Microdosing as a Potential Tool to Enhance Clinical Development of Novel Antibiotics: A Tissue and Plasma PK Feasibility Study with Ciprofloxacin

Ciprofloxacin was obtained via the hospital pharmacy from Bayer Austria (400 mg/200 mL infusion solution). [¹⁴C]Ciprofloxacin (labeled in the 2-position of the quinoline ring, molar activity: 2.2 GBq/mmol, radiochemical purity: 98.2%) was obtained from Hartmann Analytic GmbH (Braunschweig, Germany). For the i.v. injection, [¹⁴C]ciprofloxacin (7 kBq, 1.1 µg) was dissolved in 10 mL of physiological saline solution and filtered through a sterile Millex-GS filter (0.22 µm; Millipore Corporation, Bedford, MA, USA).

No formal dosimetry calculation was performed for [¹⁴C]ciprofloxacin. Because of the very high sensitivity of AMS, the administration of only very low ¹⁴C amounts is required (7 kBq in the current study). In addition, as the energy of the β⁻ particles emitted in the decay of ¹⁴C is relatively low (0.156476 MeV), AMS microdosing studies usually fall within risk category I (trivial risk, total effective dose < 0.1 mSv) according to the International Commission on Radiological Protection Publication 62 “Radiological Protection in Biomedical Research” [30]. However, an intravenously injected ¹⁸F-labeled radiotracer for PET (400 MBq) typically gives a total effective dose in the range of 5–10 mSv, corresponding to risk category IIb (minor to intermediate risk) in the International Commission on Radiological Protection publication 62 [30].

Ciprofloxacin concentrations were determined in subcutaneous tissue using microdialysis. This method allows quantification of the unbound fraction of a substance in interstitial space fluid by insertion of a microdialysis catheter into the tissue of interest. The microdialysis catheter is equipped with a semipermeable membrane at the tip and is constantly perfused at a predefined flow rate with physiologic solution [8, 9]. CMA 66 catheters (M Dialysis AB, Solna, Sweden) having a membrane with a molecular weight cut-off of 20 kDa were used in this study and perfused at a flow rate of 1.0 µL/min. The ciprofloxacin concentration that penetrated the semipermeable membrane via passive diffusion from the tissue into the dialysate was determined (C_dialysate). Because of a lack of equilibrium between extracellular tissue fluid and the perfusion medium, a probe calibration was performed after the PK sampling period using the retrodialysis method. This method assumes that the exchange process across the semipermeable membrane of the microdialysis probe is equal in both directions. After at least 30 minutes of equilibration, the catheter was perfused with a known concentration of either unlabeled ciprofloxacin (5 µg/mL) or [¹⁴C]ciprofloxacin (10.5 pg/mL) for 1 hour at a flow rate of 1.0 µL/min and the concentration of ciprofloxacin in the dialysate was determined. In vivo recovery (%) was calculated as 100 − (100 . ciprofloxacin concentration_dialysate /ciprofloxacin concentration_perfusate). The concentration in the interstitial tissue was then determined as C_dialysate/in vivo recovery.

Bronchoalveolar lavage relies on the concept that by rinsing the bronchoalveolar tree with saline, ELF can be collected [12]. Epithelial lining fluid samples were taken using BAL at 2 h, 4 h, or 8 h after [¹⁴C]ciprofloxacin administration. Because of the invasiveness of the procedure, each volunteer underwent only one BAL procedure, and timepoints were randomly assigned using envelopes to the subjects in order to obtain three samples per time point. Sedation was performed using propofol and remifentanil while vital parameters were monitored constantly. The airway was secured with a laryngeal mask and anesthesia was maintained with 1 MAC sevoflurane. After administering xylocaine^® 2% gel at the rima glottides, BAL was performed by instilling three 20-mL aliquots of saline solution.

As each BAL sample consisted of ELF in saline, the dilution of ELF was determined to convert ciprofloxacin concentrations in BAL fluid to ciprofloxacin concentrations in ELF. The urea dilution method was used to assess ELF dilution [31]. Urea freely diffuses through several body compartments, including ELF, as supported by experiments in the isolated perfused dog lung [32]. If the concentrations of urea in plasma (C_urea,plasma) and BAL fluid (C_urea,BAL) are known, concentrations of ciprofloxacin in BAL samples (C_{ciprofloxacin,BAL}) can be corrected for dilution of ELF by saline to obtain ciprofloxacin concentrations in ELF (C_{ciprofloxacin,ELF}) according to the following equation:

C_{ciprofloxacin,ELF} = C_{ciprofloxacin,BAL}/(C_urea,BAL/C_urea,plasma)

To assess plasma concentrations of urea, one plasma sample was drawn near to the timepoint of BAL. The concentration of urea in both plasma and BAL fluid was determined using a specific urea assay kit (Abnova KA1652) according to the manufacturer’s protocol. Urea reactions were performed in 96-well plates and concentrations were assessed using an EnSpire^® 2300 Multimode Plate Reader (PerkinElmer, Waltham, MA, USA) at an absorbance of 520 nm. All standards and samples were run in duplicate and the mean values were used for analysis. The limit of quantification for plasma and BAL was 0.8 µg/mL.

Healthy subjects underwent a screening visit, one study day, and a final visit. The 18 healthy subjects were randomly assigned using envelopes to one of two cohorts, each consisting of nine subjects. Both cohorts received a microdose of [¹⁴C]ciprofloxacin (7 kBq, 1.1 µg, dissolved in 10 mL of physiological saline solution) as an i.v. bolus over approximately 1 min. In cohort A, the [¹⁴C]ciprofloxacin microdose administration was preceded by a single i.v. infusion of a therapeutic dose of unlabeled ciprofloxacin (400 mg) over 60 min, while cohort B only received the [¹⁴C]ciprofloxacin microdose. On the study day, one microdialysis catheter was inserted into the subcutaneous tissue of the upper thigh, followed by the study drug administration. The i.v. bolus of the microdose of [¹⁴C]ciprofloxacin was administered immediately after the end of the infusion of the therapeutic dose of unlabeled ciprofloxacin. Blood sampling was subsequently performed at 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 7, 8, and 10 hours after administration using lithium heparin tubes (Vacuette^® FX; Greiner Bio-One, Kremsmünster, Austria). Samples were centrifuged (2000g, 4 °C, 10 min) within 1 hour after collection, snap frozen, and stored at −80 °C for determination of ciprofloxacin concentrations using LC–MS/MS. Microdialysates were sampled three times over 1 hour, respectively (from 1.5 to 2.5 h, from 3.5 to 4.5 h, and from 7.5 to 8.5 h after [¹⁴C]ciprofloxacin administration). Retrodialysis was performed after at least 30 min of equilibration and retrodialysis samples were collected from 0 to 0.5 h and from 0.5 to 1 h. Microdialysis samples were snap frozen and stored at – 80 °C for determination of ciprofloxacin concentrations using LC-MS/MS and AMS. Bronchoalveolar lavage was performed in each subject at one of three timepoints after [¹⁴C]ciprofloxacin administration (2, 4, or 8 h) as described above. The aliquots of BAL fluid were pooled, centrifuged (2000g, 4 °C, 10 min) and the supernatant was snap frozen and stored at −80 °C for determination of ciprofloxacin concentrations using LC–MS/MS and AMS. At the time of BAL, one blood sample was collected, which was used for determination of urea concentrations (see above) and for determination of ciprofloxacin concentrations in plasma using AMS.

Total ¹⁴C concentrations measured with AMS were assumed to equal the concentrations of [¹⁴C]ciprofloxacin. [¹⁴C]Ciprofloxacin concentrations in plasma and ELF and therapeutic-dose ciprofloxacin concentrations in ELF were only measured at one timepoint per subject (i.e., at 2, 4, or 8 h after administration) and concentrations from three subjects per timepoint were averaged to generate concentration–time profiles. Concentrations of [¹⁴C]ciprofloxacin (mBq/mL) were converted into mass concentrations (pg/mL) via the molar activity of [¹⁴C]ciprofloxacin (2.2 GBq/mmol). To enable comparison of therapeutic-dose and microdose ciprofloxacin concentrations in plasma, subcutaneous tissue, and ELF, values were normalized to the administered dose of ciprofloxacin (in µg) and expressed as pg equivalents per mL. To assess correlations between dose-adjusted concentrations of microdose and therapeutic-dose ciprofloxacin, Pearson’s coefficient of correlation (r) was calculated.

Pharmacokinetic analysis was performed using the Kinetica 2000 software package, version 3.0 (InnaPhase, Philadelphia, PA, USA). The following parameters were calculated for plasma, subcutaneous tissue, and ELF: maximum concentration, time to maximum concentration, area under the concentration–time curve (AUC) from time 0 to 10 h for plasma, and from 0 to 8 h for subcutaneous tissue and ELF. In addition, for plasma, the elimination half-life, apparent total body clearance from plasma, and apparent volume of distribution at steady state were calculated. To quantify tissue penetration of ciprofloxacin, the subcutaneous tissue-to-plasma (AUC_ST/AUC_plasma) and ELF-to-plasma (AUC_ELF/AUC_plasma) AUC ratios were calculated. Pharmacokinetic parameters are given as geometric mean and 95% confidence interval or median and range.