Quantitation of TAK-981 in human plasma via LC-MS/MS and its application in clinical trials

ABSTRACT

Aim

TAK-981 is a highly effective and selective inhibitor of the small ubiquitin-like modifier (SUMO) activating enzyme, and it promotes the expression of type I interferons (IFN-Is). Developing a sensitive bioanalytical assay for quantitating TAK-981 is essential in the clinical investigations for oncology drug development.

Materials & methods

TAK-981 and its stable isotope labeled compound (TAK-981¹³C₄, ¹⁵N₂) as the internal standard were employed in this LC-MS/MS assay.

Results

This assay was successfully validated from 0.1 ng/mL to 100 ng/mL with good accuracy and precision and has been applied to support clinical studies.

Conclusion

A sensitive and robust LC-MS/MS assay was validated for TAK-981 in human plasma for the first time.

Plain Language Summary

TAK-981 is a drug being studied to help treat cancer. It blocks a certain enzyme and helps the immune system work better. A special lab test was made to measure how much TAK-981 is in the blood after someone takes it. The test worked well at many different levels and gave correct and steady results. This is the first time such a reliable test has been made to measure TAK-981 in human blood for clinical research.

1. Introduction

Small ubiquitin-like modifier (SUMO) is covalently attached to lysine residues of a variety of proteins after translation, acting as a post-translational modifier within the ubiquitin family that changes the function and behavior of the target proteins [1–24]. SUMOylation is disrupted in various hematological malignancies, including multiple myeloma (MM), leukemia, and lymphoma, where it contributes to both tumorigenesis and the cancer cells’ response to treatments [16,24]. The SUMOylation pathway is frequently upregulated in MM compared to normal B cells and plasma cells from healthy individuals [16,25]. The overactivation of SUMOylation may serve a protective role in MM cells and is linked to poor patient outcomes [25]. Inhibiting the SUMOylation offers a novel therapeutic strategy for treating MM, as it not only reduces cancer cell proliferation but also affects type I interferon (IFN-I) [16,26–29]. IFN-Is, particularly IFNα and IFNβ, are key cytokines in the innate immune system that rapidly activate in response to pathogens. They influence various immune cells, including natural killer (NK) cells, by directly enhancing their cytotoxicity. Additionally, IFN-Is stimulate dendritic cells to produce interleukin-15 (IL-15), which further promotes NK cell activation and strengthens the immune response against infections [30–35]. IFN-Is also directly influence T cells, promoting their survival, clonal expansion, and the development of effector functions [36–42].

TAK-981 (Figure 1) is a first-in-class small-molecule drug targeting SUMOylation. TAK-981 is a potent and selective inhibitor of SUMO-activating enzyme (SAE), blocking SUMOylation by binding to the SAE and forming a covalent adduct with SUMO [43,44]. TAK-981 effectively traps the SUMO molecule at the active site of the SAE, prevents the transfer of SUMO to other proteins, and inhibits the process of SUMOylation [43,44]. TAK-981 blocks cancer cell progression and stimulates anti-tumor immunity by triggering interferon signaling [26,45,46]. TAK-981 demonstrated antileukemic activity by inducing apoptosis, causing cell-cycle arrest, and promoting the expression of differentiation marker in leukemic cells [47]. When combined with rituximab or 5-azacytidine, TAK-981 enhances macrophage phagocytosis and NK cell cytotoxicity, leading to synergistic antitumor activity in vivo [48,49]. When used in combination with avelumab or cetuximab, TAK-981 enhanced and extended IFN signaling and increased expression of CD86 and granzyme B in the tumor microenvironment (TME) [50]. The combination of TAK-981 and trametinib significantly induced apoptosis in various cancer cell lines and gene-engineered mouse-derived organoids [51].

Figure 1.

TAK-981 and the internal standard TAK-981-¹³C₄, ¹⁵N₂ structures.

The use of LC-MS/MS for measuring concentrations of oncology drugs has been widely published due to its sensitivity, selectivity, and specificity, etc. [52–56]. A sensitive reversed-phase LC-MS/MS assay was recently developed and validated in our lab to quantitate TAK-981 in human K₂EDTA plasma to support multiple completed and ongoing clinical trials [57–61]. This assay validation followed US Food and Drug Administration (FDA) regulatory guidance [62,63].

2. Experimental

2.1. Chemicals & reagents

Acetonitrile (ACN) (HPLC grade) and 0.1% formic acid in acetonitrile (HPLC grade) were obtained from Honeywell Burdick & Jackson (Morris Plains, NJ, USA). Dimethyl sulfoxide (DMSO) (≥99.9%) and N,N-dimethylformamide (DMF) (HPLC grade) were purchased from SigmaAldrich (St. Louis, MO, USA). Formic acid (98–100%) was produced by EMD Millipore (Burlington, MA, USA). Water was purified on-site using the Millipore Milli-Q IQ 7000 ultrapure lab waters system (Burlington, MA, USA). 2-Propanol (HPLC grade) was provided by J. T. Baker (Phillipsburg, NJ, USA). K₂EDTA human plasma was acquired from BioIVT (Westbury, NY, USA). TAK-981 and TAK-981-¹³C₄, ¹⁵N₂ were supplied by Takeda Development Center Americas, Inc. (Cambridge, MA, USA).

2.2. LC-MS/MS platform

The LC-MS/MS platform comprised an API-5000 triple quadrupole mass spectrometer (Sciex, Framingham, MA) and a Shimadzu high-performance liquid chromatography (HPLC) system (HPLC controller, CBM-20A, HPLC pumps, LC-30AD, HPLC autosampler, SIL-30AC) (Shimadzu Scientific, Columbia, MD). Data were acquired and processed using Analyst® 1.6.2 software (Sciex, Framingham, MA). An ACQUITY HSS T₃ analytical column (Waters, Milford, MA) (2.1 × 50 mm, 1.8 μm) was employed in the method validation.

2.3. Calibration curve standards and quality control samples preparation

TAK-981 reference material was accurately weighed using a microbalance, adjusted based on potency, and dissolved in DMF in polypropylene tubes (Sarstedt, Newton, NC) to generate the stock solution at 1,000 µg/mL. Further dilution of the stock solution in DMSO/ACN (50:50, v/v) produced the working solution at 5 µg/mL. When not in use, stock solutions and working solutions were kept in a −20°C freezer. Twenty microliters of the working solution was spiked into 980 µL of human plasma to prepare the highest-concentration calibration standard at 100 ng/mL. Serial dilution of this standard in human plasma produced a series of calibration standards ranging from 0.100 ng/mL to 100 ng/mL, with concentrations of 0.100, 0.200, 1.00, 4.00, 10.0, 50.0, 80.0, and 100 ng/mL in human K₂EDTA plasma. In this assay validation, calibration standard curves were prepared fresh daily and this assay required 100 μL of human K₂EDTA plasma.

Five hundred and twenty-five microliters of the quality control (QC) working solution was spiked into 34,475 µL of human plasma to prepare the HQC sample at 75.0 ng/mL. Serial dilution of the HQC sample in human plasma was used to prepare QC samples at three additional concentration levels (0.100 ng/mL (LLOQ), 0.300 ng/mL (LQC), and 30.0 ng/mL (MQC)). The intraday and inter-day accuracy and precision were evaluated in QC samples analyzed in six replicates and in three separate batches (Table 1). Additional run acceptance QC samples for supplemental validation runs were analyzed in duplicates in two separate batches and the accuracy and precision data were depicted in Supplemental Table S1. Extra LQC and HQC samples intended for stability testing were stored at −20°C and −70°C after preparation.

Table 1.

Intra-run and inter-run accuracy and precision for TAK-981 in human K₂EDTA plasma.

Run Number	LLOQ 0.100 ng/mL	LQC 0.300 ng/mL	MQC 30.0 ng/mL	HQC 75.0 ng/mL
1	0.0938	0.309	32.0	79.6
	0.0972	0.325	31.9	77.6
	0.0951	0.309	30.9	75.3
	0.104	0.317	32.9	77.5
	0.0958	0.288	30.8	78.8
	0.0991	0.307	32.2	77.6
Intra-run Mean	0.0975	0.309	31.8	77.7
Intra-run S.D.	0.00367	0.0124	0.804	1.46
Intra-run %CV	3.8	4.0	2.5	1.9
Intra-run %RE	−2.5	3.0	6.0	3.6
n	6	6	6	6
2	0.0940	0.342	28.9	76.2
	0.0929	0.256	30.2	73.3
	0.108	0.287	30.1	72.7
	0.0854	0.291	29.5	74.8
	0.102	0.289	30.0	71.1
	0.0890	0.311	29.5	77.2
Intra-run Mean	0.0952	0.296	29.7	74.2
Intra-run S.D.	0.00838	0.0286	0.494	2.28
Intra-run %CV	8.8	9.7	1.7	3.1
Intra-run %RE	−4.8	−1.3	−1.0	−1.1
n	6	6	6	6
3	0.109	0.323	31.1	76.7
	0.107	0.346*	31.6	74.1
	0.113	0.325	33.4	76.9
	0.109	0.320	32.8	76.6
	0.112	0.318	31.2	74.8
	0.105	0.334	31.1	75.3
Intra-run Mean	0.109	0.328	31.9	75.7
Intra-run S.D.	0.00299	0.0106	0.991	1.16
Intra-run %CV	2.7	3.2	3.1	1.5
Intra-run %RE	9.0	9.3	6.3	0.9
n	6	6	6	6
Inter-run Mean	0.101	0.311	31.1	75.9
Inter-run S.D.	0.00817	0.0223	1.27	2.18
Inter-run %CV	8.1	7.2	4.1	2.9
Inter-run %RE	1.0	3.7	3.7	1.2
n	18	18	18	18

Note: *: Deviation from nominal > 15%; value included in statistics.

%CV: percent coefficient of variation; %RE: percent relative error; HQC: high-quality control; LLOQ: lower limit of quantitation; LQC: low-quality control; MQC: medium-quality control; n: number; S.D.: standard deviation.

A stable isotope-labeled compound TAK-981-¹³C₄, ¹⁵N₂ was utilized as the internal standard (IS). Stock solution of the IS was prepared in DMF at 100 µg/mL and working solution of the IS was prepared in 50/50 ACN/H₂O at 20.0 ng/mL. The IS stock and working solutions were stored at -20°C when not used.

2.4. Sample preparation

Samples (standards, QCs, and study samples) were thawed using an ice bath. The analyte and IS were extracted using a protein precipitation procedure. One hundred microliters of human K₂EDTA plasma sample was pipetted into individual wells of a 1-mL LoBind 96-well plate (Eppendorf, Hamburg, Germany) and fortified with 20 µL of 20.0 ng/mL TAK-981-¹³C₄, ¹⁵N₂ IS working solution, then vortex-mixed. Five hundred microliters of 0.1% formic acid in ACN was added in each sample, and all samples were vortex-mixed then centrifuged at room temperature for 5 min. Four hundred microliters of the supernatant was transferred into a clean 1-mL LoBind 96-well plate and evaporated to dryness under a medium nitrogen stream at approximately 40°C. One hundred microliters of ACN/water (10:90, v/v) was added in each well to reconstitute the residue for each sample. The samples were vortex-mixed then centrifuged at room temperature for 1 min. A 5 µL sample of the supernatant was subjected to LC-MS/MS analysis.

2.5. Data acquisition and processing software

LC-MS/MS data acquisition and processing were performed using Analyst 1.6.2 (Sciex, Framingham, MA). A weighted linear regression model (1/x²) was applied for calibration, and statistical parameters such as mean, standard deviation (S.D.), percent relative error (%RE), and percent coefficient of variation (%CV) were calculated using Watson LIMS version 7.4.2 (Thermo Fisher Scientific, MA).

3. Results & discussion

3.1. LC-MS/MS settings

An ACQUITY HSS T₃ column (2.1 × 50 mm, 1.8 µM) was used for chromatographic separation. Water containing 0.1% formic acid (v/v) and ACN containing 0.1% formic acid (v/v) were used as mobile phase A and B, respectively. The column was kept at 60°C to ensure sharp and symmetric peaks. The flow rate was initially set to 600 µL/min, and the following gradient program was applied: 0.00 min (20% B), 2.00 min (20% B), 2.10 min (90% B), 2.90 min (90% B), 2.95 min (20% B), and 3.30 min (20% B). The flow rate was increased to 1400 µL/min from 2.12 min to 2.90 min for the purpose of saving the LC-MS run time. The run time was 3.30 min. The autosampler temperature was kept at 4°C. HPLC effluents before 2.25 min were let flow into the mass detector.

Turbo ion spray was used as the ion source in positive ion mode on a Sciex API 5000 mass spectrometer to monitor multiple reaction monitoring (MRM) with MS transitions m/z 578.1 > 481.2 for TAK-981 and m/z 584.1 > 487.2 for TAK-981-¹³C₄, ¹⁵N₂ in this assay. The resolution for both Q1 and Q3 was set to unit mode. Source temperature was kept at 600°C, ion spray voltage was set at 3000 V, curtain gas was held at 35, collision gas was optimized to be 8, GS1 and GS2 were selected at 70, declustering potential (DP) was chosen to be 90, and collision energy (CE) was maintained at 36. The structures of TAK-981 and TAK-981-¹³C₄, ¹⁵N₂ are displayed in Figure 1.

3.2. Assay range and sensitivity

TAK-981 was validated at a nominal concentration range of 0.100 to 100 ng/mL in this assay, and it was adequate to quantitate TAK-981 in human K₂EDTA plasma. Figure 2 showed representative chromatograms of TAK-981 and TAK-981-¹³C₄, ¹⁵N₂ from a matrix blank spiked with the internal standard, and representative chromatograms from an LLOQ sample in human K₂EDTA plasma was depicted in Figure 3, demonstrating adequate sensitivity to measure TAK-981 at 0.100 ng/mL in matrix.

Figure 2.

Chromatograms of TAK-981 (top) and the internal standard, TAK-981-¹³C₄, ¹⁵N₂ (bottom) from a matrix blank with internal standard in human K₂EDTA plasma sample.

Figure 3.

Chromatograms of TAK-981 (top) and is TAK-981-¹³C₄, ¹⁵N₂ (bottom) from an LLOQ sample in human K₂EDTA plasma.

3.3. Assay selectivity and specificity

The assay selectivity and specificity were evaluated to confirm the capability of this bioanalytical assay to distinguish and quantify the analyte amidst possible interfering components in the blank biological matrix [62,63]. Six individual normal human K₂EDTA plasma lots, along with one hemolyzed and one lipemic lot, were assessed in this assay, in the presence and absence of the IS, to evaluate the assay selectivity. It was observed that the analyte response ratios (interfering background peak response divided by internal standard peak response) in all the blank with IS samples were less than 20% of the response ratios in the corresponding LLOQ samples, and the IS responses in all the blank samples were below 5% of the average IS responses in the calibrators and QC samples. These results demonstrate that the assay is selective and specific for TAK-981 analysis in human plasma.

3.4. Accuracy and precision

Three accuracy and precision (A&P) runs were analyzed for TAK-981 QC samples prepared in human K₂EDTA plasma. QC samples were prepared by spiking pooled blank matrix with TAK-981, and A&P were calculated based on the results of these QC samples. Accuracy and precision were depicted as the %RE and the %CV, respectively. Accuracy signifies the measured concentration’s percentage compared to the nominal concentration, while precision indicates the proximity of replicate analyses. Six replicates at each of the four QC levels (LLOQ, LQC, MQC, and HQC) were examined to determine the intra-assay A&P. Additionally, three intra-assay A&P runs were combined to evaluate inter-assay A&P. The QC data satisfied the acceptance criteria for both intra-assay and inter-assay validation, as per regulatory guidance [62,63]. The corresponding statistical details are provided in Table 1. The intra-run %RE varied between −4.8% and 9.3% with %CV between 1.5% and 9.7% for all QC levels in human K₂EDTA plasma. The inter-run %RE ranged from 1.0% to 3.7% with %CV between 2.9% and 8.1% across all QC concentrations in human K₂EDTA plasma.

3.5. Run acceptance QC samples for supplemental validation runs

In each supplemental validation run (excluding A&P assessments), six replicates of three QC levels (LQC, MQC, and HQC) were analyzed (n = 6 per level). All QC samples across the analytical runs met the acceptance criteria, ensuring that at least 50% of the replicates at each QC level and two-thirds of all QC samples within each run falling within ± 15% for %RE and ≤15% for %CV (Supplemental Table S1).

3.6. Dilution integrity

To expand the assay’s dynamic range, dilution integrity was demonstrated by diluting a QC sample that exceeded the upper limit of quantitation (ULOQ). A human plasma sample prepared at the target concentration of 1,000 ng/mL underwent a 20-fold dilution in the same matrix. Subsequently, the diluted sample was evaluated against a newly prepared standard curve in 18 replicates over 3 days. The results were presented in Table 2. The inter-run %RE is −1.8% with a %CV 3.1% in the human K₂EDTA plasma assay. The dilution QC data satisfied the acceptance criteria outlined in the regulatory guidance [62,63]. This validation demonstrated that human plasma samples with TAK-981 concentrations higher than ULOQ and up to 1,000 ng/mL can be diluted 20-fold in human plasma and analyzed in this assay.

Table 2.

Accuracy and precision of TAK-981 in human plasma for dilution quality control samples.

Run Number	Dil 20 QC 1,000 ng/mL
1	954
	961
	953
	943
	961
	960
2	971
	1010
	1010
	1030
	998
	974
3	998
	969
	963
	1010
	966
	1050
Mean	982
S.D.	30.0
%CV	3.1
%RE	−1.8
n	18

Note: %CV: percent coefficient of variation; %RE: percent relative error; Dil 20 QC: 20-fold dilution QC; n: number; S.D.: standard deviation.

3.7. Assessment of stability

Stability test has confirmed that TAK-981 remains stable for at least 24 h in human plasma at 4°C, after undergoing 5 freeze (−20°C or −70°C)/thaw (ice bath) cycles in human plasma, for at least 215 days when stored in human plasma at −20°C or at least 552 days when stored in human plasma at −70°C, and for at least 2 h when stored in human whole blood in an ice bath. Additionally, TAK-981 sample extracts maintain stability for at least 69 h at 4°C, and reinjection of TAK-981 sample extracts is feasible after being stored at 4°C for 83 h.

TAK-981 benchtop stability was assessed by thawing a set of LQCs, HQCs, and dilution QCs and keeping them at 4°C. TAK-981 was demonstrated to be stable in human plasma at 4°C for at least 24 h (Table 3).

Table 3.

TAK-981 benchtop stability in human K₂EDTA plasma.

Run Number	LQC 0.300 ng/mL	HQC 75.0 ng/mL	Dil 20 QC 1,000 ng/mL
3	0.319	73.4	1,060
Stored at 4°C	0.315	75.9	1,070
for 24 hours	0.316	75.2	990
	0.346*	78.2	1,060
Mean	0.324	75.7	1,050
S.D.	0.0148	1.99	37.0
%CV	4.6	2.6	3.5
%RE	8.0	0.9	5.0
n	4	4	4

Note: *: Deviation from nominal > 15%; value included in statistics.

%CV: percent coefficient of variation; %RE: percent relative error; Dil 20 QC: 20-fold dilution QC; HQC: high-quality control; LQC: low-quality control; n: number; S.D.: standard deviation.

Freeze/thaw stability of TAK-981 was assessed using two sets of LQCs, HQCs, and dilution QCs subjected to five freeze/thaw cycles. One set was frozen at −20°C and the other at −70°C for each freeze cycle, followed by thawing in an ice bath. TAK-981 was shown to be stable after undergoing five freeze (−20°C or 70°C)/thaw (ice bath) cycles (Table 4).

Table 4.

TAK-981 stability in human K₂EDTA plasma after five freeze/thaw cycles.

Run Number	LQC 0.300 ng/mL	HQC 75.0 ng/mL	Dil 20 QC 1,000 ng/mL
4	0.286	68.1	931
Freeze at 20°C and	0.272	71.9	887
thaw in an ice bath	0.300	68.8	850
	0.305	72.3	900
Mean	0.291	70.3	892
S.D.	0.0149	2.13	33.5
%CV	5.1	3.0	3.8
%RE	−3.0	−6.3	−10.8
n	4	4	4
Freeze at 70°C and	0.296	69.9	929
thaw in an ice bath	0.287	67.9	888
	0.297	66.8	900
	0.288	67.0	875
Mean	0.292	67.9	898
S.D.	0.00523	1.42	23.1
%CV	1.8	2.1	2.6
%RE	−2.7	−9.5	−10.2
n	4	4	4

Note: %CV: percent coefficient of variation; %RE: percent relative error; Dil 20 QC: 20-fold dilution QC; HQC: high-quality control; LQC: low-quality control; n: number; S.D.: standard deviation.

TAK-981 long-term stability was assessed by measuring a set of LQCs, HQCs, and dilution QCs stored at −20°C or −70°C. TAK-981 was demonstrated stable for at least 215 days at −20°C and at least 552 days at −70°C (Table 5). Additional LQC, HQC, and dilution QC samples are currently stored at −20°C and −70°C for future long-term stability testing to support the clinical trial.

Table 5.

TAK-981 long-term stability in human K₂EDTA plasma.

Run Number	LQC 0.300 ng/mL	HQC 75.0 ng/mL	Dil 20 QC 1,000 ng/mL
5	0.302	66.1	991
Stored at 20°C	0.304	66.7	1,010
for 215 days	0.295	66.3	965
	0.306	65.2	994
Mean	0.302	66.1	990
S.D.	0.00479	0.634	18.6
%CV	1.6	1.0	1.9
%RE	0.7	−11.9	−1.0
n	4	4	4
Stored at 70°C	0.328	75.6	949
for 552 days	0.326	76.2	966
	0.329	76.1	966
	0.333	76.8	985
Mean	0.329	76.2	967
S.D.	0.00294	0.492	14.7
%CV	0.9	0.6	1.5
%RE	9.7	1.6	−3.3
n	4	4	4

Note: %CV: percent coefficient of variation; %RE: percent relative error; Dil 20 QC: 20-fold dilution QC; HQC: high-quality control; LQC: low-quality control; n: number; S.D.: standard deviation.

TAK-981 stability in human whole blood was assessed at 10 times the LLQC (1.0 ng/mL) and HQC (75.0 ng/mL) levels. Control samples for whole blood stability were promptly processed to plasma after blood sample collection. Non-control samples were kept in an ice bath before being processed to plasma. Stability of the analyte was concluded when the %CV was ≤15% and the %difference between the mean peak area ratios (PARs) and the respective control samples remained within ±15%. The whole blood stability testing confirmed that TAK-981 remains stable in human whole blood for up to 2 h in an ice bath before processing into plasma via centrifugation at 4°C, as all samples met the predefined acceptance criteria.

TAK-981 autosampler stability was assessed by analyzing LQCs, MQCs, and HQCs. These autosampler stability QC samples were extracted, injected, and subsequently stored at 4°C before being reanalyzed with freshly prepared calibration standards and QCs. TAK-981 sample extract was demonstrated to be stable for at least 69 h at 4°C (Supplemental Table S2).

The reproducibility of TAK-981 reinjection was assessed by re-injecting a set of calibration standards and QCs that had initially been injected and subsequently stored at 4°C for 83 h. The reinjection reproducibility was confirmed as calibration standards and QC samples met the acceptance criteria outlined in regulatory guidance in Supplemental Table S3 [62,63].

3.8. Hemolysis effect

To evaluate the effect of hemolysis on TAK-981 accuracy and precision, LLOQ and HQC samples were enriched with hemolyzed whole human blood to simulate 2% hemolysis. These samples were then analyzed in replicate (n = 3). Additionally, matrix blanks with and without IS were assessed. No significant chromatographic interferences were observed at the mass transitions and expected retention times of the analyte or the IS, confirming the absence of interference with quantitation. Hemolysis does not impact TAK-981 quantitation.

3.9. Lipemic effect

To assess the impact of lipemia on TAK-981 accuracy and precision, LLOQ and HQC samples were prepared in lipemic human plasma. These samples were then analyzed in replicate (n = 3). Additionally, matrix blanks with and without IS were assessed. No significant chromatographic interferences were observed at the mass transitions and expected retention times of the analyte or the IS, confirming the absence of interference with quantitation. Lipemia does not impact TAK-981 quantitation.

3.10. Recovery

To assess the recovery of both the analyte and the IS, responses from pre-extraction fortified samples (n = 6 per level) were compared with those from post-extraction fortified samples, considered to represent 100% recovery (n = 6 per level). The overall recovery was determined to be 91.9% for TAK-981 and 91.3% for the IS TAK-981-¹³C₄, ¹⁵N₂.

3.11. Matrix effect

The presence of endogenous substances in sample extracts may influence ionization efficiency, leading to MS signal suppression or enhancement. To evaluate matrix effects, we prepared LQCs and HQCs using six individual lots of normal human plasma, one lot of lipemic human plasma, and one lot of hemolyzed human plasma. To calculate the IS-normalized matrix factors, the peak area ratio in matrix-containing samples was divided by the average peak area ratio from matrix-free samples. As the IS-normalized matrix factor ranged from 0.93 to 1.04 across all human plasma lots tested, the matrix effects were considered consistent. There were no significant matrix effects that could undermine assay accuracy or precision.

3.12. Carryover

To assess the potential carryover of TAK-981 from high-concentration samples to the subsequent injections, an extracted matrix blank was injected immediately after the ULOQ calibration standards in each validation run. No chromatographic peaks were observed at the expected retention time of the analyte in the blank sample that exceeded 20% of the mean analyte response at the LLOQ level, indicating no significant carryover.

3.13. Run length evaluation

To replicate the duration and length of a typical analytical run, additional matrix blank samples were extracted and evaluated within a single analytical run comprising 192 injections over two 96-well plates. Each calibration standard set was included at the beginning and end of the analytical sequence, with QCs distributed across the run. The data gathered from this study revealed no indications of system performance deterioration throughout the entirety of the run, containing a total of 192 injections over two 96-well plates.

4. Assay application

This validated assay has been effectively used to analyze TAK-981 concentrations in human K₂EDTA plasma samples across five clinical trials [57–61], including TAK-981–1002 (An open label, dose-escalation, phase 1/2 study to evaluate the safety, tolerability, preliminary efficacy, and pharmacokinetics of TAK-981 in adult patients with advanced or metastatic solid tumors or relapsed/refractory hematologic malignancies) (ClinicalTrials.gov identifier No. NCT03648372). Patients received TAK-981 at doses ranging from 3 mg BIW (twice weekly) to 120 mg BIW, administered on day 1, 4, 8, and 11 of the first cycle. Pharmacokinetic (PK) plasma samples were collected on day 1 (predose, end of infusion, 0.5 h, 1 h, 2 h, 4 h, 8 h, 24 h, and 48 h), day 4 (predose and end of infusion), and day 8 (predose, end of infusion, 0.5 h, 1 h, 2 h, 4 h, 8 h, and 24 h). Figure 4 illustrates two mean concentration–time profiles of TAK-981, one at the lowest dose level (3 mg intravenous infusion, twice a week, n = 5) and the other at the highest dose level (120 mg intravenous infusion, twice a week, n = 8) on cycle 1, day 1. Similarly, Figure 5 presents two mean concentration–time profiles of TAK-981 at the lowest dose level (6 mg intravenous infusion, twice a week, n = 2) and the highest dose level (120 mg intravenous infusion, twice a week, n = 6) on cycle 1, day 8. Clinical PK analysis is not within the scope of this study. PK parameters, including C_max, T_max, AUC, and t_1/2, will be reported in a separate, future manuscript.

Figure 4.

Mean concentration-time profile of TAK-981 at the lowest dose (3 mg IV infusion, twice a week (BIW), n = 5) (top) and the highest dose (120 mg IV infusion, BIW, n = 8) (bottom) on cycle 1, day 1 (C1D1).

Figure 5.

Mean concentration-time profile of TAK-981 at the lowest dose (6 mg IV infusion, BIW, n = 2) (top) and the highest dose (120 mg IV infusion, BIW, n = 6) (bottom) on cycle 1, day 8 (C1D8).

5. Conclusion & future perspective

A sensitive and selective bioanalytical LC-MS/MS assay was validated for the quantitation of TAK-981 in human plasma. The stable isotope labeled (SIL) IS, TAK-981-¹³C₄, ¹⁵N₂, was spiked in human plasma, followed by extraction of both the analyte and the IS via protein precipitation and subsequent analysis in MRM mode on an LC-MS/MS system. A linear calibration curve employing 1/x² weighted regression was utilized in this assay validation. The intra-run %RE ranged from −4.8% to 9.3% with %CV ≤ 9.7% across all QC levels in human K₂EDTA plasma. The inter-run %RE ranged from 1.0% to 3.7% with %CV ≤ 8.1% for all QC levels in human K₂EDTA plasma. Additionally, TAK-981 can be diluted 20-fold in human plasma and reliably analyzed in this assay. TAK-981 exhibited stability in human plasma for at least 24 h at 4°C, at least 215 days when stored at −20°C, at least 552 days when stored at −70°C, and after undergoing five freeze/thaw cycles. Furthermore, TAK-981 remained stable for up to 2 h in human whole blood in an ice bath.

Briefly, a bioanalytical LC-MS/MS assay has been successfully validated in human plasma in accordance with the regulatory guidance [62,63], and this assay has been effectively applied in clinical plasma sample analysis, supporting the TAK-981 clinical trials.

Funding Statement

This paper was not funded.

Article highlights

• A sensitive LC-MS/MS assay was validated for the quantitation of TAK-981 in human K₂EDTA plasma.

• TAK-981-¹³C₄, ¹⁵N₂ (a stable isotope labeled compound) was employed as the internal standard.

• Sensitivity, selectivity, accuracy and precision were established in accordance with FDA guidance for bioanalytical method validation.

• TAK-981 was shown to be stable under various conditions, including benchtop, freeze/thaw, post-extraction, long-term, and whole blood stability studies.

• This LC-MS/MS assay has been employed to support multiple clinical trials.

Disclosure statement

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: all authors are employees and stockholders of Takeda Development Center Americas, Inc.

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.

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

Author contributions

Feng Yin: Methodology, Investigation, Validation, Writing – original draft. Ran Ye: Methodology, Investigation, Validation, Writing – review & editing. Anson Pierce: Methodology, Investigation, Writing – review & editing. John Gibbs: Resources, Investigation, Visualization, Writing – review & editing. Mike Baratta: Resources, Methodology, Investigation, Validation, Supervision, Writing – review & editing.

Ethical declaration

The authors state that they have obtained appropriate institutional review board approval or have followed the principles outlined in the Declaration of Helsinki for all human or animal experimental investigations.

Data availability statement

The data are available from the corresponding author upon sensible request.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17576180.2025.2518048