Failure of Posaconazole Therapy in a Renal Transplant Patient with Invasive Aspergillosis Due to Aspergillus fumigatus with Attenuated Susceptibility to Posaconazole

Abstract

We report the case of a kidney transplant recipient with invasive aspergillosis due to Aspergillus fumigatus resistant to voriconazole and intermediately susceptible to posaconazole who failed posaconazole therapy. Plasma posaconazole concentrations indicated an unfavorable ratio of the area under the concentration-time curve over the MIC. Posaconazole should be used with caution for invasive aspergillosis caused by strains with attenuated posaconazole susceptibility, as drug exposure may be inadequate, resulting in therapeutic failure.

TEXT

Acquired azole resistance has been reported to be an emerging problem for patients with aspergillosis (3, 11, 13, 14). In resistant Aspergillus isolates, mutations in the cyp51A gene may lead to various non-wild-type susceptibility phenotypes of the licensed azole drugs (3, 12). Itraconazole commonly shows no in vitro activity, while the activities of voriconazole and posaconazole may be attenuated; i.e., the MICs are elevated compared to those of wild-type isolates, but the drugs retain moderate in vitro activity (8, 12). Here we describe the first case of an azole-naïve patient with aspergillosis due to an Aspergillus fumigatus isolate resistant to voriconazole and with reduced susceptibility to posaconazole, for whom posaconazole treatment failed. The measurement of posaconazole concentrations in blood indicated a probable failure to achieve the pharmacodynamic target.

A 51-year-old female kidney transplant patient, weighing 104 kg, was admitted to our hospital in October 2009 with abdominal pain, fever, leukocytosis, and increased C-reactive protein (CRP) and creatinine levels (350 μmol/liter). She had received a cadaveric-kidney transplant in her left iliac fossa 9 months earlier. Her immunosuppressive regimen still included oral tacrolimus, mycophenolate mofetil, and prednisone. Nineteen days earlier, abdominal hysterectomy had been performed in another hospital because of a growing uterine myoma, with normal postoperative renal function. However, upon admittance to our hospital, a renal ultrasound showed a partial stenosis of the distal transplant ureter. The newly inserted percutaneous nephrostomy catheter dislocated, and only after four attempts was a catheter reinstalled in the nondilated renal system. Cutaneous erythema developed around the nephrostomy catheter, for which a 5-day course of ceftriaxone was given.

During the following months she remained intermittently febrile. In November, renal biopsy specimens revealed tubulitis and interstitial inflammation. A 3-day course of methylprednisolone was given for possible transplant rejection.

In January 2010, the stenotic transplant ureter was replaced by the left native ureter. Preoperative urine was sterile. Perioperatively, pus and a necrotic transplant ureter were found medially to the transplant. A Blankophor stain of the purulent material showed septate hyphae and dichotomous branching, and cultures showed A. fumigatus. Treatment was initiated with oral voriconazole at 300 mg twice a day (b.i.d.) after a loading dose of 400 mg b.i.d. on the first day. The tacrolimus dose was reduced by 50%. Antifungal susceptibility testing using EUCAST methodology (6) showed that voriconazole was inactive (MIC > 16 μg/ml); MICs of itraconazole and posaconazole were 2 and 0.5 μg/ml, respectively. The amphotericin B MIC was 0.5 μg/ml. After the MIC results became available, voriconazole was replaced by posaconazole at 200 mg four times a day (q.i.d.), aiming at an exposure above 1 μg/ml. However, after 1 week of therapy under steady-state conditions, the trough plasma posaconazole concentration measured 0.6 μg/ml. The dosing regimen was changed to 200 mg six times daily with intake after the intake of high-fat-containing food. The increased posaconazole dose still yielded the same concentration after the second week of therapy.

In February, pus drained spontaneously from the operation wound. Upon presentation, our patient was severely septicemic. Nephrostomy catheter urine showed Gram-positive rods and cocci. The patient was started on piperacillin-tazobactam, caspofungin, and liposomal amphotericin B. Posaconazole treatment was discontinued. Ultrasonography showed pus surrounding the kidney transplant. Computed tomography showed that the infection had progressed to the lower pole of the transplant and abdominal wall. Although surgical drainage was performed, the patient died shortly afterwards.

A urine culture grew Enterococcus faecalis and Corynebacterium species. A. fumigatus was cultured from the double-J catheter. In addition to morphological identification, the identification to the species level was confirmed by sequence-based analysis of the β-tubulin gene. Autopsy, performed 36 h after death, confirmed that the fungal infection had progressed only to the transplant and abdominal wound. Postmortem fungal culture results are shown in Table 1. Pulmonary thrombosis secondary to sepsis was identified as the most likely cause of death. Posaconazole concentrations were measured by a validated high-performance liquid chromatography (HPLC)-fluorescence method (1) using whole blood obtained by cardiac puncture and using tissue samples (Table 1). Sequence analysis showed two mutations in the cyp51A gene (Y121F and T289A) and a 46-bp tandem repeat in the gene promoter (13).

TABLE 1.

Culture results and posaconazole drug concentrations in blood and tissue samples obtained at autopsy^a

Site	Fungal culture result	MIC (μg/ml) (classification)b			POS level (μg/g)
		ITZ	VCZ	POS
Abscess	A. fumigatus	2 (I)	>16 (R)	0.5 (I)	5.1
Kidney (swab)	A. fumigatus	1 (S)	>16 (R)	0.5 (I)	5.9
Kidney (tissue)	A. fumigatus	1 (S)	>16 (R)	0.5 (I)	6.5
Renal fat tissue	A. fumigatus	1 (S)	>16 (R)	0.25 (S)	7.1
Liver	Negative				18.4
Spleen	Negative				5.8
Lung	Negative				4.1
Blood	Negative				1.1c

^aITZ, itraconazole; VCZ, voriconazole; POS, posaconazole; S, susceptible; I, intermediately susceptible; R, resistant.

^bClassification according to recently proposed interpretative guidelines (12).

^cIn μg/ml (whole blood).

Invasive aspergillosis occurs infrequently in kidney transplant recipients. We assume that the infection in our patient occurred through traumatic inoculation or fungal contamination during either the hysterectomy or the insertion of nephrostomy catheters. As our isolate was resistant to voriconazole, we switched to posaconazole, despite a slightly elevated MIC. A lipid formulation of amphotericin B would have been a therapeutic option, but because the patient was in excellent clinical condition and successful surgical drainage had been performed, we chose treatment with an oral drug, thus facilitating ambulant care. The presence of azole resistance in the absence of previous azole exposure suggests that the isolate had been obtained from the environment (7, 13, 15). Although treatment failure with azoles in azole-resistant aspergillosis has been reported previously (3, 8, 10, 14), the impact of in vitro resistance on the clinical response has been difficult to establish because several factors contribute to the treatment outcome, e.g., refractory underlying disease, low drug exposure, and previous exposure to other antifungal agents.

In this patient, primary therapy with posaconazole could be evaluated during infection with an A. fumigatus isolate with attenuated susceptibility. An oral posaconazole solution was given because the isolate was voriconazole resistant, the infection site was surgically drained, and the clinical condition permitted outpatient treatment. According to recently proposed interpretative breakpoints, the isolates were borderline to intermediately susceptible to posaconazole (12). We monitored posaconazole plasma concentrations and ultimately used a dosing regimen of 200 mg six times daily for maximum drug exposure. The posaconazole trough level of 0.6 μg/ml could have been suboptimal, as this signified total drug and not free drug. Posaconazole is a lipophilic agent, and it may be that in obese people there is more posaconazole in adipose tissues, but ultimately, serum concentrations or, even better, unbound serum concentrations are accepted as a surrogate for concentrations at the site of infection (5), although further research is needed concerning this issue. Furthermore, these levels were much lower than mean posaconazole levels measured in volunteers taking at least 800 mg posaconazole per day (2). Average plasma concentrations of 0.4 to 0.7 μg/ml corresponded with treatment success rates of only 53% for aspergillosis patients receiving posaconazole salvage therapy, compared to 75% for those with average concentrations of 1.2 μg/ml (16). Moreover, only 29.7% of 202 serum samples tested for posaconazole levels in a reference laboratory had levels above 0.7 μg/ml (9).

The efficacious area under the concentration-time curve (AUC) over the MIC for posaconazole is probably above 200, and conservative estimations are as high as 1,000 based on a nonneutropenic murine model of disseminated aspergillosis (4). The pharmacodynamic target of 200 could have been achieved with posaconazole plasma concentrations of 0.6 μg/ml if the infection had been caused by an A. fumigatus isolate with wild-type susceptibility. However, due to the elevated MICs, the AUC/MIC ratios for our patient were 30 to 60, at least 3.5- to 7-fold lower than the pharmacodynamic target. Plasma levels above 4 μg/ml would have been required to achieve a pharmacodynamic target of 200, which is impossible to achieve with the current formulation of posaconazole. Although the posaconazole tissue levels that we found in our patient exceeded 4 μg/ml, the calculation of the pharmacodynamic target is based on the plasma levels of the drug and not on tissue levels. Therefore, much higher tissue levels, i.e., those corresponding to a plasma level of 4 μg/ml, are probably required to achieve a high probability of treatment success.

Our case demonstrates that posaconazole should be used extremely cautiously for the treatment of patients with invasive aspergillosis caused by Aspergillus isolates with attenuated posaconazole susceptibility, due to unfavorable AUC/MIC ratios and a higher probability of treatment failure. Therapy with non-azole antifungal agents, e.g., lipid formulations of amphotericin B, may be an appropriate alternative treatment option when renal function is acceptable, although clinical experience with azole-resistant aspergillosis is still limited.