Ngianga-Bakwin Kandala 1. 2, Martin Connock 1, Amy Grove 1, Paul Sutcliffe 1, Syed Mohiuddin 3, Louise Hartley 1, Rachel Court 1, Ewen Cummins 4, Caroline Gordon 5, Aileen Clarke 1 1 Divisi Ilmu Kesehatan. Warwick Medical School, Universitas Warwick, Coventry. UK 2 KEMRI-Universitas Oxford-Wellcome Trust Collaborative Program, Malaria Kesehatan Masyarakat dan Epidemiologi Group, Pusat Kedokteran Geografis, Nairobi. Kenya 3 Ilmu Kesehatan, Universitas Manchester, Manchester. Inggris 4 McMDC Ltd, McMaster Development Consultants, Glasgow. Inggris 5 Sekolah Imunitas dan Infeksi, Fakultas Kedokteran dan Ilmu Gigi, Universitas Birmingham, Birmingham. UK Correspondence to Dr Ngianga-Bakwin Kandala NB. Kandala warwick. ac. uk Tujuan Untuk melakukan tinjauan sistematis dan meta-analisis untuk menyelidiki keefektifan klinis belimumab untuk pasien dengan lupus eritematosus sistemik (SLE) dan antinuklear dan anti-double-stranded DNA (DsDNA) autoantibodi. Metode Kami mencari delapan database elektronik dan daftar referensi untuk uji coba terkontrol acak (RCT) dari belimumab terhadap plasebo atau perawatan suportif terbaik. Penilaian kualitas dan meta analisis acak dilakukan. Rancang sebuah meta-analisis RCT. Peserta 2133 pasien SLE. Hasil primer dan sekunder mengukur SLE Responder Index (SRI) pada minggu ke 52. Hasil Tiga RCT terkontrol plasebo ganda (L02, BLISS-52 BLISS-76) menyelidiki 2133 pasien SLE. Uji coba BLISS-52 dan BLISS-76 merekrut pasien antiretroviral dan anti-dsDNA autoantibodi dan menunjukkan efektivitas belimumab untuk SRI pada minggu ke 52. Etnisitas dan lokasi geografis peserta sangat bervariasi antara uji coba BLISS. Meskipun tes untuk heterogenitas statistik negatif, hasil BLISS-52 secara sistematis lebih sesuai untuk semua hasil pengukuran. Meta-analisis hasil SRI BLISS selama 52 minggu menunjukkan keuntungan untuk belimumab (OR 1,63, 95 CI 1,27 sampai 2,09). Pada minggu ke 76, hasil utama SRI di BLISS-76 tidak signifikan secara statistik (OR 1,31, 95 CI 0,919 sampai 1,855). Tujuan kami adalah untuk mensintesis temuan dari uji coba terkontrol acak (RCT) dari belimumab untuk pasien dengan SLE dan antinuclear dan juga anti-dsDNA autoantibodi untuk membuat penilaian keseluruhan terhadap kinerja obat ini sehubungan dengan perawatan komparator dengan menggunakan SRI dan hasil lainnya (seperti Tercantum dalam gambar 1) dan untuk menilai temuan uji coba berdasarkan sampel populasi dan faktor geografis.18 Penelitian ini dilakukan sebagai bagian dari penelitian National Institute for Health Research, program Penilaian Teknologi Kesehatan (referensi dana hibah 107301 Informasi lebih lanjut tersedia dari: hta. ac. uk). Lingkup pencarian Kami mencari RCT yang menyelidiki belimumab yang diberikan secara intravena untuk pasien dengan SLE dan antinuclear andor autoantibodi anti-dsDNA. Komparator yang dipertimbangkan adalah belimumab versus plasebo dan belimumab versus perawatan suportif terbaik. Hasil mencakup semua tindakan terkait penyakit atau status kesehatan. Tidak ada batasan tahun publikasi, namun pencarian dibatasi hanya untuk referensi bahasa Inggris saja. Strategi pencarian Berikut delapan database yang dicari: Database Cochrane of Systematic Reviews the Cochrane Central Register of Controlled Trials (CENTRAL) DARE EMBASE HTA Database MEDLINE dan Pre-Medline and Science Citation Index. Strategi pencarian untuk database ini menggunakan kombinasi istilah yang berkaitan dengan populasi dan intervensi yang tercantum di atas strategi pencarian spesifik yang diberikan di lampiran lampiran tambahan online. Di MEDLINE dan EMBASE, strategi subjek dikombinasikan dengan strategi pencarian yang dirancang untuk mengidentifikasi RCT (lihat Lampiran Lampiran tambahan secara online). Studi yang tidak dipublikasikan diidentifikasi dengan uji klinis Uji Coba Terkendali Uji Coba Klinis EU Daftar Penelitian Klinis Global Studi Portofolio Penelitian Nasional Register WHO Clinical Trials Search Portal National Health Service Evidence Conference Proceedings Citation IndexScience dan Google. Selain itu, situs web tertentu digeledah: Obat-obatan dan Produk Kesehatan Regulator, Badan Obat-obatan Eropa, Food and Drug Administration (FDA) dan proses konferensi khusus berikut ini: American College of Rheumatology, British Society of Rheumatology dan Liga Eropa yang Melawan Rematik. Kriteria inklusi: Publikasi disertakan jika mereka menggambarkan hasil dari RCT dari belimumab untuk pasien SLE dengan autoantibodi positif. Dua reviewer secara independen menilai publikasi yang diambil untuk dimasukkan. Tidak ada ketidaksepakatan di antara para pengulas. Ekstraksi data: Publikasi yang berpotensi relevan diperoleh secara lengkap dan dinilai oleh dua reviewer yang sama. Satu reviewer mengekstrak data untuk semua ukuran hasil primer dan sekunder yang ditentukan, untuk kejadian buruk dan kematian. Peninjau kedua memeriksa data yang diekstrak. Evaluasi kualitas: Penilaian kualitas dan risiko bias dipandu oleh Daftar Periksa Pusat dan Daftar Diseminasi19 berdasarkan semua informasi dalam publikasi termasuk yang menentukan pelaporan pengacakan, penyembunyian alokasi, keseimbangan kelompok, penyimpangan, drop out, bias pelaporan hasil dan Apakah niat untuk mengobati analisis sudah digunakan Analisis statistik: OR yang tidak disesuaikan dan perbedaan rata-rata dihitung untuk hasil biner dan terus menerus. Heterogenitas statistik dihitung dengan menggunakan statistik I 2.20. Ada sedikit sekali penelitian untuk menganalisis bias publikasi.21 Meskipun pencarian menyeluruh kami tidak menemukan penelitian lebih lanjut, kami tidak dapat sepenuhnya mengesampingkan bahwa metode untuk menggabungkan kedua uji coba tersebut dapat menghasilkan perkiraan yang terlalu tinggi atau ukuran efek yang meremehkan karena bias publikasi. Ukuran hasil yang disesuaikan ditabulasikan dimana laporan ini dilaporkan. Efek acak meta-analisis22 dilakukan dengan menggunakan metode DerSimonian Laird di STATA V.11.23 Semua grafik disiapkan di Microsoft Excel 2010. Karakteristik studi yang disertakan Kami mengidentifikasi tiga RCT terkontrol plasebo dari belimumab versus perawatan standar: uji coba fase III yang disebut BLISS -5224 dan BLISS-7625 dan percobaan fase II (studi L02) .26 Diagram alir PRISMA menunjukkan proses identifikasi publikasi (lihat gambar 2). Kami mengidentifikasi percobaan yang sedang berlangsung di Asia.27 Ketiga percobaan yang telah selesai tampaknya berkualitas baik, rincian penyingkapan alokasi sangat kecil (tabel 1). Dalam meta-analisis, kami menyertakan uji coba dua fase III (BLISS-52 dan BLISS-76) karena populasi, rancangan percobaan dan hasil primer berbeda dalam percobaan L02. Penilaian kualitas uji coba yang disertakan Lokasi pusat yang berbeda pada uji coba multikentre BLISS 52 dan BLISS 76. Ada tambahan perbedaan populasi antara uji coba BLISS dan L02 pada saat perekrutan. Pelaporan hasil untuk pasien dengan autoantibodi anti-oksidan andor anti-dsDNA pada L02 hanya dimasukkan untuk subkelompok post hoc dan hasil primer yang diukur pada L02 tidak sebanding dengan studi BLISS. Untuk alasan ini, hasil studi L02 hanya disertakan di sini berkaitan dengan hasil keselamatan. Untuk percobaan BLISS, ukuran hasil utama komposit baru dikembangkan secara apriori dari diskusi antara FDA dan produsen dan disebut SRI (lihat gambar 1 dan tabel 3). Protokol yang menentukan titik akhir primer adalah proporsi responden SRI pada minggu ke 52. Ini diambil sebagai hasil utama dalam tinjauan sistematis ini. Khasiat menghasilkan dua uji coba BLISS untuk hasil efektivitas biner utama termasuk saat pertama kali SLE flare dan flare parah pertama dirangkum pada gambar 4. ATAU telah dihitung dengan menggunakan proporsi pasien dengan dan tanpa kejadian yang dilaporkan dalam artikel jurnal untuk ini. Percobaan.24. 25 Hasil keselamatan yang ditunjukkan pada gambar 4 dihitung setelah menggabungkan jumlah kejadian di tiga uji coba (L02, BLISS-52 dan BLISS-76) dan diambil dari dokumen FDA. HR untuk waktu untuk suar dilaporkan buruk dalam artikel jurnal dan data yang disajikan diambil dari penyerahan produsen ke FDA.28. Seperti yang ditunjukkan pada gambar 4, kedua uji coba ini memuaskan titik akhir primer ini dengan hasil yang lebih baik untuk BLISS-52. Perbedaan persentase responden pada kelompok belimumab relatif terhadap kelompok plasebo lebih besar pada BLISS-52 (14), dibandingkan dengan BLISS-76 (9,4). Ringkasan hasil untuk biner utama dan waktu untuk hasil kejadian di uji coba terkontrol acak belimumab. Untuk hasil efektifitas biner lainnya, uji coba BLISS-52 memberikan hasil yang lebih menguntungkan ke belimumab daripada pada BLISS-76, dengan hasil yang terakhir gagal mencapai tingkat signifikansi statistik konvensional kecuali untuk peningkatan 4 titik pada skor SLEDAI pada minggu 52. Pengiriman artikel dan produsen jurnal ke FDA dan ke NICE menggunakan model regresi logistik dan melaporkan OR disesuaikan dengan faktor stratifikasi yang digunakan secara acak. OR yang disesuaikan untuk respon di BLISS-52 dan di BLISS-76 masing-masing, 1,83 (95 CI 1,30 sampai 2,59 p0.0006) dan 1,52 (95 CI 1,07 sampai 2,15 p0.0207). Sekali lagi tanggapan yang superior ditemukan untuk percobaan BLISS-52. Pada minggu ke 76, tanggapan ATAU yang tidak disesuaikan untuk tanggapan SRI dalam uji coba BLISS-76 berhenti mencapai signifikansi statistik (gambar 4) ini juga berlaku untuk OR yang dilaporkan disesuaikan dengan regresi logistik (OR 1,31, 95 CI 0,92-1,87, p0.1323 ) .29 Berkenaan dengan waktu untuk suar pertama atau untuk suar berat pertama (masing-masing diperkirakan di atas 52 minggu masa tindak lanjut), tanggapan yang dilaporkan dalam penyampaian FDA sekali lagi lebih unggul untuk BLISS-52. Setiap hasil gagal mencapai signifikansi statistik konvensional untuk BLISS-76. Pengajuan FDA juga melaporkan hasil yang lebih matang yang diperkirakan berusia di atas 76 minggu setelah follow-up untuk BLISS-76, dan sekali lagi ini mengindikasikan kurangnya signifikansi statistik untuk kedua hasil (HR untuk tingkat pertama: 1,05, 95 CI 0,88 sampai 1,27 HR untuk tingkat pertama yang parah Suar 1,30, 95 CI 0,92 sampai 1,85). Gambar 4 menunjukkan hasil untuk hasil keselamatan utama. Meskipun ada kejadian buruk yang lebih serius, infeksi yang lebih serius dan lebih banyak kematian terkait dengan belimumab dibandingkan dengan plasebo, tidak ada OR untuk hasil ini yang mencapai signifikansi statistik. Ada 14 kematian selama fase terkontrol dari tiga percobaan 3 pada kelompok plasebo (n675) dan 11 pada kelompok belimumab (n1458) dengan 6 pada 10 mgkg dan 5 pada kelompok 1 mgkg, masing-masing (OR 11,7 95 CI 0,474 Sampai 6.124). Ada berbagai penyebab kematian: lima berhubungan dengan infeksi, tiga diantaranya adalah stroke, tiga kejadian kardiovaskular, dua kasus bunuh diri, satu kanker, satu dari komplikasi terkait SLE dan dua penyebabnya tidak diketahui. Hasil untuk hasil terus menerus dirangkum dalam gambar 5. Perubahan rata-rata dari baseline yang dilaporkan dalam artikel jurnal BLISS dan dalam pengantar produsen ke FDA dan NICE telah digunakan untuk menghasilkan statistik perbedaan rata-rata (kadang-kadang disebut mean weighted mean31). Ini menunjukkan superioritas respons pada BLISS-52 yang berkaitan dengan BLISS-76 untuk semua hasil yang dilaporkan, sebuah pola yang serupa dengan hasil biner. Perubahan rata-rata dari awal untuk skor FACIT-fatigue dan untuk nilai utilitas EQ-5D (tidak digambarkan) tidak mencapai signifikansi statistik dan peningkatan yang terlihat lagi pada BLISS-52 karena ini lebih unggul dari yang terlihat pada BLISS-76. Ringkasan hasil untuk hasil terus menerus utama dalam uji coba BLISS 52 dan BLISS 76. Singkatnya, BLISS-52 menunjukkan superioritas sistematik atas BLISS-76 yang dapat memberi manfaat bagi belimumab dengan berbagai respons uji (biner, waktu ke waktu dan kontinyu), yang mungkin mencerminkan perbedaan geografis antara uji coba (tabel 2 dan gambar 3 ). Hasil utama dalam BLISS-76 dicapai (disesuaikan OR 1,52, 95 CI 1,07-2,15), namun perbedaan geografis yang besar dalam BLISS-76 mencolok: tingkat 32 (46 dari 145) dan 35 (47 dari 136), untuk plasebo Dan belimumab dilaporkan untuk pasien dari Amerika Utara dan Kanada (satu tanggapan lebih besar untuk belimumab), sedangkan untuk pasien BLISS-76 di luar daerah ini, 15 respons yang lebih besar untuk belimumab mengenai plasebo dilaporkan, 71 dari 137 (51,8) untuk Belimumab dan 47 dari 130 (36,1) untuk plasebo. Sebagai perbandingan, tingkat yang sesuai untuk pasien dari Amerika Latin di BLISS-52 adalah 49 plasebo (71 dari 145), dan 61 belimumab (85 dari 140). Pengajuan produsen ke FDA dan hasil gabungan NICE dari dua uji coba BLISS dengan menggabungkan pasien dan menerapkan model regresi logistik yang dijelaskan di atas untuk hasil primer (proporsi responden SRI pada minggu ke 52), perbedaan antara kelompok belimumab dan kelompok plasebo. 11.8.28 Metode alternatif untuk menggabungkan uji coba dengan meta-analisis hasil studi dari dua percobaan BLISS menunjukkan manfaat signifikan dari belimumab untuk sebagian besar hasil utama termasuk SRI, SELENA-SLEDAI, yang memburuk pada penggunaan PGA, steroid, BILAG Skor, atau waktu untuk suar parah pertama, dan jumlah rata-rata flare dan flare parah selama 52 minggu dan minggu 2452 (gambar 6). Pengujian untuk heterogenitas statistik hasil OR dan HR tidak signifikan. Meta-analisis ini menawarkan alternatif untuk regresi logistik pabrikan dan hal ini dibenarkan untuk dua uji coba dengan ukuran besar (N577 dan N548) namun, hasil ini, dan data penyatuan data pasien dari dua percobaan sebelum regresi logistik, topeng Perbedaan sistematis antara uji coba di semua hasil. Meta-analisis hasil utama dalam dua percobaan BLISS. Diskusi Kami melakukan tinjauan sistematis terhadap keefektifan klinis belimumab, sebuah pengobatan baru yang ditargetkan pada pasien SLE dengan antigen dan anti-dsDNA autoantibodi. Kami melakukan pencarian ekstensif dan tinjauan sistematis atas percobaan yang telah selesai dan terus-menerus menggunakan sejumlah database dan dengan memeriksa daftar referensi. Data diekstraksi secara independen dan penelitian dinilai berkualitas. Meta-analisis efek acak dilakukan. Kami mengidentifikasi tiga data RCT (L02, BLISS-52, BLISS-76) yang melaporkan lebih dari 2000 pasien. Berbeda dengan uji coba BLISS, L02 merekrut pasien yang belum tentu memiliki antibodi antinuclear atau anti-dsDNA saat memulai penelitian. L02 gagal menunjukkan keefektifan klinis untuk titik akhir utamanya.26 Analisis metoda studi BLISS menunjukkan manfaat belimumab dengan hasil utama utama (SRI), menunjukkan perbaikan pada 52 minggu (OR 1,63 95 CI 1,27 sampai 2,09 plt0. 001), walaupun pada minggu ke 76 proporsi responden SRI dalam percobaan BLISS-76 berhenti mencapai signifikansi statistik (OR 1,31 95 CI 0,92 sampai 1,87 p0.1323). Tidak ada perbedaan yang signifikan antara kelompok plasebo dan kelompok intervensi dalam kualitas hidup atau kejadian buruk. Kami menemukan bahwa manfaat belimumab secara sistematis lebih besar di seluruh papan (walaupun tidak signifikan) dalam uji coba BLISS-52 dan meskipun uji heterogenitas statistik negatif, lokasi geografis pusat studi dan latar belakang ras dan etnis peserta sangat bervariasi. Jika kedua uji coba BLISS diambil dari populasi yang sama, sementara orang mungkin mengharapkan hasil berbeda, kami akan mengantisipasi bahwa ini akan terjadi secara acak di antara uji coba yang lebih baik dan beberapa lebih buruk daripada yang lain. Sangat sedikit penelitian yang secara langsung menilai keberadaan dan pentingnya perbedaan geografis dalam hasil percobaan.3234 Faktor kunci yang berkontribusi terhadap perbedaan tersebut adalah variasi karakteristik populasi pasien yang mendasari dan variasi dalam pelaksanaan studi. Vickers dkk menemukan bahwa studi Asia Timur dan Eropa Timur memiliki proporsi hasil uji positif yang lebih tinggi bila dibandingkan dengan negara lain. Hal ini terlihat pada kasus sekarang untuk hasil utama di mana tingkat respons belimumab dan plasebo pada studi BLISS 52 lebih besar daripada rasio BLISS-76 dan, tingkat respons plasebo pada BLISS-52 (49) lebih besar daripada Untuk lengan belimumab BLISS-76 (43). OShea dan DeMets35 juga melaporkan bahwa dalam Uji Hati Serangan Blocker, ada perbedaan tidak hanya pada arah tetapi juga seukuran efek pengobatan antara Kanada dan Amerika Serikat, walaupun perlu dicatat bahwa tujuan awal dari percobaan tersebut Bukan investigasi perbedaan internasional dalam efek pengobatan. Satu studi menemukan bahwa 9699 dari total varian dalam penggunaan strategi global untuk membuka arteri koroner yang tersumbat IV uji coba koroner akut (GUSTO IV ACS) dapat dipertanggungjawabkan dengan faktor tingkat pasien.36 Uji coba internasional perlu menyelaraskan pelatihan penyidik , Seleksi pasien, manajemen pengobatan, ambang batas untuk penerimaan pusat, akses terhadap fasilitas, pemastian titik akhir dan, secara implisit, hasil dari kepentingan3744 karena ada kemungkinan di pusat-pusat di berbagai negara, faktor-faktor ini mungkin berbeda secara sistematis.37 Sama, perbedaan mendasar dalam Populasi dan negara (etnisitas, genetika, status sosial ekonomi dan sistem kesehatan) dan sifat dan epidemiologi SLE menurut latar belakang etnis dapat menyebabkan perbedaan dalam melaporkan hasil dan hasil penggabungan. Hasil yang digunakan dalam uji coba BLISS tidak akan diketahui oleh sebagian besar peneliti dan mungkin saja kriteria tersebut berbeda antar negara. Secara khusus, PGA adalah elemen penting dari hasil yang diukur (lihat gambar 1). PGA diukur sebagai hasil dalam dirinya sendiri, dan juga tergabung dalam SRI. PGA menjadi perhatian karena sebagai penilaian dokter global terhadap status SLE pasien itu subjektif. Pada minggu ke 24 keuntungan belimumab atas plasebo dalam persentase perubahan skor PGA dari baseline jauh lebih besar, sekitar 10, untuk BLISS 52 daripada untuk BLISS 76 hasil yang lebih besar ini dalam satu percobaan kemungkinan memiliki pengaruh penting pada temuan ini. Untuk efektivitas relatif dalam dua percobaan. Hasil terbaru dari belimumab pada pasien dengan SLE (rancangan studi fase II, studi penyuluhan yang tidak terkontrol) melaporkan bahwa dari 449 pasien dengan pasien SLE aktif (USACanada) 177 (39,4) tetap menjalani perawatan setelah 7 tahun menjalani terapi (yaitu sekitar 1.746 pasien kumulatif - years) dan bahwa subkelompok ini menunjukkan peningkatan berkelanjutan yang berkelanjutan pada aktivitas penyakit SLE (SRI dan PGA) .30 Kesimpulan Sebagai kesimpulan, tinjauan sistematis dan efek acak meta-analisis dua RCT belimumab untuk pasien dengan SLE autoantibodi positif menunjukkan hasil positif pada Hasil utama pada minggu ke 52. Namun, mengingat populasi yang berbeda belajar di lokasi yang berbeda dalam uji coba BLISS dan hasil yang unggul secara konsisten dari satu percobaan dibandingkan dengan yang lain, kami menganggap bahwa heterogenitas populasi, perbedaan geografis dan variasi dalam uji coba dan Penilaian hasil mungkin telah memainkan peran dalam mempengaruhi hasil. Namun, generalisabilitas hasil yang dikumpulkan secara meta-analitis atau dengan regresi logistik harus dilihat dengan hati-hati dan kami menyarankan bahwa terlalu dini untuk menarik kesimpulan yang kuat dalam kasus ini. Ucapan Terimakasih Para penulis mengucapkan terima kasih kepada National Institute for Health Research, Program Penilaian Teknologi Kesehatan untuk mendanai pekerjaan ini. Kontributor N-BK berpartisipasi dalam konsepsi dan perancangan, analisis dan interpretasi data, penyusunan artikel, revisi kritis untuk konten intelektual penting dan persetujuan dari artikel akhir untuk diajukan. MC berpartisipasi dalam konsepsi dan desain, analisis dan interpretasi data, tinjauan literatur, interpretasi hasil, penyusunan artikel, revisi kritis untuk konten intelektual penting dan dalam persetujuan akhir dari artikel akhir untuk diserahkan. AG berpartisipasi dalam interpretasi hasil dan revisi kritis untuk konten intelektual yang penting. PS berpartisipasi dalam tinjauan pustaka, interpretasi hasil dan revisi kritis untuk konten intelektual yang penting. SM berpartisipasi dalam analisis data dan interpretasi, interpretasi hasil dan revisi kritis untuk konten intelektual yang penting. LH berpartisipasi dalam tinjauan pustaka, interpretasi hasil dan revisi kritis untuk konten intelektual yang penting. RC berpartisipasi dalam tinjauan literatur dan revisi kritis untuk konten intelektual penting. EC berpartisipasi dalam interpretasi hasil dan revisi kritis untuk konten intelektual penting. CG berpartisipasi dalam interpretasi hasil dan revisi kritis untuk konten intelektual yang penting. AC berpartisipasi dalam konsepsi dan desain, interpretasi hasil, penyusunan artikel, revisi kritis untuk konten intelektual penting dan dalam persetujuan artikel akhir untuk diajukan. Semua penulis membaca dan menyetujui manuskrip terakhir. Pendanaan Karya ini didukung oleh National Institute for Health Research, Program Penilaian Teknologi Kesehatan (nomor proyek 107301). Kepentingan bersaing Tidak ada. Assesment dan peer review Tidak ditugaskan peer review secara eksternal. Pernyataan berbagi data Tidak ada data tambahan yang tersedia. Informasi hak cipta: Diterbitkan oleh BMJ Publishing Group Limited. Izin untuk menggunakan (jika belum diberikan dengan lisensi) silakan masuk ke grup. bmjgrouprights-licensingpermissions Ini adalah artikel Open Access yang didistribusikan sesuai dengan lisensi Creative Commons Attribution Non Commercial (CC BY-NC 3.0), yang memungkinkan pihak lain untuk Mendistribusikan, remix, mengadaptasi, membangun atas karya ini secara tidak komersial, dan melisensikan karya turunan mereka dengan persyaratan yang berbeda, asalkan karya aslinya benar-benar dikutip dan penggunaannya tidak komersial. Lihat: creativecommons. orglicensesby-nc3.0 Referensi Bernatsky S, Joseph L, Pineau CA, dkk. Penilaian berbasis populasi kejadian lupus eritematosus sistemik dan prevalensi dan implikasi penggunaan data administratif untuk studi epidemiologi. Reumatologi (Oxford) 2007 46. 1814 18.Pons-Estel GJ, Alarcon GS, Scofield L, dkk. Memahami epidemiologi dan perkembangan sistemik lupus erythematosus. Semin Arthritis Rheum 2010 39. 257 68.Gordon C, Li CK, Isenberg DA. Lupus eritematosus sistemik. Kedokteran 2010 38. 73 80.Hochberg MC. Memperbarui kriteria reumatologi American College of Rheumatology untuk klasifikasi lupus eritematosus sistemik. Arthritis Rheum 1997 40. 1725.Tan EM, Cohen AS, Fries JF, dkk. Kriteria revisi 1982 untuk klasifikasi lupus eritematosus sistemik. Arthritis Rheum 1982 25. 1271 7.Griffiths B, Mosca M, Gordon C. Penilaian pasien dengan sistemik lupus erythematosus dan penggunaan indeks aktivitas penyakit lupus. Best Pract Res Clin Rheumatol 2005 19. 685 708.Somers EC, Thomas SL, Smeeth L, dkk. Insiden lupus eritematosus sistemik di Inggris Raya, 19901999. Arthritis Rheum 2007 57. 612 18.Pons-Estel GJ, Alarcon GS, McGwin G Jr. et al. Efek pelindung hydroxychloroquine pada kerusakan ginjal pada pasien dengan lupus nephritis: LXV, data dari kohort multietnis AS. Arthritis Rheum 2009 61. 830 9.Pons-Estel GJ, Gonzalez LA, Zhang J, dkk. Prediktor kerusakan kardiovaskular pada pasien dengan lupus eritematosus sistemik: data dari LUMINA (LXVIII), sebuah kelompok multietnis AS. Rheumatology (Oxford) 2009 48. 817 22.Bernatsky S, Boivin JF, Joseph L, dkk. Kematian pada lupus eritematosus sistemik. Arthritis Rheum 2006 54. 2550 7.Kurien BT, Scofield RH. Tekanan Autoantibody dalam diagnosis lupus eritematosus sistemik. Scand J Immunol 2006 64. 227 35.Merrill JT, Neuwelt CM, Wallace DJ, dkk. Khasiat dan keamanan rituximab pada lupus eritematosus sistemik sedang-ke-sangat aktif: evaluasi lupus erythematosus sistemik acak, double-blind, fase IIIII untuk uji coba rituximab. Arthritis Rheum 2010 62. 222 33.Summary karakteristik produk. 2012. ema. europa. eu Temukan obat Obat-obatan manusia Laporan Penilaian Publik Eropa (diakses Jan 2012). American College of Rheumatology. Belimumab untuk Lupus Erythematosus Sistemik. 2011. rheumatology. orgpublicationshotline20110315belimumab. asp (diakses Jan 2012). Buletin Obat Baru Belimumab (BenlystaHuman Genome Sciences, Inc). 2012. healthcare. utah. edupharmacybulletinsNDB228.pdf (diakses Jan 2012). Isenberg DA, Rahman A, Allen E, dkk. BILAG 2004. Pengembangan dan validasi awal dari versi terbaru dari indeks aktivitas penyakit Kepulauan Isonia Lupus Assessment untuk pasien dengan lupus eritematosus sistemik. Reumatologi 2005 44. 902 6.Gordon C, Bertsias G, Ioannidis JPA, dkk. EULAR menunjukkan perlunya melakukan uji klinis pada lupus eritematosus sistemik. Ann Rheum Dis 2009 68. 470 6.Burgos PI, McGwin G Jr. Pons-Estel GJ, dkk. Pasien AS keturunan Hispanik dan Afrika mengembangkan nefritis lupus pada awal program penyakit: data dari LUMINA, sebuah kelompok multietnis AS (LUMINA LXXIV). Ann Rheum Dis 2011 70. 393 4.Khan KS, Ter Riet G, Glanville J, dkk. Tinjauan Penelitian Efektivitas. Petunjuk CRD untuk Melaksanakan atau Menkominfo. Edisi 2 (ISBN 1900640201) Laporan CRD No. 4. York. Pusat NHS untuk Tinjauan dan Diseminasi (CRD), Universitas York. 2001. york. ac. ukinstcrdreport4.htm (diakses Januari 2012).Higgins JP, Thompson SG, Deeks JJ, dkk. Mengukur ketidakkonsistenan meta-analisis. BMJ 2003 327. 557 60.Higgins JPT, Green S. Buku Cochrane untuk Tinjauan Intervensi Sistematik Versi 5.1.0 diperbarui pada bulan Maret 2011. Kolaborasi Cochrane, 2011. 2012. cochrane-handbook. org (diakses Jan 2012).DerSimonian R, Laird N. Meta-analisis dalam uji klinis. Control Clin Trials 1986 7. 177 88.STATA Analisis statistik dan perangkat lunak. 2011. stata (diakses Apr 2012). Navarra SV, Guzman RM, Gallacher AE, dkk. Khasiat dan keamanan belimumab pada pasien dengan lupus eritematosus sistemik aktif: percobaan acak, terkontrol plasebo, fase 3. Lancet 2011 377. 721 31.Furie R, Petri M, Zamani O, dkk. Sebuah fase III, acak, studi terkontrol plasebo dari belimumab, antibodi monoklonal yang menghambat stimulator limfosit B, pada pasien dengan lupus eritematosus sistemik. Arthritis Rheum 2011 63. 3918 30.Wallace DJ, Stohl W, Furie RA, dkk. Sebuah studi fase II, acak, double-blind, placebo-controlled, dosis mulai dari belimumab pada pasien dengan lupus erythematosus sistemik aktif. Arthritis Rheum 2009 61. 1168 78.GlaxoSmithKline. GSK1550188 Sebuah studi 52 minggu tentang belimumab versus plasebo dalam pengobatan subjek dengan SLE yang terletak di Asia Timur Laut. 2011. clinicaltrials. govct2showNCT01345253 (diakses Jan 2012). Administrasi Pangan dan Obat-obatan Amerika Serikat. Risalah Rapat Komite Penasihat Arthritis untuk Pertemuan 16 November 2010: Belimumab: BENLYSTA (belimumab): Pengobatan Lupus Erythematosus Sistemik: BLA 125370. 2010. fda. govdownloadsAdvisoryCommitteesCommitteesMeetingMaterialsDrugsArthritisDrugsAdvisoryCommitteeUCM233581.pdf (diakses Jan 2012). GlaxoSmithKline. Penilaian Teknologi Tunggal (Single Technology Appraisal / STA): Belimumab untuk pengobatan lupus erythematosus lesi aktif autoantibodi aktif. 2011. guidance. nice. org. ukTAlive133075649056490.pdf (diakses Jan 2012). Egger M, GD Smith, Altman DG. Tinjauan sistematis dalam perawatan kesehatan, meta-analisis dalam konteks. 2nd edn. Buku BMJ. 2001.Pan Z, Trikalinos TA, Kavvoura FK, et al. Bias pustaka lokal dalam epidemiologi genetik: evaluasi empiris terhadap literatur Tiongkok. PLoS Med 2005 2. e334.Vickers A, Goyal N, Harland R, dkk. Apakah negara tertentu hanya menghasilkan hasil positif Tinjauan sistematis tentang percobaan terkontrol. Control Clin Trials 1998 19. 159 66.Zhang D, Freemantle N, Cheng KK. Apakah uji coba acak dilakukan di China atau India bias Analisis empiris komparatif. J Clin Epidemiol 2011 64. 90 5.OShea JC, DeMets DL. Masalah statistik yang berkaitan dengan perbedaan internasional dalam uji klinis. Am Heart J 2001 142. 21 8.Chang WC, Midodzi WK, Westerhout CM, dkk. Apakah perbedaan internasional dalam hasil sindrom koroner akut nyata atau nyata Analisis bertingkat. J Epidemiol Community Health 2005 59. 427 33.White HD. Perbedaan internasional: seleksi, kebisingan, atau jantung Eur yang sebenarnya J 2000 21. 339 42.Fox KA, Goodman S, Bigonzi F, dkk. Perbedaan dan hasil antar daerah dalam analisis angina yang tidak stabil uji coba internasional ESSENCE. Khasiat dan Keamanan Enoxaparin Subkutan dalam Acara Koroner Non-Q-wave. Eur Heart J 2000 21. 1433 9.Sinnaeve PR, Van de Werf FJ. Pola global perawatan kesehatan untuk sindrom koroner akut. Curr Opin Cardiol 2004 19. 625 30.Faergeman O, Kjekshus J, Cook T, dkk. Perbedaan dalam pengobatan penyakit jantung koroner antara negara-negara seperti yang terungkap dalam Skandinavia Simvastatin Survival Study (4S). Eur Heart J 1998 19. 1531 7.Berger JS, Stebbins A, Granger CB, dkk. Dosis aspirin awal dan hasil di antara pasien infark miokard ST-elevasi yang diobati dengan terapi fibrinolitik. Sirkulasi 2008 117. 192 9.Welsh RC, Chang W, Goldstein P, dkk. Waktu untuk pengobatan dan dampak dokter pada manajemen pra-rumah sakit terhadap infark miokard elevasi ST akut: wawasan dari uji coba ASSENT-3 PLUS. Jantung 2005 91. 1400 6.Yusuf S, Flather M, Pogue J, dkk. Variasi antar negara dalam prosedur dan hasil jantung invasif pada pasien dengan amina yang tidak stabil atau infark miokard tanpa elevasi ST awal. OASIS (Organisasi untuk Menilai Strategi untuk Sindrom Iskemik) Investigator Registri. Lancet 1998 352. 507 14.Bleyer AJ, Hylander B, Sudo H, dkk. Sebuah studi internasional tentang kepatuhan pasien terhadap hemodialisis. JAMA 1999 281. 1211 13.Mapping Genetik Lupus Erythematosus Sistemik Susceptibility Gen Swapan K. Nath Berafiliasi dengan Program Arthritis dan Imunologi, Oklahoma Medical Research Foundation. Jennifer A. Kelly Berafiliasi dengan Program Arthritis dan Imunologi, Oklahoma Medical Research Foundation. John B. Harley Berafiliasi dengan Program Arthritis dan Imunologi, Yayasan Riset Medis Oklahoma Departemen Kedokteran dan Patologi, Pusat Kesehatan Universitas Oklahoma Pusat Departemen Urusan Medis Veteran. R. Hal Scofield Berafiliasi dengan Program Arthritis dan Imunologi, Yayasan Riset Medis Oklahoma Departemen Kedokteran dan Patologi, Pusat Kesehatan Universitas Oklahoma Pusat Departemen Urusan Dokter Veteran Harga akhir dapat bervariasi sesuai dengan PPN setempat. Systemic lupus erythematosus (SLE) adalah prototipe sistemik, penyakit radang autoimun yang dapat melibatkan hampir semua jenis organ atau jaringan. Penyakit ini memiliki kecenderungan keluarga yang kuat namun, seperti kebanyakan penyakit manusia, memiliki pola pewarisan yang kompleks yang konsisten dengan gen kerentanan dan faktor risiko lingkungan. Studi asosiasi telah dilakukan, terutama untuk kompleks histokompatibilitas utama pada kromosom 6 dan untuk berbagai komponen pelengkap. Several large familial studies have begun to report results for genetic linkage. Linkage has been established for many genetic intervals. SLE is a complex clinical illness, and investigation of the genetics of the illness based on clinical manifestations revealed linkages not found without consideration of the phenotype of the disease. Autoantibodies autoantigens complement genetic association genetic linkage HLA systemic lupus erythematosus Tan, E. M. Cohen, A. S. Fries, J. F. Masi, A. T. McShane, D. J. Rothfield, N. F. et al. (1982) The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum. 25 . 12711277. PubMed CrossRef Hochberg, M. C. (1997) The epidemiology of systemic lupus erythematosus, in Dubois Lupus Erythematosus (Wallace, D. J. Hahn, B. H. eds.), Williams and Wilkins, Baltimore, MD, pp. 4965. Lawrence, R. C. Helmick, C. G. Arnett, F. C. Deyo, R. A. Felson, D. T. Giannini, E. H. et al. (1998) Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States. Arthritis Rheum. 41 . 778799. PubMed CrossRef Petri, M. (1998) The effect of race on incidence and clinical course in systemic lupus erythematosus: the Hopkins Lupus Cohort. J. Am. Med. Womens Assoc. 53 . 916. PubMed Vyse, T. J. and Todd, J. A. (1996) Genetic analysis of autoimmune disease. Cell 85 . 311318. PubMed CrossRef Deapen, D. Escalante, A. Weinrib, L. Horwitz, D. Bachman, B. Roy-Burman, P. et al. (1992) A revised estimate of twin concordance in systemic lupus erythematosus. Arthritis Rheum. 35 . 311318. PubMed CrossRef Wandstrat, A. and Wakeland, E. (2001) The genetics of complex autoimmune diseases: non-MHC susceptibility genes. Nat. Immunol. 2. 802809. PubMed CrossRef Bias, W. B. Reveille, J. D. Beaty, T. H. Meyers, D. H. and Arnett, F. C. (1986) Evidence that autoimmunity in man is a Mendelian dominant trait. Saya. J. Hum. Genet. 39 . 584602. PubMed Becker, K. G. Simon, R. M. Bailey-Wilson, J. E. Freidlin, B. Biddison, W. E. McFarland, H. F. et al. (1998) Clustering of non-major histocompatibility complex susceptibility candidate loci in human autoimmune diseases. Proc. Natl. Acad. Sci. U S A 95 . 99799984. PubMed CrossRef Botstein, D. White, R. L. Skolnick, M. and Davis, R. W. (1980) Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Saya. J. Hum. Genet. 32 . 314331. PubMed Weber, J. L. and May, P. E. (1989) Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction. Saya. J. Hum. Genet. 44 . 388396. PubMed Litt, M. and Luty, J. A. (1989) A hypervariable microsatellite revealed by in vitro amplification of a dinucleotide repeat within the cardiac muscle actin gene. Saya. J. Hum. Genet. 44 . 397401. PubMed Lindqvist, A. K. Steinsson, K. Johanneson, B. Kristjansdottir, H. Arnasson, A. Grondal, G. et al. (2000) A susceptibility locus for human systematic lupus erythematosus (hSLE1) on chromosome 2q. J. Autoimmun. 14 . 169178. PubMed CrossRef Prokunina, L. Castillejo-Lopez, C. Oberg, F. Gunnarsson, I. Berg, L. Magnusson, V. et al. (2202) A regulatory polymorphism in PDCD1 is associated with susceptibility to systemic lupus erythematosus in humans. Nature Genet. 32 . 666669. CrossRef Reveille, J. D. Bartolucci, A. and Alarcon, G. S. (1990) Prognosis in systemic lupus erythematosus. Negative impact of increasing age at onset, black race, and thrombocytopenia, as well as causes of death. Arthritis Rheum. 33 . 3748. PubMed CrossRef Scofield, R. H. Bruner, G. R. Kelly, J. A. Kilpatrick, J. Bacino, D. Nath, S. K. et al. (2003) Thrombocytopenia identifies a severe familial phenotype of systemic lupus erythematosus and reveals genetic linkages at 1q22 and 11p13. Blood 101 . 992997. PubMed CrossRef Walport, M. J. (2002) Complement and systemic lupus erythematosus. Arthritis Res. 4(Suppl. 3) . S279S293. PubMed CrossRef Kurien, B. T. Newland, J. Paczkowski, C. Moore, K. L. and Scofield, R. H. (2000) Association of neutropenia in systemic lupus erythematosus with anti-Ro and binding of an immunologically crossreactive neutrophil membrane antigen. Clin Exp Immunol. 120 . 209217. PubMed CrossRef Miki, Y. Swensen, J. Shattuck-Eidens, D. Futreal, P. A. Harshman, K. Tavtigian, S. et al. (1994) A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1 . Science 266 . 6671. PubMed CrossRef Wooster, R. Bignell, G. Lancaster, J. Swift, S. Seal, S. Mangion, J. et al. (1995) Identification of the breast cancer susceptibility gene BRCA2 . Nature 378 . 789792. PubMed CrossRef Horikawa, Y. Oda, N. Cox, N. J. Li, X. Orho-Melander, M. Hara, M. Hinokio, Y. et al. (2000) Genetic variation in the gene encoding calpain-10 is associated with type 2 diabetes mellitus. Nat Genet. 26 . 163175. PubMed CrossRef Hugot, J. P. Chamaillard, M. Zouali, H. Lesage, S. Cezard, J. P. Belaiche, J. et al. (2001) Association of NOD2 leucine-rich repeat variants with susceptibility to Crohns disease. Nature 411 . 599603. PubMed CrossRef Ogura, Y. Bonen, D. K. Inohara, N. Nicolae, D. L. Chen, F. F. Ramos, R. et al. (2001) A frameshift mutation in NOD2 associated with susceptibility to Crohns disease. Nature 411 . 603606. PubMed CrossRef Stefansson, H. Sigurdsson, E. Steinthorsdottir, V. Bjornsdottir, S. Sigmundsson, T. Ghosh, S. et al. (2002) Neuregulin 1 and susceptibility to schizophrenia. Am J Hum Genet. 71 . 877892. PubMed CrossRef Van Eerdewegh, P. Little, R. D. Dupuis, J. Del Mastro, R. G. Falls, K. et al. (2002) Association of the ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 418 . 426430. PubMed CrossRef Altmuller, J. Palmer, L. J. Fischer, G. Scherb, H. and Wjst, M. (2001) Genomewide scans of complex human diseases: true linkage is hard to find. Am J Hum Genet. 69 . 936950. PubMed CrossRef Hirschhorn, J. N. Lohmueller, K. Byrne, E. and Hirschhorn, K. (2002) A comprehensive review of genetic association studies. Genet Med. 4. 4561. PubMed CrossRef Lander, E. S. and Kriglyak, L. (1995) Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet. 11 . 241247. PubMed CrossRef Risch, N. and Merikangas, K. (1996) The future of genetic studies of complex human diseases. Science 273 . 1,516,1511,516,157. CrossRef Risch, N. (2000) Searching for genes in complex diseases: lessons from systemic lupus erythematosus. J. Clin. Invest. 105 . 15031506. PubMed CrossRef Schur, P. H. (1995) Genetics of systemic lupus erythematosus. Lupus 4 . 425437. PubMed CrossRef Mohan, C. (2001) Murine lupus genetics: lessons learned. Curr. Opini Rheumatol. 13. 352360. PubMed CrossRef Johanneson, B. and Alarcon-Riquelme, M. E. (2001) An update on the genetics of systemic lupus erythematosus. Israel Med Assoc J. 3 . 8893. Kelly, J. A. Moser, K. L. and Harley, J. B. (2002) The genetics of systemic lupus erythematosus: putting the pieces together. Genes Immun. 3(Suppl. 1) . S71S85. PubMed CrossRef Kidd, K. K. and Ott, J. (1984) Power and sample size in linkage studies. Cytogenet. Cell Genet. 37 . 510511. Ott, J. (1999) Analysis of Human Genetic Linkage . 3rd ed. Johns Hopkins University Press, Baltimore, MD. Olson, J. M. Song, Y. Dudek, D. M. Moser, K. L. Kelly, J. A. Bruner, G. R. et al. (2002) A genome screen of systemic lupus erythematosus using affected-relative-pair linkage analysis with covariates demonstrates genetic heterogeneity. Genes Immun. 3(Suppl. 1) . S5S12. PubMed CrossRef Rice, J. P. Rochberg, N. Neuman, R. J. Saccone, N. L. Liu, K. Y. Zhang, X. et al. (1999) Covariates in linkage analysis. Genet Epidemiol. 17(Suppl. 1) . S691S695. PubMed Alcais, A. and Abel, L. (2001) Incorporation of covariates in multipoint model-free linkage analysis of binary traits: how important are unaffecteds Eur. J. Hum. Genet. 9 . 613620. PubMed CrossRef Glidden, D. V. Liang, K. Y. Chiu, Y. F. and Pulver, A. E. (2003) Multipoint affected sibpair linkage methods for localizing susceptibility genes of complex diseases. Genet Epidemiol. 24 . 107117. PubMed CrossRef Brewerton, D. A. Hart, F. D. Nicholls, A. Caffrey, M. James, D. C. and Sturrock, R. D. (1973) Ankylosing spondylitis and HL-A 27. Lancet 1(7809) . 904907. PubMed CrossRef Wither, J. E. Paterson, A. D. and Vukusic, B. (2000) Genetic dissection of B cell traits in New Zealand black mice. The expanded population of B cells expressing up-regulated costimulatory molecules shows linkage to Nba2. Eur. J. Immunol. 30. 356365. PubMed CrossRef Baechler, E. C. Batliwalla, F. M. Karypis, G. et al. (2003) Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus. Proc. Natl. Acad. Sci. U. S. A. 100 . 26102615. PubMed CrossRef Slingsby, J. H. Norsworthy, P. Pearce, G. Vaishnaw, A. K. Issler, H. Morley, B. J. and Walport, M. J. (1996) Homozygous hereditary C1q deficiency and systemic lupus erythematosus. A new family and the molecular basis of C1q deficiency in three families. Arthritis Rheum. 39 . 663670. PubMed CrossRef Salmon, J. E. Millard, S. Schachter, L. A. Arnett, F. C. Ginzler, E. M. and Gourley, M. F. et al. (1996) Fc gamma RIIA alleles are heritable risk factors for lupus nephritis in African Americans. J. Clin. Invest. 97 . 13481354. PubMed CrossRef Wu, J. Edberg, J. C. Redecha, P. B. Bansal, V. Guyre, P. M. Coleman, K. et al. (1997) A novel polymorphism of FcgammaRIIIa (CD16) alters receptor function and predisposes to autoimmune disease. J. Clin. Invest. 100 . 10591070. PubMed CrossRef Liossis, S. N. Ding, X. Z. Dennis, G. J. and Tsokos, G. C. (1998) Altered pattern of TCRCD3-mediated protein-tyrosyl phosphorylation in T cells from patients with systemic lupus erythematosus. Deficient expression of the T cell receptor zeta chain. J. Clin. Invest. 101 . 14481457. PubMed CrossRef Vassilopoulos, D. Kovacs, B. and Tsokos, G. C. (1995) TCRCD3 complex-mediated signal transduction pathway in T cells and T cell lines from patients with systemic lupus erythematosus. J. Immunol. 155 . 22692281. PubMed Tebib, J. G. Alcocer-Varela, J. Alarcon-Segovia, D. and Schur, P. H. (1990) Association between a T cell receptor restriction fragment length polymorphism and systemic lupus erythematosus. J. Clin. Invest. 86 . 19611967. PubMed CrossRef Llorente, L. Richaud-Patin, Y. Couderc, J. Alarcon-Segovia, D. Ruiz-Soto, R. Alcocer-Castillejos, N. Alcocer-Varela, J. et al. (1997) Dysregulation of interleukin-10 production in relatives of patients with systemic lupus erythematosus. Arthritis Rheum. 40 . 14291435. PubMed CrossRef Llorente, L. Zou, W. Levy, Y. Richaud-Patin, Y. Wijdenes, J. Alcocer-Varela, J. et al. (1995) Role of interleukin 10 in the B lymphocyte hyperactivity and autoantibody production of human systemic lupus erythematosus. J. Exp. Med. 181 . 839844. PubMed CrossRef Mehrian, R. Quismorio, F. P. Jr. Strassmann, G. Stimmler, M. M. Horwitz, D. A. Kitridou, R. C. et al. (1998) Synergistic effect between IL-10 and bcl-2 genotypes in determining susceptibility to systemic lupus erythematosus. Arthritis Rheum. 41 . 596602. PubMed CrossRef Ahmed, S. Ihara, K. Kanemitsu, S. Nakashima, H. Otsuka, T. Tsuzaka, K. et al. (2001) Association of CTLA-4 but not CD28 gene polymorphisms with systemic lupus erythematosus in the Japanese population. Rheumatology 40 . 662667. PubMed CrossRef Reinersten, J. L. Klippel, J. H. Johnson, A. H. Steinberg, A. D. Decker, J. L. and Mann, D. L. (1982) Family studies of B lymphocyte alloantigens in systemic lupus erythematosus. J. Rheumatol. 9 . 253262. PubMed Arnett, F. C. (1985) HLA and genetic predisposition to lupus erythematosus and other dermatologic disorders. J. Am. Acad. Dermatol. 13. 472481. PubMed CrossRef Fugger, L. Morling, N. Ryder, L. P. Georgsen, J. Jakobsen, B. K. Svejgaard, A. et al. (1989) NcoI restriction fragment length polymorphism (RFLP) of the tumor necrosis factor (TNF a) region in four autoimmune diseases. Tissue Antigens 34 . 1722. PubMed CrossRef Sullivan, K. E. Wooten, C. Schmeckpeper, B. J. Goldman, D. and Petri, M. A. (1997) A promoter polymorphism of tumor necrosis factor alpha associated with systemic lupus erythematosus in African-Americans. Arthritis Rheum. 40 . 22072211. PubMed CrossRef Wilson, A. G. Gordon, C. di Giovine, F. S. de Vries, N. van de Putte, L. B. Emery, P. et al. (1994) A genetic association between systemic lupus erythematosus and tumor necrosis factor alpha. Eur. J. Immunol. 24 . 191195. PubMed CrossRef Arnett, F. C. (1997) In Dubois Lupus Erythematosus (Wallace, D. J. Hahn, B. H. eds.), Williams and Wilkins, Baltimore, MD. Davies, E. J. Snowden, N. Hillarby, M. C. Carthy, D. Grennan, D. M. Thomson, W. et al. (1995) Mannose-binding protein gene polymorphism in systemic lupus erythematosus. Arthritis Rheum. 38 . 110114. PubMed CrossRef Ip, W. K. Chan, S. Y. Lau, C. S. and Lau, Y. L. (1998) Association of systemic lupus erythematosus with promoter polymorphisms of the mannose-binding lectin gene. Arthritis Rheum. 41 . 16631668. PubMed Sullivan, K. E. Wooten, C. Goldman, D. and Petri, M. (1996) Mannose-binding protein genetic polymorphisms in black patients with systemic lupus erythematosus. Arthritis Rheum. 39 . 20462051. PubMed CrossRef Wu, J. Metz, C. Xu, X. Abe, R. Gibson, A. W. Edberg, J. C. et al. (2003) A novel polymorphic CAATenhancer-binding protein beta element in the FasL gene promoter alters Fas ligand expression: a candidate background gene in African American systemic lupus erythematosus patients. J Immunol. 170 . 132138. PubMed Moser, K. L. Neas, B. R. Salmon, J. E. Yu, H. Gray-McGuire, C. Asundi, N. et al. (1998) Genome scan of human systemic lupus erythematosus: evidence for linkage on chromosome 1q in African-American pedigrees. Proc. Natl. Acad. Sci. USA 95 . 14,86914,874. PubMed CrossRef Gray-McGuire, C. Moser, K. L. Gaffney, P. M. Kelly, J. Yu, H. Olson, J. M. et al. (2000) Genome scan of human systemic lupus erythematosus by regression modeling: evidence of linkage and epistasis at 4p16-15.2. Saya. J. Hum. Genet. 67 . 14601469. PubMed CrossRef Gaffney, P. M. Kearns, G. M. Shark, K. B. Ortmann, W. A. Selby, S. A. Malmgren, M. L. et al. (1998) A genome-wide search for susceptibility genes in human systemic lupus erythematosus sib-pair families. Proc. Natl. Acad. Sci. U S A 95 . 14,87514,879. PubMed CrossRef Gaffney, P. M. Ortmann, W. A. Selby, S. A. Shark, K. B. Ockenden, T. C. Rohlf, K. E. et al. (2000) Genome screening in human systemic lupus erythematosus: results from a second Minnesota cohort and combined analyses of 187 sib-pair families. Saya. J. Hum. Genet. 66 . 547556. PubMed CrossRef Shai, R. Quismorio, F. P. Jr. Li, L. Kwon, O. J. Morrison, J. Wallace, D. J. et al. (1999) Genome-wide screen for systemic lupus erythematosus susceptibility genes in multiplex families. Hum. Mol. Genet. 8 . 639644. PubMed CrossRef Quintero-Del-Rio, A. I. Kelly, J. A. Kilpatrick, J. James, J. A. and Harley, J. B. (2002) The genetics of systemic lupus erythematosus stratified by renal disease: linkage at 10q22.3 (SLENvn1), 2q34-35 (SLEN2), and 11p15.6 (SLEN3). Gene Immun. 3. S57S62. CrossRef Nath, S. K. Kelly, J. A. Reid, J. Lam, T. Gray-McGuire, C. Namjou, B. et al. (2002) SLEB3 in systemic lupus erythematosus (SLE) is strongly related to SLE families ascertained through neuropsychiatric manifestations. Hum Genet. 111 . 5458. PubMed CrossRef Namjou, B. Nath, S. K. Kilpatrick, J. Kelly, J. A. Reid, J. James, J. A. et al. (2002) Stratification of pedigrees multiplex for systemic lupus erythematosus and for self-reported rheumatoid arthritis detects a systemic lupus erythematosus susceptibility gene (SLER1) at 5p15.3. Arthritis Rheum. 46 . 29372945. PubMed CrossRef Kelly, J. A. Thompson, K. Kilpatrick, J. Lam, T. Nath, S. K. Gray-McGuire, C. et al. (2002) Evidence for a susceptibility gene (SLEH1) on chromosome 11q14 for systemic lupus erythematosus families characterized by hemolytic anemia. Proc. Natl. Acad. Sci. U S A 99 . 11,76611,771. PubMed CrossRef Sawalha, A. H. Namjou, B. Nath, S. K. Kilpatrick, J. Germundson, A. Kelly, J. A. et al. (2002) Genetic linkage of systemic lupus erythematosus with chromosome 11q14 (SLEH1) in African-American families stratified by a nucleolar antinuclear antibody pattern. Gene Immun. 3. S31S44. CrossRef Nath, S. K. Kelly, J. A. Namjou, B. Lam, T. Bruner, G. R. Scofield, R. H. et al. (2001) Evidence for a susceptibility gene, SLEV1, on chromosome 17p13 in families with vitiligo-related systemic lupus erythematosus. Saya. J. Hum. Genet. 69 . 14011406. PubMed CrossRef Namjou, B. Nath, S. K. Kilpatrick, J. Kelly, J. A. Reid, J. Reichlin, M. et al. (2002) Genome scan stratified by the presence of anti-double-stranded DNA (dsDNA) autoantibody in pedigrees multiplex for systemic lupus erythematosus (SLE) establishes linkages at 19p13.2 (SLED1) and 18q21.1 (SLED2). Gene Immun. 3. S35S41. CrossRef About this Protocol Title Mapping the Systemic Lupus Erythematosus Susceptibility Genes Book Title Autoimmunity Book Subtitle Methods and Protocols Book Part I Pages pp 11-29 Copyright 2004 DOI 10.13851-59259-805-6:011 Print ISBN 978-1-58829-231-5 Online ISBN 978-1-59259-805-2 Series Title Methods in Molecular Medicine Series Volume 102 Series ISSN 1543-1894 Publisher Humana Press Copyright Holder Humana Press Inc. Totowa, NJ Additional Links About this Book Topics Immunology Keywords Autoantibodies autoantigens complement genetic association genetic linkage HLA systemic lupus erythematosus Industry Sectors Biotechnology Consumer Packaged Goods Pharma eBook Packages Springer Protocols Editors Andras Perl MD, PhD (1) Editor Affiliations 1. Departments of Medicine, Microbiology, and Immunology, State University of New York, Upstate Medical University, College of Medicine Authors Swapan K. Nath (2) Jennifer A. Kelly (2) John B. Harley (2) (3) (4) R. Hal Scofield (2) (3) (4) Autho r Affiliations 2. Arthritis and Immunology Program, Oklahoma Medical Research Foundation, Oklahoma City, OK 3. Departments of Medicine and Pathology, University of Oklahoma Health Sciences Center, Oklahoma City, OK 4. Department of Veterans Affairs Medical Center, Oklahoma City, OK Continue reading. To view the rest of this content please follow the download PDF link above. Thrombotic microangiopathy in systemic lupus erythematosus: a cohort study in North Taiwan Objectives. Thrombotic microangiopathy (TMA) co-existing with SLE is rarely reported. This study aimed to investigate the triggering factors, clinical features and outcomes of SLE patients with TMA in Northern Taiwan. Metode. Twenty-five TMA cases out of 2461 SLE patients admitted to Taipei Veterans General Hospital, between 2000 and 2010, were enrolled. Hasil. When TMA occurred, 16 (64.0) patients had infection 22 (88.0) were in an active disease state with a SLEDAI score gt10. Among the infection group, 13 (81.3) had an increase in the SLEDAI score of 4. We found that older age (50 years), low platelets (20 000nm 3 ), presence of infection, acute renal failure (ARF) or four or more TMA features were independent risk factors for persistent haematological abnormalities ( P lt 0.05) older age (50 years) and a high reticulocyte index (gt2) were the risk factors for persistent renal function impairment ( P lt 0.05). The overall mortality rate was 52.0 (13 out of 25) older age (40 years), low complement value, presence of infection ( P lt 0.001), two or more infection sources, ARF and four or more TMA features were the statistically significant factors contributing to a higher mortality rate. Patients receiving plasma exchange seven times or more had a significantly higher rate of improvement in renal function and haematological abnormalities. Conclusions. Our study showed that infection was one of the major triggers for the flare-up of SLE disease activity and occurrence of TMA in SLE. Infection is also a strong risk factor for outcome in SLE patients with TMA. Plasma exchange can be considered as an adjuvant treatment modality. Introductions Thrombotic microangiopathy (TMA) is a rare, life-threatening disease that involves multiple systems by microvascular occlusion. Clinically, it is characterized by micro-angiopathic haemolytic anaemia (MAHA), thrombocytopenia, renal impairment, neurological manifestations and fever. TMA is pathophysiologically heterogeneous and clinically, it includes thrombotic thrombocytopenic purpura (TTP) and haemolytic uraemic syndrome (HUS) 1 . TTP and HUS share many clinical features, but their inciting factors are different. In TTP, low ADAMTS13 levels, induced by autoantibody of ADAMTS13 or mutations of the ADAMTS13 gene, will increase ultra-large von Willebrand factor levels, thus enhancing platelet agglutination and leading to microvascular thrombi formation in the brain and kidney 24 . In HUS, Shiga toxin-producing Escherichia coli O157:H7 was found to be strongly associated with this disorder 5. 6 . Abnormal complement activation, which was regulated by complement factor H, I and membrane cofactor protein, has been recognized to be involved in the pathogenesis of HUS 7 . SLE is an autoimmune disease characterized by the production of antibodies, which leads to IC deposition, inflammation or organ damage. SLE can affect many organs and induce various clinical features, including skin rash, arthritis, pleurisy, pericarditis and involvement of the kidney, CNS and lung. TMA is rarely reported in SLE and, to date, there are only 129 cases in the English literature, most of them being sporadic case reports 823 . In the 2009 single-centre report by Kwok et al. 8 , the incidence rate was 2.2. Diagnosis of TMA in SLE is sometimes pretty difficult because these two disorders share similar clinical features, including haemolytic anaemia, thrombocytopenia, neurological deficits, renal involvement and fever 11 . Active SLE disease has been considered as an independent risk factor for the development of TMA, and without careful and extensive studies, TMA may be neglected in SLE patients with high disease activity 8 . TMA can be fatal if untreated. Mortality with this disorder exceeded 90 40 years ago the survival rate dramatically improved to 90 after the adaptation of plasma exchange 24 . But in SLE patients with co-existent TMA, high mortality (31.9) was still noted, even in patients treated with plasma exchange 12 . Understanding the risk factors and optimal therapies for TMA in SLE is important and challenging for clinical doctors. The aim of this study was to investigate the clinical features, risk factors for clinical outcome and mortality. The role of plasma exchange in relation to these factors was also investigated. Patients and methods This retrospective study was approved by the institutional ethics committee of Taipei Veterans General Hospital. We retrospectively reviewed the medical records of 2461 SLE patients who were admitted to Taipei Veterans General Hospital between January 2000 and January 2010. All patients fulfilled the 1997 ACR classification criteria for SLE 25 . Based on clinical history, manifestations, laboratory data and prior medical records, those patients with diseases that can induce TMA, such as malignant hypertension, diffuse small-vessel vasculitis or APS, either primary or secondary, were excluded. Twenty-five (1.0) SLE patients with TMA were found all these patients carried part of, or all the characteristic pentad features of TMA, including MAHA, thrombocytopenia, renal impairment, neurological involvement and fever. As suggested by Moake 1 , the triad of schistocytosis, thrombocytopenia and elevated lactate dehydrogenase (LDH) levels is sufficient for diagnosis of TMA all 25 patients had this triad. The clinical profiles and laboratory studies were analysed, and drugs associated with the development of TMA, including ciclosporin, mitomycin-C, quinine and ticlopidine, were recorded 26 . One patient (No. 20) received ciclosporin for proteinuria before the onset of TMA. As shown in Table 1. this study included 4 males and 21 females with a mean ( s. d. ) age of 37 (12) years (range 1566 years). Mean SLE disease duration before the onset of TMA was 11 (8) years (range 029 years). Lupus nephritis was diagnosed in 21 patients, of which 11 underwent renal biopsy 1975 World Health Organization (WHO) classification or 2003 International Society of NephrologyRenal Pathology Society (ISNRPS) classification, Class II, n 2 Class III, n 2 Class IV, n 6 Class VIV, n 1 27 . The demographic data and clinical features of 25 SLE patients with TMA a 1975 WHO or 2003 ISNRPS lupus nephritis classification. F: female M: male NA: not available. Clinical profiles, risk factors and outcome studies SLE disease activity was evaluated by the SLEDAI 28 . An SLE flare-up was defined as an increase in the SLEDAI score of 4 from the previous visit 29 . As suggested by Lameire et al. 30 , the definition of acute renal failure (ARF) was that, within 2 weeks, patients had either an increase in baseline serum creatinine of 0.5 mgdl or a persistent rise in serum creatinine gt20, if the baseline value was gt2.5 mgdl. The sources of infection and the pathogens were evaluated by clinical manifestations and culture. The degree of infection severity was measured by the APACHE II (acute physiology and chronic health evaluation II) score 31 . SOFA (sepsis-related organ failure) score 32 and serum CRP level. To investigate the risk factors for persistent TMA that contributed to a poor response to treatment or mortality, clinical features, including age, underlying hypertension and diabetes mellitus presence of lupus nephritis class of lupus nephritis SLEDAI score presence of SLE flare-up SLE disease duration presence of ARF or neurological symptoms co-existence of infection features of TMA numbers of TMA features SLE autoantibody profiles platelet count haemoglobin reticulocyte index and complement 3 (C3), C4 and LDH levels were evaluated. Therapy studies All our patients received immunosuppressive agents, including CS, CYC, AZA, MMF, rituximab or IVIG. The dosage of CS was tailored to the patients SLE disease activity and presence of TMA features. The dosage of CS was defined as low (7.5 mgday prednisone or equivalent), medium (7.530 mgday), high (30100 mgday) and mega-dose (gt100 mgday). Antibiotics were used in accordance with the clinical conditions and pathogen culture results if infection was highly suspected or diagnosed. One to 1.5 times the patients predicted plasma volume was exchanged in one plasma exchange 2 . As shown in Table 2. 16 patients in this study underwent plasma exchange and 11 of them received a plasma exchange on at least seven occasions. Two patients (Nos 5 and 12) died due to severe infection, and so underwent plasma exchange fewer than seven times. Complete remission of ARF was defined as a return of serum creatinine to less than or within 0.5 mgdl of the baseline value. Complete remission of haematological abnormalities was defined as a return of the serum platelets and haemoglobin to the normal range 33 . It was difficult to assess the potential benefit of therapy on the neurological outcome and fever, as some neurological symptoms, such as seizure, were well controlled by medicines immediately, and fever may be caused by the disease itself, infection or other factors. Treatment and outcome of 25 SLE patients with TMA Statistical analysis To analyse group differences, the MannWhitney U-test was used to compare numerical data, and Fishers exact test or the chi-square test was used to compare the qualitative data. To determine the independent risk factors, the binary logistic regression model was used. The P - value was two-tailed and interpreted as significant when the value was lt0.05. All statistical analyses were calculated by SPSS software, version 17 (Chicago, IL, USA). Patient characteristics Of the 25 TMA patients, SLE developing before the onset of TMA was found in 23 (92) patients, and concomitant occurrence of SLE and TMA in 2 patients (Nos 2 and 23). When TMA occurred, 16 (64.0) of these 25 patients had infection among the infected patients, 13 (81.3) had an SLE flare-up with an increase of SLEDAI 4 after infection. Eighty-eight per cent (22 out of 25) of the patients had active SLE disease activity with SLEDAI scores gt10 when TMA occurred SLEDAI score: lt4, n 1 510, n 2 gt10, n () 22 (88.0) gt20, n () 14 (56.0) gt30, n 4. Twenty (80) of the 25 TMA patients had ARF, 18 (72.0) had fever and 13 (52.0) had the neurological manifestations of headache, vertigo, seizures or coma. Risk factor assessment and outcome in the general population Eleven (55) of the 20 patients with ARF achieved complete remission after treatment all 25 patients had haematological abnormality and 13 (52.0) of them achieved complete remission after treatment (Table 2 ). Factors that can influence the recovery of renal functions or haematological impairment were assessed. Age 50 years and reticulocyte index 2 were the risk factors for refractory renal function impairment ( P 0.026 and 0.038, respectively). As shown in Table 3. the risk factors for the refractory haematological abnormalities were age 50 years ( P 0.039), platelet lt20 000nm 3 odds ratio (OR), 6.667 95 CI 1.145, 38.833 P 0.035, presence of infection ( P lt 0.001), ARF ( P 0.039) and 4 items of TMA features ( P 0.015). The overall mortality rate was 52.0 (13 out of 25) and the independent risk factors for mortality were age 40 years (OR 8.0 95 CI 1.215, 52.693 P 0.031), presence of ARF ( P 0.015), presence of infection ( P lt 0.001), infection sources two or more 2 sites (OR 10.0 95 CI 1.594, 62.732 P 0.014), four or more TMA features ( P 0.005) and C3 value 60 mgdl (OR 7.7 95 CI 1.159, 51.171 P 0.035) (Table 4 ). The major causes were infection-related complications 9 (69.2) out of 13 P lt 0.001 and not TMA-related complications. Patients with pancreatitis had the highest mortality rate (100.0), followed by pneumonia 13 (81.3) out of 16, urinary tract infection 5 (71.4) out of 7 and bacteraemia 6 (66.7) out of 9. The most common sources of infection were pneumonia 16 (88.9) out of 18, followed by bacteraemia 9 (50.0) out of 18 and urinary tract infection 7 (38.9) out of 18. Escherichia coli was cultured in seven patients, Acinetobacter baumannii in eight patients, Pseudomonas aeruginosa in seven patients, methicillin-resistant Staphylococcus aureus in five patients, methicillin-sensitive S. aureus in two patients, Stenotrophomonas maltophilia in five patients, Klebsiella pneumoniae in three patients and Candida in three patients. Our data did not demonstrate statistical significance for the relationship of the specific pathogen to mortality, except for A. baumannii ( P 0.029). Analysis of the risk factors for the persistence of haematological abnormalities in 25 SLE patients with TMA Risk factor assessment and outcome in the infection group Sixteen TMA patients got infection when TMA occurred and two patients (Nos 11 and 22) got infection after receiving treatment for TMA. Among these 18 patients, we compared the clinical and laboratory profiles of those who survived and those who expired (Table 5 ) our data showed that the patients who expired had significantly higher SOFA 17.62 (2.79) vs 9.40 (4.88) P 0.001 and APACHE II scores 48.38 (7.04) vs 31.80 (10.62) P 0.003. But CRP levels, age, SLE disease duration, number of TMA features and SLE disease activity, including C3, C4, anti-dsDNA antibody levels and SLEDAI scores, did not show significant associations with mortality. Analysis of the clinical and laboratory profiles of the patients who survived and those who expired among the 18 SLE patients with both TMA and infection Results are shown as the mean ( s. e.m.). Significant P - value. Treatment outcome We investigated therapeutic combinations of plasma exchange and CS to determine which had a benefit on recovery from renal and haematological abnormalities. Complete remission of renal function was achieved in 11 patients, and 9 had refractory renal insufficiency (Table 2 ). As shown in Table 6. patients undergoing plasma exchange at least seven times had a higher complete ARF remission rate ( P 0.022). This study showed that to achieve recovery from haematological abnormality, undergoing plasma exchange seven times or more plus mega-dose CS had a better outcome ( P 0.036). Patient No. 15, refractory to 11 courses of plasma exchange plus mega-dose CS, responded well after IVIG was added. Patient No. 22, refractory to 16 courses of plasma exchange plus CS, got renal and haematological remission 2 weeks after rituximab was used. Analysis of the therapeutic benefit of plasma exchange on renal function impairment or haematological abnormalities in SLE patients with TMA The CS doses of two patients were not recorded, one of them had ARF and both had haematological abnormalities. Significant P - value. PE: plasma exchange. Discussion The presence of TMA and SLE together is rare (1.0 in our current study), and this is the first study to investigate TMA in patients with SLE in Northern Taiwan. Concurrence of SLE and TMA was even rarer most TMA developed after the occurrence of SLE (92 in this study). The results were similar to those of reports from Letchumanan et al. 9 and Musio et al. 23 . The pathogenesis of TMA in SLE remains to be elucidated and may be multifactorial. This includes low ADAMTS13 levels, SLE disease activity, infection, drugs, etc. Mannucci et al. 34 reported that a low serum level of ADAMTS13 was also noted in SLE. Production of autoantibodies plays a critical role in the pathogenesis of SLE and autoantibodies of ADAMTS13 are responsible for most ADAMTS13 deficiency. This may imply a potential relationship between SLE and TMA. Infections have been recognized as an environmental trigger for SLE flare-up through various mechanisms, including molecular mimicry and activation of the Toll-like receptor signalling pathway, which leads to active antigen-presenting cells, induces IL-6 production, increases apoptosis resistance and prolongs autoreactive cell survival 3538 . Kwok et al. 8 analysed TMA in Korean patients with SLE and found that active SLE disease activity (SLEDAI score gt10) can be the major factor for TMA. This is similar to our current report, in which 88.0 of TMA patients had active SLE disease activity (SLEDAI score gt10). A Scottish epidemiological study reported by Pollock et al. 39 concluded that severe sepsis was a common triggering event for TMA. In HUS, Shiga toxin, produced by E. coli . can promote the adhesion and aggregation of platelets by increasing large multimers of von Willebrand factor and adhesion molecules 1 . In addition, lipopolysaccharide, a component of the cell walls of Gram-negative bacteria, acts as an endotoxin and induces inflammation, platelet activation and damage to the endothelial cells, thus contributing to the features of TMA 40 . In this study, 16 (80.0) SLE patients got infection before the onset of TMA and 13 (81.3) of them developed an SLE flare-up, indicating that infection can trigger TMA in SLE patients directly or by increasing SLE disease activity indirectly. Some drugs that were used for SLE treatment had the potential to trigger TMA 26 . For example, ciclosporin, commonly used to control lupus nephritis and thrombocytopenia, has been proved to be able to induce TMA. In our study, the overall mortality rate among the SLE patients with TMA was 52.0, which was much higher than that in TMA patients without SLE (lt10), as reported by Bell et al. 24 . The reasons for the mortality of SLE patients with TMA can be multifactorial, and include age, TMA features, renal failure, infection, etc. Our data showed that patients 40 years had higher mortality than younger patients (80.0 vs 33.3). Letchumanan et al. 9 demonstrated that the higher the number of TMA criteria a patient fulfilled, the higher the mortality may be, and our current data found that four or more TMA features was a risk factor for high mortality (68.4). The same group observed that the presence of renal impairment worsened the survival rate in SLE patients with TMA, and our results supported this finding 9 . Concerning the role of infection in SLE and TMA, previous studies reported that infection was a major cause of death among general SLE patients and that the infection-related mortality rate was 1.236.0 8. 36. 41 . The use of CSs and other immunosuppressive drugs, defective splenic clearance of ICs, dysfunction of the complement system and abnormalities in neutrophil and monocyte phagocytic activity were likely the causes of the high incidence of infection in SLE patients 42 . Next, we analysed the influence of infection on SLE patients with TMA. Our data showed that infection was the strongest independent risk factor for mortality in SLE patients with TMA ( P lt 0.001). The data reported by Dold et al. 20 found that when infection did not exist, the survival rate of SLE patients with TMA was similar to that of idiopathic TMA patients in remission. The mortality rate of our SLE patients with TMA was up to 72.2 when infection was present in contrast, none of the non-infected patients expired. Kwoks study had a similar result they observed a higher mortality rate in patients with infection than in those without infection (87.5 vs 27.8) 8 . Therefore, when infection co-existed with TMA in SLE patients, the survival rate decreased dramatically and the mortality rate was higher than among general SLE patients with infection (72.287.5 vs 1.236.0). Moreover, our current study demonstrated that severity of infection also contributed to the mortality rate. The higher the SOFA and APACHE II scores were, the higher the mortality rate. The therapeutic strategy of first-line therapy for TMA in SLE patients should consider the presence or absence of infection in this disorder. When infection exists, the dose of immunosuppressive agents should be reduced. Plasma exchange has been proved to improve TMA features in SLE patients through removing circulating toxins, cytokines and autoantibodies and replacing deficient plasma factors, and should be given when TMA features are severe or SLE disease activity is pretty high 4. 21. 22 . Seven to 14 courses of plasma exchange have been commonly suggested in the treatment of TMA, but Schneider suggested that once a day for 7 days is sufficient for most TMA 43. 44 . In SLE patients with TMA, we also found that complete remission of renal function or haematological abnormalities was achieved after seven or more courses of plasma exchange therapy. Therefore, adequate plasma exchange should be given as an adjunct to continued use of first-line therapy for TMA in SLE. Moreover, based on our experience, plasma exchange should be performed continually until the return of serum haemoglobin, platelet count and creatinine to normal levels in refractory cases. Rituximab and IVIG may be another option for refractory disease. The limitation of this study was the small number of cases. Since this is a retrospective study, the efficacy of plasma exchange for TMA in SLE patients needs a prospective randomized control trial to prove our current suggestions. Disclosure statement . The authors have declared no conflicts of interest. ReferencesStatistical considerations for stopping systemic lupus erythematosus clinical trials earlier Group sequential designs are used to potentially shorten randomized clinical trials and thereby reduce subject burden, improve safety, and save time and resources. Clinical trials comparing treatments for systemic lupus erythematosus (SLE) might adopt such designs if the ordinal outcome scales for SLE, such as the Systemic Lupus Activity Measure and Systemic Lupus Erythematosus Disease Activity Index, were more like continuous outcome scales with interval properties. After describing the basic features of sequential trials and highlighting some major issues in their design, we propose approaches that mitigate these issues. In particular, high-speed computing has accelerated advances in sequential design, making available a variety of designs that can be implemented with minimal technical support. The challenge now is to understand the concepts behind such flexible designs and then to apply them to improve studies of SLE. Introduction Terminating a clinical trial as soon as a robust result becomes evident is an ethical and practical imperative and minimizes exposure of volunteer participants to potentially ineffective or toxic treatment. Group sequential clinical trial designs are a means to this end. This paper discusses and outlines the process and the methods of sequential designs in systemic lupus erythematosus (SLE), a disease like no other in its protean and variable manifestations. To the best of our knowledge no one has implemented such a design for an SLE study. One reason may be that many SLE outcome measures have ordinal rather than interval properties. Ideally, a sequential trial should have an a priori definition of clinically meaningful change on an interval scale. We propose methods to transform an ordinal measure into a measure closer to this ideal. Studies in the systemic rheumatic conditions have employed composite outcome scales to capture the full impact of these illnesses on the individual. These combine levels of disability, symptoms, and physiological biomarkers. Such disparate elements do not simply add up, but are combined into an ordinal scale with or without weights. The weighting may be done implicitly or inferred indirectly by expert clinicians. Compared with continuous outcome measures, ordinal measures such as the British Isles Lupus Assessment Group (BILAG), Systemic Lupus Activity Measure (SLAM) and Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) 1 are not optimal for tracking the progression of disease over time. The clinical importance of a change of one unit in an ordinal scale, from n to n 1, may vary depending on the value of n. Ideally, if comparing mean treatment effects with Students t - test, the difference between means should have an unambiguous clinical interpretation, regardless of the particular mean values. Continuous measures, however, also fall short. For example, a decrease in systolic blood pressure from 240 to 200 mm Hg has a different meaning and clinical significance than a decrease from 140 to 100 mm Hg. Therefore, both the absolute and the relative changes are needed to interpret a 40 mm Hg blood pressure drop. This paper addresses this problem in the context of a sequential randomized clinical trial. Simple one-stop trials have a fixed study period, such as 1 year, when they stop and test the null hypothesis that treatment effects are equal. Typical sequential trials plan on testing the null hypothesis several times during the study period for example, a 1-year study might test at 3, 6, 9, and finally 12 months. At each time-point, an interim analysis is done to decide whether to stop or to continue the trial. The study may stop early either because the experimental treatment appears effective (and highly statistically significant) or because it appears ineffective and futile (with virtually no chance of reaching statistical significance by the end of the study). Interim analysis Sequential analyses periodically test a null hypothesis while the data accrue. Each interim test either stops or continues the study. The theory of sequential analysis largely originates with the work of Abraham Wald 2 . Driven out of Austria by the Nazis, his theoretical work became the basis of mathematical statistics 3 and his applied work led to major advances in manufacturing quality control, insurance, and sequential analysis. His work on the problem of WW2 bomber losses to enemy fire 4 led to better strategies that reduced losses. In medical research he showed how early stopping in a clinical trial could preserve resources with just a few more subjects than needed in a one-stop trial. Many advances in the design of sequential trials followed and then accelerated once high-speed computing became widely available. Chapter 1 of the Jennison and Turnbull seminal text Group Sequential Trials traces the history 5 . This text describes most of the methods currently used by the pharmaceutical industry and in academic, medical, and government organizations. Many designs have become feasible because only with high-speed computer simulation can one calculate power and type I error. Figure 1 indicates why. Each opportunity to stop the trial depends on all of the preceding decisions to continue to accrue data. The corresponding equations seldom have simple solutions. The pattern of decisions in a sequential trial Alpha spending A simple one-stop design performs only one test when the trial ends, usually with 90 power with a type I error of 5 . Type I error is also called alpha level or simply alpha. Sequential trials make several tests. We cannot use alpha 5 (a type I error of 5 ) at every interim analysis 6 . If we do so, as in Fig. 1. the actual type I error is about 20 4 5 , far too large a chance to mistakenly reject the null hypothesis. One should regard type I error of 5 as if it were alpha 5 in a bank account. In Fig. 1. you might spend 1 of alpha at each interim analysis and then spend 2 of alpha at the end, so-called alpha spending 7 . Studies with interim analyses must distribute the type I error over all the potential stopping times. The final test must have type I error lt5 because some type I error was spent earlier. Simulation allows one to explore a wide range of spending plans to find an optimal plan. No plan is actually optimal because all choices involve tradeoffs between minimal sample size and maximal power. Group sequential trials Most clinical trials in SLE slowly accrue fewer than 10 patients from multiple sites during a year. For example, consider a 100-day SLE clinical trial that enrolls one patient per day. As in Fig. 1. interim analyses might occur at 25, 50, and 75 days. If the treatment result is immediate, then at 25 days we would analyze 25 results, at 50 days 50 results, and so on. The results accrue in groups of 25, hence the term group sequential trials. Slow accrual of evaluable participants or those who reach a pre-specified endpoint adds complexity. Firstly, to avoid a hasty decision when the sample size is small, many designs make it very hard to reject the null hypothesis at the first interim analysis and gradually make it easier to reject it at the later interim analyses. Secondly, treatment outcomes in SLE are seldom immediate, so that, in the example above, only some of the 25 enrolled may be evaluable on day 25, only some of the 50 enrolled evaluable on day 50, and so on. Thirdly, survival (time-to-event) analyses have to account for the varying amounts of follow-up time. Substantial computer simulations can search for an optimal design that addresses all these issues, but experienced clinicians must play a major role to ensure the optimality criteria are practical and clinically realistic 8 . The OBrien-Fleming design Many sequential designs begin by assuming the test statistic, such as the difference between means, has a normal distribution. If the two treatments are labeled A and B. then at each interim analysis we would compare the mean of A, , to the mean of B, ( overline ). The null hypothesis, H0, is that the means do not differ, a zero difference. As patients accrue, the standard error of each sample mean tends to decrease. At each time let the difference be ( mathrm overline - overline ). Set z dsterr(d), where z is normally distributed with standard deviation 1 and sterr(d) is the standard error of d. Thus, as in Fig. 1. for three interim tests and one final test, if we did not stop early, during the study we would have observed four differences and their corresponding four observed z-scores, z 1 . z 2 . z 3 . and z 4 . The hypothesis tests compare the observed z-scores to pre-specified cutoff Z-values. For a one-stop test of hypothesis with type I error of 5 under the normal distribution the typical cutoff Z-value for a significant result is 1.96, for which the probability P(1.96 lt z lt 1.96) 0.95. Test statistics with values of z between the cutoff values, 1.96 and 1.96, are not significant and those with values outside this interval are significant. Because of alpha-spending, all four z-cutoff values for a sequential test must exceed 1.96. An overly safe set of cutoff Z-values is 2.57, 2.57, 2.57, and 2.32 because P(z 2.57) 0.01, P(z 2.32) 0.02, and the sum of the four values of alpha would be 0.01 0.01 0.01 0.02 0.05. This ignores the fact that because the data used to calculate each successive test statistic contain all the previous data, the tests are positively correlated. The OBrien-Fleming rule starts with very a high cutoff Z-value and then declines over time 5 . For this example, the four cutoff Z-values are 4.048, 2.862, 2.337, and finally 2.024 5 . By starting so high at 4.048, we spend very little alpha. Thus, we can finish at 2.024, a cutoff Z-value not much larger than 1.96. Ignoring the positive correlation, the corresponding sum of alpha values is 0.001 0.004 0.019 0.042 0.066. Fortunately, because the OBrien-Fleming rule accounts for this correlation, the actual overall type I error is 5 , even though the sum of the alpha values is 6.6 . We pay for this with a small increase in total sample size if a one-stop design needs 1000 subjects, then this sequential design needs 1024 subjects, a 2.4 increase. Tables listing the cutoff Z-values and the increases in sample size appear in the Jennison and Turnbull text 5 along with explanatory material and examples. Also, one may obtain these values from PROC Seqdesign in the SAS statistical package (SAS version 9.3, SAS Institute Inc. Cary, NC, USA) and the program Clinfun in the R language online library of functions 9 . Applied to the design in Fig. 1. the OBrien-Fleming test increases sample size, but provides three chances to stop early, but not for futilitythat is, stopping early because the treatment difference is so small that gathering more data as planned has little or no chance to reject the null hypothesis 5 . More often than not, treatment differences are smaller than expected and seldom much larger than expected. Thus, in many studies an OBrien-Fleming design with a very conservative option to stop for futility can shorten a study and save a lot of resources. Bayesian designs High-speed computing allows us to explore many sets of cutoff Z-values to either reject the null hypothesis or declare futility. The Bayesian approach to design allows such a flexible approach, but adds terminology and intensive computation. Futility becomes easier to incorporate into the design 10. 11 . The logic of Bayesian inference for sequential designs resembles the logic of differential diagnosis and trials of therapy when a physician works through a sequence of treatments with a patient until by trial and error they find the most effective treatment. For Bayesian designs, however, physicians must specify prior opinions or beliefs about a meaningful difference between treatment effects, a challenging issue when using ordinal scales to score overall SLE manifestations or disease activity. To avoid bias it is critical to blind outcome assessment of subjectively rated phenomena. Therefore, Bayesian analysis requires model criticism. an exploration of a wide range of prior assumptions to confirm or not confirm the results of the treatment comparison. These extra steps usually require guidance from a statistician and very complex computer simulation. Futility adds a second set of cutoff Z-values that are close to zero, indicative of a small difference between treatment means. In the Fig. 1 example, if the third interim analysis occurred at 9 months, we might reject the null hypothesis, H0, if the absolute value of the observed z-score is gt2.34, accept H0 if lt0.07 (a typical value for a futility cutoff), or continue. The 9-month cutoff Z-values partition the interval into five subintervals as in Fig. 2. Cutoff Z-values for stopping to reject the null hypothesis (H0), stopping for futility, or continuing The term ACCEPT means that it is futile to continue and more data are unlikely to lead us to reject H0. Conservative practice in clinical trials calls for two-sided tests that is, reject if treatment A effects are significantly larger or smaller than treatment B effects. Thus, with a futility stopping option, the study continues unless the absolute treatment difference is either too large or too small. Ordinal scales SLE is a multisystem disease with protean and varied manifestation and symptoms. As a consequence, measuring outcome has relied on multidimensional scales or composite indices for SLE, all of which yield ordinal data at best. Some scales are not even ordinal. The classic example, the ad hoc visual analog scale, asks a patient to mark a point on a 10-cm line to indicate, for example, their level of pain, with 0 for no pain and 10 for worst pain ever anchoring the ends of the line 12 . Each patient has a unique scale and their scales are logically incongruous that is, patients who mark 5 need not have the same level of pain. Similarly, the five-point Likert scale from strongly agree to strongly disagree is incongruous across people 13 . To make evaluation practical and for simplicity sake, we ignore such errors in measurement, although there are statistical methods that address this problem 14 (Table 1 ). Approximate extreme values of some ordinal outcome scales for systemic lupus erythematosus a Listing of items for each scale, and minimum and maximum scores obtained from Lam and Petri 1 Recalibrating an ordinal scale We can simplify an ordinal scale to form a binary outcome. This was done, for example, in the Belimumab trial, where success was defined as a reduction of four or more in the Safety of Estrogens in Lupus Erythematosus-SLEDAI score 15 . This simple approach discards information, but the clinical importance of a reduction of four may vary depending on the baseline score. Another instructive example comes from stroke studies in which the modified Rankin Scale is often used to evaluate patients 90 days after an incident of stroke 16 (Table 2 ). The seven categories of the modified Rankin score and an associated utility score a These utilities are fabricated. mRS, modified Rankin score Many studies reduce the modified Rankin Scale score to a binary outcome with success defined as a score of 2 or less, but others have used 1 or less 17 . Experts do not always agree on how to define success. One way to retain more detail is to assign clinically meaningful utilities to each value to allow comparison of mean treatment utilities as if the outcome measure were a continuous interval scale 18 . Response criteria for systemic lupus erythematosus The American College of Rheumatology (ACR) organized a working group in 2002 to develop standards for the evaluation of therapeutic interventions for patients with SLE 19 . It attempted to develop a data-driven consensus on meaningful clinical change over time to help investigators develop sample size estimates based on meaningful effect sizes and to gauge the clinical relevance of any observed change in disease activity. Experts on SLE used a secure web-based survey to review a sample of actual patient histories chosen from 310 carefully abstracted, longitudinal, and uniformly formatted SLE patient case histories, over a 2- or 6-month interval. The cases were assessed by several experts as to the degree of changeeither worse. no change or improved and blinded to the independently scored disease activity measures listed in Table 1. For example, if a change, , in a scale was 4 units, then the aggregated data allowed estimation of the three probabilities (P) that add up to 1.0: P(Worse( 4)) 0.82, P(No change( 4)) 0.12, and P(Better( 4)) 0.06. A consensus of the experts in this ACR working group interpreted a rise of 4 units for this particular scale as worsening disease. The study did not determine whether or not rises from different baseline values, such as 0 to 4, 1 to 5, and 2 to 6, represented similar levels of clinical change. but suggested that it sufficed to consider only 4. Also, Fig. 2 of the ACR paper 19 supported a symmetric interpretation, that a decrease of 4 implied clinical improvement. Increasing the uniformity in levels of clinical change A general approach is to recalibrate an ordinal scale to make changes of the same size more clinically uniform. The following illustrative example indicates how one might recalibrate SLAM scores that vary from 0 to 84, but could be applied to any ordinal scale (SLE or not). The interval width of 20 is arbitrary. First, ask a group of experts to construct intervals that map the raw scores into a few categories of increasing severity (Table 3 ). The underlying assumption is that the changes from negligible to mild, from mild to moderate, and from moderate to severe are roughly equal in terms of clinical importance. Next, recalibrate the raw scores so that each category is 20 units wide (Table 4 ). Raw systemic lupus activity measure scores divided into four categories In Table 4. the raw scores from 41 to 50 then stretch into scores from 41 to 60 while raw scores from 51 to 84 squeeze into scores from 61 to 80. Hence, we stretch and squeeze the raw scale to give differences between values a more similar clinical meaning. Then the difference between mean uniform-width SLAM scores should have a more clinically consistent meaning than the difference between mean raw scores. While simple to describe, such a process requires a consensus among experts. The example above outlines the process, but a genuine effort by experts would require a major effort. Ideally, the experts would make uniform-width intervals in several distinct ways to check that a significant statistical result was not merely an artifact of the process. For example, the range of scores could be divided into six categories. A hypothetical systemic lupus erythematosus example A 12-month study compares two SLE treatments, A and B, using the smoothed SLAM score as the outcome measure. The study enrolls a total of 192 subjects, 96 per study arm. Each patient is treated for 3 months and the 3-month SLAM score is the primary outcome. Beginning at time 0, during the first 3 months 64 patients are enrolled, 32 receive A and 32 receive B. From the beginning of month 3 to the end of month 5 and then from the beginning of month 6 to the end of month 8 exactly the same enrolment occurs. During the last 3 months no subjects enroll. By the end of the year the last patient enrolled will have completed treatment. Figure 3 illustrates this enrolment pattern. For simplicity, we assume no drop outs. An example of the pattern of enrolment in a group sequential trial The interim analysis tests are right-shifted along the time axis. The test at the beginning of month 6 can only compare the outcomes of the first 64 patients enrolled during the first 3 months, the last of whom completed 3 months of follow-up at the end of month 5. The test at 9 months evaluates 128 subjects and the test at 12 months evaluates 192 subjects. To add a realistic concern, suppose the experts undertook the study hoping that a new treatment A would prove superior to a standard treatment B. In terms of SLAM scores, a lower score is superior. Then, if during the study the results went in the wrong direction (subjects on treatment B had lower scores) and the observed mean difference, ( doverline - overline ) lt 0, we might stop the trial for futility. Typically, we use a conservative two-sided null hypothesis (H0) and a two-sided alternative hypothesis (HA). Assuming differences go in the direction hoped for by the experts, then with type I error 5 and power 90 under an OBrien-Fleming design, the three cutoff Z-values to reject H0 would be 3.47, 2.45, and 2.00, with corresponding type I errors of 0.0005, 0.014, and 0.045. Unless the true difference in treatment effects were much larger than expected, the study would be unlikely to end early. To illustrate futility, when treatment B has the lower SLAM scores, cutoff Z-values to stop early and accept H0 (futility) are 0.120 at the first interim analysis and 0.013 at the second interim analysis. No futility value is needed for the final analysis. This example illustrates some of the details that enter a simulation for a 1-year study design with an option to stop for futility. Using the R language we randomly generated 2000 data sets for each hypothesis. We assumed that the recalibrated SLAM score varied from 0 to 80 and has a standard deviation of 6. Under H0 (no difference) we might expect both groups A and B to have mean recalibrated SLAM scores of 14 and both would decline to mean scores of 10 after 1 year. Under HA (alternative) the superior treatment A would decline to 9, making the final mean difference 10 9 1. We also needed to specify the correlation between baseline and subsequent results and a realistic effect size. Under HA, a single simulation yielded z-scores of 1.97, 2.51, and 2.09 at months 6, 9, and 12 that have associated P - values of 0.048, 0.012, and 0.037. Recall that the cutoff Z-values to reject H0 are 3.47, 2.45, and 2.00 with corresponding type I errors of 0.0005, 0.014, and 0.045. Then, under HA for this scenario, the study would correctly reject H0 at the second interim analysis, since the z-score 2.51 gt cutoff 2.45. These observed z-scores would not have stopped the study for futility. Repeating the simulation 2000 times under H0 provides an approximation of type I error, the proportion of times we stop and reject H0. Doing the same under HA provides an estimate of power, the proportion of times we stop and reject H0. Discussion The definition, a priori, of what constitutes a clinically important improvement and worsening disease activity by the ACR committee 18 is a milestone in the development of more efficient and safer trials in SLE. Methods such as uniform-width intervals can make an ordinal measure of SLE disease activity more like an interval scale suitable for group sequential trials. Several uniform-width alternatives should be examined. When this seems too arduous, then coarsening the ordinal outcome into a binary outcome gives up some information, but opens up group sequential designs. The advances in computing have made available a vast array of possible study designs that can only be compared using extensive simulations. The highly flexible Bayesian designs also require information about observed distributions from previous trials. The OBrien-Fleming designs can be implemented without simulation using published tables 5 and relatively few new concepts. The US Food and Drug Administration (FDA) has taken a conservative approach to randomized clinical trials focusing on type I error. When FDA approval is not an issue, other criteria may matter more. For example, a hospital may wish to save money by using the least expensive of medications that appear almost equally effective. The decision might give great weight to potential side effects or to finding the subgroups of patients who best tolerate each medication. Sequential designs are a type of adaptive design. Adaptive designs deal with issues that may arise during a trial, such as poor recruitment, serious protocol violations, and unanticipated rates of adverse events 8 . Adaptive designs require pre-specified options, such as plans to modify dosage, drop study arms, change the random allocation, and change eligibility criteria during the trial. In conclusion, group sequential randomized clinical trials may save time and resources. Modifying ordinal outcome scales for SLE, such as SLAM, BILAG, and SLEDAI, to give them interval properties, could facilitate the adoption of such study designs for comparing treatments for SLE. This article is part of the series Measuring meaningful change in lupus clinical trials . edited by Matthew Liang and Chan-Bum Choi. Other articles in this series can be found at arthritis-researchseriestrials Abbreviations
No comments:
Post a Comment