| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vol. 52, Issue 1, 1-10, March 2000
Section of Endocrinology, Veterans Affairs Medical Center, Phoenix, Arizona
I. Introduction
II. Short-Acting Insulin Analogs
A. Insulin Asp(B10)
B. Insulin Lispro
C. Insulin Aspart
III. Long-Acting Insulin Analogs
A. NovoSol Basal
B. HOE 901
C. Fatty Acid Acylated Insulins
IV. Future Directions
A. Increased Stability
B. Less Variability
C. Selective Action
D. Ultra Rapid Onset
E. Ultra Long Activity
F. Benefit without Metabolic Activity?
V. Closing Thoughts
References
 |
Abstract |
|---|
 Top References
|
|---|
Tight glucose control is essential to minimize complications in diabetic patients. However, the pharmacokinetic characteristics of the currently available rapid-, intermediate-, and long-acting preparations of human insulin make it almost impossible to achieve sustained normoglycemia. Until recently, improvements in insulin formulations were seriously limited as advances were only achieved in insulin purity, species, and characteristics of the retarding agent. The availability of molecular genetic techniques opened new windows to create insulin analogs by changing the structure of the native protein and to improve the therapeutic properties. The first clinically available insulin analog, Lispro, confirmed the hopes by showing that improved glycemic control can be achieved without an increase in hypoglycemic events. This requires, however, optimal basal insulin replacement, either by multiple daily injections of neutral protein Hagedorn (NPH) insulin or by insulin pump. Evidence suggests that short-acting insulin analogs would be better matched by a true basal insulin than by the erratically absorbed and rather short-acting NPH insulin. Therefore, future availability of long-acting analogs raises the hope to realize the true potential benefits of the currently available short-acting analog, Lispro, and of those still awaiting approval. The introduction of new short-acting and the first truly long-acting analogs, the development of analogs with increased stability, less variability and perhaps selective action will help to develop more individualized treatment strategies targeted to specific patient characteristics and to achieve further improvements in glycemic control.
| |
I. Introduction |
|---|
|
TopNextReferences
|
|---|
In nondiabetic
individuals, ingestion of food results in a rise of serum insulin
concentration to a maximum after 30 to 45 min, followed by a decline to
basal levels after 2 to 3 h. The pharmacokinetic characteristics
of the currently available rapid-, intermediate-, and long-acting
preparations of human insulin make it almost impossible to achieve
sustained normoglycemia. The onset of action of s.c. injected regular
insulin is too slow, and the duration of its action is too long to
mimic the insulin secretion pattern of a healthy individual during a
carbohydrate-containing meal (Heinemann et al., 1992
). As a result,
early postprandial hyperglycemia followed by an increased risk for
hypoglycemia before the next meal are present. Similarly, the available
intermediate/long-acting insulin preparations are unable to provide a
stable, continuous baseline insulin level. Instead, they cause peak
serum insulin levels at 3 to 4 h after s.c. injection and show
considerable inter- and intrasubject variations in their
bioavailability. The Diabetes Control and Complications Trial confirmed
the link between glycemic control and the complications of diabetes
(DCCT Research Group, 1993). Therefore, to achieve improved glucose
control, the need for new insulin preparations with a faster onset and shorter duration of action, and long-acting preparations with a more
flat time-action profile and less variable bioavailability became
apparent in the 1990s (Berger, 1989). Until recently, however, improvements in insulin formulations were seriously limited, because advances were only achieved in insulin purity, species, and
characteristics of the retarding agent. The availability of molecular
genetic techniques opened new windows to create insulin analogs by
changing the structure of the native protein and to improve the
therapeutic properties of it.
| |
II. Short-Acting Insulin Analogs |
|---|
|
TopPreviousNextReferences
|
|---|
Regular insulin has additional disadvantages. Because of its
relatively slow onset of action, regular insulin is optimally administered 30 to 60 min before meals. Due to the inconvenience and
difficulties with predicting the time of the meal, most patients do not
follow this advice, even when adequately instructed (Jorgensen and
Nielsen, 1990; Heinemann, 1995). Therefore, short-acting analogs that
can be injected immediately before meals could improve compliance with
treatment recommendations and the patients' overall satisfaction with
the regimen. Patients consider the opportunity to inject insulin
immediately before the meal an advantage, as it can increase flexibility and freedom in daily activities (Desmet et al., 1994). However, this cannot be achieved by human insulin, since above physiologic concentrations such as those present in the injectable preparations native human insulin forms dimers and hexamers, which inhibits its rapid absorption from the injection site (Mosekilde et
al., 1989). Therefore, a possible approach to facilitate
absorption and achieve rapid action is to develop analogs with a
decreased tendency to self-association. This can be accomplished by
changing the amino acid sequence of human insulin (Fig.
1).
Because of faster absorption, a substantial reduction in the
postprandial glucose excursion is expected with such analogs, and the
more rapid decline in the serum concentration of the analog should result in a reduced risk of late hypoglycemia compared with regular human insulin (Brange, 1997).
|
|
A. Insulin Asp(B10)
Elucidating the genetic basis for a case of familial
hyperproinsulinemia (it involves a single point mutation in the
proinsulin gene resulting in the substitution of aspartic acid for the
naturally occurring histidine for residue 10 of the B chain of insulin) led to the development of insulin Asp(B10), one of the first insulin analogs (Schwartz et al., 1987). Insulin Asp(B10) is absorbed twice as
rapidly as regular insulin and offers potential therapeutic benefits
(Nielsen et al., 1995). However, studies with Asp(B10) pointed out that
a potential problem with altering the amino acid sequence of human
insulin is that it can change the three-dimensional structure of the
molecule in a way that results in altered interaction with the insulin
receptor and the insulin growth factor
(IGF2)-I receptor.
In relative terms, the IGF-I receptor mediates cell growth to a greater
extent than the insulin receptor. Insulin analogs, which have a greater
affinity for the IGF-I receptor than human insulin, would therefore
have increased growth and mitogenic effects compared with the native
insulin molecule. The analog Asp(B10) has been demonstrated to have an
increased affinity for the IGF-I receptor and a decreased rate of
dissociation from the insulin receptor, as well as prolonged cellular
processing (Bornfeldt et al., 1991; Hamel et al., 1999). Asp(B10) has
been shown to have an increased cellular residence time, which results from an increased binding and internalization without a concomitant increase in intracellular degradation by insulin-degrading enzyme (Drejer, 1992; Hansen et al., 1996; Hamel et al., 1999). The
consequence is cellular retention of biologically active insulin. These
properties of Asp(B10) result in a greater metabolic effect compared
with human insulin, which would be a potential advantage. However, the
above characteristics also lead to increased mitogenic activity, and as
a result, suprapharmacological doses of Asp(B10) cause a dose-dependent
increase in the incidence of adenocarcinomas in laboratory animals
(Jorgensen et al., 1992; Drejer, 1992). Further clinical studies with
this analog were therefore halted.
B. Insulin Lispro
The first genetically engineered rapid-acting analog to become
available for the clinician was insulin Lispro, which was
approved for clinical use in Europe in April 1996 and in the United
States in June 1996. In insulin Lispro the normal sequence of proline at position 28 of the B chain and lysine at position 29 is reversed (LysB28,ProB29; Lispro) (Fig. 1). This reversal causes a decreased tendency for self-association, and as a result, faster absorption, higher peak serum levels, and shorter duration of action can be observed with insulin Lispro compared to regular insulin (Howey et al.,
1994). Importantly, the amino acid sequence changes in Lispro do not
affect its receptor-binding domain. Therefore, the affinity to the
insulin receptor of insulin Lispro is similar to that of regular
insulin, whereas Lispro's affinity for the IGF-I receptor is slightly
higher but not enough to cause a difference in its cell
growth-stimulating activity compared with regular insulin (Slieker et
al., 1991, 1994). Pharmacokinetic studies indicate that insulin Lispro
acts within 15 min, peaks in approximately 1 h, and disappears
within 2 to 4 h after s.c. injection (Howey et al., 1994; Torlone
et al., 1994). In clinical studies, as expected from a short-acting
analog, insulin Lispro achieved significant improvements in
postprandial glucose levels with a lower rate of hypoglycemic events
compared with regular insulin (Pfutzner et al., 1996; Anderson et al.,
1997; Brunelle et al., 1998). This can be observed even if insulin
Lispro is administered immediately before meals, and regular insulin is
injected 30 to 45 min before meals. Unfortunately, in most cases these
beneficial effects were not accompanied by improvements in glycosylated
hemoglobin values (Pfutzner et al., 1996; Anderson et al., 1997).
Besides the decrease in hypoglycemic events, the most likely
explanation for this is the inability of the currently used long-acting
insulins to provide true basal coverage. Therefore, increased
preprandial plasma glucose concentrations are present in patients on
insulin Lispro. Supporting this theory, a clinically and statistically
significant decrease of hemoglobin A1c levels was
seen when insulin Lispro was used with two or more instead of one daily
injections of NPH insulin (Karsidag et al., 1996; Ciofetta et al.,
1999). Therefore, adding a few units of NPH to Lispro at each meal,
combined with bedtime NPH (Del Sindaco et al., 1998; Ciofetta et al.,
1999; Lalli et al., 1999) can be recommended for the intensive therapy
of diabetes by multiple daily injections. This regimen may even
improve unawareness of, and impaired counter-regulation to hypoglycemia
(Lalli et al., 1999).
Similarly, because continuous s.c. insulin infusion (CSII) systems are
able to provide a reasonable basal insulin substitution, improved
glycosylated hemoglobin values would be expected with pump treatment
using insulin Lispro. After the stability of Lispro in insulin pump
systems had been confirmed (Lougheed et al., 1997), clinical trials
began to assess its effectiveness in CSII treatment. As assumed,
results with insulin Lispro in patients receiving CSII are promising as
evidenced by lower glycosylated hemoglobin values and improved
postprandial glucose levels when compared with patients receiving pump
treatment with regular insulin (Zinman et al., 1997; Renner et al.,
1999). Importantly, the improved glycemic control is achieved without
an increase in the number of hypoglycemic events. A potential
disadvantage of using insulin Lispro in pump systems as opposed to
regular insulin is that because of its more rapid disappearance,
patients might be at more risk for developing ketoacidosis in case of
catheter occlusion or pump malfunction (Pen et al., 1996). However,
this was not confirmed by a recent study (Attia et al., 1998) in which
no difference was found between patients receiving CSII treatment with
Lispro or regular insulin with respect to the rate of rise in plasma glucose or serum ketone levels after disrupting s.c. infusion. The
frequency of catheter occlusion or other site-related problems is
similar with Lispro and buffered regular insulin (Zinman et al., 1997;
Renner et al., 1999).
A protamine formulation of insulin Lispro with prolonged action
(neutral protamine Lispro) has been developed and shown to be
suitable as an intermediate-acting agent or as part of premixed preparations of Lispro and neutral protamine Lispro (25/75 and 50/50)
(Janssen et al., 1997; Roach et al., 1999). Compared with human insulin
mixtures, twice-daily administration of insulin Lispro mixtures
resulted in improved postprandial glycemic control, similar overall
glycemic control, and less nocturnal hypoglycemia, as well as offering
the convenience of dosing closer to meals.
Managing diabetes in patients with end-stage renal disease is often
problematic, because renal failure interferes with the metabolism of
glucose and insulin. Many of these diabetics have wide fluctuations in
their daily blood glucose profile. The action of regular insulin may be
prolonged as a consequence of the failure of renal insulin degradation,
making the dose-effect profile of insulin difficult to control, and
hypoglycemia more likely. There is evidence that using insulin Lispro
may make the calculation of insulin requirements easier and help to
avoid large fluctuations in blood glucose levels of these patients
(Jehle et al., 1999).
Based on the limited available data on its long-term effectiveness, it
appears that insulin Lispro remains effective in treating diabetic
patients up to 5.4 years of treatment (Garg et al., 1999). No
differences have been reported between insulin Lispro and regular insulin in the likelihood of developing allergic reactions, adverse events, or abnormal laboratory values (Anderson et al., 1994). The
immunogenicity of insulin Lispro is similar to that of regular insulin
(Fineberg et al., 1996). Antibodies specific against insulin Lispro
rarely develop and do not affect dose requirements (Roach et al., 1996;
Garg et al., 1999). Interestingly, patients have been reported in whom
severe resistance to human insulin because of antibody formation was
successfully overcome by switching them to insulin Lispro (Henrichs et
al., 1996; Lahtela et al., 1997).
Despite the difficulties with standardizing quality of life
assessments, the available data are surprisingly consistent and show a
greatly increased treatment satisfaction among patients receiving
Lispro by CSII or as multiple injections (Pfutzner et al., 1996;
Kotsanos et al., 1997; Renner et al., 1999). This can improve patient
motivation and compliance, which are very important components of
treatment success in diabetic patients.
C. Insulin Aspart
The next example of changing the amino acid sequence of the
insulin molecule to achieve short-acting insulin analogs is insulin aspart (AspB28), in which substitution of proline with the charged aspartic acid is carried out to reduce self-association of the molecule
(Brange et al., 1988) (Fig. 1). Preclinical studies of insulin aspart
have demonstrated that receptor interaction kinetics with the insulin
receptor and with the IGF-I receptor are equivalent to those seen with
human insulin (Drejer, 1992), and an equivalent metabolic effect of
insulin aspart and human insulin has been shown with i.v.
administration (Kang et al., 1991). When administered i.v., insulin
aspart shows a similar safety profile as human insulin (Dall, 1999).
Insulin aspart has been shown to be absorbed twice as fast as human
insulin and to reach maximum concentrations twice as high, whereas its
duration of action is shorter (Heinemann et al., 1998; Lindholm et al.,
1999). As expected, the postprandial glucose control achieved with this
analog is superior to regular human insulin, although their
bioavailability is comparable (Lindholm et al., 1999). These results
are consistent with those reported with the other short-acting analog,
Lispro, but there is evidence that the improvement in postprandial
control can be achieved without deterioration of late postprandial
plasma glucose concentrations (Home et al., 1998). The expectation of
lower rates of hypoglycemia also seems to have been met with insulin
aspart as evidenced by a recent multicenter trial of type 1 diabetic
patients, who showed more than a 50% reduction in major hypoglycemic
events compared with human insulin (Home et al., 1998). Importantly,
this analog retains its beneficial pharmacodynamic properties in a
stable 30/70 premixed formulation as it shows a significantly greater metabolic effect in the first 4 h than the 30/70 mixture of human insulin (Weyer et al., 1997). Because of its promising characteristics, studies are presently underway to evaluate long-term metabolic control
with insulin aspart.
| |
III. Long-Acting Insulin Analogs |
|---|
|
TopPreviousNextReferences
|
|---|
A number of alterations of the insulin molecule by genetic
engineering are currently being tested to retard and stabilize absorption kinetics of long-acting insulin preparations. One
possibility to prolong insulin action is to elevate the isoelectric
point of human insulin from pH 5.4 toward neutral by developing analogs with more positively charged amino acids (Rosskamp and Park, 1999). This will make the analog less soluble at the neutral pH of the injection site, and the injection of the analog into the s.c. tissue
will result in crystallization of the molecules, causing delayed
absorption into the circulation.
A. NovoSol Basal
One of the first analogs developed by protein-engineering
technology based on the above therapeutic goals was NovoSol Basal (B27Arg, A21Gly, B30Thr-NH2). Whereas the task of prolonged absorption was successfully completed with that analog as evidenced by longer T50 than that of Ultratard HM insulin,
nearly two times higher doses of the analog were required to achieve
compatible glucose control. Also, whereas NovoSol Basal showed less
intraindividual variability in its action, the interindividual
variation remained high. Therefore, and because of its reduced
bioavailability NovoSol Basal was withdrawn from further studies
(Jorgensen et al., 1989; Jorgensen and Drejer, 1990).
B. HOE 901
HOE 901 (insulin glargine) is a new long-acting biosynthetic human
insulin analog currently being investigated by Hoechst, which results
from elongation of the C-terminal end of the insulin B chain by two
glycine residues, as well as substitution of the A21 asparagine residue
with glycine (Fig. 1). This analog behaves with respect to insulin
receptor binding, receptor autophosphorylation, phosphorylation of
signaling elements, and promotion of mitogenesis like regular human
insulin (Berti et al., 1998). Moreover, the IGF-I receptor-mediated
growth promoting activity of HOE 901 in muscle cells and the maximal
metabolic activity of this analog are not different from those of
native human insulin (Bahr et al., 1997). However, its therapeutic
properties and potentials are remarkable and different from human
insulin. HOE 901 was shown to exert a glucose-lowering effect for
24 h after a single daily injection without a pronounced plasma
peak (Dreyer et al., 1994). In one of the first small short-term
clinical studies investigating this analog in 1996, once-daily
injections of HOE 901 resulted in similar glycemic control as four
daily injections of the same total units of NPH in type 1 diabetics
(Talaulicar et al., 1996). Later on, the characteristics of HOE 901 have been investigated in both type 1 and type 2 diabetic patients. In
Phase II trials conducted in Europe and the United States with type 1 diabetics, once-daily injections of HOE 901 along with premeal regular
insulin achieved significantly lower fasting plasma glucose levels
(Rosenstock et al., 1998) and hemoglobin A1c
values compared with patients on NPH and regular insulin (Pieber et
al., 1998). Remarkably, the better glucose control was associated with
similar or even lower incidence of hypoglycemia. Studies of type 2 diabetic subjects showed similar fasting plasma glucose values with one
injection of HOE 901 as those found with one or two injections of NPH
insulin. Again, the incidence of hypoglycemia was similar or lower
among patients on HOE 901 (Matthews and Pfeiffer, 1998; Raskin
et al., 1998; Rosenstock et al., 1999). The technical difficulties with blinding the studies comparing NPH and HOE 901 should be noted, as the
two preparations can be easily identified, because HOE 901 is a clear
solution as opposed to the cloudy solution of NPH. A potential problem
with altering the structure of the insulin molecule is increasing the
risk of antibody development and adverse reactions at the site of
injection. Importantly, adverse events and injection-site reactions
associated with HOE 901 were not different from those found with NPH
insulin, and antibody formation was also similar with the two preparations.
C. Fatty Acid Acylated Insulins
Another way of prolonging insulin action is to couple the insulin
molecule to nonesterified fatty acids, which bind to albumin. Albumin
serves as a multifunctional transport protein that binds a wide variety
of endogenous substances and drugs. Albumin is present in the s.c.
tissue fluid with a slow disappearance rate. Binding insulin to albumin
can therefore retard the absorption of the molecule and prolong its
action. The binding to albumin apparently involves both nonpolar and
ionic interactions with the protein (Kurtzhals et al., 1995). Acylation
of the insulin molecule is usually performed in the side chain of
lysine at position 29 of the B chain. Such insulin analogs are
currently being studied by Lilly [N(
)-palmitoyl
Lys(B29); WW99-S32] and Novo Nordisk [LysB29-tetradecanoyl, des(B30);
NN304] (Table 1).
In animal studies, the time for 50% disappearance from the s.c. space
of NN304 was 14.3 h, significantly longer than that of NPH insulin
(10.5 h) and with significantly less interanimal variation (Markussen
et al., 1996). In healthy volunteers the metabolic response induced by
s.c. injection of NN304 does not show the pronounced peak seen with NPH
insulin in an identical dose. NN304 also shows a slower onset of
action, as indicated by a significantly higher
tmax compared with NPH insulin
(Heinemann et al., 1999). Importantly, the binding of NN304 has been
shown to be independent of the binding of drugs in the two major
binding pockets that are located in domains IIA and IIIA of the albumin molecule. Thus, NN304 is unlikely to be involved in clinically significant drug interactions at the albumin binding level (Kurtzhals et al., 1997).
In a diabetic animal model, the duration of action of the other analog
WW99-S32 human insulin administered i.v. was nearly twice that of
unmodified human insulin, and the plasma half-life was nearly 7-fold
that of the unmodified protein. Administered s.c., WW99-S32
human insulin had a longer duration of action, a flatter, more basal,
plasma insulin profile, and a lower intersubject variability of
response than the intermediate-acting insulin suspension Humulin L
(Myers et al., 1997). In healthy volunteers, this analog showed a
highly reproducible linear pharmacokinetic profile but less potency
when compared with NPH (Howey et al., 1997). The latter finding was
subsequently confirmed in C-peptide negative patients (Radziuk et al.,
1998). Based on the results with insulin acylation, derivatization with
albumin-binding ligands could be applicable generally to prolong the
action profile of peptide drugs (Kurtzhals et al., 1995).
| |
IV. Future Directions |
|---|
|
TopPreviousNextReferences
|
|---|
A. Increased Stability
Insulin is not a stable chemical entity. A variety of chemical
changes of the primary structure affect insulin during handling, storage, and even use. Insulin decomposition is mainly due to two
categories of chemical reactions: hydrolysis and intermolecular transformation leading to covalent insulin dimers. Identification of
the residues undergoing chemical changes during storage allows designing insulin analogs with improved stability. The advantage of
such analogs would be prolonged shelf-life and more convenient storage
conditions. Improved stability is also essential for pump usage. The
above discussed Asp(B10) analog has increased stability but is
unfortunately not suitable for clinical use (Brems et al., 1992).
Substitution of AsnB3 by Gln, and AsnA21 by Ala or Gly results in
analogs with 30 times less deamination and 10 times reduced formation
of covalent dimers (Brange 1997). Designing and testing more analogs
with increased stability remains an important task for the future.
B. Less Variability
The high intra-and interindividual variability of the response to
identical insulin doses is a serious problem for patients and their
clinicians as well, and can hamper the achievement of reasonable
glycemic control without the risk for hypoglycemic events (Heinemann et
al., 1998). There are two explanations for the variability of insulin
responsiveness. Pharmacokinetic variability can result from variations
in insulin absorption, leading to different plasma concentrations of
insulin after s.c. injection of the same doses (Lauritzen et al., 1979;
Binder et al., 1984). Pharmacodynamic variability, on the other hand,
can be caused by differences in insulin action, causing different
metabolic effects by similar plasma insulin concentrations (Ziel et
al., 1988). The above discussed long-acting analog NovoSol Basal shows
less intraindividual variation in its pharmacokinetics than the longest
acting currently available human insulin preparation Ultratard HM
(Jorgensen et al., 1989). Similarly, a decreased variability in serum
insulin concentrations compared with regular human insulin has been
shown after s.c. injections of the analog insulin Lispro (Antsiferov et
al., 1995). Nevertheless, developing insulin analogs with lower
pharmacokinetic and pharmacodynamic variability remains to be an
important task.
C. Selective Action
Insulin influences glucose metabolism by inhibiting hepatic
glucose production and stimulating peripheral glucose disposal. Insulin
analogs with relatively greater effect on hepatic glucose production
could offer potential therapeutic benefits for selected patients.
Proinsulin, the single chain precursor of insulin is more effective in
the liver than in the periphery (Revers et al., 1984; Lavelle-Jones et
al., 1987). Reasons for this selectivity are not fully
understood, but the increased molecular size of proinsulin compared
with insulin has been proposed as a potential mechanism. Endothelial
cells in peripheral tissues limit the transfer of substances from the
circulation into the tissues with a rate inversely related to the
molecular size of the transferred substance. However, hepatocytes are
freely in contact with all blood constituents in the hepatic sinusoids.
Dose requirements for proinsulin are approximately 4-fold higher than
for human insulin, and there is a possible association between its use
and myocardial infarction (Spradlin et al., 1990; Galloway et al.,
1992). Proinsulin was therefore withdrawn from clinical trials, but the
recognition of its selective action has stimulated the search for
analogs with greater hepatic effects. Two insulin analogs with
increased molecular size due to covalent dimerization have been shown
to have a greater effect on hepatic glucose production than peripheral glucose disposal after i.v. administration (Shojaee-Moradie et al.,
1995). These dimeric analogs [N(
)B1,
N()B'1,-suberoyl insulin dimer, and N()B29,
N()B'29,-suberoyl insulin dimer] are probably not
suitable for clinical use because of their relatively low potency, but
they confirm the possibility that analogs with selective action due to
increased molecular size might be developed. Another interesting
finding is that N
1-thyroxyl insulin and
N1-thyroxyl-aminohexanoyl insulin also show greater
selectivity for hepatic glucose production (Shojaee-Moradie et al.,
1998). These insulin analogs bind thyroid hormone-binding proteins to
form high molecular weight complexes. These findings provide further
support to the theory that the reduced peripheral insulin-like effect
could be due to reduced transcapillary access to peripheral insulin
receptor sites, which results from high molecular weight.
Another possibility for selectivity of an analog, either to a specific
tissue or for a specific action (e.g., increased mitogenicity compared
with metabolic effects) would be altered cellular metabolism of the
analog. Reduced degradation would prolong cellular residence of the
material and alter activity profile. The analog Asp(B10) is an example
of this (Hamel et al., 1999). For future development of specific
analogs more information is needed on properties of the insulin
molecule important for different biological activities, e.g.,
carbohydrate versus fat versus protein metabolic effects (Duckworth,
1997).
D. Ultra Rapid Onset
Although significant improvements in postprandial plasma glucose
levels can be achieved with the presently available short-acting analog, insulin Lispro, even when it is injected immediately before meals, there is evidence that its optimal administration would be
actually 15 to 30 min before meals (Rassam et al., 1999). When administered at least 15 min before meals, Lispro achieves a greater improvement in postprandial values as opposed to being injected immediately before meals. This suggests that developing even more rapidly absorbed short-acting analogs could offer potential benefits.
E. Ultra Long Activity
Some insulin-requiring patients simply do not have the background
or resources needed for insulin treatment. They may not have access to
a refrigerator or are unable to use insulin without getting help
because of disabilities. Potentially these patients could use ultra
long-acting analogs that could be injected once weekly or for even
longer periods of time. This type of preparation obviously would not
provide good control but could offer basal coverage sufficient to
prevent ketoacidosis or other acute complications. The concept may seem
utopistic at first sight, but a recent study reported that a single
s.c. injection of a new analog, in which two 9-fluorenylmethoxycarbonyl
moieties are covalently linked to the phenylalanine at position B1 and
to the lysine at B29 of human insulin, normalized blood sugar levels of
rats for 2 to 3 days with streptozotocin-induced diabetes (Gershonov et
al., 1999). The analog itself has only 1 to 2% of the biological
potency of insulin, but undergoes a time-dependent spontaneous
conversion to fully active insulin. The conversion takes place slowly
under physiological conditions, with a
t1/2 of 12 (!) days.
F. Benefit without Metabolic Activity?
The insulin analog Asp(B25) does not bind to the insulin receptor
or IGF-I receptor and has no hypoglycemic effect (Drejer et al., 1991).
However, this analog has been shown to prevent diabetes in an animal
model of spontaneous diabetes that shares many features of human type 1 diabetes (Karounos et al., 1997). The analog prevented diabetes in the
animals even when it was initiated after the onset of extensive
lymphocytic infiltration of the pancreatic islets. The
mechanism
because it did not involve metabolic effectsappears to be
immunological. Although it is yet unclear, whether Asp(B25) can be used
for preventing diabetes in prediabetic children and young adults, the
theory of using analogs without the potentially harmful hypoglycemic
effects for diabetes prevention is certainly an interesting one.
| |
V. Closing Thoughts |
|---|
|
TopPreviousNextReferences
|
|---|
The need for nearly optimal glucose control in diabetics to
minimize complications clearly exists. Without insulin analogs, however, this can only be accomplished at the expense of hypoglycemic reactions. The DCCT trial demonstrated that a 10% improvement of
glycosylated hemoglobin levels results in a 43% improvement of
retinopathy but is accompanied by an 18% increase of severe hypoglycemic episodes (DCCT Research Group, 1993). The first clinically available insulin analog, Lispro, opened new hopes by showing that
improved glycemic control can be achieved without an increase in
hypoglycemic events. This requires, however, optimal basal insulin
replacement, either by multiple daily injections of NPH or by CSII.
Evidence suggests that short-acting insulin analogs would be better
matched by a true basal insulin than by the erratically absorbed and
rather short-acting NPH insulin (Home et al., 1998). Therefore, future
availability of long-acting analogs raises the hope of realizing the
true potential benefits of the currently available short-acting analog,
Lispro, and of those still awaiting approval. The introduction of new
short-acting and the first truly long-acting analogs, the development
of analogs with increased stability, less variability, and perhaps
selective action will help to develop more individualized treatment
strategies targeted to specific patient characteristics and to achieve
further improvements in glycemic control.
| |
Footnotes |
|---|
1 Address for correspondence: William C. Duckworth, Section of Endocrinology, Veterans Affairs Medical Center, 650 E. Indian School Rd., 111E, Phoenix, AZ 85012.
| |
Abbreviations |
|---|
IGF, insulin growth factor;
NPH, neutral protein Hagedorn;
CSII, continuous s.c. insulin infusion;
NN304, LysB29-tetradecanoyl, des(B30);
WW99-532, N()-palmitoyl Lys(B29);
HOE 901, insulin glargine.
| |
References |
|---|
|
TopPrevious |
|---|
randomization failure or HPI effect (Abstract)
Diabetologia
33 (Suppl):
A60.
0031-6997/00/5201-0001$03.00/0
PHARMACOLOGICAL REVIEWS
Copyright © 2000 by The American Society for Pharmacology and Experimental Therapeutics
This article has been cited by other articles:
|
|
 |
|
  T. P. Ciaraldi, S. A. Phillips, L. Carter, V. Aroda, S. Mudaliar, and R. R. Henry Effects of the Rapid-Acting Insulin Analog Glulisine on Cultured Human Skeletal Muscle Cells: Comparisons with Insulin and Insulin-Like Growth Factor I J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5551 - 5558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Peter, S. D. Luzio, G. Dunseath, A. Miles, B. Hare, K. Backx, V. Pauvaday, and D. R. Owens Effects of Exercise on the Absorption of Insulin Glargine in Patients With Type 1 Diabetes Diabetes Care, March 1, 2005; 28(3): 560 - 565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Duttaroy, P. Kanakaraj, B. L. Osborn, H. Schneider, O. K. Pickeral, C. Chen, G. Zhang, S. Kaithamana, M. Singh, R. Schulingkamp, et al. Development of a Long-Acting Insulin Analog Using Albumin Fusion Technology Diabetes, January 1, 2005; 54(1): 251 - 258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Devendra, E. Liu, and G. S Eisenbarth Type 1 diabetes: recent developments BMJ, March 27, 2004; 328(7442): 750 - 754. [Full Text] [PDF]
|
|
|
|
|
|
I. Rakatzi, S. Ramrath, D. Ledwig, O. Dransfeld, T. Bartels, G. Seipke, and J. Eckel A Novel Insulin Analog With Unique Properties: LysB3,GluB29 Insulin Induces Prominent Activation of Insulin Receptor Substrate 2, but Marginal Phosphorylation of Insulin Receptor Substrate 1 Diabetes, September 1, 2003; 52(9): 2227 - 2238. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Doyle and J. M. Egan Pharmacological Agents That Directly Modulate Insulin Secretion Pharmacol. Rev., March 1, 2003; 55(1): 105 - 131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. P. Ciaraldi, L. Carter, G. Seipke, S. Mudaliar, and R. R. Henry Effects of the Long-Acting Insulin Analog Insulin Glargine on Cultured Human Skeletal Muscle Cells: Comparisons to Insulin and IGF-I J. Clin. Endocrinol. Metab., December 1, 2001; 86(12): 5838 - 5847. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||
