基礎胰島素聯合GLP-1受體激動劑：一種機制互補的降糖方案

文章來源：中華內分泌代謝雜誌, 2018,34(4) : 274-280

作者：Frías Juan Pablo

摘要

時至今日，我們已擁有多種降糖藥物來應對2型糖尿病的不同病理缺陷。基於豐富的循證證據和臨床實踐經驗，近期的糖尿病治療指南推薦及早採用機制互補的藥物聯合方案，以幫助患者持續達到個體化的血糖控制目標。對於已使用基礎胰島素但血糖控制尚未達標的2型糖尿病患者，多項臨床試驗證明對其加用胰升糖素樣肽1(GLP-1)受體激動劑有助於安全有效地控制血糖。基礎胰島素聯合GLP-1受體激動劑治療方案，通過互補的作用機制，在優化血糖控制的同時，較基礎餐時胰島素等方案顯著降低低血糖風險，改善體重。因此，對於採用基礎胰島素治療但血糖控制不佳的2型糖尿病患者應考慮增加GLP-1受體激動劑。

正文

眾所周知，嚴格控制血糖使之接近甚至達到正常範圍，對2型糖尿病管理非常重要[1,2,3,4]。雖然循證證據已充分證明改善HbA1C可降低糖尿病併發症進展的風險，但即使在醫療系統最發達的國家中，也有超過50%確診的2型糖尿病患者HbA1C>7.0%，即血糖控制不達標[5,6,7]。甚至不少患者HbA1C>9.0%，處於糖尿病長期併發症的高風險中[7]。

血糖控制難以達標和(或)難以持續達標的原因有多個，包括疾病複雜的病理生理機制及其本身進行性惡化的特性(例如β細胞功能障礙)，以及一系列的醫護人員相關因素和患者相關因素[8,9,10,11]。與其他慢性疾病類似，患者的態度、治療依從性、潛在的藥物不良反應、糖尿病治療方案的複雜程度(尤其在有高血壓、高血脂、心血管疾病等多個共患病同時需要藥物治療的情況下)等，均是影響疾病長期管理效果的難題[10]。從醫生的角度而言，由於門診時間非常有限，往往難以充分有效地應對患有多種慢性疾病的患者，由此可導致所謂的"臨床惰性" ，延遲強化治療方案的採用，也延誤了血糖的及時控制達標[11]。解決這些阻礙患者血糖達標的障礙的關鍵因素之一在於選擇恰當的治療方案。恰當的治療方案應該具有良好的療效，較少的不良反應，潛在的降糖外獲益(例如減重、降壓等)，以及相對簡單的給藥方案便於醫生指導，也便於患者用藥。

時至今日，我們已擁有多種降糖藥物來應對2型糖尿病的不同病理生理缺陷[9]。相應的，各大指南也推薦選擇作用機制互補的藥物進行聯合治療[12,13]。儘管幾乎所有的指南都推薦沒有禁忌證的2型糖尿病患者首先起始二甲雙胍治療，但隨著時間推移，絕大多數的患者都需要加用至少1種降糖藥物來使血糖達標和(或)維持達標。鑒於胰島β細胞功能缺陷在2型糖尿病發病及進展過程中的重要性和外源性胰島素治療在降血糖治療中的有效性，胰島素治療，尤其是長效(基礎)胰島素治療，在口服藥治療失效患者中的應用是非常普遍的[12,13]。自從10年前基礎胰島素類似物(甘精胰島素U-100/來得時和地特胰島素/諾和平)的出現和治療達標(treat-to-target)研究結果[14,15]的發表，越來越多的臨床治療開始採用每日1次基礎胰島素控制空腹血糖，並加用口服降糖藥物[14,15]。目前在2型糖尿病患者中，基礎胰島素加口服藥物的聯合治療仍然是應用最廣泛，也是最有效的治療方案之一。

但對於已使用基礎胰島素聯合口服藥物治療但血糖仍然無法達標的患者，應如何治療呢？美國糖尿病協會(ADA)指南對這個問題作了全面的闡釋[12]。當已應用基礎胰島素使空腹血糖達標，而HbA1C尚未達標時，應考慮餐後高血糖的影響。治療選擇可有3種：(1)在當天的主餐中加入速效胰島素(基礎聯合方案)然後根據需要轉為每日多次的基礎-餐時方案，或(2)從基礎胰島素轉為每日2次注射的預混胰島素方案，或(3)添加GLP-1受體激動劑[12]。對於第3種方案，楊文英教授團隊近期發表了1項最新研究，評估在使用基礎胰島素和口服藥物治療血糖仍控制不佳的患者中，加用短效GLP-1受體激動劑利司那肽的治療效果[16]。

2型糖尿病的高血糖是由多種病理缺陷共同作用的結果。GLP-1受體激動劑和基礎胰島素這兩種藥物，能以互補的方式分別糾正其中的大部分病理缺陷。基礎胰島素能增強以骨骼肌為主的外周組織中的葡萄糖攝取，同時在肝臟抑制肝糖原生成[17]。因此其主要作用於空腹和餐前高血糖。GLP-1受體激動劑的作用機制有多種，包括葡萄糖依賴的刺激胰島素分泌，抑制餐後胰升糖素分泌(糖尿病患者的餐後胰升糖素分泌往往不被抑制甚至反常升高)，延緩胃排空，以及在中樞神經系統誘導產生飽腹感，從而減輕體重[18]。GLP-1受體激動劑的這些作用機制主要引起餐後血糖降低。由此可見，GLP-1受體激動劑是基礎胰島素最佳的聯合用藥選擇。

事實上目前已有多個研究探索了GLP-1受體激動劑和基礎胰島素聯合方案的治療效果[19,20,21]。與基礎胰島素加速效胰島素方案相比，加用GLP-1受體激動劑的降血糖效果相似甚至更好，低血糖風險更低，還能顯著減輕體重[19,20,21]。其不良反應以噁心、嘔吐、腹瀉等胃腸道不良反應常見[19,20,21](見表1)。雖然有部分患者因不良反應而停葯，但這些不良反應大多為輕到中度，多在治療的早期發生，且能隨時間推移逐漸緩解。從臨床醫生的角度而言，用藥前的醫患溝通非常重要，應事先告知患者這些不良反應的相關信息，提醒他們不要過度進食，以及胃腸道不良反應發生後，會在短期內逐漸緩解。同時也有若干研究在基礎胰島素治療血糖控制不佳的患者中，對加用GLP-1受體激動劑和加用速效胰島素方案的成本效益進行評估比較[22,23]。儘管這些研究僅在特定國家進行，但其結果均證明，與加用每日1次至3次餐時胰島素的強化胰島素治療方案相比，在基礎胰島素基礎上加用GLP-1受體激動劑可改善血糖控制，降低低血糖、體重增加等風險，降低併發症累積發生率，由此改善預期質量調整生命年並減少總體醫療支出[22,23]。在選擇治療方案時，醫療成本是一項關鍵的驅動因素。因此在國家、地區或區域範圍內進行治療方案相關醫療決策時，也應該著重考慮方案的成本效益。

GLP-1受體激動劑可分為長效和短效兩大類[24]。短效GLP-1受體激動劑以每日2次注射的艾塞那肽和每日1次注射的利司那肽為代表，其半壽期約為2 h，具有非常顯著的降低餐後血糖的療效，但對空腹血糖影響有限。相反的是，長效GLP-1受體激動劑包括利拉魯肽、度拉糖肽、索瑪魯肽等，可能由於快速耐受胃排空作用，對餐後血糖影響較小，而對空腹血糖有明顯的療效。一項在2型糖尿病患者中評估利司那肽和利拉魯肽餐後降糖效果的短期研究結果，清晰地顯示了這兩種藥物的葯代動力學差異[25]。該研究採用標準早餐，熱量451 kcal，含60.6%碳水化合物，12.4%蛋白質，26.9%脂肪。短效GLP-1受體激動劑利司那肽可在標準早餐後4h，使葡萄糖曲線下面積(AUC)自基線降低(12.6±0.6)h·mmol / L，而長效GLP-1受體激動劑利拉魯肽僅能降低(4.0±0.6)h·mmol/L。二者估計差異可達－8.6 h·mmol/L (95%CI－10.0, －7.1 h·mmol/L,P[25]。同時楊文英教授團隊的研究也證實，利司那肽可顯著改善餐後血糖波動，但對空腹血糖影響較小[16,19,20]。長效GLP-1受體激動劑的空腹降糖療效也在多個研究中得到驗證。儘管目前尚未有臨床研究在長期應用基礎胰島素的糖尿病患者中，對長效和短效GLP-1受體激動劑進行正式比較，但就臨床經驗而言，對於餐後高血糖影響HbA1C達標的患者，應選擇短效GLP-1受體激動劑，而對於積極調整胰島素劑量仍不能使空腹血糖達標的患者，應選擇長效GLP-1受體激動劑。

最近，基礎胰島素和GLP-1受體激動劑的固定劑量復方製劑已在美國和全球多個國家及地區上市。該類復方製劑主要有2種，一種是甘精胰島素U-100和利司那肽的復方製劑(iGlarLixi，美國商品名為Soliqua?100/33，其他國家及地區商品名為Suliqua?)；另一種是德谷胰島素U-100和利拉魯肽的復方製劑(iDegLira，美國商品名為Xultophy?100/3.6，其他國家及地區商品名為Xultophy?)。在美國，每支SoloSTAR?注射筆含3 ml iGlarLixi，其中每毫升含有100單位的甘精胰島素和33 μg利司那肽。在美國以外的國家和地區還有另一種規格，含100單位的甘精胰島素和50 μg利司那肽。每支FlexTouch?注射筆含有3 ml iDegLira，其中每毫升含有100單位的德谷胰島素和3.6 mg利拉魯肽。在僅服用口服降糖葯或口服降糖葯聯合基礎胰島素治療血糖控制不佳的患者中進行隨機對照臨床研究，這兩種復方製劑都比胰島素或GLP-1受體激動劑單獨治療能更好地控制血糖[26,27,28,29]。值得注意的是，iGlarLixi和iDegLira這兩種復方製劑均在不增加低血糖風險的前提下改善血糖控制。與基礎胰島素單獨治療相比，復方製劑不影響或輕微降低體重；與GLP-1受體激動劑單獨治療相比，復方製劑的胃腸道不良反應顯著減輕[26,27,28,29]。綜上所述，與基礎胰島素和GLP-1受體激動劑單獨治療相比，固定劑量復方製劑不僅增強降糖療效，還能減輕各自主要不良反應。同時採用安全易用的注射筆每日1次注射，為患者用藥帶來便捷。這些具有顯著降糖療效(以及降糖外獲益)，良好安全性特徵，給藥方便的藥物製劑的迅速研發上市，對醫生和患者都是值得慶幸的消息。

考慮到我們已擁有多種不同作用機制的降糖藥物來應對血糖管理，當下很有必要思考的一個問題是，如何合理地選擇聯合治療方案，以便這些藥物的作用機制互補，由此以更安全的方式改善血糖控制。藥物機制研究數據，臨床研究數據，以及臨床應用經驗都清楚表明，基礎胰島素加GLP-1受體激動劑是一種行之有效的機制互補的聯合治療方案。雖然糖尿病管理必須要個體化，沒有任何一種藥物能普遍適用於所有患者，但對於已用基礎胰島素治療但血糖控制不佳的患者而言，應首先考慮這種聯合治療方案。

The importance of normal or near normal glycemic control in the treatment of type 2 diabetes has been well established for many years[1,2,3,4]. Despite irrefutable evidence demonstrating that improvements in glycemic control, as measured by hemoglobin A1C(HbA1C), reduce the risk of development and/or progression of diabetes-related complications, even in countries with the most advanced medical systems greater than 50% of patients with diagnosed type 2 diabetes generally have a HbA1Cabove 7.0%, the commonly accepted glycemic target[5,6,7]. Perhaps more importantly, many patients have a HbA1Cabove 9.0%, clearly placing them at heightened risk of long-term diabetes complications[7]．

The reasons for the difficulty in achieving and/or maintaining good glycemic control are several-fold and include the complex pathophysiology of the disease and its progressive nature(i.e., beta-cell dysfunction), as well as factors related to both patients and healthcare providers[8,9,10,11]. Like any chronic disease, there are often issues with patient motivation, adherence with therapeutic regimens, potential treatment side effects, and the complexity of diabetes treatment regimens, particularly in the face of multiple other co-morbid conditions requiring pharmacotherapy(e.g., hypertension, dyslipidemia, cardiovascular disease)[10]. From a healthcare provider perspective, often with significant time constraints during office visits, dealing effectively with patients with multiple chronic conditions can be challenging. This often leads to so-called clinical inertia, with delays in treatment intensification and attainment of glycemic targets[11]. Therapeutic regimens that are effective, have limited side effects and potential benefits(e.g., weight loss, reduction in blood pressure), and that are relatively simple for healthcare providers to teach and for patients to administer are critical in helping address some of these barriers which so often prevent our patients from achieving glycemic targets.

Today, we are fortunate to have pharmacological agents that have effects on the multiple defects contributing to the pathophysiology of type 2 diabetes[9]. As such, it makes sense, and is recommended by various guidelines, to treat patients with combinations of agents that have complementary mechanisms of action[12,13]. Guidelines unanimously recommend initiating therapy with metformin in all patients with type 2 diabetes who have no contraindications to this agent. Most patients require the addition of one or more therapeutic agents over time in order to attain and/or maintain good glycemic control. Given the importance of the pancreatic beta-cell defect in type 2 diabetes and its progressive nature, as well as the effectiveness of exogenous insulin therapy in reducing plasma glucose, treatment with insulin, particularly long-acting(basal)insulin, is very common when combinations of oral agents fail to achieve glycemic targets[12,13]. This treatment regimen of oral agents plus a daily application of basal insulin, aimed primarily at addressing fasting hyperglycemia, became widely used with the advent of basal insulin analogs[insulin glargine U-100 (Lantus), and insulin detemir (Levemir)]and the publication of the so-called " treat-to-target" trials over 10 years ago[14,15]. It remains a very common and effective therapeutic regimen in patients with type 2 diabetes.

But what should be done when patients fail to achieve adequate glycemic control with combinations of oral agents and basal insulin? This question is addressed comprehensively by the American Diabetes Association(ADA)guidelines[12]. In this situation, when fasting glucose has reached its target by appropriate titration of basal insulin and yet the HbA1Cremains elevated, postprandial hyperglycemia is generally " responsible" for the inability to achieve target HbA1C. Treatment options are three-fold：(1)add a rapid-acting(prandial)insulin at the largest meal of the day(a " basal plus" regimen)and then, if needed, move to a full basal-bolus regimen with rapid-acting insulin administered at multiple meals, or(2)Switch from basal insulin to a twice daily regimen of pre-mixed insulin containing both basal and prandial insulin, or(3)add a Glucagon-like peptide 1 receptor agonist(GLP-1 RA)[12]. In the recent issue of the Diabetes, Obesity and Metabolism, Yang et al. present the findings from a clinical trial assessing the latter of these 3 choices, the addition of a GLP-1 RA, in this case the short-acting GLP-1 RA lixisenatide, to patients not achieving glycemic targets with oral agents and optimally-titrated basal insulin[16].

When looking at the various pathophysiologic defects that contribute to hyperglycemia in type 2 diabetes, these two agents, basal insulin and GLP-1 RAs, address a number of these defects and do so in a complementary fashion. Basal insulin acts to enhance insulin-stimulated glucose uptake in peripheral tissues, primarily skeletal muscle, and, at the level of the liver, suppresses hepatic glucose production[17]. By doing so, its primary effect is on fasting and pre-prandial glucose. GLP-1 RAs have several mechanisms of action including enhancing glucose-stimulated insulin secretion, suppressing postprandial glucagon secretion(which is often not suppressed or paradoxically elevated in persons with diabetes), and slowing gastric emptying. Additionally, there is a central effect that induces satiety resulting in weight loss[18]. These actions result in a reduction in postprandial glucose, thus making GLP-1 RAs an excellent choice for combination with basal insulin; with basal insulin having its primary effect on fasting glucose and GLP-1 RA targeting postprandial glucose.

Indeed, the multiple GLP-1 RAs that are now commercially available throughout the world, have all been studied in combination with basal insulin[19,20,21]. Compared to adding prandial, rapid-acting insulin to basal insulin, the addition of a GLP-1 RA results in similar or sometimes better glycemic control, but with significantly less hypoglycemia and with body weight loss compared to a significant body weight gain with a basal-bolus insulin regimen[19,20,21](see Table 1). As expected, given the side effect profile of GLP-1 RAs, gastrointestinal(GI)-related side effects(nausea, vomiting and diarrhea)are more common in a regimen including a GLP-1 RA[19,20]. Although some patients experience GI side effects that preclude the use of these agents, most of these side effects are mild to moderate in intensity, occur early in the course of therapy and resolve over time, thus allowing the patient to continue their use. From a clinical perspective, it is very important to communicate these potential side effects to patients and to tell them not to over eat and that these symptoms, should they occur, will likely resolve in a relatively short period of time. The cost-effectiveness of adding a GLP-1 RA versus adding rapid-acting insulin to patients inadequately controlled with basal insulin therapy has been assessed[22,23]. Although these analyses are country-specific, adding GLP-1RA to basal insulin has been associated with improvements in quality-adjusted life expectancy and total cost savings compared with intensification of therapy with daily or three times daily mealtime insulin due to improvement of glycemic control and reductions of risk factors such as hypoglycemia, BMI, and cumulative incidence of complications[22,23]. Cost is often a critical driver of utilization, and cost-effectiveness considerations, which should be made at the country, regional or local level, play an important role in decision-making related to therapeutic regimens.

Tab 1Comparison of combination injection regimens

An important distinction to make between the various GLP-1 RAs available today is whether they are short or long-acting[24]. The short-acting GLP-1 RA, like exenatide BID and lixisenatide, have a plasma half-life of approximately 2 hours and have a very profound postprandial-lowering effect with little or no effect on fasting glucose. By contrast, the long-acting GLP-1 RA such as liraglutide, dulaglutide, and semaglutide, have less of a postprandial effect(likely due to tachyphylaxis to the effect on gastric emptying), but do have a significant effect on fasting glucose. In a short-term study assessing the postprandial effects of lixisenatide versus liraglutide in patients with type 2 diabetes, this difference in pharmacodynamic effect(based on the pharmacokinetics of each agent)was well demonstrated[25]. The postprandial glucose area under the curve(AUC)during the 4 hours following the start of a standard breakfast meal(451 kcal; 60.6% carbohydrate, 12.4% protein, 26.9% fat)was reduced from baseline by(12.6±0.6)h·mmol/L with lixisenatide, the short-acting GLP-1 RA, and by(4.0±0.6)h·mmol/L with the long-acting GLP-1 RA liraglutide [estimated treatment difference －8.6 h·mmol/L (95%CI：－10.0, －7.1 h·mmol/L),P[25]. On the contrary, numerous studies have shown the fasting glucose lowering capacity of the long-acting GLP-1 receptor agonists, while studies like the one by Yang et al. demonstrate that lixisenatide had a significant effect on postprandial glucose excursions, with no discernible effect on fasting glucose concentrations[16,19,20]. Although these two " classes" of GLP-1 RA(short- and long-acting)have not been formally compared in long-term trials in basal insulin-using patients, from a clinical perspective one could envision selecting a short-acting agent in patients where significant postprandial hyperglycemia is the primary issue preventing attainment of target HbA1C, and a long-acting agent when fasting glucose remains difficult to control despite aggressive basal insulin titration.

More recently, fixed-ratio combinations of basal insulin and GLP-1 RAs have become commercially available in the United States and other parts of the world. There are currently two such agents, one combining insulin glargine U-100 and lixisenatide(iGlarLixi, Soliqua?100/33 in US and Suliqua?in the rest of the world)and the other combining insulin degludec U-100 and liraglutide(IDegLira, Xultophy?100/3.6 in the US and Xultophy?in the rest of the world). In the US, iGlarLixi is available as a combination of 100 units insulin glargine and 33 μg lixisenatide per ml, in a 3 ml SoloSTAR?pen(outside of the US, there is also a combination of 100 units insulin glargine and 50 μg lixisenatide per ml). IDegLira is available in a 3 ml FlexTouch?pen as a combination of 100 units insulin degludec and 3.6 mg liraglutide per ml. In randomized, controlled clinical trials in patients suboptimally controlled on oral agents alone or on oral agents plus basal insulin, both fixed-ratio combination agents have been shown to significantly improve glycemic control compared to basal insulin or the GLP-1 RA component alone[26,27,28,29]. Importantly, both iGlarLixi and IDegLira, improved glycemic control without increasing the risk of hypoglycemia, with either weight neutrality or weight loss(compared to weight gain with basal insulin alone), and with significantly less GI side effects compared to the GLP-1 RA component alone[26,27,28,29]. Therefore, the fixed-ratio combinations of basal insulin and GLP-1 RAs improve efficacy compared with their mono-components while mitigating key side effects of the individual components. They are administered once daily as a single injection with well-established and easy-to-use pen devices, adding convenience and ease-of use for patients. In the future, we may very well have safe and effective oral GLP-1 receptor agonists to add to even longer-acting basal insulins with " flatter" and more consistent pharmacokinetic/pharmacodynamic profiles. Thankfully for us and for the patients we care for, development of new agents with improved glycemic(and extra-glycemic)efficacy, good safety profiles, and ease of administration continues at a rapid pace.

Given the importance of achieving and maintaining good glycemic control and the fact that we now have various classes of pharmacological agents at our disposal, it is imperative that we use rational combinations of these agents, when needed, so that their complementary mechanisms of action can help improve glycemic control in a safe manner. Mechanistic data, data from clinical trials, as well as clinical experience, clearly indicate that basal insulin plus a GLP-1 RA is one such combination. Although no single agent or combination of agents is appropriate for every patient, and diabetes care must be individualized, this combination should always be given consideration when caring for a patient with type 2 diabetes not achieving glycemic targets with basal insulin.

參考文獻 （略）

掃碼下載雜誌官方App

喜歡這篇文章嗎？立刻分享出去讓更多人知道吧！