Daptomycin (CUBICIN^®) belongs to a new class of cyclic lipopeptide antibiotics (Heidary et al., 2018). It is approved for the treatment of complicated skin and skin structure infections (cSSSI), Staphylococcus aureus bloodstream infections (S. aureus bacteremia) in adults and pediatric patients 1–17 years of age, and right-sided infective endocarditis in adults (FDA, 2020).

Although the use of vancomycin is associated with a high failure rate in the treatment of methicillin-resistant S. aureus (MRSA) blood infections (Kalil et al., 2014; van et al., 2012; Shariati et al., 2020), it is still the mainstream clinical treatment method. In contrast, daptomycin treatment is associated with lower clinical failure and 30-days mortality rates than vancomycin treatment as reported in MRSA blood-infection studies (Claeys et al., 2016). In addition, daptomycin is an effective option for treating vancomycin-resistant enterococci (Melese et al., 2020; Shi et al., 2020). Multiple guidelines also recommend daptomycin for the treatment of S. aureus bacteremia or cSSSI (Stevens et al., 2014; Habib et al., 2015; Baddour et al., 2015; Sartelli et al., 2018).

The recommended daptomycin dosing regimens in adults are 4 mg/kg q24 h (once every 24 h) for the treatment of cSSSI and 6 mg/kg q24 h for the treatment of S. aureus bacteremia by infusion over 30 min or injection over 2 min. The dosing interval should be prolonged to q48 h (once every 48 h) in adult patients when the creatinine clearance rate (CL_CR) is <30 ml/min. However, the dosing regimens for pediatric patients are more complex because daptomycin clearance is higher in younger than in older children (Principi et al., 2015). Hence, daptomycin dosing varies among different age groups. However, daptomycin dosage adjustments for pediatric patients with renal impairment have not been established (FDA, 2020).

The pharmacokinetics (PK) of daptomycin is generally linear, and a steady state is reached after 3 days of intravenous, once-daily administration (Hair and Keam, 2007). Daptomycin is primarily bound to serum albumin with a concentration-independent protein binding rate of 90%–93%. The volume of distribution at steady-state (V_ss) of daptomycin in healthy adult subjects is dose-independent and approximately 0.1 L/kg. The mean plasma half-life of daptomycin is approximately 8–9 h. It tends to be prolonged by deteriorated renal function, for even up to 28 h, because renal excretion is the primary elimination route. Approximately 5.7% of the dose is found in feces, but according to in vitro studies, daptomycin is not metabolized by human liver microsomes (FDA, 2020). Therefore, significant drug-drug interactions between daptomycin and drugs metabolized by these systems are not expected. In addition, the antibiotic can be injected within approximately 2 min, making it a simple and safe choice for adult outpatients.

Based on PK studies of daptomycin in adult subjects with various degrees of renal function (Yabuno et al., 2013; Chaves et al., 2014; Kullar et al., 2014; Xu et al., 2017; Xie et al., 2020), dose adjustments are recommended in adult patients whose CL_CR is less than 30 ml/min, whereas no dosage adjustment is necessary for adults with mild-to-moderate renal impairment. However, there are no data for pediatric patients with renal impairment. To complement the lack of research in this area, our research team used physiologically based pharmacokinetic software to combine the drug parameters of daptomycin with the physiological parameters of children with renal impairment. Based on several studies conducted by our research team and others on PBPK modeling for successfully predicting drug concentrations and drug-drug and drug-disease interactions in special populations (Zhang et al., 2019; Chen et al., 2020; Ye et al., 2020; You et al., 2020; Li et al., 2021; Liu et al., 2021), we established a new PBPK model of daptomycin for simulating the daptomycin PK in pediatric patients with renal impairment and evaluating the drug’s pharmacodynamics (PD).

The primary objective of this study was to predict the PK of daptomycin in pediatric patients with renal impairment by developing PBPK models. The resulting PK simulations address a lack of reports on the use of daptomycin in those populations owing to the difficulties in conducting clinical daptomycin PK studies in children with renal impairment for various reasons, including ethical restrictions.

We established and optimized the basic PBPK model of daptomycin in healthy populations. Our modeling approach generated excellent simulation results for both healthy adults and healthy children based on the fold error values of less than 1.24. During the modeling process, renal and non-renal elimination pathways were incorporated into the daptomycin model. Because the non-renal elimination pathway has not been clearly established, we assumed that it occurred via biliary excretion, which we considered as liver clearance. Renal daptomycin clearance is commonly described as a filtration process owing to a lack of data showing the contribution of transporters to renal excretion in humans, and the non-renal clearance was estimated by subtracting the renal clearance from the total clearance. This model optimization method is similar to the published vancomycin PBPK modeling approach (Emoto et al., 2018). Studies also report that daptomycin is subjected to bile excretion, which shows that our hypothesis is reasonable (Tascini et al., 2011).

Owing to the limited pharmacokinetic studies of daptomycin in children, we only obtained specific data points measured at a dose of 4 mg/kg (Abdel-Rahman et al., 2008). To circumvent this lack of published data, we incorporated various recommended dosing regimens for different age groups from the FDA label into the model and compared the predicted values with the FDA data to further verify the reliability of the model (FDA, 2020). Remarkably, our prediction results met the 2-fold error threshold criterion with a mean error fold of 1.09. The daptomycin clearance, V_ss, and half-life (T_half) in our model are similar to those reported in the literature (Supplementary Table S3). In the group comprising the youngest children, the clearance rate was higher and the T_half was shorter than in the groups with older children. The trend for these two parameters was consistent with the results reported by other research groups (FDA, 2020; Abdel-Rahman et al., 2008; Durand et al., 2014). In summary, our model was validated by comparing various pharmacokinetic parameters.

Remarkably, although the GFR decreased in our model with the decline of renal function, the biliary excretion did not change, which is also consistent with the reported assessment by Nolin et al. (2008). In contrast, some reports indicatethat bile excretion changes when renal function declines (Nolin et al., 2008; Schijvens et al., 2020). The controversy on this aspect may require further research. We hypothesize that this may be related to the specificity of the drug and the expression of the transporter involved in the excretion pathway. Therefore, the biliary excretion in our model remained unchanged to obtain predicted values that were closer to the measured data. The fup value changed only slightly, with 8% in healthy condition, 9% in mild renal impairment, and 10% in moderate renal impairment to end-stage renal disease, which is the same as previously reported (Gregoire et al., 2019). This change made not only the simulated total plasma concentration closer to the measured value, but also made the predicted free concentration, which is related to fup, closer to the measured value. However, in the PBPK model of adults with renal impairment, we corrected the fup to 9% in healthy status to make the prediction result closer to the measured value because the included population had an average median age of approximately 67 years, and the serum albumin is generally low in the older population (Gom et al., 2007). We made assumptions on the Blood-to-plasma ratio (RBP) of daptomycin based on the following information: Daptomycin is a highly protein-bound antibacterial drug, mainly bound to serum albumin, so its drug concentration in plasma is much higher than that in whole blood. Furthermore, because of its low volume of distribution, of only 0.1 L/kg and about 7 L in adults, it is mainly distributed in the plasma, and to a lower extent in the extracellular water space fluid. Given its extremely hydrophilic nature, it is very unlikely that daptomycin has the ability to penetrate human cells in significant amounts (Thallinger et al., 2008). Therefore it has a low potential to penetrate blood cells. In contrast to daptomycin, several hundred liters are the distribution volumes of the macrolides class of antibiotics such as azithromycinfor which the RBP was estimated to be 1 (Johnson et al., 2016; Guimarães et al., 2021). On the other hand, it is possible that we used 1.2 as optimized logP value, which might be quite a bit higher than experimental, resulting in the need to use a lower RBP to satisfy the daptomycin distribution. Further study is required on this front.

In this study, we performed Monte Carlo simulations using the combination of Crystal Ball software and population simulation from GastroPlus^®. This method combines human parameters in a physiological-based population simulation with mathematical-statistical methods to yield results compatible with the real world. The optimized dosing recommendations (Table 6) for daptomycin in pediatric patients with renal impairment could potentially achieve the same AUC values as those in healthy children. For each dosage regimen, 10,000 random simulations were performed to obtain the PTA and CFR predictions (Figure 5 and Table 5). Based on the CFR results, the recommended dose for S. aureus bacteremia on the FDA label and the adjusted dosage derived from our PBPK model can reach a CFR of more than 90% for MRSA or E. faecium. A comparison with the published Monte Carlo simulation data of different daptomycin doses in children indicated that our PTA and CFR data for healthy children were similar (Wei et al., 2020). The similarity of these data also increases the credibility of our PBPK model.

Elevation of creatinine phosphokinase (CPK) levels is an adverse reaction of daptomycin, and the severe form can develop into rhabdomyolysis (FDA, 2020). The increase in CPK is associated with high-dose daptomycin treatment and a declining renal function (Tai et al., 2018). Additionally, the plasma trough concentration (C_min) of daptomycin ≥24.3 mg/L is reportedly associated with elevated CPK levels (Bhavnani et al., 2010). This adverse reaction should be especially considered for children with renal impairment. Therefore, we simulated the C_min at steady state in children with different renal functions under the adjusted daptomycin dose and evaluated the safety of the adjustment to a certain extent (Figure 5). The administration instructions indicate that a healthy population can reach a steady state after the third dose, but a decline in renal function prolongs the time for the drug concentration to reach a steady state correspondingly. In children aged 7–17 years with an onset of severe renal impairment, the steady state was attained after the fourth dose (q48 h), and in end-stage renal disease, the fifth dose (q48 h) was administered before reaching a steady state. Four daptomycin doses (q48 h) were needed to reach the steady state in children aged 1–6 years with severe renal impairment or end-stage renal disease. Clearance was higher in younger children than in older children, decreasing from around 20 ml/h/kg to around 11 ml/h/kg as age increased (Supplementary Table S3). Even if a higher dose was used in the younger age group, the steady-state trough concentration was far less than 24.3 mg/L. Furthermore, in children aged 7–11 and 12–17 years with end-stage renal disease, the daptomycin steady-state trough concentration was slightly higher than 24.3 mg/L, indicating that when renal function deteriorated, even if the dosage is adjusted, the CPK level should be closely monitored more than once a weekto prevent the occurrence of rhabdomyolysis (Kullar et al., 2014).

This study had several limitations. Firstly, there were no data sets available for daptomycin in pediatric patients with renal impairment to validate the model. Thus, for model improvement, it will be necessary to accumulate new clinical data. Secondly, daptomycin is not recommended in pediatric patients younger than 1 year of age because of the risk of potential adverse effects on the muscular, neuromuscular, and nervous systems observed in neonatal dogs (FDA, 2020). Although some scholars have conducted related studies on this population, we currently have no plans to establish a PBPK model for children less than 1 year old. In the future, we will use new research results as they become available for supplementing our model. Lastly, the MIC distributions of the pathogen isolates were obtained from EUCAST, which represents Europe but no other areas. However, compared with related PD studies (Falcone et al., 2013; Soraluce et al., 2018; Urakami et al., 2019; Wei et al., 2020; Yamada et al., 2020), the data did not differ substantially, and an impact on the results is not expected. Thus, our study can fill the gap in clinical data to a certain extent.

In summary, we established, validated, and optimized PBPK models of daptomycin for healthy adults, healthy children, and adults with impaired renal function, which were used to derive and construct PBPK models for children with renal impairment. We used PBPK modelingto evaluate the pharmacokinetic changes of daptomycin in this special population and create adjusted dosing regimens. After evaluating the PD and adverse reactions, we obtained the recommended daptomycin doses for children with renal impairment. In pediatric patients with mild-to-moderate renal impairment, no dosage adjustments were needed. In pediatric patients with severe renal impairment, the dosing interval should be extended (from q24 h to q48 h). In patients with end-stage renal disease combined with cSSSI, the daptomycin dose of 9 mg/kg q24 h should be adjusted to 7 mg/kg q48 h in children aged 2–6 years, and 10 mg/kg q24 h should be adjusted to 6 mg/kg q48 h in children aged 1–2 years. For S. aureus bacteremia, 12 mg/kg q24 h should be adjusted to 8 mg/kg q48 h in children aged 1–6 years. Importantly, CPK should be closely monitored more than once a weekin pediatric patients with renal impairment. Our model may serve as a useful tool for predicting the PK of daptomycin and support dose adjustments or other relevant decisions in clinical settings.