Invasive fungal infections (IFIs) are a common complication of chemotherapy and transplantation in immunocompromised patients. Candida and Aspergillus species are the most common causes of IFIs, but other yeasts and filamentous fungi including Fusarium, Scedosporium, and Purpureocillum are emerging pathogens (Gullo, 2009). Given the limited clinical manifestations and the risk of developing multiresistant strains, the methodology for early diagnosis and treatment of IFI is very limited (Casucci et al., 2014). Once diagnosed, early initiation of appropriate antifungal therapy is essential for reducing the extremely high morbidity and mortality rates (Gullo, 2009). Since the early 1950s, amphotericin B deoxycholate has been the first choice of drug to treat IFIs (Fluckiger et al., 2006), but several alternative treatment options have emerged in recent years. Lipid-based formulations of amphotericin B and expanded-spectrum triazoles (i.e., voriconazole and posaconazole) and echinocandins have been introduced in clinical practice for both the prevention and treatment of IFIs (Pound et al., 2011). Breakthrough of fungal infections and treatment failures may well be associated with low serum concentration of these drugs (Lerolle et al., 2014; Kim et al., 2017). In contrast, high serum concentration of these drugs are associated with toxic effects (Andes et al., 2009).

Posaconazole (PCZ) is an extended-spectrum triazole with broad activity against a variety of fungal pathogens, including yeasts and molds (Mehta and Langston, 2009). PCZ has been used for the prophylaxis of IFI in patients undergoing chemotherapy or HSCT (Leclerc et al., 2018). Previous studies have reported a significant association between a steady-state lower serum concentration of PCZ and a higher breakthrough rate of IFIs (Dolton et al., 2012; Moore et al., 2015; Chin et al., 2017). The absorption of PCZ is significantly increased when administered with a (high-fat) meal (Krishna et al., 2009). PCZ is not metabolized by CYP450 isoenzymes and is mostly excreted unchanged in the feces (Spencer et al., 2011). Between 20 and 30% of the PCZ dose is glucuronidated by the phase II enzyme, UGT1A4. Therefore, inhibitors or inducers of this elimination pathway may affect the serum concentration of PCZ (Dolton et al., 2012). With unchanged administration of phenytoin (to prevent seizures), on the 10th day of a steady-state serum concentration, the PCZ level was reduced by as much as 50%, which corresponds to a 90% increase in the steady-state PCZ clearance rate (Leung et al., 2015). At present, routine measurements of the serum concentration of PCZ is not carried out in clinical practice in China.

Proton pump inhibitors (PPIs) are widely used for the treatment of peptic ulcers and gastrointestinal hemorrhage and are often co-administered with antifungal agents in patients at a high risk of contracting IFIs (Yan et al., 2018). However, PPIs are likely to inhibit the metabolism of PCZ and thus decrease the serum concentration (Chayakulkeeree et al., 2015; Yan et al., 2018). Therefore, in the present study, the effect of a PPI on the serum concentration of PCZ was investigated. It is noteworthy that PCZ has a long half-life of approximately 35 h and the steady-state concentration is achieved after 7 days of administration. There is minimum fluctuation around the mean values once the steady-state concentration is reached (Chae et al., 2015). Hence, the serum concentration on day 7 was regarded as the average steady-state PCZ serum concentration for each subject. A study of 4,192 Chinese patients in 2015 demonstrated that antifungal prophylaxis produced an independent protective effect, but that this therapy was not commonly used, even in high-risk patients (Sun et al., 2015). PCZ showed good efficacy as a prophylaxis agent against IFIs in high-risk neutropenic Chinese patients and was well tolerated during long-term use (Huang et al., 2012; Shen et al., 2013). It is imperative to establish a definitive cut-off value to provide guidance for PCZ administration and thus improve its prophylaxis efficacy.

Therefore, we explored the relationship between the serum concentration of PCZ and its prophylaxis action, to determine whether there are predictable risk factors for prophylaxis therapy failure, to provide unequivocal evidence for a reduced breakthrough rate for IFIs and to promote diagnosis of fungal infections in patients with neutropenia or those who are immunocompromised.

The present study explored the correlation between the serum concentration of PCZ and the breakthrough rate of IFIs in hematology patients who received chemotherapy or allogeneic HDCTs. The data presented is based on 74 enrolled patients who underwent PCZ prophylaxis. The results revealed that 10 patients (13.5%) in total had progression of fungal infections including five patients who received chemotherapy and five who underwent transplantation, which is in good agreement with previously published data. For example, a study in the United States reported that 15% of patients who received PCZ tablets for prophylaxis had breakthrough IFIs (Pagano and Mayor, 2018). However, a single-center study of patients with acute myeloid leukemia documented four breakthrough IFIs (4/84, 4.8%) in the PCZ group, which was lower than the normal range (Ozkocaman et al., 2018). In addition, different formulations of PCZ may be related to the breakthrough rate of IFIs, since one study noted that Aspergillus infections occurred in nine (8.7%) patients in the PCZ suspension group and in no patient who took the tablet formulation (Leclerc et al., 2018). Other antibiotic drugs were evaluated with regard to prophylaxis efficacy and showed similar rates. A 12% breakthrough rate of IFIs was found in hematologic malignancy patients prescribed isavuconazole for prophylaxis (DeVoe et al., 2018). Invasive Aspergillus breakthrough occurred in 6/46 (13%) of patients treated with caspofungin, with symptoms that included persistent fever and neutropenia (Lafaurie et al., 2010). Overall, our study has demonstrated a similar prophylaxis efficacy of PCZ, but it is still unacceptably low and needs to be considerably improved.

In the present study, a cut-off value of 0.47 μg/ml has been proposed, which provides a critical evaluating index during PCZ prophylaxis. Given the high risk of mortality from IFIs and inter- and intra-subject variability in PCZ pharmacokinetics, therapeutic drug monitoring (TDM) should be a strategy that should be employed to optimize drug therapy (Yi et al., 2017). A serum level of PCZ > 0.70 μg/ml is normally the target concentration for effective prophylaxis (Maleki et al., 2018). A meta-analysis of 28 studies that involved 1,930 patients found that patients with PCZ serum concentrations >0.5 mg/L were twice as likely to achieve successful responses compared to those with lower concentrations (Chen et al., 2018). In our study, a cut-off concentration value of 0.47 μg/ml is recommended, which provides a critical evaluating index for PCZ prophylaxis during the TDM process in China.

In addition, our results showed that the breakthrough point in male subjects was 10.4% and in female subjects 19.2%, which implies that gender may significantly influence drug action: males (median = 521.50 ng/ml) exhibited greater PCZ serum concentrations than females (median = 376.50 ng/ml, P = 0.028) among the 172 patients who contracted IFIs (Allegra et al., 2017). The study subjects’ ages, body mass index (BMI), and the administered PCZ dose did not significantly affect PCZ pharmacokinetics (Allegra et al., 2017). However, a study that evaluated the risk factors for sub-therapeutic levels of PCZ tablets found that male gender was one factor associated with PCZ troughs (<0.7 μg/ml) (Tang et al., 2017). Our data showed that the breakthrough rate in male subjects (10.4%) was lower than for female subjects (19.2%). Further investigation with a larger cohort of subjects will be required before an unequivocal conclusion can be reached.

Proton pump inhibitors are among the most frequent drugs administered to hematology patients (Neubauer et al., 2010). Patients who are likely to require antifungal therapy often have abnormal gastric pH levels and present with gastrointestinal disorders, which are treated with PPIs (Krishna et al., 2009). Patients who receive PPIs are more likely to have lower concentrations of PCZ in their serum (Cojutti et al., 2013; Tang et al., 2017), a finding in agreement with the present study (Table 5). However, multivariable regression analysis showed that the administration of PPIs had a limited influence on the failure rate of prophylactic antifungal treatment (Table 6). One possible reason is that PPIs can effectively prevent gastrointestinal mucosal inflammation elicited by chemotherapy or transplantation, and a complete mucosal barrier can reduce the chance of acquiring fungal infections. Additionally, the median PCZ serum concentration in PPI combined with PCZ-treated subjects was above the cut-off value (0.47 μg/ml), which may also explain the non-significant difference in the breakthrough rate between the PPI and non-PPI-treated groups.

Apart from the small number of subjects in the chemotherapy group, the limitations of the present study were its single-center design with limited generalizability, and there may also have been selection biases that affected the extrapolation of research results. However, our study used continuous enrollment to reduce selection biases to some extent. In addition, it is fair to say that confounding biases are an unavoidable problem in observational research, but a multifactorial model was used to analyze the influencing factors for fungal infection prophylaxis failure in order to control these.

In conclusion, we established a cut-off concentration value of PCZ in Chinese hematology patients treated with chemotherapy and HSCT who were at a significant risk of IFIs. We found that PPI medication decreased the PCZ serum concentration when PCZ prophylaxis was combined with PPI treatment. However, the risk of onset of IFIs occurred only when the serum blood concentration was <0.47 μg/ml. It is strongly recommended that PCZ serum concentrations should be monitored, especially in patients at a high risk of contracting IFIs.

The datasets generated for this study are available on request to the corresponding author.

The studies involving human participants were reviewed and approved by the Medical Ethics Committee of the First Affiliated Hospital of Suzhou University approved the study (2015 IRB119). Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

WL and FX conceptualized the study. HQ carried out the analysis and interpretation. WL and HZ contributed to the data collection and analysis. WL wrote the original draft. HZ, DW, XM, and AS wrote, reviewed and edited the manuscript. All authors read and approved the submitted manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.