Invasive fungal infections pose a continuous and serious health problem in immunocompromised patients, particularly those who have become neutropenic while receiving aggressive chemotherapy. Despite recent advances in antifungal pharmacology, invasive fungal infections contribute to more than 1.5 million deaths worldwide each year (1). Even when patients receive the appropriate drug and regimen, their antifungal drug levels at the infection site still may not be effective due to inter- and intra-individual pharmacokinetic and pharmacodynamic variabilities, potentially leading to therapeutic failure.
Merely increasing the dosage in nonresponsive patients is not a solution, as elevated drug concentrations can result in toxicity. Many antifungal drugs not only have narrow therapeutic windows but also exhibit significant differences in blood drug levels due to different absorption, distribution, metabolism, elimination, and/or interaction with concomitant drugs.
Consequently, antifungal therapeutic drug monitoring (TDM) to detect drug exposures plays an important role in optimally delivering antifungal therapy in patients with invasive infections. This review highlights the utility of antifungal TDM and explains how to perform testing using liquid chromatography tandem-mass spectrometry (LC-MS/MS).
Clinical Case Part 1
A clinician diagnosed an allogeneic hematopoietic stem cell transplant recipient with invasive aspergillosis and started treatment with intravenous voriconazole according to standard recommendations. The laboratory obtained a trough sample 5 days after treatment began and sent it to a reference laboratory for therapeutic drug monitoring (TDM). The reference lab used a threshold of >5.5 µg/mL for voriconazole toxicity. At this time, with the patient stable, the daily dosage was continued. However, 2 days later, the patient complained of visual disturbances and subsequently presented mental changes consistent with voriconazole toxicity. The result of the initial TDM sample was not yet available. The laboratory obtained a new trough sample, and the clinician adjusted the dosage based on the clinical findings. A result for the first TDM sample showing a voriconazole concentration of 6.2 µg/mL arrived 2 days after the patient presented symptoms of toxicity. The second sample showed a voriconazole concentration of 8.6 µg/mL. Could this patient’s voriconazole toxicity have been prevented?
Antifungal Drugs and Therapeutic Ranges
Currently four antifungal drug classes have been approved for managing invasive fungal infections: the polyene class (amphotericin B); the triazole class (itraconazole, fluconazole, voriconazole, and posaconazole); the echinocandins (caspofungin, micafungin, and anidulafungin); and the 5-fluorocytosine class (flucytosine). Guidelines recommend TDM when the drug presents unpredictable pharmacokinetics, and has a narrow therapeutic window and defined therapeutic range. Based on these criteria, laboratories should consider the triazoles for TDM. Although fluconazole belongs to this class, its interpatient variability relates to differences in renal function, and therefore the dose would be better adjusted according to creatinine values (2). However, TDM is highly recommended for itraconazole, voriconazole, and posaconazole.
Voriconazole is a first-line drug for treating invasive aspergillosis, an infection associated with high mortality in hematopoietic stem cell transplant (HSCT) patients. Studies have suggested that voriconazole serum trough concentrations <1µg/mL are associated with higher treatment failures, while voriconazole trough concentrations >5.5 µg/mL have been associated with toxicity (2, 3). Drug interactions are important to consider as all triazoles are inhibitors of CYP3A4. In addition, voriconazole is a strong inhibitor of CYP2C19 (3). Manifestations of voriconazole toxicity include visual disturbances, neurotoxicity, and liver dysfunction (4).
Clinicians increasingly use posaconazole, another agent of the triazole class, for prophylaxis against invasive fungal infection in HSCT patients. Studies have suggested that to effectively prevent and treat invasive infections, the posaconazole trough concentration should be >0.7 µg/mL and >1.25 µg/mL, respectively. Posaconazole serum concentrations associated with adverse effects have not been well-defined.
Amphotericin B and the echinocandins have more predictable pharmacokinetics than the triazole agents, so TDM may not be required in all patients receiving these agents.
At our institution, we’ve seen a steady trend of HSCT increasing every year, so antifungal drugs are used routinely for prophylaxis and treatment of invasive fungal infections. As a result, we decided to offer TDM in-house. Below we summarize the methodology available for TDM and the parameters that laboratories need to consider before implementing TDM for antifungal drugs.
Antifungal TDM Assays
As shown in Table 1, clinical laboratories have developed several antifungal TDM assays to measure blood concentrations of antifungal drugs, including microbiological assays (bioassays), high-performance liquid chromatography (HPLC) with ultraviolet or fluorescence detectors, and LC-MS/MS (4–6). LC-MS/MS is a relatively new method with increased sensitivity, specificity, and rapid throughput compared to other methods.
Verification and Validation of LC-MS/MS for Antifungal TDM
As a highly challenging and complex technique, LC-MS/MS requires experienced technical expertise to develop and validate methods as well as maintain and troubleshoot instruments. When developing an LC-MS/MS method for antifungal TDM, laboratories must take into account and optimize at least seven conditions and parameters:
1) Chemical and physical properties such as chemical structure, solubility and stability in matrix, and acidity.
2) Sample preparation: The most common sample types used routinely for antifungal TDM are plasma and serum. There is a general requirement to remove proteins and preferably other interference materials that can cause matrix effects—the main obstacle of quantitative LC-MS/MS methods.
3) LC column: Even with a very sophisticated MS/MS instrument, an optimized LC column with better separation efficiency is very useful to remove interferences from samples that affect ionization and induce matrix effects.
4) Mobile phase: Solvents and buffers for making the mobile phase have to be compatible with MS.
5) Internal standards: Laboratories must select internal standards carefully because they are critical to compensating for sample preparation variation, ionization efficiency differences, and matrix effects with LC-MS/MS detection. The ideal internal standards are isotope-labeled target analytes, which have the exact same physical and chemical properties as the target analytes.
6) Calibrator and QC: When available, standardized and commercial calibrators and QC should be the first choice for LC-MS/MS method development.
7) CLIA requirements: Laboratories must develop and validate their method according to CLIA requirements. Specific recommendations for validation can be found in Clinical and Laboratory Standards Institute (CLSI) C62-Liquid Chromatography-Mass Spectrometry Methods; Approved Guideline.
When validating an LC-MS/MS method, laboratories should include sensitivity and specificity, matrix effects, carryover, linearity and dilution, trueness (accuracy), imprecision, recovery, and interference. Labs also must establish acceptance criteria for each component of the method validation before starting data collection.
Acceptable criteria may be based on biological variation, established clinical guidelines, and local, regional, and/or national regulatory requirements (7). In Table 2 we present typical validation parameters and the requirements summarized from CLSI (C62-A). We also recommend evaluating the trueness of developed LC-MS/MS methods by cross-validating them, when available, with reference standards, reference methods, and/or reference labs.
Unfortunately, proficiency testing programs specifically for TDM using LC-MS/MS instrumentation are not readily commercially available. Laboratories performing antifungal TDM often struggle to find ways to fulfill this requirement. Some laboratories participate in an international proficiency testing program for measuring antifungal concentrations organized by the Dutch Association for Quality Assessment in Therapeutic Drug Monitoring and Clinical Toxicology (8), while others send blind samples to a reference laboratory using the same methodology to compare results to their in-house testing.
Application of LC-MS/MS in Antifungal TDM
The triazoles are the most common antifungal drugs in clinical practice, used often for both prophylaxis and treatment of invasive infections due to their broad spectrum activities. Although LC-MS/MS for TDM works well with all the agents in the triazole class—itraconazole, fluconazole, voriconazole, and posaconazole—laboratories most frequently employ LC-MS/MS for voriconazole and posaconazole because of their increasing use in clinical practice. Laboratories have developed and validated rapid, simple, flexible, and multiplexed assays, including all four of the triazoles.
As noted previously, questions linger about the need for TDM for amphotericin B and the echinocandins, although TDM for these agents has been determined in human blood using LC-MS/MS assays (9, 10). Finally, when reporting results, laboratories should provide information about the recommended therapeutic range as well as the possible toxic threshold of each drug tested.
Clinical Case Part 2
The patient had trough voriconazole concentrations that were clearly above the 5.5 µg/mL threshold. The higher concentration was present 1 or 2 days before the clinical manifestations; however, because the la-boratory was sending the sample to a reference laboratory for testing, the turnaround time reached 3–5 days, depending on the day of sample collection. As a result, the dosage was not adjusted until the pa-tient’s symptoms appeared. Since our institution started performing TDM in-house, we have had fewer critical values to report because we release results in a timely manner and the clinical team is able to make adjustments quickly if neces-sary. We recommend that labs use serum trough samples for TDM of antifungal drugs. Most of the time, the drug is administered by mouth. In cases where the drug is administered intravenously, care should be tak-en that the sample is not collected from the same line through which the antifungal was administered, or if the same line is used, that the line is flushed with saline first to avoid contamination with residual drug in the line.
In antifungal TDM, accurately detecting drugs and/or their metabolites is extremely important because of the drugs’ narrow therapeutic windows. LC-MS/MS’s superior selectivity, high sensitivity, and simultaneous quantitation of multiple drugs make it a possible reference method for antifungal TDM. However, LC-MS/MS also presents some challenges for labs.
Instruments and maintenance are expensive, and laboratories need experienced personnel to develop a method. LC-MS/MS methods also suffer from high inter-lab and inter-instrument variation due to a lack of method standardization, commercial calibrators and QC, and stable-labeled isotypes. Future development should focus on simplifying the technology and making these complex instruments more affordable and robust, easier to use, and more open to automation.
Further improvements in sensitivity, specificity, and throughput, as well as the commercialization of LC-MS/MS methods, will enable broad LC-MS/MS applications in antifungal TDM. In the near future, highly complex LC-MS/MS, once limited to specialized labs, will become widely utilized not only in commercial labs and large clinical centers, but also in physicians’ office labs.
Without a doubt, using LC-MS/MS for antifungal TDM will increase clinical efficacy, decrease toxicity risk, and improve treatment response in invasive
1. Campoy S, Adrio JL. Antifungals. Biochem Pharmacol November 2016;133:86–96.
2. Ashbee HR, Gilleece MH. Has the era of individualised medicine arrived for antifungals? A review of antifungal pharmacogenomics. Bone Marrow Transplant 2012;47:881–94.
3. Andes D. Optimizing antifungal choice and administration. Curr Med Res Opin 2013;29 Suppl 4:13–8.
4. Ashbee HR, Barnes RA, Johnson EM, et al. Therapeutic drug monitoring (TDM) of antifungal agents: guidelines from the British Society for Medical Mycology. J Antimicrob Chemother 2014;69:1162–76.
5. Müller C, Arndt M, Queckenberg C, et al. HPLC analysis of the antifungal agent posaconazole in patients with haematological diseases. Mycoses 2006;49 Suppl 1:17–22.
6. Chhun S, Rey E, Tran A, et al. Simultaneous quantification of voriconazole and posaconazole in human plasma by high-performance liquid chromatography with ultra-violet detection. J Chromatogr B Analyt Technol Biomed Life Sci 2007;852:223–8.
7. Clinical and Laboratory Standards Institute (CLSI) C62-A. Liquid chromatography-mass spectrometry methods; Approved guideline. 2014;34(16):1–71.
8. Lempers VJ, Alffenaar JW, Touw DJ, et al. Five year results of an international proficiency testing programme for measurement of antifungal drug concentrations. J Antimicrob Chemother 2014;6:988–94.
9. Decosterd LA, Rochat B, Pesse B, et al. Multiplex ultra-performance liquid chromatography-tandem mass spectrometry method for simultaneous quantification in human plasma of fluconazole, itraconazole, hydroxyitraconazole, posaconazole, voriconazole, voriconazole-N-oxide, anidulafungin, and caspofungin. Antimicrob Agents Chemother 2010;54:5303–15.
10. Farowski F, Cornely OA, Vehreschild JJ, et al. Quantitation of azoles and echinocandins in compartments of peripheral blood by liquid chromatography-tandem mass spectrometry. Antimicrob Agents Chemother 2010;54:1815–9.
Bing Pang, MD, PhD, is a research associate working on mass spectrometry at Wake Forest University School of Medicine in Winston-Salem, North Carolina.+Email: email@example.com
Steven H. Wong, PhD, is a professor of pathology and director of clinical chemistry and toxicology and co-director of the Clinical and Translational Mass Spectrometry Center at Wake Forest University School of Medicine in Winston-Salem, North Carolina.+Email: firstname.lastname@example.org
Elizabeth Palavecino, MD, is a professor of pathology and director of clinical microbiology and co-director of the Clinical and Translational Mass Spectrometry Center at Wake Forest University School of Medicine in Winston-Salem, North Carolina.+Email: email@example.com