Development of a predictive model for meglumine antimoniate treatment failure in patients with cutaneous leishmaniasis: Aretrospective cohort study

Susana Ríos-Echavarría; Gilma Hernández Herrera; Hector Ivan García García; Liliana López-Carvajal; Lina Maria Serna-Higuita

doi:10.1371/journal.pntd.0014606

. 2026 Aug 4;20(8):e0014606. doi: 10.1371/journal.pntd.0014606

Development of a predictive model for meglumine antimoniate treatment failure in patients with cutaneous leishmaniasis: Aretrospective cohort study

Susana Ríos-Echavarría ^1,², Gilma Hernández Herrera ², Hector Ivan García García ², Liliana López-Carvajal ¹, Lina Maria Serna-Higuita ^3,^*

Editor: Guilherme L Werneck⁴

PMCID: PMC13436726 PMID: 42550827

Abstract

Background

Although alternative therapies for cutaneous leishmaniasis (CL) are available, systemic meglumine antimoniate (MA) remains the first-line treatment in many endemic regions. Its use is nevertheless associated with serious adverse effects and a high risk of treatment failure (TF). This study aimed to identify clinical and sociodemographic risk factors for TF following systemic MA therapy, which may enhance therapeutic decision-making and improve clinical outcomes.

Methodology

We evaluated a retrospective cohort of 296 patients with CL treated with MA between 2007 and 2024. A multivariable logistic regression model was performed using candidate variables selected via clinical relevance, biological plausibility, stepwise selection and least absolute shrinkage and selection operator (Lasso) regression. Model performance was assessed through discrimination, calibration, and internal validation. Results were reported as odds ratios, 95% confidence intervals and p-values.

Principal Findings

All included patients received first line MA therapy and completed at least six months of follow-up. Independent predictors associated with TF were age, occupational activity, size, number and anatomical location of lesions, clinical form, regional lymphadenopathy, and prior history of leishmaniasis. The final model showed moderate overall performance with a Hosmer-Lemeshow p value = 0.667, AUC = 0.691, and a Brier score of 22.8. Internal validation yielded a Harrel C = 0.608.

Conclusions

The present study demonstrates an association between TF and socio-demographic, clinical variables. Identifying these risk factors may support clinical decision-making and contribute to optimizing treatment outcomes.

Author summary

Cutaneous leishmaniasis is a parasitic disease that causes skin lesions that remains an important health problem in many parts of the world. Although alternatives therapies exist, systemic meglumine antimoniate remains the most used treatment in many countries. However, its clinical utility is frequently limited by severe adverse effects and a high risk of therapeutic failure we evaluated a historical cohort of patients treated with meglumine antimoniate between 2007 and 2024 to identify predictors of treatment failure. Our findings show that treatment was less successful in patients who were younger, living in rural place, prolonged disease duration, with multiple lesions, and prior history of leishmaniasis. Recognizing these factors can guide clinician in selecting optimized treatments and improve care for people with cutaneous leishmaniasis.

Introduction

Leishmaniasis is a zoonotic disease caused by parasites of the genus Leishmania, and transmitted by Lutzomyia vector of the subfamily Phelobominae [1,2]. In Colombia, around 11 million people are at risk of Leishmaniasis infection, with rural workers with low socioeconomic conditions being the most affect. This population additionally faces health care barriers which limit the access to early diagnosis and treatment availability [3–7]. Among the various clinical manifestations, cutaneous leishmaniasis (CL) is the most prevalent form [8], with an incidence of 18.6 cases per 100,000 inhabitants in 2022 [9].

Systemic meglumine antimoniate (MA) has been used as first-line treatment in patients with CL for over 70 years. However, its use is related with a wide range of side effects [2], including cardiac, renal, hepatic and pancreatic toxicity [10,11]. MA has been also associated with an increasing risk of therapeutic failure (TF), with an estimated prevalence of 23.4% across Latin America [12–14]. Previous studies have identified several factor associated with TF like age, [13,15,16], lesion size number [17] and duration of skin lesion [12], Leishmania species [2,13,17–20], geographical location [2], compliance therapy [12], presence of lymphadenopathies [12], weight and previous history of leishmaniasis [20,21].

The Pan American Health organization (PAHO) currently recommends classifying CL treatment as local or systemic based on lesion characteristics and the feasibility of patient follow up [22]. Recent studies has reported adequate cure rates with the use of miltefosine as the first systemic treatment option in patients with CL [23]. However, MA remains widely used in our Colombia due to political and logistical difficulties, as well as the lack of knowledge about the infection among healthcare professionals [24].

Identifying socio-demographic, clinical, and environmental risk factors for TF can improve clinical decision-making and patient outcomes in CL. A predictive model in Peru using polymerase chain reaction (PCR)-based Leishmania species identification showed high accuracy, however, that study excluded non-ulcerated or mixed lesions and which often have higher parasite loads and may be linked to TF [18]. Additionally, PCR is not widely accessible in primary healthcare settings, limiting the model’s practical utility [18]. Our study aimed to identify risk factors related to TF to MA, using variables easily collected by healthcare providers. Which enables the development of a predictive risk model to support clinical decision-making and optimize therapy selection for patients diagnosed with CL.

Materials and methods

Ethics statements

This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the International Council for Harmonization Good Clinical Practice Guidelines. The study protocol was reviewed and approved by the Review Board (IRB) approval of the University of Antioquia. As part of their routine medical care, all patients provided written informed consent that included authorization for the secondary use and analysis of their clinical data for research purposes. Any such secondary use, including future related research or pooled analyses, complies with Institutional ethics standards and relevant data protection regulations. To ensure participant confidentiality and privacy, all data were fully anonymized prior to analysis, and no identifiable personal information was accessed or stored.

Study design and population

This single-center retrospective cohort analysis included patients with diagnosed with CL who were treated at PECET (Programa de Estudio y Control de Enfermedades Tropicales, Antioquia University, Medellín - Colombia) between 2007 and 2024. All patients received MA (20 mg/kg/day) as first-line therapy for up to 20 days. Patients with mucosal or visceral Leishmaniasis, those who did not agree to sign the informed consent and patients with post treatment follow-up less than 3 months were excluded.

Data collection

Socio-demographic, clinical, and post treatment follow up information was extracted from the electronic health records of PECET. Sociodemographic variables included age, gender, economic activity, place of residence and comorbidities. Clinical variables comprised the site of CL infection, duration of skin lesion in days, number and lesion size (area). For ulcerated lesions, the area was calculated using the rhomboid formula (major diagonal × minor diagonal/ 2), whereas for non-ulcerated lesions, the area was calculated using the rectangle formula (base × height). Additional variables included anatomical location of lesions, presence of regional lymphadenopathy, and history of previous leishmaniasis episodes. Occupational activity was categorized as high-risk occupations—including military personnel and agricultural workers—and low-risk occupations—including individuals with higher education, independent workers such as drivers, miners, self-employed individuals, and construction workers. Adherence to treatment was defined as receiving at least 90% of the prescribed regimen, corresponding to a complete treatment dose of 20 mg/kg/day (maximum 15 ml per day) for 18 days. Adherence was evaluated at the end-of-treatment visit, during which patients returned all medication blisters to their physician for verification.

Outcomes

The primary outcome of this study was TF to MA, defined as the presence of at least one of the following criteria three months post treatment: 1) a ≥ 50% increase in lesion size, 2) lesion progression after initial regression (including enlargement of a plaque, nodule, or ulcer), or 3) incomplete re-epithelialization and/or persistent inflammatory signs such edema or erythema [9,25].

Sample size estimation

Following the recommendation proposed by Riley et al., [18,25,26], an assuming an R² Nagelkerke of 0.54 derived from a previous prognostic model (estimated by the Cox-Snell R²) [18,25,26], a shrinkage factor of 0.9, a TF prevalence of 23.4% [12–14], and a maximum of 12 candidate predictors. The required sample size was estimated to be 286 participants.

Statistical analysis

Continuous variables were summarized as mean and standard deviation (SD) or median and interquartile range (IQR) depending on their distribution. The distribution of continuous variables was assessed using kurtosis, skewness, Q-Q plots, and histograms. Categorical variables were reported as absolute and relative frequency.

Variable selection was guided by univariate binary logistic regression model and clinical reasoning [26,27]. Additionally, least absolute shrinkage and selection operator (Lasso) regression model was used to identify relevant predictors by shrinking coefficients toward zero [26,28,29]. Multivariable binary logistic regression models were carried out based on the selected variables. A total of five models were evaluated before selecting the final model. The linearity relationship between continuous predictors with the log-odds of the outcome was evaluated using spline regression models [30] (S1 Fig). Multicollinearity among predictors was assessed using the variance inflation factor (VIF) [31]. Results are reported as crude (univariate regression model) and adjusted (multiple regression model) odds-ratios (OR) with their corresponding 95% confidence intervals (CI). Based on the final model, a risk score was developed. Each variable included in the final model was assigned a weight according to its beta coefficient, which was multiplied by 10 and rounded to the nearest integer to obtain the corresponding point value.

Model fitting was evaluated using the Hosmer-Lemeshow [34,35], R² Nagelkerke, Akaike information criterion (AIC) and the chi-square goodness of fit statistic [32]. Discrimination was evaluated by calculating the receiver operator characteristics (ROC) and the concordance Harrell’s C-index [26,30]. Calibration was evaluated using calibration curves and Brier scores [30]. Internal validation of the final predictive model was conducted via bootstrapping with 1,000 resamples [27].

Missing data were assumed to be at random [27] and multiple chained equation imputation was used to replace lost data with plausible values, based on the observed data. Fifty imputations were implemented and the estimates were pooled according to Rubin’s Rules [33,34]. For all analyses, a 2-tailed P value of <0.05 was considered statistically significant. SPSS software (version 28.0; IBM, Chicago, IL) and R (version 4.3) software were used.

Results

A total of 1181 patients were evaluated between 2007 and 2024. Of these, 885 patients were excluded for reasons outlined in Fig 1. A total of 296 patients were included in this study. The baseline characteristics of the cohort are presented in Table 1. Among 296 enrolled patients, 44.2% (n = 131) had TF. The mean age was 26.2 years (±12.2 years, min – max: 0.5 – 66 years), with males predominating at 86.8%. CL was frequently diagnosed via direct parasitological examination (273/296 patients), whereas only 28 patients underwent PCR to identify the infectious species. When comparing patients with and without TF, differences were observed in age (p = 0.012), and duration of skin lesion (p = 0.029). Regarding occupational activity, individual working in professional fields (Administrative, social, human health, and animal health sciences fields) exhibited a trend toward higher cure rates (p = 0.066). No differences were found regarding gender, weight, geographic area of infection, comorbidities, number of lesions, size of the largest lesion, or anatomical location of lesions. The majority of patients resided in rural areas (95.5%), with the Andina, Orinoco, and Amazonia regions accounting for the highest number of patients with systemic MA for CL.

Table 1. Baseline characteristics stratified according to the presence of TF to systemic MA.

	Cure (N = 165)	Therapeutic failure (N = 131)	Total (N = 296)	P value
Age (years)
mean (SD)	27.8 (12.3)	24.2 (11.9)	26.2 (12.2)	0.012^TT
Min – max	3 – 66	0.5 – 64	0.5 - 66
Age less 18 years	18 (10.9%)	20 (15.3%)	38 (12.8%)
Gender
Male n (%)	141 (85.5%)	116 (88.5%)	257 (86.8%)	0.542^Chi
Female n (%)	24 (14.5%)	15 (11.5%)	39 (13.2%)
Weight (Kg)
(SD)	67.6 (14.4)	65.1 (20.3)	66.6 (17.2)	0.254^TT
Missing values	8 (4.8%)	10 (7.6%)	18 (6.1%)
Economic activity ^α
military personnel n (%)	76 (46.1%)	71 (54.2%)	147 (49.7%)	0.066^Chi
Agriculture n (%)	24 (14.5%)	19 (14.5%)	43 (14.5%)
Independent worker n (%)	13 (7.9%)	9 (6.9%)	22 (7.4%)
Higher education n (%)	18 (10.9%)	3 (2.3%)	21 (7.1%)
Others n (%)	34 (20.6%)	29 (22.1%)	63 (21.3%)
Place of residence
Rural n (%)	156 (94.5%)	123 (96.9%)	279 (95.5%)	0.403^Fi
Urban n (%)	9 (5.5%)	4 (3.1%)	13 (4.5%)
Geographic regions
Amazon n (%)	43 (26.2%)	44 (33.8%)	87 (29.6%)	0.430^Fi
Andean n (%)	66 (40.2%)	51 (39.2%)	117 (39.8%)
Caribbean n (%)	5 (3.0%)	2 (1.5%)	7 (2.4%)
Orinoquia n (%)	29 (17.7%)	23 (17.7%)	52 (17.7%)
Pacific n (%)	21 (12.8%)	10 (7.7%)	31 (10.5%)
Missing values	1 (0.6%)	1 (0.8%)	2 (0.7%)
LC diagnosis
Biopsy n (%)	10 (6.1%)	1 (0.8%)	11 (3.8%)	0.0443^Fi
Culture n (%)	1 (0.6%)	2 (1.6%)	3 (1.0%)
Direct n (%)	152 (92.7%)	121 (96.8%)	273 (94.5%)
IDR n (%)^b	1 (0.6%)	1 (0.8%)	2 (0.7%)
Missing values	1 (0.6%)	6 (4.6%)	7 (2.4%)
Leishmania spp. ^c
L.braziliensis n (%)	0 (0%)	1 (8.3%)	1 (3.6%)
L.panamensis n (%)	16 (100%)	11 (91.7%)	27 (96.4%)
Missing values	149 (90.3%)	119 (90.8%)	268 (90.5%)
Comorbidities ^d
Yes n (%)	15 (9.1%)	8 (6.1%)	23 (7.8%)	0.3881^Chi
No n (%)	150 (90.9%)	123 (93.9%)	273 (92.2%)
Duration of skin lesion in days
Median (RIQ)	60 (42- 91)	56 (30- 84)	56 (31.5- 90)	0.029^MW
Missing values	8 (4.8%)	6 (4.6%)	14 (4.7%)
Size of lesion (mm²) ^e
Median (RIQ)	177.9 (47.12 - 588.70)	148.83 (39.25- 462.55)	171.17 (43.79- 500.50)	0.493^MW
Missing values	11 (6.7%)	25 (19.1%)	36 (12.2%)
Number of lesions
1 n (%)	90 (55.2%)	81 (62.8%)	171 (58.6%)	0.506^Chi
2 n (%)	26 (16%)	20 (15.5%)	46 (15.8%)
3 n (%)	15 (9.2%)	11 (8.5%)	26 (8.9%)
4 n (%)	9 (5.5%)	7 (5.4%)	16 (5.5%)
5 or more n (%)	23 (14.1%)	10 (7.8%)	33 (11.3%)
Missing values	2 (1.2%)	2 (1.5%)	4 (1.4%)
Anatomical location of lesions
Head and neck n (%)	24 (14.9%)	25 (19.7%)	49 (17%)	0.428^Chi
Upper limbs n (%)	62 (38.5%)	43 (33.9%)	105 (36.5%)
Lower limbs n (%)	22 (13.7%)	25 (19.7%)	47 (16.3%)
Trunk n (%)	14 (8.7%)	9 (7.1%)	23 (8%)
More than one lesion n (%)	39 (24.2%)	25 (19.7%)	64 (22.2%)
Missing values	4 (2.4%)	4 (3.1%)	8 (2.7%)
Clinical lesion type ^f
Ulcerated n (%)	133 (83.1%)	110 (84.6%)	243 (83.8%)	0.716^Chi
Non-ulcerated n (%)	9 (5.6%)	9 (6.9%)	18 (6.2%)
Mixed n (%)	18 (11.3%)	11 (8.5%)	29 (10%)
Missing values	5 (3.0%)	1 (0.8%)	6 (2.0%)
Presence of local adenopathy
Yes n (%)	46 (33.1%)	31 (27.7%)	77 (30.7%)	0.431^Chi
No n (%)	93 (66.9%)	81 (72.3%)	174 (69.3%)
Missing values	26 (15.8%)	19 (14.5%)	45 (15.2%)
Doses of AM (20 mg/kg/day) ^g
Insufficient n (%)	2 (1.2%)	5 (3.9%)	7 (2.4%)	0.231^Chi
Sufficient n (%)	149 (91.4%)	111 (86%)	260 (89%)
High doses n (%)	12 (7.4%)	13 (10.1%)	25 (8.6%)
Missing values	2 (1.2%)	2 (1.5%)	4 (1.4%)
Adherence ^h
< 90%	0 (%)	5 (3.8%)	5 (1.7%)	0.016^Fi
≥ 90%	165 (100%)	126 (96.2%)	291 (98.3%)
Prior leishmaniasis infection
Yes n (%)	31 (18.9%)	29 (22.3%)	60 (20.4%)	0.560^Chi
No n (%)	133 (81.1%)	101 (77.7%)	234 (79.6%)
Missing values	1 (0.6%)	1 (0.8%)	2 (0.7%)

Open in a new tab

^aThe military personnel group includes members of the army, police, and security personnel. The independent workers group comprises drivers, miners, self-employed individuals, and construction workers. Higher education includes individuals who have completed university studies.

^bIDR: intradermal reaction of Montenegro. The diagnosis was made based on a positive IDR, lesions suggestive of CL, and a proven epidemiological link.

^cSpecies detected by PCR for Leishmania spp

^dThe following were considered risk comorbidities in patients with CL: immunosuppression and cardiovascular, respiratory, metabolic, hepatic, renal, thyroid, neoplastic diseases.

^eThe area of ulcerated lesions corresponds to the rhombus area (major diagonal x minor diagonal/2) and the area of non-ulcerated lesions corresponds to the rectangles area (base x height).

f A mixed lesion is the presence of both ulcerated and non-ulcerated lesions in the same clinical picture, in any anatomical location on the patient.

^gThe appropriate dose of MA should be based on the patient’s weight: 20 mg/kg/day, maximum 15 cc per day.

^hThe length of systemic MA for CL treatment is 20 days. Adherence to treatment means receiving complete treatment dose of 20 mg/kg/day, maximum 15 ml per day for 18 days, which means 90% of adherence. This is evaluated at the end of treatment visit when the patient brought all blister to their physician.

S1 Table shows the univariate analysis of the candidate predictors. Age (p value = 0.015) and duration of skin lesion (p value = 0.019), were significantly associated with TF. The interaction between lesion area and number of lesions was evaluated (S2 Fig). Patients presenting with more than three lesions showed an increased risk of TF as the size of the largest lesion increased, however this interaction was not significant (p-value = 0.108). Therefore, the interaction variable was not included in the multivariable model.

Lasso model

The optimal penalty value determined by the regression method was a lambda of 0.0235 (log lambda = –3.751) with an expected shrinkage (ES) of 0.054 (log ES = –2.918) (S3 Fig). The resulting model identified the following variables as potential predictors: age, duration of skin lesion, occupation, number of lesions, and presence of local lymphadenopathy. S2 Table presents these variables along with their respective coefficients after penalization.

Model development

Five models were developed (S3 Table). Table 2 shows the performance of the five models. Among them, model 3 demonstrated better discrimination and calibration performance (See S4 and S5 Figs). Model 5 was chosen as the most suitable and readily applicable model. This model included ten variables: age, history of leishmaniasis, duration of skin lesion (≤ 60 days vs. > 60 days), occupation (categorized as high-risk occupations—including military personnel and agricultural workers—and low-risk occupations—including individuals with high education, independent workers such as drivers, miners, self-employed individuals, and construction workers), place of residence (urban or rural), lesion type (ulcerated, mixed, or non-ulcerated), anatomical location of lesions, number of lesions (≤ 3 lesion vs. > 3 lesions), and size of the major lesion (≤ 900 mm² vs. > 900 mm²), in this model age (p value = 0.006), high risk occupation (p value = 0.036) were significantly related with the risk of TF. S4 Table shows the VIF of the selected model, no evidence of multicollinearity was observed among the included predictors. S4 Fig shows the Brier scores for all models, each remaining within the optimal range, indicating good overall predictive accuracy. The odds ratios, 95% CI, and p-values for the final model are presented in Fig 2 and S3 Table (Model 5).

Table 2. Performance (calibration and discrimination) of 5 candidate models.

	HL	R²N	AUC/ROC (IC_95%)	Deviance	Brier score
Model 1	0.406	0.059	0.620 (0.549 - 0.691)	334.7	23.3
Model 2	0.150	0.142	0.688 (0.620 - 0.756)	336.3	22.4
Model 3	0.767	0.171	0.705 (0.634 - 0.777)	297.3	22.6
Model 4	0.389	0.152	0.694 (0.622 - 0.765)	299.7	22.6
Model 5	0.667	0.150	0.691 (0.619 - 0.764)	297.3	22.8

Open in a new tab

HL = Hosmer and Lemeshow goodness of fit (GOF) test, R²N = R² Nagelkerke.

Fig 2 — High-risk occupations—including military personnel and agricultural workers.

The endpoint score was calculated using the selected multivariable model. Each variable included in the final model was assigned a weight based on its beta coefficient, which was multiplied by 10 and rounded to the nearest integer to derive the point value. The resulting scores were then categorized into quartiles to estimate the probability of FT occurrence. Within the study cohort, endpoint scores ranged from 10.6 to 44.1, with a mean of 30.4 (SD ± 6.21). As shown in Fig 3, patients with higher scores had an increased risk of TF compared to those with lower scores. Additionally, when the score was included in a logistic regression model, it significantly predicted TF risk, yielding an odds ratio (OR) of 1.105 (95% CI: 1.06–1.15; p < 0.001).

Table 3 shows sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the final model at various cutoff thresholds. At a predicted probability of 50%, the model achieved a sensitivity of 47.3% and a specificity of 72.7%, with a PPV of 57.9%. When the cutoff was lowered to below 25%, sensitivity increased at the expense of specificity, resulting in an NPV of 85.2%.

Table 3. Operating characteristics of the selected model.

Probability	Sens	Spec	VPP	VPN
> 25%	96.9%	13.9%	47.2%	85.2%
> 50%	47.3%	72.7%	57.9%	63.5%
> 75%	0.8%	100%	100%	55.9%

Open in a new tab

Internal validation of the model

Internal validation yielded Harrell’s C-statistic optimism of 0.12, with an apparent C-index of 0.73 and a bias-corrected (validated) C-index of 0.61. Similarly, the optimism for the calibration slope was 0.509, with an apparent calibration slope of 1.0 and a corrected slope of 0.491.

We then calculated a prediction by using a nomogram prediction. The total score of the patients was obtained by summing the scores corresponding to each subvariable to obtain the predicted probability of TF (Fig 4).

Fig 4 — The nomogram was composed of 10 variables. The nomogram is used as follows: the first horizontal line displays the points; the reader locates how many points each predictor has according to the patient’s characteristics, and then adds up the points and locates the total number of points on the last horizontal line. Finally, the reader draws a vertical line from the total number of points and obtains the probability of TF. High-risk occupations: including military personnel and agricultural workers.

Discussion

A clinical prediction model for TF to MA was developed using a retrospective cohort of patients with CL who received treatment at PECET, located in Medellín- Colombia. The final model includes ten easily obtainable sociodemographic and clinical variables, demonstrates moderate overall performance, acceptable discriminatory power (AUC = 0.691) and adequate calibration (Hosmer-Lemeshow test, p = 0.667). This study represents a significant advance toward risk stratification in clinical practice, particularly in resource-limited settings where molecular tests for the identification of Leishmania species are not routinely available.

A key finding of our study is the high incidence of TF, which reached 44.2% (n = 131) in our cohort. This number significantly exceeds the estimated prevalence of 23.4% in Latin America that was used to calculate the sample size and highlights the growing crisis in the efficacy of systemic pentavalent antimonial [2,11,13,20,35]. However, our observation is consistent with recent reports from other countries in the region that also document an increase in parasite resistance and TF rates [8,18]. This trend has been one of the main reasons for recent updates to PAHO’s CL management guidelines. Additionally, it is plausible that the high incidence of TF observed in our study is partially magnified by selection bias, given that PECET is a national referral center that tends to concentrate cases of greater clinical complexity, with longer duration and possible previous TF.

Our multivariate analysis identified predictors of TF that are consistent with both existing evidence and biological plausibility. Younger age (OR: 0.968; 95% CI: 0.946-0.991) and high-risk occupation (OR: 1.927; 95% CI: 1.043-3.560) were significantly associated with an increased risk of TF. This inverse relationship between age and TF has been previously reported [12,13,36,37]. While the precise mechanisms are not yet fully understood, evidence indicates that treatment response is multifactorial, related to differences in child immune response, drug pharmacokinetics, and antigen exposure [13,38]. Immunologically, the risk of TF in children could be due as the percentage of CD4 + T cells is lower during infancy and adolescence before rising again later in life [13,39]. CD4 + T cells are known to be important factor for Th1 proinflammatory response which drives macrophage activation, parasite clearance, and subsequent lesion healing [37,39]. Furthermore, a short exposure to parasite antigens and sand fly saliva in children may induce lower protective immunity [13,38]. Age-specific pharmacokinetic profiles in pediatric patients can explain also the risk for TF. Children exhibit a significantly higher clearance rate of AM compared to adults [40], resulting in a lower maximum plasma concentration and a reduced area under the curve [41]. These pharmacological differences highlight the critical need to improve the current clinical guidelines for pediatric CL by dose adjustment or using alternative therapeutic regimens [12].

High-risk job, which in this cohort predominantly include military personnel and agricultural workers, are associated with more intense and frequent exposure to the to the vector [18]. This occupational vulnerability is further compounded by adverse socioeconomic conditions such as poor sanitation, which led a higher incidence of FT [42,43]. Furthermore, although they did not reach statistical significance, other variables showed clinically relevant trends. Longer duration of skin lesion (OR: 1.572; p = 0.079) and location in the lower limbs (OR: 2.547; p = 0.092) emerged as important risk factors. Longer duration of skin lesion may correlate with greater parasite load, tissue fibrosis, and bacterial superinfection, factors that complicate the response to treatment [12,13,17]. Similarly, lesions in the lower extremities tend to heal more slowly due to anatomical and physiological factors, such as reduced vascularization, which has been observed in other studies as a therapeutic challenge [22].

When comparing our model with previous, the predictive model developed by Valencia et al. in Peru stands out [18]. This model achieved superior discrimination (AUC = 0.93) with an explanatory variability of 54% (R² Nagelkerke = 0.54), sensitivity of 78%, and specificity of 92.5%, however, its main limitation for applicability in general clinical practice was the inclusion of Leishmania species identification by PCR as a key predictive variable. In many primary care and rural centers from Colombia, this technology is not available, which makes our model, based exclusively on clinical and anamnestic data, considerably more pragmatic and immediately implementable. Also, our study broadens the generalization of the findings by including patients with non-ulcerated and mixed lesions, clinical lesion type that were excluded in the Peruvian study but are associated with a higher parasite load and potentially, a higher risk of TF.

The number and size of lesions are two of the main characteristics of the infection that must be considered to make the best therapeutic decision for patients with CL [8,18,22]. Our study suggested the presence of an interaction between both variables, since in patients with more than 3 lesions, a larger lesion area increased the probability of TF. This was not observed when the number of lesions was less than 3; however, this finding did not reach statistical significance, but it is consistent with biological plausibility and supporting the recommendations to perform systemic treatment in patients with CL who have three or more lesions, at least one of which is larger than 3 cm.

The main strength of our study is its pragmatic approach: A score has been developed to be used by healthcare personnel at any level of care to estimate the risk of MA failure through a targeted medical history and physical examination. Secondly, the statistical methodology was robust, using variable selection methods such as Lasso Regression to identify the most relevant predictors and internal validation using Bootstrapping to quantify and correct model optimism, a bias inherent in the development of prognostic models. Finally, this work is aligned with PAHO guidelines [22], which advocate for a more stratified approach to CL management, considering clinical and epidemiological characteristics to guide therapeutic choices.

Our study has several limitations that must be acknowledged. First, its retrospective design makes it susceptible to information and selection biases, although strategies were implemented to minimize these, such as multiple imputations for handling missing data. Second, the model demonstrated only moderate discriminative performance, with a bias-corrected Harrell’s C-index of 0.61 after internal validation, indicating that the model has limited discriminatory power and should be considered a support tool rather than an absolute determinant of clinical decision-making. Third, the width of the 95% CI for several of the variables included in the final model suggests uncertainty in the estimates, probably due to an insufficient sample size relative to the number of predictors evaluated, which could have contributed to a certain degree of overfitting. Finally, the study cohort consists mainly of young men in the armed forces, which limits the extrapolation of the results to the female population, pediatric age groups, and other demographic contexts.

Conclusion

This study describes the development of a predictive model for TF to systemic MA in CL, derived from routinely available clinical variables. In a scenario of TF to MA greater than 40%, the ability to identify a priori patients at higher risk of TF is crucial. Although the model demonstrated moderate performance, it remains at an early stage of development and should be interpreted with caution. External validation and evaluation in independent populations and diverse clinical settings are necessary before considering its implementation in routine practice.

Nevertheless, because the model is based on variables that are readily obtainable during a standard medical consultation and do not require specialized diagnostic tests, it may serve as a preliminary framework for future research. If validated, a risk-stratified therapeutic approach informed by such a model could potentially contribute to improving clinical decision-making, reducing unnecessary drug exposure and associated toxicity, and supporting more efficient allocation of healthcare resources in Colombia and other endemic regions

Supporting information

S1 Fig. Linear relationship to the logit of the outcome (Therapeutic failure).

(DOCX)

pntd.0014606.s001.docx^{(2.5MB, docx)}

S2 Fig. Prediction of the model including interaction between number of lesions and lesion area after log transformation.

(DOCX)

pntd.0014606.s002.docx^{(198.3KB, docx)}

S3 Fig. Lasso regression model (variable selection).

(DOCX)

pntd.0014606.s003.docx^{(118.1KB, docx)}

S4 Fig. Calibration curves of 5 models.

(DOCX)

pntd.0014606.s004.docx^{(158.2KB, docx)}

S5 Fig. ROC curve of five models.

(DOCX)

pntd.0014606.s005.docx^{(83.9KB, docx)}

S1 Table. Univariate analysis (binary logistic regression).

(DOCX)

pntd.0014606.s006.docx^{(17.2KB, docx)}

S2 Table. Regression coefficients after Lasso regression.

(DOCX)

pntd.0014606.s007.docx^{(13.2KB, docx)}

S3 Table. Model predictions developed.

(DOCX)

pntd.0014606.s008.docx^{(18.6KB, docx)}

S4 Table. Checking multicollinearity (VIF) of selected model.

(DOCX)

pntd.0014606.s009.docx^{(14.1KB, docx)}

S1 Data. Description and definitions of variables in the raw dataset.

(DOCX)

pntd.0014606.s010.docx^{(15.1KB, docx)}

Acknowledgments

We gratefully acknowledge the PECET group for their support throughout this project. We also extend our sincere thanks to Professor Daniel Aguirre for his valuable contributions to methodology and study design. His expertise and commitment greatly enriched the development of this study.

Data Availability

All relevant data are within the manuscript and its Supporting Information files.

Funding Statement

The author(s) received no specific funding for this work.

References

1.Muñoz Morales D, Suarez Daza F, Franco Betancur O, Martinez Guevara D, Liscano Y. The Impact of Climatological Factors on the Incidence of Cutaneous Leishmaniasis (CL) in Colombian Municipalities from 2017 to 2019. Pathogens. 2024;13(6):462. doi: 10.3390/pathogens13060462 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.García-Bustos MF, González-Prieto G, Paniz-Mondolfi AE, Parodi C, Beckar J, Monroig S, et al. Risk factors for antimony treatment failure in American Cutaneous Leishmaniasis in Northwestern-Argentina. PLoS Negl Trop Dis. 2021;15(1):e0009003. doi: 10.1371/journal.pntd.0009003 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Alvar J, Vélez ID, Bern C, Herrero M, Desjeux P, Cano J, et al. Leishmaniasis Worldwide and Global Estimates of Its Incidence. PLOS ONE. 2012;7(5):e35671. doi: 10.1371/journal.pone.0035671 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Centers for Disease Control (CDC) PHS. Parasites - Leishmaniasis. 2013. https://www.cdc.gov/parasites/leishmaniasis/gen_info/index.html
5.Carrillo-Bonilla LM, Trujillo JJ, Alvarez-Salas L, Vélez-Bernal ID. Study of knowledge, attitudes, and practices related to leishmaniasis: evidence of government neglect in the Colombian Darién. Cad Saude Publica. 2014;30(10):2134–44. doi: 10.1590/0102-311x00139713 [DOI] [PubMed] [Google Scholar]
6.Castro MDM, Erber AC, Arana B, Cota G, Denkinger CM, Harrison N, et al. Involving patients in drug development for Neglected Tropical Diseases (NTDs): A qualitative study exploring and incorporating preferences of patients with cutaneous leishmaniasis into Target Product Profile development. PLoS Negl Trop Dis. 2024;18(2):e0011975. doi: 10.1371/journal.pntd.0011975 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Garcia Bustos MF, Barrio A, Parodi C, Beckar J, Moreno S, Basombrio MA. Miltefosine versus meglumine antimoniate in the treatment of mucosal leishmaniasis. Medicina (B Aires). 2014;74(5):371–7. [PubMed] [Google Scholar]
8.Manual de procedimientos para la vigilancia y el control de las leishmaniasis en la región de las Américas. Washington, D.C.: Organización Panamericana de la Salud (OPS). 2022. [Google Scholar]
9.Leishmaniasis, semana epidemiológica 25 19 al 25 de junio de 2022. Boletin Epidemológic Semanal Insituto Nacional de Salud Colombia. 2022. [Google Scholar]
10.Vásquez L. Terapéutica antileishmania: revisando el pasado, el presente y el futuro. Gaceta Médica de Caracas. 2009;117:93–111. [Google Scholar]
11.Pasquau F, Ena J, Sanchez R, Cuadrado JM, Amador C, Flores J, et al. Leishmaniasis as an opportunistic infection in HIV-infected patients: determinants of relapse and mortality in a collaborative study of 228 episodes in a Mediterreanean region. Eur J Clin Microbiol Infect Dis. 2005;24(6):411–8. doi: 10.1007/s10096-005-1342-6 [DOI] [PubMed] [Google Scholar]
12.Castro MDM, Cossio A, Velasco C, Osorio L. Risk factors for therapeutic failure to meglumine antimoniate and miltefosine in adults and children with cutaneous leishmaniasis in Colombia: A cohort study. PLoS Negl Trop Dis. 2017;11(4):e0005515. doi: 10.1371/journal.pntd.0005515 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Llanos-Cuentas A, Tulliano G, Araujo-Castillo R, Miranda-Verastegui C, Santamaria-Castrellon G, Ramirez L, et al. Clinical and parasite species risk factors for pentavalent antimonial treatment failure in cutaneous leishmaniasis in Peru. Clin Infect Dis. 2008;46(2):223–31. doi: 10.1086/524042 [DOI] [PubMed] [Google Scholar]
14.Tuon FF, Amato VS, Graf ME, Siqueira AM, Nicodemo AC, Amato Neto V. Treatment of New World cutaneous leishmaniasis--a systematic review with a meta-analysis. Int J Dermatol. 2008;47(2):109–24. doi: 10.1111/j.1365-4632.2008.03417.x [DOI] [PubMed] [Google Scholar]
15.Mohammadzadeh M, Behnaz F, Golshan Z. Efficacy of glucantime for treatment of cutaneous leishmaniasis in Central Iran. J Infect Public Health. 2013;6(2):120–4. doi: 10.1016/j.jiph.2012.11.003 [DOI] [PubMed] [Google Scholar]
16.Vásquez L, Scorza Dagert JV, Scorza JV, Vicuña-Fernández N, de Peña YP, López S, et al. Pharmacokinetics of experimental pentavalent antimony after intramuscular administration in adult volunteers. Curr Ther Res Clin Exp. 2006;67(3):193–203. doi: 10.1016/j.curtheres.2006.06.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Suprien C, Rocha PN, Teixeira M, Carvalho LP, Guimarães LH, Bonvoisin T, et al. Clinical Presentation and Response to Therapy in Children with Cutaneous Leishmaniasis. Am J Trop Med Hyg. 2020;102(4):777–81. doi: 10.4269/ajtmh.19-0531 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Valencia C, Arévalo J, Dujardin JC, Llanos-Cuentas A, Chappuis F, Zimic M. Prediction score for antimony treatment failure in patients with ulcerative leishmaniasis lesions. PLoS Negl Trop Dis. 2012;6(6):e1656. doi: 10.1371/journal.pntd.0001656 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Romero GA, Guerra MV, Paes MG, Macêdo VO. Comparison of cutaneous leishmaniasis due to Leishmania (Viannia) braziliensis and L. (V.) guyanensis in Brazil: therapeutic response to meglumine antimoniate. Am J Trop Med Hyg. 2001;65(5):456–65. doi: 10.4269/ajtmh.2001.65.456 [DOI] [PubMed] [Google Scholar]
20.Rodrigues AM, Hueb M, dos Santos TARR, Fontes CJF. Factors associated with treatment failure of cutaneous leishmaniasis with meglumine antimoniate. Rev Soc Bras Med Trop. 2006;39(2):139–45. doi: 10.1590/s0037-86822006000200001 [DOI] [PubMed] [Google Scholar]
21.Lago T, Carvalho LP, Nascimento M, Guimarães LH, Lago J, Castellucci L, et al. Influence of Obesity on Clinical Manifestations and Response to Therapy in Cutaneous Leishmaniasis Caused by Leishmania braziliensis. Clin Infect Dis. 2021;73(6):1020–6. doi: 10.1093/cid/ciab236 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Pan American Health Organization PAHO. Guideline for the Treatment of Leishmaniasis in the Americas. 2022. https://iris.paho.org/bitstream/handle/10665.2/56120/guidelineleishmaniasisannex_eng.pdf?sequence=14
23.Directrices para el tratamiento de las leishmaniasis en la región de las Américas. Organización Panamericana de la Salud. 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Mendigaña Paez FA. Plan estratégico leishmaniasis 2018 - 2022. 2022. https://www.minsalud.gov.co/sites/rid/Lists/BibliotecaDigital/RIDE/VS/PP/ET/plan-estrategico-leishmaniasis-2018-2022.pdf
25.Riley RD, Ensor J, Snell KIE, Harrell FE Jr, Martin GP, Reitsma JB, et al. Calculating the sample size required for developing a clinical prediction model. BMJ. 2020;368:m441. doi: 10.1136/bmj.m441 [DOI] [PubMed] [Google Scholar]
26.Steyerberg EW. Clinical prediction models, a practical approach to development, validation, and updating. 2019.
27.Ensor J. Sample size for development of a prediction model. 2023.
28.Steyerberg EW, Eijkemans MJ, Harrell FE Jr, Habbema JD. Prognostic modelling with logistic regression analysis: a comparison of selection and estimation methods in small data sets. Stat Med. 2000;19(8):1059–79. doi: 10.1002/(sici)1097-0258(20000430)19:8<1059::aid-sim412>3.0.co;2-0 [DOI] [PubMed] [Google Scholar]
29.Tibshirani R. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological). 1996;58(1):267–88. [Google Scholar]
30.Harrell F. Regression Modeling Strategies with Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. 2 ed. Statistics SS i. 2015.
31.Senaviratna NAMR, A. Cooray TMJ. Diagnosing Multicollinearity of Logistic Regression Model. AJPAS. 2019;:1–9. doi: 10.9734/ajpas/2019/v5i230132 [DOI] [Google Scholar]
32.Jiang J. Applied Medical Statistics. Blackwell W. 2022.
33.van Buuren S. Flexible imputation of missing data. 2nd ed. Chapman and Hall/CRC. 2018. [Google Scholar]
34.Marshall A, Altman DG, Holder RL, Royston P. Combining estimates of interest in prognostic modelling studies after multiple imputation: current practice and guidelines. BMC Med Res Methodol. 2009;9:57. doi: 10.1186/1471-2288-9-57 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Santos AGA, Sousa JM, de Aguiar AHBM, Torres KCS, Coelho AJS, Ferreira AL, et al. Systematic Review of Treatment Failure and Clinical Relapses in Leishmaniasis from a Multifactorial Perspective: Clinical Aspects, Factors Associated with the Parasite and Host. Trop Med Infect Dis. 2023;8(9):430. doi: 10.3390/tropicalmed8090430 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Bamorovat M, Sharifi I, Dabiri S, Shamsi Meymandi S, Karamoozian A, Amiri R, et al. Major risk factors and histopathological profile of treatment failure, relapse and chronic patients with anthroponotic cutaneous leishmaniasis: A prospective case-control study on treatment outcome and their medical importance. PLoS Negl Trop Dis. 2021;15(1):e0009089. doi: 10.1371/journal.pntd.0009089 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Silva H, Liyanage A, Deerasinghe T, Chandrasekara V, Chellappan K, Karunaweera ND. Treatment failure to sodium stibogluconate in cutaneous leishmaniasis: A challenge to infection control and disease elimination. PLoS One. 2021;16(10):e0259009. doi: 10.1371/journal.pone.0259009 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Davies CR, Llanos-Cuentas EA, Pyke SD, Dye C. Cutaneous leishmaniasis in the Peruvian Andes: an epidemiological study of infection and immunity. Epidemiol Infect. 1995;114(2):297–318. doi: 10.1017/s0950268800057964 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Kawiak J, Rokicka-Milewska R, Zeman K, Hoser G, Derulska G, Fornalczyk-Wachowska E, et al. Peripheral blood leukocytes and lymphocyte subpopulations as determined by flow cytometric measurements in healthy children. Folia Histochem Cytobiol. 1995;33(1):33–8. [PubMed] [Google Scholar]
40.Older Adults “Chronic Disease Indicators”. US Center for Disease Control and Prevention. https://www.cdc.gov/cdi/indicator-definitions/older-adults.html?
41.Castro MDM, Gomez MA, Kip AE, Cossio A, Ortiz E, Navas A, et al. Pharmacokinetics of Miltefosine in Children and Adults with Cutaneous Leishmaniasis. Antimicrob Agents Chemother. 2017;61(3):e02198-16. doi: 10.1128/AAC.02198-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Cruz A, Rainey PM, Herwaldt BL, Stagni G, Palacios R, Trujillo R, et al. Pharmacokinetics of antimony in children treated for leishmaniasis with meglumine antimoniate. J Infect Dis. 2007;195(4):602–8. doi: 10.1086/510860 [DOI] [PubMed] [Google Scholar]
43.Paniz-Mondolfi AE, Talhari C, García Bustos MF, Rosales T, Villamil-Gomez WE, Marquez M, et al. American cutaneous leishmaniasis in infancy and childhood. Int J Dermatol. 2017;56(12):1328–41. doi: 10.1111/ijd.13664 [DOI] [PubMed] [Google Scholar]

PLoS Negl Trop Dis. doi: 10.1371/journal.pntd.0014606.r001

Decision Letter 0

Guilherme L Werneck, Elham Kazemirad

1 Feb 2026

-->PNTD-D-25-02094

Development and Internal Validation of a Predictive Model for Meglumine Antimoniate Treatment Failure in Patients with Cutaneous Leishmaniasis: A Retrospective Cohort Study

PLOS Neglected Tropical Diseases

Dear Dr. Serna-Higuita,

Thank you for submitting your manuscript to PLOS Neglected Tropical Diseases. After careful consideration, we feel that it has merit but does not fully meet PLOS Neglected Tropical Diseases's publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript within by Apr 02 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosntds@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pntd/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

* A letter that responds to each point raised by the editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. This file does not need to include responses to any formatting updates and technical items listed in the 'Journal Requirements' section below.

* A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

* An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, competing interests statement, or data availability statement, please make these updates within the submission form at the time of resubmission. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

We look forward to receiving your revised manuscript.

Kind regards,

Elham Kazemirad, Ph.D

Academic Editor

PLOS Neglected Tropical Diseases

Guilherme Werneck

Section Editor

PLOS Neglected Tropical Diseases

Shaden Kamhawi

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

orcid.org/0000-0003-4304-636XX

Paul Brindley

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

orcid.org/0000-0003-1765-0002

Journal Requirements:

1) Please provide an Author Summary. This should appear in your manuscript between the Abstract (if applicable) and the Introduction, and should be 150-200 words long. The aim should be to make your findings accessible to a wide audience that includes both scientists and non-scientists. Sample summaries can be found on our website under Submission Guidelines:

https://journals.plos.org/plosntds/s/submission-guidelines#loc-parts-of-a-submission

2) Please upload all main figures as separate Figure files in .tif or .eps format. For more information about how to convert and format your figure files please see our guidelines:

https://journals.plos.org/plosntds/s/figures

3) In the online submission form, you indicated that The raw data underlying the results presented in the study are available from the PECET (Programa de Estudio y Control de Enfermedades Tropicales) for researchers who meet the criteria for access to confidential data.. All PLOS journals now require all data underlying the findings described in their manuscript to be freely available to other researchers, either

1. In a public repository

2. Within the manuscript itself

3. Uploaded as supplementary information.

This policy applies to all data except where public deposition would breach compliance with the protocol approved by your research ethics board. If your data cannot be made publicly available for ethical or legal reasons (e.g., public availability would compromise patient privacy), please explain your reasons by return email and your exemption request will be escalated to the editor for approval. Your exemption request will be handled independently and will not hold up the peer review process, but will need to be resolved should your manuscript be accepted for publication. One of the Editorial team will then be in touch if there are any issues.

Reviewers' Comments:

Reviewer's Responses to Questions

Key Review Criteria Required for Acceptance?

As you describe the new analyses required for acceptance, please consider the following:

Methods

-Are the objectives of the study clearly articulated with a clear testable hypothesis stated?

-Is the study design appropriate to address the stated objectives?

-Is the population clearly described and appropriate for the hypothesis being tested?

-Is the sample size sufficient to ensure adequate power to address the hypothesis being tested?

-Were correct statistical analysis used to support conclusions?

-Are there concerns about ethical or regulatory requirements being met?

Reviewer #1: - L4, aims should be stated in the past tense.

- This study considered only treatment failure cases. What happens to the relapse cases that occur after six months of follow-up assessment?

- Scientific names should be written in italic font (L27, Leishmania, L28 Lutzomyia, etc.)

- Prevalently used terms should be abbreviated (L3, meglumine antimoniate; MA) and treatment failure (TF L4,5,19 and it should be defined at first appearance).

- CL was already abbreviated (L31); it should not be used in complete form /or defined (L58, 59, 276, etc.), or TF is defined in many places (L275).

- Treatment failure and case definition should be presented in M&M.

- L83, Nagekerke is printed in small first letter?

- Since a few cases were identified to species level by PCR, how could we be sure that the remaining (n=268) are L. braziliens?

Reviewer #2: Overall, the methodological section is clearly presented. The study objectives are explicitly stated, and both the design and the analytic strategy appear appropriate for addressing the research aims. The population is generally well described, and the study setting seems adequate for testing the hypotheses posed. Nevertheless, some methodological aspects would benefit from clarification or further detail:

1. Some key variables require clearer operational definitions. For example, the term “high-risk occupations” is later presented as an important finding in the results, yet the data collection section lists only “age, gender, economic activity, educational level, place of residence, and comorbidities”, without specifying how occupational risk categories were defined.

2. The manuscript states that “informed consent was obtained”, but does not explain how this was feasible given the retrospective nature of the study and the extended time frame (dating back to 2007). Clarification is needed on the consent procedures, if applicable, or whether waivers were granted. Additionally, the ethics approval number is not described.

3. The description of the outcome variable, particularly the timing of the three criteria used to determine treatment failure, should be improved. According to the harmonized criteria for cutaneous leishmaniasis treatment outcomes, the timing of assessments is critical (Olliaro, et al). Considering that retrospective studies often face challenges related to missing or inconsistently documented follow-up, it would be good for authors to describe how they handled variability in follow-up time points, how outcome classification was standardized, and whether any assumptions or imputation strategies were applied. A general description of imputations for missing values was presented, but it is not clear whether that also applies to the outcome measures.

Reviewer #3: I am not sure that this could be called a retrospective cohort study. I think it is more fair to call it a retrospective cohort analysis.

**********

Results

-Does the analysis presented match the analysis plan?

-Are the results clearly and completely presented?

-Are the figures (Tables, Images) of sufficient quality for clarity?

Reviewer #1: Results are alright. - Treatment failure cases were patients who demonstrated active lesion(s) or relapse at a defined period (6 months) of follow-up examination.

- Relapse is the follow-up assessment. A lesion following cure. Here, the authors mentioned nothing about the relapse rate post-treatment follow-up visits. The reason one visits the CL patients following treatment is to see whether the cases have relapsed or not.

- It is not clear the number of variables that were significant in the result section.

Reviewer #2: The results are generally aligned with the analytical approach described in the methods, and the figures and tables are of good quality. Below I list some aspects that require clarification to improve clarity and interpretation of the findings.

1. Age is presented only as a mean with standard deviation. Given the well-established association between younger age and treatment outcomes in cutaneous leishmaniasis, it would be important to report the age range and explicitly indicate the proportion of pediatric patients in the cohort. This information is essential for interpreting the external validity of the findings.

2. Table 1:

o Age range: Please report the minimum and maximum ages represented.

o Economic activity: The category “professional” appears more closely related to educational attainment than to occupational classification.

o IDR: Spell out the acronym (presumably intradermal reaction or Montenegro skin test) at first mention.

o Lesion area: The method used to calculate lesion area should be described, including the geometric formula applied.

o Treatment adherence: The description requires further detail. For example, reporting “18/20 doses” is insufficient without specifying the timeframe (e.g., over how many days) and how adherence was assessed relative to prescribed dosing. Although retrospective designs impose limitations on quality f adherence data, clarity regarding dose administration patterns is important given the short elimination half-life of antimonials and their potential sensitivity to gaps between doses.

o Lesion duration: Kindly check whether the time unit is months or days. Although chronic lesions can persist for extended periods, a median duration of five years (60 months) would contradict existing literature.

3. Line 174: The “endpoint score” is introduced without any explanation in the Methods. The authors should clarify what this score represents and whether it was pre-specified or created post hoc.

4. Although the authors describe the modeling steps thoroughly, the clinical meaning of the results is insufficiently addressed. Several included predictors are not defined, making interpretation difficult. Examples include:

o High-risk occupation: The term is not defined.

o Non-local adenopathy: The rationale for using this category instead of standard regional adenopathy is unclear.

Of the ten predictors in the final model, only age and high-risk occupation were significant, but their clinical implications are not discussed in depth. The supplementary figures help clarify some associations, but this is not reflected in the main text.

Minor comments:

1. Kindly check the most appropriate term in the first paragraph (patients or records), since it is a study with retrospective evaluation of electronic health data. Are all the records from one PECET or multiple facilities? If multiple facilities, kindly indicate whether clinical management practices were comparable across sites.

2. The model development details may be more appropriate in the Methods. As currently written, they break the flow of the Results.

Reviewer #3: The internal validation results sections is very short and simple. Since this is announced in the title as one of the strengths of this study, this section should be larger and more detailed. Another option is to remove this information from the title.

**********

Conclusions

-Are the conclusions supported by the data presented?

-Are the limitations of analysis clearly described?

-Do the authors discuss how these data can be helpful to advance our understanding of the topic under study?

-Is public health relevance addressed?

Reviewer #1: Suitable.

Reviewer #2: The discussion summarizes the main findings, and the text is overall well written. My main concern is, however, that the clinical use and interpretation of the finding is limited. It is important that authors explicitly address how these findings could be used in clinical settings and provide stronger justification for the selected variables. Some relevant points are elaborated below:

1. The authors attribute the association between younger age and treatment failure primarily to host immunity. However, other well-known explanations such as age-related differences in antimonial pharmacokinetics, should be acknowledged. In general, providing a more balanced interpretation would strengthen this section.

2. In addition, greater clarity on the age distribution would help readers understand the observed patterns. The supplementary figure suggests a U-shaped risk curve, with wider confidence intervals at age extremes, likely due to small sample sizes. The discussion should state how many very young and very old participants were included, as this can affect the interpretation of the finding.

3. The lesion duration is inconsistent across tables. It is unclear whether lesion duration was recorded in days or months. Some tables and figures appear inconsistent. This needs verification, since this affects interpretation and the plausibility of the results.

4. The final statement claims that the model provides a “valuable tool” to optimize management, yet the manuscript does not explain how it could be used in practice. Beyond a broad statement, the discussion does not elaborate on how clinicians could apply the model. Only two of the ten predictors were statistically significant, and the clinical utility of the remaining variables is not addressed.

Given the model’s moderate performance and lack of external validation, the conclusions should be more cautious and specify the steps needed before clinical implementation.

Reviewer #3: The authors should moderate their optimism about the potential use of the predictive model in clinical practice. As they recognize, the model has limitations because of its moderate overall performance and moderate sensitivity and specificity. On the other hand, 10 variables could be considered a very large number of variables to be assessed by a clinician with each patient. Also, before considering the potential use of any model, it needs to be operationalized and adapted to be the clinical practice settings. The authors should expand a little bit more this aspect in the discussion and conclusion.

**********

Editorial and Data Presentation Modifications?

Use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity. If the only modifications needed are minor and/or editorial, you may wish to recommend “Minor Revision” or “Accept”.

Reviewer #1: The manuscript is well prepared, clinically relevant, fairly reliable methods and appropriate modeling approaches (multivariate logistic regression). Some concerns about the referral center cases that are mixed (it is a solitary center retrospective cohort study). I believe the manuscript can be suitable for publication if the authors clearly complete and respond to the above questions. There are two types of unresponsiveness to treatment: 1- treatment failure, which is the cases that do not respond to the assigned drug basically; 2- Resistant cases are those that initially respond to the action of the drug and get cured but later on recur, and that is why, 6 months following treatment, they should be re-examined and managed with other drugs. The authors ignored such relapsed cases.

Reviewer #2: Comments are described across the review text for each section.

Reviewer #3: There are editorial aspects that need to be reviewed. There are some missing words and inconsistency in the use of some abbreviations.

**********

Summary and General Comments

Use this section to provide overall comments, discuss strengths/weaknesses of the study, novelty, significance, general execution and scholarship. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. If requesting major revision, please articulate the new experiments that are needed.

Reviewer #1: (No Response)

Reviewer #2: The manuscript addresses a relevant topic and presents a dataset with potential to inform clinical practice. However, the interpretation of the results appear insufficiently developed in their current form. The Results and Discussion sections rely heavily on statistical descriptions, with limited attention to the clinical meaning of the findings, variable definitions, and practical implications of the model. These gaps make it difficult to fully assess the model’s potential and the contribution it brings to the existing literature. With additional clarification and more emphasis on interpretation, the work could become a more solid contribution, but substantial revision is needed before its strengths can be fully appreciated

Reviewer #3: This is a very interesting article, based on a study conducted by a research group with extensive experience in cutaneous leishmaniasis research, resulting in one of the best clinical datasets on this disease in the region. The development and validation of models that can predict glucantime treatment failure, and that can be used by primary care personnel in endemic areas, is crucial, given that this is a neglected tropical disease and that laboratory resources are very limited for evaluating patients and predicting treatment failure. This study contributes to that goal.

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Figure resubmission:

While revising your submission, we strongly recommend that you use PLOS’s NAAS tool (https://ngplosjournals.pagemajik.ai/artanalysis) to test your figure files. NAAS can convert your figure files to the TIFF file type and meet basic requirements (such as print size, resolution), or provide you with a report on issues that do not meet our requirements and that NAAS cannot fix.

-->

Reproducibility:

To enhance the reproducibility of your results, we recommend that authors of applicable studies deposit laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

-->

PLoS Negl Trop Dis. 2026 Aug 4;20(8):e0014606. doi: 10.1371/journal.pntd.0014606.r002

Author response to Decision Letter 1

16 Mar 2026

Attachment

Submitted filename: Answer Letter S-R_LMSH 24-FEB-26.docx

pntd.0014606.s011.docx^{(38.1KB, docx)}

PLoS Negl Trop Dis. doi: 10.1371/journal.pntd.0014606.r003

Decision Letter 1

Guilherme L Werneck, Elham Kazemirad

8 Jun 2026

-->PNTD-D-25-02094R1-->-->Development of a Predictive Model for Meglumine Antimoniate Treatment Failure in Patients with Cutaneous Leishmaniasis: A Retrospective Cohort Study-->-->PLOS Neglected Tropical Diseases-->--> -->-->Dear Dr. Serna-Higuita,-->--> -->-->Thank you for submitting your manuscript to PLOS Neglected Tropical Diseases. After careful consideration, we feel that it has merit but does not fully meet PLOS Neglected Tropical Diseases's publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.-->--> -->-->Please submit your revised manuscript by Jul 08 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosntds@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pntd/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.-->--> -->-->Please include the following items when submitting your revised manuscript:-->-->* A letter that responds to each point raised by the editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. This file does not need to include responses to any formatting updates and technical items listed in the 'Journal Requirements' section below.-->-->* A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.-->-->* An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.-->--> -->-->If you would like to make changes to your financial disclosure, competing interests statement, or data availability statement, please make these updates within the submission form at the time of resubmission. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.-->--> -->-->As the corresponding author, your ORCID iD is verified in the submission system and will appear in the published article. PLOS supports the use of ORCID, and we encourage all coauthors to register for an ORCID iD and use it as well. Please encourage your coauthors to verify their ORCID iD within the submission system before final acceptance, as unverified ORCID iDs will not appear in the published article. Only the individual author can complete the verification step; PLOS staff cannot verify ORCID iDs on behalf of authors.-->-->-->-->We look forward to receiving your revised manuscript.-->--> -->-->Kind regards,-->--> -->-->Elham Kazemirad, Ph.D-->-->Academic Editor-->-->PLOS Neglected Tropical Diseases-->--> -->-->Guilherme Werneck-->-->Section Editor-->-->PLOS Neglected Tropical Diseases-->-->

Shaden Kamhawi

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

orcid.org/0000-0003-4304-636XX

Paul Brindley

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

orcid.org/0000-0003-1765-0002

-->--> -->-->Additional Editor Comments (if provided): -->--> -->--> -->--> -->--> -->-->Journal Requirements:

If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.-->--> -->-->1) Please upload all main figures as separate Figure files in .tif or .eps format. For more information about how to convert and format your figure files please see our guidelines:

https://journals.plos.org/plosntds/s/figures.

-->--> -->-->Reviewers' comments: -->--> -->-->Reviewer's Responses to Questions

Key Review Criteria Required for Acceptance?

As you describe the new analyses required for acceptance, please consider the following:

Methods

-Are the objectives of the study clearly articulated with a clear testable hypothesis stated?

-Is the study design appropriate to address the stated objectives?

-Is the population clearly described and appropriate for the hypothesis being tested?

-Is the sample size sufficient to ensure adequate power to address the hypothesis being tested?

-Were correct statistical analysis used to support conclusions?

-Are there concerns about ethical or regulatory requirements being met?

Reviewer #2: The methods are overall clear and aligned with the study aims. Some recommendations to improve this section include:

Lines 107-108: please consider adding a citation to support the use of Rubin’s Rules for pooling estimates after multiple imputation. There also appears to be a minor typo in this sentence: “the estimated were pooled”, which should read “the estimates were pooled.”

The statement “All participants provided written informed consent” may be misleading in the context of a retrospective study, as it could be interpreted as consent specifically for participation in this study. Based on the authors’ response and the manuscript context, it appears that patients provided consent for medical care and possibly for secondary use of their clinical data. Please revise the wording to clarify the nature and scope of the consent obtained, and distinguish this from study-specific informed consent, if applicable.

Reviewer #3: Yes. The study objetive is clearly articulated with the study results. Authors improved this section after the first revision and took into account the reviewer comments.

**********

Results

-Does the analysis presented match the analysis plan?

-Are the results clearly and completely presented?

-Are the figures (Tables, Images) of sufficient quality for clarity?

Reviewer #2: The Results section is clear, and the presentation has improved. In particular, removing some of the model development text from this section has improved readability. However, some additional comments remain, mostly related to proofreading and clarity of writing rather than to the quality of the study. More broadly, the text may benefit from proofreading by a native English speaker or a professional language editing service.

Some of the comments are listed below:

Lines 118 – 120: “When comparing patients with and without TF, differences were found in terms of occupation, with professionals having higher cure rates (p=0.06) and lower adherence to therapy (less than 90%).”

This sentence is difficult to interpret and would benefit from revision. In its current form, it is unclear whether the higher cure rates and lower adherence to therapy refer to professionals or to patients with TF. Please clarify the comparison being made and the direction of the reported differences.

204: Leishmania should be italicized.

Figure 2: Please note the term “evolution time” is not commonly used in English in this context and may reflect a direct translation from Spanish. Consider revising it to a more standard term, such as “disease duration,” “duration of symptoms,” or “time since symptom onset,” depending on the intended meaning.

A similar observation applies to the phrase “localization of single lesion”. Authors may consider revising it to “location of single lesion,” “site of single lesion,” or “anatomical location of single lesion.”

Reviewer #3: Yes. The results are clearly and completely presented. Tables and figures are adequate.

**********

Conclusions

-Are the conclusions supported by the data presented?

-Are the limitations of analysis clearly described?

-Do the authors discuss how these data can be helpful to advance our understanding of the topic under study?

-Is public health relevance addressed?

Reviewer #2: The Discussion and Conclusion sections are generally clear and aligned with the data presented in the results; however, some statements would benefit from stronger grounding in the existing literature and further clarification. Comments and suggestions to improve these sections are listed below.

Authors state that “PECET is a national referral center that tends to concentrate cases of greater clinical complexity, with longer duration and possible previous TF.” Please clarify whether this was observed in the study population. Specifically, how many participants had documented previous therapeutic failure, and how might this have influenced the results and the interpretation of the model?

Lines 217–219: “The association with age has been previously reported, although the underlying mechanisms, possibly related to the host immune response, are not yet fully understood.”

Please note that this issue, raised in the previous review, does not appear to have been adequately addressed. The association between age and therapeutic response to antileishmanial treatment has been reported in multiple studies, and several hypotheses have been proposed to explain this relationship. However, the revised text still provides a broad statement about host immune response without citing the relevant literature or engaging with other proposed explanations. Please revise this section to include appropriate citations and a more complete discussion of the possible mechanisms underlying the association with age.

In the conclusion, the claim that this is “the first model on the continent” appears overstated and potentially inaccurate. A prior model from Peru is already mentioned in the Discussion, which would seem to preclude describing the present study in this way. Please revise the statement to avoid overclaiming novelty and to clearly distinguish the contribution of this study from previously published work.

The clinical applicability of the proposed score remains unclear. The model includes many variables, 10 in total, yet the manuscript does not explain how clinicians are expected to calculate, interpret, or act on the score. In addition, given that the therapeutic failure rate in this cohort is higher than that reported in other settings in the region, please discuss how this may affect the generalizability of the score, particularly in populations with different baseline risks.

Minor comment: Line 220: The term “militaries,” when referring to individuals in high-risk occupations, should be revised. Please consider using alternatives such as “military personnel,” “service members,” or “members of the armed forces,” as appropriate.

Reviewer #3: Yes.

**********

Editorial and Data Presentation Modifications?

Reviewer #2: (No Response)

Reviewer #3: No further comments.

**********

Summary and General Comments

Reviewer #2: Overall, the manuscript has improved, particularly in the presentation of the Results section and the readability of the model development description. However, I recommend further revision. The Discussion and Conclusion sections still require additional work, particularly to better ground key statements in the existing literature, clarify the interpretation and applicability of the proposed model, and address concerns regarding generalizability and clinical use of the score. I would also recommend proofreading by a native English speaker or copyediting services.

Reviewer #3: Although not many people care about treatment failure in cutaneous leishmaniasis, this is a big problem for patients. Being a neglected disease, it is important for them and the doctors and nurses who are responsible of treating this patients and be aware of clinical and epidemiological patient characteristics that could predict treatment failure. Therefore, this paper is valuable. Further work is needed to validate and improve this model, but this is a good start.

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Reviewer #3: No

-->--> -->-->[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]-->--> -->-->Figure resubmission: -->--> -->-->-->While revising your submission, we strongly recommend that you use PLOS’s NAAS tool (https://ngplosjournals.pagemajik.ai/artanalysis) to test your figure files. NAAS can convert your figure files to the TIFF file type and meet basic requirements (such as print size, resolution), or provide you with a report on issues that do not meet our requirements and that NAAS cannot fix.-->-->

After uploading your figures to PLOS’s NAAS tool - https://ngplosjournals.pagemajik.ai/artanalysis, NAAS will process the files provided and display the results in the "Uploaded Files" section of the page as the processing is complete. If the uploaded figures meet our requirements (or NAAS is able to fix the files to meet our requirements), the figure will be marked as "fixed" above. If NAAS is unable to fix the files, a red "failed" label will appear above. When NAAS has confirmed that the figure files meet our requirements, please download the file via the download option, and include these NAAS processed figure files when submitting your revised manuscript.-->-->--> -->-->Reproducibility: -->--> -->-->To enhance the reproducibility of your results, we recommend that authors of applicable studies deposit laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols-->

PLoS Negl Trop Dis. 2026 Aug 4;20(8):e0014606. doi: 10.1371/journal.pntd.0014606.r004

Author response to Decision Letter 2

17 Jun 2026

Attachment

Submitted filename: Answer Letter 16-Jun-26.docx

pntd.0014606.s012.docx^{(25.3KB, docx)}

PLoS Negl Trop Dis. doi: 10.1371/journal.pntd.0014606.r005

Decision Letter 2

Guilherme L Werneck

22 Jul 2026

Dear Dr Serna-Higuita,

We are pleased to inform you that your manuscript 'Development of a Predictive Model for Meglumine Antimoniate Treatment Failure in Patients with Cutaneous Leishmaniasis: A Retrospective Cohort Study' has been provisionally accepted for publication in PLOS Neglected Tropical Diseases.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Neglected Tropical Diseases.

Best regards,

Guilherme Werneck

Section Editor

PLOS Neglected Tropical Diseases

Shaden Kamhawi

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

orcid.org/0000-0003-4304-636XX

Paul Brindley

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

orcid.org/0000-0003-1765-0002

***********************************************************

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; line-height: 16.0px; font: 14.0px Arial; color: #323333; -webkit-text-stroke: #323333}span.s1 {font-kerning: none

Reviewer's Responses to Questions

Key Review Criteria Required for Acceptance?

As you describe the new analyses required for acceptance, please consider the following:

Methods

-Are the objectives of the study clearly articulated with a clear testable hypothesis stated?

-Is the study design appropriate to address the stated objectives?

-Is the population clearly described and appropriate for the hypothesis being tested?

-Is the sample size sufficient to ensure adequate power to address the hypothesis being tested?

-Were correct statistical analysis used to support conclusions?

-Are there concerns about ethical or regulatory requirements being met?

Reviewer #2: (No Response)

**********

Results

-Does the analysis presented match the analysis plan?

-Are the results clearly and completely presented?

-Are the figures (Tables, Images) of sufficient quality for clarity?

Reviewer #2: (No Response)

**********

Conclusions

-Are the conclusions supported by the data presented?

-Are the limitations of analysis clearly described?

-Do the authors discuss how these data can be helpful to advance our understanding of the topic under study?

-Is public health relevance addressed?

Reviewer #2: (No Response)

**********

Editorial and Data Presentation Modifications?

Reviewer #2: (No Response)

**********

Summary and General Comments

Reviewer #2: The manuscript is clear, and its overall quality has improved. The authors have addressed the reviewers’ comments, particularly through revisions to the Discussion and Conclusions sections. I recommend a final proofread to correct the remaining minor typographical errors throughout the text.

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

PLoS Negl Trop Dis. doi: 10.1371/journal.pntd.0014606.r006

Acceptance letter

Guilherme L Werneck

Dear Dr Serna-Higuita,

We are delighted to inform you that your manuscript, "Development of a Predictive Model for Meglumine Antimoniate Treatment Failure in Patients with Cutaneous Leishmaniasis: A Retrospective Cohort Study," has been formally accepted for publication in PLOS Neglected Tropical Diseases.

We have now passed your article onto the PLOS Production Department who will complete the rest of the publication process. All authors will receive a confirmation email upon publication.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any scientific or type-setting errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript. Note: Proofs for Front Matter articles (Editorial, Viewpoint, Symposium, Review, etc...) are generated on a different schedule and may not be made available as quickly.

Soon after your final files are uploaded, the early version of your manuscript will be published online unless you opted out of this process. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

For Research Articles, you will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Neglected Tropical Diseases.

Best regards,

Shaden Kamhawi

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

Paul Brindley

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Fig. Linear relationship to the logit of the outcome (Therapeutic failure).

(DOCX)

pntd.0014606.s001.docx^{(2.5MB, docx)}

S2 Fig. Prediction of the model including interaction between number of lesions and lesion area after log transformation.

(DOCX)

pntd.0014606.s002.docx^{(198.3KB, docx)}

S3 Fig. Lasso regression model (variable selection).

(DOCX)

pntd.0014606.s003.docx^{(118.1KB, docx)}

S4 Fig. Calibration curves of 5 models.

(DOCX)

pntd.0014606.s004.docx^{(158.2KB, docx)}

S5 Fig. ROC curve of five models.

(DOCX)

pntd.0014606.s005.docx^{(83.9KB, docx)}

S1 Table. Univariate analysis (binary logistic regression).

(DOCX)

pntd.0014606.s006.docx^{(17.2KB, docx)}

S2 Table. Regression coefficients after Lasso regression.

(DOCX)

pntd.0014606.s007.docx^{(13.2KB, docx)}

S3 Table. Model predictions developed.

(DOCX)

pntd.0014606.s008.docx^{(18.6KB, docx)}

S4 Table. Checking multicollinearity (VIF) of selected model.

(DOCX)

pntd.0014606.s009.docx^{(14.1KB, docx)}

S1 Data. Description and definitions of variables in the raw dataset.

(DOCX)

pntd.0014606.s010.docx^{(15.1KB, docx)}

Attachment

Submitted filename: Answer Letter S-R_LMSH 24-FEB-26.docx

pntd.0014606.s011.docx^{(38.1KB, docx)}

Attachment

Submitted filename: Answer Letter 16-Jun-26.docx

pntd.0014606.s012.docx^{(25.3KB, docx)}

Data Availability Statement

All relevant data are within the manuscript and its Supporting Information files.

[pntd.0014606.ref001] 1.Muñoz Morales D, Suarez Daza F, Franco Betancur O, Martinez Guevara D, Liscano Y. The Impact of Climatological Factors on the Incidence of Cutaneous Leishmaniasis (CL) in Colombian Municipalities from 2017 to 2019. Pathogens. 2024;13(6):462. doi: 10.3390/pathogens13060462 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref002] 2.García-Bustos MF, González-Prieto G, Paniz-Mondolfi AE, Parodi C, Beckar J, Monroig S, et al. Risk factors for antimony treatment failure in American Cutaneous Leishmaniasis in Northwestern-Argentina. PLoS Negl Trop Dis. 2021;15(1):e0009003. doi: 10.1371/journal.pntd.0009003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref003] 3.Alvar J, Vélez ID, Bern C, Herrero M, Desjeux P, Cano J, et al. Leishmaniasis Worldwide and Global Estimates of Its Incidence. PLOS ONE. 2012;7(5):e35671. doi: 10.1371/journal.pone.0035671 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref004] 4.Centers for Disease Control (CDC) PHS. Parasites - Leishmaniasis. 2013. https://www.cdc.gov/parasites/leishmaniasis/gen_info/index.html

[pntd.0014606.ref005] 5.Carrillo-Bonilla LM, Trujillo JJ, Alvarez-Salas L, Vélez-Bernal ID. Study of knowledge, attitudes, and practices related to leishmaniasis: evidence of government neglect in the Colombian Darién. Cad Saude Publica. 2014;30(10):2134–44. doi: 10.1590/0102-311x00139713 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref006] 6.Castro MDM, Erber AC, Arana B, Cota G, Denkinger CM, Harrison N, et al. Involving patients in drug development for Neglected Tropical Diseases (NTDs): A qualitative study exploring and incorporating preferences of patients with cutaneous leishmaniasis into Target Product Profile development. PLoS Negl Trop Dis. 2024;18(2):e0011975. doi: 10.1371/journal.pntd.0011975 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref007] 7.Garcia Bustos MF, Barrio A, Parodi C, Beckar J, Moreno S, Basombrio MA. Miltefosine versus meglumine antimoniate in the treatment of mucosal leishmaniasis. Medicina (B Aires). 2014;74(5):371–7. [PubMed] [Google Scholar]

[pntd.0014606.ref008] 8.Manual de procedimientos para la vigilancia y el control de las leishmaniasis en la región de las Américas. Washington, D.C.: Organización Panamericana de la Salud (OPS). 2022. [Google Scholar]

[pntd.0014606.ref009] 9.Leishmaniasis, semana epidemiológica 25 19 al 25 de junio de 2022. Boletin Epidemológic Semanal Insituto Nacional de Salud Colombia. 2022. [Google Scholar]

[pntd.0014606.ref010] 10.Vásquez L. Terapéutica antileishmania: revisando el pasado, el presente y el futuro. Gaceta Médica de Caracas. 2009;117:93–111. [Google Scholar]

[pntd.0014606.ref011] 11.Pasquau F, Ena J, Sanchez R, Cuadrado JM, Amador C, Flores J, et al. Leishmaniasis as an opportunistic infection in HIV-infected patients: determinants of relapse and mortality in a collaborative study of 228 episodes in a Mediterreanean region. Eur J Clin Microbiol Infect Dis. 2005;24(6):411–8. doi: 10.1007/s10096-005-1342-6 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref012] 12.Castro MDM, Cossio A, Velasco C, Osorio L. Risk factors for therapeutic failure to meglumine antimoniate and miltefosine in adults and children with cutaneous leishmaniasis in Colombia: A cohort study. PLoS Negl Trop Dis. 2017;11(4):e0005515. doi: 10.1371/journal.pntd.0005515 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref013] 13.Llanos-Cuentas A, Tulliano G, Araujo-Castillo R, Miranda-Verastegui C, Santamaria-Castrellon G, Ramirez L, et al. Clinical and parasite species risk factors for pentavalent antimonial treatment failure in cutaneous leishmaniasis in Peru. Clin Infect Dis. 2008;46(2):223–31. doi: 10.1086/524042 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref014] 14.Tuon FF, Amato VS, Graf ME, Siqueira AM, Nicodemo AC, Amato Neto V. Treatment of New World cutaneous leishmaniasis--a systematic review with a meta-analysis. Int J Dermatol. 2008;47(2):109–24. doi: 10.1111/j.1365-4632.2008.03417.x [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref015] 15.Mohammadzadeh M, Behnaz F, Golshan Z. Efficacy of glucantime for treatment of cutaneous leishmaniasis in Central Iran. J Infect Public Health. 2013;6(2):120–4. doi: 10.1016/j.jiph.2012.11.003 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref016] 16.Vásquez L, Scorza Dagert JV, Scorza JV, Vicuña-Fernández N, de Peña YP, López S, et al. Pharmacokinetics of experimental pentavalent antimony after intramuscular administration in adult volunteers. Curr Ther Res Clin Exp. 2006;67(3):193–203. doi: 10.1016/j.curtheres.2006.06.005 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref017] 17.Suprien C, Rocha PN, Teixeira M, Carvalho LP, Guimarães LH, Bonvoisin T, et al. Clinical Presentation and Response to Therapy in Children with Cutaneous Leishmaniasis. Am J Trop Med Hyg. 2020;102(4):777–81. doi: 10.4269/ajtmh.19-0531 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref018] 18.Valencia C, Arévalo J, Dujardin JC, Llanos-Cuentas A, Chappuis F, Zimic M. Prediction score for antimony treatment failure in patients with ulcerative leishmaniasis lesions. PLoS Negl Trop Dis. 2012;6(6):e1656. doi: 10.1371/journal.pntd.0001656 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref019] 19.Romero GA, Guerra MV, Paes MG, Macêdo VO. Comparison of cutaneous leishmaniasis due to Leishmania (Viannia) braziliensis and L. (V.) guyanensis in Brazil: therapeutic response to meglumine antimoniate. Am J Trop Med Hyg. 2001;65(5):456–65. doi: 10.4269/ajtmh.2001.65.456 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref020] 20.Rodrigues AM, Hueb M, dos Santos TARR, Fontes CJF. Factors associated with treatment failure of cutaneous leishmaniasis with meglumine antimoniate. Rev Soc Bras Med Trop. 2006;39(2):139–45. doi: 10.1590/s0037-86822006000200001 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref021] 21.Lago T, Carvalho LP, Nascimento M, Guimarães LH, Lago J, Castellucci L, et al. Influence of Obesity on Clinical Manifestations and Response to Therapy in Cutaneous Leishmaniasis Caused by Leishmania braziliensis. Clin Infect Dis. 2021;73(6):1020–6. doi: 10.1093/cid/ciab236 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref022] 22.Pan American Health Organization PAHO. Guideline for the Treatment of Leishmaniasis in the Americas. 2022. https://iris.paho.org/bitstream/handle/10665.2/56120/guidelineleishmaniasisannex_eng.pdf?sequence=14

[pntd.0014606.ref023] 23.Directrices para el tratamiento de las leishmaniasis en la región de las Américas. Organización Panamericana de la Salud. 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref024] 24.Mendigaña Paez FA. Plan estratégico leishmaniasis 2018 - 2022. 2022. https://www.minsalud.gov.co/sites/rid/Lists/BibliotecaDigital/RIDE/VS/PP/ET/plan-estrategico-leishmaniasis-2018-2022.pdf

[pntd.0014606.ref025] 25.Riley RD, Ensor J, Snell KIE, Harrell FE Jr, Martin GP, Reitsma JB, et al. Calculating the sample size required for developing a clinical prediction model. BMJ. 2020;368:m441. doi: 10.1136/bmj.m441 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref026] 26.Steyerberg EW. Clinical prediction models, a practical approach to development, validation, and updating. 2019.

[pntd.0014606.ref027] 27.Ensor J. Sample size for development of a prediction model. 2023.

[pntd.0014606.ref028] 28.Steyerberg EW, Eijkemans MJ, Harrell FE Jr, Habbema JD. Prognostic modelling with logistic regression analysis: a comparison of selection and estimation methods in small data sets. Stat Med. 2000;19(8):1059–79. doi: 10.1002/(sici)1097-0258(20000430)19:8<1059::aid-sim412>3.0.co;2-0 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref029] 29.Tibshirani R. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological). 1996;58(1):267–88. [Google Scholar]

[pntd.0014606.ref030] 30.Harrell F. Regression Modeling Strategies with Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. 2 ed. Statistics SS i. 2015.

[pntd.0014606.ref031] 31.Senaviratna NAMR, A. Cooray TMJ. Diagnosing Multicollinearity of Logistic Regression Model. AJPAS. 2019;:1–9. doi: 10.9734/ajpas/2019/v5i230132 [DOI] [Google Scholar]

[pntd.0014606.ref032] 32.Jiang J. Applied Medical Statistics. Blackwell W. 2022.

[pntd.0014606.ref033] 33.van Buuren S. Flexible imputation of missing data. 2nd ed. Chapman and Hall/CRC. 2018. [Google Scholar]

[pntd.0014606.ref034] 34.Marshall A, Altman DG, Holder RL, Royston P. Combining estimates of interest in prognostic modelling studies after multiple imputation: current practice and guidelines. BMC Med Res Methodol. 2009;9:57. doi: 10.1186/1471-2288-9-57 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref035] 35.Santos AGA, Sousa JM, de Aguiar AHBM, Torres KCS, Coelho AJS, Ferreira AL, et al. Systematic Review of Treatment Failure and Clinical Relapses in Leishmaniasis from a Multifactorial Perspective: Clinical Aspects, Factors Associated with the Parasite and Host. Trop Med Infect Dis. 2023;8(9):430. doi: 10.3390/tropicalmed8090430 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref036] 36.Bamorovat M, Sharifi I, Dabiri S, Shamsi Meymandi S, Karamoozian A, Amiri R, et al. Major risk factors and histopathological profile of treatment failure, relapse and chronic patients with anthroponotic cutaneous leishmaniasis: A prospective case-control study on treatment outcome and their medical importance. PLoS Negl Trop Dis. 2021;15(1):e0009089. doi: 10.1371/journal.pntd.0009089 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref037] 37.Silva H, Liyanage A, Deerasinghe T, Chandrasekara V, Chellappan K, Karunaweera ND. Treatment failure to sodium stibogluconate in cutaneous leishmaniasis: A challenge to infection control and disease elimination. PLoS One. 2021;16(10):e0259009. doi: 10.1371/journal.pone.0259009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref038] 38.Davies CR, Llanos-Cuentas EA, Pyke SD, Dye C. Cutaneous leishmaniasis in the Peruvian Andes: an epidemiological study of infection and immunity. Epidemiol Infect. 1995;114(2):297–318. doi: 10.1017/s0950268800057964 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref039] 39.Kawiak J, Rokicka-Milewska R, Zeman K, Hoser G, Derulska G, Fornalczyk-Wachowska E, et al. Peripheral blood leukocytes and lymphocyte subpopulations as determined by flow cytometric measurements in healthy children. Folia Histochem Cytobiol. 1995;33(1):33–8. [PubMed] [Google Scholar]

[pntd.0014606.ref040] 40.Older Adults “Chronic Disease Indicators”. US Center for Disease Control and Prevention. https://www.cdc.gov/cdi/indicator-definitions/older-adults.html?

[pntd.0014606.ref041] 41.Castro MDM, Gomez MA, Kip AE, Cossio A, Ortiz E, Navas A, et al. Pharmacokinetics of Miltefosine in Children and Adults with Cutaneous Leishmaniasis. Antimicrob Agents Chemother. 2017;61(3):e02198-16. doi: 10.1128/AAC.02198-16 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pntd.0014606.ref042] 42.Cruz A, Rainey PM, Herwaldt BL, Stagni G, Palacios R, Trujillo R, et al. Pharmacokinetics of antimony in children treated for leishmaniasis with meglumine antimoniate. J Infect Dis. 2007;195(4):602–8. doi: 10.1086/510860 [DOI] [PubMed] [Google Scholar]

[pntd.0014606.ref043] 43.Paniz-Mondolfi AE, Talhari C, García Bustos MF, Rosales T, Villamil-Gomez WE, Marquez M, et al. American cutaneous leishmaniasis in infancy and childhood. Int J Dermatol. 2017;56(12):1328–41. doi: 10.1111/ijd.13664 [DOI] [PubMed] [Google Scholar]

PERMALINK

Development of a predictive model for meglumine antimoniate treatment failure in patients with cutaneous leishmaniasis: Aretrospective cohort study

Susana Ríos-Echavarría

Gilma Hernández Herrera

Hector Ivan García García

Liliana López-Carvajal

Lina Maria Serna-Higuita

Roles

Abstract

Background

Methodology

Principal Findings

Conclusions

Author summary

Introduction

Materials and methods

Ethics statements

Study design and population

Data collection

Outcomes

Sample size estimation

Statistical analysis

Results

Fig 1. Patient selection flow chart.

Table 1. Baseline characteristics stratified according to the presence of TF to systemic MA.

Lasso model

Model development

Table 2. Performance (calibration and discrimination) of 5 candidate models.

Fig 2. Forest plot of selected model.

Fig 3. Distribution of the calculated score divided into quartiles, stratified by the presence or absence of TF.

Table 3. Operating characteristics of the selected model.

Internal validation of the model

Fig 4. Nomogram for predicting TF.

Discussion

Conclusion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Guilherme L Werneck

Elham Kazemirad

Roles

Author response to Decision Letter 1

Decision Letter 1

Guilherme L Werneck

Elham Kazemirad

Roles

Author response to Decision Letter 2

Decision Letter 2

Guilherme L Werneck

Roles

Acceptance letter

Guilherme L Werneck

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases