The pharmacokinetics, efficacy, and safety of intravenous (iv) itraconazole (2 days at 400 mg/day, 12 days at 200 mg/day), followed by 12 weeks of oral capsules (400 mg/day) were studied in 31 immunocompromised patients with pulmonary invasive aspergillosis. All patients received iv itraconazole (median duration, 14 days), and 26 then received oral itraconazole (median duration, 78.5 days). After receiving iv itraconazole, concentrations increased rapidly, with trough plasma levels ⩾250 ng/mL in 91% of patients and in all patients by day 7. Concentrations ⩾500 ng/mL were observed in 64% of patients by day 2. Mean trough concentrations after 2 and 14 days were 670 and 850 ng/mL, respectively. Therapeutic levels were maintained after switching to oral capsules. A complete or partial response was seen at the last on-treatment assessment in 15 (48%) of 31 patients, with 6 (19%) showing stable disease. Itraconazole was well tolerated, with no unexpected effects. Overall iv/oral itraconazole was safe and effective in invasive aspergillosis.
Invasive fungal infections have increased in incidence over the past 30 years and are now among the most frequent causes of serious morbidity and mortality in patients with neutropenia [1, 2]. For example, evidence of invasive fungal infection at autopsy has been found in 20%–50% of patients with hematologic malignancies [3, 4]. The causative organisms are usually Candida species, but aspergillosis is often more common in bone marrow transplant (BMT) recipients and other patients with severe neutropenia. In 1 study in a BMT unit, Aspergillus species was isolated from 36% of patients with nosocomial pneumonia [5]. The attributable mortality rate of invasive aspergillosis is typically very high and has been estimated at 13%–95%, with higher rates found in patients undergoing allogeneic BMTs [5]. Pulmonary aspergillosis is generally acquired by inhalation [6]. Sinus and rhinocerebral involvement is generally seen in only 5%–10% of cases. Infection is frequently associated with vascular events, such as thrombosis or infarction of surrounding tissue and pulmonary or cerebral hemorrhage. Dissemination to deep organs may also occur [7, 8].
Amphotericin B is considered to be the “gold standard” for the treatment of systemic fungal infections; however, it is associated with significant toxicity and requires iv administration. Nausea, fever, and chills are common during iv infusion of amphotericin B, and renal dysfunction may be severe enough to necessitate withdrawal of treatment [9]. Triazole antifungal drugs are currently one of the alternatives to amphotericin B [10], but they vary in their spectrum of activity and pharmacokinetic properties. Itraconazole has a broad spectrum of activity against both Aspergillus species and Candida species and is well tolerated, with a favorable safety profile [11]. Open-label studies of treatment with itraconazole capsules have demonstrated high response rates in solid-organ transplant recipients and patients with pulmonary aspergillosis and neutropenia [12, 13]. A small comparative study of oral itraconazole (capsules, 400 mg/day) and parenteral amphotericin B (0.6 mg/kg daily) revealed greater efficacy of oral itraconazole against Aspergillus species infections [14].
Many severely immunocompromised patients, however, are unable to tolerate oral therapy. In addition, the gastrointestinal absorption of the capsule formulation of itraconazole is often poor and variable in severely ill patients [15, 16]. To overcome these limitations, an iv formulation has been developed by using a 40% hydroxypropyl-β-cyclodextrin (HP-β-CD) solution to increase the solubility of itraconazole. In this study, the proposed dosage regimen (2 days of iv itraconazole 400 mg/day, 12 days of iv itraconazole 200 mg/day, and then 12 weeks of oral itraconazole capsules, 400 mg/day) aimed to rapidly produce therapeutic concentrations of itraconazole in the blood; these concentrations could then be maintained with oral follow-up therapy. The objectives of this study were to determine the efficacy and safety of this iv/oral itraconazole dosing regimen in the treatment of pulmonary aspergillosis in immunocompromised patients and to evaluate plasma concentrations of itraconazole.
Male and female patients with active invasive pulmonary aspergillosis were enrolled at 13 centers worldwide. Invasive pulmonary aspergillosis was classified according to at least 1 of the criteria in table 1.
Number of patients demonstrating a complete response, partial response, stable disease, progressive disease, or treatment failure during itraconazole therapy.
Number of patients demonstrating mild, moderate, or severe clinical symptoms at baseline and at study endpoint.
Patients receiving terfenadine, astemizole, phenytoin, phenobarbital, rifampicin, oral midazolam, triazolam, cisapride, or hydroxymethylglutaryl coenzyme A reductase inhibitors or who had received phenytoin, phenobarbital, or rifampicin within the 2-week period before study entry were excluded. Other exclusion criteria were liver disease (serum glutamic-oxaloacetate transaminase or serum glutamate pyruvate transaminase ⩾5 times the upper limit of normal or bilirubin ⩾50 mM), renal insufficiency (creatinine clearance, <30 mL/min), administration of investigational drugs (other than anticancer regimens) within a 1-month period before study entry, hypersensitivity to azole antifungal drugs, definitive evidence of cerebral aspergillosis, or intent to surgically resect the only evaluable pulmonary lesion within 7 days. In addition, patients were excluded if the diagnosis of proven aspergillosis had been made >3 days before entry into the trial.
Study design. The study was an open, international, multicenter trial designed to investigate the efficacy and safety of iv followed by oral itraconazole capsules in the treatment of patients with invasive pulmonary aspergillosis. Plasma concentrations of itraconazole, hydroxy-itraconazole, and HP-β-CD were determined. Patients received 200 mg itraconazole by an iv infusion in a 40% HP-β-CD solution over 60–90 min every 12 h for the first 2 days. For the following 12 days, 200 mg itraconazole was administered by iv infusion once daily. Oral itraconazole capsules (200-mg) was then administered twice daily (morning and evening with a meal) from weeks 3–14.
With the exception of topical agents, oral (nonabsorbable) amphotericin B and nystatin antifungal treatments were not permitted during the study period. Because itraconazole can inhibit the metabolism of certain drugs metabolized by the cytochrome 3A4 family (cyclosporin A, digoxin, oral anticoagulants, systemic methylprednisolone, vinca-alkaloids, and tacrolimus), their dosages were reduced during the study if necessary. Surgical resection of invasive pulmonary aspergillosis was available to all deteriorating patients in this study for ethical reasons; however, whenever possible, the surgery was not performed until 7 days after study entry. In all patients for whom surgery was performed, a CT scan of the chest was performed within the 24-h period before surgery, and specimens were processed for fungal culture and histologic diagnosis. Patients who underwent surgery remained in the study.
Study assessments. At study entry, a complete clinical evaluation was made of the patient's condition, and relevant medical and surgical history was taken. Clinical signs and symptoms, including productive and nonproductive cough, pleuritic chest pain, hemoptysis, and others, were scored according to severity. Neutrophil counts were performed at baseline and during the study, and all patients underwent a chest X-ray within 24 h before study entry and a CT scan within 48 h after study entry, as well as on day 7, at the end of iv therapy (day 14), and at the end of oral therapy (week 14). Circumscribed lesions were measured. All patients also underwent BAL at study entry, and the fluid was processed for cytologic examination and culture to detect Aspergillus species Biopsy was performed on any abnormal areas identified unless prevented by thrombocytopenia. Samples from the anterior nares and sputum were cultured for fungal growth, and plasma samples were taken for Aspergillus antigen detection. These tests were performed before administration of itraconazole, on days 7 and 14, and at the end of weeks 8 and 14.
Safety assessments. Each patient was evaluated for adverse events throughout the study. All adverse events and laboratory abnormalities were recorded. Laboratory measurements were performed before study entry, on days 2, 7, and 14, and at the end of the study. These included hematology, blood biochemistry, and urinalysis assessments. Creatinine clearance was calculated on days 0, 4, 7, 11, and 14 (or at the end of the iv administration). In addition, patients were instructed to immediately contact the treatment center if adverse events suggestive of hepatic dysfunction were experienced (dark urine, jaundice, or persistent nausea and vomiting).
Pharmacokinetic monitoring. Venous blood samples (10 mL) were obtained immediately before the first infusion of iv itraconazole (at baseline) and at all subsequent visits immediately before the administration of the study medication. Blood samples collected were centrifuged (10 at 1000 g) within 2 h of collection. Separated plasma was stored at <-18°C before assay. Plasma concentrations of itraconazole and hydroxy-itraconazole were determined by HPLC with ultraviolet detection (limit of quantification, 100 ng/mL) [17, 18]. Plasma concentrations of HP-β-CD were determined by size-exclusion chromatography with postcolumn complexation (limit of quantification 5–50 µg/mL, depending on the available plasma volume) [18].
Efficacy parameters. An assessment of response to therapy was performed on day 7, at the end of iv therapy (day 14), at week 8, and at the end of oral therapy (week 14). The primary analysis was an intent-to-treat analysis (ITT), which included all entered patients with at least 1 drug administration. The primary outcomes were the overall response (the number of patients who showed a complete or partial response) in the ITT population at the end of iv treatment and at the end of follow-up (study endpoint). Because invasive aspergillosis without antifungal treatment and some resolution of neutropenia is always a progressive disease, an overall evaluation of stable disease was considered a positive effect of treatment. Therefore, the incidence of patients with a complete or partial response or stable disease was recorded.
The overall response to treatment was defined according to the following criteria. “Complete response” was defined as resolution of all attributable symptoms, signs, and radiographic or bronchoscopic abnormalities present at enrollment. “Partial response” was defined as major improvement (usually nearly complete) in attributable symptoms, signs, and radiographic or bronchoscopic abnormalities present at enrollment. “Stable disease” was considered to be minor or no improvement in attributable symptoms, signs, and radiographic or bronchoscopic abnormalities, but patient continued on therapy without deterioration. “Progressing disease” was defined as new disease sites or original disease worsened, but patient continued study medication. “Failure” was considered deterioration in attributable clinical or radiographic abnormalities, necessitating alternative antifungal therapy or resulting in death.
The secondary efficacy variables were the median time to achieve a response and microbiological results from the anterior nares and sputum fungal culture analyses. Exploratory analyses were also performed to examine a possible effect of a number of demographic and baseline characteristics on response. The factors chosen were based on a publication by the National Institute of Allergy and Infectious Diseases (NIAID) Mycoses Study Group [19].
Statistical analyses. The primary analysis (overall response) was based on the ITT population that included all subjects who received at least 1 dose of itraconazole. Descriptive statistics were performed for all demographic and baseline data, and were calculated at all time points for the neutrophil count and changes in symptoms. Changes in each group were analyzed with the Friedman test for exploratory purposes [20]. Time to response was calculated with the Kaplan-Meier analysis [21].
Descriptive statistics (mean ± SD, median, minimum-maximum) were calculated for the trough serum levels of itraconazole, hydroxy-itraconazole, and HP-β-CD at days 0, 2, 7, and 14 and the end of weeks 8 and 14. For creatinine and clearance of creatinine, descriptive statistics and changes from baseline were calculated.
Study population and treatments. A total of 31 patients were enrolled at 13 sites in 5 countries (Australia, Canada, France, Greece, and the United Kingdom). Of the 31 patients enrolled, 20 (65%) were men, and the median age was 48 years (table 2). Twenty-seven patients (87%) had hematologic malignancies, and the major predisposing risk factor was neutropenia (<0.5 × 109/L; 19 patients [61%]; table 2).
Nineteen patients (61%) had received ⩾1 drugs for antifungal prophylaxis before study entry (amphotericin B, n = 14; lipid formulations of amphotericin B, n = 1; fluconazole, n = 5; itraconazole, n = 2; nystatin, n = 1). Prior empiric antifungal treatment had been administered to 14 patients (45%; amphotericin B, n = 10; lipid formulations of amphotericin B, n = 2; fluconazole, n = 3; itraconazole, n = 3), and 4 patients had received prior antifungal therapy for a proven infection (amphotericin B, n = 2; fluconazole, n = 1; itraconazole, n = 1; table 2). Concomitant iv or oral medications were given to 29 (94%) subjects (table 2). These included antibiotics, analgesics, diuretics, plasma substitutes, perfusion solutions, corticosteroids, antacids, and drugs for the treatment of peptic ulcers, antihistamines, psycholeptics, and oral antivirals.
During the study, 31 patients received iv itraconazole for a median duration of 14 days (range, 4–28 days). Among these, 26 patients were then switched to oral itraconazole capsules for a median duration of 78.5 days (range, 1–90 days).
Pharmacokinetic evaluation. Plasma concentration data were available for 22 subjects, although itraconazole and hydroxy-itraconazole were detectable before the start of treatment in 7 of 22 patients. According to their medication history, 5 of these patients had received itraconazole before inclusion in the study; however, detectable levels in the other 2 patients remain unexplained and for 1 patient, the levels were very low and close to the limits of detection. Summary statistics were therefore calculated at time 0 for the patients who had no detectable levels at the start of treatment (table 3). The median trough itraconazole plasma concentration was 620 ng/mL after 2 days of iv treatment and increased to 644 and 1000 ng/mL after 7 and 14 days of iv therapy, respectively. In total, 20 (91%) of 22 patients had concentrations of itraconazole ⩾250 ng/mL by day 2, and 14 (64%) of 22 patients had concentrations ⩾500 ng/mL. The average ratio of hydroxy-itraconazole:itraconazole was 1.41 ± 0.62 during the iv phase of the study. In the oral capsule follow-up phase, the trough plasma concentrations of itraconazole continued to increase, reaching 2877 ng/mL after 8 weeks, whereas the ratio of hydroxy-itraconazole:itraconazole decreased to 1.32 ± 0.31 during the oral capsule phase of the study.
Trough plasma concentrations of itraconazole, hydroxy-itraconazole, and hydroxypropyl-β-cyclodextrin (excluding subjects with quantifiable predose levels).
HP-β-CD levels could be quantified only during the iv phase of the study, and a high degree of variability was seen. HP-β-CD concentrations were quantifiable in 19 (86%) of 22 patients tested after the loading doses, whereas concentrations were detectable in only 6 of 20 patients tested after 7 days and 7 of 16 patients tested on day 14 (table 4). There was no sign of HP-β-CD accumulation in this patient population. The highest mean concentration (89.9 µg/mL) was achieved after the initial 2-day loading doses, when the iv infusions were given twice daily. After this time, mean HP-β-CD levels continued to decrease during dosing to 18.8 µg/mL at day 14.
Efficacy analysis. The primary efficacy parameter was the global response to treatment with itraconazole at the end of each treatment period (figure 1). At the end of the iv treatment period (day 14), 10 (32.3%) of 31 patients had experienced a complete or partial response to itraconazole treatment, and by the end of oral capsule treatment (week 14), 12 (38.7%) of 31 patients showed a complete (n = 3) or partial (n = 9) response. At the study endpoint (last on treatment assessment), 15 (48%) of 31 patients had experienced a complete (n = 8) or partial (n = 7) response to itraconazole treatment. Overall, 18 (58%) of 31 experienced a complete or partial response at any point during the trial. The estimated time to a response (complete or partial) was attained over a median time period of 55 days (95% CI, 14–57 days). Four of the 10 subjects who experienced at least 1 episode of neutropenia (absolute neutrophil count <0.5 × 109 neutrophils/L) during the trial completely or partially responded to treatment at study endpoint.
When stable disease was also considered a positive response to treatment, the success rate was 67.7% (21 of 31 patients) at day 14 after the iv treatment period and 45.2% (14 of 31 patients) at the end of the oral phase. At the study endpoint (last on-treatment assessment), 21 (68%) of 31 had responded or achieved stable disease. In addition, 27 (87%) of 31 patients achieved a complete response, partial response, or stable disease at any time point during the study.
Six (19.4%) patients died during the study. The estimated time to death was 92 days in the first quartile of subjects. Of the 6 patients who received <2 weeks of itraconazole therapy, 1 had a partial response, 1 had stable disease, 2 showed signs of progressive disease, and 2 were did not respond to treatment (they died), whereas 77% of patients treated for at least 2 weeks showed a complete response, a partial response, or stable disease.
Cultures obtained from various body sites (mainly BAL fluid or sputum) were positive for Aspergillus species in 14 of 31 patients (19 positive cultures, including BAL, in 6 patients and sputum in 6 patients) at baseline (table 5). By the end of the study, positive cultures had been performed from a further 3 of 31 patients. At week 8, however, only 4 cultures were positive for Aspergillus species, and 1 culture was positive for Geotrichum candidum (table 5). No positive culture was performed after this time point.
Microbiological culture results at baseline and after treatment of 14 patients with itraconazole.
Surgical excision of pulmonary lesions was performed in 7 patients during the study and in 2 of these patients on 2 separate occasions. At baseline, these patients experienced severe signs or symptoms that were potentially attributable to pulmonary fungal infection (figure 2). All signs and symptoms improved during the study, with only 1 severe rating noted at the last on-treatment assessment (hemoptysis). The improvements seen by day 14 for productive cough and pleuritic chest pain were statistically significant (P < .001).
Safety analysis. The median duration of treatment was 45 days, with a range of 4–104 days. For iv therapy, the median duration of treatment was 14 days, (range, 4–28 days), and the median duration of oral capsule treatment was 78.5 days (range, 1–90 days).
Overall, iv and oral itraconazole were well tolerated. Adverse events occurred in 28 (90%) of 31 of patients during iv therapy, but most were mild-to-moderate in nature. The most frequent adverse events were fever (7 events [23%]), dyspnea (6 [19%]), diarrhea (5 [16%]), pulmonary edema (5 [16%]), rash (5 [16%]), cough (4 [13%]), fatigue (4 [13%]), and nausea and vomiting (4 [13%]). However, only 2 events were considered definitely related to the study drug: rash in 1 patient and rigors during drug infusion in another. Thirteen patients (42%) experienced adverse events possibly related to the study drug. The main events were diarrhea (10%), fever (10%), increased nonprotein nitrogen (6%), and nausea (6%). Two (6% overall) of these 13 patients were withdrawn from iv therapy because of these events (1 because of increased nonprotein nitrogen and 1 with decreased creatinine clearance). Two others were considered serious but did not require withdrawal. Three patients (30%) died during the iv phase of the study: 1 of gastrointestinal hemorrhage, 1 of myocardial infarction, and 1 of fungal infection. None of the deaths was thought be related to study medication.
No consistent clinically relevant changes in laboratory parameters were observed. There was a tendency toward a slight decrease in plasma creatinine during iv treatment. For example, the mean plasma creatinine level decreased from 97.01 µM at baseline to 78.68 µM at day 14. There were no consistent changes in any of the urine parameters.
During the oral capsule treatment phase, adverse events occurred in 19 (73%) of 26 patients. Again, most adverse events were mild-to-moderate in severity. The adverse events most frequently reported (>10%) were nausea, fungal infection, fever, rash, injury, and pneumonia. Nine patients (35%) experienced adverse events that were considered possibly related to treatment during the oral capsule phase. The most frequent of these events were diarrhea, fever, increase in nonprotein nitrogen, and nausea. Treatment was withdrawn in 7 patients (22%) because of adverse events during oral capsule therapy. Of the 7 patients withdrawn from oral capsule itraconazole treatment, the adverse events in 4 of the patients (fungal infection [n = 2], increased liver enzymes [n = 1], and abnormal gait/hemoptysis [n = 1]) were considered by the investigators to be possibly related to the administration of study medication. In 3 patients, the adverse events were considered severe but not related to the study medication. Serious adverse events that did not lead to death were reported in 10 patients (during oral capsule treatment). The nonfatal serious adverse events that each occurred in 2 patients in the oral capsule treatment phase included fever, fungal infection, and hemoptysis. Adverse events leading to death during the oral capsule treatment period occurred in 3 subjects: condition aggravated, hepatic failure, and circulatory failure and pneumonia. These were not related to the study medication.
The optimal approach to treatment of invasive pulmonary aspergillosis has yet to be determined. Amphotericin B, the “gold standard” treatment for aspergillosis, is associated with significant toxicity. Even in the best of circumstances, responses to amphotericin B are limited, with only ∼20%–30% of patients responding in most clinical situations [12]. Higher doses may be useful but are often limited by the toxicity of amphotericin B. More recent studies have shown AmBisome (Gilead Sciences) to be an effective treatment of invasive aspergillosis, with similar efficacy rates to the parent compound but with less nephrotoxicity [22,23-24]. However, in 2-year review, Fisher et al. [25] suggested that the use of AmBisome in the treatment of confirmed aspergillosis in liver transplant recipients was of uncertain value. Overall, 4 of 5 patients with confirmed aspergillosis died despite treatment with AmBisome. In addition, the cost of AmBisome remains high and may represent a limiting factor to its widespread use. Therefore, the recent introduction of new formulations of itraconazole (oral solution and iv) with more favorable safety profiles has expanded the therapeutic choice for the treatment of invasive aspergillosis.
One of the aims of this study was to achieve therapeutic plasma concentrations of itraconazole as rapidly as possible by using a loading dose of 200 mg itraconazole q12h for the first 2 days of treatment. Data from this study confirm that after 2 days, 91% of patients assessed (20 of 22) achieved trough plasma concentrations of itraconazole (>250 ng/mL), which is considered to be the minimum level required for efficacy [26]. Itraconazole trough concentrations of 250 ng/mL were achieved for all patients at 1 week of therapy and were maintained during the oral capsule follow-up phase. The mean predose concentrations achieved after 2 days of iv treatment were similar or slightly higher than those previously reported [27], as were the concentrations achieved at the end of iv and oral capsule therapy. Trough concentrations of itraconazole and hydroxy-itraconazole in plasma increased up to 8 weeks and reached stable levels of 3000–4000 ng/mL.
Low levels of HP-β-CD were detectable only during the iv phase of treatment in a few patients; this was expected, because it is not present in the capsule formulation of itraconazole. The highest mean HP-β-CD concentrations (90 µg/mL) were observed during the initial 2 days of high-dose iv therapy, with lower mean concentrations obtained after 7 and 14 days. Acute and chronic toxicity studies have shown that animals can tolerate very large doses (up to 800-fold the dose in humans) of HP-β-CD administered orally or parenterally. Histologic observations in mice, rats, and dogs have mainly found minimal changes in the epithelial cells of the urinary bladder, kidney tubular cells, and liver [28, 29]. These changes were reversible and were not considered to represent a toxic effect of HP-β-CD. Extensive testing of HP-β-CD in humans has also shown no significant toxicity, other than occasional mild diarrhea when it is taken by mouth [30]. The half-life of HP-β-CD is ∼2.5 h in healthy volunteers, and 90% of it is cleared within ∼12 h of dosing; therefore, no accumulation beyond 14 days is expected in patients with normal renal function.
The majority of patients enrolled in this study had a hematologic malignancy, with neutropenia as the main predisposing factor. During the study, one-third of the patients assessed (10 of 30) experienced at least 1 episode of neutropenia (<0.5 × 109/L). Despite this, 48% of patients with invasive pulmonary aspergillosis showed a complete or partial response to itraconazole at their last on-treatment assessment. In addition, infection was stabilized in another 20% of the study population. The level of response to treatment increased as the study progressed. For example, no patient demonstrated a complete response after the first week of treatment, and indeed, progression of disease or not responding to treatment was observed in 20% of patients. However, at the study endpoint, 68% of patients showed either a complete (25.8%) or partial response (22.6%) or stabilization of the infection (19.4%). This suggests that, if treatment can be continued for at least 14 weeks, there is a high chance of a positive response. Eighteen patients (58%) experienced a complete or partial response to itraconazole treatment at some time during the study, and the median time to achieve this was 55 days. When stable disease was included as a response to treatment, 87% of patients responded to itraconazole treatment at some time during the study.
In invasive aspergillosis, assessing the response to therapy is always problematic. Previously published data from Europe have shown response rates of 63%–73% for aspergillosis treated with oral capsule itraconazole [31, 32]. In addition, a multicenter study conducted by the NIAID Mycoses Study Group reported an overall response rate of 39% for aspergillosis [32], although patients with pulmonary aspergillosis responded better than patients with disseminated aspergillosis. In this study, despite the higher risk of these patients than in previous studies [31, 32], the outcome was relatively good and similar to that observed by the National Mycosis Study Group [33].
Overall, itraconazole was well tolerated and was associated with a favorable safety profile in these severely ill patients. Adverse events occurred in 90% of patients during iv itraconazole treatment and in 73% during oral capsule treatment; however, the majority of these events were considered related to the underlying disease. Adverse events that were considered definitely related to itraconazole treatment were reported in only 2 patients. Itraconazole treatment was permanently discontinued in 11 patients because of adverse events, of which 5 events were possibly related to treatment. None of these adverse events was unexpected in view of the safety profile of itraconazole and the disease status of the patient population. It was therefore concluded that this regimen of itraconazole was safe, well tolerated, and effective in patients with invasive pulmonary aspergillosis. These results are particularly encouraging considering that amphotericin B is associated with considerable toxicity, which may result in the withdrawal of treatment or the use of subtherapeutic dosing regimens.
All patients provided written, informed consent. The study was approved by ethics committees at all participating centers.
IDSA Members: For your free access to this journal, log in via the IDSA members area.
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy