ltraconazole : Precautions regarding drug interactions and bioavailability

OBJECTIVE: To describe patients wilh fungal infections in whom drug interactions and bioavailabil ity problems were associated with itraconazole treatment failure. DESIGN: Retrospective chart review. SETTING: Two te r tiary care university-affil iated leaching hospitals. PoPULATION STUDIED: llraconazole-trealed patients between 1990 and 1992. MAIN REsULTs : Four palicnls treated with itraconazole were identified in whom s uboptimal serum itraconazole levels were associated with relapse of disseminated h istoplasmosis, treatment failure of invasive aspergillosis or development of superficia l mycoses. Low serum itraconazole levels were associated with concurrent therapy with rifampin or agents inh ib iting gastric acid secretion, with administration of itraconazole capsule contents through a nasogastric tube or with lhe presence of AIDS. CoNCLUSIONS: It is recommended lhal serum itraconazole levels be measured in palienls wilh potentially life-threatening mycoses if coexistent factors may interfere with itraconazole lherapy.


A MPHOTERICIN B HAS BEEN THE MAINSTAY OF TREATMENT
for invasive mycoses since its introduction in 1956; however. it must be given intravenously, and adverse effects are common (1). Since the late 1970s, a number of orally administered azole antifungal agents, including ketoconazole, fluconazole and itraconazole. has been developed, and they offer therapeutic alternatives to amphotericin B for certain mycoses (2). Itraconazole is a recently licensed triazole that is well tolerated (2)(3)(4) and has broad spectrum antifungal activity (2,5). Clin ical trials have shown encouraging results in the treatment of aspergillosis (6. 7). histoplasmosis (8), and other systemic and superficial mycoses (3,(9)(10)(11)(12). Its tolerance and efficacy indicate t11at it will be an important addition to antifungal chemotherapy.
Drug interactions resulting in suboptimal serum azole levels and failed t11erapy have been a greater problem with ketoconazole tl1an with fluconazole (13 -15). That such problems also complicate itraconazole t11erapy may not be widely appreciated, despite a recent report ( 16). There have been few other reports of drug interactions or other factors associated with low serum itraconazole levels and treatment failures . Four patients are presented in whom drug interactions or other factors resulted in enhanced metabolism or reduced bioavailability of itraconazole. possibly resulting in diminished clinical efficacy.

PATIENTS AND METHODS
Itraconazole was an investigational drug in Canada until late 1993. In the present study, four patients were identified in whom suboptimal to unmeasurable serum itraconazole levels were documented. All were treated with itraconazole at university teaching hospitals in Vancouver. British Columbia between August 1991 and December 1992. Patients were instructed to take itraconazole capsules with meals. Itraconazole levels in all patients were measured by selective bioassay or high pedormance liquid chromatography (Michael Rinaldi, PhD, Mycology Testing Laboratory, Audie Murphy VA Hospital, San Antonio, Texas), and were obtained in all but one case approximately 4 h after the previous dose. The first level in patient 3 was obtained 18 h after the previous dose on day 14 of itraconazole therapy.
Patients' medical records were examined in detail and tl1e following data extracted: age and sex of patient, underlying disease, fungal infection. site and method of diagnosis, other infections, dosage and route of administration of itraconazole, duration of therapy and timing of serum levels. concu rrent medications with dosages and clinical course.

RESULTS
Patient characteristics and itraconazole levels are summarized in Table 1. Case histories-Case one: A 45-year-old man with AIDS presented with constitutional symptoms, anemia. rash 180 and hepatosple nomegaly. Disseminated histoplasmosis and Mycobacterium avium infection were diagnosed from blood cultures . Remission of the histoplasmosis was induced with a cumulative dose of 1.4 g amphotericin B, followed by maintenance therapy witl1 itraconazole 400 mg/day. The M avium infection was treated with rifampin, ethambutol. clofazamine and ciprofloxacin. The patient presented again three mont11s later with a blood culture-documented relapse of disseminated histoplasmosis and was noted to have extensive tl1rush. A serum itraconazole level was unmeasurable. Itraconazole was discontinued, and he was re-treated with amphotericin B. During amphotericin B maintenance therapy (1 mg/kg once weekly), tl1e patient experienced a second relapse of histoplasmosis . Itraconazole 400 mg/day was restarted, and rifampin was discontinued. No other medications were changed. Three days later, a serum itraconazole level was 1.2 flg / mL. During the subsequent six months the patient remained clinically free from histoplasmosis. Itraconazole was then discontinued because of persistent thrush, which was also poorly responsive to oral fluconazole and topical amphotericin B. He died seven months later of an acute encephalopathy of uncertain etiology. and an autopsy was not perlormed. Case two: A 58-year-old man with AIDS and disseminated Mycobacterium kansasii infection developed biopsy-and culture-proven extrapulmonary histoplasmosis of the right nostril. A cumulative dose of 1 g amphotericin B was associated with resolution of the nasal lesion. This was followed by maintenance therapy with itraconazole 400 mg/clay. The patient was receiving concurrent rifan1pin, isoniazid and ethambutol. Thrush and dermatophytosis developed after seven mont11s of itraconazole therapy. A serum itraconazole level was unmeasurable. Rifampin was discontinued. and after three clays a repeat serum itraconazole level was 0.9 flg/mL. The superficial mycoses resolved. and t11e patient remained free from histoplasmosis relapse until his death from staphylococcal septicemia nine months later. Persistent thrush developed one montl1 before death, and did not respond to fluconazole. Case three: A 31-year-olcl man with AIDS and invasive pulmonary aspergillosis diagnosed by percutaneous needle aspirate was initially treated with an1photericin B 80 mg intravenously tl1ree times weekly, with partial response. Neutrophil count at the onset of symptoms was 0.5x10 9 /L. Itraconazole 400 mg/day was started and amphotericin was continued in addition until adequate absorption was documented by serum levels. The patient was receiving multiple other medications including omeprazole 20 mg/clay and antacids as needed. The highest itraconazole serum level at steady state was 0.42 flg/mL. Despite increase in the itraconazole dose to 800 mg/clay, continuation of amphotericin B and stopping omeprazole and antacids, progressive disease with pulmonary cavitation and intractable chest wall  She did not respond to treatment and died of disseminated aspergillosis two weeks after starting itraconazole. despite achieving persistently nom1al white b lood cell and neutrophil counts. A serum itraconazole level four days after starting treatment was 0.6 J.!g/mL.

DISCUSSI ON
Factors that result in low itraconazole levels do so either by reducing absorption or by increasing hepatic metabolism. ltraconazole is a wealc base that is ionized only under conditions of low pH (17) ; gastric acidity is important for the drug to be soluble and this enhances absorption from the gastrointestinal tract. The absorption of itraconazole from capsules in the fasting state is only 40% of that obtained postprandially (18). Various classes of drugs , such as histamine receptor antagonists , antacids and proton pump inhibitors raise gastric pH and thus impair absorption of itraconazole. Didanosine (ddi) is formulated with buffers to prevent CAN J INFECT DIS VOL 5 No 4 JULY/ AUGUST 1994 its destruction by gastric acid. Coadministration of ddi and itraconazole is not recommended (19), because of the likeliliood that ddi interferes with absorption of the latter. If both drugs are used in a given patient, then itraconazole should be administered at least 2 h before ddi (20). Gastric achlorhydria impairs absorption of ketoconazole (21). Low serum itraconazole levels have been documented in patients with NOS (22), in whom achlorhydria is known to be common (23).
Administration of itraconazole capsules through a nasogastric tube may be associated with low serum levels and treatment failure. Denning eta! (6) described two patients with invasive aspergillosis in whom itraconazole capsules were administered through either a nasogastr;c tube or a gastrostomy tube. One patient had low serum levels and died of progressive disease; the other had a serum level of 0.6 J.!g/mL after five days despite a dose of 600 mg/day. The use of interfering medications s u ch as antacids was not mentioned in these two cases. Reduced absorption was attributed to "agglutination of particles clogging even a large-calibre tube". Infon11ation from the manufacturer suggests that in the nonacid environment of a nasogasmc tube, the coating on individual itraconazole particles may agglutinate to form an insoluble complex that cannot be absorbed, regardless of any mechanical clogging of the tube (personal communication) . If itraconazole must be administered through a nasogastric tube, the suspension fonn of the drug is recommended .
Itraconazole undergoes hepatic metabolism; little is excreted unchanged in urine (17,24). Several drugs that induce hepatic metabolism have been noted to decrease itraconazole levels; these include rifampin, phenytoin, carbamazepine and possibly phenobarbital (7 , 9,16,25,26). The interaction with rifampin may be problematic in AIDS patients, who frequently have con-current mycobacteria l and fungal infections. The interaction wilh anliconvu lsants may complicate treatment of fungal infections in patients with seizure disorders .
The main factor responsible for undcleclable ilraconazole levels in cases 1 and 2 appeared to be the coadm inistration of rifamp in . Drug-induced achlorhydria and possibly nasogaslric tube administration contributed lo the low serum level in case 4. Low serum ilraconazole levels associated with AJDS likely accounted for the results obtained in case 3. because levels did not rise significanlly despite an increase in the ilraconazole dose to 800 mg/day a nd elimination of the omeprazolc and antacids. Poor compliance is unlikely to have caused low serum ilraconazole levels in the cases presented. Cases 1 and 2 denied noncompliance . and serum levels rose promplly after rifampin was stopped. Cases 3 and 4 had levels measured while receiving ilraconazo le in hospital.
Limited data a re available regarding optimal serum itraconazole levels. Although there was s ignificant interindividual variation, the mean peak serum level on 400 mg/day (standard dose for invasive mycoses) at steady slate was 1.981-!g/mL in healthy volunteers (24) . Leve ls a re approximately 50% lower in patients with AIDS (22) . Denning et al (7) found that treatment failure of invasive aspergillosis was associated with steady stale itraconazole levels below 7 pmol/L (5 pg/mL) .
The consequences of low serum itraconazole levels due to enhanced metabolism are variable. Tucker el al ( 16) described five palien ts with C!yplococcosis and mycobacterial disease treated with itraconazole and rifampin . Four patients responded despite decreased itraconazole levels. a lthough two had a worsening of seborrheic dermatitis: one patient had worsening c!yplococcal disease. The combination of itraconazole and rifampin was synergistic against cryptococcus in vitro. Reduced serum itraconazole leve ls have been assoc iated with treatment failures against other fungal infections (3.7.9.16.25.26) and failures of prophylaxis (2 7) . In our patients. low ilraconazole levels were associated with a re lapse of disseminated histoplasmosis in one of two patients and development of superficial mycoses in both . The superficial mycoses resolved promplly in lhe palienl whose rifampin was stopped. Two patients with invasive aspergillosis failed to respond to itraconazole. a lthough neither had previously responded to amphotericin B. and one died within two weeks of starling ilraconazole .
Certainly. factors other than low itraconazole levels likely played a role in treatment failure of the cases presented. including severity of immunocompromised stale and underlying disease, and possibly the development of azole resistance, which has been well documented in human immunodeficiency virus-related thrush (28). There are no controlled trials of itraconazole in invasive aspergillosis (6, 7). and thus its efficacy is not clearly defined. Both patients with invasive aspergillosis received concomitant itraconazole and amphotericin B while ilraconazole levels were pending. Amphotericin B was to be continued in these patients unlil adequate oral absorption of itraconazole was documented by serum levels. Although the mechanisms of action of polyenes and azoles suggest potential antagonism (29). the clinical significance is unknown. Recenlly in an aspergillosis animal model, Patterson et al (30) showed no evidence of antagonism with amphotericin plus fluconazole combination therapy.

CONCLUSIONS
We have described four patients in whom suboptimal to unmeasurable serum ilraconazole levels were documented. Contributing factors included: concurrent therapy with drugs known to enhance itraconazole metabolism or drugs raising gastric pH: and possibly the administration of the contents of itraconazole capsules through a nasogastric tube. Low levels were associated with relapse of disseminated histoplasmosis. development of superficial mycoses and treatment failure in invasive aspergillosis. We recommend lhal serum ilraconazole levels be routinely measured when treating serious fungal infections in patients in whom potential interfering factors are identified .