Quantitation of anacetrapib in human and animal adipose by liquid chromatography with mass spectrometric detection

ABSTRACT

Cholesteryl ester transfer protein (CETP) inhibitor is a target for both lowering low-density lipoproteins and raising high-density lipoproteins. Anacetrapib was the lead compound in our cholesteryl ester transfer protein inhibitor program. Preclinical studies were initiated to support the safety of anacetrapib deposition in adipose tissue, followed by a clinical trial to evaluate the effects of anacetrapib in people with vascular disease. An ultra-high performance liquid chromatography/tandem mass spectrometry method was developed to determine tissue anacetrapib concentrations in the adipose of three animal species and humans. The assays were validated in the concentration ranges of 5–5000 ng/ml and 0.1–100 μg/ml. The anacetrapib concentrations in adipose tissue from preclinical and clinical studies were determined.

Plain language summary

Article highlights.

Experimental

• A bioanalytical method using LC-MS/MS was developed to quantitatively determine Anacetrapib in adipose tissue. Two different calibration ranges were established for both preclinical and clinical assays based on the dose levels.

• Adipose tissue was homogenized in plasma to ensure uniformity and facilitate liquid-liquid extraction of the drug during sample preparation.

Results

• The method was validated in two different calibration ranges and was successfully implemented for the analysis of Anacetrapib in both preclinical and clinical adipose samples.

• Pharmacokinetic studies of Anacetrapib, with analysis performed on over 100 animal adipose samples from three different species and over 500 human adipose samples were supported.

1. Introduction

Clinical trials have shown that lowering low-density lipoprotein cholesterol reduces the risks of coronary events and strokes [1]. Studies have also demonstrated that lower concentrations of high-density lipoproteins (HDL) cholesterol and of apolipoprotein (apo) A1 are related to a higher risk of coronary heart disease. Increased HDL cholesterol and reduced low-density lipoprotein cholesterol levels can reduce the risk of coronary heart disease [2–5].

Cholesteryl ester transfer protein (CETP) is a plasma protein that facilitates the exchange of triglycerides and cholesteryl esters between HDL and the atherogenic apolipoprotein B-containing lipoproteins. Anacetrapib is a CETP inhibitor that raises HDL and lowers low-density lipoprotein. It was also well tolerated and did not raise blood pressure in the clinical studies [6–8].

In man, adipose tissue expresses substantial amounts of CETP mRNA [9,10] and may contribute significantly to human CETP production. Studies have shown that human adipose tissue maintained in organ culture synthesizes and secretes CETP [10].

During the anacetrapib development program, preclinical studies were initiated in rats, mice and non-human primates to characterize the deposition of anacetrapib in adipose tissue. Rodents (Sprague Dawley rats and CD-1 mice) were administered anacetrapib for one year at dosing levels consistent with those used in the carcinogenicity studies. During these year-long studies, adipose concentrations were designated to be determined at several time points. A similar study in non-human primates was conducted, which utilized collection techniques planned to be used during clinical studies.

Thus, methods to analyze anacetrapib in adipose tissue were needed to support the preclinical work and subsequent clinical studies. The development and validation of methods for determining tissue concentrations of anacetrapib in the three animal species and humans are described herein.

2. Experimental

2.1. Materials

Anacetrapib and its stable isotope-labeled internal standard [13C 2H3] anacetrapib (Figure 1), were obtained internally. HPLC grade acetonitrile (ACN), HPLC grade Isopropanol (IPA), HPLC grade hexane, HPLC grade Methyl tertbutyl ether (MTBE), and laboratory grade formic acid (90%) were purchased from Fisher Scientific (Pittsburgh, PA, USA). Dimethyl sulfoxide (DMSO) ACS reagent grade (99.9%) was purchased from Sigma-Aldrich (Allentown, PA, USA). Control human adipose tissue was purchased from Analytical Biological Services Inc. (Wilmington, DE, USA). Rat, mouse, dog and monkey control adipose were obtained internally. Water was purified by a Milli-Q ultra-pure water system from Millipore (Bedford, MA, USA).

Figure 1.

Representative chromatogram of rat adipose homogenate (1:9) spiked with anacetrapib at LLOQ level and ISTD.

ISTD: Internal standard; LLOQ: Lower limit of quantitation.

2.2. Instruments

A Waters Acquity UPLC system (Framingham, MA, USA) consisting of a UPLC Binary Solvent Manager, Sample Manager, and Sample Organizer was used for HPLC separation. As a detector, a Sciex API 4000 triple quadrupole mass spectrometer with a Sciex Turbo Ion Spray Interface (AB Sciex, Toronto, Canada) was used. The data were collected and processed through Analyst 1.4 software.

2.3. Chromatographic conditions

Extracted samples, prepared as described below, were analyzed on a BEH Shield RP18 (50 × 2.1 mm × 1.7 um) column from Waters (Framingham, MA, USA) with a 1 μl sample injection. A gradient elution was used with a mixture of water with 0.1% formic acid (Solvent A) and ACN with 0.1% formic acid (Solvent B). The flow rate was 0.6 ml/min. The gradient program was as follows: 70% B for 0.1 min; ramp to 90% B for 1.4 min; hold at 90% B for 0.5 min; return to 70% B after 0.1 min; then hold for 0.4 min (total run time of 2.5 min). The column was maintained at 40°C, and the samples were kept at 10°C in the autosampler.

2.4. Mass spectrometry detection & calculation

A PE Sciex API 4000 triple-quadrupole mass spectrometer (MS/MS) with a turbo-ion spray interface ionization source operated in a positive ion mode was used to quantify anacetrapib. The ion pairs (precursor ion → product ion) m/z 638 → 283 for anacetrapib and m/z 642 → 287 for the internal standard (ISTD) were selected for multiple reaction monitoring (MRM). The instrument setting was adjusted to maximize the response for the analyte. The turbo gas temperature was 450°C. The settings of nebulizing gas (GS1, nitrogen), heater gas (GS2, nitrogen), collision gas (nitrogen) and curtain gas (nitrogen) were 40, 50, 4 and 40. The optimized declustering potential, entrance potential, collision energy and collision cell exit potential were 65, 10, 31 and 9 v. The dwell time was 100 milliseconds for anacetrapib and internal standard. Both Q1 and Q3 quadrupoles were set at unit resolution. The total run time for each injection was 2.5 min. Raw peak areas generated in the SCIEX Analyst (ver. 1.4.2) software package are exported into the WATSON system for quantitation. A calibration curve was obtained by weighted (1/x²) least squares linear regression of the peak area ratio of the analyte to the internal standard versus the analyte's nominal concentration (x).

2.5. Adipose-plasma homogenate preparation

Adipose samples were homogenized in plasma to extract anacetrapib from the tissue. Because of the high anacetrapib concentrations, a 1:29 adipose: plasma ratio was used for animal studies. For human studies, an adipose: plasma ratio of 1:9 was initially used to prepare human adipose samples, then changed to a 1:29 adipose: plasma ratio when anacetrapib concentrations in adipose were found to increase in later studies.

To prepare a consistent adipose-plasma homogenate, adipose samples were cut into fine pieces using disposable scalpels and then transferred to polypropylene tubes. After determining the weight of the tissue, the volume of control plasma (μl) was calculated (weight of tissue (g) X 9 × 1000 for 1:9 ratio, or weight of tissue (g) X 29 × 1000 for 1:29 ratio) and added to the sample tube. Stainless steel beads were then added to each tube and homogenized using a Geno Grinder for 10 min at 1500 RPM.

2.6. Calibration standards & quality control samples

Two stock solutions for anacetrapib were prepared from two separate weighings and dissolved in a diluent, respectively. The stock solutions were prepared at 200 μg/ml in ACN for the low-range calibration standards and at 5 mg/ml in DMSO for high-range calibration curve standards. Calibration standards were prepared from one set of analyte stock solutions, and quality control (QC) samples were prepared from another set. Working standards were prepared by serial dilutions of the standard stock with ACN/water (50/50, v/v) to yield the following concentration: 5, 10, 25, 50, 250, 500, 2500 and 5000 ng/ml for the low calibration curve, and 0.1, 0.2, 0.5, 1, 5, 20, 85 and 100 μg/ml for the high calibration curve. Working standards were stored at 4°C in amber glass vials. Internal standard stock solutions at 100 μg/ml in ACN and 1 mg/ml in DMSO were prepared for low and high curves, respectively.

The ISTD working solutions were prepared at 200 ng/ml and 2 μg/ml in ACN/water (50/50, v/v) for low and high curves, respectively. QC samples were prepared at 15, 150 and 400 ng/ml in adipose-plasma homogenate for the low curve and at 0.3, 10 and 75 μg/ml for the high curve. The aliquots were stored in a -20°C freezer following preparation.

2.7. Sample extraction

QCs and control adipose-plasma homogenate were thawed at room temperature and mixed well.

For the low curve range of 5–5000 ng/ml, an aliquot of 50 μl unknown and QC homogenate samples were transferred into a 96-deep well plate (2.4 ml well volume). An aliquot of 25 μl of ISTD working solution and 25 μl 50% ACN were sequentially added to each well and mixed. The calibration standards were prepared using the same procedures, except the sample was substituted with a control adipose-plasma homogenate, and 50% ACN was substituted with working standard solutions. 50 μl of pH 9.8 carbonate buffer was added to all wells, and the plate was vortexed for 30 s. Around 1050 μl extraction solvent (20/80 Isopropanol/Hexane) was added to all samples, which were then mixed via rotation at a speed of 70 rpm for 15 min, followed by centrifugation for 5 min at 3000 rpm (1643 g). Around 600 μl of the supernatant was transferred to a clean deep-well plate and dried at 40°C with nitrogen. The dried samples were reconstituted in 200 μl 50% ACN with 0.1% formic acid. After centrifuging at 3500 rpm for 5 min, 4 μl of reconstituted sample was injected into the LC-MS/MS system for analysis.

For the high calibration curve range of 0.1–100 μg/ml, an aliquot of 50 μl unknown and QC homogenate samples were added into wells in a 96-deep-well plate. Around 50 μl of ISTD working solution and 50 μl 50% ACN were added to each well and mixed well. The same procedures were used to prepare the calibration standards, except for substituting the sample with a control adipose-plasma homogenate and substituting 50% ACN with working standard solutions. 50 μl of pH 9.8 carbonate buffer was added to all wells, and the plate was vortexed for 30 s. Around 1050 μl extraction solvent (20/80 Isopropanol/Hexane) was added to all samples. The samples were then mixed via rotation at a speed of 70 rpm for 15 min, followed by centrifuging for 5 min at 3000 rpm (1643 g). Around 200 μl of the supernatant was transferred to a clean deep-well plate and dried at 40°C with nitrogen. The samples were then reconstituted in 500 μl 50% ACN with 0.1% formic acid. After centrifuging at 3500 rpm for 5 min, 1 μl was injected into the LC-MS/MS system for analysis.

2.8. Study samples for analysis

Rodent and non-human primate samples for analysis were obtained from studies reviewed and approved by the Merck IACUC – Institutionalized Animal Care and Use Committee. The study approval numbers were #12067336020169, #13060026020172, #13020026020099 and #13020026020092.

Clinical samples for analysis were obtained from subjects with informed consent. The clinical study (EudraCT Number: 2010-023467-18) from which samples were obtained was approved by the institutional review boards of the clinical sites.

3. Results

The performance of the assay was characterized through assessments that included establishing the standard calibration range, determining intra-assay accuracy and precision, and evaluating analyte and internal standard interference, carryover, dilution integrity, matrix stability (short-term stability, freeze-thaw stability and long-term frozen stability) and stock and working solution stability.

3.1. Accuracy & precision assessment

The assay was evaluated over the concentration range of 5–5000 ng/ml and 0.1–100 μg/ml for human adipose and 0.1–100 μg/ml for animal adipose. The calibration curves were generated using the peak area ratios of the analyte to its internal standard versus standard concentrations, with a weighted (1/x2, where x is the concentration of analyte) least-squares regression. Assay accuracy was 96.0–107.0% of nominal values for the low concentration curve of 5–5000 ng/ml anacetrapib in human adipose homogenate and 95.0–106.0% of nominal values for the high concentration range of 0.10–100 μg/ml anacetrapib in rat, mouse, monkey and human adipose homogenate (Table 1). The lower limit of quantitation (LLOQ) was 5 ng/ml for the low concentration range and 0.1 μg/ml for the high concentration range, using 0.05 ml of adipose homogenate. Representative chromatograms of rat adipose homogenate (1:9) spiked with anacetrapib at LLOQ level, and ISTD are displayed in Figure 1. Mouse, monkey and human adipose homogenate samples had similar chromatography.

Table 1.

Intraday accuracy for the determination of anacetrapib in human adipose-plasma homogenate.

Nominal concentration (ng/ml)	Human (1:9)	Nominal concentration (μg/ml)	Human (1:29)	Rat (1:9)	Rat (1:29)	Mouse (1:29)	Monkey (1:29)
	Accuracya (%)		Accuracya (%)	Accuracya (%)	Accuracya (%)	Accuracya (%)	Accuracya (%)
5	96.2	0.1	97.3	100.0	100.0	100.0	100.0
10	107.0	0.2	106.0	100.0	105.0	100.0	95.0
25	101.6	0.5	97.4	102.0	100.0	100.0	100.0
50	101.4	1	101.0	103.0	102.0	102.0	99.0
250	100.4	5	100.6	99.8	103.0	94.6	99.8
500	97.6	20	100.5	100.7	100.8	101.1	100.5
2500	96.0	85	100.2	100.2	96.3	101.4	105.8
5000	100.0	100	96.4	96.2	95.6	99.9	97.8

^aExpressed as [(mean observed concentration)/(nominal concentration)] × 100 (%).

3.2. Interference & crosstalk assessment

No peak was observed at the analyte or internal standard retention times in control adipose-plasma homogenates of human and all animal species.

The absence of “crosstalk” between the analyte channels was demonstrated by the analysis of adipose-plasma homogenate samples containing ISTD of 0.1–2 μg/ml without analyte and the analysis of adipose-plasma homogenate samples containing anacetrapib at the ULOQ of 5 or 100 μg/ml without ISTD, for low and high curve, respectively. In the ISTD-only samples, the chromatographic response at the expected retention time of the analyte was less than 33% of the peak response resulting from the analysis of the LLOQ standard, as well as for ULOQ-only sample response at the expected retention time of the internal standard was less than 5% of the peak response of the internal standard in the LLOQ standard.

3.3. Carryover

Carryover was evaluated by analyzing a control blank injected after a ULOQ sample. For all the species, the responses in the control blank samples used for carryover assessment were less than 100% of the LLOQ response of the analyte standard and less than 5% of the corresponding IS responses.

3.4. Matrix stability

The matrix stability of the analyte (short-term, freeze-thaw, and long-term frozen) was evaluated using low and high QCs.

The short-term stability of anacetrapib in the adipose homogenate was assessed by placing the thawed QC samples at room temperature for at least 2 h. Assay accuracy was 88.0–102.7% of nominal value in samples from all species.

Freeze/thaw stability was assessed after at least one freeze cycle and thaw after storage at -20°C. Assay accuracy for the freeze/thaw was 88.0–101.7% of nominal in all species.

Long-term frozen stability was evaluated by analyzing the QCs stored at -20°C for varying periods. Long-term frozen stability at -20°C in adipose homogenate was assessed for all species from 1 to 725 days. Assay accuracy was found to be 89.6–106.8% of nominal. The results are presented in Tables 2 & 3.

Table 2.

The adipose homogenate low-range QCs stability.

	Matrix	Ambinent Temp. 3 h		Freeze-thaw 5 cycle		Long-term stability 423 days at -20°C
		15 ng/ml	4000 ng/ml	15 ng/ml	4000 ng/ml	15 ng/ml	4000 ng/ml
Accuracya (%) (%CVb) n = 5	Human (1:9)	92.0 (2.5)	88.0 (1.0)	93.0 (1.6)	90.0 (1.3)	91.4 (2.5)	89.6 (5.3)

Numbers in parentheses are coefficients of variation (%CV).

^aExpressed as [(mean observed concentration)/(nominal concentration)] × 100 (%).

^bCoefficient of variation.

Table 3.

The adipose homogenate high-range QCs stability.

Matrix	Ambinent Temp. 2 h		Ambinent Temp. 3 h		Freeze-thaw 1 cycle		Freeze-thaw 3 cycle		Long-term stability
	0.3 μg/ml	75 μg/ml	0.3 μg/ml	75 μg/ml	0.3 μg/ml	75 μg/ml	0.3 μg/ml	75 μg/ml	0.3 μg/ml	75 μg/ml
Human (1:29)	n/a	n/a	101.0 (4.1)	102.7 (1.7)	n/a	n/a	100.0a (1.5)	101.7a (2.5)	101.0a (4.1)c	101.7a (2.5)c
Rat (1:9)	96.7a (3.5)	95.9a (1.4)	n/a	n/a	93.3a (3.6)	96.1a (1.6)	n/a	n/a	93.3a (3.6)d	96.1a (1.6)d
Rat (1:29)	96.7a (3.5)	100.4a (2.4)	n/a	n/a	100.0a (3.3)	98.5a (1.1)	n/a	n/a	96.7a (3.5)c	100.4a (2.2)c
Mouse (1:29)	93.3a (3.6)	97.2a (1.2)	n/a	n/a	90.0a (3.7)	97.3a (2.7)	n/a	n/a	96.7a (0.0)e	94.6a (2.0)e
Monkey (1:29)	n/a	n/a	96.7a (3.5)	99.4a (3.7)	96.7a (3.5)	101.3a (1.7)	n/a	n/a	102.5a (3.1)f	106.8a (1.3)f

Numbers in parentheses are coefficients of variation (%CV).

^aExpressed as [(mean observed concentration)/(nominal concentration)] × 100 (%).

^bCoefficient of variation.

^cDistored at -20°C for 3 days.

^dStored at -20°C for 1 day.

^eStored at -20°C for 343 days.

^fStored at -20°C for 725 days.

3.5. Dilution integrity

Dilution integrity was evaluated for each species by analyzing dilution QCs containing anacetrapib 3–16-times higher than the ULOQ. The samples were diluted with control adipose homogenate before extraction. Assay accuracy was found to be 86.5–96.4% of nominal values. The precision of five replicates of all species was less than 3% C.V. The results are presented in Table 4.

Table 4.

The adipose homogenate QCs dilution integrity.

	Matrix	QC Nominal Concentration (μg/ml)
		300	80,000
		4x dilution	20x dilution
Accuracya (%) (%CVb) n = 5	Human (1:9)	n/a	86.5 (2.1)
	Human (1:29)	n/a	n/a
	(1:9)	n/a	n/a
	Rat (1:29)	94.9 (2.9)	n/a
	Mouse (1:29)	96.4 (3.0)	n/a
	Monkey (1:29)	95.4 (1.6)	n/a

Numbers in parentheses are coefficients of variation (%CV).

^aExpressed as [(mean observed concentration)/(nominal concentration)] × 100 (%).

^bCoefficient of variation.

3.6. Solution stability

The stability of the analyte stock and working solutions was evaluated after they were stored at 4°C for at least 37 days and moved to room temperature for 6 h before being analyzed with freshly prepared stock and working solution for comparison. The difference between initial and fresh preparations was 5.3% at maximum.

3.7. Clinical & preclinical sample analysis

During the 52-week preclinical studies, rats, mice, and monkeys were given an oral dose of 50, 2000 and 500 mg/kg/day anacetrapib. The representative-determined concentrations in rat and monkey adipose are shown in Figures 2 & 3.

Figure 2.

Representative adipose concentrations from rats after an oral dose of 50 mg/kg/day anacetrapib in a 52-week study.

Figure 3.

Representative adipose concentrations from monkeys after an oral dose of 500 mg/kg/day anacetrapib in a 52-week study.

After the preclinical studies were completed, a large-scale, randomized, placebo-controlled clinical trial of the clinical effects of anacetrapib among people with established vascular disease was conducted. This was an 8–12-week study with a once-daily oral dose of 100 mg anacetrapib. Representative-determined average anacetrapib concentration from 10 clinical samples was 170 ng/ml with a CV of 27%.

4. Discussion

An extended plasma half-life of Anacetrapib was observed in early pre-clinical and clinical studies [11–13]. A potential explanation for this observation was that Anacetrapib was accumulating in adipose tissue during dosing due to its highly lipophilic nature. The drug was released slowly from this depot into plasma after administration was stopped. To test this hypothesis, methods for determining Anacetrapib in adipose tissue from pre-clinical species and humans were required.

Solid tissue sample analysis generally requires homogenization in a solvent or matrix from which the target analytes can be readily extracted. Previously, we have described a method for the determination of Anacetrapib and several of its metabolites in human plasma. To leverage the conditions from this previous work, we investigated using plasma to homogenize the adipose tissue samples.

Homogenization of adipose with plasma in either a 1:9 or a 1:29 ratio of adipose to plasma was found to solubilize the anacetrapib in the adipose samples. Following pH adjustment, the adipose in the solubilized samples could be extracted with a 20/80 v/v% mixture of isopropanol/hexane. After evaporation and reconstitution, the samples were analyzed using conditions similar to those previously reported for the plasma analysis method [14].

Using the methodology described herein, it was confirmed that Anacetrapib was found in high concentrations in adipose tissue, and thus, deposition in adipose was the likely reason why the drug could be detected in plasma for up to 1 year following the discontinuation of Anacetrapib administration [13].

5. Conclusion

The bioanalytical methods for quantitative determination of anacetrapib in adipose tissue by LC-MS/MS were developed and validated over two different calibration ranges, which covered concentration ranges from 5 ng/ml to 100 μg/ml. The QCs of all species provided good accuracy and precision. The methods were implemented in both preclinical and clinical adipose sample analysis to support anacetrapib pharmacokinetics studies. Over 100 animal adipose samples from three species and over 500 human adipose samples were analyzed. The results demonstrated that the assays had excellent reliability and suitability for different adipose homogenate matrices.

Funding Statement

The work described herein was funded by Merck & Co., Rahway, NJ, USA.

Financial disclosure

The work described herein was funded by Merck & Co., Rahway, NJ, USA. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

All authors are employees of Merck & Co., Rahway, NJ, USA. The authors have no other competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript apart from those disclosed.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Ethical conduct of research

The clinical study was approved by National Research Ethics Service Committee South Central – Oxford B. Trial registration number: EudraCT 2010-023467, Date: 2011-01-25. Written consent was obtained from the participants. The preclinical studies were reviewed and approved by the Merck IACUC – Institutionalized Animal Care and Use Committee. The study approval numbers were #12067336020169, #13060026020172, #13020026020099, and #13020026020092.

Data availability statement

The authors certify that this manuscript reports original clinical trial data. The data will not be made publicly available.