Regimen comprising clarithromycin, clofazimine and bedaquiline is more efficacious than monotherapy in a mouse model of chronicMycobacterium aviumlung infection

14 Mycobacterium avium , a leading non -tuberculous mycobacterium (NTM) pathogen, causes chronic 15 pulmonary infections, particularly in individuals with underlying lung conditions or 16 immunosuppression. Current treatments involve prolonged multi-drug regimens with poor outcomes and 17 significant side effects, highlighting the urgent need for improved therapies. Using a BALB/c mouse 18 model of chronic M. avium pulmonary disease, we evaluated the efficacy of individual antibiotics —19 clarithromycin, clofazimine, and rifabutin —and combination regimens including 20 clarithromycin+bedaquiline and clarithromycin+clofazimine+bedaquiline. Clarithromycin 21 demonstrated potent bactericidal activity, reducing lung bacterial burden by 2.2 log 10 CFU, while 22 clofazimine transitioned from bacteriostatic to bactericidal, achieving a 1.7 log 10 CFU reduction. 23 Rifabutin was bacteriostatic against M. avium MAC 101 but ineffective against MAC 104. The triple -24 drug regimen of clarithromycin+clofazimine+bedaquiline was the most effective, achieving a 3.3 log 10 25 CFU reduction in bacterial load, with 98% clearance within the first week and continued efficacy over 26 eight weeks. Gross pathology confirmed these results, with granulomatous lesions observed only in 27 untreated or rifabutin-treated mice. Combination therapy demonstrated enhanced efficacy compared to 28 monotherapy. The findings underscore the potential of oral clarithromycin+clofazimine+bedaquiline or 29 clarithromycin+clofazimine regimen as a promising therapeutic strategy for M. avium pulmonary 30 disease. 31 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint

32 Mycobacterium avium is a slow-growing, non-tuberculous mycobacterium (NTM) commonly found in 33 water and soil (1). It is the most prevalent NTM pathogen in humans and a member of the Mycobacterium 34 avium complex (MAC), which includes closely related species often indistinguishable using standard 35 clinical microbiology staining techniques (2). M. avium primarily causes opportunistic lung infections, 36 particularly in individuals with underlying lung comorbidities such as bronchiectasis, cystic fibrosis, or 37 chronic obstructive pulmonary disease (COPD), as well as those with compromised immune systems 38 (3). Most infections result from environmental exposure, and patients typically present with symptoms 39 resembling bronchiectasis or tuberculosis (TB) (2). Relative to TB, caused by a related mycobacterium, 40 treatment for M. avium pulmonary disease is challenging, with limited options and low cure rates (4). 41 The current standard of care involves multi -drug regimens comprising three or more antibiotics that 42 inhibit essential functions in M. avium (5–8). Treatment typically lasts at least 18 months but may be 43 further prolonged and is complicated by significant side effects, requiring frequent monitoring and 44 adjustments. Despite these efforts, treatment outcomes remain poor. The increasing global prevalence 45 of M. avium pulmonary disease underscores the urgent need for more effective and tolerable therapies. 46 Only one drug, amikacin, has been approved for treating M. avium pulmonary disease based on a clinical 47 trial (9). In contrast, TB drug development has progressed more rapidly, in part due to preclinical testing 48 in animal models, particularly mouse models. These models have been critical in informing clinical trials 49 for TB and other mycobacterial diseases (10). 50 To address this gap, Andrejak et al. developed a chronic M. avium pulmonary disease model using mice 51 infected via aerosol exposure to mimic the natural infection route in humans (11). This model reproduces 52 lung pathology in humans and has been validated with standard antibiotics such as clarithromycin, 53 clofazimine, ethambutol, and rifampin, showing efficacy patterns consistent with human outcomes (11, 54 12). It has also been used to test experimental agents against M. avium (13). Using the BALB/c mouse 55 model, we evaluated the efficacy of select antibiotics with in vitro activity against M. avium but uncertain 56 effectiveness for lung disease. These included bedaquiline, clofazimine and rifabutin. Additionally, we 57 assessed the efficacies of select drug combinations, as M. avium pulmonary disease typically requires 58 regimens of three or more antibiotics. These included a two -drug combination of clarithromycin and 59 bedaquiline and a three-drug combination of clarithromycin, bedaquiline, and clofazimine. The efficacy 60 of the combination clarithromycin and clofazimine was not considered as it has been described using the 61 same mouse model (12). Our study aims to identify more effective therapeutic options, addressing the 62 critical need for improved treatments for M. avium pulmonary disease. 63 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint

64 Monotherapy efficacies: Clarithromycin and clofazimine are bactericidal, rifabutin lacks 65 efficacy 66 We assessed the efficacy of three antibiotics—100 mg/kg clarithromycin, 25 mg/kg clofazimine, and 20 67 mg/kg rifabutin—administered orally once daily to mice infected with MAC 101 ( Figure 1a). At the 68 time of infection, the mean lung bacterial load was 4.4 log10 CFU, which remained stable for four weeks 69 before treatment began. In the control group treated with PBS (the solvent for the test antibiotics), the 70 mean lung burden steadily increased over 12 weeks, resulting in a net increase of 0.95 log10 CFU, 71 reflecting a steady, chronic infection. 72 In the rifabutin -treated group, the mean lung burden of MAC 101 remained stable throughout the 73 treatment period, leading to a negligible net reduction of 0.09 log 10 CFU after eight weeks. Thus, 74 rifabutin displayed bacteriostatic activity against MAC 101. Statistical comparisons of the mean lung 75 burden among treatment groups are provided in Table S1. 76 For clofazimine -treated mice, the lung burden remained unchanged after one week of treatment. 77 However, at the end of four and eight weeks, net reductions in lung burden were 1.4 log 10 CFU and 1.7 78 log10 CFU, respectively. This indicates that clofazimine initially exhibited bacteriostatic activity but 79 became bactericidal with prolonged treatment. Clarithromycin demonstrated bactericidal activity from 80 the start of treatment, achieving a net reduction of 2.2 log 10 CFU by the end of the study. Among the 81 three antibiotics, clarithromycin was the most effective against MAC 101. 82 A parallel experiment was conducted with mice infected with MAC 104 to validate the findings using 83 an independent isolate (Figure 1b). At the time of infection, the mean lung burden was 4.7 log 10 CFU, 84 which increased by 1.5 log 10 CFU over 12 weeks in the PBS -treated control group, consistent with a 85 chronic infection. In rifabutin -treated mice, the lung burden followed a trajectory similar to the PBS 86 group, indicating that rifabutin was ineffective against MAC 104. Clofazimine exhibited bacteriostatic 87 activity during the first week of treatment but became bactericidal over time, producing a net reduction 88 of 1.9 log10 CFU after eight weeks. Clarithromycin again demonstrated bactericidal activity throughout 89 the treatment period, with a net reduction of 1.9 log10 CFU, matching the efficacy of clofazimine. 90 Gross pathological examination revealed granulomatous lesions in the lungs of mice treated with PBS 91 or rifabutin, which were absent in mice treated with clarithromycin or clofazimine ( Figure 1c). These 92 pathological findings aligned with the microbiological results. In summary, rifabutin was bacteriostatic 93 against MAC 101 but showed no activity against MAC 104. In contrast, clofazimine and clarithromycin 94 were effective against both isolates, with clarithromycin being the most potent overall. 95 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint Efficacy of regimen comprising clarithromycin, clofazimine and bedaquiline 96 The treatment of M. avium disease requires a multi-drug regimen to enhance efficacy and reduce the risk 97 of selecting drug-resistant mutants (5–8). Consequently, neither clarithromycin nor clofazimine is used 98 as monotherapy for this condition. However, given their strong anti-M. avium activity, we evaluated the 99 efficacy of a regimen combining clarithromycin and clofazimine with a third agent, bedaquiline, in line 100 with the current guideline recommendations to treat MAC lung infection with regimens comprising three 101 or more agents. (5–8). 102 We tested a triple -drug regimen comprising 100 mg/kg clarithromycin, 25 mg/kg clofazimine, and 25 103 mg/kg bedaquiline against MAC 101 using the same protocol as described above (Figure 2a ). In 104 untreated mice, the lung burden of MAC 101 increased steadily, similar to the first study. The 105 combination clarithromycin+clofazimine+bedaquiline demonstrated bactericidal activity throughout the 106 treatment period, achieving a net 3.3 log 10 CFU reduction in the lung burden of MAC 101. This 107 represented a 98% reduction in bacterial load at the conclusion of the first week of treatment ( Figure 108 2b). Of the remaining bacteria, 94% were cleared during the second to fourth weeks, and 54% of the 109 survivors were eliminated in the final four weeks of treatment. 110 Monotherapy with clarithromycin, clofazimine, or bedaquiline also reduced the MAC 101 lung burden, 111 but at a slower rate compared to the triple -drug regimen ( Figure 2a and 1a ). The combination 112 clarithromycin+bedaquiline was bactericidal throughout the treatment period, leading to a 3.1 log10 CFU 113 reduction in lung MAC 101 burden. During the first four weeks of treatment, the addition of clofazimine 114 significantly enhanced the potency of clarithromycin+bedaquiline, resulting in a greater reduction in 115 lung burden. However, after eight weeks, both regimens produced statistically similar reductions in lung 116 MAC 101 burden. This indicates that clofazimine primarily enhances the efficacy of 117 clarithromycin+bedaquiline during the early stages of treatment , although paradoxically clofazimine 118 monotherapy is bacteriostatic during this treatment stage. 119 Gross pathological examination at the end of the study revealed consolidated granulomas in the lungs of 120 untreated mice (Figure 2c). These granulomas, a hallmark of M. avium lung disease in both mice (11) 121 and humans (14), were absent in the lungs of mice treated with clarithromycin, bedaquiline, 122 clarithromycin+bedaquiline, or clarithromycin+clofazimine+bedaquiline . Notably, the lungs of mice 123 treated with the triple -drug regimen exhibited a reddish -yellow pigmentation, likely attributable to 124 clofazimine, which is known to cause such pigmentation (15). Mice receiving antibiotics appeared 125 healthy and showed no signs of sickness or lethargy throughout the study. In contrast, untreated mice 126 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint became lethargic during the final stages of the study. Importantly, no deaths occurred in any of the 127 treatment groups. 128 129

130 Current treatment for MAC pulmonary infections is protracted and frequently complicated by the poor 131 tolerability of complex regimens (4). Effective clinical decision -making, particularly when initiating 132 treatment or modifying regimens to manage side effects, depends on a robust understanding of the 133 bactericidal versus bacteriostatic efficacy of individual drugs and drug combinations. Unfortunately, 134 such data has historically been more limited for MAC compared to TB (11, 12, 16–19). The kinetics of 135 treatment response are critical clinical considerations, as therapy for chronic infections like MAC is often 136 divided into distinct phases: a rapid-killing “induction” phase, followed by less intensive “consolidation” 137 and “maintenance” phases. Each phase requires a dynamic balance between bactericidal efficacy, disease 138 symptom management, mitigation of treatment side effects, and the logistical complexity of the regimen. 139 Optimizing therapy to align with these shifting priorities at each phase has the potential to significantly 140 enhance both patient experience and overall treatment outcomes. 141 Two distinct MAC isolates were included in this study to identify variations in drug efficacy, such as the 142 differential activity of rifabutin, as well as instances where similar efficacies across isolates may allow 143 for broader generalization of the findings to other strains. The dose and dosing frequency of bedaquiline, 144 clarithromycin, clofazimine and rifabutin used in mice approximate their exposures in humans using 145 approved doses. The treatment period was limited to eight weeks and was not designed to determine the 146 duration required to achieve lung sterilization in mice. As such, the findings primarily offer valuable 147 insights into the trajectory of early bactericidal activity associated with various regimens. This study 148 focused on assessing drug efficacy against MAC isolates that are susceptible to bedaquiline, 149 clarithromycin, clofazimine, and rifabutin. Furthermore, the main focus was to assess the efficacies of 150 the dual combination clarithromycin+bedaquiline and the triple combination 151 clarithromycin+clofazimine+bedaquiline that have not been evaluated before. 152 Consistent with clinical observations, rifabutin as a monotherapy displayed limited efficacy, showing 153 only bacteriostatic activity at best against MAC 101 and no observable effect against MAC 104 (20, 21). 154 On the other hand, Clarithromycin and clofazimine exhibited bactericidal activity against both MAC 155 strains and were therefore tested in combination with bedaquiline. Again, consistent with prior 156 observations for other mycobacteria, clofazimine as a monotherapy showed an initial bacteriostatic effect 157 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint followed by delayed bactericidal activity (22, 23) . Bedaquiline monotherapy closely paralleled the 158 bactericidal trajectory of clarithromycin monotherapy by week four, although it showed comparably 159 reduced bactericidal activity during the early stages of treatment. When clarithromycin was combined 160 with bedaquiline, the regimen demonstrated early bactericidal activity similar to clarithromycin 161 monotherapy, but with slightly more sustained bactericidal effects by week eight, indicating added 162 benefit from the combination during later treatment stages. 163 The triple-drug combination of clofazimine, clarithromycin, and bedaquiline demonstrated a more rapid 164 bactericidal effect against MAC 101 than was expected based on the effects of clofazimine monotherapy 165 or the clarithromycin + bedaquiline dual therapy. This triple combination led to a greater than 1 log10 166 reduction in lung CFU burden at both weeks one and four, translating to a 98% reduction in organisms 167 within the first week of treatment. This rapid early bactericidal activity contrasts sharply with the delayed 168 bactericidal effect observed with clofazimine monotherapy against both MAC 101 and 104. However, 169 this study did not evaluate clofazimine in two -drug combinations with either bedaquiline or 170 clarithromycin, and as such, we cannot speculate whether these dual regimens might achieve comparable 171 bactericidal timing to the three-drug combination. Previous study by Lanoix et al. suggested a synergistic 172 relationship between clarithromycin and clofazimine based on their inclusion in more complex regimens 173 alongside ethambutol and rifampin, though direct testing of clarithromycin and clofazimine as a 174 standalone pair was not conducted (12). Future studies are needed to assess the efficacy of bedaquiline 175 and clarithromycin in pairwise combinations with clofazimine. 176 The bactericidal trajectories observed in the two- and three-drug regimens in this study are both striking 177 and clinically informative. At the conclusion of eight weeks of treatment , the 178 clarithromycin+bedaquiline dual therapy achieved a level of bactericidal activity comparable to that of 179 the clofazimine+clarithromycin+bedaquiline combination, although its early bactericidal effect was not 180 as potent as that of the triple therapy. This finding suggests that clofazimine could be strategically added 181 to or removed from a clarithromycin -based backbone, with or without bedaquiline, to tailor treatment 182 across different phases. One potential approach would involve using the three -drug combination for its 183 strong early bactericidal activity during the induction phase, then transitioning to 184 clarithromycin+bedaquiline for the maintenance phase. Alternatively, if adverse side effects or drug -185 drug interactions pose significant concerns during the stabilization period, clofazimine could be 186 introduced to a clarithromycin+bedaquiline regimen after stabilization. While beyond the scope of this 187 study, future research on transitioning between such regimens at six to eight weeks of treatment could 188 help further optimize bactericidal effects and inform clinical management strategies. 189 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint

190 Bacterial strains, growth media and growth conditions. Mycobacterium avium strain ATCC 700898, 191 historically known as MAC 101, was purchased from American Type Culture Collection (Manassas, 192 Virginia). Mycobacterium avium strain MAC 104 was a gift from Jacques Grosset laboratory, Johns 193 Hopkins University, and used in the development of the mouse model of M. avium pulmonary disease 194 (11). To infect mice, MAC 101 and MAC 104 were grown in Middlebrook 7H9 broth (Difco, catalog 195 no. 271310) supplemented with 0.5% glycerol, 0.05% Tween-80 and 10% oleic acid-albumin-dextrose-196 catalase enrichment as described (24) in an orbital shaker at 220 RPM, 37 oC. MAC 101 and MAC 104 197 in the lungs of mice were grown by inoculating 10 -fold serial dilutions of lung homogenates onto 198 Middlebrook 7H11 selective agar (Difco, catalog no. 283810) supplemented with 0.5% glycerol, 0.05% 199 Tween-80 and 10% oleic acid -albumin-dextrose-catalase enrichment (BD, catalog no. 212351), 50 200 μg/mL cycloheximide (Sigma -Aldrich, catalog no. C7698), and 50 μg/mL carbenicillin (Research 201 Products International, catalog no. C46000). 202 Antibiotics. All antibiotics preparations were made under sterile conditions. For clarithromycin (Sigma-203 Aldrich, catalog no. C9742), the amounts of the powder form necessary for each week of administration 204 to mice were weighed into 50 ml polypropylene tubes prior to treatment initiation and stored at 4˚C. At 205 the beginning of each week, the weekly aliquot was retrieved, mixed with 0.05% agarose at 4 ˚C to 206 prepare a concentration of 10 mg/mL, and vortexed for 5 minutes. This preparation appears as white 207 homogeneous suspension. The aliquot necessary for each day was transferred to 5 ml tubes and stored 208 at 4˚C until use. An 0.05% agarose solution was prepared by adding 50 mg Bacto agar (BD, catalog no. 209 214010) to 100 mL 1x phosphate buffered saline (PBS), pH 7.4 (Quality Biologicals, catalog no. 114-210 058-101), autoclaving for 10 min at 121°C and stored at 4˚C until use. 211 For clofazimine (Sigma-Aldrich, catalog no. C8895), the weekly amount of powder was weighed into 212 50 mL polypropylene tubes and stored at 4˚C. At the beginning of each week, the weekly aliquot was 213 retrieved, mixed with 0.05% agarose at 4˚C to prepare a concentration of 2.5 mg/mL, and vortexed for 214 5 minutes. This suspension was then sonicated at 50% power for 15 seconds per cycle, with 2 -3 cycles, 215 until a matte red, opaque, homogeneous colloidal suspension was achieved. Aliquots necessary for each 216 day were transferred to 5 ml tubes and stored at 4˚C until use. 217 For rifabutin (Sigma-Aldrich, catalog no. R3530), the amounts of powder necessary for each week were 218 weighed into 50 ml polypropylene tubes prior to treatment initiation and stored at 4˚C. At the beginning 219 of each week, the weekly aliquot was retrieved, mixed with 0.05% agarose at 4 ˚C to prepare a 220 concentration of 2 mg/mL, and vortexed for 5 minutes. This preparation appears as dark red 221 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint homogeneous suspension. The aliquot necessary for each day was transferred to 5 ml tubes and stored 222 at 4˚C until use. 223 For bedaquiline, powdered form bedaquiline fumarate (CAS no. 845533-86-0, Octagon Chemicals Ltd) 224 was used. The amounts of the powder necessary for each week were weighed into a 100 ml borosilicate 225 bottle, the precise volume of 20% 2-hydroxypropyl-β-cyclodextrin solution was added and dissolved by 226 stirring with a magnetic stirrer for three hours at 4 ˚C to prepare 2.5 mg/mL solution which appears 227 transparent. Aliquots necessary for each day were transferred to 5 ml tubes and stored at 4 ˚C until use. 228 A 20% 2 -hydroxypropyl-β-cyclodextrin (HPCD) (Sigma-Aldrich, catalog no. 332593) solution was 229 prepared as described (25). Briefly, 20 g of HPCD powder was transferred to a 100 -mL borosilicate 230 bottle, and 75 mL of sterile deionized water was added and stirred with a magnetic stirrer until a clear 231 solution was obtained (~30 min). Approximately 1.5-mL of 1 N HCl was added to bring pH to 2.0, and 232 the final volume was brought to 100 mL by adding sterile DI water. This solution was filtered through a 233 0.22-mm acetate cellulose filter and stored at 4°C until use. 234 Infection and antibiotics efficacy assessment in mice. Three different cohorts of four -five weeks old 235 female BALB/c mice were procured from the Charles River Laboratory (Wilmington, Massachusetts, 236 USA) and housed in biosafety level 2 vivarium. Following arrival in our vivarium, mice were allowed 237 to acclimatize for 7-10 days prior to initiating the studies. Mice were infected with MAC 101 or MAC 238 104 as described by Andrejak et al in a mouse model of M. avium lung infection (11). To infect mice, a 239 fresh MAC 101 or MAC 104 culture at exponential phase, A 600nm of 1.00 -1.60, was diluted in 240 Middlebrook 7H9 broth to A 600nm of 1.0. 10 ml of this suspension was aerosolized with a nebulizer 241 attached to Glas -Col Inhalation Exposure System A4212 (Glas -Col, Terre Haute, Indiana) into the 242 chamber where all mice in an infection cohort were held. The infection sequence comprised of 15 243 minutes of pre -heat, 30 minutes of Mab suspension aerosolization into the chamber, 30 minutes of 244 aerosol decay, and 15 minutes of surface decontamination with ultraviolet light. All mice in each study 245 were infected simultaneously by natural breathing of the same M. avium-carrying aerosol for one hour. 246 To determine M. avium implantation in the lungs, five mice were sacrificed one day post infection 247 (designated ‘week -4’), lungs were extracted aseptically, homogenized in 1xPBS with 2 mm glass beads 248 by bead-beating for 30 seconds at 4,000 rounds -per-minute (Minilys, Bertin Instruments), 0.1 ml of 249 appropriate 10-fold dilutions were inoculated onto selective Middlebrook 7H11 agar, incubated at 37 oC 250 for 14 days and colony forming units were enumerated. Similarly, five mice were sacrificed at one-, two-251 , three- and four-weeks post infection (designated as weeks -3, -2, -1 and 0, respectively, in the figures) 252 and lung M. avium burden was determined. Timepoint designated as ‘week 0’ represents the day 253 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint antibiotics treatment was initiated and marks the conclusion of four weeks of infection. Lung M. avium 254 burden was determined at the completion of one -, four- and eight-weeks of treatment (designated as 255 ‘week+1, +4 and +8’, respectively) from five mice per treatment group, per timepoint. 256 Bedaquiline, clarithromycin, clofazimine and rifabutin were administered to deliver 25 mg/kg, 100 257 mg/kg, 25 mg/kg and 20 mg/kg of the antibiotics, respectively, per mouse, once daily, seven days a week 258 for eight weeks. To achieve this, 0.2 ml bolus of 2.5 mg/ml bedaquiline, 10 mg/ml clarithromycin, 2.5 259 mg/ml clofazimine, and 2.0 mg/ml rifabutin preparations described were administered to each mouse by 260 oral gavage using a 22 -gauge curved gavage needle, with a 2 -mm tip diameter (Gavageneedle.com; 261 AFN2425C) fitted to a 1-mLslip-tip syringe (Becton & Dickinson, 309659). 262 Ethics statement . Animal procedures described here were performed in adherence to the national 263 guidelines and to the Johns Hopkins University Animal Care and Use committee approved protocol 264 MO23M163. 265 Lung Gross Pathology. In two efficacy assessment studies, one against MAC 101 and one against MAC 266 104, one half of the lungs from two mice from each treatment group at the final time point were allocated 267 for lung gross pathology. Respective lungs were extracted, submerged in 5 ml 1x PBS for 48 hours and 268 in 5 ml 10% buffered-formalin for 72 hours. The lungs were air dried and photographed. 269 Data analysis. Raw lung CFU data were analyzed, and the mean ± standard deviation was calculated for 270 each group at each timepoint. These results were graphed as dot plots. To assess the variance between 271 treatment groups at each timepoint, a one -way ANOVA multi comparison was performed ( Table S1), 272 with significance determined at the 95% confidence level. A p-value of ≤ 0.05 was considered indicative 273 of a non-random event, signifying significant differences in CFU burden between groups. 274 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint FIGURES 275 Figure 1: M. avium MAC 101 (A) and MAC 104 (B) burden in the lungs of BALB/c mice. Time point 276 week -4 represents 24 h after infection with respective strain via the aerosol route. Time point week 0 277 represents conclusion of four weeks of infection and the day of antibiotic treatment initiation. Time 278 points week - 1, 4 and 8 represent the end of 1, 4 and 8 weeks of once daily oral administration of 279 phosphate-buffered saline (PBS), 100 mg/kg clarithromycin (CLR), 25 mg/kg clofazimine (CFZ), and 280 20 mg/kg rifabutin (RFB). Mean CFU per lung and standard deviation are shown (n=5 per time point 281 per group). (C) Gross pathology of the lungs of mice infected with MAC 104 from each treatment 282 group, two mice per group at the conclusion of treatment (week 8) are shown. 283 Figure 2: (A) M. avium MAC 101 burden in the lungs of BALB/c mice. Time point week -4 represents 284 24 h after infection via the aerosol route. Time point week 0 represents conclusion of four weeks of 285 infection and the day of antibiotic treatment initiation. Time points week- 1, 4 and 8 represent the end of 286 1, 4 and 8 weeks of once daily oral administration of phosphate -buffered saline (No treatment), 100 287 mg/kg clarithromycin (CLR), 25 mg/kg bedaquiline (BDQ) and 25 mg/kg clofazimine (CFZ). Mean 288 CFU per lung and standard deviation are shown (n=5 per time point per group). (B) Percentage 289 reductions in the mean MAC 101 burden in the lungs of mice treated with 100 mg/kg clarithromycin + 290 25 mg/kg clofazimine + 25 mg/kg bedaquiline during the first week, second -fourth week, and fifth -291 eighth week are shown. (C) Gross pathology of the lungs of mice infected with MAC 101 from each 292 treatment group, two mice per group, at the conclusion of treatment (week 8) are shown. 293 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint FUNDING 294 This study was supported by NIH award R01 AI 155664. Ruth Howe was supported by the Sherrilyn 295 and Ken Fisher Center for Environmental Infectious Diseases, Division of Infectious Diseases, Johns 296 Hopkins University. 297 298 AUTHOR CONTRIBUTIONS 299 BR: methodology, study design, investigation, data analysis and interpretation, manuscript preparation. 300 RAH: data interpretation and manuscript preparation. CMP. Methodology and investigation. GL: study 301 conception, study design, project admin istration, data interpretation, manuscript preparation, and 302 funding acquisition. 303 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint

304 1. Falkinham JO. 2022. Nontuberculous mycobacteria in the environment. Tuberculosis 305 137:102267. 306 2. Adjemian J, Daniel-Wayman S, Ricotta E, Prevots D. 2018. Epidemiology of Nontuberculous 307 Mycobacteriosis. Semin Respir Crit Care Med 39:325–335. 308 3. van Ingen J, Obradovic M, Hassan M, Lesher B, Hart E, Chatterjee A, Daley CL. 2021. 309 Nontuberculous mycobacterial lung disease caused by Mycobacterium avium complex - disease 310 burden, unmet needs, and advances in treatment developments. Expert Rev Respir Med 15:1387–311 1401. 312 4. Zo S, Choe J, Kim DH, Kim S-Y, Jhun BW. 2024. Long-term clinical course of Mycobacterium 313 avium complex pulmonary disease patients with treatment failure. Antimicrob Agents Chemother 314 68. 315 5. Daley CL, Iaccarino JM, Lange C, Cambau E, Wallace RJ, Andrejak C, Böttger EC, Brozek J, 316 Griffith DE, Guglielmetti L, Huitt GA, Knight SL, Leitman P, Marras TK, Olivier KN, Santin M, 317 Stout JE, Tortoli E, van Ingen J, Wagner D, Winthrop KL. 2020. Treatment of Nontuberculous 318 Mycobacterial Pulmonary Disease: An Official ATS/ERS/ESCMID/IDSA Clinical Practice 319 Guideline. Clin Infect Dis 71:e1–e36. 320 6. Nguyen M -VH, Daley CL. 2023. Treatment of Mycobacterium avium Complex Pulmonary 321 Disease: When Should I Treat and What Therapy Should I Start? Clin Chest Med 44:771–783. 322 7. Varley CD, Winthrop KL. 2022. Nontuberculous Mycobacteria. Clin Chest Med Diagnosis Treat 323 43:89–98. 324 8. Haworth CS, Banks J, Capstick T, Fisher AJ, Gorsuch T, Laurenson IF, Leitch A, Loebinger MR, 325 Milburn HJ, Nightingale M, Ormerod P, Shingadia D, Smith D, Whitehead N, Wilson R, Floto 326 RA. 2017. British Thoracic Society guidelines for the management of non -tuberculous 327 mycobacterial pulmonary disease (NTM-PD). Thorax 72:ii1–ii64. 328 9. Winthrop KL, Flume PA, Thomson R, Mange KC, Yuen DW, Ciesielska M, Morimoto K, Ruoss 329 SJ, Codecasa LR, Yim J -J, Marras TK, van Ingen J, Wallace RJ, Brown -Elliott BA, Coulter C, 330 Griffith DE. 2021. Amikacin Liposome Inhalation Suspension for Mycobacterium avium 331 Complex Lung Disease: A 12-Month Open-Label Extension Clinical Trial. Ann Am Thorac Soc 332 18:1147–1157. 333 10. Sarkar S, Heise MT. 2019. Mouse Models as Resources for Studying Infectious Diseases. Clin 334 Ther 41:1912–1922. 335 11. Andréjak C, Almeida D V., Tyagi S, Converse PJ, Ammerman NC, Grosset JH. 2015. 336 Characterization of mouse models of Mycobacterium avium complex infection and evaluation of 337 drug combinations. Antimicrob Agents Chemother 59:2129–2135. 338 12. Lanoix J -P, Joseph C, Peltier F, Castelain S, Andréjak C. 2020. Synergistic Activity of 339 Clofazimine and Clarithromycin in an Aerosol Mouse Model of Mycobacterium avium Infection. 340 Antimicrob Agents Chemother 64. 341 13. Froment A, Delomez J, Da Nascimento S, Dassonville -Klimpt A, Andréjak C, Peltier F, Joseph 342 C, Sonnet P, Lanoix J-P. 2024. Efficacy of mefloquine and its enantiomers in a murine model of 343 Mycobacterium avium infection. PLoS One 19:e0311167. 344 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint 14. Kalayjian RC, Toossi Z, Tomashefski JF, Carey JT, Ross JA, Tomford JW, Blinkhorn RJ. 1995. 345 Pulmonary Disease Due to Infection by Mycobacterium avium Complex in Patients with AIDS. 346 Clin Infect Dis 20:1186–1194. 347 15. Murashov MD, LaLone V, Rzeczycki PM, Keswani RK, Yoon GS, Sud S, Rajeswaran W, Larsen 348 S, Stringer KA, Rosania GR. 2018. The Physicochemical Basis of Clofazimine -Induced Skin 349 Pigmentation. J Invest Dermatol 138:697–703. 350 16. Ji B, Lounis N, Truffot-Pernot C, Grosset J. 1994. Effectiveness of various antimicrobial agents 351 against Mycobacterium avium complex in the beige mouse model. Antimicrob Agents Chemother 352 38:2521–2529. 353 17. Lee JM, Kim L-H, Kim S-Y, Jhun BW, Lee W, Shin SJ. 2023. Intracellular and in vivo activities 354 of oxazolidinone drugs against Mycobacterium avium complex infection. Sci Rep 13:20631. 355 18. Zheng L, Qi X, Zhang W, Wang H, Fu L, Wang B, Chen X, Chen X, Lu Y. 2023. Efficacy of 356 PBTZ169 and pretomanid against Mycobacterium avium, Mycobacterium abscessus, 357 Mycobacterium chelonae, and Mycobacterium fortuitum in BALB/c mice models. Front Cell 358 Infect Microbiol 13:1115530. 359 19. Offman EM, Leestemaker-Palmer A, Fathi R, Keefe B, Bibliowicz A, Raday G, Bermudez LE. 360 2024. Triple -Antibiotic Combination Exerts Effective Activity against Mycobacterium avium 361 subsp. hominissuis Biofilm and Airway Infection in an In Vivo Murine Model. Antibiotics 362 13:475. 363 20. Nightingale SD, Cameron DW, Gordin FM, Sullam PM, Cohn DL, Chaisson RE, Eron LJ, Sparti 364 PD, Bihari B, Kaufman DL, Stern JJ, Pearce DD, Weinberg WG, LaMarca A, Siegal FP. 1993. 365 Two Controlled Trials of Rifabutin Prophylaxis against Mycobacterium avium Complex Infection 366 in AIDS. N Engl J Med 329:828–833. 367 21. Gordin FM, Sullam PM, Shafran SD, Cohn DL, Wynne B, Paxton L, Perry K, Horsburgh, Jr. CR. 368 1999. A Randomized, Placebo‐Controlled Study of Rifabutin Added to a Regimen of 369 Clarithromycin and Ethambutol for Treatment of Disseminated Infection with Mycobacterium 370 avium Complex. Clin Infect Dis 28:1080–1085. 371 22. Ammerman NC, Swanson R V, Tapley A, Moodley C, Ngcobo B, Adamson J, Dorasamy A, 372 Moodley S, Mgaga Z, Bester LA, Singh SD, Almeida D V, Grosset JH. 2017. Clofazimine has 373 delayed antimicrobial activity against Mycobacterium tuberculosis both in vitro and in vivo. J 374 Antimicrob Chemother 72:455–461. 375 23. Sriram D, Wahi R, Maggioncalda EC, Panthi CM, Lamichhane G. 2022. Clofazimine as a 376 comparator for preclinical efficacy evaluations of experimental therapeutics against pulmonary 377 M. abscessus infection in mice. Tuberculosis 137:102268. 378 24. Larsen M. 2000. Some Common Methods in Mycobacterial Genetics, p. 313–320. In Hatfull, GF, 379 Jacobs, W. R., J (eds.), Molecular Genetics of Mycobacteria. American Society for Microbiology, 380 Washington DC. 381 25. Lounis N, Veziris N, Chauffour A, Truffot -Pernot C, Andries K, Jarlier V. 2006. Combinations 382 of R207910 with Drugs Used To Treat Multidrug -Resistant Tuberculosis Have the Potential To 383 Shorten Treatment Duration. Antimicrob Agents Chemother 50:3543–3547. 384 385 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted December 11, 2024. ; https://doi.org/10.1101/2024.12.11.627976doi: bioRxiv preprint