These pages are best viewed with Netscape version 3.0 or higher or Internet Explorer version 3.0 or higher. When viewed with other browsers, some characters or attributes may not be rendered correctly.

F E A T U R E  A R T I C L E

Insulin Lispro: The Next Step

Christian D. Herter, MD, CDE

It has only been a few years since our understanding of the natural history of diabetes was broadened by results of the Diabetes Control and Complications Trial (DCCT).¹ Applying intensive therapy to our patients with type I diabetes will reduce their risk of developing complications. However, using currently available technology to achieve this worthy goal can prove challenging, even to seasoned diabetologists.

Researchers participating in the DCCT had an arsenal of dietitians, nurses, endocrinologists, and funding to help carry out strategies designed to achieve and maintain the euglycemic state for a decade. Proper timing of pre-prandial short-acting insulin was necessarily high on the list of teaching goals, which were continually reinforced, and rightly so. Those of us involved in diabetes care know all too well that most of our patients, even those who have completed a thorough regimen of diabetes education, commonly dose their pre-meal insulin immediately before eating.

Regular insulin does not work properly when given this way.^2-4 Although the time to peak action, as well as total duration of activity, of a single subcutaneously administered dose of regular varies with dose,⁵ a minimum waiting time of 30 minutes is required after injection before a meal should be eaten. As a patient with diabetes myself, I realize that this is not always possible or prudent. When taken immediately before eating, the intensive effect of the insulin is lost, even if the calculated amount exactly matches what would be required for the carbohydrate load to be consumed.

Our attempt to mimic the insulin secretory effect of the nondiabetic pancreas is thus often thwarted by the unnatural action of our therapeutic agent. Not only is the insulin deposited in a location far removed from the portal circulation, but its pharmacokinetic action is altered radically by the method of delivery. Insulin levels in patients taking the exogenous hormone tend to be higher than in nondiabetic individuals of the same weight, an effect probably related to the peripheral release.⁶ Insulin lispro directly addresses the absorption problem.

Molecular Structure

Regular insulin lacks speed. Absorption takes far too long, and attainment of peak serum activity occurs so late that its ideal use is nearly impossible for most people with type I diabetes. To discover why this is so requires a basic understanding of fundamental biochemical structure theory.

The insulin molecule is relatively small, by biochemical standards, and is composed of two amino acid subchains, one longer than the other. A consequence of the electronegative quality of this arrangement is an inherent tendency for the molecule to self-associate into a hexameric superstructure.⁷ Unfortunately for everyone, it is not this multiple isomeric form of insulin that is absorbed and biologically active. Within the subcutaneous depots, the hexamer must be slowly broken apart into dimeric and monomeric sub-units, which find their way into circulation. The sluggishness of this process is responsible for the onset of action, time to peak action, and duration of activity of regular insulin. Absorption is, therefore, the rate-limiting step.

Insulin lispro was designed for speed. Its fundamental structure permits rapid dissociation from the hexomeric form to dimmer and monomeric sub-units. Lispro is a synthetic product made by exchanging the amino acids proline and Iysine at positions 28 and 29, respectively, on the large beta chain of the molecule (Figure 1). The idea came from observations of a related and similar structural cousin, insulin-like growth factor 1, which exists as a monomer. The resulting product, initially designated "Insulin Analogue [Lys(B28), Pro(B29)]," soon became shortened to "insulin Lys-Pro," and, finally, to insulin lispro (pronounced "lice pro"). Eli Lilly and Co. markets this new therapeutic agent as Humulog ™.

Insulin lispro does associate into a weakly stable hexomeric form when suspended in an aqueous solution containing zinc and phenol.8 This provides a stable storage medium and interferes little with rapid dissociation when injected subcutaneously.⁹ The concept of this seemingly unusual transport milieu is far from esoteric, since it is the same aqueous matrix used for years to protect the stability of regular insulin. The biological activity of lispro is compared with regular insulin in Table 1.

Some variation is seen when reviewing the literature for information regarding the time to peak biological activity. In their studies of 10 C-peptide negative patients with type I diabetes, Torlone and associates showed that the time to maximal plasma insulin concentration after injection with lispro was 41 + 4 minutes. In 10 similar patients, Howey and associates⁹ demonstrated peak biological activity at 42 minutes post-injection for pure lispro but 53 minutes when it was mixed with zinc and phenol, as in the commercially available preparation (which was what Torlone had used). Compared with regular insulin, both studies confirmed the very rapid action of lispro.

Insulin lispro is currently recommended only for subcutaneous use. Some experimental data, however, suggest that regular and insulin lispro act exactly the same when given intravenously.¹¹ Future investigations into this area may help broaden usage guidelines. For the present, regular insulin should be considered the preparation of choice for clinical applications requiring intravenous injection.

Immunologic Concerns

Individuals who develop antibodies to their injected insulin products will show a later onset of action, blunted and delayed time to peak action, and prolonged total duration of activity.¹² Such a response to this new insulin, with its synthetically altered beta chain tantalizing suspicious cells of the human immune system, would reduce its appeal significantly. Fortunately, no increase in immunogenic response to lispro has been reported in any trials. A recent clinical investigation¹³ suggested that insulin lispro is no more likely to trigger an antibody response than is regular insulin in type I and type II diabetes patients who had been previously exposed to the latter.

Clinical Application

Perhaps not surprisingly after reviewing the pharmacodynamic data, current research has not demonstrated an improvement in hemoglobin A_1c when lispro insulin was used with NPH, with the longer-acting insulin given either at bedtime or twice a day.^14,15 The rapid-acting nature of lispro is achieved at the expense of relative biological endurance. After a strong, early surge following injection, it fades from the system rapidly and is completely eliminated in 4 hours. Its application to clinical practice, therefore, must include attention to providing consistent basal insulin between meals.

Using lispro with NPH. Two concerns must be considered when using this familiar, intermediate-acting product with lispro. Unlike regular insulin, which is only slightly affected by mixing with NPH, lispro undergoes some moderation of character. When the two are combined and injected in a syringe, Torlone and associates demonstrated that the resulting time to peak plasma insulin concentration is increased slightly.'° More importantly, they was able to observe a late stabilizing effect seen with this combination, noted at 7 hours post-injection. This effect is more pronounced than that occurring with regular and NPH. (Figures 2 and 3). This discovery prompted Torlone and her colleagues to suggest a novel dosing regimen, whereby a combination of 20-40% NPH and 60 80% lispro is employed before breakfast, lunch, and dinner. This system effectively deals with the issue of low plasma insulin levels that could occur when lispro is used in twice-daily or bedtime NPH dosing regimens.

Using lispro with ultralente. Theoretically, the judicious use of this very long-acting product should allow patients to achieve and maintain fairly continuous and consistent basal serum insulin levels. It would seem, therefore, ideally suited for use with lispro. Unlike regular, which interacts with ultralente when the two are combined, the action of lispro is minimally affected by such a mixture.

The dosing schedule for ultralente insulin, however, needs to be more carefully planned when this product is used with lispro. For example, some patients who use multiple-dose programs take a single bedtime injection of ultralente and have excellent control, providing regular insulin is used at least every 6 hours during the next day. The longer-acting nature of regular insulin makes this possible.

Insulin lispro would need some help during the daytime hours if used only with nighttime ultralente. Since basal insulin levels would be low during the day following evening ultralente administration, patients would have to inject about every 3 hours to prevent a potentially serious fall in serum insulin. By dosing the ultralente twice a day, more consistent basal insulin levels would be predicted, and the lispro could be used for meals as needed. Individuals with a prominent dawn phenomenon might require, in addition to their ultralente regimen, an evening dose of NPH or lente to help control the resulting morning hyperglycemia.

Despite the intuitive nature of this idea, studies comparing the effectiveness of twice-daily ultralente and intermediate-acting insulins with pre-meal regular have not shown an improvement in HbA₁ values. ^16-17 The same trials did note improved fasting blood glucose levels on ultralente, which suggests that some other factor, perhaps postprandial hyperglycemia, may have canceled the beneficial effect of correctly administered basal insulin. Substituting lispro for regular insulin in such a program should help further reduce total daily serum glucose excursions, hopefully resulting in better overall glycemic control.

Using lispro in CSII. To date, only one study has compared the use of lispro to regular in continuous subcutaneous insulin infusion (CSII). Zinman and colleagues ¹⁸ demonstrated better 1-hour postprandial blood glucose values and significantly improved HbA₁ levels in their cohort taking lispro.

They did not discover a higher rate of infusion site failure with this form of insulin, which would be expected, since it is phosphate-buffered in the commercially available preparation. pH-stabilized insulins are preferred for use in CSII since the risk of infusion catheter occlusion is lower.¹⁹There was no greater incidence of hypoglycemia in patients receiving the insulin analog, which suggests that lispro should be considered the preferred form of insulin for use in CSII.

Supplemental use of lispro. With its rapid pharmacokinetics, lispro insulin is ideally suited for use as an adjunct to many treatment regimens. Heinemann's truly practical trial comparing the effects of regular and lispro insulin injected immediately before a high-carbohydrate meal²⁰ demonstrated the remarkable difference between these two products when given under near"real world" conditions. In a double-blind, crossover design, study participants were given one or the other type of insulin just before a meal that consisted of tiramisu, pizza, and regular Coca Cola™. When plotted, the blood glucose response to both types of insulin is clearly evident (Figure 4). The area beneath the curve describing the action of lispro, an indication of duration and extent of hyperglycemia, is 78% less than the corresponding relationship observed when regular insulin was given.

Patients using intensive therapy will often experience hyperglycemia during the course of their self-management. How a particular individual deals with this depends somewhat on the type of maintenance program he or she is using. For example, people on multiple-dose regimens frequently use additional regular insulin, given on a sliding scale to cover glucose excursions outside of their therapeutic goal range. Holleman and associates showed in a recent study²¹ that lispro is much more effective at reducing experimental hyperglycemia in type I diabetes patients than is regular insulin. This lends credibility to the use of the analog for prompt treatment of incidental hyperglycemia.

Injection site kinetics. Absorption characteristics of regular insulin from areas of injection are known to be site dependent. Subcutaneous administration in abdominal distributions results in more rapid onset of action and shorter time to peak action.

This effect is not nearly as dramatic with insulin lispro. Recently, Ter Braak and associates²² demonstrated no significant difference in onset of action when lispro was injected subcutaneously into abdominal, deltoid, or femoral regions of healthy male volunteers. Duration of action was increased by up to 1.5 hours, however, for extremity sites. Regular insulin administered in deltoid and femoral areas demonstrated a significant increase both in time to peak action and in total duration of activity, and the latter was more pronounced than that observed with the analog. This strongly supports previous recommendations that cite the abdomen as a preferred area for subcutaneous administration of insulin when rapid action is desired.

Use in pregnancy. Insulin lispro is considered a "class C" drug, just as all other forms of insulin. This is because there are no studies currently available assessing the safety of this therapeutic agent in terms of teratogenic potential. Consequently, the question of applicability to management strategies for type I diabetes and pregnancy or for treatment of gestational diabetes is a somewhat complicated issue.

Even when glycemic control is exceptional, fetal macrosomia is common in pregnancies complicated by maternal type I diabetes. This is believed to be a result of exogenous insulin acting on fetal glucose metabolism. Mechanisms have been proposed that allow for movement of antibody-bound insulin across the placental membranes from mother to fetus.²³

Structurally, insulin lispro is closely related to regular insulin. It is possible, therefore, that the same mechanisms that usher the latter across maternal-fetal boundaries would also act on lispro, thus delivering the analog directly to fetal tissues. Until data regarding the effects of lispro on fetal tissues becomes available, caution is warranted when using this drug in pregnancy.

Clinical Concerns Hypoglycemia. Obviously, the rapid action of lispro could work against patients not aware of the significant differences between the traditional and newer insulins. Of particular concern would be those individuals who take their dose and then become distracted before being able to eat.

Despite the potential for trouble here, practitioners should be reassured by the lower incidence of hypoglycemia demonstrated in clinical comparison trials. However, because the kinetics of this new product are so different from anything previously available, healthcare professionals should be sure to review the properties of this new insulin with their patients, especially those who have been using regular insulin for many years and are accustomed to its much slower biological characteristics.

Hyperglycemia. Individuals using lispro on either multiple daily injection (MDI) or CSII programs need to be acutely aware of the short duration of action of the analog. With CSII, for example, any circumstance that interrupts the supply of insulin, such as catheter occlusion, site failure, or simply running out of insulin, will lead to diabetic ketoacidosis much more rapidly with lispro.

The need to evaluate infusion sites regularly and carry out self-management of blood glucose often should be reinforced aggressively whenever CSII patients are seen. Additionally, in the case of site failure when establishing a new site is not immediately possible, lispro insulin injections would need to be given frequently to prevent hypoinsulinemia.

Individuals who use an MDI program should be reminded that lispro alone is not the best alternative when long-acting insulin is accidentally omitted. It may be reasonable advice for

both groups of patients to have regular insulin on hand for emergency situations. In this case, the longer duration of action can be helpful in maintaining more physiological insulin levels until the usual treatment program can be reestablished.

Exercise. Experimentally, lispro given with a meal before exercise produced more hypoglycemia than did regular insulin when administered 40 minutes before exertion, but less if the exercise was postponed for 180 minutes.²⁴ This is certainly expected based on what is known about the pharmacodynamics of both insulins.

Advice regarding exercise in type I diabetes is a complicated issue. It needs to be considered very carefully and be based on information about the specific type of activity, physical conditioning and general health of the patient, nutritional factors, and the other types of insulin preparations used. Generally, its always best for patients to time their activity as far removed from the last dose of insulin as is practical. Because of lispro's shorter duration of action, patients taking lispro may be able to begin exercise sooner after dosing compared with individuals using regular insulin.

Conclusion People with type I diabetes live in an imperfect world. There is no substitute for their missing ß-cells yet, and delivering insulin to the body by injecting it into adipose depots only feebly approximates the delicate interplay between blood glucose levels and the healthy pancreas. Insulin lispro can help tighten the gap by allowing more control over postprandial hyperglycemia, even when it is administered immediately before a meal.

The rapid action comes with a cost, however, in that the analog does not persist in the system as long as other insulins. Therefore, we must be caution when prescribing it, paying particular attention to providing adequate basal insulin.

Intensive programs using human ultralente given twice daily can be effective. Torlone's intriguing schedule of multiple-dose NPH has been suggested as a way to deal with the problem of erratic absorption of ultralente, which can occur in some individuals.²⁵

Other problems can occur as well, requiring the crafty physician to design creative insulin dosing schemes, perhaps using the properties of various preparations to flatten the glucose response curve for a given patient. An individual with recalcitrant fasting hyperglycemia, for example, may benefit from the addition of nighttime NPH to her twice-daily ultralente regimen.

The use of this new product in CSII may be the easiest of all, provided that our patients understand the more urgent problems associated with catheter occlusion.

We are now just a bit closer to being able to prescribe insulin in a way that mimics the non-diabetic state. As other short- and long-acting analogs emerge, there may come a time when the use of "real" insulin becomes unusual. Until then, our success in helping our patients approach euglycemia depends on the artful application of many types of insulin products. There is no standard, orthodox way of using any of them, if the goal is to help our patients reduce the magnitude and frequency of their glucose excursions. Instead, the program we design for a particular individual will depend on all the variables that influence glucose metabolism for that one person.

We must take each of these variables into consideration when choosing which intermediate or long-acting insulins to use, deciding when to apply them throughout the day, and determining on which schedule the shorter-acting agents should be added. Insulin lispro can be extremely helpful in this regard, since it allows us more precision in attenuating postprandial hyperglycemia with less risk of hypoglycemia.

REFERENCES

1 The DCCT Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N. Engl J Med 329:977-86, 1993.

2 Dimitriadis G. Gerich J: Importance of timing of preprandial subcutaneous insulin administration in the management of diabetes mellitus. Diabetes Care 6:374-77, 1983.

3 Binder C, Lauritzen T. Faber O. Pramming S: Insulin pharmacokinetics. Diabetes Care 7:188-99, 1984.

4 Heinemann L, Chantelau E: Pharmacokinetics and pharmacodynamics of subcutaneously administered U40 and U100 formulations of regular human insulin. Diabete Metab 18:21-24, 1992.

5 Santiago JV: "Results of recent clinical trials with lispro. Overcoming barriers to intensive glucose control: the inadequacies of conventional insulin therapy." Clinical Symposium, San Francisco, June 10, 1996.

6 Gale EA: Insulin lispro: the first insulin analogue to reach the market. Prac Diabetes Int 13:122-24, 1996.

7 Brange J, Langkjoer L: Insulin structure and stability. Pharm Biotechnol 5:315-50, 1993.

8 Ciszak E, Beals JM, Frank BH, Baker JC, Carter ND, Smith GD: Role of C-terminal Bchain residues in insulin assembly: the structure of hexameric LysB28ProB29-human insulin. Structure 3:615-22, 1995.

9 Howey CD, Bowsher RR, Brunelle RL, Woodworth JR: [Lys(B28), Pro(B29)]-human insulin—a rapidly absorbed analogue of human insulin. Diabetes 43:396-402, 1994.

10 Torlone E, Pampanelli S, Lalli C, Del Sindaco P, Di Vincenzo A, Rambotti AM, Modarelli F Epifano L, Kassi G, Perriello G, Brunetti P, Bolii G: Effects of the short acting insulin analog [Lys(B28),Pro(B29)] on postprandial blood glucose control in IDDM. Diabetes Care 19:945-52, 1996

11 DiMarchi RD: "insulin lispro-an insulin design for improved mealtime glucose control. Overcoming barriers to intensive glucose control: the inadequacies of conventional insulin therapy." Clinical Symposium, San Francisco, June 10, 1996.

12 Francis AJ, Hanning I, Alberti KGGM: The influence of insulin antibody levels on the plasma profiles and action of subcutaneously injected human and bovine short acting insulins. Diabetologia 28:330-34, 1985.

13 Fineberg NS, Fineberg SE, Anderson JH, Kirkett MA, Gibson RG Hufferd S: Immunologic effects of insulin lispro[Lys(B28),Pro(B29) Human Insulin] in IDDM and NIDDM patients previously treated with insulin. Diabetes 45:1750-54 1996.

14 Pfutzner A Kustner E Forst T, SchulzeSchleppinghoff B, Trautmann ME, Haslbeck M, Schatz H, Beyer J: Intensive insulin therapy with insulin lispro in patients with type I diabetes reduces the frequency of hypoglycemic episodes. Exper Clin Endocrinol & Diabetes 104:25-30, 1996.

15 Garg SK, Carmain JA, Braddy KC, Anderson JH, Vignaati L, Jennings MK, Chase HP: Pre-meal insulin analogue insulin lispro vs. Humulin R insulin treatment in young subjects with type I diabetes. Diabetic Med 13:47-52, 1995.

16 Wolfsdorf JI, Laffel LM, Pasquarello C, Vemon A, Herskowitz RD: Split-mixed insulin regimen with human ultralente before supper and NPH (isophane) before breakfast in children and adolescents with IDDM. Diabetes Care 14:1100-106, 1991.

17 Johnson NB, Kronz KK, Fineberg NS, Golden MP: Twice-daily humulin ultralente insulin decreases morning fasting hyperglycemia. Diabetes Care 15:1031-33, 1992.

18 Zinman B. Chiasson JL, Tildesley H. Tsui E, Strack TR: Insulin lispro in CSII: results of a double blind cross-over study. (Abstract) Diabetes 45 (Suppl 2):28A, 1996.

19 Walter HM, Timmler R. Mehnert H: Stabilized human insulin prevents catheter occlusion during continuous subcutaneous insulin infusion. Diabetes Res 13:75-77, 1990.

20 Heinemann L Heise T. Wahl LC, Trautman ME, Ampudia J. Starke AAR, Berge M: Prandial glycaemia after a carbohydrate-rich meal in type I diabetic patients using the rapid acting insulin analogue [Lys(B28),Pro(B29)] human insulin. Diabetic Med 13:625 -30, 1996.

21 Holleman F. Van Den Brand JJG, Hoven RARA, Van Der Linden JM, Van Der Tweel I, Hoekstra JBL, Erkelens D W: Comparison of LysB28, ProB29-human insulin analog and regular human insulin in the correction of incidental hyperglycemia. Diabetes Care 19:1426-29, 1996.

22 Ter Braak E, Woodwroth JR, Bianchi R. Cerimele B. Erkelens DW, Thisjssen JHH, Kurtz D: Injection site effects on the pharmacokinetics and glucodynamics of insulin lispro and regular insulin. Diabetes Care 19: 1437-40, 1996.

23 Menon RK, Cohen RM, Sperling MA, Cutfield WS, Mimouni F. Khoury JC: Transplacental passage of insulin in pregnant women with insulin-dependent diabetes mellitus: its role in fetal macrosomia. N Engl J Med 323 :309- 15, 1990.

24 Tuominen JA, Karonen SL, Melamies L, Bolli G Koivisto VA: Exercise-induced hypoglycaemia in IDDM patients treated with a short-acting insulin analogue. Diabetologia 38:106-11, 1995.

25 Freeman SL, O'Brien PC, Rizza RA: Use of human ultralente as the basal insulin component in treatment of patients with IDDM. Diabetes Res Clin Prac 12:187-92, 1991.

26 Shaw WN, Su KSE: Biological aspects of a new human insulin analogue: [Lys (B28), Pro (B29)]-human insulin (Abstract). Diabetes 40 (Suppl 1): 464A, 1991.

Christian D. Herter, MD, CDE, is an assistant professor of medicine in the Diabetes Care Center and the Family Medicine Center at the University of Washington, in Seattle.

Return To American Diabetes Association Home Page

Copyright © 1997 American Diabetes Association

Last updated: 6/3/97
For ADA Related Issues contact CustomerService@diabetes.org

For Technical Issues contact webmaster@diabetes.org