The present invention relates to a specific and sensitive bio-analytical method for
detection of insulin or insulin analogues in plasma, serum or any other biological
fluid. It particularly relates to labelling of biological compound with stable
isotopic nitrogen and bio-analytical method using 5 solid phase extraction and
liquid chromatography with tandem mass spectrometric detection.
BACKGROUND OF INVENTION:
For proteins based biological drugs, it is necessary to detect proteins such as
insulin contained in the plasma, serum or any other biological fluid, and measure
10 them quantitatively in order to elucidate the function of the protein, or for
testing and screening of the protein, or to measure the systemic exposure of the
protein. Therefore, if a more general method for quantifying the amount of
insulin with a high sensitivity is established, it becomes possible to estimate drug
kinetics during preclinical testing or in clinical trials. Thus, it would be possible to
15 estimate drug kinetics in human at an early stage of drug development.
Typically, radioimmunoassay (RIA) or ELISA methods have been used for
quantifying insulin in biological samples. Attempts to generate antibodies
specific to Insulin analogue ‘IN-105’ were not successful, as the immunisation
cycles yielded antibodies which showed identical binding to human insulin and
20 IN-105, indicating that the addition of the small alkyl polyethylene glycol (PEG)
does not alter the structure significantly enough to differentiate the molecule
immunologically. The difference in molecular weight between the two molecules
is 217 Daltons due to the alkyl PEG in IN-105. Hence, a specific ELISA based
estimation assay could not be developed for IN-105.
25 Conventionally, a method using electrophoresis such as two-dimensional
electrophoresis was conducted for quantifying biogenic proteins. With this
method, detection and quantification were performed by staining the protein to
3
be quantified, or by autoradiography, or by using an antibody specific to a
particular protein (western blot). However, it was difficult to apply these
methods for quantifying a high molecular protein like insulin or insulin analogue
since it is not possible to electrophorese insoluble proteins or high molecular
5 proteins.
On the other hand, there have been significant advances in the field of mass
spectrometry, and the technique have been considered and used for detecting or
measuring various biological materials. Typically an LC/MS/MS instrumentation
involve liquid chromatography (LC) for separation of analytes, followed by
10 ionisation chamber where the sample undergoes desolvation and ionisation and
the masses are then detected by mass spectrometer. Advancements in the
design of mass spectrometers supported consistent fragmentation of such a
large molecule in its intact form, allowing the use of multiple reaction monitoring
(MRM) mode, making the method highly specific to IN‐105.
15 The human insulin protein is composed of 51 amino acids, and has a molecular
mass of 5808 Da. It is a dimer of an A‐chain and a B‐chain, which are linked
together by disulphide bonds. Several analogues of human insulin are available
such as IN‐105, Insulin Aspart, Insulin Lispro, and Insulin Glargine. These insulin
analogues are closely related to the human insulin structure, and were
20 developed for specific aspects of glycaemic control in terms of fast action
(prandial insulin) and long action (basal insulin).
A specific LC/MS/MS method could be developed for measurement of intact IN‐
105 molecule in plasma. Sample preparations could be carried out using offline
solid phase extraction of the analyte, followed by online solid phase extraction
25 before analysis of the samples by LC/MS/MS.
4
OBJECT OF INVENTION:
The object of present invention is to provide a sensitive analytical method for the
detection and quantification of insulin or insulin analogues IN-105 in human
plasma at the concentration range of 0.20 ng/ml to 50.0 ng/ml.
5 SUMMARY OF INVENTION:
The present invention relates to a bio-analytical method for the detection of
insulin or insulin analogues such as IN-105 in human plasma.
The method comprises labelling of insulin or insulin analogue by a stable isotope
of nitrogen i.e. 15N. Labelling is attained by stable isotope of nitrogen (15N) by
10 providing labelled nitrogen in growth and fermentation medium. The cells are
grown in labelled medium therefore almost all nitrogen atoms in insulin or
insulin analogues are substituted with stable isotope 15N. The labelled nitrogen
source used for the method is ammonium sulphate [(15NH4)2SO4].
Another aspect of the present invention relates to determination of intact
15 molecule of the labelled insulin or insulin analogues which is further determined
using solid phase extraction and liquid chromatography with mass spectrometric
detection.
BRIEF DESCRIPTION OF DRAWINGS:
Fig 1 is flowchart of scheme of overall process
20 Fig 2 is flowchart of upstream process
Fig 3 is process of conversion of IN-105 precursor into IN-105 through trypsin and
carboxypeptidase B treatment.
Fig 4 is chromatography peak for Reagent Blank
Fig 5 is chromatography peak for Double Blank
25 Fig 6 is chromatography peak for Matrix Blank
5
Fig 7 is Calibration Standard at 0.200 ng/mL LLOQ (Lower Limit of Quantification)
Fig 8 is Calibration Standard 50.0 ng/mL ULOQ (Upper Limit of Quantification)
Fig 9 is QC 0.200 ng/mL LLOQ
Fig 10 is QC 0.600 ng/mL LoQC
5 Fig 11 is QC 20.0 ng/mL MeQC
Fig 12 is QC 40.0 ng/mL HiQC
Fig 13 is AUCt v DOSE graph of pharmacokinetic study
Fig 14 is Cmax v DOSE graph of pharmacokinetic study
Fig 15 is Plot of Mean Plasma Concentration for IN-105 30mg
10 DESCRIPTION OF INVENTION:
Present invention relates to bio-analytical method for detection of insulin or
insulin analogues using liquid chromatography with tandem mass spectrometry.
In particular, the present invention relates to a highly specific method for
quantification of IN-105 in biological matrix.
15 In one of the embodiments, biological matrix is whole blood, blood serum, blood
plasma, urine, fermentation broth or buffer.
The present method allows us to determine IN-105 exposure under variety of
disease conditions. Method can distinguish IN-105 from co-administered
insulin/insulin analogue and endogenous insulin. Due to the use of mass
20 spectrometry, the method can be used to determine different insulin analogues
simultaneously as long as there is a difference in the molecular weight. The
method can be further developed to additionally determine C-peptide levels
along with insulin or its analogues. The method in general provides a useful
alternative to immunological methods.
25 The scheme of the process is depicted in Fig 1.
6
IN-105, is a novel ‘rapid acting insulin analogue’ being developed for oral delivery
as disclosed and described in US 9,101,596. The molecule has an amino acid
sequence identical to that of human insulin, except that it has been conjugated
with a short chain alkyl-polyethylene glycol molecule at the ε-amino group of
lysine at the 29th position 5 of the B-chain.
Due to the high specificity of the method, IN-105 could be selectively quantified
in presence of any human insulin or co-administered insulin like insulin Glargine.
This makes the method very useful in determining pharmacokinetic exposure of
IN-105 following oral administration as demonstrated from the pharmacokinetic
10 profile obtained from 10 mg dosing in patients with T1D (Type 1 Diabetes).
Upstream process involves labelling of insulin or insulin analogues by stable
isotopic non-metal elements such as nitrogen (15N), sulphur (34S), carbon (13C),
oxygen, or hydrogen. The isotopic nitrogen (15N) source in the culture medium is
at least one of ammonium sulphate, Ammonium phosphate, Ammonium
hydroxide, Methylamine, Urea; the isotopic carbon (1315 C) source in the culture
medium is at least one of Methanol, Glycerol, Glucose, Sorbitol; isotopic sulphur
(34S) source in the culture medium is at least one of Calcium sulphate,
Magnesium sulphate, Potassium sulphate; D20 for hydrogen and H2O18 for
oxygen labelling. The scheme of the process is depicted in Fig 2.
20 Downstream process involves validation method for determining
pharmacokinetic exposure in normal and disease plasma from human subjects.
The validation study included linearity, accuracy and precision, selectivity and
specificity, repeatability and reproducibility.
The upstream process begins with seed culture preparation, which involves
25 inoculation of seed flasks from the culture of cell bank prepared by using a
recombinant strain, Pichia pastoris. The strain carries a gene which codes for the
expression of insulin precursor. The seed flasks were grown over a period to
7
increase the cell mass which will be used as the inoculum to start the production
fermenter. Fermentation is carried out in two phases viz. batch phase and fed
batch phase. During batch phase, the fermentation is performed to increase the
cell mass using glycerol as carbon source and during fed batch phase methanol is
used to induce the secretion of insulin precursor 5 along with 15N labelled
ammonium sulphate as the only nitrogen source during process. The insulin
precursor protein with labelled 15N is released into the supernatant during
fermentation.
After fermentation, the downstream purification process serves to isolate the
10 insulin precursor from the cell-free supernatant and convert it into a Methoxytriethyleneglycol-
propionyl-NεB29 recombinant insulin precursor, which through
a trypsin and carboxypeptidase B catalysed reaction gets converted into insulin.
Multiple HPLC steps are used to resolve close product-related impurities during
this process. Multiple chromatography steps are used to resolve product related
15 impurities. After the purification steps, the product is crystallized and lyophilized.
Estimation methods specific to an insulin analogue is useful in clinical
investigations in differentiating the exogenously given insulin analogue from
endogenous insulin. In the context of IN-105, a method for specifically estimating
IN-105 provides unequivocal evidence for absorption of the insulin analogue
20 upon oral delivery. The implications of such a finding are of immense importance
in clinical development of oral insulin. IN-105 being a rapid acting insulin, it is
expected to be co-administered with a basal insulin e.g. Glargine, under specific
therapy conditions. Therefore, specificity of the LC/MS/MS method was tested in
presence of basal insulin Glargine.
Upon obtaining 1525 N labelled product of insulin or insulin analogue, the product
was used as internal standard in analytical method for determination of IN-105 in
plasma. Analytical method can be used for the determination of insulin or insulin
8
analogues in human plasma or other biological fluids containing K2EDTA or
K3EDTA (either Di-potassium or Tri-potassium EDTA).
Analysis employs liquid chromatography - tandem mass spectrometry using
TurbolonSpray, in positive ion, multiple reaction monitoring mode. Samples are
prepared using off-line Solid-phase extraction (5 SPE) extraction, followed by online
SPE in 96-well format.
The method relates to determination of insulin or insulin analogues in human
plasma. Preferably, the method is employed to determine insulin or insulin
analogues over the concentration range 0.200 ng/mL to 50.0 ng/mL.
10 UPSTEAM PROCESS:
Inoculum Flask: The procedure for labelling begins with seeding of cell line in
inoculum flask (either directly or through seed flask) that contain medium,
designed for same as shown in table 1.
1 Composition
of Medium:
Yeast nitrogen base (YNB) without amino acid and
ammonium sulphate (2.01g) + Ammonium sulphate (6g)
The above contents were added and volume was made
up to 360ml with miliQ water.
2 Composition
of
Phosphate
buffer:
K2HPO4 (2.1g) + KH2PO4 (6.5g)
The above contents were added and volume was made
up to 120ml with miliQ water.
3 Glycerol:
Glycerol (7.5g)
Glycerol was added and volume was made up to 120ml
with miliQ water.
Table 1: medium composition for inoculum flask
15 1. All the three components were separately autoclaved at 121°C for 60min.
9
2. After autoclaving, the three components were mixed in 2000mL flask and
was inoculated with one culture vial.
3. The flask was kept in shaker incubator for incubation at 30°C± 1°C for 24 ±
2 hr.
5 Fermentation Procedure
When OD (Optical Density) reached to desired level (10), the inoculum from
inoculum flask was transferred further for fermentation. Fermentation involved
batch phase and methanol fed-batch.
Composition of fermentation medium described in table 2-4.
Raw Material Conc. (g/L) Conc. (g/2.5L)
Glycerol 40 100.0
H3PO4
(Ortho Phosphoric Acid)
15.7 39.3
K2SO4 18.2 45.5
MgSO4.7H2O 14.9 37.3
KOH 2.1 5.3
CaSO4.2H2O 0.5 1.3
Ammonium sulphate
(15N labeled)
20 50.0
10
Table 2: Composition of Fermentation Medium
10
Age (h)
20% (w/v) Ammonium Sulphate (15N labeled)
Feed Rate (g/h)
22-60 4±2
60-EOF 12±2
Table 3: Ammonium sulphate feed stock
Sr.
No.
Composition
Quantity (gm)
for 1000ml
1 Copper Sulphate (CuSO4.5H2O) 6.0
2
Manganese Sulphate
(MnSO₄·H₂O)
3.0
3 Sodium Iodide (NaI) 0.08
4 Zinc Chloride (ZnCl2) 20.0
5
Sodium Molybdate
(NaMoO4.2H2O)
0.2
6 Boric Acid (H3BO3) 0.02
7 Cobalt Chloride (CoCl2) 0.5
8 Iron Sulphate (FeSO4.7H2O) 65.0
9 Sulphuric Acid (H2SO4) 5.0 ml
Table 4: Trace Element Composition
11
Biotin solution preparation:
Biotin (0.2g/L) was dissolved in potable water and filter sterilized through 0.2μ
before use.
Procedure:
1. Fermentation medium was then autoclaved 5 at 121°C for 60min.
2. Fermentation started with 30°C with 350 rpm.
3. After connecting the fermenter, pH was adjusted to 4.9 with 20% w/w
NaOH (sodium hydroxide) solution.
4. Sterile trace salt solution and biotin was added (4.35 ml/L of each). As
10 soon as trace salt was added, DO was dropped.
5. 12.0 mL of trace salt solution, 12mL of D-biotin solutions and 2.5gm of
20% H2SO4 (w/w) are added per litre of methanol. Methanol is filter
sterilized through 0.2μ before feeding into fermenter.
6. 6. Once the OD in the Inoculum flask was reached to 10-15, 250 ml of
15 inoculum was added.
7. 7. RPM was increased from 350 to 800 in 7-8 steps.
Batch Phase
Parameters of batch phase fermentation are
1. pH: 5.0±0.2
20 2. Temperature: 30±2°C
3. Dissolved Oxygen (DO): 30%
DO was first maintained by increasing the RPM manually. When DO was reached
to 40%, RPM was increased by 200 in one step up to maximum.
12
Once RPM was reached to maximum, DO (Dissolved Oxygen) was put in cascade
of gas mix only starting with minimum of 10% and maximum of 30% and then
increased as per the requirement.
When DO started to shoot up and pH started rising, sample was taken and
analysis for Wet Cell Weight was done. pH was then 5 changed to set point 6.0
whereas, temperature was adjusted to 23°C.
When batch phase was over, after one hour methanol phase was started.
Methanol Fed-Batch
Parameters of methanol fed-batch fermentation (Methanol induction phase) are
10 1. pH: 6.0±0.2
2. Temperature: 23±2°C
Once the DO was raised in batch phase, methanol addition was started. (Refer
table 4).
Methanol flow rate was checked periodically by using balance (0.8 density
15 factor). Balance reading were recorded continuously.
The fermentation was checked periodically for methanol accumulation (1-5 min)
during methanol induction phase as mentioned in the table 5.
MIP Age (hrs.) Methanol Feed Rate (g/hrs.)
0-2 2
3-6 3-4
6-8 8-9
8-10 13-15
13
10-14 17-19
14-16 20-22
16-Till EOF 24-25
Table 5: Methanol Induction Phase (MIP) against Methanol Feed Rate.
The insulin or insulin analogue precursor protein with Labelled 15N is released
into the supernatant during fermentation.
Treatment of precursor
After fermentation, pH of final broth is adjusted to pH 2.5 5 with ortho-phosphoric
acid followed by centrifugation to get cell free supernatant which further
subjected to the downstream purification process to isolate the insulin or insulin
analogue precursor from the cell-free supernatant and convert it into a Methoxytriethyleneglycol-
propionyl-N-ε-B29 recombinant insulin or insulin analogue
10 precursor, which through a trypsin and carboxypeptidase B catalysed reaction
gets converted into insulin or insulin analogue. The scheme of the process of
conversion of IN-105 precursor is depicted in Fig 3.
DOWNSTEAM PROCESS:
Preparation of stock solution, calibration standards and sample
15 Control Plasma:
Control Human Plasma (K2EDTA) from normal healthy volunteers (NHV) was
obtained.
Control Human Plasma (K2EDTA or K3EDTA) from patients diagnosed with either
type one (T1DM) or type two diabetes (T2DM) were also obtained. Plasma from
20 patients with type 1 and type 2 diabetes was procured from external supplier
14
and was available either Di-potassium EDTA (K2EDTA) or Tri-potassium EDTA
(K3EDTA).
Control Human Blood (K2EDTA or K3EDTA) was obtained from volunteers for use
in whole blood stability experiment and in preparation of the matrix used in the
assessment of haemolysis 5 (2% whole blood in plasma).
Lipaemic plasma was prepared by the addition of intralipid to control human
plasma (K2EDTA) in a ratio 1:9, was used to assess the impact of lipaemia on the
method. Haemolysed (2% whole blood in plasma) plasma was also obtained.
All the plasma obtained, either from external supplier or in-house volunteers,
10 was stored at -20 °C prior to use and whole blood was stored at 4 °C.
Solution preparation:
Two separate stock solutions were prepared by dissolving 2 mg of IN-105 and
dissolving in 1 mL of 2% acetic acid containing methanol:water in the ratio of
30:70 v/v. Thus, the stock solution of 2000 μg/mL was prepared.
15 One was named “calibration stock” and other was named “quality control stock”.
The stock solution was stored at -20°C for 40 days.
Calibration stock solution was prepared by adding 50 μL of stock solution and
diluting it in 450 μL of diluent (0.1% TFA in acetonitrile:water 50:50 v/v).
Quality control (QC) samples were prepared as per table 6 in manner having final
20 concentration of 0.2 (LLOQ QC), 0.6 (LoQC), 20 (MeQC), 40 (HiQC) and 50 (ULOQ
QC). These QC samples were stored at -20 °C and -80 °C for a period of 73 days
and were used only for validation purpose.
Quality Control Samples
Test Samples
QC Design Conc (ng/mL)
15
LLOQ-QC
6 Replicates
0.200
LoQC 0.600
MeQC 20.0
HiQC 40.0
ULOQ-QC 50.0
Table 6: Quality Control Samples
Preparation of standards
Various standards prepared for validation study that includes analytical standard,
internal standard, co-administered standard and calibration standard.
 Analytical Standard: Standards IN-105 at 96.3% 5 purity were used within
its known stability period.
 Internal Standard: Stock solution of internal standard was used in its
known stability period and prepared using 15N labelled IN-105 in the same
manner as calibration solutions. The final concentration of internal
10 standard solution was 100 ng/mL at 96.1% purity.
 Co-administered Standard: The co-administered standard ‘Glargine’ at
3.64mg/mL concentration and 100% purity was used within its known
stability period.
 Calibration Standard: The calibration standards were prepared freshly
15 from “calibration stock solution”. The calibration standards were
prepared using appropriate volume of calibration stock and diluting it
with plasma from normal healthy volunteers containing K2EDTA or
K3EDTA. The calibration standards can be prepared in bulk and stored at -
200C or -800C for up to 244 days and used within the known stability of
20 IN-105 in human plasma. Table 7 elaborates details of calibration
standards used for procedure and its preparation.
16
The final calibration standard concentrations were 0.2, 0.4, 1, 5, 10, 20, 45 and
50 ng/mL.
Test Samples
Preparation
Acceptance criteria based on
back Concentration calculated concentrations
(ng/mL)
Replicates
0.200 2 Calibration standards
1. Prepared from
calibration stock
solution
2. prepared in control
human plasma
3. prepared freshly for
each run
4. used within the
known stability
Overall ≥75% of the calibration
standard should be within ±15%
of nominal (±20% at the LLOQ).
1. Calibration standards within
±15% of nominal concentration
must be included within the
calibration line, and
2. Calibration standards with a
concentration exceeding ±15%
of nominal concentration must
be excluded from the
calibration line
0.400 2
1.00 2
5.00 2
10.0 2
20.0 2
45.0 2
50.0 2
Table 7: Calibration Standard
Sampling
5 The samples obtained using labelled insulin or insulin analogue in biological
matrix by off-line solid phase extraction (SPE) followed by on-line SPE in 96 well
formation.
Sampling was done for sterility, wet cell weight (WCW), Titre, pH and
conductivity on every day basis (every 24 hrs.).
10 Quality control samples prepared in quality control working solution in TFA
(0.1%) in MeCN:H2O 50:50, freshly on the day of analysis.
17
Aliquot (300 μL each) of samples were prepared in a 2 mL 96 well plate. To these
aliquots, 25 μL of internal standard (100 ng/mL) was added. None of the blanks
were spiked with internal standard. The plate was vortexed for 2 minutes at 1000
rpm. To this, 50 μL of 6 M guanidine HCl was added and the plates were vortexed
for 2 minutes at 1000 rpm followed by sonication 5 for 10 minutes. To the wells,
300 μL of 10 mM tris buffer was added and plate was vortexed for 2 min at 750
rpm for mixing.
In one embodiment nitrogen isotope labelled insulin having purity 96.3% and
96.1% respectively, was used as analytical standard and internal standard for
10 further analysis. Glargine (3.64 mg/mL) of 100% purity was used as coadministered
standard in the analytical procedure. The stability studies to
understand storage stability and solution stability were carried out.
Stability Study
In one embodiment, for stability assessment of IN-105 in whole blood, IN-105
15 was spiked in whole blood samples at LoQC and HiQC concentrations. The
samples were sub-aliquoted and stored at room temperature and on ice.
Aliquots of whole blood were taken up to 2 hrs. The samples were processed for
analysis by centrifugation at 4000 rpm at 20°C for 10 minutes. Separated plasma
was harvested and stored at -80°C until the time of analysis.
20 Assessment of sample stability on storage (storage stability) was carried
according to table (8), Acceptance criterion was defined as a recovered
concentration within ±15% of the nominal spiked concentration. All standards
were used within its stability period.
1 stability in plasma
stored at room
temperature
analysed for 24 hr
stability
plasma was spiked at
LoQC, HiQC, and
dilution QC
concentrations
18
2 stability in plasma
exposed to
repeated freezethaw
cycles
assessed for six
cycles of freeze-thaw
plasma was spiked at
LoQC, HiQC, and
dilution QC
concentrations
3 stability in frozen
plasma at -20 and
-80 °C
assessed for a
storage period of
244 days
plasma was spiked at
LoQC and HiQC
concentrations
Table (8): Stability Study
System suitability was tested prior to run. The signal at the LLOQ should be at
least 5 times greater than that of the noise and the chromatography and
retention time should be consistent with that seen previously. Carry-over greater
than 20% of the LLOQ has been observed within the 5 validation and therefore
carry-over within the system suitability samples is anticipated but should be no
greater than 50% of the LLOQ when directly after the ULOQ and less than 20% of
the LLOQ in the blank injection. The samples with be injected such that each type
of validation sample, at each concentration level, was distributed throughout the
10 run. Table 9 elaborates system suitability checked prior to run.
Table 9: elaborates system suitability checked prior to run.
SPE extraction:
Solid phase extraction was carried out using waters Oasis Max (10 mg) SPE plate.
15 Samples were prepared using off-line SPE extraction followed by on-line SPE.
19
Upon obtaining labelled insulin, calibration standard and quality control
standards were prepared.
Prior to each validation run, system suitability samples were analysed. These
samples typically consisted of samples at the LLOQ, ULOQ and two blank matrix
samples thus allowing qualitative assessment 5 of the quality of the
chromatography, sensitivity and carry over; with the exception of analytical runs
1 to 7 where only one matrix blank sample was injected after the ULOQ sample
(as mentioned in table 7). Although this was a deviation from the Analytical
Method, any carry-over observed was not considered to have impacted on the
10 analytical runs.
SPE conditions:
The SPE cartridges were conditioned with methanol (300 μL) and equilibrated
with 300 μL of water. Samples were loaded on to the plate, for ensuring
adequate mixing they were aspirated and dispensed. Plate was washed with 500
15 μL 1M ammonium acetate in MeCN:water (1:10:90 v/v/v), followed by washing
with 500 μL 1M ammonium acetate in MeCN:water (1:50:50 v/v/v). Packing
material was dried using maximum pressure. 0.1% Triton X-100 was added to a
fresh 1 mL well plate. Samples were eluated with 250 μL of 2% formic acid in
MeCN:water (50:50 v/v) to the Triton X-100 containing plate. To each of these
20 eluates 300 μL of 0.1% TFA in water was added and mixed with aspiration and
dispensing. Samples were capped and taken for analysis.
LC/MS/MS conditions:
Chromatographic conditions:
Mobile phase A was TFA 0.02% in acetonitrile: water (10:90 v/v); Mobile phase B
25 was TFA 0.02% in acetonitrile:water (90:10 v/v). Initial concentration of mobile
phase A was 85% till 30s, from 30s to 4 min 30s, mobile phase A changed from
20
85% to 60% in a linear fashion, over next 10s mobile phase A changed to 10%
and stayed at 10% till 5 min 30s. For the equilibration, over next 10s mobile
phase A changed back to 85%. The run time was completed at 6 min 30s.
Temperature of elution was maintained at 50 °C. Elution was carried out using
0.5 mL/min flow rate. Injection volume 5 was 500 μL. Samples were maintained at
10 °C during the sequence. Wash solvent 1 was 0.01% TFA in acetonitrile:water
(10:90 v/v), Wash solvent 2 was 0.01% TFA in methanol:water: Triton-X100
(70:30:0.05 v/v/v), wash solvent 3 was Ammonia (0.5% of 28%) in
methanol:water (80:20 v/v).
10 Online extraction conditions:
Cartridges were conditioned using 500 μL methanol at 5 mL/min flow rate.
Online cartridges were equilibrated with 500 μL of (0.1% TFA) in MeCN: water
(10:90 v/v). Backflush wash was given using 1 mL of (0.1% TFA) in MeCN: water
(10:90 v/v) at 2 ml/min flow rate. Following that cartridges were washed using 1
15 mL of (0.1% TFA) in MeCN: water (20:80 v/v) at 5 ml/min flow rate. Clamp flush 1
was carried out using 500 μL of (0.1% TFA) in MeCN: water (10:90 v/v) at 5
mL/min flow rate, clamp flush 2 was carried out using (0.1% TFA) in MeCN: water
(90:10 v/v) at 5 ml/min flow rate and Clamp flush 3 was carried out using 500 μL
of (0.1% TFA) in MeCN: water (10:90 v/v) at 5 ml/min flow rate.
20 MRM transition:
For IN-105, 1506.9 m/z a four charge state was isolated as parent ion and its
transition to 1820 m/z daughter ion was monitored. For 15N labelled IN-105,
1520 m/z was isolated and its transition to 1838.5 m/z daughter ion was
monitored for quantification. The molecular weights and charge states of human
25 insulin along with transitions monitored in the MRM mode and summarized as
follows in table 10.
21
Species type
Average
Molecular
weight (Da)
(M + 4H)+4
(Da)
(M + 5H)+5
(Da)
Ion Transition
monitored (Da)
Human insulin 5808 1453 1162.6 Not Monitored
IN-105 6025 1507.25 1206 1507.25 > 1822
[15N] IN-105 6076 1520 1216 1520 > 1838.5
Table (10): molecular weights and charge states of human insulin along with
transitions monitored in the MRM mode
Validation
The validation study included linearity, accuracy and precision, selectivity and
specificity, repeatability 5 and reproducibility.
Validation consisted of intra run precision and bias, inter-run precision and bias,
dilution integrity, matrix related modification of ionisation, auto-injector carry
over, stability in whole blood at room temperature and on ice, freeze thaw
stability of plasma over 5 cycles, 24 h room temperature stability of plasma,
10 frozen stability in plasma at -20 and -80 °C for 73 days, recovery and extract
stability at 10 °C.
Additional validation runs were analysed to carry out additional experiments that
would not fit within the runs used to assess precision and bias. The runs to
assess precision and bias were designed to be as large as a potential study
15 sample run (96 samples). Where additional samples were required to make a
validation run as large as a study sample run, then control human plasma
samples were used for this purpose.
Validation Study Design
Following table 11 elaborate parameters of validation study.
22
Parameter investigated
Nominal Insulin concentrations Number of
0 0.200 0.600 20.0 40.0 50.0 500* replicates
Intra-run precision and
bias (investigated in
three runs)
X X X X X 6
Inter-run precision and
bias (investigated in
three runs)
X X X X X 18
Dilution integrity X 6
Matrix related
modification of
ionisationΔ
X X 6
Auto injector carry-over
(investigated in all runs)
X 4
Stability in whole blood
at room temperature
and on ice
X X 3
Freeze/thaw stability in
plasma (5 cycles)
X X X 6
Room temperature
stability in plasma (24
hours)
X X X 6
Frozen stability in
plasmaΔ (-20°C and -
80°C for 73 days)
X X 3
Recovery X X X 3
Extract stability at 10°C X X X X X 6
Table 11: Summarized validation study *
analysed after 20-fold dilution with control plasma. Δ Investigated in plasma
from normal healthy volunteers (NHV) & patients with T1DM or T2DM
In the first analytical run, a calibration standard which was prepared at a
5 concentration of 40.0 ng/mL and not at 45.0 ng/mL as per the study plan. All
23
other runs contained calibration standards at the concentrations specified within
the study plan. The mean intra-run precision was found to be less than, or equal
to 11.1% and the mean intra-run bias was between 0.4% and 3.0%. The inter-run
precision and bias was 16.5% and 2.5% at the LLOQ QC concentration, at all
other levels the inter-run precision was equal to or 5 less than 8.9% and the interrun
bias was between 0.3% and 2.0%.
In addition to the experiments described above, selectivity of the method in
plasma (K2EDTA/ K3EDTA) from different sources, effect of lipaemia and
haemolysis, selectivity and modification of ionisation of insulin in the presence of
10 Glargine and stability of solutions of insulin or insulin analogues investigated.
Thorough check was performed for stability of plasma at various conditions.
For a calibration curve to be acceptable, 12 (75%) of the calibration standards
had to have a back-calculated concentration of IN-105 within ±15% of the
nominal value (±20% for the LLOQ). For a given calibration curve, calibration
15 standards had to be included in the regression if the back-calculated
concentration deviated from the nominal by less than, or equal to, 15% (20% at
the LLOQ). A calibration standard had to be omitted from the regression if the
calculated concentration of IN-105 deviated by more than 15% from the nominal
(20% at the LLOQ). Up to four calibration standards (25%) could be omitted from
20 the calibration curve, to achieve the acceptance criteria; otherwise the run was
rejected. The acceptance criteria for mean intra-run and the inter-run precision
was less than, or equal to, 15% at all levels, other than at the LLOQ, at which
precision had to be better than, or equal to, 20%. The mean intra-run and
inter-run bias had to be within ±15% of the nominal concentration at all levels,
25 other than at the LLOQ, at which bias had to be within ±20% of the nominal
concentration.
The study design for these additional validation experiments are detailed in
Table 12 below:
24
Table 12: Additional validation procedure Δ
Investigated in plasma from normal healthy volunteers (NHV)
*Though not part of the study design to determine the validity of the
modification made to the Analytical Method was investigated using the modified
5 Analytical Method
Fig 4-12 show the peaks obtained during LC/MS/MS of samples.
The method was not affected by inter-individual variability, lipaemia, and
haemolysis or by presence of Glargine. The observation infers that method is
sufficiently accurate and precise and have sufficient selectivity and reliability that
Parameter investigated
Nominal Insulin concentrations Number
of
replicat
es
0
0.20
0
0.60
0
20.0 40.0 50.0
Intra-run precision and bias
(investigated in one run)
X X X X X 6
Matrix related modification
of ionisationΔ
X X 6
Selectivity
(with and without internal
standard)
X 6
Auto injector carry-over
(investigated in all
additional runs)
X 4
Frozen stability in plasma*
(-20°C and -80°C for
244 days)
X X 3
Extract stability at 10°C X X X X X 6
25
allow determination of insulin in human plasma samples over the examined
range.
Insulin was found to be stable in plasma when stored at room temperature for
up to 24 hours, after five freeze/thaw cycles and after 244 days storage at both -
20°C and -80°C. Also found to be stable in extracts 5 when stored at 10°C for
approximately 85 hours.
Labelled insulin was stable in acetic acid (2%) in methanol: water (30:70 v/v)
when stored at -20°C for up to 40 days and when stored at room temperature for
up to 24 hours.
10 Example 1: The detection of isotope nitrogen labelled IN-105
Blood samples were collected in tubes containing K2-EDTA. 15N-labelled IN-105
was used as an internal reference standard (ISTD) and added to tubes after
separation of serum or plasma. The ISTD-spiked samples were subjected to a
mixed-mode anion exchange-reverse phase solid phase extraction process in a
15 96-well format. The eluate from the mixed-mode SPE are transferred to a
reverse-phase (C8) SPE set up in a 96-well format. This RP SPE was setup online
with an analytical reverse phase HPLC system. The eluate from the RP SPE were
transferred online to the C18 analytical chromatography column and subjected
to reverse phase HPLC.
20 The eluate from the analytical chromatography column passes into a triple
quadrupole mass spectrometer, where an ESI process generates gas-phase ions
from the eluate. These gas phase ions were then analysed in MRM mode as
follows.
The ions pass into the first quadrupole, where ions with m/z values in a narrow
25 range are selected. Ions within this narrow m/z range are allowed to pass
through to the second quadrupole.
26
At the second quadrupole, intact IN-105 and ISTD molecules (and any other ions
with m/z values in the same range) are subjected to a collision-induced
fragmentation process in presence of an inert gas. This generates fragment ions
characteristic of the parent species that were allowed to pass into the second
quadrupole. These fragmented species then pass to 5 the third quadrupole.
In the MRM mode, the third quadrupole will be set to select for ions having m/z
values characteristic of fragments generated from IN-105 and ISTD. Fragments
from other species that happened to have the same intact m/z as IN-105 and
ISTD will be eliminated at this stage.
10 Only the fragments allowed through the third quadrupole are then detected as
an ion current. The peak in the extracted Ion Chromatogram (XIC) for IN-105 and
ISTD were both found to be integrated. The ratio of the integrated area of IN-105
is to ISTD was used as a measure of the quantity of IN-105 present in the sample.
Example 2: Pharmacokinetics Study
15 The method was considered validated successfully since it passed all the predetermined
criteria in the validation protocol. The validated method was
subsequently used for measurement of IN-105 levels in a euglycemic clamp study
carried out in patients with Type 1 diabetes. The method for the determination
of IN-105 in human plasma has been validated successfully over the
20 concentration range 0.200 ng/ml to 50.0 ng/mL.
Plasma concentration data for each patient and treatment was analysed by a
non-compartmental method. The area under plasma level curve for AUC0-t was
calculated by the trapezoidal rule. The primary pharmacokinetic parameters
(mean ± SD) were Cmax, AUClast, Tmax and PD parameters were Tmin, Cmin, AUClast.
25 Ratios and 90% CIs of geometric means were calculated for PK and PD
parameters from mixed effects model with fixed effects for sequence, period and
treatment, and patients within sequence as a random effect for log transformed
27
Cmax and AUC. The maximal plasma concentration of IN-105 as well as the area
under the PK curve are linearly correlated significantly (p < 0.05) with the dose
employed (as per table 16 below).
The average plasma concentration time profile of IN-105 obtained after
administration of IN-105 tablets in volunteers is 5 shown in figures 13-14.
Function of dose slope* Intercept**
Regression
coefficient (R2)
p-value
AUClast
(AUCt) (ng.h/ml)
26.427 -243.525 0.907 0.0476
Cmax (ng/ml) 0.431 -2.421 0.9434 0.0287
Table 16 pharmacokinetic parameters with slope points and p-value
*area units / mg for AUClast and concentration units/mg for Cmax
**area units for AUClast and concentration units for Cmax
Plots of mean plasma concentration as a function of time after dosing shows the
10 expected time profile of drug concentration in plasma. The plot for the 30 mg/kg
dose is shown in figure 15, where a well-defined peak concentration and
subsequent clearance is evident.
28
We Claim:
1. A method for detection and quantification of insulin or insulin analogue in a
biological matrix wherein the method comprising the steps of:
i. labelling known amount of insulin or insulin analogue by a stable isotope
to form a labelled 5 insulin or insulin analogue;
ii. introducing the labelled insulin or insulin analogue to biological matrix;
and
iii. Analysing the biological matrix for intact insulin or insulin analogues by
liquid phase chromatography-tandem mass spectrometry (LC-MS/MS).
10 2. The method of claim 1, where the labelled insulin analogue is IN-105 has the
following structure
3. The method of claim 1, wherein the step of labelling a known amount of
insulin or insulin analogue by a stable isotope to form labelled insulin or
15 insulin analogue comprises a culture medium comprising a recombinant
strain of Pichia pastoris carrying a gene which codes for expression of a
proinsulin or proinsulin analogue during a fermentation process.
4. The method according to any of claims 1-3, wherein isotope is at least one
amongst nitrogen (15N), sulphur (34S), oxygen (18O), hydrogen (2H) and
carbon (1320 C).
29
5. The method of claim 4, wherein the isotopic nitrogen (15N) source in the
culture medium is at least one of ammonium sulphate, ammonium
phosphate, ammonium hydroxide, methylamine, urea.
6. The method of claim 4, wherein the isotopic carbon (13C) source in the
culture medium is at least one of methanol, 5 glycerol, glucose, sorbitol.
7. The method of claim 4, wherein the isotopic sulphur (34S) source in the
culture medium is at least one of calcium sulphate, magnesium sulphate,
potassium sulphate.
8. The method of claim 3, wherein the fermentation process is two phase, the
10 first phase is a batch phase to increase cell mass with the inclusion of
glycerol as a carbon source and a fed batch phase using methanol to induce
secretion of a precursor of the labelled insulin or insulin analogue and
isotope source for inclusion into the expressed labelled insulin or insulin
analogue.
15 9. A method of claim 8, wherein the precursor of labelled insulin or insulin
analogue is treated with trypsin and carboxypeptidase B to form the labelled
insulin or insulin analogue.
10. The method of claim 1, wherein the biological matrix is whole blood, blood
serum, blood plasma, urine, fermentation broth or buffer.
20 11. The method of claim 1, wherein the biological matrix is combined with dipotassium
or tri-potassium EDTA.
12. The method of claim 1, wherein the biological matrix is stored at -20°C or at -
80°C or freshly prepared for every cycle.
13. The method of claim 12, wherein the storage period was from 1 day to 250
25 days.
30
14. The method of claim 1, wherein step ii) further comprises sample processing
using off-line solid-phase extraction (SPE) followed by on-line SPE in 96 well
formation.
15. The method of claim 1, wherein the labelled insulin or insulin analogue is
used to differentiate from exogenously delivered 5 insulin analogue from
endogenous insulin.
16. The method of claim 15, wherein the method determines a concentration of
isotope labelled insulin or insulin analogue ranging from 0.200 ng/mL to 50
ng/mL in the biological matrix.17. A method for determining the stability of IN-105 insulin in a biological matrix
comprising at least one additional type of insulin or an insulin analogue, the
method comprising:
i) labelling a known amount of IN-105 by a stable isotope of nitrogen to
form a labelled IN-105;
15 ii) introducing the labelled IN-105 to the biological matrix comprising the
at least one additional insulin or insulin analogue;
iii) subjecting the solution obtained from step ii) to a mixed-mode anion
exchange-reverse phase solid phase extraction process using off-line solid-phase
extraction (SPE) followed by on-line SPE in 96 well formation.
20 iv) analysing the intact labelled IN-105 relative to that of the at least one
additional type of insulin or insulin analogue by liquid phase chromatographytandem
mass spectrometry (LC-MS/MS).
18. The method of claim 17, wherein labelled IN-105 was stable for 24 hours at
room temperature.
19. The method of claim 17, where IN-105 has the following structure