهم لـ SSIBs. نتائج هذه الدراسة لها تأثير إيجابي على سلامة الأدوية للمرضى الذين يعانون من داء السكري من النوع 2 أو السمنة، وهو أمر مهم في الحفاظ على الصحة العامة وتقديم دعم قوي لقرارات الأدوية السريرية.
ردود الفعل السلبية للأدوية
سلوكيات انتحارية أو ذاتية الإصابة
الوكالة الأوروبية للأدوية
إدارة الغذاء والدواء
نظام الإبلاغ عن الأحداث السلبية التابع لإدارة الغذاء والدواء
مدرا – المعجم الطبي للأنشطة التنظيمية
وقت بدء TTO
فترة الثقة
مكون معلومات IC

معلومات إضافية

تحتوي النسخة الإلكترونية على مواد إضافية متاحة علىhttps://doi. org/10.1186/s12916-024-03274-6.
الملف الإضافي 1: الشكل S1. الخطوات الرئيسية في معالجة قاعدة بيانات FAERS. الشكل S2. ملخص سير العمل لحساب الحساسية. الشكل S3. التغيرات في قيم IC025 لكل GLP-1 RA بعد التوسع التدريجي في عدد الحالات. الشكل S4. معلومات عن 81 حالة لها تاريخ في استخدام الأدوية النفسية. الشكل S5. مقارنة IC025 لكل GLP-1 RA في مجموعات عمرية مختلفة من البالغين. الشكل S6. مقارنة IC025 للانتحار أو إيذاء النفس. الجدول S1. قائمة ردود الفعل السلبية لـ SSIBs المدرجة في هذه الدراسة. الجدول S2. جدول الطوارئ لاكتشاف الإشارة لتفاعل الأدوية. الجدول S3. حالات العلاج الأحادي والعلاج المتعدد مع مثبطات GLP-1. الجدول S4. اختبار أداء ملاءمة النماذج التوزيعية الأربعة المعلمية. الجدول S5. تحليل الوقت حتى ظهور SSIBs المرتبطة بمثبطات GLP-1 في قاعدة بيانات FAERS. الجدول S6. IC025 و من بين أفضل 23 دواء. الجدول S7. البيانات الخام لتحليل وقت بدء الأعراض في 52 حالة صالحة. الجدول S8. البيانات الخام من التحليلات الأولية، السلبية الكاذبة، الحساسية وتحليلات المجموعات الفرعية في حساب عدم التناسب.

شكر وتقدير

شكرًا لإدارة الغذاء والدواء الأمريكية على توفير مصدر مجاني للبيانات للدراسة.

العرض السابق

لم يتم نشر هذه المخطوطة أو تقديمها في مكان آخر جزئيًا أو كليًا، وهي ليست قيد النظر من قبل مجلة أخرى.

مساهمات المؤلفين

صمم X.M.W. و M.B.L. الدراسة. بحث J.X.Z. و Y.Z. و B.H.X. البيانات، وساهموا في المناقشة، وكتبوا المسودة الأولى من المخطوطة. قدم Y.F.Z. و S.J.L. كود التحليل. قام L.E.Z. و Y.H.L. بإجراء تحليل الأدوية المشتركة. قام C.L.L. بتنقيح لغة المخطوطة. X.M.W. هو الضامن لهذا العمل، وبالتالي كان لديه وصول كامل إلى جميع البيانات في الدراسة ويتحمل مسؤولية سلامة البيانات ودقة تحليل البيانات. قرأ جميع المؤلفين ووافقوا على المخطوطة النهائية.

حسابات تويتر للمؤلفين

@JX_ZHOU_FJMU

تمويل

لم يتم توفير أي تمويل لهذه الدراسة.

توفر البيانات والمواد

تتوفر مجموعات البيانات التي تم إنشاؤها و/أو تحليلها خلال الدراسة الحالية في قاعدة بيانات FAERS الأمريكية (https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.htmlالشفرة التي تم إنشاؤها و/أو تحليلها في الدراسة الحالية متاحة من المؤلف المراسل عند الطلب المعقول.

الإعلانات

لم يكن من الضروري الحصول على موافقة أخلاقية حيث تم إجراء الدراسة باستخدام بيانات عامة غير محددة الهوية.
غير قابل للتطبيق

الاختصارات

ناهضات مستقبلات الببتيد الشبيه بالجلوكاجون 1
الببتيد الشبيه بالجلوكاجون 1

المصالح المتنافسة

يعلن المؤلفون أنهم ليس لديهم مصالح متنافسة.

تفاصيل المؤلف

قسم الصيدلة، مستشفى اتحاد جامعة فوجيان الطبية، فوزهو، فوجيان، الصين. كلية الصيدلة، جامعة فوجيان الطبية، فوزهو، فوجيان، الصين. مدرسة التصوير الطبي، جامعة فوجيان الطبية، فوزهو، فوجيان، الصين. قسم الجهاز التنفسي، مستشفى شنغهاي للطاقة الكهربائية، شنغهاي، الصين. معهد شنغهاي للطب الحيوي، الأكاديمية الصينية للعلوم، شنغهاي، الصين.
تاريخ الاستلام: 18 أغسطس 2023 تاريخ القبول: 25 يناير 2024
نُشر على الإنترنت: 14 فبراير 2024

References

  1. Lingvay I, Leiter LA. Use of GLP-1 RAs in cardiovascular disease prevention: a practical guide. Circulation. 2018;137:2200-2.
  2. Patorno E, Htoo PT, Glynn RJ, Schneeweiss S, Wexler DJ, Pawar A, et al. Sodium-glucose cotransporter-2 inhibitors versus glucagon-like peptide-1 receptor agonists and the risk for cardiovascular outcomes in routine care patients with diabetes across categories of cardiovascular disease. Ann Intern Med. 2021;174:1528-41.
  3. Ma X, Liu Z, Ilyas I, Little PJ, Kamato D, Sahebka A, et al. GLP-1 receptor agonists (GLP-1RAs): cardiovascular actions and therapeutic potential. Int J Biol Sci. 2021;17:2050-68.
  4. Barritt AS, Marshman E, Noureddin M. Review article: role of glucagonlike peptide-1 receptor agonists in non-alcoholic steatohepatitis, obesity and diabetes-what hepatologists need to know. Aliment Pharmacol Ther. 2022;55:944-59.
  5. Davies MJ, Aroda VR, Collins BS, Gabbay RA, Green J, Maruthur NM, et al. Management of Hyperglycemia in Type 2 Diabetes, A consensus report by the american diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2022;45(11):2753-86.
  6. Jastreboff AM, Kushner RF. New frontiers in obesity treatment: GLP-1 and nascent nutrient-stimulated hormone-based therapeutics. Annu Rev Med. 2023;74:125-39.
  7. Bailey CJ, Flatt PR, Conlon JM. An update on peptide-based therapies for type 2 diabetes and obesity. Peptides. 2023;161:170939.
  8. O’Neil PM, Birkenfeld AL, McGowan B, Mosenzon O, Pedersen SD, Wharton S, et al. Efficacy and safety of semaglutide compared with liraglutide and placebo for weight loss in patients with obesity: a randomised, double-blind, placebo and active controlled, dose-ranging, phase 2 trial. Lancet. 2018;392:637-49.
  9. Chadda KR, Cheng TS, Ong KK. GLP-1 agonists for obesity and type 2 diabetes in children: systematic review and meta-analysis. Obes Rev. 2021;22:e13177.
  10. Tan B, Pan X-H, Chew HSJ, Goh RSJ, Lin C, Anand VV, et al. Efficacy and safety of tirzepatide for treatment of overweight or obesity. A systematic review and meta-analysis. Int J Obes (Lond). 2023;47:677-85.
  11. Pfizer Provides Update on GLP-1-RA Clinical Development Program for Adults with Obesity and Type 2 Diabetes Mellitus | Pfizer. https://www. pfizer.com/news/press-release/press-release-detail/pfizer-provides-update-glp-1-ra-clinical-development. Accessed 8 Aug 2023.
  12. Bezin J, Gouverneur A, Pénichon M, Mathieu C, Garrel R, Hillaire-Buys D, et al. GLP-1 receptor agonists and the risk of thyroid cancer. Diabetes Care. 2023;46:384-90.
  13. He L, Wang J, Ping F, Yang N, Huang J, Li Y, et al. Association of glucagonlike peptide-1 receptor agonist use with risk of gallbladder and biliary diseases: a systematic review and meta-analysis of randomized clinical trials. JAMA Intern Med. 2022;182:513-9.
  14. EMA. EMA statement on ongoing review of GLP-1 receptor agonists. European Medicines Agency. 2023. https://www.ema.europa.eu/en/ news/ema-statement-ongoing-review-glp-1-receptor-agonists. Accessed 20 Jul 2023.
  15. Fine KL , Rickert ‘Reilly LM , Sujan AC, Boersma K, Chang Z, et al. Initiation of opioid prescription and risk of suicidal behavior among youth and young adults. Pediatrics. 2022;149:e2020049750.
  16. Takeuchi T, Takenoshita S, Taka F, Nakao M, Nomura K. The relationship between psychotropic drug use and suicidal behavior in Japan: Japanese adverse drug event report. Pharmacopsychiatry. 2017;50:69-73.
  17. Gonda , Dome , Serafini . How to save a life: From neurobiological underpinnings to psychopharmacotherapies in the prevention of suicide. Pharmacol Ther. 2023;244:108390.
  18. Hughes JL, Horowitz LM, Ackerman JP, Adrian MC, Campo JV, Bridge JA. Suicide in young people: screening, risk assessment, and intervention. BMJ. 2023;381:e070630.
  19. Jing Y, Liu J, Ye Y, Pan L, Deng H, Wang Y, et al. Multi-omics prediction of immune-related adverse events during checkpoint immunotherapy. Nat Commun. 2020;11:4946.
  20. Raschi E, Fusaroli M, Giunchi V, Repaci A, Pelusi C, Mollica V, et al. adrenal insufficiency with anticancer tyrosine kinase inhibitors targeting vascular endothelial growth factor receptor: analysis of the FDA adverse event reporting system. Cancers (Basel). 2022;14:4610.
  21. Xia S, Gong H, Zhao Y, Guo L, Wang Y, Ma R, et al. Tumor lysis syndrome associated with monoclonal antibodies in patients with multiple myeloma: a pharmacovigilance study based on the FAERS database. Clin Pharmacol Ther. 2023;114:211-9.
  22. Zhou J, Wei Z, Xu B, Liu M, Xu R, Wu X. Pharmacovigilance of triazole antifungal agents: analysis of the FDA adverse event reporting system (FAERS) database. Front Pharmacol. 2022;13:1039867.
  23. Ren W, Wang W, Guo Y. Analysis of adverse reactions of aspirin in prophylaxis medication Based on FAERS database. Comput Math Methods Med. 2022;2022:7882277.
  24. Yu RJ, Krantz MS, Phillips EJ, Stone CA. Emerging causes of drug-induced anaphylaxis: a review of anaphylaxis-associated reports in the fda adverse event reporting system (FAERS). J Allergy Clin Immunol Pract. 2021;9:819829.e2.
  25. Hu Y, Bai Z, Tang Y, Liu R, Zhao B, Gong J, et al. Fournier gangrene associated with sodium-glucose cotransporter-2 inhibitors: a pharmacovigilance study with data from the U.S. FDA Adverse event reporting system. J Diabetes Res. 2020;2020:3695101.
  26. Liu L, Chen J, Wang L, Chen C, Chen L. Association between different GLP-1 receptor agonists and gastrointestinal adverse reactions: a realworld disproportionality study based on FDA adverse event reporting system database. Front Endocrinol (Lausanne). 2022;13:1043789.
  27. Hetrick SE, McKenzie JE, Bailey AP, Sharma V, Moller CI, Badcock PB, et al. New generation antidepressants for depression in children and adolescents: a network meta-analysis. Cochrane Database Syst Rev. 2021;5:CD013674.
  28. Sauzet O, Carvajal A, Escudero A, Molokhia M, Cornelius VR. Illustration of the weibull shape parameter signal detection tool using electronic healthcare record data. Drug Saf. 2013;36:995-1006.
  29. van Puijenbroek EP, Bate A, Leufkens HGM, Lindquist M, Orre R, Egberts ACG. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Pharmacoepidemiol Drug Saf. 2002;11:3-10.
  30. Norén GN, Hopstadius J, Bate A. Shrinkage observed-to-expected ratios for robust and transparent large-scale pattern discovery. Stat Methods Med Res. 2013;22:57-69.
  31. Trillenberg P, Sprenger A, Machner B. Sensitivity and specificity in signal detection with the reporting odds ratio and the information component. Pharmacoepidemiol Drug Saf. 2023;32:910-7.
  32. Noguchi Y, Tachi T, Teramachi H. Comparison of signal detection algorithms based on frequency statistical model for drug-drug interaction using spontaneous reporting systems. Pharm Res. 2020;37:86.
  33. Anderson N, Borlak J. Correlation versus causation? Pharmacovigilance of the analgesic flupirtine exemplifies the need for refined spontaneous ADR reporting. PLoS One. 2011;6:e25221.
  34. Fusaroli M, Raschi E, Giunchi V, Menchetti M, Rimondini Giorgini R, De Ponti F, et al. Impulse control disorders by dopamine partial agonists: a pharmacovigilance-pharmacodynamic assessment through the FDA adverse event reporting system. Int J Neuropsychopharmacol. 2022;25:727-36.
  35. Raschi E, Fusaroli M, Ardizzoni A, Poluzzi E, De Ponti F. Cyclin-dependent kinase 4/6 inhibitors and interstitial lung disease in the FDA adverse event reporting system: a pharmacovigilance assessment. Breast Cancer Res Treat. 2021;186:219-27.
  36. Horska K, Ruda-Kucerova J, Skrede S. GLP-1 agonists: superior for mind and body in antipsychotic-treated patients? Trends Endocrinol Metab. 2022;33:628-38.
  37. Kelly AS, Auerbach P, Barrientos-Perez M, Gies I, Hale PM, Marcus C, et al. A randomized, controlled trial of liraglutide for adolescents with obesity. N Engl J Med. 2020;382:2117-28.
  38. O’Neil PM, Aroda VR, Astrup A, Kushner R, Lau DCW, Wadden TA, et al. Neuropsychiatric safety with liraglutide 3.0 mg for weight management: Results from randomized controlled phase 2 and 3a trials. Diabetes Obes Metab. 2017;19:1529-36.
  39. Li J-R, Cao J, Wei J, Geng W. Case report: Semaglutide-associated depression: a report of two cases. Front Psychiatry. 2023;14:1238353.
  40. Klinitzke G, Steinig J, Blüher M, Kersting A, Wagner B. Obesity and suicide risk in adults-a systematic review. J Affect Disord. 2013;145:277-84.
  41. da Silva Bandeira BE, Dos Santos Júnior A, Dalgalarrondo P, de Azevedo RCS, Celeri EHVR. Nonsuicidal self-injury in undergraduate students: a cross-sectional study and association with suicidal behavior. Psychiatry Res. 2022;318:114917.
  42. Yan Y, Gong Y, Jiang M, Gao Y, Guo S, Huo J, et al. Utilization of glucagonlike peptide-1 receptor agonists in children and adolescents in China: a real-world study. Front Endocrinol (Lausanne). 2023;14:1170127.
  43. Dubé JP, Smith MM, Sherry SB, Hewitt PL, Stewart SH. Suicide behaviors during the COVID-19 pandemic: a meta-analysis of 54 studies. Psychiatry Res. 2021;301:113998.
  44. Yaribeygi H, Sathyapalan T, Sahebkar A. Molecular mechanisms by which GLP-1 RA and DPP-4i induce insulin sensitivity. Life Sci. 2019;234:116776.
  45. Jespersen MJ, Knop FK, Christensen M. GLP-1 agonists for type 2 diabetes: pharmacokinetic and toxicological considerations. Expert Opin Drug Metab Toxicol. 2013;9:17-29.
  46. Gliatto MF, Rai AK. Evaluation and treatment of patients with suicidal ideation. Am Fam Physician. 1999;59:1500-6.
  47. Battini V, Van Manen RP, Gringeri M, Mosini G, Guarnieri G, Bombelli A, et al. The potential antidepressant effect of antidiabetic agents: New insights from a pharmacovigilance study based on data from the reporting system databases FAERS and VigiBase. Front Pharmacol. 2023;14:1128387.
  48. Darwish AB, El Sayed NS, Salama AAA, Saad MA. Dulaglutide impedes depressive-like behavior persuaded by chronic social defeat stress model in male C57BL/6 mice: Implications on GLP-1R and cAMP/PKA signaling pathway in the hippocampus. Life Sci. 2023;320:121546.
  49. Rubino DM, Greenway FL, Khalid U, O’Neil PM, Rosenstock J, Sørrig R, et al. Effect of weekly subcutaneous semaglutide vs daily liraglutide on body weight in adults with overweight or obesity without diabetes: the STEP 8 randomized clinical trial. JAMA. 2022;327:138-50.
  50. Rubino D, Abrahamsson N, Davies M, Hesse D, Greenway FL, Jensen C, et al. Effect of continued weekly subcutaneous semaglutide vs placebo on weight loss maintenance in adults with overweight or obesity: the STEP 4 randomized clinical trial. JAMA. 2021;325:1414-25.
  51. Wadden TA, Bailey TS, Billings LK, Davies M, Frias JP, Koroleva A, et al. Effect of subcutaneous semaglutide vs placebo as an adjunct to intensive behavioral therapy on body weight in adults with overweight or obesity: the STEP 3 randomized clinical trial. JAMA. 2021;325:1403-13.
  52. Davies M, Færch L, Jeppesen OK, Pakseresht A, Pedersen SD, Perreault L, et al. Semaglutide 2.4 mg once a week in adults with overweight or obesity, and type 2 diabetes (STEP 2): a randomised, double-blind, doubledummy, placebo-controlled, phase 3 trial. Lancet. 2021;397:971-84.
  53. Sakaeda T, Tamon A, Kadoyama K, Okuno Y. Data mining of the public version of the FDA adverse event reporting system. Int J Med Sci. 2013;10:796-803.
  54. Khaleel MA, Khan AH, Ghadzi SMS, Adnan AS, Abdallah QM. A Standardized dataset of a spontaneous adverse event reporting system. Healthcare (Basel). 2022;10:420.
  55. McIntyre RS, Mansur RB, Rosenblat JD, Kwan ATH. The association between glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and suicidality: reports to the Food and Drug Administration Adverse Event Reporting System (FAERS). Expert Opin Drug Saf. 2023;1-9. Online ahead of print.
  56. Chen C, Zhou R, Fu F, Xiao J. Postmarket safety profile of suicide/ self-injury for GLP-1 receptor agonist: a real-world pharmacovigilance analysis. Eur Psychiatry. 2023;66:e99.
  57. Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Association of semaglutide with risk of suicidal ideation in a real-world cohort. Nat Med. 2024. https://doi.org/10.1038/s41591-023-02672-2.

ملاحظة الناشر

تظل شركة سبرينجر ناتشر محايدة فيما يتعلق بالمطالبات القضائية في الخرائط المنشورة والانتماءات المؤسسية.

  1. ماوباي ليو وشيوماي وو قدما مساهمة متساوية في هذا العمل ككاتبين متعاونين.
    *المراسلة:
    ماوباي ليو
    liumb0591@fjmu.edu.cn
    شيو مي وو
    wuxuemei@fjmu.edu.cn
    قائمة كاملة بمعلومات المؤلف متاحة في نهاية المقال

Journal: BMC Medicine, Volume: 22, Issue: 1
DOI: https://doi.org/10.1186/s12916-024-03274-6
PMID: https://pubmed.ncbi.nlm.nih.gov/38355513
Publication Date: 2024-02-14

Check for updates
Exploration of the potential association between GLP-1 receptor agonists and suicidal or self-injurious behaviors: a pharmacovigilance study based on the FDA Adverse Event Reporting System database

Jianxing Zhou , You Zheng , Baohua Xu , Songjun Long , Li-e Zhu , Yunhui Liu , Chengliang Li , Yifan Zhang , Maobai Liu and Xuemei Wu (D)

Abstract

Background Establishing whether there is a potential relationship between glucagon-like peptide 1 receptor agonists (GLP-1RAs) and suicidal or self-injurious behaviors (SSIBs) is crucial for public safety. This study investigated the potential association between GLP-1RAs and SSIBs by exploring the FDA Adverse Event Reporting System (FAERS) database.

Methods A disproportionality analysis was conducted using post-marketing data from the FAERS repository (2018 Q1 to 2022 Q4). SSIB cases associated with GLP-1RAs were identified and analyzed through disproportionality analysis using the information component. The parametric distribution with a goodness-of-fit test was employed to analyze the time-to-onset, and the shrinkage was used to evaluate the potential effect of co-medication on the occurrence of SSIBs.

Results In total, 204 cases of SSIBs associated with GLP-1RAs, including semaglutide, liraglutide, dulaglutide, exenatide, and albiglutide, were identified in the FAERS database. Time-of-onset analysis revealed no consistent mechanism for the latency of SSIBs in patients receiving GLP-1RAs. The disproportionality analysis did not indicate an association between GLP-1RAs and SSIBs. Co-medication analysis revealed 81 cases with antidepressants, antipsychotics, and benzodiazepines, which may be proxies of mental health comorbidities. Conclusions We found no signal of disproportionate reporting of an association between GLP-1RA use and SSIBs. Clinicians need to maintain heightened vigilance on patients premedicated with neuropsychotropic drugs. This contributes to the greater acceptance of GLP-1RAs in patients with type 2 diabetes mellitus or obesity.

Highlights

  • To determine whether there is a potential relationship between glucagon-like peptide 1 receptor agonists (GLP1RAs) and suicidal or self-injurious behaviors (SSIBs).
  • Is there a direct association between GLP-1RAs and SSIBs? No evidence reasonably suggests an association between GLP-1RAs and SSIBs based on clinical characteristics, time-onset, disproportionality, and co-medication analysis.
  • Clinicians should pay more attention to the psychiatric status of patients with a history of neuropsychotropic drugs, and more comprehensive monitoring is needed to consider their susceptibility to SSIBs carefully.
    Keywords GLP-1RAs, Suicidal or self-injurious behaviors, Pharmacovigilance, FAERS database, Type 2 diabetes, Obesity

Graphical Abstract

HIGHLIGHT: The current study found no signal of disproportionate reporting of an association between GLP-1RA use and SSIBs. Clinicians need to maintain heightened vigilance on patients premedicated with neuropsychotropic drugs.
A disproportionality pharmacovigilance study of post-marketing data in the FDA adverse event reporting system database
The utilization of GLP-1RAs resulted in a total of 91,431 reported cases spanning from 2018 to 2022
Drugs of interest included semaglutide, liraglutide, dulaglutide, exenatide, lixisenatide, and albiglutide
204 cases of suicidal or self-injurious behaviors
Exploration of the Potential Association between GLP-1 Receptor Agonists and Suicidal or Self-Injurious Behaviors: A Pharmacovigilance Study Based on the FDA Adverse Event Reporting System Database
Jianxing Zhou, You Zheng, Baohua Xu, Songjun Long, Li-e Zhu, Yunhui Liu, Chengliang Li, Yifan Zhang, Maobai Liu, Xuemei Wu*

Time to onset analysis

No single mechanism can explains the temporal association of onset between GLP-1RAs and SSIBs
Exploration of the Potential Association between GLP-1 Receptor Agonists and Suicidal or Self-Injurious Behaviors: A Pharmacovigilance Study Based on the FDA Adverse Event Reporting System Database
Jianxing Zhou, You Zheng, Baohua Xu, Songjun Long, Li-e Zhu, Yunhui Liu, Chengliang Li, Yifan Zhang, Maobai Liu, Xuemei Wu*

Descriptive analysis

All IC , no significant association between GLP-1RAs and SSIBs
Exploration of the Potential Association between GLP-1 Receptor Agonists and Suicidal or Self-Injurious Behaviors: A Pharmacovigilance Study Based on the FDA Adverse Event Reporting System Database
Jianxing Zhou, You Zheng, Baohua Xu, Songjun Long, Li-e Zhu, Yunhui Liu, Chengliang Li, Yifan Zhang, Maobai Liu, Xuemei Wu*

Co-medication analyses

SSIBs were more likely attributed to the patient’s mental state rather than caused by GLP-1RAs directly
Exploration of the Potential Association between GLP-1 Receptor Agonists and Suicidal or Self-Injurious Behaviors: A Pharmacovigilance Study Based on the FDA Adverse Event Reporting System Database
Jianxing Zhou, You Zheng, Baohua Xu, Songjun Long, Li-e Zhu, Yunhui Liu, Chengliang Li, Yifan Zhang, Maobai Liu, Xuemei Wu*
Causal link global assessment
Globally, Bradford Hill Criteria were unfulfilled

Background

Glucagon-like peptide 1 receptor agonists (GLP-1RAs) are modified derivatives of glucagon-like peptide 1 (GLP-1) that simulate the regulatory effect of GLP-1 on blood glucose levels. The primary advantages of GLP1RAs lie in glucose reduction, cardiorenal benefits, and amelioration of metabolic syndrome, all of which are substantiated by a plethora of evidence-based studies [1-4]. Therefore, GLP-1RAs have emerged as one of the recommended pharmacological interventions for type 2 diabetes mellitus, especially for patients with cardiovascular risk factors and coexisting comorbidities (e.g.,
chronic kidney disease) [5]. Furthermore, GLP-1RAs exhibit substantial weight reduction effects by facilitating glucose uptake and suppressing appetite [6]. According to the Centers for Disease Control and Prevention, diabetes and obesity rank among the top 10 most economically burdensome chronic diseases in the USA. Given the escalating global prevalence of these diseases, there is an unprecedented opportunity to expand the global market for GLP-1RAs.
Since the approval of exenatide for the treatment of type 2 diabetes by the Food and Drug Administration (FDA) in 2005, research on GLP-1 RAs has become a hot
topic and new analogs are constantly being introduced to the market. In December 2014, the FDA approved liraglutide as a weight loss drug. Right after this, semaglutide became available with the indication of diabetes in 2017, and then in 2020, its treatment for obesity was also authorized. Global demand for semaglutide and liraglutide exceeds the present supply, which might be related to their off label uses and misuse [7]. Despite the significant potential and advantages of GLP-1RAs, they are inevitably accompanied by a range of possible adverse drug reactions (ADRs), as is the case with any medication. The most prevalent ADRs associated with GLP1RA therapy encompass gastrointestinal events such as nausea, vomiting, and diarrhea. However, these nonserious ADRs are tolerable in individuals with diabetes and obesity [8-10]. What raises concerns are some of the recent reports of serious ADRs associated with GLP1RAs: (i) Pfizer’s disclosure of elevated aminotransferases in users of lotiglipron (a novel oral GLP-1RA) is indicative of hepatocellular injury [11]; (ii) Studies revealed an elevated risk of thyroid cancer associated and cholecystitis with the use of GLP-1RAs [12, 13]; (iii) The Icelandic Medicines Agency has submitted 150 reports of GLP-1RAs-related suicidal or self-injurious behaviors (SSIBs) to the European Medicines Agency (EMA) [14]. Following the aforementioned incidents, the development of lotiglipron was discontinued, and drug regulatory agencies in several countries classified GLP-1RAs as a potential risk factor for thyroid cancer, gallbladder, and biliary diseases. However, no studies have investigated the association between GLP-1RAs and SSIBs. SSIBs, recognized as significant ADRs, have garnered widespread acknowledgment as critical factors that may jeopardize patient safety [15]. Studies have demonstrated a consistent year-on-year escalation in patient suicides associated with medication use, intensifying concerns among the general public and clinical professionals regarding this phenomenon [16-18]. Therefore, in the context of the widespread global utilization of GLP-1RAs, it is imperative to investigate whether a potential relationship exists between GLP-1RAs and SSIBs, which is of paramount importance for public and clinical medication safety.
The FDA Adverse Event Reporting System (FAERS) is a comprehensive database of ADRs specifically designed by the FDA to facilitate postmarketing surveillance of drugs and therapeutic biologics [19]. The database includes all ADRs and medication errors documented by the FDA. It facilitates the identification and quantitative analysis of signals indicating disproportionate reporting of ADRs, thereby aiding in the recognition of associations between specific drugs and particular ADRs [20, 21].
This study aimed to explore the morbidity characteristics of SSIBs associated with GLP-1RAs by mining the
FAERS database and measuring the association between GLP-1RAs and SSIBs through metrics such as time of onset, disproportionality analysis, and co-medication analysis. The results of this study will have a positive impact on medication safety for patients with type 2 diabetes or obesity, which offering substantial backing for clinical medication choices.

Methods

Data sources and study design

The FAERS, one of the largest publicly accessible databases for ADRs, provides researchers with raw data from the FDA website (https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html). To process the extracted raw data from the FAERS database, we utilized Microsoft SQL Server (version 2019; Microsoft Corporation, Redmond, WA, USA). Considering duplicate or implausible reports can lead to significant errors [22], extensive cleaning, and normalization were conducted to ensure interpretable data (detailed process is provided in Additional file 1: Fig S1[23-25]).
The study year was selected on the basis of drug and data availability. At the commencement of this study, the FDA only published data for the first quarter of 2023. Tirzepatide (marketed in September 2022) was excluded due to insufficient data, and semaglutide was the latest drug of interest to be marketed (December 2017). Therefore, we performed a disproportionality analysis of postmarketing data from the FAERS repository to investigate the potential association of SSIBs with GLP-1RAs. Data were selected for each complete year since the latest drug was marketed from 2018 quarter 1 to 2022 quarter 4. Ethical approval was not required as the study was conducted using de-identified publicly available data.

Definition of cases and drugs of interest

In the FAERS database, ADRs are classified according to the Medical Dictionary for Regulatory Activities (MedDRA) terminology in terms of signs and symptoms, which are called preferred terms (PTs). The ADRs of interest include suicidal ideation, self-injurious ideation, suicidal behaviors, and self-injurious behaviors (see Additional file 1: Table S1 for a detailed list), which are homogenously attributed to a high-level group term (HLGT) named suicidal and self-injurious behavior by MedDRA (version 26.0). The drugs of interest included FDA-approved GLP-1RAs [26]: semaglutide, liraglutide, dulaglutide, exenatide, lixisenatide, and albiglutide. In this study, venlafaxine was used as a positive control because it is significantly associated with SSIBs [27]. Orlistat and empagliflozin were selected as negative controls based on their similarity in the population indicated for GLP-1RAs, without any reported evidence of self-harm
or suicide potential in the specification or related studies. The current study encompassed all cases in which GLP1RAs were administered and listed as primary suspect drugs, secondary suspect drugs, or concomitant drugs for SSIBs.

Descriptive analysis

Patient demographics (age, sex, and reporter type) and clinical aspects (latency, outcome, and indication) were documented in reports of SSIBs with semaglutide, liraglutide, dulaglutide, exenatide, lixisenatide, and albiglutide.

Reported timeto-onset analysis

To describe the latency of SSIBs with GLP-1RAs, we performed a time-to-onset (TTO) analysis by selecting the parametric distribution model that demonstrated the best result in the goodness-of-fit test among the Weibull, log-normal, gamma, and exponential distributions. The parametric distribution is characterized by the scale parameter and shape parameter [28]. The scale parameter represents the magnitude of ADRs occurring at the quantile within the distribution function. Depending on the value of the shape parameter , the upper limit of the confidence interval (CI) for indicates an initial increase, followed by a decrease in the hazard rate (early-failure type). Conversely, the lower limit of the CI for demonstrates an increasing hazard over time (wear-out failure type), whereas the CI includes 1 for , suggesting a constant hazard rate throughout the exposure period (random failure type). More details about TTO analysis and classification of can be found in the Additional file 1, sections 4-5.

Disproportionality analysis

A sequential approach was adopted to systematically address major confounders:
(1) An exploratory disproportionality approach comparing GLP-1RAs with all other drugs reported in the FAERS database, using venlafaxine as the positive control and empagliflozin/orlistat as the negative control. The definition of the disproportionality approach can be found in Additional file 1: section 4 [29].
(2) We utilized the Bayesian Information Component (IC) to calculate the lower limit of the CI ( ). This approach is more precise than relying on the report odds ratio, particularly in situations with limited cases [30]. The IC calculation procedure can be found in Additional file 1: section 5.
(3) False-negative analysis was conducted by artificially augmenting the reported incidence of GLP-1RA-
associated SSIBs by , aiming to ascertain whether the absence of positive findings could be attributed to our exclusion of some cases involving the concurrent use of multiple GLP-1RAs. This meticulous examination effectively substantiates the credibility of the negative results.
(4) Given the inherent heterogeneity and potential reporting bias in the FAERS database, we conducted a series of sensitivity analyses. First, we excluded case and non-case reports with gastrointestinal events (all PTs belonged to gastrointestinal disorders in the system organ class) in the dataset. This step aimed to reduce the masking effect of gastrointestinal events to avoid possible competitive bias. Furthermore, the main indications in the included cases were type 2 diabetes mellitus and obesity, which are potential risk factors for SSIBs. Therefore, we narrowed the analysis dataset to subject with type 2 diabetes mellitus or weight loss through the indi_pt field (which represents the indication). Previous studies have shown that restricting the analyzed population can mitigate indication bias [21]. IC was calculated using the processed dataset to conduct the above sensitivity analyses.
(5) Considering a low number of expected cases could lead to an insufficient sensitivity to detect disproportionality of relevant strength [31], we reported the sensitivity of a representative to measure the reliability of negative results. The result was deemed confident when the sensitivity to detect representative . The calculation procedure can be found in Additional file 1: Fig S2.

Co-medication analysis

A case-by-case analysis was performed by two independent pharmacists to identify the combined medications by searching DRUG and THER files of the included 204 cases. When discrepancies existed, a third pharmacist was introduced to make the final decision. We conducted a co-medication analysis of GLP-1RAs and the top 20 most frequently used drugs in cases of SSIBs associated with GLP-1RAs. Calculation of for these drugs is consistent with the procedure described in section 2.5. We used the shrinkage to measure drug-drug interactions because a previous study showed that it is the most conservative method among multiple algorithms [32]. The detection criterion is the lower limit of the CI of the . The calculation process of and the list of drugs were described in Additional file 1: Table S2. When at least one neuropsychiatric drug (such as antipsychotics, antidepressants, and anxiolytics) was recorded in the report, the patient was defined to have a medication history.

Global assessment of the evidence

The Modified Bradford Hill Criteria was used to assess the potential relationships among various available evidence [33-35]. The biological plausibility can assess whether there is a mechanism to support the potential relationship between ADRs and the studied drug. We also considered the strength of the evidence (the magnitude of observed effects) and the consistency of findings across various studies or sources. Additionally, specificity was utilized to assess whether some events were associated with specific factors, and coherence was to examine the temporal sequence of events that aligned with a potential relationship. Lastly, other similar drugs (empagliflozin and Orlistat) were used as a negative control to support the final conclusion.

Statistical analysis

The disproportionality threshold of the reporting signal was defined as , and a positive drug-drug interaction was defined as . R (version 4.3.1; R Foundation for Statistical Computing, Vienna, Austria) was used for plotting, TTO analysis, and calculation of IC and . The magnified percentage used in the false-negative analysis was , determined based on a step-by-step strategy (Additional file 1: Fig S3, and Table S3).

Results

Descriptive analysis

From 2018 quarter 1 to 2022 quarter 4 of 2018, 57, 74, 51, 20, and 2 reports of SSIBs associated with semaglutide, liraglutide, dulaglutide, exenatide, and albiglutide, respectively, were identified in the FAERS database. Three patients were excluded because of the concomitant use of multiple GLP-1RAs, and no SSIB reports for lixisenatide were detected. The distribution of the reported years exhibited a generally even pattern, with a noticeable upward trend observed in 2022. The majority (54.90%) of the reports on SSIBs utilizing GLP-1RAs were contributed by healthcare professionals. Despite the small amount of missing or unknown data ( ), the proportion of reports in females was significantly higher than that in males ( 64.22 vs. , respectively). Among the reports with available data, half of the cases were observed in adults (18-65 years). Type 2 diabetes ( ) and weight loss ( ) emerged as the most predominant indications, while others accounted for , and unknown or missing constituted (Table 1).

Reported time-to-onset analysis

A total valid case of 52 was used for TTO analysis (Fig. 1). Among them, three cases reported SSIBs on 1 day, 1 day, and 3 days after the cessation of GLP-1RAs ( 1 case for
liraglutide and 2 cases for semaglutide), which might be related to the delayed ADR caused by the interested drugs.
In analysis of TTO based on parameter distributions and the valid cases, the median TTOs (Interquartile range) for SSIBs associated with semaglutide, liraglutide, dulaglutide, and exenatide were , , and 2.5 (1.5-169.5, ) days, respectively (albiglutide was not included in the TTO analysis due to insufficient data). The results of goodness-of-fit performance test (Additional file 1: Table S4) indicated that a log-normal model was the best one to describe the latency of liraglutideand exenatide-related SSIBs. A Gamma model was suitable for semaglutide-related SSIBs, and a Weibull model described the latency of dulaglutide-related SSIBs well. Semaglutide and dulaglutide were classified as early failure types, exenatide as a random failure type, and liraglutide as a wear-out failure type (Additional file 1: Table S5 for raw data).

Disproportionality analysis

First, primary analysis was conducted to compare GLP1RAs with other drugs in the FAERS database. The evaluated GLP-1RAs were semaglutide ( ), liraglutide ( ), dulaglutide ( ), exenatide ( ), and albiglutide ( 3.91), none of which exhibited a signal of disproportionate reporting. Negative (empagliflozin, ; orlistat, ) and positive (venlafaxine, ) controls confirmed the internal validity of the database. We then performed a false-negative analysis in which the results of the disproportionate analysis of all GLP-1 drugs with SSIBs did not satisfy after expanding the number of reports by , and the negative and positive control results remained valid (Fig. 2). Finally, was also observed in the sensitivity analyses (Fig. 3). The sensitivity to detect this was for all GLP-1 RAs in primary and false-negative analyses. In some subgroup analyses, there are some sensitivities to detect this was < 0.8 for albiglutide and exenatide. The raw data for the above analysis can be found in the Additional file 1: Table S8.

Co-medication analysis

We conducted a co-medication analysis of the top 20 medications with the highest reported frequency of use for all GLP-1RA-associated SSIBs (considering the same reported cases, 23 medications were included). Multiple neuropsychiatric drugs, including bupropion, quetiapine, and aripiprazole, were identified, and disproportionality analysis with SSIBs yielded a reporting signal. Further examination revealed that 81 patients had a history of
Table 1 Cases characteristics of SSIBs associated with GLP-1RA in the FAERS database
Categories Semaglutide (%) Liraglutide (%) Dulaglutide (%) Exenatide (%) Albiglutide (%) Total (%)
Reports of SSIBs 57 74 51 20 2 204
Report year
2018 2 (3.51) 16 (21.62) 12 (23.53) 4 (20.00) 1 (50.00) 35 (17.16)
2019 4 (7.02) 11 (14.86) 7 (13.73) 7 (35.00) 0 29 (14.22)
2020 8 (14.04) 19 (25.68) 10 (19.61) 4 (20.00) 1 (50.00) 42 (20.59)
2021 16 (28.07) 12 (16.22) 8 (15.69) 2 (10.00) 0 38 (18.63)
2022 27 (47.37) 16 (21.62) 14 (27.45) 3 (15.00) 0 60 (29.41)
Reporter
Healthcare professional 30 (52.63) 53 (71.62) 19 (37.25) 9 (45.00) 1 (50.00) 112 (54.90)
Nonhealthcare professional 26 (45.61) 21 (28.38) 31 (60.78) 6 (30.00) 1 (50.00) 85 (41.67)
Unknown or missing 1 (1.75) 0 1 (1.96) 5 (25.00) 0 7 (3.43)
Sex
Male 18 (31.58) 20 (27.03) 19 (37.25) 9 (45.00) 1 (50.00) 67 (32.84)
Female 38 (66.67) 52 (70.27) 29 (56.86) 11 (55.00) 1 (50.00) 131 (64.22)
Unknown or missing 1 (1.75) 2 (2.70) 3 (5.88) 0 0 6 (2.94)
Age category, years
Juvenile (< 18) 0 0 0 3 (15.00) 0 3 (1.47)
Adult (18-65) 28 (49.12) 43 (58.11) 24 (47.06) 7 (35.00) 0 102 (50.00)
Seniors (>65) 1 (1.75) 7 (9.46) 10 (19.61) 2 (10.00) 0 20 (9.80)
Unknown or missing 28 (49.12) 24 (32.43) 17 (33.33) 8 (40.00) 2 (100.00) 79 (38.73)
Median (IQR) 43 (35-52) 47 (36-57) 55 (46-67) 40 (25-54) / /
Indication
Type 2 diabetes mellitus 13 (22.81) 14 (18.92) 30 (58.82) 13 (65.00) 1 (50.00) 71 (34.80)
Weight loss 15 (26.32) 15 (20.27) 0 0 0 30 (14.71)
Others 1 (1.75) 4 (5.41) 0 3 (15.00) 0 8 (3.92)
Unknown or missing 28 (49.12) 41 (55.41) 21 (41.18) 4 (20.00) 1 (50.00) 95 (46.57)
Outcomes
Death 1 (1.75) 17 (22.97) 3 (5.88) 1 (5.00) 0 22 (10.78)
Life-threatening 6 (10.53) 6 (8.11) 3 (5.88) 5 (25.00) 0 20 (9.80)
Hospitalization 8 (14.04) 20 (27.03) 19 (37.25) 7 (35.00) 0 54 (26.47)
Disability 2 (3.51) 1 (1.35) 2 (3.92) 1 (5.00) 0 6 (2.94)
Other serious illness 52 (91.23) 55 (74.32) 37 (72.55) 12 (60.00) 2 (100) 158 (77.45)
IQR interquartile range, SSIBs Suicidal and Self-Injurious Behaviors, GLP-1RAs GLP-1 receptor agonists
Healthcare professionals including reporters such as physicians and pharmacists; nonhealthcare professionals including reporters such as consumer and lawyer
Since a case may experience different clinical outcomes during drug therapy, it is reasonable that the sum percentage of the outcome under this item may exceed 100%
neuropsychiatric drugs, and 63 patients used more than one neuropsychiatric drug (Additional file 1: Fig S4). Subsequent shrinkage results indicated no potential for drug-drug interactions between all GLP-1RAs and neuropsychiatric drugs (all ; refer to Additional file 1: Table S6 for the list of top 20 medications’ , and ).

Causal relationship global assessment

Globally, the Bradford Hill Criteria were not fulfilled, as indicated by the strength of the disproportionality and its consistency throughout the analysis, temporal
relationships, and biological plausibility, thus unsupporting a likely causal association between GLP-1RAs (semaglutide, liraglutide, dulaglutide, exenatide, lixisenatide, and albiglutide) and SSIBs (Table 2).

Discussion

The current study conducted a comprehensive analysis comprising four key objectives: first, we characterized the clinical features of SSIB cases related to GLP-1RA by analyzing postmarketing data from the FAERS database; second, we developed a latency model for these cases; third, we concluded the lack of disproportional reporting
Fig. 1 TTO analysis of SSIBs reported for each GLP-1 RA. Reported TTO analysis and duration of treatment of SSIB associated with a liraglutide, b semaglutide, c dulaglutide, and d exenatide. Black diamonds represent the TTO of SSIBs following the administration of GLP-1RAs, while gray bars indicate the duration of GLP-1 treatment for each case. Due to limitations in data availability, the duration of therapy could only be plotted for a subset of the cases. Goodness-of-fit test of SSIB associated with liraglutide, semaglutide, dulaglutide, and exenatide. Positioned above the plot is the quantile boxplot. The green bars correspond to the case number of occurrences within the distribution, while the green line represents the fitted curve for the model exhibiting the most optimal outcomes. we adopted log-normal distribution to describe the latency of liraglutide and exenatide, gamma distribution for semaglutide, and Weibull distribution for dulaglutide. The raw data of TTO analysis in the 52 valid cases can be found in Additional file 1: Table S7. TTO, time-to-onset; SSIBs, Suicidal and Self-Injurious Behaviors; GLP-1 RA, GLP-1 receptor agonist
signals regarding GLP-1RAs and SSIBs; finally, we examined co-administration patterns of GLP-1RAs and potential drug-drug interactions with other medications to explore their impact on the reporting frequency of SSIBs.
Initially, we compared the clinical characteristics of patients treated with GLP-1RAs and SSIBs. Reports from 2018 to 2021 exhibited an overall average distribution. However, an uptrend in both ADR with case and
non-case was observed in 2022, possibly reflecting the increased prescription of GLP-1RAs in recent years. Among the available data on adverse events, a significantly higher proportion of females (64.22%) than males reported SSIBs. Adults constituted the primary reporting population, accounting for the majority of all reports (juvenile: , adults: , and the elderly: ). The aforementioned sex disparities are
Fig. of SSIBs associated with GLP-1 RA in the primary analysis and false negative analysis. When , a disproportionate reporting signal was detected. FAERS, US Food and Drug Administration Adverse Event Reporting System; IC, information component; SSIBs, Suicidal and Self-Injurious Behaviors; GLP-1 RA, GLP-1 receptor agonist
Fig. of SSIBs associated with GLP-1 RA in the sensitivity analysis. When , a disproportionate reporting signal was detected. FAERS, US Food and Drug Administration Adverse Event Reporting System; IC, information component; SSIBs, Suicidal and Self-Injurious Behaviors; GLP-1 RA, GLP-1 receptor agonist
also evident in epidemiological investigations of suicide attempts and suicidal ideation, which have demonstrated a higher prevalence of self-harming behaviors among
females than among males. Notably, this disparity is particularly pronounced in obese individuals [40, 41]. With regard to age, physicians should exert heightened caution
Table 2 Global assessment through adapted Bradford Hill Criteria
Criteria Description Source/method
Strength of the association Although ICs are not measures of risk, the strength of the disproportionality both in primary (vs. all other drugs) and false negative analysis suggests a negative signal Disproportionality analysis
Analogy The irrelevance was also demonstrated for other anti-diabetic/anti-obesity drugs (empagliflozin and Orlistat), which were used as a negative control in this study Disproportionality analysis and labels
Biological plausibility/ empirical evidence GLP-1RAs are also proposed to have pro-cognitive effects. Particularly in terms of dual therapeutic mechanisms potentially improving both central nervous system deficits and metabolic burden [36]. There is no evidence to support that GLP-1RA will cause SSIBs Disproportionality analysis and literature
Consistency Results of disproportionality approaches were consistent in false negative Disproportionality analysis
Coherence A randomized, controlled trial reports three adolescent cases associated with suicidal ideation/behavior using liraglutide. However, the participant who committed suicide, who was in the liraglutide group, had a history of attention deficit-hyperactivity disorder and there was one suicide attempt in the liraglutide and placebo groups, respectively [37]. An exploratory pooled analysis reported 34/3291 suicidal ideation with liraglutide. But no between-treatment imbalances in suicidal ideation/behavior or depression were noted through prospective questionnaire assessments [38]. A case report describes two instances of depression associated with semaglutide [39]. literature search
Specificity The results of primary and false-negative analysis showed no association between GLP-1RAs and SSIBs. The co-medication analysis indicated that the occurrence of SSIBs was more likely to be related to the patient’s own mental state. Disproportionality and co-medication analysis
Temporal relationship Available data suggested that there were cases of SSIBs after discontinuation of GLP-1RA, but further studies could not be performed due to missing data Time-to-onset analysis
Reversibility This criterion is of limited value here as there is not enough data on rechallenge and de-challenge in the FAERS database Descriptive
when considering the use of such medications in pediatric and geriatric populations, given their safety profiles [42]. Previous studies have also indicated that adults may exhibit a higher susceptibility to suicidal and self-harming tendencies than other age groups [43], potentially contributing to a higher number of reported cases among adults. Furthermore, between 2018 and 2022, 91,431 cases involving GLP-1RAs were reported, with the proportions of females, males, and unknown being 53.77, 38.79 , and , respectively. The proportion of children, adults, elderly individuals, and those of unknown age were , and , respectively. We believe that the higher reporting by females and adults may also be related to the distribution of reports. By analyzing these data patterns and trends, we gained preliminary insights into the potential association between GLP-1RAs and SSIBs.
Next, we delved into the potential association between GLP-1RAs and TTO of SSIBs. Our analysis aimed to determine whether these drugs might influence the development of SSIBs by some common mechanism. GLP-1RAs are a class of drugs used to treat type 2 diabetes that can simulate the regulatory effect of GLP-1 on
blood glucose levels. Although these GLP-1RAs differ in molecular structure and duration of action, their pharmacological mechanisms are similar [44, 45]. Therefore, we hypothesized that if GLP-1RAs had potential mechanistic associations with SSIBs, their pathogenic patterns would be similar. However, our results revealed unexpected findings. We observed three different pathogenic models for the drugs in the TTO analysis. For example, semaglutide and dulaglutide were categorized as earlyfailure types, exenatide was determined to be a random failure type, and liraglutide was classified as a wear-out failure type. This suggests that no single mechanism (tolerance effect/accumulation effect) explains the temporal association of onset between GLP-1RAs and SSIBs. Thus, the diversity of the onset patterns observed for GLP1RAs appears to challenge the above assumptions, making it difficult to conclude a clear relationship between them.
We then performed an exploratory analysis of GLP1RAs against all other drugs in the FAERS database using the same methodology used in previous pharmacovigilance studies [20,21]. Negative and positive controls were used to determine the internal validity of the
database. Orlistat and empagliflozin served as negative controls, and venlafaxine served as a positive control. The results for these control drugs were consistent with the information available in the literature and labels, confirming the reliability of our study in terms of methodology and data analysis. The results of the disproportionate analysis showed that none of the GLP-1RAs showed signaling with SSIBs. These negative results imply no direct association between GLP-1RAs and SSIBs. Considering that we discarded three cases involving the concurrent use of multiple GLP-1RAs, we further performed a falsenegative analysis to avoid false-negative results. However, even with a increase in the number of reports, the results of the disproportionate analysis between all GLP-1RAs and SSIBs did not meet the criterion of IC . Considering the masking effect of gastrointestinal events and the occurrence of SSIBs caused by underlying diseases, sensitivity analyses were performed. After restricting case and non-case according to the screening conditions, the results remained negative. In addition, further subgroup analyses for different ages and ADR also yielded negative results (Additional file 1: Fig S5-6). The sensitivity to detect this IC of GLP-1 RAs was > 0.8, suggesting that the absence of a signal is meaningful. However, in some subgroup analyses, the absence of a signal (sensitivity < 0.8) for albiglutide and exenatide is possibly attributed to the reduced number of drugs and ADRs for screening conditions. Validation of the above analysis further confirmed the lack of evidence of an association between GLP-1RAs and SSIBs.
We further analyzed the top 20 concomitant medications with the highest frequency reported in SSIBs associated with GLP-1RAs for co-medication analysis. Multiple neuropsychiatric drugs, including bupropion, quetiapine, and aripiprazole, were included among these medications. Further delving into the use of neuropsychiatric drugs in the cases included in this study, we found that 81 individuals with GLP-1RA-related SSIBs had a history of neuropsychiatric drug use. The results of disproportionality analysis performed for these medications and SSIBs showed a reporting signal, and studies have suggested that individuals who use these medications may be at a higher risk of SSIBs [46]. These results indicate an association between neuropsychiatric drugs and GLP-1RA-related SSIBs. Increased adverse metabolic effects (diabetes, weight gain, dyslipidemia, and increased cardiovascular risk) are common during neuropsychiatric drug treatment [36]. Therefore, the potential comedication of neuropsychiatric drugs and GLP-1RAs is high. We further explored potential drug interactions between GLP-1RAs and neuropsychiatric drugs by performing an shrinkage analysis. The results showed that none of the GLP-1RAs showed possible drug interactions
with neuropsychiatric drugs. In combination with neuropsychiatric drugs, GLP-1RAs did not increase the risk of developing SSIBs. Moreover, GLP-1RAs have cogni-tion-promoting effects and activate the GLP-1R/cAMP/ PKA pathway, thus providing a new intervention for the treatment of depression and reduction of SSIBs [36, 47, 48]. These studies provide a comprehensive and accurate biological perspective for a deeper understanding of the association between GLP-1RAs and SSIBs. Our view is beginning to lean toward the idea that the occurrence of SSIBs is more likely to be related to the mental state of the patient rather than GLP-1RAs.
Based on the above results and assessment of available research information, we adopted the adjusted Bradford Hill Criteria from previous publications and assessed the point-by-point link between GLP-1RAs and SSIBs. In general, the current evidence does not fully meet the adjusted Bradford Hill Criteria. Studies have reported liraglutide-associated suicide and semaglutide-associated depression in the coherence criterion [37-39]. However, the researchers concluded that liraglutide was not associated with the occurrence of SSIBs. In the randomized controlled trial (RCT) by Kelly et al. [37], a suicide case in the liraglutide group was reported to have a history of attention deficit hyperactivity disorder, which is a potential factor of SSIBs. Meanwhile, one adolescent attempted suicide in each group (the placebo group and the liraglutide group), highlighting that adolescents are a vulnerable population that should be monitored carefully. In another report of semaglutide-associated depression [39], the individual had a long history of depression, making it challenging to rule out the interference of disease recurrence. Furthermore, in the program of Semaglutide Treatment Effect in People with Obesity [49-52], no SSIBs were reported except for some mild or moderate psychiatric disorders(e.g., insomnia, anxiety) were observed. Through a comprehensive retrospective analysis of the strength of the association, analogy, and consistency metrics of GLP-1RAs and SSIBs, we concluded that there was no evidence reasonably suggesting an association between GLP-1RAs and SSIBs. However, the RCT studies, case reports, and the FAERS database case all highlighted that we should be aware of co-administered neuropsychotropic drugs that may associated with SSIBs. Clinicians should recognize that patients undergoing polypharmacy require heightened vigilance, comprehensive monitoring, and specialized counseling.
Our study has some limitations. First, as a spontaneous reporting system, the FAERS has inherent shortcomings, such as duplicate records with variable information quality. Despite manual corrections and deletions, only a few duplicate cases may exist [53,54]. Second, due to the limited availability of TTO data, the number of valid cases
may be inadequate. Consequently, the conclusions drawn from the TTO analysis should be regarded as low-quality evidence. Third, false-negative analysis is a test that we propose based on the specific situation of this study. However, it is not yet widely used in the field of pharmacovigilance. Therefore, its application warrants careful consideration. Fourth, the SSIBs defined in our study were PTs included for Suicidal and self-injurious behaviors (HLGT level). Given the differences in the MedDRA’s categorization rules, some potential PTs that were not included may exist, e.g., depression suicidal (belonging to HLGT level, depressed mood disorders and disturbances) and intentional overdose (belonging to HLGT level, overdoses, and underdoses NEC).
The FAERS database has become the most widely used ADR database in recent years. By mining and analyzing the FAERS database, this study addressed concerns about the potential association between GLP-1RAs and SSIBs and helped dispel public concerns regarding drug safety. The majority of SSIB cases reported in the current RCT and post-marketing study are not considered to be related to GLP-1 RAs. The disproportionate analyses conducted concurrently with our study, based on the FAERS database with varying year spans, also indicate no direct association between GLP-1RAs and SSIBs [55, 56]. In a retrospective analysis utilizing electronic health records, researchers noted that semaglutide can even reduce the risk of suicide compared to non-GLP-1 RAs anti-obesity medications [57]. However, given the rarity of the occurrence of SSIBs, the size of the RCT, and the limitations of the retrospective studies, higher-quality and larger prospective trials are needed to determine the credibility of the current conclusions.

Conclusions

This study explored the relationship between GLP-1RAs and SSIBs through disproportionality analysis. This study found no evidence reasonably suggesting an association between GLP-1RA and SSIBs based on clinical characteristics, TTO, disproportionality, and co-medication analysis. Comparatively, clinicians should pay more attention to the psychiatric status of patients with a history of neuropsychotropic drugs, and more comprehensive monitoring is needed to consider their susceptibility to SSIBs carefully. The results of this study have a positive impact on medication safety for patients with type 2 diabetes or obesity, which is important in maintaining public health and providing strong support for clinical medication decisions.
ADRs Adverse drug reactions
SSIBs Suicidal or self-injurious behaviors
EMA European Medicines Agency
FDA The Food and Drug Administration
FAERS The Food and Drug Administration Adverse Event Reporting System
MedDRA The Medical Dictionary for Regulatory Activities
TTO Time-to-onset
Cl Confidence interval
IC Information Component

Supplementary Information

The online version contains supplementary material available at https://doi. org/10.1186/s12916-024-03274-6.
Additional file 1: Fig S1. The main steps in the processing of the FAERS database. Fig S2. Summary of the workflow for calculation of the sensitivity. Fig S3. Changes in IC025 values for each GLP-1 RA after gradual expansion of case number. Fig S4. Information on 81 cases with a history of psychotic drug use. Fig S5. Comparison of IC025 for Each GLP-1 RA in Different Age Groups of Adults. Fig S6. Comparison of IC025 for suicide or self-injury. Table S1. List of adverse reactions of SSIBs included in this study. Table S2. The contingency table for signal detection of drug-drug interaction. Table S3. Cases of monotherapy and polytherapy with GLP-1 RAs. Table S4. The performance test of goodness-of-fit among four parametric distribution models. Table S5. Time-to-onset analysis of SSIBs associated with GLP-1 RA in the FAERS database. Table S6. IC025 and of top 23 medications. Table S7. The raw data of time to onset analysis in the 52 valid cases. Table S8. The raw data from primary, falsenegative, sensitivity and subgroup analyses in calculate disproportionality.

Acknowledgements

Thanks to the US FDA for providing a free source of data for the study.

Prior presentation

This manuscript has not been published or presented elsewhere in part or in entirety and is not under consideration by another journal.

Authors’ contributions

X.M.W. and M.B.L. designed the study. J.X.Z., Y.Z., and B.H.X. researched data, contributed to the discussion, and wrote the first draft of the manuscript. Y.F.Z. and S.J.L. provide the code of analysis. L.E.Z. and Y.H.L. performed the co-medication analysis. C.L.L. polished the language of the manuscript. X.M.W. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. All authors read and approved the final manuscript.

Authors’ Twitter handles

@JX_ZHOU_FJMU

Funding

No funding was provided for this study.

Availability of data and materials

The datasets generated and/or analyzed during the current study are available in the US FAERS database (https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html). The code generated and/or analyzed in the current study is available from the corresponding author on reasonable request.

Declarations

Ethical approval was not required as the study was conducted using deidentified publicly available data.
Not applicable

Abbreviations

GLP-1RAs Glucagon-like peptide 1 receptor agonists
GLP-1 Glucagon-like peptide 1

Competing interests

The authors declare that they have no competing interests.

Author details

Department of Pharmacy, Fujian Medical University Union Hospital, Fuzhou, Fujian, China. School of Pharmacy, Fujian Medical University, Fuzhou, Fujian, China. School of Medical Imaging, Fujian Medical University, Fuzhou, Fujian, China. Department of Respiratory, Shanghai Electric Power Hospital, Shanghai, China. Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai, China.
Received: 18 August 2023 Accepted: 25 January 2024
Published online: 14 February 2024

References

  1. Lingvay I, Leiter LA. Use of GLP-1 RAs in cardiovascular disease prevention: a practical guide. Circulation. 2018;137:2200-2.
  2. Patorno E, Htoo PT, Glynn RJ, Schneeweiss S, Wexler DJ, Pawar A, et al. Sodium-glucose cotransporter-2 inhibitors versus glucagon-like peptide-1 receptor agonists and the risk for cardiovascular outcomes in routine care patients with diabetes across categories of cardiovascular disease. Ann Intern Med. 2021;174:1528-41.
  3. Ma X, Liu Z, Ilyas I, Little PJ, Kamato D, Sahebka A, et al. GLP-1 receptor agonists (GLP-1RAs): cardiovascular actions and therapeutic potential. Int J Biol Sci. 2021;17:2050-68.
  4. Barritt AS, Marshman E, Noureddin M. Review article: role of glucagonlike peptide-1 receptor agonists in non-alcoholic steatohepatitis, obesity and diabetes-what hepatologists need to know. Aliment Pharmacol Ther. 2022;55:944-59.
  5. Davies MJ, Aroda VR, Collins BS, Gabbay RA, Green J, Maruthur NM, et al. Management of Hyperglycemia in Type 2 Diabetes, A consensus report by the american diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2022;45(11):2753-86.
  6. Jastreboff AM, Kushner RF. New frontiers in obesity treatment: GLP-1 and nascent nutrient-stimulated hormone-based therapeutics. Annu Rev Med. 2023;74:125-39.
  7. Bailey CJ, Flatt PR, Conlon JM. An update on peptide-based therapies for type 2 diabetes and obesity. Peptides. 2023;161:170939.
  8. O’Neil PM, Birkenfeld AL, McGowan B, Mosenzon O, Pedersen SD, Wharton S, et al. Efficacy and safety of semaglutide compared with liraglutide and placebo for weight loss in patients with obesity: a randomised, double-blind, placebo and active controlled, dose-ranging, phase 2 trial. Lancet. 2018;392:637-49.
  9. Chadda KR, Cheng TS, Ong KK. GLP-1 agonists for obesity and type 2 diabetes in children: systematic review and meta-analysis. Obes Rev. 2021;22:e13177.
  10. Tan B, Pan X-H, Chew HSJ, Goh RSJ, Lin C, Anand VV, et al. Efficacy and safety of tirzepatide for treatment of overweight or obesity. A systematic review and meta-analysis. Int J Obes (Lond). 2023;47:677-85.
  11. Pfizer Provides Update on GLP-1-RA Clinical Development Program for Adults with Obesity and Type 2 Diabetes Mellitus | Pfizer. https://www. pfizer.com/news/press-release/press-release-detail/pfizer-provides-update-glp-1-ra-clinical-development. Accessed 8 Aug 2023.
  12. Bezin J, Gouverneur A, Pénichon M, Mathieu C, Garrel R, Hillaire-Buys D, et al. GLP-1 receptor agonists and the risk of thyroid cancer. Diabetes Care. 2023;46:384-90.
  13. He L, Wang J, Ping F, Yang N, Huang J, Li Y, et al. Association of glucagonlike peptide-1 receptor agonist use with risk of gallbladder and biliary diseases: a systematic review and meta-analysis of randomized clinical trials. JAMA Intern Med. 2022;182:513-9.
  14. EMA. EMA statement on ongoing review of GLP-1 receptor agonists. European Medicines Agency. 2023. https://www.ema.europa.eu/en/ news/ema-statement-ongoing-review-glp-1-receptor-agonists. Accessed 20 Jul 2023.
  15. Fine KL , Rickert ‘Reilly LM , Sujan AC, Boersma K, Chang Z, et al. Initiation of opioid prescription and risk of suicidal behavior among youth and young adults. Pediatrics. 2022;149:e2020049750.
  16. Takeuchi T, Takenoshita S, Taka F, Nakao M, Nomura K. The relationship between psychotropic drug use and suicidal behavior in Japan: Japanese adverse drug event report. Pharmacopsychiatry. 2017;50:69-73.
  17. Gonda , Dome , Serafini . How to save a life: From neurobiological underpinnings to psychopharmacotherapies in the prevention of suicide. Pharmacol Ther. 2023;244:108390.
  18. Hughes JL, Horowitz LM, Ackerman JP, Adrian MC, Campo JV, Bridge JA. Suicide in young people: screening, risk assessment, and intervention. BMJ. 2023;381:e070630.
  19. Jing Y, Liu J, Ye Y, Pan L, Deng H, Wang Y, et al. Multi-omics prediction of immune-related adverse events during checkpoint immunotherapy. Nat Commun. 2020;11:4946.
  20. Raschi E, Fusaroli M, Giunchi V, Repaci A, Pelusi C, Mollica V, et al. adrenal insufficiency with anticancer tyrosine kinase inhibitors targeting vascular endothelial growth factor receptor: analysis of the FDA adverse event reporting system. Cancers (Basel). 2022;14:4610.
  21. Xia S, Gong H, Zhao Y, Guo L, Wang Y, Ma R, et al. Tumor lysis syndrome associated with monoclonal antibodies in patients with multiple myeloma: a pharmacovigilance study based on the FAERS database. Clin Pharmacol Ther. 2023;114:211-9.
  22. Zhou J, Wei Z, Xu B, Liu M, Xu R, Wu X. Pharmacovigilance of triazole antifungal agents: analysis of the FDA adverse event reporting system (FAERS) database. Front Pharmacol. 2022;13:1039867.
  23. Ren W, Wang W, Guo Y. Analysis of adverse reactions of aspirin in prophylaxis medication Based on FAERS database. Comput Math Methods Med. 2022;2022:7882277.
  24. Yu RJ, Krantz MS, Phillips EJ, Stone CA. Emerging causes of drug-induced anaphylaxis: a review of anaphylaxis-associated reports in the fda adverse event reporting system (FAERS). J Allergy Clin Immunol Pract. 2021;9:819829.e2.
  25. Hu Y, Bai Z, Tang Y, Liu R, Zhao B, Gong J, et al. Fournier gangrene associated with sodium-glucose cotransporter-2 inhibitors: a pharmacovigilance study with data from the U.S. FDA Adverse event reporting system. J Diabetes Res. 2020;2020:3695101.
  26. Liu L, Chen J, Wang L, Chen C, Chen L. Association between different GLP-1 receptor agonists and gastrointestinal adverse reactions: a realworld disproportionality study based on FDA adverse event reporting system database. Front Endocrinol (Lausanne). 2022;13:1043789.
  27. Hetrick SE, McKenzie JE, Bailey AP, Sharma V, Moller CI, Badcock PB, et al. New generation antidepressants for depression in children and adolescents: a network meta-analysis. Cochrane Database Syst Rev. 2021;5:CD013674.
  28. Sauzet O, Carvajal A, Escudero A, Molokhia M, Cornelius VR. Illustration of the weibull shape parameter signal detection tool using electronic healthcare record data. Drug Saf. 2013;36:995-1006.
  29. van Puijenbroek EP, Bate A, Leufkens HGM, Lindquist M, Orre R, Egberts ACG. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Pharmacoepidemiol Drug Saf. 2002;11:3-10.
  30. Norén GN, Hopstadius J, Bate A. Shrinkage observed-to-expected ratios for robust and transparent large-scale pattern discovery. Stat Methods Med Res. 2013;22:57-69.
  31. Trillenberg P, Sprenger A, Machner B. Sensitivity and specificity in signal detection with the reporting odds ratio and the information component. Pharmacoepidemiol Drug Saf. 2023;32:910-7.
  32. Noguchi Y, Tachi T, Teramachi H. Comparison of signal detection algorithms based on frequency statistical model for drug-drug interaction using spontaneous reporting systems. Pharm Res. 2020;37:86.
  33. Anderson N, Borlak J. Correlation versus causation? Pharmacovigilance of the analgesic flupirtine exemplifies the need for refined spontaneous ADR reporting. PLoS One. 2011;6:e25221.
  34. Fusaroli M, Raschi E, Giunchi V, Menchetti M, Rimondini Giorgini R, De Ponti F, et al. Impulse control disorders by dopamine partial agonists: a pharmacovigilance-pharmacodynamic assessment through the FDA adverse event reporting system. Int J Neuropsychopharmacol. 2022;25:727-36.
  35. Raschi E, Fusaroli M, Ardizzoni A, Poluzzi E, De Ponti F. Cyclin-dependent kinase 4/6 inhibitors and interstitial lung disease in the FDA adverse event reporting system: a pharmacovigilance assessment. Breast Cancer Res Treat. 2021;186:219-27.
  36. Horska K, Ruda-Kucerova J, Skrede S. GLP-1 agonists: superior for mind and body in antipsychotic-treated patients? Trends Endocrinol Metab. 2022;33:628-38.
  37. Kelly AS, Auerbach P, Barrientos-Perez M, Gies I, Hale PM, Marcus C, et al. A randomized, controlled trial of liraglutide for adolescents with obesity. N Engl J Med. 2020;382:2117-28.
  38. O’Neil PM, Aroda VR, Astrup A, Kushner R, Lau DCW, Wadden TA, et al. Neuropsychiatric safety with liraglutide 3.0 mg for weight management: Results from randomized controlled phase 2 and 3a trials. Diabetes Obes Metab. 2017;19:1529-36.
  39. Li J-R, Cao J, Wei J, Geng W. Case report: Semaglutide-associated depression: a report of two cases. Front Psychiatry. 2023;14:1238353.
  40. Klinitzke G, Steinig J, Blüher M, Kersting A, Wagner B. Obesity and suicide risk in adults-a systematic review. J Affect Disord. 2013;145:277-84.
  41. da Silva Bandeira BE, Dos Santos Júnior A, Dalgalarrondo P, de Azevedo RCS, Celeri EHVR. Nonsuicidal self-injury in undergraduate students: a cross-sectional study and association with suicidal behavior. Psychiatry Res. 2022;318:114917.
  42. Yan Y, Gong Y, Jiang M, Gao Y, Guo S, Huo J, et al. Utilization of glucagonlike peptide-1 receptor agonists in children and adolescents in China: a real-world study. Front Endocrinol (Lausanne). 2023;14:1170127.
  43. Dubé JP, Smith MM, Sherry SB, Hewitt PL, Stewart SH. Suicide behaviors during the COVID-19 pandemic: a meta-analysis of 54 studies. Psychiatry Res. 2021;301:113998.
  44. Yaribeygi H, Sathyapalan T, Sahebkar A. Molecular mechanisms by which GLP-1 RA and DPP-4i induce insulin sensitivity. Life Sci. 2019;234:116776.
  45. Jespersen MJ, Knop FK, Christensen M. GLP-1 agonists for type 2 diabetes: pharmacokinetic and toxicological considerations. Expert Opin Drug Metab Toxicol. 2013;9:17-29.
  46. Gliatto MF, Rai AK. Evaluation and treatment of patients with suicidal ideation. Am Fam Physician. 1999;59:1500-6.
  47. Battini V, Van Manen RP, Gringeri M, Mosini G, Guarnieri G, Bombelli A, et al. The potential antidepressant effect of antidiabetic agents: New insights from a pharmacovigilance study based on data from the reporting system databases FAERS and VigiBase. Front Pharmacol. 2023;14:1128387.
  48. Darwish AB, El Sayed NS, Salama AAA, Saad MA. Dulaglutide impedes depressive-like behavior persuaded by chronic social defeat stress model in male C57BL/6 mice: Implications on GLP-1R and cAMP/PKA signaling pathway in the hippocampus. Life Sci. 2023;320:121546.
  49. Rubino DM, Greenway FL, Khalid U, O’Neil PM, Rosenstock J, Sørrig R, et al. Effect of weekly subcutaneous semaglutide vs daily liraglutide on body weight in adults with overweight or obesity without diabetes: the STEP 8 randomized clinical trial. JAMA. 2022;327:138-50.
  50. Rubino D, Abrahamsson N, Davies M, Hesse D, Greenway FL, Jensen C, et al. Effect of continued weekly subcutaneous semaglutide vs placebo on weight loss maintenance in adults with overweight or obesity: the STEP 4 randomized clinical trial. JAMA. 2021;325:1414-25.
  51. Wadden TA, Bailey TS, Billings LK, Davies M, Frias JP, Koroleva A, et al. Effect of subcutaneous semaglutide vs placebo as an adjunct to intensive behavioral therapy on body weight in adults with overweight or obesity: the STEP 3 randomized clinical trial. JAMA. 2021;325:1403-13.
  52. Davies M, Færch L, Jeppesen OK, Pakseresht A, Pedersen SD, Perreault L, et al. Semaglutide 2.4 mg once a week in adults with overweight or obesity, and type 2 diabetes (STEP 2): a randomised, double-blind, doubledummy, placebo-controlled, phase 3 trial. Lancet. 2021;397:971-84.
  53. Sakaeda T, Tamon A, Kadoyama K, Okuno Y. Data mining of the public version of the FDA adverse event reporting system. Int J Med Sci. 2013;10:796-803.
  54. Khaleel MA, Khan AH, Ghadzi SMS, Adnan AS, Abdallah QM. A Standardized dataset of a spontaneous adverse event reporting system. Healthcare (Basel). 2022;10:420.
  55. McIntyre RS, Mansur RB, Rosenblat JD, Kwan ATH. The association between glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and suicidality: reports to the Food and Drug Administration Adverse Event Reporting System (FAERS). Expert Opin Drug Saf. 2023;1-9. Online ahead of print.
  56. Chen C, Zhou R, Fu F, Xiao J. Postmarket safety profile of suicide/ self-injury for GLP-1 receptor agonist: a real-world pharmacovigilance analysis. Eur Psychiatry. 2023;66:e99.
  57. Wang W, Volkow ND, Berger NA, Davis PB, Kaelber DC, Xu R. Association of semaglutide with risk of suicidal ideation in a real-world cohort. Nat Med. 2024. https://doi.org/10.1038/s41591-023-02672-2.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

  1. Maobai Liu and Xuemei Wu contributed equally to this work as cocorresponding authors.
    *Correspondence:
    Maobai Liu
    liumb0591@fjmu.edu.cn
    Xuemei Wu
    wuxuemei@fjmu.edu.cn
    Full list of author information is available at the end of the article