A Novel Encochleated Formulation Improves Atovaquone Activity in a Murine Model of Pneumocystis Pneumonia

Abstract

Although atovaquone is effective in treating and preventing Pneumocystis pneumonia (PCP), its use is limited by nonlinear absorption and adverse events. The current study was undertaken to examine the activity of encochleated atovaquone (eATQ), a novel lipid-crystal nanoparticle formulation, in a mouse model of PCP. eATQ 100–200 mg was superior to commercially available atovaquone at 14 days in decreasing total Pneumocystis nuclei and asci. eATQ plus anidulafungin reduced nuclei significantly better than commercial atovaquone plus anidulafungin. eATQ is a novel formulation of atovaquone that warrants further evaluation for treatment and prevention of PCP.

In a mouse model of Pneumocystis pneumonia, a novel encochleated nanoparticle formulation of atovaquone was superior to commercially available atovaquone in clearance of infection when each drug was administered alone as well as when combined with anidulafungin.

Pneumocystis pneumonia (PCP) continues to be a life-threatening infection in immunosuppressed patients, especially people living with human immunodeficiency virus (HIV), patients with hematologic malignancies, and solid organ transplant recipients. Prophylaxis administered to at-risk populations is highly effective in preventing PCP. Several drugs, including trimethoprim/sulfamethoxazole (TMP/SMX), dapsone, atovaquone, and aerosol pentamidine, have demonstrated activity in preventing infection in randomized clinical trials, which were conducted primarily in HIV-infected populations. TMP/SMX, dapsone (combined with trimethoprim), atovaquone, clindamycin/primaquine, and intravenous pentamidine also have demonstrated efficacy in treating PCP.

TMP/SMX is the most effective prophylactic regimen and is typically the recommended first-line regimen in published guidelines [1, 2]. However, TMP/SMX can be associated with a high incidence of adverse reactions that often require its discontinuation, especially in people with HIV. Moreover, due to the increased risk of bone marrow suppression associated with TMP/SMX [3], many physicians are reluctant to use it in certain populations such as hematopoietic stem cell transplant recipients.

Atovaquone, a hydroxynaphthoquinone that inhibits mitochondrial electron transport, has similar efficacy as dapsone and aerosol pentamidine in PCP prophylaxis in HIV-infected patients [4, 5]. Several factors, however, have limited its use. The current commercially available formulation of atovaquone, a suspension of microfine particles, has limited gastrointestinal absorption, resulting in plasma concentrations that are not proportional to the dose administered, and it must be taken with meals to maximize absorption [6]. Although optimal plasma atovaquone levels for prophylaxis are unknown, plasma levels correlate with efficacy in treating PCP as well as cerebral toxoplasmosis [7, 8]. Furthermore, gastrointestinal symptoms, intolerance of the taste, and other side effects can frequently lead to discontinuation of therapy [9]. Thus there is a need to develop an alternative, highly effective, and well-tolerated formulation of atovaquone that can consistently achieve predictable plasma (and target tissue) levels.

Cochleates are a stable, lipid-crystal, nanoparticle drug formulation platform, composed primarily of phosphatidylserine and calcium, that are orally bioavailable and can potentially deliver a broad range of bioactive molecules with predictable pharmacokinetics, including some that currently can only be administered parenterally [10]. Orally administered encochleated amphotericin B, for example, has been shown to be active in mouse models of Candida albicans and Cryptococcus neoformans infection [11, 12] and is currently undergoing clinical evaluation in phase 2 studies (eg, ClinicalTrials.gov NCT02971007 and NCT04031833). Encochleated drugs may target uptake by macrophages, which then gradually release the drug, possibly resulting in higher or more sustained tissue concentrations than the native drug. Given the potential benefits of encochleation as a well-tolerated drug delivery system with improved absorption compared to the available microparticle suspension, we undertook to evaluate the pharmacokinetics, safety, and efficacy of an encochleated formulation of atovaquone in a mouse model of PCP.

MATERIALS AND METHODS

Animals

Male C3H/HeN mice (National Cancer Institute Mouse Repository, Charles River Laboratories) were handled in strict accordance with good animal practice, as defined by the University of Cincinnati and Cincinnati Veterans Affairs Medical Center Institutional Animal Care and Use Committee. Mice were housed under barrier conditions with autoclaved food, acidified water, and bedding in sterilized shoebox cages equipped with sterile microfilter lids.

In Vivo Mouse Studies

For therapy studies, mice were infected by exposure to Pneumocystis murina–infected mice for 2 weeks and begun on an immunosuppressive regimen the first day of exposure (4 µg/mL dexamethasone in drinking water) and continued on this regimen for the remainder of the study. After 5 weeks, mice were randomly divided into treatment and control groups of 8 mice each, and begun on the treatment regimen or vehicle control. Encochleated atovaquone (eATQ; 25–200 mg/kg), commercial atovaquone (100 mg/kg; Mepron, GlaxoSmithKline, Research Triangle Park, North Carolina), and TMP/SMX (50/250 mg/kg; Teva Pharmaceuticals, North Wales, Pennsylvania) or vehicle alone (empty cochleates) were administered by oral gavage for 14 or 21 days. Anidulafungin (1 mg/kg; Eraxis, Pfizer, New York, New York) was administered by intraperitoneal injection 3 times per week for 14 days.

For pharmacokinetic studies, mice were infected as described above. After 5 weeks, mice received a single 100 mg/kg dose of eATQ or Mepron by oral gavage. Three mice were killed at each of 10 timepoints: prior to dosing (baseline) and 2, 4, 8, 10, 12, 24, 48, 72, and 96 hours postdose. Blood was collected and placed in tubes containing 0.3 M K₂ ethylenediaminetetraacetic acid and immediately centrifuged to collect plasma. Lungs from each mouse were also collected, weighed, and flash frozen.

Analysis of Pharmacokinetic Samples

Plasma and lung samples were extracted using a novel solid-phase extraction method employing Oasis Prime HLB 30 mg/well, 96-well plates (Waters Corporation). Additional details are provided in the Supplementary Methods. Pharmacokinetic parameters were calculated via noncompartmental analysis (Phoenix Winnonlin version 9.0).

Microscopic Enumeration of Organism Burdens

Lungs were dissociated in 10 mL of phosphate-buffered saline (PBS) using gentleMACS Dissociator (Miltenyi Biotec, Auburn, California), then filtered through a 40-μm pore mesh, centrifuged at 3400g for 10 minutes and resuspended in 2.0 mL PBS. Ten microliters of homogenate was placed in triplicate onto glass slides, which were air dried, heat fixed, and stained with either cresyl echt violet to enumerate asci or Hema 3 (Fisher Scientific Co), a rapid Wright–Giemsa stain, to enumerate total nuclei in both asci and trophic forms. Asci or nuclei were counted in 30 microscopic fields under oil immersion; limit of detection is approximately 1.75 × 10⁴ (log₁₀ 4.24/lung) cysts or nuclei per lung.

Histological Evaluation of Organs

Lung tissue (all mice) and other organs (5 mice per group) from the second study were fixed in 10% neutral buffered formalin and paraffin embedded. Slides were stained with hematoxylin and eosin. Lung inflammation and consolidation were scored based on level of involvement as previously described (Supplementary Methods) [13].

Statistical Analyses

Microscopic counts were log transformed and group outcomes were compared by 1-way analysis of variance followed by Tukey multiple comparison test using GraphPad Prism version 8 (GraphPad Software, San Diego, California). Survival curves were based on the determined treatment period and compared by the log-rank test using GraphPad Prism. For all analyses, P values < .05 were considered significant.

RESULTS

Significant Reduction in Organism Burden and Increased Survival Were Observed in eATQ-Treated Mice

Daily treatment of infected mice with eATQ for 14 or 21 days significantly reduced both nuclei and asci burden vs untreated controls at all doses (Figure 1A). There was no difference in efficacy between eATQ at 200 mg/kg vs the positive control TMP/SMX at reducing asci burden at either timepoint. Of note, eATQ significantly reduced both nuclei and asci burden vs clinically used atovaquone (Mepron) at day 14 when mice were treated with 100 mg/kg of each. A consistent dose response was seen between eATQ dose levels at both timepoints.

Figure 1.

A, Log₁₀ mean nuclei and asci counts after 14 (top row) and 21 days (bottom row) of treatment with encochleated atovaquone (eATQ; 25–200 mg/kg/day), Mepron (100 mg/kg/day), and trimethoprim/sulfamethoxazole (TMP/SMX; 50/250 mg/kg/day). Mice (8 per group) received indicated treatment by gavage, and nuclei and cysts were determined by counting aliquots of lung homogenates. By analysis of variance, groups were statistically significantly different for each panel (P < .0001 for all). Inserts show P values for comparison of each pair of groups, using Tukey multiple comparisons test. Significant differences are color coded as follows: pink, P < .05–.01; blue, P < .01–.0001; yellow, P < .0001). Bars represent mean and standard deviation. B, Survival curves after 14 and 21 days of treatment with eATQ, Mepron, and TMP/SMX. These data represent the survival curves for mice included in (A). P = .03 by log-rank (Mantel–Cox) test for the group analysis for 14-day survival, though no significant differences were seen when comparing each drug treatment group to the control group. For 21-day survival, P = .005 for the group analysis. *The following treatment groups had significantly better survival (P = .01–.015) when compared to the control group: eATQ 25 mg/kg and 100 mg/kg; Mepron; TMP/SMX. The early deaths (days 1 to 2) in the 50- and 200-mg eATQ groups suggest that those animals had advanced disease and therapy was begun too late to impact survival. Control indicates vehicle-treated negative controls.

Survival Rates Were Increased in eATQ-Treated Mice

Enhanced survival vs untreated controls in the 14-day study was seen in all eATQ dose groups except the 25 mg/kg group (Figure 1B). Consistent with this, eATQ at 25 mg/kg and 100 mg/kg showed a significant improvement in survival vs untreated controls at day 21. The 3 deceased mice in the 50 and 200 mg/kg eATQ dose groups were found at days 1 and 2 of dosing, suggesting that therapy was started too late to benefit them.

Combination Therapy With Anidulafungin Enhances Efficacy of eATQ

Anidulafungin is an echinocandin, a class of β-1,3-glucan synthase inhibitors of fungi that target the asci of Pneumocystis. Treatment with echinocandins results in decreased inflammation in mouse models of PCP [14, 15]. Given that Pneumocystis β-1,3-glucans may be responsible for much of the inflammation seen in PCP, combination therapy with eATQ and anidulafungin may provide additional benefits to eATQ monotherapy. Daily therapeutic treatment of infected mice with eATQ at 100 mg/kg in combination with anidulafungin significantly reduced both nuclei and asci burdens after 14 days (Figure 2A). The combination of eATQ and anidulafungin performed significantly better than eATQ, Mepron, or anidulafungin, each given as single agents, as well as the combination of Mepron and anidulafungin, in reducing nuclei burden when using the same dose levels. This study also reconfirmed the superiority of eATQ alone vs Mepron. There were no differences in survival between any groups.

Figure 2.

A, Log₁₀ mean nuclei and asci counts after 14 days of treatment with encochleated atovaquone (eATQ; 100 mg/kg/day), Mepron (100 mg/kg/day), and anidulafungin (1 mg/kg given 3 times per week), given alone or in combination. Mice (8 per group) received indicated treatment by gavage (eATQ and Mepron) or intraperitoneal injection (anidulafungin), and nuclei and cysts were determined by counting aliquots of lung homogenates. By analysis of variance (ANOVA), groups were statistically significantly different (P < .0001 for both panels). Inserts show P values for comparison of each pair of groups, using Tukey multiple comparisons test. The combination of eATQ plus anidulafungin showed a significantly greater reduction in nuclei counts compared to all other groups. Additional groups (not included in this analysis) included empty cochleates combined with anidulafungin, Mepron, and anidulafungin plus Mepron; results were concordant to those seen without empty cochleates. Significant differences are color coded as follows: pink, P < .05–.01; blue, P < .01–.0001; yellow, P < .0001. Bars represent mean and standard deviation. B, Histology score of lung inflammation and consolidation after 14 days of treatment for the mice included in (A). By ANOVA, groups were statistically significantly different (P < .0001 for inflammation; P = .01 for consolidation). Insert shows P values for comparison of each pair of groups, using Tukey multiple comparisons test. All treatment groups had significantly less inflammation than the control group, but only eATQ demonstrated less consolidation than the controls. There were no significant differences among the drug treatment groups in either inflammation or consolidation. Significant differences are color coded as follows: blue, P < .01–.0001; yellow, P < .0001). Bars represent mean and standard deviation. Control indicates vehicle-treated negative controls.

Reduced Inflammation and Consolidation Seen in eATQ-Treated Mice

Histological evaluation of lung pathology in the combination therapy experiment is summarized in Figure 2B. All treated groups showed a significant reduction in inflammation vs the untreated controls, likely due to decreased organism burden and reduced β-glucan levels in the anidulafungin-treated groups. Interestingly, only the eATQ alone group significantly reduced consolidation vs untreated controls.

Pharmacokinetic Results

After single 100 mg/kg eATQ or Mepron dose administration, maximum atovaquone concentrations (C_max) in both plasma and lung were observed at 12 hours postdose in eATQ-treated mice compared with 10 hours postdose with Mepron. With both formulations, there was a linear decline in concentrations, which remained detectable through 96 postdose (Supplementary Figure 1). The C_max was approximately 52.8 µg/L in plasma and 36.4 µg/g in lungs in eATQ-treated mice and 62.3 µg/L in plasma and 52.7 µg/g in lungs in Mepron-treated mice. The terminal elimination half-life was approximately 13 hours in plasma and 50 hours in lungs following eATQ treatment and 12 hours in plasma and 32 hours in lungs following Mepron treatment. Overall exposure, as measured by the area under the curve, was approximately 1360 µg × hour/mL and 1640 µg × hour/mL in lungs and 1240 µg/mL and 1040 µg/mL in plasma of eATQ-treated and Mepron-treated mice, respectively.

Histopathology

Gross physical examination of mouse organs showed no signs of overt toxicity. Histopathologic analysis of the kidney, liver, spleen, and brain from 5 animals per group in the second experiment showed no abnormal pathology in any group.

DISCUSSION

This study has demonstrated that an encochleated form of atovaquone is equal to or more effective than the commercially available formulation of atovaquone in treating PCP in a mouse model, and at the highest dose tested (200 mg/kg/day) showed similar activity to TMP/SMX at 14, though not 21, days. Furthermore, it showed superior activity to Mepron when each was coadministered with anidulafungin during a 14-day study. Thus, eATQ is a potentially attractive alternative to the currently available commercial formulation of atovaquone for the treatment and prophylaxis of PCP.

Although atovaquone has well-documented anti-Pneumocystis activity, its broad utilization in both treatment and prophylaxis has been hampered by several factors: limited gastrointestinal absorption, gastrointestinal toxicities, and intolerance of the taste of atovaquone. Highlighting this, in a study of atovaquone for treatment of toxoplasmosis, 9 of 40 subjects discontinued therapy due to nausea and vomiting or intolerance of the taste [9].

TMP/SMX is the recommended first-line regimen for prophylaxis of PCP in people with HIV and in non-HIV-infected patients, including hematopoietic stem cell transplant recipients, though some physicians are reluctant to use it due to delayed engraftment and a higher incidence of neutropenia [1, 2]. Alternatives include dapsone and aerosol pentamidine, which, in clinical trials in people with HIV, had similar efficacy as Mepron. eATQ has potential advantages over Mepron, including better tolerability and improved bioavailability, and like Mepron should also have activity against toxoplasmosis. While the current study was an efficacy study in treating PCP, the total approved daily dose for Mepron (1500 mg) is identical for treatment and prophylaxis for PCP, and thus eATQ should have activity in both settings.

A major concern with treatment of PCP is the progressive hypoxia that can develop soon after initiating therapy, which can be ameliorated by corticosteroid therapy. This appears to be triggered by death of Pneumocystis organisms and likely is related at least in part to the release of β-glucans, which are known to trigger inflammatory responses [14, 15]. While echinocandins alone are not indicated for treatment of PCP, given that they target the asci but not trophic forms, combining them with other anti-Pneumocystis agents may lead to decreased inflammation during therapy. We found that eATQ combined with anidulafungin resulted in a lower nuclei count compared to either drug alone, or Mepron plus anidulafungin. Although we did not examine inflammation in detail, by histology eATQ plus anidulafungin had the lowest inflammation score, though this was not statistically significantly different from the other treated groups.

In summary, this novel encochleated atovaquone shows promise as an improved formulation that may provide improved bioavailability, better tolerance, and long-term clinical benefit.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Acknowledgments. These studies were administered by the Cincinnati Education and Research Services Foundation at the Cincinnati Veterans Affairs Medical Center (VAMC). We are grateful to Ronald Brubaker, MD, Greater Cincinnati Pathologists, the Christ Hospital Health Network, for evaluation of the histopathology. We thank Chris Lambros for his advice on study design and valuable administrative support.

Disclaimer. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government.

Financial support. This project has been funded in whole or in part with federal funds from the Intramural Research Program of the United States National Institutes of Health (NIH) Clinical Center and the National Institute of Allergy and Infectious Diseases (NIAID), NIH. This work utilized funding from the NIAID suite of preclinical services for in vivo assessment (contract number HHSN2722010000291; task order HHSN27200004). M. T. C. is supported in part by a Biomedical Laboratory Research and Development Senior Research Career Scientist Award from the US Department of Veterans Affairs (award number IK6BX005232).

Potential conflicts of interest. J. A. K. was principal investigator on a cooperative research and development agreement, now expired, between the NIH and Matinas BioPharma Inc to support clinical trials of encochleated drugs. R. M. and R. L. are employees of Matinas BioPharma Inc. All other authors report no potential conflicts of interest.

All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.