Potential interactions between triazole antifungal agents and lorlatinib based on ultra-performance liquid chromatography-tandem mass spectrometry in rat plasma

Abstract

Aim

Our study is to investigate the effects of triazole antifungal drugs on the pharmacokinetics of lorlatinib in rats.

Methods

The samples were precipitated with methanol. Chromatographic separation was performed on a ultra-performance liquid chromatography (UPLC) system using a BEH C18 column. The mobile phase consisted of 0.1% formic acid water and methanol. Lorlatinib and crizotinib (internal standard) were detected in multiple reaction monitoring mode. The fragment ions were 407.3–228.07 for lorlatinib and m/z 450.3–260.0 for crizotinib. Lorlatinib and different triazole antifungal drugs were given to Sprague Dawley rats by gavage, and blood was collected from the tail vein at a certain time point. The validated UPLC–MS/MS method was applied to a drug interaction study of ketoconazole, voriconazole, itraconazole, and posaconazole with lorlatinib in rats.

Results

Ketoconazole and voriconazole significantly inhibited lorlatinib metabolism. When administration with ketoconazole and voriconazole, the area under the curve from time zero to infinity of lorlatinib increased by 49.0% and 104.3%, respectively; the clearance decreased by 40.0% and 40.0%, respectively. While itraconazole and posaconazole did not affect lorlatinib pharmacokinetics.

Conclusion

The UPLC-MS/MS-based assay is helpful to further understand the pharmacokinetics of lorlatinib in rats, and confirmed the findings that the combination of lorlatinib with CYP3A inhibitors should be avoided as predicted by our pre-clinical studies.

Introduction

Non-small cell lung cancer (NSCLC) is the most common cause of cancer-related deaths worldwide annually. Its occurrence is associated with several driver gene mutations [1]. Anaplastic lymphoma kinase (ALK) mutation is usually described as a diamond mutation because of its low incidence. Only 3%–8% of lung cancer patients have ALK mutations [2]. However, these patients benefit from tyrosine kinase inhibitors (TKIs). Unfortunately, the problem of drug resistance has not yet been resolved. Acquired drug resistance has been a problem for the marketed ALK-TKI including crizotinib, alectinib, ceritinib, and brigatinib [3]. For this reason, the third-generation ALK-TKI lorlatinib has been designed targeting these drug-resistant mutations. Lorlatinib was marketed in 2018 and is used for newly diagnosed ALK mutation-advanced NSCLC or patients who have progressed after receiving other ALK-TKIs (Fig. 1a). It is also effective for patients with brain metastasis. Several clinical studies showed the significant clinical efficacy of lorlatinib on ALK-positive NSCLC [4–8].

The pharmacokinetics and metabolic features of lorlatinib have been characterized. It occurs rapid absorption after a 100-mg administration of lorlatinib. Peak plasma concentrations occur from 0.5 to 4 h. The absolute bioavailability is 81%, and the half-life is 24 h [9]. The N-demethylation and N-oxidation mediated via Cytochrome P450 (CYP) 3A, and N-glucuronidation via uridine diphosphate glucuronic acid transferase 1A4 are primarily responsible for lorlatinib metabolism [10]. That is to say, CYP3A inducers and inhibitors may change the metabolism of lorlatinib, resulting in exposure fluctuations in vivo.

Triazole antifungal agents are first-line drugs for preventing and treating systemic mycoses. Ketoconazole acts against dermatophytes, candida, and aspergillus [11]. Voriconazole is used for preventing and curing invasive and superficial yeast fungal infections [12]. Itraconazole treats fungal skin, dematiaceous fungi, acute and chronic aspergillosis, and allergic bronchopulmonary aspergillosis [13]. Posaconazole is usually applied for oropharyngeal candidiasis refractory to itraconazole [14]. In the long-term treatment of NSCLC, fungal infections may cause significant complications, which makes possible for simultaneous use of triazole antifungals and lorlatinib in the clinic. However, triazole antifungal agents are CYP3A4 inhibitors and may cause drug-drug interactions with lorlatinib, leading to toxicity and adverse drug reactions. Therefore, effective methods are required to evaluate triazole antifungal agents’ effects on lorlatinib metabolism.

Few studies evaluate the drug–drug interactions among lorlatinib and triazole antifungals. Therefore, we developed an ultra-performance liquid chromatography-tandem mass spectrometry (UPLC–MS/MS) approach for assaying lorlatinib plasma concentration and apply it for estimating the impacts of ketoconazole, voriconazole, itraconazole, and posaconazole to lorlatinib pharmacokinetics.

Reagents and methods

Reagents

Lorlatinib (purity > 98%) was obtained from Aladdin Biochemical Technology Co., Ltd (Shanghai, China). Crizotinib (purity > 99%, internal standard, IS) was obtained from Energy Chemical (Guangzhou, China, Fig. 1b). Ketoconazole and itraconazole (purity > 98%) were purchased from Tixiae Chemical Industrial Development Co., Ltd (Shanghai, China). Posaconazole (purity > 98%) was obtained from Energy Chemical (Guangzhou, China). Voriconazole (purity > 99%) was obtained from Yimiao Chemical Technology (Shanghai, China). Chromatographic grades of formic acid and methanol were purchased from Merck KGaA (Darmstadt, Germany).

Instrument condition

The determination of analytes was conducted on an UPLC–MS/MS system, which is made up of an ACQUITY UPLC Ⅰ-Class system and a Xevo TQ-S triple quadrupole tandem mass spectrometer (Milford, MA, USA). All detected data were acquired and processed using MassLynx and TargetLynx software (Milford, MA, USA).

A BEH C18 column (2.1 × 100 mm,1.7 μm; Waters) was applied to separate the analysts. The chromatographic separation was performed with a mobile phase made up of methanol (A) and 0.1% formic acid (B). The gradient elution program was as follows: 0–0.5 min, 70% A to 30% A; 0.5–1.0 min, 30% A to 30% A; 1.0–2.5 min, 30% A to 70% A; 2.5–3.0 min, 70% A to 70% A; flow rate was set of 0.30 ml/min; total run time was 3.0 min. The volume of each injection was 1.0 μl. The temperature of autosampler and column were 10°C and 40°C, respectively.

Lorlatinib and IS were measured with multiple reaction monitoring mode, and the precursor-to-product ion transitions were m/z 407.3–228.07 for lorlatinib quantitation, m/z 407.3–180.05 for lorlatinib quantification, and m/z 450.3–260.0 for IS quantitation. The cone voltages were 54 V, 54 V, and 50 V, respectively; the collision energies were 18 eV, 20 eV, and 18 eV, respectively.

Sample treatment

The Eppendorf tube was filled with 90 μl of rat plasma, followed by 10 μl of IS solution, and then 500 μl of methanol. After the full shock, the mixture was centrifuged at 13 000 rpm, 4°C for 5 min. Then 100 μl of supernatant was transferred to the injection bottle, and 1 μl of supernatant was injected for analysis.

Method validation

Specificity

The specificity of the method was assessed by comparing the chromatograms of blank plasma from rats, blank plasma added with lorlatinib and IS, and plasma samples from rats orally administered lorlatinib with triazole antifungal.

Standard curve

Lorlatinib was dissolved in methanol to prepare 1.0 mg/ml of stock solution. The working solutions were produced by diluting the stock solution with methanol. The standard curve solutions were prepared by spiking 10 μl of working solutions into 80 μl of blank rat plasma and the concentrations were set 1, 5, 50, 500, 1000, and 5000 ng/ml. IS solution (200 ng/ml) was dissolved in a mixed solution of methanol and water (1:1).

Quality control standards

The concentrations of quality control (QC) samples for the method were set at 1 ng/ml, 3 ng/ml, 2500 ng/ml and 4000 ng/ml for the lower limit of quantification (LLOQ), low-quality control (LQC), mid-quality control (MQC), and high-quality control (HQC), respectively.

Accuracy and precision

Four levels of QC samples in five replicates were analyzed on the same day to determine the intra-day precision and accuracy and on three consecutive days to determine the inter-day precision and accuracy. The accuracy and precision shall be guaranteed to be less than 15% (± 20% for LLOQ).

Matrix effect and recovery

The matrix effect was assessed by comparing the peak areas of analyte added in the processed blank plasma with those of the analyte in methanol at the QC levels. The recovery was assessed by comparing the peak areas of analyte from the QC samples with those of the analyte added in the processed blank plasma at the corresponding concentration levels. Five replicates of LQC, MQC, and HQC samples were used to evaluate the matrix effect and recovery. The matrix effect should be within 85%–115%.

Stability

Stability was evaluated under room temperature for 6 h, in autosampler (10°C) for 7 h, under 4°C for 24 h, after triplicate complete freeze-thaw processes, and –80°C for 1 month at LQC and HQC concentrations. The stability should be below 15%.

Drug–drug interaction

Male Sprague Dawley rats (285 ± 22 g) were obtained from the Laboratory Animal Center. Rats were housed at a chamber with a constant temperature of 25°C and humidity of 40%–60% and a 12-h light–dark cycle for 7 days to minimize their suffering. The study was approved by the Animal Protection and Use Committee (ethic number: wyyy-aec-2023-008; date: 2 August 2023).

We designed drug–drug interaction studies between lorlatinib and triazole antifungals with the use of a single dose referred to published researches [15–17]. Lorlatinib, ketoconazole, posaconazole, voriconazole, and itraconazole were prepared with the solution of sodium carboxymethyl cellulose (0.5% CMC-Na solution). Rats were kept in fasting condition for 12-h overnight and randomly divided into 5 groups (n = 6 each): group A (CMC-Na), group B (ketoconazole), group C (posaconazole), group D (voriconazole), and group E (itraconazole). Each rat received the corresponding triazole antifungal with dose of 20 mg/kg or 0.5% CMC-Na with intragastric administration. 30 min later, all rats received lorlatinib (10 mg/kg) by gavage. At 0.167, 0.333, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 9, 12, 24, and 48 h after lorlatinib was given, 300 μl of blood was collected from the tail vein and centrifugated at 13 000 rpm, 4°C for 10 min, and 90 μl of plasma was obtained.

Data analysis

The pharmacokinetic parameters of lorlatinib were calculated by DAS 3.0 (Bontz Inc., Beijing, China) with non-compartmental analysis. The drug concentration–time curve was constructed using Prism 7.0 (GraphPad Software Inc., San Diego, USA). The comparisons of the pharmacokinetic parameters between group A and other groups were performed by using SPSS software 23.0 (IBM Corp., Chicago, USA). The analysis of multiple variance comparisons followed by LSD tests (when the group data were homogeneous) or Kruskal–Wallis rank sum test (when variance between groups was not homogeneous) was performed. P < 0.05 expressed statistical significance.

Results and discussion

Assay establishment and optimization

The specificity of the analyte was maximized by optimizing the mass spectrometry conditions. The response of lorlatinib in positive ion mode was more favorable than negative ion mode with a better signal-to-noise ratio. Therefore, positive ion ESI was selected to monitor lorlatinib. More than three transitions of precursor-product ion were compared and those with the highest intensity were used as diagnostic. According to the charge state of precursor and product ions, the optimum ion pairs for lorlatinib and IS were m/z 407.3–228.07 and m/z 450.3–260.0, respectively.

Protein precipitation was used for sample preparation because of its ease of operation and ability to use for high-throughput sample processing, and we found lorlatinib expressed higher signal intensity with using methanol as precipitation agent by comparing with acetonitrile. Various proportions of acetonitrile, methanol, and water with different pH values were tested to optimize the chromatographic separation and peak shapes of lorlatinib and IS, and we found that the mobile phase composed of 0.1% formic acid solution and methanol produced the optimal peak shape with appropriate retention time.

Method validation

Specificity

Fig. 2 shows chromatograms of blank rat plasma (A), blank plasma added with lorlatinib and IS (B), and drug–drug interaction samples (C), respectively. Lorlatinib and IS had retention durations of 1.91 and 1.82 min, respectively, and no endogenous interference was identified during the experiment. Furthermore, the parent and fragment ions of lorlatinib and four triazole antifungals used in this study are significantly different (Supplementary Table 1), so no chromatographic interactions between lorlatinib and the antifungals would exit. These results suggest that the method has good specificity for determining lorlatinib in rat plasma.

Standard curve

The standard curve was determined by linear regression between the peak area ratio of lorlatinib versus IS and the concentration of lorlatinib (weighted 1/square of lorlatinib concentration). The standard curve displayed excellent linearity with the concentration of 1–5000 ng/ml, and coefficient R² for the determined drug was not less than 0.995, with the regression equation of Y = 5.9864X + 4.4485. The LLOQ was 1 ng/ml, and chromatographic signal of analyte lorlatinib from LLOQ was more than 10-fold that of the blank plasma noise in the same time.

Accuracy and precision

The accuracies and precisions of the method were assessed by measuring the four levels of QC samples at three independent days and three batches of QC samples on same day. The intra-day and inter-day accuracies were between −4.60% and 11.81%, the RSD of precisions were less than 7.19% (Table 1). These findings suggest our method has good accuracy and precision for the assay of lorlatinib.

Matrix effect and recovery

The matrix effect range of lorlatinib was 88.62%–102.9%, and the recovery rate of lorlatinib was between 90.11 and 96.22% (Table 2), suggesting that it has no significant matrix effects on lorlatinib analysis in rat plasma and extraction recovery of lorlatinib in this method complies with the FDA guidance.

Stability

The result (Table 3) indicated lorlatinib was stable at room temperature for no less than 6 h, in an autosampler (10°C) for 7 h, at 4°C for 24 h, triplicate complete freeze-thaw (from −80°C to ambient temperature), and at −80°C for 1 month.

Drug–drug interaction

Based on the developed analytical method, we measured plasma lorlatinib levels under the combined action of various antifungals. We evaluated the impacts of the four triazole antifungals on lorlatinib metabolism. The average concentration–time curves of five groups are displayed in Fig. 3 and pharmacokinetic parameters are listed in Table 4. When lorlatinib is combined with posaconazole or itraconazole, its pharmacokinetic parameters and curves showed no significant differences with lorlatinib treatment alone. By contrast, when lorlatinib was combined with ketoconazole, its pharmacokinetic parameters and pharmacokinetic curve showed a statistically significant increase compared with lorlatinib treatment alone, AUC_0-∞ increased by 49.0%, clearance decreased by 40.0%. When lorlatinib was combined with voriconazole, the AUC_0-∞ of lorlatinib increased by 104.3%, T_max prolonged 1.2 h, and clearance decreased by 40.0 % (P < 0.05). The results express ketoconazole and voriconazole inhibit lorlatinib metabolism in vivo. Either ketoconazole or voriconazole is a potent inhibitor of CYP3A4 metabolizing enzymes. When they were combined with lorlatinib, the lorlatinib metabolism slows down, and exposure levels elevated, which may lead to toxicity and adverse reactions. Therefore, the combination of lorlatinib with ketoconazole or voriconazole may pose an unexpected risk of poor outcomes.

Although ketoconazole, itraconazole, posaconazole, and voriconazole were CYP3A inhibitors, their effects on lorlatinib metabolism were different, which might be due to their different bioavailability. Actually, the bioavailability of posaconazole and itraconazole are lower than that of ketoconazole and voriconazole [18]. In the case of single-dose administration, the concentration levels of posaconazole and itraconazole in vivo might not be enough to affect the metabolism of lorlatinib, resulting in lack of interaction with lorlatinib. Therefore, posaconazole and itraconazole failed to interact effectively with lorlatinib in a single dose mode.

However, there are still some limitations in the design of this study, such as the lack of detection and analysis of M8, a metabolite of lorlatinib; drug–drug interaction studies between lorlatinib and triazole antifungals were only designed with using of single dose, without evaluating the effects of multiple doses of triazole antifungals on lorlatinib metabolism [19]. More researches needs to be further performed and analyzed.

Conclusion

Based on UPLC–MS/MS, we established an efficient and accurate bioassay to measure plasma lorlatinib in rats. This method was verified following FDA standards and applied to estimate the interactions between lorlatinib and four triazole antifungals in rats. Ketoconazole and voriconazole significantly inhibited lorlatinib metabolism, while itraconazole and posaconazole had no effects on lorlatinib pharmacokinetics. These pre-clinical results will be helpful for clinical use of these four triazole antifungals with lorlatinib. Future clinical study is required to verify the metabolic interaction of lorlatinib with ketoconazole or voriconazole.

Author contributions

Z.Z. and X.Z. designed the study and revised the manuscript; Z.Y., C.W., R.L., C.C., J.Y., and Y.C. performed the experiments; J.F., T.Z., M.J., and A.H. analyzed the experimental data; Z.Y. wrote the manuscript.

Conflict of interest

None declared.

Funding

This work was supported by Zhejiang Provincial Natural Science Foundation of China under grant no. LYY19H310007 and Medical Health Science and Technology Project of Zhejiang Provincial Health Commission under grant no. 2019KY448 and Project of Wenzhou Science and Technology Bureau under grant nos. Y2020184 and Y20210221.

Data availability

The data will be shared on reasonable request to the corresponding author.

References

Author notes