Thank you for visiting Nature.com. You are using a browser version with limited CSS support. For the best experience, we recommend that you use an updated browser (or disable Compatibility Mode in Internet Explorer). In addition, to ensure ongoing support, we show the site without styles and JavaScript.
Sliders showing three articles per slide. Use the back and next buttons to move through the slides, or the slide controller buttons at the end to move through each slide.
An extended-release platform for HIV pre-exposure prophylaxis (PrEP) can increase adherence and maximize public health benefits. We report an injectable, biodegradable and in situ shaped removable implant (ISFI) that releases above a protective baseline the integrase inhibitor cabotegravir (CAB) when administered subcutaneously for more than 6 months. CAB ISFI was well tolerated in female mice and female macaques, showing no signs of toxicity or chronic inflammation. In rhesus monkeys, median plasma CAB concentrations exceeded established protective baseline levels of PrEP for 3 weeks and conferred complete protection against recurrent rectal SHIV infections. Removal of the implants through a small incision in 2 rhesus monkeys at 12 weeks resulted in a 7 to 48-fold decrease in plasma CAB levels within 72 hours. Modified dose modeling of CAB ISFI showed that a 3 ml injection would exceed baseline human protection levels for up to 5 months post-administration. Our results confirm the clinical progress of CAB ISFI in human long-acting PrEP.
As of 2021, 38.4 million people worldwide are living with HIV and 40.1 million people worldwide have died from AIDS-related illnesses since the start of the AIDS epidemic1. Daily oral pre-exposure prophylaxis (PrEP) including emtricitabine (FTC) and tenofovir disoproxil fumarate (TDF) or tenofovir alafenamide if adherence is high. Very effective in HIV infection2,3. While high adherence leads to high PrEP efficacy, maintaining high adherence to daily oral PrEP remains a major challenge4-6. Poor treatment adherence and subsequent infections can also lead to the development of drug-resistant viruses7. In addition, adherence is particularly low among young women in sub-Saharan Africa due to high levels of stigma, low product acceptance and/or failure to communicate product use to sexual partners5. Alleviating adherence problems in sub-Saharan Africa, especially among women, is key to the success of PrEP, as sub-Saharan Africa accounts for more than 60% of all new HIV infections worldwide1,5. To this end, the HIV prevention program is moving towards the development of long-acting PrEP (PrEP) products that do not require frequent dosing and may address some of the compliance problems associated with daily oral PrEP. In fact, studies have shown that long-acting drugs have higher adherence and tolerability when taken daily by mouth5,8,9,10,11, especially in populations with the highest HIV prevalence. Thus, increased adherence to PrEP and the acceptability of long-acting PrEP may ultimately reduce HIV transmission.
The FDA approved the long-acting integrase inhibitor cabotegravir (CAB LA) injectable in late 2021 for PrEP12 in both men and women. The approval of CAB LA followed the results of the HPTN 083 and 084 trials, which found CAB LA to be safer and more effective than daily oral FTC/TDF, which may reflect the adherence advantage of long-acting PrEP13,14,15. These studies also determined that the plasma concentration of CAB required for protection should be four times the 90% protein-adjusted inhibitory concentration (4x PA-IC90, 664 ng/mL16). CAB LA is administered intramuscularly at 3 ml monthly for an initial period of 2 months and then every two months. Despite significant progress in HIV PrEP and CAB LA treatment, large injection volumes (3 ml per injection), injection site reactions14,17 and the inability to self-administer or remove the drug to stop treatment if needed all need to be addressed. In addition, CAB LA has a long pharmacological tail due to its long terminal half-life (>40 days)16 and failure to eliminate after administration, with higher levels of residual plasma CAB up to 15 months after treatment cessation Low but detectable18,19 These suboptimal levels of CAB can lead to breakthrough infection and the development of drug-resistant HIV, requiring additional oral PrEP to cover the tail to prevent future infection19. To overcome these limitations, efforts are currently being made to develop detachable, self-contained (eg, subcutaneous) ultra-long-acting forms of CAB that respond to long dosing intervals (eg, every 6 months or more) Maintaining protective plasma levels of CAB. Such formulation can facilitate large-scale implementation and maximize cost-effectiveness and public health benefits in resource-poor and resource-rich countries alike.
Forming in situ implants (ISFIs) have desirable properties for ultra-tape CAB formulations, including long dosing intervals, low injection volumes, and retrievability 20, 21 . ISFI consists of hydrophobic and biodegradable polymers such as poly(lactic glycolic acid) (PLGA), biocompatible water-soluble organic solvents such as N-methyl-2-pyrrolidone (NMP) or bismethyl sulfoxide (DMSO)) and an active pharmaceutical ingredient (API). )22, 23, 24 are formulated together to obtain a homogeneous and injectable liquid solution or suspension. After intramuscular or subcutaneous injection, water-soluble organic solvents diffuse into the aqueous medium, causing a phase inversion to form a solid or semi-solid depot containing API23,24,25,26 trapped in the precipitating polymer matrix. The API is released from the depot by diffusion of the polymer matrix and extensive degradation of the polymer over time.
ISFI containing PLGA and NMP has been extensively studied for sustained release of dolutegravir (DTG), an analogue of the HIV integrase inhibitor CAB20,21. In total gamma-knockout NOD scid mice and bone marrow-liver-thymus (BLT) mice, ISFI releases DTG for 11 months at levels 10-100 times that of PA-IC90. These levels have been associated with high protection against multiple high-dose vaginal HIV infection and suppression of viremia in infected mice 20, 21 . In addition, although the implant did not need to be removed due to the biodegradable nature of PLGA, this study demonstrated that the reservoir could be easily removed through a small skin incision. After depletion, plasma DTG levels fell below PA-IC90 within 24 hours and below the detection limit after 7 days20. Taken together, these observations highlight the potential of ISFI as a new ultra-long acting delivery system and support an extended evaluation of ISFI for CAB delivery.
Here, we have developed and characterized a high drug loading CAB ISFI formulation suitable for small volume subcutaneous administration. We have determined in vitro and in vivo stability, microstructure, injectability and release kinetics. We show that this drug is safe in female mice and non-human primates and releases CAB over 6–11 months at levels above the established baseline for protecting PrEP in macaques and humans. We have associated prolonged release of CAB in ISFI with long-term protection against SHIV infection in a macaque PrEP model that predicted the clinical efficacy of CAB LA and other approved oral PrEP regimens. Our study identifies a promising platform for sustained release of CAB at levels known to be associated with PrEP protection in humans.
Many factors affect drug release kinetics from ISFI, including polymer type and molecular weight (MW), solvent, polymer to solvent ratio, miscibility of drug and polymer to solvent, and drug physicochemical properties20,24,27. Here, we investigated the effect of solvent, drug loading, polymer molecular weight, and polymer to solvent ratio to develop ISFI with maximum CAB loading and high release rate in vitro. First, we examined the saturation solubility of CAB in various solvents to determine the ideal solvent system for the formulation to achieve maximum drug loading (Supplementary Table 1). NMP and DMSO are commonly used water-miscible organic solvents in ISFI24 formulations, so these solvents were tested in various weight ratios. The addition of excipients such as Tween20, Pluronics and Gelucire have also been investigated as they may improve the water solubility of poorly soluble drugs28. Based on the solubility of CAB in these various solvents, a 1:1 (w/w) weight ratio of NMP:DMSO (referred to as “solvent”) showed the highest solubility of CAB (167.12±12.04mg/mL at saturation) and thus , it is useful to prepare the CAB as a stable ISFI suspension at a solubility above this saturation.
During the optimization and building of our previous work, all CAB ISFI formulations were created using an FDA-approved biodegradable polymer composed of 50:50 PLGA, allowing antiretroviral drugs to be released from ISFI20,21,29, extremely sustained release and good quality. degradation properties, including initiating complete degradation over several months30,31. This degradation profile is ideal for ensuring complete degradation of the polymer prior to subsequent ISFI injections to prevent polymer accumulation. In addition, all formulations consisted of 50:50 low molecular weight 10 kDa or 27 kDa PLGA, as lower molecular weight PLGA can withstand higher drug loading than higher molecular weight PLGA, causing stronger drug release. low viscosity to provide injectability31 for ISFI systems24,32,33.
A 35-day cumulative in vitro release study of seven CAB ISFI formulations (Figure 1a) was performed to examine the effect of drug loading, PLGA molecular weight, and polymer to solvent ratio to determine the agent with the highest loading and drug release. , Optimal cooking speed (Figure 1 and Supplementary Figure 1). The results showed that the CAB release rate increased with (1) increasing drug loading (Fig. 1b), (2) increasing solvent amount, and (3) decreasing PLGA molecular weight (Fig. 1c). All formulations showed very low burst release (<3% release in 24 hours), sustained release over one month, and expected in vitro release over 6 months (Figure 1d). Contains 349 mg/ml CAB (1:4 w/w PLGA (27 kDa): solvent) (recipe 4), 500 mg/ml CAB (1:4 w/w PLGA (27 kDa): solvent) (recipe 5) ISFI) and 500 mg/ml CAB (1:4 w/w PLGA (10 kDa):solvent) (Formulation 7) showed zero order release kinetics during the first 35 days. In addition, CAB release significantly differed (p < 0.05) with changes in drug loading (Fig. 1b) and PLGA molecular weight (Fig. 1c). Notably, ISFI contains 500 mg/ml CAB (1:4 w/w PLGA (10 kDa): diluent) (recipe 7) and 500 mg/ml (1:4 w/w PLGA (27 kDa) : solvent) (formulation 5) showed comparable release kinetics (p > 0.05) despite differences in PLGA molecular weight.
Cumulative release of CAB ISFI formulations. b Effect of drug loading on cumulative CAB release. c Effect of PLGA molecular weight on cumulative CAB release. All in vitro release studies were performed in phosphate buffered saline (PBS, pH 7.4 containing 2% Solutol) at 37°C. Data are presented as mean ± standard deviation for n = 3 samples. Raw data are provided in the form of raw data files. Statistical analysis. Two-way ANOVA with Tukey’s multiple comparison test was used to compare drug release with respect to composition and time point in b and c. *p < 0.05, ***p < 0.001, ****p < 0.0001, ns at p > 0.05 (not significant). d Summary table of release kinetics of CAB ISFI formulations. Solvent = 1:1 (w/w) NMP:DMSO.
The similar release rates between formulations 5 and 7 may be due to the rather high drug loading (500 mg/mL CAB) in the formulations, which determined the release profile of CAB regardless of differences in PLGA molecular weight. In both formulations, the ratio (w/w) of PLGA to CAB was 1:3.5 w/w. (41.2% CAB and 11.8% PLGA). Therefore, the difference in PLGA molecular weight between formulations 5 and 7 did not significantly affect the release of CAB compared to other formulations with different PLGA molecular weight, such as formulations 3 and 6 (PLGA:CAB ratio 1:0.75 by weight). ).
Based on the similar and promising drug release profiles of formulations 5 and 7, the in vitro release of these formulations was extended to 180 days (Supplementary Fig. 2a). Although we observed higher in vitro release of CAB from Formulation 5 compared to Formulation 7 at approximately 2 months, an experimental study in mice showed that there was no difference in plasma concentrations of CAB in cells (Supplementary Fig. 2b). Due to the lower molecular weight of PLGA (10 kDa), formulation 7 was chosen for further studies. The lower molecular weight of PLGA resulted in a lower viscosity (Supplementary Table 2) allowing for injection. In addition, 10 kDa PLGA (recipe 7) was expected to degrade faster compared to 27 kDa PLGA (recipe 5), so this PLGA was chosen to ensure complete polymer degradation and reduce polymer accumulation at subsequent doses24,32,33. .
Finally, 500 mg/mL CAB (1:4 w/w PLGA (10 kDa): solvent) ISFI (formulation 7) (abbreviated as CAB ISFI) was further characterized according to storage microstructure due to its high drug loading capacity. agent and in vitro release rate. It is well known that the kinetics of drug release is affected by storage microstructure, which in turn depends on the polymer, solvent, physicochemical properties of the drug, and miscibility of the drug with the solvent20,24. As shown in Supplementary Fig. 3, the microstructure of CAB ISFI was qualitatively analyzed by scanning electron microscopy (SEM) and formed densely packed reservoirs with CAB crystals that remained unchanged during the 90 day incubation period. There are no visible pores in the microstructural form of ISFI. in vitro release medium. In addition, we performed scanning electron microscopy (SEM EDX) energy dispersive X-ray analysis of CAB ISFI and placebo ISFI (no CAB) to confirm CAB crystals visible on SEM images (Supplementary Fig. 4). The SEM EDX results showed no crystals in placebo ISFI and confirmed that the crystals consisted of CAB in CAB ISFI by elemental analysis (fluorine groups present, Supplementary Fig. 4). Ultimately, these results confirmed the kinetics of CAB release in vitro and in vivo over 90 days.
To determine the shelf life of the optimized ISFI CAB, we performed 30-day and 90-day stability studies under two storage conditions (40°C/75% RH and 4°C) and then analyzed after release. situation in the test tube. Stability was determined based on the appearance of the ISFI formulation (color, viscosity, phase separation) as well as the stability and concentration of the drug after storage.
CAB ISFI formulations showed no visual differences in appearance (color, injectability, phase separation) after 30 days at 4°C or 40°C/75% RH and 90 days at 4°C. Drug concentrations were comparable (<5% difference) to baseline concentration (t = 0), and no drug degradation peaks were observed by high performance liquid chromatography (HPLC) (Supplementary Fig. 5a). After 90 days at 40 °C/75% RH, the formulation was no longer injectable or homogeneous (Supplementary Fig. 5b), preventing post-storage release kinetics from being analyzed.
Figures 2a and b show the release kinetics after in vitro storage obtained after 30 days at 4°C or 40°C/75% RH and after 90 days at 4°C. Regardless of storage conditions, CAB release was slower and significantly different (p < 0.05) after storage compared to the initial formulation (t = 0). In particular, in vitro CAB release was significantly different from baseline at 7 days (p = 0.0042) and 21 days (p = 0.0004) when stored at 40°C/75% RH and 4°C, respectively. Finally, after 90 days of in vitro release, all formulations were significantly different (p<0.0001) from baseline (Figure 2a).
a Cumulative in vitro release kinetics of CAB ISFI (500 mg/ml CAB (1:4 w/w PLGA (10 kDa):solvent)) at baseline (t = 0), 30 days and 90 days after storage after 4 hours. °C and 40 °C/75% RH. All in vitro release studies were performed in phosphate buffered saline (PBS, pH 7.4 containing 2% Solutol) at 37°C. Data are presented as mean ± standard deviation for n = 3 samples. Statistical Analysis: Two-way analysis of variance with Tukey’s multiple comparisons comparing CAB release in terms of composition and time point. By day 90 after in vitro storage, CAB release from all stored formulations was significantly different and slower than baseline (p < 0.0001). b Summary of in vitro CAB ISFI release kinetics after storage a. c Effect of storage conditions on weight average molecular weight of PLGA placebo formulation as measured by gel permeation chromatography (GPC) analysis. d Summary table of PLGA degradation in placebo formulations after 30 and 90 days at 4°C and 40°C/75% RH. Solvent = 1:1 w/w NMP:DMSO. Raw data are provided in the form of raw data files.
This reduction in release kinetics can be affected by PLGA hydrolysis under stable storage conditions. PLGA degraders can crystallize during hydrolysis24 resulting in a more crystalline polymer network that delays drug release. To confirm degradation, gel permeation chromatography (GPC) analysis of the ISFI placebo formulation (1:4 w/w PLGA (10 kDa):solvent) showed a 2.3% reduction in molecular weight when stored at 4°C, while the decrease in molecular weight of MW during storage for 90 days at 40°C/75% relative humidity increased by 36.1% compared to the MW of the original PLGA (t = 0) (Fig. 2c and d). In addition, the pH of the placebo formulation was measured before and after storage. At baseline and after 90 days of storage at 4°C, the pH of the placebo formulation was neutral (pH = 6–7), while after 90 days of storage at 40°C/75 the pH became more acidic (pH = 4). ) % RH, further confirming the degradation of PLGA to its acidic by-products, especially after storage at 40°C/75% RH. Overall, CAB ISFI was more stable when stored at 4°C, as evidenced by minimal degradation of PLGA (2.3%) and the ability to maintain injectability and formulation homogeneity after 90 days.
An in vivo safety study of CAB ISFI was conducted in female BALB/c mice to assess local and systemic inflammation after injection compared to control mice that received no treatment or injections. The results of the study indicated that CAB ISFI was well tolerated and there were no overt signs of toxicity, behavioral changes, or weight loss in mice (Supplementary Fig. 6). Histopathological analysis of the resected implants and the surrounding subcutaneous tissue showed that CAB ISFI exhibited mild to moderate local inflammation manifested by immune cell infiltration around the depot (Fig. 3a). On days 3 and 7, the mean microscopic skin inflammation score was 3 (moderate inflammation), probably due to the initial immune response to the injection, and a decrease was observed in 2 mice on day 30 (Fig. 3b).
a Local inflammation of excised depots and surrounding subcutaneous tissues harvested on days 3, 7, and 30 post-injection (n = 3/time point for mice treated with CAB ISFI, n = 3/time point for control (non-injected) mice = 1/ point in time) and use it. Asterisks indicate CAB implants. Arrows indicate infiltrating immune cells and areas of inflammation. All scale bars represent 1 mm. b Inflammation in the subcutaneous tissue surrounding the depot was assessed by light microscopy and performed blind by a certified pathologist. The black bars represent the mean inflammation score at each time point (n = 3 per time point). Inflammation score: 0: inflammatory cells within expected ranges, 1: mild inflammation, slight increase, scattered immune cells, 2: mild inflammation, small clusters of immune cells, thin or localized inflammatory traces, or slight increase in cells, diffuse 3. Moderate, dense or multifocal inflammation, or a moderate number of cells diffuse around the reservoir; 4. Severe, combined inflammatory traces large enough to replace normal tissue architecture, or large numbers of cells scattered around the reservoir; 5. There is marked inflammation, an expanding area that is a replacement for normal tissue architecture. c Plasma TNF-α concentrations (pg/mL) determined by ELISA on day 3 (n=3), day 7 (n=3) and day 30 (n=5) after injection. d Plasma IL-6 concentrations (pg/ml) determined by ELISA on day 3 (n = 3), day 7 (n = 3) and day 30 (n = 5) after injection. e Plasma CAB concentrations (1215 mg/kg) over 90 days (n = 6 for each time point) (mean ± SD). The 1× and 4× PA-IC90 values for CAB are indicated by dotted lines (166 ng/mL and 664 ng/mL, respectively). Plasma samples from individual mice are shown in Supplementary Fig. 7. Raw data are provided in the form of raw data files.
Systemic inflammation was assessed by enzyme-linked immunosorbent assay (ELISA) to quantify the pro-inflammatory cytokines TNF-α and IL-6 in plasma. The results showed no systemic acute or chronic inflammation. TNF-α ranged from 0 to 20 pg/ml, which is comparable to the control group without injections (p = 0.5521) (Fig. 3c). Plasma IL-6 levels ranged from 0 to 45 pg/mL (Fig. 3d) and were comparable to those in the control group without injections [p = 0.4188 (two-way multiple comparison ANOVA)]. Variability in inflammation scores or pro-inflammatory cytokine levels can be explained by inter-individual variability in mice or hormonal cycle variability. Overall, these results indicate that CAB ISFI was generally well tolerated and considered safe, with no significant signs of toxicity or chronic inflammation.
A pharmacokinetic (PK) study was performed for 90 days in female BALB/c mice to evaluate the in vivo drug release kinetics of CAB ISFI. Plasma CAB concentrations were quantified using high performance liquid chromatography and LC/MS-MS mass spectrometry and plotted over 90 days (Figure 3e). In addition, we evaluated the kinetics of CAB release according to three mathematical models (zero order, first order, and controlled diffusion35). We determined that the observed release of CAB in mice during the 90-day pharmacokinetic study was most consistent with a zero order model (zero order model R2 = 0.97; first order model R2 = 0.87; diffusion controlled model R2 = 0.86) . ). In addition, mean plasma CAB concentrations of all mice were 6–53 times higher than 4×PA-IC90 (664 ng/mL16) during the 90-day study period (Fig. 3e and Supplementary Fig. 7).
Based on promising safety and pharmacokinetic data in BALB/c mice, the CAB ISFI formulation was selected to evaluate safety, pharmacokinetics, and efficacy in rhesus monkeys. However, since the injection volume in macaques (two 1 ml injections) is much higher than in mice (50 µl), it is critical to ensure the injectability of large formulations in vitro before scaling up studies in macaques. We used a polyacrylamide hydrogel36 to evaluate the injectability of the optimized CAB ISFI formulation. Polyacrylamide hydrogels have been shown to mimic in vivo subcutaneous tissue mechanical properties at the injection site and provide a better correlation with in vivo ISFI release when directly injected into a PBS bath36,37 compared to standard in vitro release techniques. Injectability evaluation is critical for the CAB ISFI formulation due to its rapid phase inversion properties upon injection, which is associated with the high miscibility of organic solvents with water and the low molecular weight of PLGA24. If the phase inversion occurs too quickly, the drug may solidify between the syringe and needle and block the flow of the injection.
We therefore examined the injectability of several placebo formulations with different polymer to solvent ratios and PLGA molecular weights, as well as an optimized formulation of CAB ISFI (1:4 w/w PLGA (10 kDa): solvent). The injectability of formulations in polyacrylamide hydrogels was studied using 16 gauge (G), 18 G and 19 G needles with an injection volume of 1 ml (Supplementary Table 3). 1 ml was injected into the hydrogel matrix with a 19 G PLGA 1:4 needle by weight (10 kDa) as shown in Supplementary Table 3: placebo diluent or CAB ISFI formulation due to rapid phase inversion and rapid depot formation. drug flow through the needle. This may be due to the combination of low molecular weight PLGA (10 kDa) and high solvent (1:4 w/w PLGA:solvent). On the other hand, 1 ml placebo ISFI prepared with higher molecular weight PLGA (27 kDa) or lower solvent loading (1:3 and 1:2 w/w PLGA:solvent) was successfully incorporated into the hydrogel matrix. In addition, optimized CAB ISFI (500 mg/ml 1:4 PLGA (10 kDa): diluent) was successfully injected into 1 ml hydrogel matrix using an 18G or 16G needle. Based on these results, the CAB ISFI formulation was easily administered with a 16 G needle in a macaque study.
We performed CAB ISFI on six female rhesus monkeys. All animals received two separate 1 ml injections (total 1000 mg CAB), except for one macaque (RH-1080) which received 1.0 ml and 0.5 ml ISFI or a total of 750 mg CAB. Animals received between 72.8 and 143.9 mg/kg CAB depending on body weight (median = 113.8 mg/kg). RH-42012 rhesus monkey, an otherwise healthy SHIV-infected animal from another study, was used for pharmacokinetic purposes only. On fig. 4a shows the longitudinal plasma concentration of CAB. Overall, plasma CAB concentrations at week 4 exceeded 4×PA-IC90 in all macaques except for RH-1073, which reached baseline concentrations (1230 ng/mL) at week 24. The median (range) plasma concentrations of CAB at weeks 4, 8, and 12 were 982 [406–1977], 1950 [578–5627], and 2127 [522–2552] ng/mL, or approximately 1.5–3.2 times higher. 4× PA-IC90. Removal of both CAB ISFI in RH-1097 and RH-1093 rhesus monkeys at week 12 resulted in a 7-fold and 48-fold reduction in plasma CAB levels at 72 h, respectively, and approximately 10-100-fold by week 2 . weeks after removal (Fig. 4a). Similarly, removal of one of the two ISFIs still palpable in RH-42012 rhesus monkeys at week 14 resulted in an approximately two-fold decrease in plasma CAB within one week (1550–765 ng/mL); CAB concentrations remained at 4.×PA – IC90 above 3 weeks. In the remaining three animals with intact CAB ISFI, mean plasma CAB concentrations were 1923 [534–2082], 2227 [646–2827], 1230 [971–1585] at 16, 20, 24, and 28 weeks, respectively, and 886 [886]. 473–1415] ng/mL, or 1.3–3.4 times higher than 4× PA-IC90 (Fig. 4b). Notably, plasma CAB levels in one animal (RH-1048) remained above 4x PA-IC90 (838 ng/mL) at week 47. Taken together, these results indicate that two 1 ml injections of the optimized ISFI CAB formulation can release CAB at levels above the protective threshold for up to 6-11 months in macaques.
Longitudinal assessment of plasma CAB concentration. Two CAB ISFIs were removed from macaque RH-1097 and 1093 (blue and purple solid circles) at week 12, and one of two CAB ISFIs was removed from macaque RH-42012 at week 14 (green solid circles). Dashed lines show CAB levels after removal. b An asterisk (*) indicates animals in which ISFI was removed at 12-14 weeks. Data after implant removal were not included in the calculation of the median. c CAB levels detected in plasma, rectal tissue, and vaginal tissue of three rhesus monkeys (RH-42012, RA-1048, and RA1080) using CAB ISFI. Samples were taken from the same animal at three different time points (4, 8 and 12 weeks) after implantation. The black bars represent the medians. d Ratio of CAB to plasma concentration in vaginal tissues (VT) and rectal tissues (RT).
CAB concentrations were also measured in the vaginal and rectal tissues of three rhesus monkeys (RH-42012, RA-1048 and RA-1080) at 4, 8 and 12 weeks. On fig. 4c shows that CAB was consistently detected in both tissues, with the exception of macaque RH-1048, in which CAB was not detected in vaginal tissues after 4 weeks. Median CAB concentrations in rectal tissue increased approximately 3-fold between weeks 4 and 8 (333–1004 ng/g, respectively) and decreased slightly at week 12 (713 ng/g). Median vaginal CAB concentrations also increased approximately 2.8-fold from weeks 4 to 8 (293-849 ng/g, respectively) and remained at 823 ng/g at week 12. The tissue-to-plasma ratio (Fig. 4d) remained stable over time and was similar in the vaginal [median = 0.18 (0.15-0.32)] and rectal [median = 0.42 (0.23-0.32) .43)].
To investigate whether CAB delivered from ISFI could confer rectal protection, we performed a series of experiments with SHIV infection at different times after implantation. We first assessed short-term protection in two rhesus monkeys (RH-1093 and RH-1097) challenged twice a week between 4 and 8 weeks (total 8 infections per animal) (Fig. 5a). Both animals were protected from SHIV challenge compared to untreated live control animals (RH-1092) infected after a single exposure to SHIV. Long-term protection was assessed in two additional monkeys (RH-1048 and RH-1080) exposed to SHIV twice a week between 14 and 18 weeks (8 infections per animal) (Figure 5b). One animal (RH-1048) that maintained plasma CAB levels above 4×PA-IC90 received six additional SHIV infections between 25 and 28 weeks. Two CAB-treated animals were protected from infection, while an untreated live control (RH-1084) became infected after a single exposure to SHIV (Figure 5b). Overall, one ISFI treatment completely protected 4 macaques from a total of 38 SHIV rectal exposures for 27 weeks.
Short-term protection CAB ISFI. Two CAB-treated macaques (RH-1093 and RH-1097) and one untreated macaque (RH-1092) were exposed rectally to SHIV between 4 and 8 weeks post-implantation. Each animal was challenged rectally with SHIV twice a week for up to 4 weeks (8 exposures in total). Arrows indicate exposure to SHIV. ISFI was surgically resected at week 12 (blue and purple solid circles). b Plasma levels of SHIV RNA in animals were determined by real-time PCR during the challenge period and the 32-week observation period. c Long term CAB ISFI protection. Two CAB-treated macaques (RH-1080 and RH-1048) and one untreated macaque (RH-1084) were subjected to rectal SHIV stimulation twice a week between 14 and 18 weeks post-implantation (maximum 8 exposures) . The RH-1048 animal underwent six additional SHIV exposures between 25 and 28 weeks (14 exposures in total). Arrows indicate exposure to SHIV. d Animal plasma levels of SHIV RNA by real-time PCR during the challenge period and during the observation period of 24 to 36 weeks.
To assess safety and tolerability, we examined the area around the implant weekly for 12 weeks or until the implant was removed. The evaluation included six rhesus monkeys receiving two injections for a cumulative analysis of 12 implantation sites and a total of 144 clinical observations. During the 12-week study period, all implant sites were normal and showed no signs of local skin reactions according to the Draize score (Fig. 6a and Supplementary Fig. 8). At the time of implant removal (12–14 weeks post-implantation), a semi-quantitative histopathological evaluation was performed on 3-mm-thick skin biopsies taken from the site of implantation in three monkeys. For comparison, 3 mm thick biopsies were taken from animals that did not receive CAB ISFI. Skin biopsy showed no or minimal lymphoplasmacytic infiltration in the animals (including controls), and there were no signs of infection or foreign bodies (residual implants) in any tissue sections (Figure 6b and Supplementary Table 4).
a Heat map of local skin reactions at the implant site in six ISFI-treated animals. Local skin reactions were assessed according to the Draize scale (from 0, no to 4, severe). b Histopathology of the skin. When implanted in treated animals RH-1093, RH-1097, RH-42012 (n=3) and untreated controls (n=1), one skin biopsy was taken per animal. All scales correspond to 1 mm (H&E, original magnification × 4).
To assess the retrievability of implants after in vivo studies, we performed CAB ISFI removed from mice through a small skin incision. ISFIs were successfully removed from all animals without fibrous tissue surrounding the depot (Fig. 7a). Depots removed on day 30 (n = 6), day 60 (n = 6), and day 90 (n = 6) were further processed to evaluate polymer degradation (n = 3/time point) and residual CAB concentrations (n = 3/time). After 90 days in mice, the results of these assays showed a 59.5 ± 13.4% reduction in depot mass (Figure 7b) and a 49.7 ± 5.4% reduction in PLGA molecular weight (Figure 7c). Importantly, 61.6 ± 6.5% CAB remained in the implants extracted from mice at day 90, demonstrating that ISFI-released CAB can persist for more than 90 days (Figure 7d).
CAB ISFI images obtained from BALB/c mice 30, 60 and 90 days post-injection. b CAB ISFI mass (mean ± SD; n = 6/time point) at 30, 60 and 90 days post-injection in mice compared to baseline ISFI mass (day 0) for 50 µl (60.75 mg) injection. weight on day 0 was calculated from the injection volume of 50 μl, and the density of the formulation (1.215 g/ml) was used to determine the approximate weight at the time of injection. c Quantification of residual CAB in ISFI in mice at 30, 60, and 90 days post-injection compared to the starting dose (25 mg in 50 μL injections) (mean ± standard deviation of n=3 samples). d Reduction in molecular weight of PLGA in CAB ISFI at 30, 60, and 90 days post-injection in mice compared to pure PLGA (10 kDa) (mean ± SD of n = 3 samples). e Quantification of drug residue from implants (placed in the left upper back and right upper back) taken from three rhesus monkeys 84 and 98 days after injection. Raw data are provided in the form of raw data files.
In addition, two CAB ISFIs were taken from two monkeys (RH-1097 and RH-1093) on day 84 post-injection, and one ISFI was taken from a third monkey (RH-42012) on day 98 post-injection. As shown in Figure 7e, after removal from storage, an average of 48.32 ± 11.44% CAB remained per implant.
Individual CAB clearance was estimated from our extravascular PK profiles in six monkeys (Supplementary Fig. 9) given 1.5–2 ml subcutaneous CAB ISFI [median (IQR) = 15.9 (9.1–25.2) ml/ (h * kg)] and nine reference macaques treated intramuscularly with 50 mg/kg CAB LA 7 and 1 day before PK sampling [median (IQR) = 12.4 (11.7–14.3) ml/ (h*kg)]38. The administered doses for the two long-acting formulations were calculated by multiplying the observed plasma concentrations by the body weight of our six ISFI-treated monkeys at the time of dosing, or the estimated clearance of the respective animals adjusted after adoption of the historical reference macaque body. a weight of 8 kg was obtained (Fig. 8). Mean parameter estimates for 39 simulated human plasma models using the published PC models of the CAB population (CL = 151 ml/h, V2 = 5270 ml, Q = 507 ml/h and V3 = 2430 ml ) were valid Target infusion rates for clinical doses Concentrations at various intravenous infusion rates. Infusion rates of 3 and 0.75 mg/day resulted in plasma concentrations above 4x PA-IC90 and 1x PA-IC90, respectively. The mean infusion rate of CAB ISFI observed in our macaque study exceeded this clinical efficacy threshold of 3 mg/day on days 28 to 140 post-dose. Assuming the infusion rate increases in proportion to the volume administered, an injection of 3 ml CAB ISFI will reach this threshold between days 21 and 168, or 5.6 months post-dose. Compared to macaques that received two intramuscular injections of CAB LA 50 mg/kg 6 days apart, CAB ISFI persisted above the predicted protective threshold for an average of 97 days (Fig. 8b).
The infusion rate of CAB ISFI was estimated by multiplying the observed plasma concentration by the clearance rate (determined from the extravascular PK profile) for each respective animal. The dotted and dotted control lines indicate infusion rates of 3 and 0.75 mg/day, which are expected to result in plasma concentrations above 4× PA-IC90 and PA-IC90 in humans, respectively. b Median (±IQR) CAB ISFI injection rate observed in this study for a (green line) superimposed on the predicted median rate for a 3 ml injection volume (assuming a proportional increase in injection rate with volume; purple line) and n. UP value = 9 control monkeys treated with 50 mpk (mg/kg) CAB LA38 intramuscularly 7 days and 1 day prior to PK sampling.
We report an ultra-long-acting CAB ISFI that extends dosing intervals at low injection volumes, provides strong protection against SHIV rectal infection, and allows product reconstitution. We demonstrate CAB release above the established PrEP baseline protection for up to 6–11 months in rhesus monkeys and use this data to assess clinical impact. We found that injections of 2 ml or 3 ml of our preparation maintained plasma concentrations above 4× PA-IC90 in humans for 4 or 5 months, respectively.
CAB ISFI remained injectable, showed no drug degradation, and maintained a constant CAB concentration for up to 30 days at 40°C/75% RH and up to 90 days at 4°C, demonstrating formulation stability. Although release studies after in vitro storage showed slower release of CAB under both storage conditions, the difference in cumulative release of CAB between baseline and after 90 days of storage at 4°C was only about 3%. We also conducted studies on the injectability of ISFI in polyacrylamide hydrogels to investigate whether the rapid phase inversion property of ISFI leading to potential implant curing could be a limitation associated with high injection volumes (≥1 ml). We have demonstrated that 18 G or 16 G needles can be used for injection volumes ≥ 1 ml. In BALB/c mice, plasma CAB concentrations were 53 times higher than 4x PA-IC90 for 90 days. In rhesus monkeys, CAB ISFI maintained plasma CAB concentrations above 4×PA-IC90 for 6-11 months, in contrast to previously described long-acting CAB preparations that recorded plasma CAB levels above protective baseline. The time was much shorter40,41,42,43. Our in vivo safety studies in female BALB/c mice and female rhesus monkeys showed that CAB ISFI was well tolerated, with no significant local or systemic levels of inflammation. In addition, the implants remained palpable in mice and macaques up to 90 days post-injection, demonstrating that they can be easily removed in the event of an adverse event, which may eliminate the need to cover the long pharmacological tail with oral PrEP, as in this case, the current CAB LA formula is also used. . We also noted that approximately 60% of CAB remains in implants in mice and between 30% and 60% in each implant in macaques after recovery at 84-90 days post-injection, suggesting that ISFI CAB may be released for longer periods. time. time.
In rhesus monkeys, a dose of 50 mg/kg CAB LA for injection provided plasma concentrations of CAB above the established PrEP baseline protective level for about 6 weeks 44 . Here we show that two 1 ml CAB ISFI injections containing only 1.5-2.9 times CAB per kg can maintain plasma CAB levels above baseline for 6-11 months. In rectal and vaginal tissues, CAB concentrations were also in the range observed in rhesus monkeys treated with CAB LA and were approximately 2 to 4 times higher than in humans achieved with a clinical dose of 600 mg CAB LA, despite potential limitations. our analysis included only the first 12 weeks after implantation44,45. In addition, ultra long acting CAB ISFI has significant advantages over other long acting CAB formulations in pre-clinical development. Karunakalan et al. reported a long acting hydrophilic poly(etherurethane) radiopaque CAB implant that exceeded 1x plasma PA-IC90 for 12 weeks in rhesus monkeys but less than 4xPA-IC90 at about 2 weeks after 6 implants per macaque. 41. Each implant contains approximately 274 mg of CAB, releasing an average of 348 micrograms of CAB41 per day in the body. Zhou et al. and Kulkarni et al. reported an injectable nanoformulated CAB prodrug that demonstrated sustained release of CAB in macaque monkeys (45 mg CAB equivalent/kg, IM volume ~2 ml), however plasma CAB concentrations were below 4× PA-IC90 and PA-IC90 after ~60 days. 42 and 200 days 43 respectively. The injectable and biodegradable CAB ISFI described in our study demonstrated plasma CAB levels above 4x PA-IC90 in macaque monkeys for 6-11 months and was 100% protected with just two 1 ml subcutaneous injections. Protected against multiple rectal SHIV.
Our results, showing 100% protection in macaques over several months, represent the longest recorded PrEP activity observed with a single CAB administration to our knowledge. We limited SHIV exposure to periods when plasma CAB levels exceeded 4×PA-IC90, as this concentration represents the accepted benchmark for PrEP protection in macaques and humans. Thus, the overall protection observed despite a cumulative 38 rectal SHIV exposures was not entirely unexpected. Further determination of the duration of action of PrEP CAB ISFI is important as levels above 1x PA-IC90 have also been associated with significant (>95%) protection of macaques against rectal SHIV challenge40,46. This analysis may also provide information on potential reduction in doses or volumes of CAB ISFI injections and potential extension of protection beyond the predicted 5.6 months after 3 ml injection in humans. However, evidence of rare breakthrough infections in people receiving CAB LA and maintaining target levels may argue against lowering CAB doses 47, 48 .
Analysis of plasma CAB concentration in rhesus monkeys after implant removal at 3 months also provided important information on ISFI recovery. In two monkeys, removal of both implants resulted in a rapid 10–100-fold decrease in plasma CAB levels over a 2-week period, although continuous monitoring of CAB showed significant residual implant material in one animal with plasma concentrations higher than PA-IC90. In another macaque, removal of one of the two implants resulted in a rapid two-fold drop in plasma CAB levels within a week. In all cases, approximately half of the CAB dose remained in the reconstituted ISFI, indicating the possibility of a longer release of CAB ISFI. Further research is needed to determine the window of opportunity for deletion and to understand when CAB is depleted from ISFI. Although the proposed CAB ISFI has shown great potential for use in HIV PrEP, its formulation still has some limitations. For example, although we did not observe any signs of toxicity or adverse events in our safety study in mice, our results were limited due to the study’s small sample size (n = 3). Therefore, future studies will include comprehensive safety studies in mice with large sample sizes and long-term safety evaluations up to 180 days. Another limitation is that CAB concentrations in rhesus monkeys begin to decline during the first two weeks after injection, and PrEP activity may be lost when levels fall below 4×PA-IC90. If this low release period is confirmed in a human, this would mean an alternative HIV prophylaxis recommended 28 days after the first injection of 2 ml CAB ISFI or 21 days after the 3 ml injection. Current CDC guidelines on PrEP recommend using alternative HIV prevention methods for 7–20 days after starting daily oral PrEP, depending on the potential route of transmission49. Another potential limitation is that the formulation contains organic solvents which can cause toxicity in large amounts. When administered intravenously to mice, the median lethal dose (LD50) and no observed effect level (NOEL) of NMP were 3600 mg/kg and 257 mg/kg50, respectively. Alternatively, in non-human primates, the LD50 and NOEL for DMSO are 4000 mg/kg and 3300 mg/kg, respectively, while the LD50 in mice is 14,000 mg/kg51 in blood. In our study, mice (700 mg/kg of each vehicle) and rhesus monkeys (71 mg/kg of each vehicle) were administered approximately 300 mg/mL of each vehicle with no evidence of chronic or overt toxicity. Although ISFI was well below the LD50 and NOEL limits for DMSO and NMP in macaques, future research should include solvent reduction efforts to reduce potential concerns about toxicity.
Additional studies to address these limitations and further advance this study include (1) increasing the release, dosing, or oral administration of CAB ISFI to ensure that plasma concentrations remain above 4×PA-IC90 during treatment and after injection. the first month (2) to determine the time to complete release of CAB and complete degradation of the polymer to assess the complete pharmacokinetic profile in vivo and (3) to add barium sulfate to create a radiopaque implant for x-ray monitoring. To facilitate extraction, if necessary, and investigate the migration of the implant. In addition, we have demonstrated the ability to load multiple ARVs into a single ISFI, resulting in ultra-long release kinetics in mice. Thus, additional future research may include adding another antiretroviral drug to the currently proposed CAB ISFI formula for therapeutic applications against HIV.
Taken together, these results highlight the promise of CAB ISFI as an ultra-long-acting platform for delivering PrEP and supporting its clinical progression in humans. The observed preclinical safety and extended pharmacokinetic bioequivalence of CAB LA is very promising and, if proven clinically, could lead to a fast track to regulatory approval without the need for efficacy testing. We expect CAB ISFI to be very popular and committed as it transitions to a clinical setting. Studies have shown that long-acting injectables generate the highest acceptance and adherence among users in areas with the highest HIV prevalence compared to other potential dosage forms (eg, daily oral tablets and vaginal rings)5,10. ease of use and privacy. In addition, due to subcutaneous injection, the proposed formulation can be self-administered, which can further enhance user acceptance and ease of use. In general, ultra-long acting CAB ISFI can be administered two to three times a year in a clinical setting and can advance the HIV field and expand prevention opportunities worldwide.
50:50 Poly(DL-lactide-co-glycolide, PLGA) was purchased from LACTEL (Birmingham, Alabama; catalog number B6010-1P, lot number A17-142, molecular weight 27.2 kDa, intrinsic viscosity 0.26-0 .54 dl/g, catalog number B6017-1G, batch number A15-108, molecular weight 10 kDa, intrinsic viscosity 0.15-0.25 dl/g). N-methyl-2-pyrrolidone (NMP, USP) was purchased from ASHLAND (Wilmington, Delaware, product code 851263, 100% NMP). Dimethyl sulfoxide (DMSO, >99.7%) was purchased from Fisher Scientific (Waltham, MA). Solutol-HS 15, phosphate buffered saline (0.01 M PBS, pH 7.4) and acetonitrile (ACN) for HPLC and water were purchased from Sigma Aldrich (St. Louis, MO). Gelucire 44/14 (catalog number 1356950) was purchased from Sigma Aldrich (St. Louis, MO). Tween 20 was purchased from Fisher Scientific (Hampton, NJ; p/n BP337-100). Tetrahydrofuran (THF) was purchased from Sigma Aldrich (St. Louis, MO; catalog number SHBF9530V). 98.5% acrylamide (AM) and ammonium persulfate (APS) were purchased from Acros Organics (Carlsbad, CA; AM p/n 16435000, APS p/n 40116-1000) and a 2% solution of bisacrylamide and ‘, N’- tetramethylethylenediamine (TEMED) was purchased from Fisher Scientific (Hampton, NJ; bisacrylamide cat. no. BP150-20, TEMED cat. no. BP1404-250). High purity CAB (≥99%) was purchased from Selleckchem (Houston, TX; catalog number S7766) and sterile filtered DMSO (≥99.7%) was purchased from Sigma-Aldrich (St. Louis, MO; lot number RNBH5297) .
Reverse phase HPLC analysis was performed on an Agilent 1260 HPLC system (Agilent Technologies, Santa Clara, CA, USA) equipped with a diode array detector and an autosampler LC pump. The stationary phase used for CAB analysis was an Inertsil ODS-3 column (5 µm, 4.6 × 150 mm, 100 Å, [GL Sciences, Torrance, CA]) maintained at 40°C. Chromatographic separation was achieved by gradient elution using a mobile phase consisting of 0.1% trifluoroacetic acid in water and acetonitrile (H2O/ACN 95:5 v/v). The flow rate was 1.0 ml/min and the total assay time was 25 minutes per 25 ml injection. CAB was analyzed at 254 nm with a retention time of 11.4 minutes. The assay has a calibration range of 0.48-250 µg/mL. HPLC data were collected using Agilent OpenLab software (version C.01.08).
CAB in various weight ratios (w/w) NMP, DMSO, NMP:DMSO (1:1, 9:1 and 2:8 w/w), NMP:Gelucire 44/14 9:1 (w/w), 9 :1 w/w (NMP:DMSO): Tween 20 (Supplementary Table 1) to determine the optimal solvent in an ISFI system for maximum CAB loading. CAB saturation solubility was also measured in release medium (PBS with 2% Solutol) to ensure immersion conditions (less than 1/552 of maximum solubility) during in vitro release studies (Supplementary Table 1). The addition of Solutol to the release medium has been shown to increase the solubility of hydrophobic drugs in the release medium, ensuring that uptake conditions are maintained during in vitro release studies20. The addition of 2% Solutol to PBS increased the CAB saturation solubility by 4.3 times compared to PBS alone.
To determine the saturation solubility of CAB in the following solvents shown in Supplementary Table 1, 100 mg of CAB were weighed into separate vials and 100 mg of each solvent was added. The mixture was stirred at 37°C for 24 hours. Samples were centrifuged at 16,000 g (Eppendorf Centrifuge 5415R, USA) for 30 min to remove excess undissolved drug. Sample aliquots (n = 3) were collected from the saturated supernatant and diluted with acetonitrile. Drug concentrations in saturated aliquots were determined by HPLC analysis.
50:50 poly(DL-lactide-co-glycolide) (PLGA) (molecular weight 27 kDa or 10 kDa) with a weight ratio of 1:1 (w/w) NMP:DMSO in a ratio of 1:2 or 1:4 Mix with w in 7 ml scintillation vials and mix until dissolved at room temperature to obtain a homogeneous placebo formulation. Cabotegravir (CAB; 100-400 mg/g) was then added to an appropriate polymer/solvent solution and the drug was allowed to dissolve and obtain a stable ISFI solution or suspension at its target drug level. To ensure complete dissolution of the drug in the placebo formulation and homogeneity of the drug solution or suspension, sample aliquots (1–2 mg, n = 4) were taken from the drug formulation and dissolved in ACN. Drug concentrations were quantified by HPLC analysis.
Post time: Mar-20-2023