Therapeutic Drug Monitoring of Beta-Lactams in Cystic Fibrosis: Unattained Target in Standard Antibiotic Dosing

Abstract:

Acute exacerbations of cystic fibrosis (CF) diminish quality of life and, if inadequately treated, can be life-threatening. The pathophysiological alterations associated with CF result in modified antibiotic pharmacokinetics. Moreover, the viscous mucus in the lungs limits pathogen exposure to drugs, rendering successful antibiotic treatment challenging. A 23-year-old female patient with CF was repeatedly admitted for intravenous antibiotic therapy for acute exacerbation of Pseudomonas aeruginosa infection. In the context of altered pharmacokinetics in CF, therapeutic drug monitoring of meropenem and piperacillin revealed consistently low plasma levels of both drugs. Targeted plasma levels were ultimately achieved through continuous high-dose infusions, based on therapeutic drug monitoring and subsequent dose adjustments.

CLINICIAN

A 23-year-old woman was diagnosed with cystic fibrosis (CF); the pathogenic variants Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) dele2,3 (21 kb)/R553X were detected. The patient exhibited no impaired pulmonary function or bronchiectasis until the age of 20 years and was chronically infected with Staphylococcus aureus. In May 2021, the patient experienced severe exacerbation with a predicted FEV1 decline of 38%. Subsequently, her condition deteriorated and pulmonary function rapidly declined. The patient experienced frequent exacerbations and developed a chronic mucoid Pseudomonas aeruginosa (PSAE) infection in November 2022, necessitating frequent hospitalizations. The standard treatment regimen consisted of a combination of meropenem and aminoglycosides; however, the clinical improvement was unsatisfactory, and her pulmonary function remained low. In October 2023, she experienced another PSAE exacerbation (FEV1 1.27 L, ie, 45% predicted). Acute infections were treated early with a combination of 2 classes of intravenous antibiotics (for PSAE infections, typically beta-lactams and aminoglycosides). This combined approach reduces the potential for promoting bacterial resistance, whereas different mechanisms of action enable synergy against the pathogen.^1,2

TDM CONSULTANT

Beta-lactams with anti-pseudomonal activity include meropenem and piperacillin/tazobactam, which are time-dependent antibiotics. The best results were achieved with 100% fT >4–8× minimum inhibitory concentration (MIC).^3–5 The target concentration should be attained at the site of infection. In healthy subjects, meropenem levels in the lungs are approximately half the plasma levels, whereas piperacillin levels are approximately one-third of the plasma levels.⁶ In this case, the target pathogen was PSAE (mucoid phenotype); however, the MIC was not determined. According to The European Committee on Antimicrobial Susceptibility Testing (EUCAST) data, PSAE is susceptible to meropenem when the MIC is <8 mg/L and to piperacillin/tazobactam when the MIC is <16 mg/L.⁷ Owing to CF-specific metabolism and the elimination of antibiotics, achieving the target of 100% fT >4–8× MIC is challenging. Based on general susceptibility breakpoints for meropenem in PSAE (well-susceptible: MIC ≤2 mg/L),⁷ and the aim of 100% fT >4–8× MIC, the target meropenem plasma concentrations would be at least 8.0–16.0 mg/L for intermittent dosing.⁸

The patient was administered meropenem and gentamicin. The therapy was initiated with meropenem 2000 mg every 8 hours (total body weight, 47 kg; relative eGFR of creatinine 101 mL/min/1.73 m²). For meropenem therapy, Meropenem Kabi (Fresenius Kabi, Bad Homburg, Germany) was used. Based on the manufacturer's labelling, the physical and chemical stability is reported to be 6 hours when diluted with 0.9% saline solution and stored at room temperature. Based on this, the 2-g infusions of meropenem were changed every 6 hours.

Initially, meropenem plasma concentration was below the lower limit of quantification (1.1 mg/L). Meropenem plasma concentration was determined using high-performance liquid chromatography (HPLC) as a validated method at the Department of Clinical Biochemistry, University Hospital Brno (kit Antibiotics in Serum/Plasma—HPLC; Chromsystem, Munich, Germany); analyzer: HPLC Agilent 1100, Agilent Technologies (Waldbronn, Germany).

The dose of meropenem was subsequently adjusted to 2000 mg every 6 hours, with each dose infused over 2 hours. After the dose adjustment, the measured meropenem concentration remained low (3.4 mg/L); consequently, the meropenem regimen was changed to 8000 mg/d by continuous infusion. At this dose, meropenem levels were finally 29.4 mg/L. After 2 days of therapy with the same dosing regimen, we measured the value of 28.7 mg/L using the same method. The gentamicin plasma levels were monitored based on a local protocol, necessitating a dose-dependent increase after the initial measurements. The initial dose of gentamicin (6.8 mg/kg) resulted in a trough concentration below 0.3 mg/L and a peak concentration of 11.5 mg/L. The peak concentration was considered insufficient (15–20 mg/L), and the dose was increased to 7.6 mg/kg. The subsequent trough concentration remained unchanged, whereas the peak concentration increased to 14.4 mg/L.⁹

These concentrations meet the target and can be considered safe because the threshold for meropenem neurotoxicity is higher than 64 mg/L and nephrotoxicity is higher than 44 mg/L, even when considered for intermittent treatment.¹⁰ All measured levels of meropenem were compared with the levels predicted by the pharmacokinetic software (MWPharm Online, version 1.7.8.4.; Mediware, Prague, Czech Republic); refer to Figure 1. Meropenem (and piperacillin) plasma levels were measured at steady-state (at least 48 hours after therapy initiation or after dose change) using High-Performance Liquid Chromatography-Ultraviolet (HPLC/UV).

FIGURE 1.

Prediction of meropenem and piperacillin plasma levels (mg/L) and comparison with actual measured levels (MWPharm Online, Mediware a.s., version 1.7.8.4.).

CLINICIAN

After 14 days of continuous meropenem infusion, sputum quantity and quality improved and shortness of breath disappeared. FEV1 measurements indicated an increase from 540 mL to 64% of the predicted value. No adverse effects were observed, although a minor decrease in renal function (relative eGFR of creatinine 71 mL/min/1.73 m²) was noted, which subsequently normalized at the follow-up ambulatory examination.

Unfortunately, several weeks later, the patient was admitted because of another PSAE exacerbation. The treatment consisted of amikacin and piperacillin/tazobactam; the initial dosage was 4.5 g of piperacillin/tazobactam administered every 6 hours (for a total body weight of 53 kg, relative eGFR of creatinine 99 mL/min/1.73 m²).

TDM CONSULTANT

For piperacillin, the target plasma level was considered 32.0–64.0 mg/L, based on EUCAST MIC susceptibilities and toxicity evidence.^7,10 According to new microbiological screening, PSAE exhibited intermediate susceptibility to meropenem, which could be attributed to insufficient meropenem exposure during the previous antibiotic course.

The first trough-plasma level of piperacillin 2.0 mg/L was significantly lower than the target. The daily dose of piperacillin/tazobactam remained unchanged (18 g/d); however, the administration method was changed to continuous infusion. Subsequent measurements revealed a piperacillin level of 6.0 mg/L, which was below the target. Consequently, the daily dose was increased to 27 g/d with continuous infusion, and the therapy was completed with a determined plasma level of 58.0 mg/L. Minimal and maximal amikacin levels were regularly monitored to achieve the desired targets. The initial dose of amikacin was set to 30 mg/kg, with a concentration below 2 mg/L and a peak concentration of 41.7 mg/L. These concentrations were within the desired target range (35–50 mg/L) and the dose was maintained throughout the course of treatment.⁹

CLINICIAN

Sputum volume decreased, and shortness of breath and chest pain resolved. Her pulmonary function improved (FEV1 by 10%, to 1.56 L, ie, 55% pred.). No adverse effects were observed (specifically, the potential synergistic nephrotoxicity of amikacin and piperacillin/tazobactam was not detected; the relative eGFR of creatinine remained at 89 mL/min/1.73 m² or piperacillin/tazobactam concentration-dependent neurotoxicity, threshold 157 mg/L for continuous administration).¹¹

CONCLUSION

As demonstrated in this case of a patient with CF chronically infected with PSAE, with preserved renal function and normal body mass index, standard (or on-label) doses of antibiotics may be insufficient for the treatment of acute infectious exacerbations. One of the primary reasons for this observation is a well-established phenomenon. The pharmacokinetics of antibiotics are altered in patients with CF compared with healthy populations. Notably, the factors contributing to this alteration include a larger volume of distribution, decreased plasma protein binding owing to compromised nutritional status (lower albumin levels), altered number and activity of enzymes/transporters, and increased renal elimination. Furthermore, a thicker mucus layer in the gastrointestinal tract, fat malabsorption, reduced intestinal bicarbonate secretion, and a slower gastrointestinal passage can affect the absorption of orally administered drugs. Specifically, a higher volume of distribution and faster clearance (both renal and nonrenal) have been observed.^12,13 Moreover, Katz et al¹⁴ concluded that therapeutic drug monitoring of beta-lactams among patients with CF decreased exacerbations per year and prolonged the exacerbation-free interval.

As a result of various factors associated with the primary diagnosis, individual patient characteristics, and the pathogen, as well as other antibiotics (eg, aminoglycosides or vancomycin), therapeutic drug monitoring (TDM) of beta-lactams in patients with CF should be considered for each intravenous antibiotic course. TDM-guided antibiotic therapy (intermittent or continuous) enhances the attainment of the pharmacokinetic and pharmacodynamic targets. This enables effective treatment with higher antibiotic doses while minimizing the risk of adverse effects. Owing to the differences in antibiotic pharmacokinetics in patients with CF, population-specific pharmacokinetic models are crucial for the utility of beta-lactam TDM in clinical practice because the models for the general population do not accurately represent the pharmacokinetics of patients with CF (Fig. 1).

The main limitation of this study is that the MIC was not determined for our patient. Nowadays, MIC assessment is not a standard test in our hospital. We acknowledge that determining antibiotic levels without an individual MIC value may be biased, and we are currently in discussions with the Society of Microbiology regarding routine MIC determination in pharmacokinetically challenging patients. Second, the sputum concentration of antibiotics was calculated based on the plasma concentration and general penetration. Because knowledge of beta-lactam lung penetration for CF patients with CF is insufficient,¹⁵ it would be a valuable part of future studies (but is not feasible in our hospital settings).