Association of adverse effects with high serum posaconazole concentrations

Abstract

Posaconazole therapeutic drug monitoring (TDM) is widely utilized to assess therapeutic efficacy and safety; however, clinical effects of very high serum concentrations are unknown. A retrospective review of 90 patients receiving posaconazole for treatment or prophylaxis of invasive fungal infections with serum concentrations ≥3000 ng/mL from 1/1/2019 to 4/30/2021 evaluated the incidence and type of adverse drug reactions (ADRs). Symptomatic ADRs were very common in patients with posaconazole concentrations of ≥5000 ng/mL and 3000–4999 ng/mL (80% vs. 58.8%; P = 0.31). Posaconazole TDM should be performed for both treatment and prophylaxis indications and dose decrease for serum concentrations >3000 ng/mL should be considered.

Background

Posaconazole is an extended-spectrum azole antifungal commonly utilized for treatment and prophylaxis of fungal infections, particularly molds. Historically, adequate serum concentrations proved challenging with posaconazole suspension due to erratic, fat dependent absorption; however, delayed-release tablets have demonstrated improved bioavailability generally resulting in higher serum concentrations.¹ Posaconazole therapeutic drug monitoring (TDM) is widely utilized to assess therapeutic efficacy and safety [i.e., hepatotoxicity, corrected QT interval (QTc) prolongation] given drug-interactions and substantial variability in bioavailability between individual patients.² While standards for TDM are not well established and reference intervals have been debated in the literature, trough concentrations of >700 ng/mL for prophylaxis, >1000–1250 ng/mL for treatment, and an upper limit of 3745 ng/mL have been suggested.²

Currently, little is known regarding clinical effects of very high serum posaconazole serum concentrations (i.e., ≥5000 ng/mL). In clinical trials, treatment emergent and treatment related adverse drug reactions (ADR) were 98% and 30%, respectively, with increased alanine transaminase (ALT) or aspartate transferase (AST), nausea and vomiting, and hypokalemia frequently observed.³ While ADR are common, higher concentrations have additionally been associated with hypokalemia, secondary hypertension, and pseudohyperaldosteronism.^4,5 Data regarding associations between serum drug concentrations and toxicity are mixed.⁵ Thus, we sought to identify incidence of ADR types in patients with serum posaconazole concentrations approaching and above the proposed upper threshold of 3745 ng/mL.

Methods

We conducted a retrospective study comparing incidence of ADR in patients with serum posaconazole concentrations of 3000–4999 ng/mL to concentrations of ≥5000 ng/mL. Patients with serum concentration ≥3000 ng/mL from 1/1/2019 to 4/30/2021 at any Mayo Clinic site (3 centers of Rochester, Florida, and Arizona and the Mayo Clinic Health System) were included. For patients with multiple concentrations during this timeframe, outcomes were censored after first serum concentration ≥3000 ng/mL. Patients <18 years of age and those who declined Minnesota Research Authorization were excluded. This study was approved by the Mayo Clinic Institutional Review Board (#21-005253).

The laboratory information system (LIS) was utilized to identify patients with posaconazole concentrations meeting criteria for inclusion. Chart review was performed, and patient baseline characteristics, concomitant hepatotoxic drugs, potential interacting drugs, metabolic laboratory results, endocrine parameters, and posaconazole regimen and TDM were collected.

The primary outcome was incidence of symptomatic ADR documented at first serum concentration ≥3000 ng/mL. Secondary outcomes were laboratory defined hepatoxicity, electrolyte and adrenal laboratory abnormalities, QTc changes, and dose changes in response to TDM. Patients with posaconazole concentrations of 3000–4999 ng/mL were compared to ≥5000 ng/mL using Fisher's exact tests for categorical data and Wilcoxon rank sum tests for continuous data. All tests were two-sided with P-values of <0.05 considered statistically significant.

Results

In total, 90 patients met inclusion criteria, 80 (88.9%) with concentration of 3000–4999 ng/mL, and 10 (11.1%) with concentration of ≥5000 ng/mL occurring at a median 91 days [Interquartile range (IQR), 26–443] and 27 days [IQR, 12–45] from drug initiation, respectively. Baseline characteristics were similar between groups, except for age, weight, and body mass index (BMI). Patients were predominantly female, white, and immunocompromised, with a mix of definitive treatment, empiric treatment, and prophylaxis indications.

Symptomatic ADRs were very common in both groups with posaconazole concentrations of ≥5000 ng/mL and 3000–4999 ng/mL (80% vs. 58.8%; P = 0.31). For the overall cohort, reported ADRs were primarily neurologic (49.1%), followed by gastrointestinal (32.7%), cardiac (27.3%), respiratory (27.3%), and other (36.4%), with no posaconazole-induced pseudohyperaldosteronism observed (Table 1). Electrolytes and QTc results were similar between groups, but median overall QTc was high (456 ms [IQR, 435–479]) at time of supratherapeutic concentration (Table 2).

Table 1.

Baseline characteristics and posaconazole concentrations.

	Concentration 3000–4999 ng/mL (n = 80)	Concentration ≥5000 ng/mL (n = 10)	Total cohort (n = 90)	P-value
	Median [IQR] or n (%)
Age (years), median, [IQR]	62 [56–67]	50 [34–55]	61 [54–67]	0.015
Gender				0.50
Male	37 (46.3%)	3 (30%)	40 (44.4%)
Female	43 (53.8%)	7 (70%)	50 (55.6%)
Race				0.58
White	73 (91.3%)	9 (90%)	82 (91.1%)
Black	1 (1.3%)	0 (0%)	1 (1.1%)
Asian	2 (2.5%)	0 (0%)	2 (2.2%)
Other/missing	4 (5%)	1 (10%)	5 (5.6%)
Height (cm), median, [IQR]	170 [165–177]	155 [153–176]	170 [163–177]	0.057
Weight (kg) median, [IQR]	73 [59–87]	53 [49–69]	69 [57–85]	0.007
BMI, median, [IQR]	25.0 [21.6–29.3]	20.7 [19.7–23.5]	24.5 [20.7–28.6]	0.032
Solid organ transplant	23 (28.8%)	1 (10%)	24 (26.7%)	0.28
Hematopoietic stem cell transplant	22 (27.5%)	4 (40%)	26 (28.9%)	0.47
Active malignancy	23 (28.8%)	3 (30%)	26 (28.9%)	0.99
Other immunocompromised state	4 (5.0%)	1 (10%)	5 (5.6%)	0.45
Cirrhosis	1 (1.3%)	0 (0%)	1 (1.1%)	0.99
Systemic hypertension	30 (37.5%)	2 (20%)	32 (35.6%)	0.48
Indication				0.78
Prophylaxis	34 (42.5%)	4 (40%)	38 (42.2%)
Empiric/culture negative	7 (8.8%)	0 (0%)	7 (7.8%)
Treatment	39 (48.8%)	6 (60%)	45 (50%)
Formulation				0.30
Delayed release tablet	78 (97.5%)	9 (90%)	87 (96.7%)
Liquid	1 (1.3%)	0 (0%)	1 (1.1%)
IV	1 (1.3%)	1 (10%)	2 (2.2%)
Supratherapeutic level (ng/mL), median, [IQR]	3580 [3150–3960]	6140 [5540–7040]	3660 [3210–4310]	<0.001
Time from initiation (days), median, (IQR]	91 [26–443]	27 [12–45]	80 [22–411]	0.066
Loading dose (mg)	22 (28.2%)	2 (20%)	24 (27.3%)	0.72
Initial dose (mg), median, [IQR]	300 [300–300]	300 [300–300]	300 [300–300]	0.23
Current dose (mg), median, (range)	300 (200–800)	300 (200–600)	300 (200–800)	0.43
Concentration prior to supratherapeutic draw	41 (52.6%)	4 (44.4%)	45 (51.7%)	0.73
Prior concentration (ng/mL), median, [IQR]	2080 [1410–2400]	2360 [1528–4440]	2080 [1410–2440]	0.45
Symptomatic at supratherapeutic concentration	47 (58.8%)	8 (80%)	55 (61.1%)	0.31
Neurologic symptoms	21 (44.7%)	6 (75%)	27 (49.1%)	0.14
Gastrointestinal symptoms	15 (31.9%)	3 (37.5%)	18 (32.7%)	0.99
Cardiac symptoms	13 (27.7%)	2 (25%)	15 (27.3%)	0.99
Respiratory symptoms	14 (29.8%)	1 (12.5%)	15 (27.3%)	0.42
Other symptoms	18 (38.3%)	2 (25%)	20 (36.4%)	0.70
Suspect possible hepatotoxicityb	18 (23.4%)	4 (40.0%)	22 (25.3%)	0.26
Intervention				0.94
Decrease dose	27 (33.8%)	3 (30%)	30 (33.3%)
Discontinue antifungal	9 (11.3%)	1 (10%)	10 (11.1%)
Continue without change	39 (48.8%)	6 (60%)	45 (50%)
Other medication changec	5 (6.3%)	0 (0%)	5 (5.6%)
Next concentration (ng/mL), median, [IQR]	2760 [1880–3910]	2245 [823–3835]	2760 [1740–3910]	0.43
Dose at next concentration (mg), median, [IQR]	300 [200–300]	300 [200–300]	300 [200–300]	0.87

^aInherited immunodeficiency, use of immunosuppressant medications (adalimumab and mycophenolate), uncontrolled type 2 diabetes, and untreated Gleason 6 adenocarcinoma of the prostate.

^bCholestatic injury pattern for the majority (82%, n = 18; 2 excluded, 2 unlikely, 11 possible, 2 probable, and 2 highly probable), followed by mixed (14%, n = 3; 1 unlikely and, 1 probable, 1 possible), and hepatocellular (5%, n = 1; 1 highly likely).

^cPosaconazole formulation change, held and then later re-initiated, or dose change occurring prior to obtaining concentration due to other laboratory abnormalities.

IQR, interquartile range; BMI, body mass index.

Table 2.

Laboratory and ECG Monitoring.

Labs	Baseline	Supratherapeutic concentration	N at baseline and supratherapeutic concentration
Potassium (mmol/L), median [IQR]	4.3 [3.9–4.5]	4.0 [3.6–4.6]	76, 85
Bicarbonate (mmol/L), median [IQR]	23 [22–26]	25 [23–27]	59, 69
ALT (U/L), median [IQR]	22 [16–36]	31 [20–44]	82, 86
AST (U/L), median [IQR]	23 [19–37]	29 [22–43]	79, 82
Total bilirubin (mg/dL), median [IQR]	0.4 [0.2–0.5]	0.5 [0.3–0.6]	79, 78
Alk phos (U/L), median [IQR]	87 [70–121]	90 [74–130	80, 84
QTc (ms), median [IQR]	442 [427–463]	456 [435–479]	47, 48

ECG, electrocardiogram; IQR, interquartile range; ALT, alanine aminotransferase; AST, aspartate aminotransferase; Alk Phos, alkaline phosphatase; QTc, corrected QT interval.

Hepatotoxicity was also commonly observed (≥5000 ng/mL 40% vs 3000–4999 ng/mL 23.4%, P = 0.26) with likelihood of attribution of hepatic injury to posaconazole exposure assessed using the Roussel Uclaf Causality Assessment Method (RUCAM).^6,7 Of 22 patients with suspected hepatotoxicity, cholestatic injury pattern was most common (Table 1).

An association between weight and BMI and supratherapeutic concentrations ≥5000 ng/mL was observed (Table 1). However, symptomatic ADR and suspected hepatotoxicity were not associated with weight (72.5 kg vs. 69 kg, P = 0.51; and 72.3 kg vs. 67.5 kg, P = 0.16) or BMI (25.2 vs. 24.1, P = 0.72; and 25.2 vs. 22, P = .055), respectively.

Of the total cohort, 96.7% received posaconazole delayed-release tablets at initial median dose of 300 mg daily, with 27.3% of patients receiving a loading dose. About half of patients (51.7%) had serum assay results prior to index supratherapeutic concentration. At time of supratherapeutic concentration, the median daily dose remained 300 mg for patients with both concentrations of 3000–4999 ng/mL and ≥5000 ng/mL, with median posaconazole concentrations of 3580 ng/mL and 6140 ng/mL observed, respectively. In response to supratherapeutic concentrations, 50% of posaconazole regimens were continued without change, 33% decreased dose, 11% were discontinued, and 6% resulted in other changes. Repeat TDM was completed in 70% of patients following index supratherapeutic concentration (Table 1).

Discussion

Our study evaluated frequency of ADRs in high (3000–4999 ng/mL) and very high (≥5000 ng/mL) concentrations, finding a high rate of symptomatic ADRs and illuminating safety concerns for patients with posaconazole concentrations ≥3000 ng/mL. Interestingly, we observed an association between lower body weight and BMI with concentrations ≥5000 ng/mL. However, our analysis did not show a statistically significant association between weight or BMI and symptomatic ADR or possible hepatic toxicity, though a trend toward an association between BMI and hepatotoxicity was noted. While posaconazole is not currently dosed based on body weight, further studies are warranted to better understand the relationship between BMI and supratherapeutic concentrations.

Previous studies have demonstrated an association of higher posaconazole concentrations with hypokalemia, secondary hypertension, and occasionally pseudohyperaldosteronism.^4,5 Pseudohyperaldosteronism was not observed in our cohort, but appeared to be considered infrequently, with aldosterone and renin, deoxycortisol, and morning cortisol levels ordered in four, one, and four patients, respectively. Elevations in QTc interval as compared with baseline, however, were observed in our total cohort at the time of supratherapeutic concentration, underscoring the importance of cardiac monitoring.

This study also revealed that, despite supratherapeutic posaconazole concentrations, regimens were frequently continued without a change in dosing strategy. Furthermore, repeat posaconazole TDM was only ordered in 70% of patients subsequent to supratherapeutic concentration despite only 11.1% of patients discontinuing posaconazole. At our institution, azole concentrations are not highly protocolized and are obtained at treatment team discretion. While our institutional guidance provides recommendations for timing and reference ranges, there is currently no guidance provided for ordering frequency or protocolized dose changes based on TDM results. Our findings emphasize the importance of posaconazole dose adjustments and repeat TDM in response to serum concentrations, clinical response assessment, and symptomatic ADRs.

A key limitation of our study was its retrospective nature, and thus relied on medical record documentation for reporting of symptomatic ADRs. It is not possible to confirm if additional adverse events occurred without being documented in the notes, subjecting our study to reporting bias and limiting generalizability. Additionally, posaconazole concentrations may have only been ordered for patients with either suspected ADRs or at high risk of toxicity (i.e., sampling bias).

Conclusions

Our findings emphasize that posaconazole concentrations should be monitored for both treatment and prophylaxis indications, and, importantly, dose adjusted according to serum concentration and patient symptoms. Patients with lower body weight may be at risk for very high posaconazole concentrations and subsequent adverse effects, further underscoring the need for TDM in this subgroup.

Author contributions

Kelsey Jensen (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing), Omar Abu Saleh (Conceptualization, Data curation, Project administration, Visualization, Writing – review & editing), Supavit Chesdachai (Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Resources, Visualization, Writing – review & editing), Paul J. Jannetto (Conceptualization, Investigation, Methodology, Project administration, Resources, Validation, Writing – review & editing), Kristin C. Mara (Data curation, Formal analysis, Investigation, Validation, Writing – review & editing), Zachary A. Yetmar (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – review & editing), and Christina G. Rivera (Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Supervision, Validation, Writing – review & editing)

Funding

This work was supported by the Pharmacy Research Committee from the Department of Pharmacy at Mayo Clinic. This project was supported by Grant Number UL1 TR002377 from the National Center for Advancing Translational Sciences (NCATS). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

Disclosure of conflict of interest

Authors have no conflicts of interest to disclose.