Treatment of Bolivian Leishmania braziliensis Cutaneous and Mucosal Leishmaniasis

ABSTRACT.

Although infection with Leishmania braziliensis is perhaps the key reason to treat New World cutaneous leishmaniasis (CL) and mucosal leishmaniasis (ML), the total literature contains relatively few reported cases. With the aim of supplementing the meager clinical information available, we searched the records of Jorochito (Dermatology) Hospital, Bolivia, for the years 1999–2020 and identified treatment records for 1,696 naive CL patients and 355 naive ML patients. Because follow-up was poor for this real-world treatment experience in the developing world, only 255 CL patients (15%) and 114 ML patients (32%) attended follow-up at Hospital. We therefore engaged in an Active Search for “lost” patients, located a further 542 CL patients (32%) and 142 ML patients (44%), thus eventually accomplished follow up on 697 CL patients (41%) and 256 ML patients (72%). Granular adverse event data derived from hospital records is listed for the 902 CL and 86 ML patients administered Glucantime intramuscularly, the 401 CL and 202 ML patients administered Glucantime intravenously, and the 163 CL and 89 ML patients administered miltefosine orally. Efficacy was obtained from hospital records for patients seen at hospital and from patient recall communicated by telephone for the patients found in the Active Search. The overall CL cure rate was 508 of 697 CL patients (73%) with follow-up: intramuscular Glucantime—196/293 (67%); intravenous Glucantime—90/126 (71%); intralesional Glucantime—34/54 (63%); oral miltefosine—52/69 (75%). The overall ML cure rate was 161 of 256 ML patients (63%) with follow-up: intramuscular Glucantime—26/48 (54%); intravenous Glucantime—66/104 (63%); intravenous amphotericin B deoxycholate—19/35 (54%); oral miltefosine—50/71 (70%). We offer this extensive adverse event and efficacy experience as useful guides for clinicians presented with a L. braziliensis infection. The cure rates also illustrate the quandary of New World CL and ML chemotherapy: sufficiently high to be useful but nevertheless needing augmentation with new agents.

INTRODUCTION

Cutaneous leishmaniasis (CL) is endemic in the New World from the United States–Mexico border through Central America and the northern part of South America and down to the region of Rio de Janeiro. The disease, which is characterized by papules that enlarge into ulcers, can be caused by members of the Leishmania vianna subgenus, such as L.v. braziliensis (Lb), L.v. panamensis, and L.v. guayanensis; and members of the Leishmania mexicana complex. ¹ Of these, CL due to Lb in the Andean region is the most feared ² due to the perceived increased frequency of dissemination of Lb CL to the mucus membranes of the nasopharynx causing mucosal leishmaniasis (ML). Mucosal leishmaniasis, which almost always involves the nasal mucosa, ³ can progress to involve the nasal septum, palate, larynx, and pharynx, ³ the cartilage of which slowly erodes over time.

Cutaneous leishmaniasis is treated to speed cure of the disease itself and to attempt to prevent mucosal dissemination. Mucosal leishmaniasis is treated to prevent further destruction of the nasal-oro-pharynx. Standard treatment of both CL and ML has been with parenteral pentavalent antimony (Glucantime) although oral miltefosine has recently been added to the armamentarium. Intravenous (IV) amphotericin B, either deoxycholate or more recently liposomes, is used in recalcitrant cases. As the leading cause of ML, Lb can be argued to be the prime driver of programs to improve chemotherapy of New World CL and ML, but in spite of the centrality of Lb to leishmaniasis chemotherapy, there are relatively few Lb patients described in the literature. The very recent (2020) Cochrane review of “Interventions for New World Cutaneous and Mucocutaneous Leishmaniasis” ⁴ summarizes the data on controlled studies of, by our count, a total of 608 Lb CL patients in 22 studies given Sb, 96 Lb CL patients in two studies given miltefosine, 138 Lb ML patients in seven studies given Sb, and 20 Lb ML patients in one study given miltefosine. To these 704 Lb CL patients and 158 Lb ML patients need to be added reports of uncontrolled studies; nevertheless, the total number of Lb patients described in the literature is relatively few.

Jorochito hospital located in Santa Cruz, Bolivia, is a referral hospital for dermatology patients from a 50 to 700 km catchment area where Lb leishmaniasis is endemic. To attempt to improve understanding of Lb patients and their response to chemotherapy, the hospital records for all 2,050 CL patients (1,696 naïve and 354 relapsed) and all 475 ML patients (355 naïve and 120 relapsed) from 1999 to 2020 were retrospectively reviewed in 2021. Because relatively few treated patients attended Hospital follow-up, we contacted 542 of the CL patients and 142 of the ML patients at their residences thus achieved follow-up of a substantial number of the originally treated patients.

METHODS

Cutaneous leishmaniasis.

Identification of CL patients: All Jorochito hospital patients for whom the diagnosis was CL for the years 1999–2020 were flagged.

CL Case report form: The case report form for CL patients is available upon request to Dr. J Soto. Data on the initial encounter with the patient was derived from the Jorochito hospital records. Data on treatment follow-up was first attempted from the hospital records. If follow-up data was not available in the hospital records, an Active Search was initiated in which a study team member attempted to contact each patient by either of the cell phone numbers included in the hospital records. If the contact was successful, permission was requested to conduct telephone interviews to populate the fields on the case reform form.

Efficacy outcomes: lesion “cure” was defined as “lesion completely re-epithelialized with or without a scar.” Lesion “noncure” was defined as “lesion did not completely re-epithelialize.” Lesion relapse was defined as “lesion enlarged after complete or partial re-epitheliazation.” Lesion “Failure” was defined as the sum of lesion noncures plus lesion relapses.

Mucosal leishmaniasis.

Identification of ML patients: All Jorochito hospital patients for whom the diagnosis was ML for the years 1999–2020 were flagged.

Mucosal leishmaniasis Case report form: The case report form for ML patients is available upon request to Dr. J. Soto. As for CL patients, data on the initial encounter with the patient was derived from the Jorochito hospital records and follow-up data was derived from hospital records plus an Active Search for data not in the hospital records.

CL and ML parasitology.

At Jorochito hospital, biopsies were cultured ⁵ and/or subjected to polymerase chain reaction (PCR). ⁶ Cultured parasites were speciated. ⁷ Smears were Giemsa-stained.

Treatment regimens for CL and ML patients.

Glucantime (Sanofi Aventis, Gentilly, France) was administered either intravenously or intramuscularly at a target dose of 20 mg/kg/day for 20 days (CL) or 28 days (ML), as labeled. Initial patients received no upper limit on daily dose; after 2015, when national guidelines changed, patients received an upper limit of three ampules per day. Since one ampule contains 5 mL of 81 mg Sb/mL, three ampules per day = 1,215 mg Sb which is 20 mg/kg/day for a 60 kg person. Miltefosine (Knight Therapeutics Inc., Rehinburg, Germany) was administered orally at a target dose of 2.5 mg/kg/day for 28 days, with a maximum of 150 mg/day, as labeled—150 mg/day equates to 2.5 mg/kg/day for a 60 kg person. The intralesional Glucantime regiment for CL was five doses, one every other day, of 650 μg Sb/mm² of lesion area per dose. ⁸

Adverse event recording for both CL and ML patients.

Patients administered Glucantime intramuscularly or intravenously were assessed for the following adverse events: subjective adverse events each day of drug administration; and alterations in electrocardiogram (EKG), lipase, amylase, serum glutamic-oxaloacetic transaminase (SGOT), creatinine, and complete blood count measured at the mid-point and at the end of the treatment period. Patients administered miltefosine orally were assessed for the following adverse events: subjective adverse events each day of drug administration; and alterations in SGOT and creatinine at the mid-point and at the end of the treatment period.

Ethical review.

The protocol was approved by the “Comite de Bioetica de la Facultad de Medicina de la Universidad Mayor de San Simón.”

RESULTS

Cutaneous leishmaniasis.

2,050 patients were seen at Jorochito hospital between 1999 and 2020.

Entrance characteristics ( Tables 1 – 5).

Table 1.

Presenting characteristics of CL patients: Ages and genders of CL patients

	Naive patients (N = 1,626)				Relapse patients (N = 342)
	1,302 Males		324 Females		289 Males		53 Females
Age (years)	n	%	n	%	n	%	n	%
0–4	8	0.6	3	0.9	2	0.7	1	2.1
5–14	97	7.5	9	2.8	4	1.3	3	6.4
15--19	364	28	85	26.2	91	30.2	10	21.3
20–44	592	45.5	187	57.7	78	25.9	17	36.2
45–64	221	17	35	10.8	81	26.9	15	31.9
≥ 65	20	1.5	5	1.5	33	11	1	2.1

CL = cutaneous leishmaniasis; n = number.

There were 1,696 naive patients (patients whose lesion had not previously been investigated) and 364 relapsed patients (patients whose lesion had relapsed after previous treatment).

Data on the age and genders of 1,626 of the naive patients (96% of the naïve patients) and on 342 of the relapsed patients (94% of the relapsed patients) are shown in Table 1. About 80% of naive patients and 85% of relapsed patients were male. Most patients were 15–64 years of age, with the most common age being 20–44 years. The gender discrepancy and age range is similar to that of patients we have previously reported. ^{8 – 10}

The appearance of both naïve and relapsed lesions is given in Table 2. The majority of naïve lesions (75%) presented as ulcers, whereas the majority of relapsed lesions (55%) appeared as plaques. Lesion sizes are given in Table 3. The most common lesion size was 401–900 mm² for naive patients and also for relapsed patients. This lesion size is larger than we have seen in previous studies ^{8 – 10} in which patients were actively recruited and lesions recognized earlier.

Table 2.

Presenting characteristics of CL patients: Appearance of CL lesion

	Naive lesions (N = 1,466)		Relapsed lesions (N = 322)
	n	%	n	%
Ulcer	1,106	75.4	102	31.7
Plaque	257	17.5	177	55
Papule/Nodule	101	6.9	36	11.2
Scar	2	0.1	7	2.2

CL = cutaneous leishmaniasis; n = number.

Table 3.

Presenting characteristics of CL patients: CL lesion sizes

	Naive (N = 1,783)		Relapse (692)
Size (mm2)	n	%	n	%
0–100	214	12	66	9.5
101–400	404	22.7	105	15.2
401–900	794	44.5	279	40.3
901–1,600	263	14.8	206	29.8
More than 1,600	108	6.1	36	5.2

CL = cutaneous leishmaniasis; n = number.

The time from first notice of a lesion to presentation to Jorochito hospital is shown in Table 4. Whereas naïve patients generally presented within 6 months, relapsed patients presented at times approximately evenly distributed between 0 months and > 36 months.

Table 4.

Presenting characteristics of CL patients: CL evolution times

	Naive lesions (N = 1,234)		Relapsed lesions (N = 325)
Months	n	%	n	%
0–3	222	18	53	16.3
4–6	741	60	85	26.2
7–12	140	11.3	80	24.6
12–36	88	7.1	63	19.4
> 36	43	3.5	44	13.5

CL = cutaneous leishmaniasis; n = number.

Speciation could be performed on specimens that were culture-positive or that were PCR-positive. (The relatively low number of cultured specimens reflects the lack of emphasis on cultures in routine evaluation of CL) Ninety-one percent were identified as L braziliensis and 9% were untypable. Biopsy, skin test, smear, and PCR were each positive in 64–92% of attempted naïve cases (Table 5). For relapsed lesions, parasitological methods were generally positive although not as often as for naïve lesions: biopsy, smear, and PCR were positive in 48–76% of attempted cases (Table 5).

Table 5.

Presenting characteristics of CL patients: CL parasitology

	Naive lesions		Relapsed lesions
	N	Positive (%)	N	Positive (%)
Smear	1,546	1,098 (71)	332	170 (52.8)
Culture	310	163 (52.6)	91	36 (39.5)
Speciation	205	183 (89.2)	40	35 (87.2)
PCR (from biopsy)	123	113 (91.8)	29	22 (75.8)
Skin test	325	264 (81.2)	230	198 (86.1)
Biopsy	119	76 (63.9)	285	137 (48.1)

CL = cutaneous leishmaniasis; PCR = polymerase chain reaction. The number of lesions is greater than the number of patients because some patients had more than one lesion. For 1,447 naïve patients: 1,147 (79%) had one lesion; 188 (13%) had two lesions; 59 (4%) had three lesions; 53 (4%) had four lesions.

Response of naive CL patients to chemotherapy.

1,303 of the 1,696 naïve patients (77%) were treated with Glucantime intramuscularly (902 patients) or intravenously (401 patients). In addition, 130 patients received intralesional Glucantime and 163 patients received oral miltefosine (mean of 2.2 mg/kg/day for 28 days).

Adverse events (Tables 6–8).

Table 6.

Adverse events for CL patients: Adverse events in 902 naive patients administered Glucantime intramuscularly

Adverse event	N	%	Initiation day: mean (range)	Duration (days): mean (range)	Severity: grade* and % patients with that grade
Local pain	541	60	3 (1–11)	12 (5–15)	1 = 73, 2 = 25, 3 = 2
Myalgias	406	45	7 (5–14)	8 (5–14)	1 = 70, 2 = 26, 3 = 4
Headache	207	23	11 (5–16)	3 (1–7)	1 = 85, 2 = 15, 3 = 0
Abdominal pain	99	11	9 (3–13)	3 (1–4)	1 = 88, 2 = 12, 3 = 0
Rash	30	3	7 (3–10)	4 (2–5)	1 = 95, 2 = 5, 3 = 0
EKG alterations	54	6	9 (5–14)	7 (3–11)	1 = 79, 2 = 19, 3 = 2
Leukopenia	45	5	7 (5–8)	7 (4–9)	1 = 89, 2 = 11, 3 = 0
Low platelets	27	3	7 (5–8)	7 (4–9)	1 = 92, 2 = 8, 3 = 0
SGOT increase	153	17	8 (6–10)	11 (5–13)	1 = 50, 2 = 44, 3 = 6
Lipase/amylase increase	180	20	4 (3–7)	8 (5–11)	1 = 47, 2 = 50, 3 = 3
Creatinine increase	189	21	7 (5–8)	13 (5–12)	1 = 53, 2 = 46, 3 = 1
Electrolyte alteration	126	14	7 (5–8)	13 (5–12)	1 = 60, 2 = 33, 3 = 7

CL = cutaneous leishmaniasis; EKG = electrocardiogram; SGOT = serum glutamic-oxaloacetic transaminase.

^*Severity grades: 1 = not substantially effect the patient, 2 = some intervention needed, 3 = therapy stopped at least transiently For EKG changes: 1 = not substantially effect the patient, 2 = therapy stopped transiently, 3 = therapy terminated.

Table 7.

Adverse events for CL patients: Adverse events in 401 naive patients administered Glucantime intravenously

Adverse event	N	%	Initiation day: mean (range)	Duration (days): mean (range)	Severity: grade* and % patients with that grade
Local pain	3	0.7	8 (4–15)	7 (5 − 12)	1 = 0, 2 = 66, 3 = 33
Myalgias	148	37	6 (3–12)	8 (5–13)	1 = 66, 2 = 30, 3 = 4
Headache	72	18	7 (3 − 15)	4 (2–7)	1 = 90, 2 = 10, 3 = 0
Abdominal pain	28	7	11 (6–16)	5 (3–10)	1 = 82, 2 = 18, 3 = 0
Rash	12	3	7 (3–10)	6 (3–9)	1 = 100, 2 = 0, 3 = 0
EKG alterations	20	5	7 (4–10)	7 (5–14)	1 = 75, 2 = 20, 3 = 5
Leukopenia	16	4	7 (5–8)	7 (5–8)	1 = 100, 2 = 0, 3 = 0
Low platelets	16	4	7 (5–8)	7 (5–8)	1 = 100, 2 = 0, 3 = 0
SGOT increase	80	20	7 (5–8)	11 (5–13)	1 = 72, 2 = 23, 3 = 5
Lipase/amylase increase	88	22	5 (3–9)	4 (2–8)	1 = 53, 2 = 45, 3 = 2
Creatinine increase	72	18	7 (5–8)	8 (5–12)	1 = 69, 2 = 28, 3 = 3
Electrolyte alteration	60	15	7 (5–8)	8 (5–12)	1 = 75, 2 = 20, 3 = 5

CL = cutaneous leishmaniasis; EKG = electrocardiogram; SGOT = serum glutamic-oxaloacetic transaminase.

^*Severity grades: 1 = not substantially effect the patient, 2 = some intervention needed, 3 = therapy stopped at least transiently For EKG changes: 1 = not substantially effect the patient, 2 = therapy stopped transiently, 3 = therapy terminated.

Table 8.

Adverse events for CL patients: Adverse events for 163 naïve patients administered oral miltefosine

	n	%	Initiation day: mean (range)	Duration (days): mean (range)	Severity: grade* and % patients with that grade
Nausea	48	29.4	3 (1–20)	20 (1–25)	1 = 83, 2 = 15, 3 = 2
Vomit	26	16	3 (1–15)	7 (1–14)	1 = 69, 2 = 23, 3 = 8
Diarrhea	16	9.8	5 (3–10)	5 (1–9)	1 = 69, 2 = 31, 3 = 0
Abdominal pain	8	4.9	10 (4–16)	4 (1–6)	1 = 100, 2 = 0, 3 = 0
Motion sickness	14	8.6	11 (4–16)	3 (1–6)	1 = 93, 2 = 7, 3 = 0
Scrotal pain	3	1.8	17 (7–22)	4 (3–5)	1 = 100, 2 = 0, 3 = 0
SGOT increase	61	37.4	6 (4–22)	11 (7–21)	1 = 85, 2 = 15, 3 = 0
Creatinine increase	33	20.2	6 (4–19)	14 (7–20)	1 = 76, 2 = 24, 3 = 0

CL = cutaneous leishmaniasis; n = number; SGOT = serum glutamic-oxaloacetic transaminase.

^*Severity grades: 1 = not substantially effect the patient, 2 = some intervention needed, 3 = therapy stopped at least transiently.

Adverse events are given in Table 6 for intramuscular (IM) Glucantime patients and in Table 7 for IV Glucantime patients. For both IM and IV Glucantime, myalgias, headache, and increases in SGOT/pancreatic enzymes/creatinine were commonly seen. There were 10 IM patients (1%) and four IV patients (1%) who experienced QT prolongation that were grade 2 meaning that therapy was stopped transiently then restarted when the abnormality soon reversed. There was one IM patient (0.1%) and one IV patient (0.2%) who had grade 3 EKG findings meaning that therapy was terminated because the EKG findings did not rapidly resolve. No other patient had to permanently terminated therapy.

Adverse events for oral miltefosine are shown in Table 8. For miltefosine patients, at least one instance of nausea, vomiting, and diarrhea was commonly experienced, and led to one instance of nausea grade 3 and two instances of vomiting grade 3 (transiently stopping treatment). Some increase in SGOT and creatinine was also commonly seen, but was not sufficiently high to warrant transiently stopping treatment in any patient. No patient permanently stopped treatment due to adverse events. Spontaneously reported scrotal pain was mentioned by three patients, beginning on days 17–22 of treatment, lasting for a mean of 4 days, and resolving without medical intervention.

Efficacy.

For this experience in which patients were treated under real-world conditions rather than in tightly controlled Pharma or Academic clinical trials, patients might or might not return to Hospital for follow-up. In fact, of the 1,696 naive patients, only 255 (15%) returned to hospital for posttreatment follow-up.

To generate a larger corpus of posttreatment efficacy data for naïve lesions, we attempted to contact each of the 1,696 patients in 2021. In this Active Search, 542 of the 1,696 patients (32%) could be contacted by a study nurse, generally via telephone although sometimes by an in-person visit. The time between initial treatment and study nurse interview is shown in Table 9 and the time needed to cure as remembered by the patients is shown in Table 10.

Table 9.

CL patient time periods: Time between treatment and interview for the 542 contacted Active Search CL patients

Time (years)	n	%
0–4	304	56.1
5–9	152	28
10–14	60	11.1
≥15	26	4.8

CL = cutaneous leishmaniasis; n = number.

Table 10.

CL patient time periods: Time to cure for the 372 cured Active Search CL patients

Time to cure (months)	n	%
1	77	20.7
2	168	45.2
3	66	17.7
4–6	40	10.8
> 6	21	5.6

CL = cutaneous leishmaniasis; n = number.

Although most patients were interviewed by 4 years after treatment, a considerable percentage of patients were interviewed between 5 and 14 years after treatment and a few were contacted more than 14 years after treatment. We nevertheless believe that the Active Search patients’ recall of their experience with CL was accurate because the lesion and the treatment (IM injections × 20 days, IV injections × 20 days, oral administration × 28 days, intralesional injections) are noteworthy.

For the 155 patients followed at hospital, 19 failed which signifies a 7.5% failure rate for that population. Of the 542 Active Search patients who were contacted, 372 (69%) reported being cured without relapse. According to the patients’ memory, complete re-epithelialization was generally rapid (within 3 months) such that only a few lesions required more than 6 months to completely cure (Table 10).

There were 170 Active Search patients (31%) who did not cure. The failure rate in each RX group was: IM Glucantime—97 of 293 patients (33%); IV Glucantime—36 of 126 patients (29%); intralesional Glucantime—20 of 54 patients (37%); oral miltefosine—17 of 69 patients (25%). Because the patients were not randomly assigned to treatment groups, comparison of the failure rates is problematic, although it can be noted that the failure rate did not differ substantially between the different treatment regimens.

Although this report focuses on cure rate, it is worth noting that CL cures with a scar, which could be functionally or aesthetically unpleasing. For both males and females, functional embarrassment was infrequent, but for females, aesthetic embarrassment was appreciable. Nineteen of 70 female patients (27%) versus 22 of 478 male patients (5%) reported “severe” aesthetic embarrassment.

Further follow-up of relapsed patients.

For the total of 189 patients who failed, most (157 patients) were recorded as being re-treated with IM Glucantime at facilities other than Jorochito hospital. Although 97 of these patients (62%) reported that they were cured by this second administration of drug, 60 (38%) reported that they were not cured. If these IM Glucantime failures are combined with the data of other second therapies, a total of 68 patients (36% of 189) failed their second treatment.

Mucosal leishmaniasis.

475 ML patients were treated at Jorochito hospital, of whom 355 were naive cases and 120 were relapses.

Entrance characteristics (Tables 11–13).

Table 11.

ML presenting characteristics: Symptoms

	Naive (355)		Relapse (120)
	n	%	n	%
Nasal pruritus	355	100.0	89	74.2
Nasal pain	170	47.9	102	85.0
Rhionorrhea	309	87.0	114	95.0
Bleeding	131	36.9	72	60.0
Nasal obstruction	142	40.0	83	69.2
Oral pain	43	12.1	36	30.0
Disphony	32	9.0	27	22.5

ML = mucosal leishmaniasis; n = number.

Table 12.

ML presenting characteristics: Signs

	Naive (355)		Relapse (120)
	n	%	n	%
Nasal mucosa	339	95.5	115	95.8
Nasal skin	156	43.9	81	67.5
Upper lip	142	40.0	84	70
Cheeks	24	6.8	36	30
Oropharynx	89	25.1	90	75
Soft palate	125	35.2	93	77.5
Hard palate	28	7.9	66	55
Larynx	32	9.0	27	22.5

ML = mucosal leishmaniasis; n = number.

Table 13.

ML presenting characteristics: Parasitology

	Naive (355)			Relapse (120)
	Done	Positive		Done	Positive
	n	n	%	n	n	%
Smear	320	141	44.1	105	31	29.5
Culture	38	18	47.4	30	11	36.7
Speciation	16	12	75	24	19	79.2
PCR	11	9	81.8	17	14	82.4
Skin test	144	129	89.6	90	77	85.6
Biopsy	16	7	43.8	20	8	40

ML = mucosal leishmaniasis; n = number; PCR = polymerase chain reaction.

Tabulated data of ML symptoms and signs upon presentation of naïve and relapsed patients at Jorochito hospital are in Tables 11 and 12, respectively. All patients reported nasal symptoms such as pruritis, rhinorrhea, obstruction, or bleeding, and > 95% of patients had lesions of the nasal mucosa. In addition to the almost universal involvement of nasal or oral disease, some naïve patients (9–35%) had symptoms and signs relegated to the pharynx or larynx whereas a larger percentage of relapsed patients (22–75%) had a pharyngeal or laryngeal presentation.

Parasitology (Table 13).

The percentage of samples that were parasitologically positive was similar for smear, culture, and biopsy for naive lesions and also for relapsed lesions. Polymerase chain reaction was positive in a high percentage of samples from the relatively few patients from whom samples were collected. A parasitological diagnosis was made on 44% of naive patients and 36% of relapsed patients. Speciation was performed on relatively few cultures; all parasites were L braziliensis.

Naïve cases: Relationship to prior CL.

A history of antecedent CL was reported by 404 of 435 patients queried (93%) and a scar suggestive of healed CL was seen on 366 of 416 patients examined (88%).

The time period between prior CL and the onset of naive ML as reported by the patient is shown in Table 14. Mucosal leishmaniasis succeeded CL by 1–5 years in about 60% of patients and was > 5 years in about 20% of patients. For relapsed patients, the time between the first and second episodes of ML is shown in Table 15. Although the time period could be as short as 6 months or longer than 36 months, the time between initial ML and relapse was 6–36 months in 83% of patients.

Table 14.

ML patient times: Time between ML and antecedent CL

	Naive patients (355)
Time (years)	n	%
< 1	26	7.3
1–2	93	26.2
3–5	113	31.8
5–10	55	15.5
> 11	28	7.9
No data	40	11.3

CL = cutaneous leishmaniasis; ML = mucosal leishmaniasis; n = number.

Table 15.

ML patient times: Time between first ML and relapse ML

	Relapses (120)
Time (months)	n	%
< 6	13	10.8
6–12	44	36.7
13–24	41	34.2
25–36	15	12.5
> 36	7	5.8

ML = mucosal leishmaniasis; n = number.

The time between initial ML symptoms and first seeking medical attention is shown in Table 16. A full 50% of naive patients (157 of 312) did not seek medical attention until symptoms had been present for more than 3 years. This very long time period may reflect the indolent nature of initial nasal symptoms, and the infrequency of ML disease not just for the developed world but also for the young physicians to whom ML patients present in endemic regions.

Table 16.

ML patient times: Time between initial ML symptoms and first seeking medical attention

	Naive (355)		Relapse (120)
Time (months)	n	%	n	%
0–3	0	0	0	0
4–6	7	2	12	10
7–12	44	12.4	32	26.7
12–36	104	29.3	29	24.2
> 36	157	44.2	33	27.5
No data	43	12.1	14	11.7

ML = mucosal leishmaniasis; n = number.

Response of ML cases to chemotherapy.

All 475 cases were treated at Jorochito hospital. Most treatments used Glucantime intravenously (202 cases) or intramuscularly (86 cases); some patients were treated with oral miltefosine (89 cases) or IV amphotericin B deoxycholate (66 cases).

Adverse effects (Tables 17–19).

Table 17.

Adverse events for ML patients: Treatment with meglumine antimoniate IV (N = 202)

Adverse event	n	%	Initiation day: mean (range)	Duration (days): mean (range)	Severity: grade and number patients with that grade
Local pain	13	6.4	7 (4–17)	6 (4–12)	1 = 10, 2 = 1, 3 = 1
Myalgias	143	70.8	6 (4–15)	9 (4–22)	1 = 112, 2 = 30, 3 = 1
Headache	88	43.6	5 (3 − 19)	3 (1–24)	1 = 77, 2 = 9, 3 = 2
Abdominal pain	26	12.9	8 (2–17)	3 (1–6)	1 = 20, 2 = 6, 3 = 0
Rash	11	5.4	10 (6–15)	4 (1–7)	1 = 100, 2 = 0, 3 = 0
EKG alterations	23	11.4	7 (5–18)	7 (5–14)	1 = 14, 2 = 7, 3 = 2
Leukopenia	27	13.4	7 (4–8)	6 (4–14)	1 = 21, 2 = 6, 3 = 0
Low platelets	7	3.5	7 (4–8)	7 (4–14)	1 = 7, 2 = 0, 3 = 0
SGOT	91	45.0	7 (5–16)	14 (7–30)	1 = 73, 2 = 16, 3 = 2
Lipase/amylase	84	41.6	4 (3–10)	7 (3–15)	1 = 64, 2 = 14, 3 = 6
Creatinine	44	21.8	8 (5–20)	8 (5–21)	1 = 29, 2 = 13, 3 = 2
Electrolytes	38	18.8	9 (5–20)	8 (5–21)	1 = 31, 2 = 2, 3 = 5

EKG = electrocardiogram; IV = intravenous; ML= mucosal leishmaniasis; n = number; SGOT= serum glutamic-oxaloacetic transaminase.

Table 18.

Adverse events for ML patients: Treatment with meglumine antimoniate IM (N = 86)

Adverse event	n	%	Initiation day: mean (range)	Duration (days): mean (range)	Severity: grade and number patients with that grade
Local pain	65	75.6	4 (1–30)	14 (1–30)	1 = 56, 2 = 7, 3 = 2
Myalgias	71	82.6	5 (3–11)	11 (3–22)	1 = 59, 2 = 11, 3 = 1
Headache	45	52.3	7 (4–9)	5 (2–14)	1 = 39, 2 = 6, 3 = 0
Abdominal pain	19	22.1	5 (3–9)	3 (1–5)	1 = 19, 2 = 0, 3 = 0
Rash	8	9.3	7 (3–11)	3 (1–5)	1 = 8, 2 = 0, 3 = 0
EKG alterations	12	14.0	7 (4–15)	10 (7–21)	1 = 7, 2 = 3, 3 = 2
Leukopenia	13	15.1	8 (7–14)	8 (5–24)	1 = 10, 2 = 3, 3 = 0
Low platelets	2	2.3	8 (7–14)	8 (5–11)	1 = 2, 2 = 0, 3 = 0
SGOT	38	44.2	7 (5–20)	9 (7–17)	1 = 23, 2 = 13, 3 = 2
Lipase/amylase	31	36.0	4 (3 − 11)	6 (3–15)	1 = 16, 2 = 13, 3 = 2
Creatinine	27	31.4	9 (7–21)	14 (7–20)	1 = 20, 2 = 6, 3 = 1
Electrolytes	19	22.1	9 (7–21)	12 (7–20)	1 = 18, 2 = 1, 3 = 0

EKG = electrocardiogram; IM = intramuscular; ML= mucosal leishmaniasis; n = number; SGOT = serum glutamic-oxaloacetic transaminase.

Table 19.

Adverse events for ML patients: Treatment with miltefosine PO (N = 89)

Adverse event	n	%	Initiation day: mean (range)	Duration (days): mean (range)	Severity: grade and number of patients with that grade
Nausea	34	38.2	4 (2–16)	11 (1–28)	1 = 27, 2 = 3, 3 = 4
Vomit	29	32.6	5 (2–17)	6 (1–11)	1 = 20, 2 = 7, 3 = 2
Diarrhea	17	19.1	6 (3–13)	5 (1–10)	1 = 15, 2 = 1, 3 = 1
Abdominal pain	11	12.4	8 (2–17)	8 (1–14)	1 = 11, 2 = 0, 3 = 0
Motion sickness	3	3.4	9 (7–11)	5 (1–9)	1 = 3, 2 = 0, 3 = 0
Scrotal pain	0	0	–	–	1 = 0, 2 = 0, 3 = 0
SGOT	33	37.1	7 (5–16)	12 (7–21)	1 = 16, 2 = 13, 3 = 4
Creatinine	28	31.5	8 (4–22)	12 (7–20)	1 = 16, 2 = 9 3 = 3

ML= mucosal leishmaniasis; n = number; PO = per os; SGOT = serum glutamic-oxaloacetic transaminase.

For ML treated with Glucantime IV (Table 17) and Glucantime IM (Table 18), the percentage of patients who experienced adverse effects at some point during therapy was: myalgias (71% and 83%, respectively), headache (44% and 52%), abdominal pain (26% and 22%), increased SGOT (45% and 44%), increased lipase or amylase (42% and 36%), increased creatinine (22% and 31%).

With respect to adverse events of special interest: Of the 202 Glucantime IV and 86 Glucantime IM patients, seven (3.5%) and three (3.5%) respectively experienced grade 2 EKG findings that led to transiently stopping therapy, and two (1%) and two (2%) respectively experienced grade 3 EKG findings that led to termination of therapy. None of the 89 courses of oral miltefosine generated complaints of scrotal pain (Table 19).

Efficacy of treatment of naïve cases (Tables 20, 21).

Table 20.

ML efficacy: Efficacy for 114 Naive ML patients followed at Jorochito hospital

	Treated (N = 355)	Followed at Jorochito
Treatment	n	n	Cured at 6–12 months (%)
Glucantime intramuscular	78	20	12 (65%)
Glucantime intravenous	161	39	30 (77%)
Amphotericin B Deoxycholate	50	19	12 (63%)
Miltefosine oral	66	36	28 (78%)

ML = mucosal leishmaniasis; n = number.

Table 21.

ML efficacy: Efficacy for 142 naive ML patients contacted in Active Search

Treatment	Contacted	Cured (%)
Glucantime intramuscular	28	14 (50%)
Glucantime intravenous	65	36 (55%)
Amphotericin B Deoxycholate	16	7 (44%)
Miltefosine oral	33	22 (67%)

ML = mucosal leishmaniasis.

114 of the 355 naive cases (32%) attended short-term (6–12 month) follow-up at the Hospital. Efficacy for those patients is shown in Table 20. The overall cure rate at this follow-up time was 82 of 114 (72%). The 28% “failure” rate at 6–12 months probably means “lack of initial cure” although it is possible that some of the failures were “relapses” in the sense of substantial initial healing followed by clinical worsening.

To supplement the relatively short-term follow-up of only 32% of patients at hospital, we engaged in an Active Search for the mucosal patients at their residences. Of the 324 naive ML patients who were not already known to have failed via Hospital follow-up, we located 142 patients (44%). Efficacy for those patients is shown in Table 21. Of these 142 patients, 79 (56%) cured by self-report. As for CL, the patients were not randomly assigned to treatment groups so comparison of the cure rates is problematic, but it is noteworthy that there was a considerable failure rate for each treatment.

For the 63 Active Search patients who failed, the reported time to failure was: 1 year—seven patients; 2–3 years—31 patients; 4–5 years—15 patients; 6–10 years—six patients; > 10 years—four patients. Given these long periods of time posttreatment, these “failures” can be considered “relapses” rather than “initial failure to cure.”

The total failure rate for treatment of naive ML lesions was 32 of 114 patients seen at Jorochito plus 63 of 142 patients located in the Active Search or 95 of 256 (37%). Of these 95 failures, 56 (59%) were reported as occurring > 1 year after treatment.

DISCUSSION

Of the common forms of New World cutaneous and ML, the most feared is that caused by L braziliensis due to the infrequent natural cures of CL, ¹¹ the apparently substantial frequency that cutaneous Lb spreads to the nasal-oro mucosa, ¹² and the slow but relentless progression of Lb mucosal disease. ¹³ Yet little about the response of Lb to chemotherapy is clear even in controlled trials about which the Cochrane review summarizes “Evidence certainty was mostly moderate or low, due to methodological shortcomings, which precluded conclusive results . . . ” ⁴ Our aim was simple: to record and describe what happens to Lb patients administered chemotherapeutic agents.

We retrieved data from 1,696 naive Lb CL patients and from 355 naïve Lb ML patients. Our ability to access data from these large numbers of patients was due to the retrospective nature of our data acquisition, and the quality of our datasets needs to be discussed with that methodology in mind. Treatment was performed at Jorochito hospital, and the records were largely complete in that entrance data was available for ≥ 73% of patients and adverse event data was available for all patients. We view the entrance and adverse event data as reliable and of quality comparable to the mostly academic studies in the Cochrane review. ⁴ Although evaluation of CL and ML efficacy requires long periods of follow-up, relatively few patients returned to hospital: 255 (15%) for CL and 114 (32%) for ML. To supplement hospital follow-up data, we engaged in an extensive Active Search for “lost” CL and ML patients, ultimately locating a further 542 (32%) of CL patients and 142 (40%) of ML patients by telephone. We interviewed the Active Search patients by asking each question in our questionnaire, which for CL was aided by electronic transmission of pictures of typical CL lesions. We consider that the patients correctly answered the questions via telephone, and, because the method of contacting patients was independent of subsequently determined treatment outcome, that the efficacy data we recorded well represents our patients’ experiences. An added value of our Active Search effort was that it permits comparison of data for “lost” patients to those who attended scheduled follow-up, a comparison that by definition is not possible for “lost” patients in clinical trials who remain truly lost.

Cutaneous leishmaniasis entrance characteristics: CL patients were generally male between 20 and 44 years of age, characteristics reflecting the acquisition of infection via activity out-of-doors. The majority of naïve lesions presented as an ulcer, as is well-known for Lb, ¹⁴ whereas the majority of relapsed lesions presented as plaques, something that is not well-known. Whereas naive patients generally presented within 6 months, relapsed patients presented at times approximately evenly distributed between 0 and 3 months and > 36 months. Combined with the surprisingly large size of relapsed lesions, this data indicates that our relapsed patients considerably delayed re-seeking medical attention when a lesion relapsed. The delay in seeking medical attention may reflect the treatment situation under real-world conditions in an endemic region, where medical visits have to compete with the requirements of daily living.

Cutaneous leishmaniasis efficacy: For naive cases, 19 of 155 of patients (7.5%) seen in hospital did not cure, whereas 170 of 542 Active Search patients (31%) did not cure. This 4-fold discrepancy indicates that follow-up at the treatment facility was infrequent and strongly skewed to those who cured. However, the assumption in Intent-To-Treat analyses that all “lost” patients are “treatment failures” is also not correct, since 69% of “lost” patients found in our Active Search self-reported as cured. The overall failure rate in patients examined in Hospital plus contacted via telephone was 189/697 or 27%. This figure is similar to the failure rate in the Cochrane 2020 review for the predominately used agent, parenteral Sb (175/489 [36%] in 16 studies by our count ⁴) and suggests that approximately 73% of Lb CL cases are likely to cure with initial therapy. In our study, the cure rate for each of the four major chemotherapeutic agents—IM Glucantime, IV Glucantime, intralesional Glucantime, oral miltefosine—ranged between 63% and 75%. Because the patients were not randomly assigned to treatment groups, the cure rates cannot be directly compared, although it can be noted that the failure rate did not differ substantially between the different treatment regimens. When the patients with initial failure were re-treated, most with IM Glucantime, 64% reported cure. This re-treatment cure rate is similar to the naïve-treatment cure rate, and suggests that at least for Lb CL in Bolivia, parasite resistance to chemotherapy was not generated by the previous treatment course.

Mucosal leishmaniasis entrance characteristics: Nasal involvement in almost all naïve and relapsed ML patients is not surprising. We were surprised, however, by the preponderance of relapsed patients who had a pharyngeal or laryngeal presentation which suggests an augmented tendency to relapse at those sites. The time period between antecedent CL and the first presentation of ML was generally 1–5 years (similar to that described in a large retrospective experience of 327 Brazilian patients ¹⁵); the time between naïve ML and relapsed ML seen at Hospital was generally 6–24 months. These time periods in our study may reflect the indolent and often overlooked symptoms of initial ML versus a heightened awareness of ML symptoms by patients who avail themselves of hospital follow-up.

Mucosal leishmaniasis efficacy: The failure rate for treatment of naïve ML lesions was 32 of 114 patients seen at Jorochito (28%) versus 63 of 142 patients (44%) located in the Active Search. The > 2 year time period at which 18% of the relapses occurred in our study is notable. The realities of clinical trials is that follow-up may be for 90 days, ¹⁵ 1 year, ^{3, 16} or initially for 1 year ¹⁷ with an extension to 2 years, ¹⁸ though Sampaio et al. were able to follow a modest number of patients (32) for 4 years. ¹⁹ Although our Active Search produced patient-reported retrospective data, the finding that almost 20% of failures occurred ≥ 2 years after treatment suggests that clinical trials are unlikely to record the long time periods at which Lb ML may relapse and the true Lb ML relapse rate. As for CL, the ML failure rates for each drug were not strikingly different. The overall ML failure rate of 37% signifies that approximately 63% of ML patients will cure without relapse. The overall Lb failure rates of approximately 27% for CL and 37% for ML are noteworthy because they are based on 697 CL and 256 ML cases.

Adverse effects: Sb: The common occurrence of myalgias, headache, and increases in SGOT/pancreatic enzymes/creatinine in treatment of CL or ML with either IM or IV Glucantime is not a new finding. For example, a review of adverse effects of anti-leishmanial chemotherapy in 2000 lists myalgias/arthralgias, gastrointestinal symptoms, headache, and increase in liver function tests as the most frequently seen adverse reactions in 39–79% publications out of 28 that were reviewed. ²⁰ The granular patient-level data we report for each adverse reaction—the percentage of patients affected, day of initiation, duration, and severity—can however provide more specific guidance for treatment physicians. Electrocardiogram changes are the most worrisome of adverse reactions to Sb. Our CL and ML data shows that a small but not-negligible percentage of patients had to stop therapy transiently (grade 2 reactions: 1–3.5%) or permanently (grade 3 reactions: 0.1–2%) due to QTc prolongation thus the requirement to perform EKGs in the middle and end of Sb therapy should be maintained.

Adverse effects: miltefosine: The prominent frequency of gastrointestinal symptoms and increases in SGOT and creatinine are consistent with the product brochure. ²¹ Since neither ML nor CL manifests systemically, the incidence of systemic adverse effects for the two patient populations should be similar thus we have no explanation for the somewhat higher incidence of gastrointestinal symptoms and creatinine elevations in our ML versus CL patients. The miltefosine (Impavido) product brochure was recently revised in May 2021 ²¹ with particular focus on adverse effects on the male reproductive system. Scrotal pain, decreased ejaculate volume, and absent ejaculation had previously been reported in observational studies, and a prospective study to fulfill a postmarketing requirement found reductions in all sperm parameters at the end of treatment. All sperm parameters, except for sperm concentration, recovered on follow-up assessments at 3 and 6 months after treatment completion. For sperm concentration, small mean decreases persisted on follow-up assessments at 3 and 6 months after treatment completion. In our present real-world experience, only three of 252 CL and ML patients (1%) spontaneously mentioned scrotal pain, grade 1 in each instance, and no male spontaneously complained about their reproductive capacity.

Overall, we offer this extensive experience as a guide to what we consider is likely to occur when patients are infected with Lb and administered chemotherapy. The granular adverse event data may be useful for clinicians preparing to treat something they rarely see, infection with Leishmania braziliensis or indeed infection with any cause of CL or ML. The substantial but not overwhelming cure rates—73% for Lb CL, 63% for Lb ML—may also be useful guides for the clinician, and illustrates the quandary of New World CL and ML chemotherapy generally: sufficiently high to be valuable but nevertheless sufficiently low to entice development of new agents.