High dose fluconazole in salvage therapy for HIV-uninfected cryptococcal meningitis

This is a retrospective cohort study conducted among HIV-uninfected CM patients from January 2011 to December 2016 at Huashan Hospital, Fudan University (a tertiary health care center in Shanghai, China, with approximately 1200 hospital beds and 60,000 admissions per year). Patients who had been refractory to or intolerant of prior antifungal drugs and who then switched to salvage therapy with fluconazole 800 mg/day were included. This study was reviewed and approved by the local medical ethics committee.

A proven diagnosis of CM was made if the patient met any of the following criteria: (1) positive culture of Cryptococcus from cerebrospinal fluid (CSF), (2) positive CSF ink smear, (3) cryptococcal capsular polysaccharide antigen detected in CSF on cryptococcal antigen (CrAg) lateral flow assay (IMMY, Inc., Norman, Oklahoma, USA) in CSF or (4) compatible histopathological findings, which are 5–10 μm encapsulated yeasts in brain tissue. Refractory CM was considered if any 2 of the following conditions were present after an adequate period of antifungal therapy (≥14 days): (1) persistently positive cultures of Cryptococcus in CSF, (2) deterioration of clinical signs and symptoms of disease, (3) new sites of disease or worsening of pre-existing lesions radiologically, and (4) decreasing level of glucose and increasing level of protein in CSF [21, 22]. Patients were considered as intolerance to antifungal drugs if they suffered severe or life-threatening toxicity from amphotericin B (AmB) based initial therapy [22].

Patients treated with high dose fluconazole for ≥1 week were included for efficacy evaluation. Response to salvage therapy was assessed at 2 weeks after initiation of salvage therapy, at the end of salvage therapy, and at the end of antifungal therapy. Efficacy of antifungal treatment was categorized as success (complete or partial response) or failure (stable response, disease progression, or death during the study period regardless of any cause) on the basis of clinical, radiological, and microbiological data according to previous criteria [21]. Adverse events (AEs) occurring during salvage therapy were recorded. The relationships of AEs and high dose fluconazole were evaluated with Naranjo probability scale [23]. High dose fluconazole related AEs were defined as one with a possible, probable or certain relationship.

In total, 44 patients with proven CM were included in our study. Thirty-two also had pulmonary involvement and 6 cryptococcemia. The median age was 44 years, and 63.6% were male. Predisposing factors co-existed in 16 patients (36%). Detailed information is summarized in Table 1. Headache (100%) was the most common symptom, and 72.7% of them presented with fever, vomiting and meningeal irritation. Thirty-four patients presented with further neurological symptoms and signs, including altered mental status, epileptic seizures, cranial nerve deficits, limb weakness, uracratia and dysuria (Table 1).

Cranial magnetic resonance imaging (MRI) was performed in all patients, with 41/44 (93.2%) yielding abnormal results. Local parenchymal lesions (86.4%) were most frequently observed, and common sites included frontal lobe (68.2%), parietal lobe (59.1%), periventricular region (29.5%) and basal ganglion (22.7%). Other abnormalities included meningeal enhancement (25 patients; 56.8%) and ventricular enlargement (12 patients; 27.3%). Most patients had 2 or more sites of involvement. Abnormalities on chest computer tomography (CT) scans were found in 32 of 44 patients (72.3%). The common findings were nodules and masses (47.7%), patches and strips (43.9%), pleural thickening (22.7%), pleural effusion (20.5%), lymph node enlargement (6.8%), air bronchogram sign (4.5%) and ground glass attenuation (2.3%). Some patients had mixed types of lesion on chest CT.

Before initiating salvage therapy, CSF examinations were performed in 43 patients including 42 samples from lumbar punctures, 1 from continuous lumbar cerebrospinal fluid drainage. CSF cultures were positive for Cryptococcus in 4 patients, and ink smears were positive in 18 patients. CSF CrAg tests were positive in all 43 patients, and 47.7% of them (21/43) had CrAg titer of at least 1:640. Median white blood cell count (WBC) of CSF was 29 /mm³ (range, 0–550 /mm³). Median levels of CSF protein and glucose were 824 mg/dL and 34 mg/dL, respectively. Of the 42 patients who had lumbar punctures, 21 had CSF opening pressure above 25 cm H₂O.

As listed in Fig. 1, 19 patients (43.2%) were refractory to initial antifungal therapy, 8 (18.2%) were refractory to consolidation therapy, and 17 (38.6%) were intolerant of prior antifungals. AmB-based regimens were initially administered in all 44 patients, and daily dosage of AmB ranged from 20 to 40 mg (0.4–0.7 mg/kg/day). Median cumulative dose and duration of AmB were 1205 mg (range, 205–5775 mg) and 48.5 days (range, 9–212 days).

Of the 19 patients refractory to initial therapy, 11 received high dose fluconazole in combination with AmB and flucytosine for salvage, 6 received high dose fluconazole and flucytosine, and the remaining 2 received high dose fluconazole alone. For patients who were refractory to consolidation therapy, fluconazole dosage was increased from 400 mg/day to 800 mg/day and flucytosine co-administered in all but one patient. Of the 17 antifungal-intolerant patients, 10 received high dose fluconazole alone, and 7 others were treated with high dose fluconazole and flucytosine. Overall, triple therapy of high dose fluconazole, AmB and flucytosine was administered in 11 patients, 20 received dual therapy of high dose fluconazole and flucytosine, 13 received monotherapy of high dose fluconazole. Median duration of high dose fluconazole in salvage therapy for the 44 patients was 136.5 days (range, 1–667 days). Among the 31 patients (70.5%) receiving flucytosine, 27 were treated with flucytosine throughout the course, and the remaining 4 discontinued the drug for suspected side effects (3 with anemia, 1 with elevated transaminase). Median duration of flucytosine in salvage therapy for the 31 patients was 142 days (range, 22–561 days).

Except 1 patient who received high dose fluconazole less than 7 days and died of multiple organ failure (MOF), the other 43 patients were all included into efficacy evaluation. Two weeks after salvage therapy, partial response in 31 patients (72.1%), stable response in 11 patients (25.6%), and death caused by brain hernia in 1 patient (2.3%) were observed. At the end of salvage therapy, 3 patients (7.0%) achieved complete response, 33 (76.7%) achieved partial response, 2 (4.7%) achieved stable response, and 5 (11.6%) died. The overall efficacy rate was 83.7%. Of the 2 stable patients, 1 discontinued fluconazole, reinstituted with AmB and flucytosine and achieved partial response; the other was lost for follow-up 40 days after initiation of salvage therapy. Of the 4 patients who died after 2 weeks of salvage therapy, 1 was considered to have died of CM, 2 died from secondary infection and MOF, and 1 died from aggravation of lymphoma. Thirty-seven of 43 patients (86.0%) achieved clinical success and stopped antifungal drugs, 5 (11.6%) died and 1 (2.3%) was lost for follow-up. As shown in Table 2, we compared the response rates between patients treated with high dose fluconazole with or without flucytosie and those with combinations of AmB and high dose fluconazole with or without flucytosine. Although a significant higher 2-week response rate was observed in the triple combination group (62.5% vs. 100%, P = 0.019), there was no apparent differences in efficacy at the end of salvage therapy (84.4% vs. 81.8%, P >  0.999) and the end of antifungal therapy (84.4% vs. 90.9%, P >  0.999). At 1-year follow-up after stopping antifungal drugs, all of the 37 successfully treated patients survived, and no relapses were observed.

All 44 patients were included in the safety assessment. Possible high-dose fluconazole related AEs were found in 54.5% (24/44) of them. No grade 3 or 4 AEs were observed. As shown in Table 3, the most frequent AE was elevated transaminase level in 11 patients. Transient grade 2 neutropenia and grade 1 thrombocytopenia occurred in 1 patient treated with fluconazole monotherapy. One patient presented with grade 2 anemia in the first week of salvage therapy, but his hemoglobin level recovered without discontinuance of fluconazole or flucytosine. Hypokalemia occurred in 3 patients, all of whom returned to normal condition after potassium supplementation (AmB were not co-administrated). There were no increased creatinine levels related to high dose fluconazole in our cases.

In terms of clinical manifestations, gastrointestinal side effects were common complaints. Skin lesions were documented in 2 patients. Photosensitive dermatitis occurred in 1 patient soon after initiation of fluconazole, and improved without fluconazole discontinuance. Rash developed in another patient several weeks after the salvage therapy and lasted until CM recovered and antifungal treatment finished. We also observed 1 patient with elevated D-dimer and deep venous thrombosis which was relieved after treatment with oral warfarin. None of these patients discontinued high dose fluconazole due to intolerance.

From the beginning of salvage therapy, 1-year mortality related to all cause was 13.6% (6 of 44 patients) with no significant difference among 3 treatment groups (Fig. 2, P = 0.258). We assessed factors with a potential impact on all-cause mortality by means of univariate and multivariate analyses. In the univariate model, factors significantly associated with 1-year mortality were age ≥ 60 years (P = 0.001); serum albumin level < 35 g/L (P = 0.001); estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73m² (P = 0.015); and CSF WBC count < 10 /mm³ (P = 0.002). CSF CrAg titer ≥1280 was more frequently presented in mortality group but it did not reach a statistic significance (P = 0.058). By multivariate analysis, as shown in Table 4, factors significantly correlated with decreased survival in the overall population were serum albumin level < 35 g/L (P = 0.014) and CSF CrAg titer ≥1280 (P = 0.048).