Optimizing dosing strategies for post-kala-azar dermal leishmaniasis: a geographical comparison of systemic and skin pharmacokinetics and pharmacodynamics of antileishmanial drugs
Wan-Yu Chu (1)(2), Eugenia Carrillo (3), Fabiana Alves (4), Alwin D.R. Huitema (1)(5)(6), Thomas P.C. Dorlo (1)(2)
(1) Department of Pharmacy and Pharmacology, Netherlands Cancer Institute, Amsterdam, the Netherlands; (2) Department of Pharmacy, Uppsala University, Uppsala, Sweden; (3) WHO Collaborating Centre for Leishmaniasis, National Centre for Microbiology, Instituto de Salud Carlos III, Madrid, Spain; (4) Drugs for Neglected Diseases initiative (DNDi), Geneva, Switzerland; (5) Department of Pharmacology, Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands; (6) Department of Clinical Pharmacy, University Medical Centre Utrecht, Utrecht University, Utrecht, the Netherlands
Objectives:
Post-kala-azar dermal leishmaniasis (PKDL) is a dermatological complication that arises after the successful treatment of the neglected tropical disease visceral leishmaniasis (VL) as a result of evading Leishmania parasites to the skin. The skin lesions in PKDL patients serve as reservoir for transmission of leishmaniasis, representing a major public health issue. Treating PKDL remains challenging due to limited understanding of disease pathology, inadequate knowledge on drug exposure in skin, and the lengthy treatment periods leading to poor patient compliance. Moreover, discrepancies in clinical manifestations across geographical regions further complicate disease management. In Sudan, patients mainly have papular and nodular lesions, with a large number resolving spontaneously, while on the Indian subcontinent (ISC), most patients present with macular lesions that require drug treatment [1].
Two unique clinical trials were recently conducted in Sudan and the ISC, investigating shortened regimens combining oral miltefosine (MF) with parenteral liposomal amphotericin B (LAmB) or paromomycin (PM). These trials marked the first evidence of target site drug exposure in the skin. In Sudan, the trial achieved an initial one-year cure rate of 80%, whereas in the ISC, the corresponding rate was 30% [2][3]. The reasons for geographical differences in treatment response remain unclear, possibly due to variations in drug exposure, parasite clearance, and/or pathophysiology and host response. To elucidate these geographical differences in drug response, a drug-parasite-host pharmacokinetic-pharmacodynamic (PK-PD) model was developed, aiming to:
1) Characterize the systemic and skin target-site PK of MF, LAmB and PM.
2) Identify the PK-PD of combination regimens on parasite clearances in the skin, and subsequent clinical skin lesion recovery.
3) Assess potential geographical differences in PK and PD characteristics between Sudan and the ISC.
Methods:
PK and PD data originated from clinical trials conducted in Sudan [2] and the ISC [3]. In Sudan, patients received IV LAmB (total 20mg/kg, 4 doses in 7 days) plus oral MF (allometric dosing, 28 days) or IM PM (20 mg/kg, 14 days) plus oral MF (allometric dosing, 42 days). In the ISC, patients received only IV LAmB (total 20mg/kg, 5 doses in 15 days) or a combination of LAmB with oral MF (allometric dosing, 21 days). In both trials, one skin biopsy for PK assessment was taken per patient at the end of the treatment (EOT). PD markers were evaluated at baseline and during the 12-month follow-up, including skin parasite load measured by quantitative real-time PCR (qPCR) and PKDL lesion size quantified using the referred scoring system representing the affected areas [4].
Previously developed PK models for MF, LAmB and PM in patients with PKDL were used as starting points for model development. The plasma PK of MF was described by a two-compartment model, incorporating an effect of cumulative dose on bioavailability (F). The plasma PK of LAmB was characterized by a two-compartment model, with a saturable distribution towards the mononuclear phagocyte system, affected by the parasite dynamics. The plasma PK of PM was described by a three-compartment model, with clearance mediated by glomerular filtration rate and saturable tubular reabsorption through megalin receptors. Population PK-PD analysis was performed with NONMEM 7.5.
Results:
The only geographical difference found in plasma PK of all three drugs was in the baseline F of MF, which was 14% lower in patients from Sudan compared to those from the ISC. Concentrations of MF and LAmB in skin at the EOT were comparable between Sudan and the ISC. The PK within the skin target-site was modeled as an effect compartment, characterized by the skin-to-plasma concentration ratio (Rs-p) and the distribution rate constant from plasma to skin (kp-s). For MF, LAmB, and PM, Rs-p were 1.48, 0.98, and 0.45, respectively. kp-s was estimated at 0.0055 h-1 for LAmB, and was fixed to a high value for MF and PM, assuming an immediate distribution.
Baseline skin parasite qPCR was estimated at 93 and 390 p/µgDNA in patients from Sudan and the ISC, respectively, with a high between subject variability (BSV) of 200%. Within the treatment period, a 100-fold decrease in skin parasite qPCR was generally observed across all treatment arms. At the EOT, skin parasite qPCR was undetectable in 95% and 48% of patients in Sudan and the ISC, respectively. A turnover model was used to describe the dynamics of parasites, wherein drug-induced parasite clearances were associated with drug concentrations in the skin, scaled by a drug-specific slope factor (λdrug). λMF, λAmB and λPM were 1*10-4, 9.9*10-4,and 6.8*10-4 h-1*mg/L, respectively, with no difference identified between regions.
The median initial lesion size score was 23 units in Sudan and 165 units in the ISC. No clear relationship between skin parasite qPCR and skin lesion size was observed at baseline and during follow-up. Therefore, the healing of lesions was hypothesized to be inhibited by the presence of parasites, with subsequent drug-induced clearance of parasites leading to long-term slow skin recovery. Besides, the potential effects of drugs on skin lesion recovery were evaluated. The inhibitory effect of parasites on lesion recovery was characterized using a sigmoid Emax function. In the absence of parasites, the half-life of lesion recovery was estimated at 15 days in Sudan and 176 days in the ISC. Within the same region, no significant difference in the rate of lesion recovery was found across lesion types. In both regions, the median parasite AUC in the skin until EOT was around 25% lower in patients clinically cured at one year compared to those not cured. During the one-year follow-up period, the skin parasite load remained below the potential transmission threshold of 416 p/μgDNA for 90% of the time in the LAmB monotherapy arm, while it exceeded 99% for all the combination treatment arms.
Conclusions:
For the first time, the PK of MF, LAmB, and PM within skin tissue were described, suggesting comparable target-site exposures between PKDL patients in Sudan and the ISC. Notably, MF showed the highest skin-to-plasma ratio and LAmB had the longest residence half-life in skin. All treatment combination effectively reduced skin parasites, while at EOT, the parasite load remained higher in ISC patients due higher baseline levels. Geographical variations in treatment response are likely due to differences in disease manifestation and host responses associated with subsequent healing of the skin lesions. This raises concerns about the use of similar PKDL lesion quantification system and treatment endpoints across regions. Nevertheless, shortened treatments in Sudan and the ISC adequately prevent parasite transmission and restored skin lesion recovery, suggesting promising alternatives to current lengthy PKDL treatments. Moreover, skin parasite AUC might be a potential marker for earlier assessment of drug effects.
References:
[1] Zijlstra, E. E. et al. Post-kala-azar dermal leishmaniasis. Lancet Infect. Dis. 3, 87–98 (2003).
[2] Younis B.M. et al. Safety and efficacy of paromomycin/miltefosine/liposomal amphotericin B combinations for the treatment of post-kala-azar dermal leishmaniasis in Sudan. PLoS Negl Trop Dis. 17(11):e0011780 (2023).
[3] A phase II, non-comparative randomised trial of two treatments involving liposomal amphotericin B and miltefosine for post-kala-azar dermal leishmaniasis in India and Bangladesh. https://dndi.org/research-development/portfolio/new-treatments-pkdl.
[4] Mondal, D. et al. Study on the safety and efficacy of miltefosine for the treatment of children and adolescents with post-kala-azar dermal leishmaniasis in Bangladesh. BMJ Open 6, e010050 (2016).