18th Expert Committee on the Selection and Use of Essential Medicines (Accra, Ghana, 21 to 25 March 2011) REVIEW OF THE

EVIDENCE COMPARING INSULIN (HUMAN OR ANIMAL) WITH ANALOGUE INSULINS FEBRUARY 2011

While there are a number of new trials (post 2007) available assessing analogue insulins in comparison to regular human insulin, there is little data available within these trials to update previous meta-analyses. For the analyses that could be updated there was little change in results, which demonstrated little or no differences between analogue and regular human insulins. In analyses that indicated statistically significant advantages for analogue insulins for glycaemic control, the differences between analogues and regular human insulin remain very small (ie 0.09%) and do not constitute clinically important differences. Consequently, the available evidence indicates that analogue insulins have no advantage over regular human insulins for the outcome of glycaemic control. Regarding the occurrence of hypoglycaemic events, analogue insulins appear to have statistically significant advantages compared to regular human insulins, but these advantages are not consistent across types of insulin (rapid or long-acting) or types of diabetes, and the clinical importance of these differences is not clear. In addition, many trials which demonstrated a difference between analogue insulin and regular human insulin for the occurrence of hypoglycaemia excluded patients with a history of recurrent major hypoglycaemia (Singh et al., 2009), therefore it may not be appropriate to assume such advantages will be observed across all patients. It should be noted that the trials assessing analogue insulins are of relatively low quality, given the lack of blinding of participants and outcome assessors, with resultant potential for bias with patient-reported outcomes (ie hypoglycaemia). Also, most trials are linked to the pharmaceutical industry, suggesting that there may be some degree of publication bias. Costeffectiveness estimates of analogue insulins vary widely, from just over 500 to greater than 412,000 per QALY gained. Estimates indicating cost-effectiveness are generally specific to a particular population and regimen, however the broader and more comprehensive analyses indicate that analogue insulins appear to lack cost-effectiveness. Recent reviews of the potential link between analogue insulin use and cancer raise a number of methodological and statistical questions and indicate that further evidence is required before firm conclusions can be drawn. There remains a lack of evidence addressing longer-

term outcomes of diabetes such as mortality and long-term complications. Given the lack of clear benefits for analogue insulins for glycaemic control as well as the inconsistent and clinically debatable benefits for occurrence of hypoglycaemia, along with concerns about trial quality, the current evidence does not indicate a strong advantage for analogue insulins compared to regular human insulin for both Type 1 and 2 diabetes.

Translate: Meskipun ada beberapa percobaan baru (pasca 2007) tersedia menilai insulin analog dibandingkan dengan insulin manusia biasa, ada sedikit data yang tersedia dalam uji coba ini untuk memperbarui meta-analisis sebelumnya. Untuk analisis yang dapat diperbarui ada sedikit perubahan dalam hasil, yang menunjukkan sedikit atau tidak ada perbedaan antara insulin analog dan human insulin reguler Dalam analisis yang menunjukkan keuntungan yang signifikan secara statistik untuk insulin analog untuk kontrol glikemik, perbedaan antara analog dan human insulin reguler tetap sangat kecil (yaitu 0,09 %) dan bukan merupakan perbedaan klinis yang penting. Akibatnya, bukti yang ada menunjukkan bahwa insulin analog tidak memiliki keuntungan lebih dari human insulin reguler untuk hasil kontrol glikemik. Mengenai terjadinya peristiwa hipoglikemik, insulin analog tampaknya memiliki keunggulan yang signifikan secara statistik dibandingkan dengan insulin manusia biasa, tapi keuntungan ini tidak konsisten di seluruh jenis insulin (rapid atau long-acting) atau tipe diabetes, dan pentingnya klinis perbedaan ini tidak jelas. Selain itu, banyak percobaan yang menunjukkan perbedaan antara insulin analog dan human insulin reguler untuk terjadinya hipoglikemia dikecualikan pasien dengan riwayat hipoglikemia mayor berulang (Singh dkk, 2009) , oleh karena itu mungkin tidak tepat untuk menganggap keuntungan seperti akan diamati di semua pasien.

Perlu dicatat bahwa percobaan menilai insulin analog adalah secara kualitas relatif rendah, Perkiraan efektivitas biaya dari insulin analog bervariasi. Perkiraan menunjukkan efektivitas biaya umumnya spesifik untuk populasi tertentu dan cara hidup, namun pada analisis yang lebih luas dan lebih komprehensif menunjukkan bahwa insulin analog tampaknya kurang efektivitas biaya . Ulasan terbaru dari hubungan potensial antara penggunaan insulin analog dan kanker memunculkan sejumlah pertanyaan metodologis dan statistik dan menunjukkan bahwa buktibukti lebih lanjut diperlukan sebelum kesimpulan dapat ditarik . Masih ada kurangnya bukti menangani hasil jangka panjang dari diabetes seperti kematian dan komplikasi jangka panjang. Mengingat kurangnya manfaat yang jelas bagi insulin analog untuk kontrol glikemik serta manfaat klinis konsisten dan diperdebatkan untuk terjadinya hipoglikemia , bersama dengan kekhawatiran tentang kualitas persidangan , bukti saat ini tidak menunjukkan keuntungan yang kuat untuk insulin analog dibandingkan dengan insulin manusia biasa untuk diabetes tipe 1 dan 2 .

Short-Acting Insulin Analogues vs. Human Insulin for Diabetes

MICHAEL SCHOOFF, M.D., and KRISTA EHLERS, M.D., Clarkson Family Medicine, Omaha, Nebraska Am Fam Physician. 2005 Sep 1;72(5):805-807. http://www.aafp.org/afp/2005/0901/p805.html

Insulin Types and Duration of Action

INSULIN TYPE ONSET (MINUTES) Immediate-acting Insulin lispro solution Insulin aspart solution Rapid-acting Regular Prompt insulin zinc solution Intermediate-acting Isophane NPH Lente 60 to 150 7 to 15 24 insulin suspension 60 to 150 4 to 12 24 30 to 60 60 to 90 2 to 4 5 to 10 8 to 12 12 to 16 15 15 0.5 to 1.5 1 to 3 2 to 5 3 to 5 PEAK (HOURS) DURATION (HOURS)

Long-acting Ultralente Lantus 240 to 480 60 10 to 30 5 20 to 36 24 or more

NPH = neutral protamine Hagedorn.

Traditional methods of insulin therapy use neutral protamine Hagedorn (NPH) and regular insulin in combination at a ratio of 70:30 taken before morning and evening meals. NPH often is moved to bedtime with regular insulin still taken before the evening meal.2 Another option is prandial insulin with meals in addition to basal insulin once or twice per day. Insulin pumps commonly are used to give continuous subcutaneous insulin.2 One of the benefits of the newer short-acting analogues is their fast onset, which allows patients to take insulin immediately before eating instead of 30 minutes in advance. Also, they enable patients to dose insulin after eating, if needed.1 Disadvantages of short-acting insulin are the high cost and potential long-term side effects. Short-acting insulins are approximately double the cost of regular insulin.2 There also is some concern about the potential mitogenic effects of analogues, and more information is needed on the long-term safety and side effects of these drugs.1

Cochrane Abstract
Background. In short-acting insulin analogues the dissociation of hexamers is facilitated, achieving peak plasma concentrations about twice as high and within approximately one half the time compared with regular human insulin. According to these properties, this profile resembles the shape of patients without diabetes more than that of regular human insulins. Despite this theoretical superiority of short-acting insulin analogues over regular human insulin, the risk-benefit ratio of short-acting insulin analogues in the treatment of patients with diabetes is still unclear. Objectives. To assess the effect of treatment with short-acting insulin analogues versus regular human insulin. Search Strategy. A highly sensitive search for randomized controlled trials combined with key terms for identifying studies of short-acting insulin analogues versus regular human insulin was performed using the Cochrane Library (issue 4, 2003), MEDLINE, and EMBASE. Date of last search was December 2003. Selection Criteria. The authors1 included randomized controlled trials with diabetic patients of all ages that compared short-acting insulin analogues with regular human insulin. Intervention duration had to be at least four weeks. Data Collection and Analysis. Trial selection as well as evaluation of study quality was done by two independent reviewers. The quality of reporting of each trial was assessed according to a modification of the quality criteria as specified by Schulz (JAMA 1995;273:40812) and Jadad (Controlled Clin Trials 1996;17:112). Primary Results. Altogether 7,933 participants took part in 42 randomized controlled studies. Most studies were of poor methodological quality. In patients with type 1 diabetes, the weighted mean difference (WMD) of A1C was 0.1 percent (95% confidence interval [CI], 0.2 to 0.1) in favor of insulin analogue, whereas in patients with type 2 diabetes the WMD was estimated to be 0.0 percent (95% CI, 0.1 to 0.1). In subgroup analyses of different types of interventions in patients with type 1 diabetes, the WMD in A1C

was 0.2 percent (95% CI, 0.3 to 0.1) in favor of insulin analogue in studies using continuous subcutaneous insulin injections (CSII), whereas for conventional intensified insulin therapy (IIT) studies the WMD in A1C was 0.1 percent (95% CI, 0.2 to 0.0). The WMD of the overall mean hypoglycemic episodes per patient per month was 0.2 percent (95% CI, 1.2 to 0.9) and 0.2 (95% CI, 0.5 to 0.1) for analogues in comparison with regular insulin in patients with type 1 and type 2 diabetes, respectively. For studies in patients with type 1 diabetes, the incidence of severe hypoglycemia ranged from zero to 247.3 (median 20.3) episodes per 100 person-years for insulin analogues and from zero to 544 (median 37.2) for regular insulin. In patients with type 2 diabetes, the incidence ranged from zero to 30.3 (median 0.6) episodes per 100 person-years for insulin analogues and from zero to 50.4 (median 2.8) for regular insulin. No study was designed to investigate possible long-term effects (e.g., mortality, diabetic complications), in particular in patients with diabetes-related complications. Reviewers Conclusions. The authors conclude that their analysis suggests only a minor benefit of shortacting insulin analogues in the majority of patients with diabetes who are treated with insulin. Until longterm efficacy and safety data are available, the authors suggest a cautious response to the vigorous promotion of insulin analogues. Because of fears of potentially carcinogenic and proliferative effects, most studies to date have excluded patients with advanced diabetic complications. For safety purposes, we need a long-term follow-up of large numbers of patients who use short-acting insulin analogues. Furthermore, we need well-designed studies in pregnant women to determine the safety profile for the mother and the unborn child.

In this Cochrane review,1 Siebenhofer and colleagues looked at regular insulin in comparison with short-acting analogues (lispro, aspart). Overall, they found that short-acting insulin analogues were as effective as regular human insulin in long-term glycemic control, and that they were similarly associated with episodes of hypoglycemia. The only difference found was in patients with type 1 diabetes, who had a small decrease in their A1C levels (about 0.1 percent; 0.2 percent in those who used continuous subcutaneous insulin infusions). It is unlikely that this difference is clinically significant. However, 10 percent of the included studies were only 28 days in duration, and 36 percent of the studies lasted fewer than 90 days. With intensive therapy, A1C levels can begin to drop in three to five weeks,5 but the full effect of a change in therapy may not occur until the end of the 120-day life span of the red blood cells. Thus, more than one third of the included studies may have had insufficient follow-up time to determine the full effect of the intervention on A1C levels. No studies were found that compared regular insulin with short-acting analogues in terms of their effects on the long-term complications of diabetes. When the results from the Diabetes Control and Complications Trial Research Group study6 are extrapolated, a 0.1 percent reduction in A1C levels translates into a number needed to treat of 650 to prevent the development of diabetic retinopathy, and 765 to halt its progression. Patients can achieve effective long-term control of blood sugar with regular insulin or a shortacting analogue. Regular and short-acting analogues are similar in achieving glycemic control in studies performed to date. Short-acting analogues may be advantageous in allowing for greater

flexibility and convenience. For patients with type 1 diabetes, insulin analogues are slightly more effective than regular insulin, especially if the patient uses an insulin pump.

Translate:

Kesimpulan Reviewer ' . Para penulis menyimpulkan bahwa analisis mereka menunjukkan hanya manfaat kecil analog insulin short-acting pada kebanyakan pasien dengan diabetes yang diobati dengan insulin. Sampai efikasi jangka panjang dan keamanan data yang tersedia, penulis menyarankan respon berhati-hati terhadap promosi yang kuat dari analog insulin . Karena kekhawatiran efek berpotensi karsinogenik dan proliferasi , kebanyakan penelitian hingga saat ini dikecualikan pasien dengan komplikasi diabetes lanjut . Untuk tujuan keamanan, kita perlu tindak lanjut jangka panjang dari sejumlah besar pasien yang menggunakan analog insulin short-acting .

Dalam review Cochrane ini, Siebenhofer dan koleganya melihat insulin reguler dibandingkan dengan analog short-acting ( lispro , ASPART ) . Secara keseluruhan , mereka menemukan bahwa analog insulin short-acting sama efektifnya dengan insulin manusia biasa dalam kontrol glikemik jangka panjang , dan bahwa mereka sama-sama terkait dengan episode hipoglikemia . Satu-satunya perbedaan yang ditemukan adalah pada pasien dengan diabetes tipe 1 , yang mengalami penurunan kecil dalam tingkat A1C mereka (sekitar 0,1 persen, 0,2 persen pada mereka yang digunakan terus menerus subkutan insulin infus ) . Hal ini tidak mungkin bahwa perbedaan ini secara klinis signifikan. Namun, 10 persen dari studi termasuk yang hanya 28 hari dalam durasi , dan 36 persen dari studi berlangsung kurang dari 90 hari . Dengan terapi yang intensif , kadar A1C dapat mulai drop dalam tiga sampai lima minggu, tetapi efek penuh perubahan dalam terapi mungkin tidak terjadi sampai akhir masa hidup 120 hari dari sel-sel darah merah . Dengan demikian , lebih dari sepertiga dari studi termasuk mungkin memiliki cukup waktu tindak lanjut untuk menentukan efek dari intervensi pada tingkat A1C . Tidak ada studi yang menemukan bahwa dibandingkan insulin reguler dengan analog short-acting dalam hal pengaruhnya terhadap komplikasi jangka panjang dari diabetes. Ketika hasil dari Diabetes Control and Complication Trial Research Group study diekstrapolasi , penurunan 0,1 persen pada tingkat A1C diterjemahkan dalam jumlah yang diperlukan untuk mengobati dari 650 untuk mencegah perkembangan retinopati diabetes , dan 765 untuk menghentikan perkembangannya. Pasien dapat mencapai kontrol gula darah jangka panjang yang efektif dengan insulin reguler atau analog short-acting . Analog short-acting dan Reguler sama dalam mencapai kontrol glikemik pada studi yang dilakukan sampai saat ini. Analog short-acting mungkin menguntungkan dalam memungkinkan untuk fleksibilitas dan kenyamanan yang lebih besar . Untuk pasien dengan diabetes

tipe 1 , analog insulin yang sedikit lebih efektif daripada insulin reguler , terutama jika pasien menggunakan pompa insulin .