Nifurtimox (NF) is one of the only two drugs currently available for Chagas disease (ChD) treatment. However, data on NF safety are scarce, and many physicians defer or refuse NF treatment because of concerns about drug tolerance.
KEYWORDS: Trypanosoma cruzi, nifurtimox, children, adults, adverse drug reactions, adverse drug effects
ABSTRACT
Nifurtimox (NF) is one of the only two drugs currently available for Chagas disease (ChD) treatment. However, data on NF safety are scarce, and many physicians defer or refuse NF treatment because of concerns about drug tolerance. In a retrospective study of adverse drug reactions (ADRs) associated with NF treatment of ChD, children received NF doses of 10 to 15 mg/kg/day for 60 to 90 days, and adults received 8 to 10 mg/kg/day for 30 days. A total of 215 children (median age, 2.6 years; range, 0 to 17 years) and 105 adults (median age, 34 years; range, 18 to 57 years) were enrolled. Overall, 127/320 (39.7%) patients developed ADRs, with an incidence of 64/105 adults and 63/215 children (odds ratio [OR] = 3.7; 95% confidence interval [CI], 2.2 to 6.3). We observed 215 ADRs, 131 in adults (median, 2 events/patient; interquartile range for the 25th to 75th percentiles [IQR25–75], 1 to 3) and 84 in children (median, 1 event/patient; IQR25–75 = 1 to 1.5) (Padjusted < 0.001). ADRs were mainly mild and moderate. Severe ADRs were infrequent (1.2% in children and 0.9% in adults). Nutritional, central nervous, and digestive systems were the most frequently affected, without differences between groups. Treatment was discontinued in 31/320 (9.7%) patients without differences between groups. However, ADR-related discontinuations occurred more frequently in adults than in children (OR = 5.5, 95% CI = 1.5 to 24). Our study supports the safety of NF for ChD treatment. Delaying NF treatment due to safety concerns does not seem to be supported by the evidence. (This study has been registered in ClinicalTrials.gov under identifier NCT04274101.)
INTRODUCTION
Chagas disease (ChD) is a silent but devastating disease caused by infection with the parasite Trypanosoma cruzi. The disease is endemic to the Americas, from the United States to Argentina, with over 7 million people currently infected in Latin America. ChD has been transmitted to many countries of the world via immigration, with most cases reported in Europe, North America, Australia, and Japan (1).
Most patients are asymptomatic during acute ChD. The acute phase is followed by a chronic asymptomatic stage that eventually leads to irreversible heart disease in up to 30% of the infected patients many years later (2). Over 7,000 deaths occur yearly due to complications of Chagas disease.
The current treatment for ChD is limited to two nitro-heterocyclic drugs, nifurtimox (NF) and benznidazole (BZ), which have similar effectiveness (3). Despite both drugs having been available since the early 1970s, treatment recommendations vary significantly from country to country, and the evidence base for the current treatment regimens is limited. This failure to treat may possibly be explained by many obstacles, including health care providers’ low awareness of the disease and its treatment options, overblown concerns about drug side effects, limited access to health care for many patients, lack of an optimal straightforward test of treatment response, widespread drug shortages and irregular supplies, and regulatory barriers.
The most commonly observed NF adverse drug reactions (ADRs) are anorexia and weight loss, irritability, sleepiness, and other nervous system signs and symptoms (4, 5). NF is also associated with rash, pruritus, and drug-associated hepatitis but less frequently than BZ. Depression, peripheral neuropathy, and psychiatric symptoms have also been reported. Similar to BZ, NF-associated ADRs seem much more common and severe in adults (6) and are usually mild in children, including neonates (7, 8). However, current tolerability data come mainly from small-cohort studies. The pharmacological basis for the differences in the incidence of adverse events remain to be studied (9).
Here, we present results from a large cohort of ChD patients which included infants, children, and adults treated with NF, describing and comparing safety among adults and pediatric patients.
RESULTS
Population characteristics.
Medical records of ChD patients treated at our institution were reviewed, and 372 patients who were prescribed NF were identified. However, 52 patients were excluded because they did not start treatment (i.e., did not fill the NF prescription). The remaining 320 patients were included in the study. A total of 215 pediatric patients (0 to 17 years) and 105 adults were included. Among children, 56 were 0 to 7 months old, 43 were 8 months to 1 year old, 44 were 2 to 6 years old, 37 were 7 to 11 years old, and 35 were 12 to 17 years old. A low rate of loss to follow-up was observed, as only 16/320 (5%) patients abandoned the study (Fig. 1).
FIG 1.
Study flowchart summarizing the medical records included in this study.
In general, male and female subjects were well balanced in children but not in adults, where 87.6% of subjects were female (most of them were mothers of children assisted by our service). The route of infection was congenital in 131, undetermined in 154, vectorial in 32, and by blood transfusion in 3 cases (Table 1).
TABLE 1.
Demographic data
| Parametera | Children | Adults | Total |
|---|---|---|---|
| Gender [no. (%)] | |||
| F | 109 (50.7) | 92 (87.6) | 201 (62.8) |
| M | 106 (49.3) | 13 (12.4) | 119 (37.2) |
| Age | |||
| Median (Q1, Q3) | 31.0 (6.00, 108) mo | 34 (29, 38) yr | |
| Mean (SD) | 59.5 (63.1) mo | 34.5 (7.37) yr | |
| Min–max | 1.00–215 mo | 19–57 yr | |
| Route of infection [no. (%) of patients] | |||
| Vector | 22 (10.2) | 10 (9.5) | 32 (10.0) |
| Congenital | 120 (55.8) | 11 (10.5) | 131 (40.9) |
| Blood transfusion | 1 (0.5) | 2 (1.9) | 3 (0.9) |
| Undetermined | 72 (33.5) | 82 (78.1) | 154 (48.1) |
| Clinical examination at diagnosis [no. (%) of patients] | |||
| Asymptomatic | 185 (86.0) | 104 (99.0) | 289 (90.3) |
| Symptomatic | 30 (14.0) | 1 (1.0) | 31 (9.7) |
| Total [no. (%) of patients] | 215 (67) | 105 (33) | 320 (100) |
F, female; M, male; Q1 and Q3 denote the first and third quartiles, respectively.
Overall, most patients (289/320 [90.3%]) were asymptomatic, with only 31/320 (9.7%) being symptomatic. With regard to route of infection, patients infected by the vector route were predominantly symptomatic 12/32 (37.5%), and the most frequent symptom was ocular chagoma (11/12 cases). In patients infected by the remaining routes (congenital, undetermined, and blood transfusion), symptomatic cases were infrequent 19/288 (6.6%). Symptomatic cases were observed mainly in infants under 2 years of age (16/19 [84.2%]), and the most affected organ was the liver (12/19 cases).
A clinical improvement was observed in all but one symptomatic patient during treatment. Only one 3-month-old infant, coinfected with Trypanosoma cruzi and HIV by the transplacental route, did not show clinical improvement. The patient developed encephalitis and myocarditis related to HIV infection and died during NF treatment due to respiratory complications.
ADRs incidence and relationship.
Overall, 127/320 patients (39.7%) developed ADRs, with incidences of 64/105 (60.9%) in adults and 63/215 (29.3%) in children (odds ratio [OR] = 3.7; 95% confidence interval [CI] = 2.2 to 6.3; Padjusted < 0.001).
A total of 215 ADRs in 127 patients were observed. In 64 adults, 131 ADRs were observed, with a median of 2 events per patient (interquartile range for the 25th to 75th percentiles [IQR25–75] = 1 to 3), and in 63 children, 84 ADRs were observed, with a median of 1 event per patient (IQR25–75 = 1 to 1.5) (Padjusted < 0.001).
NF-related ADRs were more frequent in adults (79.7%) than in children (7.9%) (OR = 9.9, 95% CI = 3.7 to 33; Padjusted < 0.001) (Table 2). No significant differences were observed in the numbers of ADRs between patients during the acute (vectorial and infants younger than 8 months of age) and the chronic phases of infection (Padjusted = 0.4).
TABLE 2.
Adverse events classified by severity and their relationship to treatmenta
| ADR classification | No. (%) among: |
|||
|---|---|---|---|---|
| Children |
Adults |
|||
| Patients (n = 63) | ADRs (n = 84) | Patients (n = 64) | ADRs (n = 131) | |
| Severity | ||||
| Mild | 47 (74.6) | 60 (71.4) | 54 (84.4) | 101 (77.1) |
| Moderate | 8 (12.7) | 8 (9.5) | 14 (21.9) | 27 (20.6) |
| Severe | 1 (1.6) | 1 (1.2) | 1 (1.6) | 1 (0.8) |
| No data | 11 (17.5) | 15 (17.9) | 1 (1.6) | 2 (1.5) |
| Relationship to treatment | ||||
| None | 5 (7.9) | 10 (11.9) | 2 (3.1) | 2 (1.5) |
| Unlikely | 1 (1.6) | 1 (1.2) | 1 (1.6) | 1 (0.8) |
| Probable | 47 (74.6) | 53 (63.1) | 20 (31.2) | 31 (23.7) |
| Certain | 5 (7.9) | 5 (6) | 51 (79.7) | 95 (72.5) |
| No data | 11 (17.5) | 15 (17.9) | 1 (1.6) | 2 (1.5) |
| Total | 63 (100.0) | 84 (100.0) | 64 (100.0) | 131 (100.0) |
Relationship classification was recorded according to the World Health Organization criteria. For each age group, two columns are shown. “Patients” shows the number of patients presenting at least one ADR and its corresponding percentage. “ADRs” shows the observed number of ADRs and its corresponding percentage. Notice that patients could present more than one ADR belonging to different categories (i.e., mild and moderate).
The number of ADRs was associated with incomplete treatment: 2.58 events/patient in subjects that discontinued treatment versus 1.55 events/patient in those with completed treatment (Padjusted < 0.001). Moreover, differences remained when the cohorts were considered separately (adults, Padjusted = 0.028; children, Padjusted = 0.016).
The profile of the 215 ADRs is shown in Table 3. The systems most commonly affected were the nutritional system (75/215 [34.9%]), central nervous system (CNS) (61/215 [28.4%]), and digestive system (38/215 [17.7%]), without differences between adults and children. Few adverse skin effects (20/215 [9.3%]) were observed in both groups, and hematological ADRs (7/215 events) were observed only in children (OR = ∞, 95% CI = 2.3 to ∞; Padjusted = 0.005).
TABLE 3.
ADR occurrence and patient incidence by site or organ systema
| Site or system and ADR | No. (%) among: |
|||
|---|---|---|---|---|
| Children |
Adults |
|||
| Patients (n = 63) | ADRs (n = 84) | Patients (n = 64) | ADRs (n = 131) | |
| Body as a whole | 1 (1.6) | 1 (1.2) | 4 (6.2) | 4 (3.1) |
| Fever | 1 (1.6) | 1 (1.2) | 2 (3.1) | 2 (1.5) |
| Asthenia | 2 (3.1) | 2 (1.5) | ||
| Cardiovascular | 2 (3.1) | 2 (1.5) | ||
| Syncope | 1 (1.6) | 1 (0.8) | ||
| Tachycardia | 1 (1.6) | 1 (0.8) | ||
| Digestive | 11 (17.5) | 12 (14.3) | 21 (32.8) | 26 (19.8) |
| Vomiting | 8 (12.7) | 8 (9.5) | 4 (6.2) | 4 (3.1) |
| Nausea | 2 (3.2) | 2 (2.4) | 8 (12.5) | 8 (6.1) |
| Dyspepsia | 1 (1.6) | 1 (1.2) | 8 (12.5) | 8 (6.1) |
| Abdominal pain | 1 (1.6) | 1 (1.2) | 4 (6.2) | 4 (3.1) |
| Others | 2 (3.1) | 2 (1.5) | ||
| Hematological | 6 (9.5) | 7 (8.3) | ||
| Eosinophilia | 3 (4.8) | 3 (3.6) | ||
| Leukopenia | 3 (4.8) | 3 (3.6) | ||
| Thrombocytopenia | 1 (1.6) | 1 (1.2) | ||
| Nutritional | 30 (47.6) | 31 (36.9) | 40 (62.5) | 44 (33.6) |
| Wt loss | 13 (20.6) | 13 (15.5) | 33 (51.6) | 33 (25.2) |
| Hyporexia | 18 (28.6) | 18 (21.4) | 11 (17.2) | 11 (8.4) |
| Musculoskeletal | 1 (1.6) | 1 (1.2) | 2 (3.1) | 2 (1.5) |
| Myalgias | 2 (3.1) | 2 (1.5) | ||
| Chest pain | 1 (1.6) | 1 (1.2) | ||
| CNS | 19 (30.2) | 22 (26.2) | 28 (43.8) | 39 (29.8) |
| Headache | 6 (9.5) | 6 (7.1) | 20 (31.2) | 20 (15.3) |
| Irritability | 15 (23.8) | 15 (17.9) | 7 (10.9) | 7 (5.3) |
| Dizziness | 5 (7.8) | 5 (3.8) | ||
| Others | 1 (1.6) | 1 (1.2) | 5 (7.8) | 7 (5.3) |
| Respiratory | 1 (1.6) | 3 (3.6) | ||
| Acute bronchitis | 1 (1.6) | 1 (1.2) | ||
| Rhinorrhea | 1 (1.6) | 1 (1.2) | ||
| Influenza syndrome | 1 (1.6) | 1 (1.2) | ||
| Skin | 6 (9.5) | 7 (8.3) | 10 (15.6) | 13 (9.9) |
| Rash | 6 (9.5) | 7 (8.3) | 6 (9.4) | 6 (4.6) |
| Urticaria | 4 (6.2) | 4 (3.1) | ||
| Others | 3 (4.7) | 3 (2.3) | ||
| Psychiatric | 1 (1.6) | 1 (0.8) | ||
| Depression | 1 (1.6) | 1 (0.8) | ||
| Total | 63 (100.0) | 84 (100.0) | 64 (100.0) | 131 (100.0) |
A total of 215 ADRs occurred in the 127 patients. Low-frequency symptoms were grouped into the general category “others” and include the following: for the digestive system, epigastralgia and pyrrhosis; for the CNS, depression, nightmare, dysarthria, insomnia, loss of memory panic attacks, and tinnitus; for the skin, facial edema and pruritus.
Time of onset of ADRs was recorded for 130/215 ADRs in 64/127 patients (50.4%). Overall, 93.8% of ADRs appeared within 30 days of treatment. The median times to onset of ADRs (IQR25–75) were 5 (2.5 to 10.2) days for the digestive system, 5.5 (1.5 to 11) days for CNS, 8 (0.75 to 26) days for the nutritional system, and 13 (10 to 20) days for skin.
ADRs had an earlier onset in adults, who presented a median onset time of 6.5 days (IQR25–75 = 1.5 to 9), than in children, who presented a median onset time of 12 days (IQR25–75 = 9.7 to 21; Padjusted <0.001).
Severity.
ADR severity is described in Table 2. Most ADRs were mild (74.9%) or moderate (16.3%) and resolved without sequelae. Severe ADRs were infrequent (2/215 [0.9%]) and occurred in 1 adult and in 1 child (Table 2). The adult was a woman in her 30s who developed a headache with a defined relationship to NF. ADRs resolved without consequences, but NF treatment was discontinued. The child was in the age range of 8 months to 2 years and presented with severe leukopenia, which also resolved without consequences, and he was able to complete treatment.
The severity of ADRs was associated with treatment discontinuation in adults, with 67% of discontinuations in the 15 subjects that presented moderate or severe ADRs compared to 6.1% in the 49 subjects that presented mild ADRs (OR = 27.8; 95% CI = 5.1 to 212; Padjusted < 0.001). In children, no association was observed (OR = 2.4; 95% CI = 0.04 to 35; Padjusted = 0.4).
Serious events were observed in 2 patients. One patient showed a serious event unrelated to NF. The patient died due to complications related to HIV infection, and this case was described previously. The other patient, another female in her 30s (different from the one described above), presented tremors, dysarthria, and panic attacks which required hospitalization. This patient made a full recovery, but treatment was discontinued.
Treatment completion.
Overall treatment was completed in 289/320 patients (90.3%), and the completion rates were not different between children (92.6%) and adults (85.7%) (Table 4). Treatment was discontinued in 31/320 (9.7%) patients, but only 14/320 (4.4%) discontinuations were related to ADRs (Table 5). ADR-related discontinuations occurred more frequently in adults (9.5%) than in children (1.9%) (OR = 5.5; 95% CI = 1.5 to 24; Padjusted = 0.008). Notably, the main cause of treatment discontinuation in children was moderate skin ADRs (3/4 children) (Tables 4 and 5).
TABLE 4.
Treatment discontinuation and interruption
| Treatment status | No. (%) of: |
||
|---|---|---|---|
| Children | Adults | Total patients | |
| Completed treatment | |||
| Yes | 199 (92.6) | 90 (85.7) | 289 (90.3) |
| No | 16 (7.4) | 15 (14.3) | 31 (9.7) |
| Discontinuation cause | |||
| Patient decision | 6 (2.8) | 1 (1.0) | 7 (2.2) |
| Adverse effect | 4 (1.9) | 10 (9.5) | 14 (4.4) |
| Death | 1 (0.5) | 1 (0.3) | |
| Lost to follow-up | 5 (2.3) | 4 (3.8) | 9 (2.8) |
| Treatment complete | 199 (92.6) | 90 (85.7) | 289 (90.3) |
| Temporary interruption | |||
| Yes | 10 (4.7) | 10 (9.5%) | 20 (6.2) |
| No | 205 (95.3) | 95 (90.5) | 300 (93.8) |
| Total | 215 (67.0) | 105 (33.0) | 320 (100.0) |
TABLE 5.
ADRs causing treatment discontinuationa
| Patient group | Age range (yr) | Gender | Symptom(s) | Treatment length (days) | Second treatment |
|---|---|---|---|---|---|
| Children | 0–1 | F | Irritability | 15 | NF; completed |
| 7–17 | M | Rash | 15 | ||
| 7–17 | F | Rash | 27 | ||
| 2–6 | M | Abdominal pain, rash, fever, eosinophilia | 21 | ||
| Adults | 30–39 | F | Headache, irritability, nausea | 3 | BZ; discontinued because of ADR |
| 40–49 | F | Hyporexia | 23 | BZ; completed | |
| 50–59 | F | Hyporexia | 11 | BZ; completed | |
| 30–39 | F | Syncope, dizziness, headache, dyspepsia, nausea | 12 | BZ; completed | |
| 40–49 | F | Headache, abdominal pain | 15 | ||
| 20–29 | F | Abdominal pain, vomiting, myalgias | 11 | ||
| 30–39 | F | Wt loss, irritability, tremors, dysarthria, panic attack | 18 | BZ; discontinued because of ADR | |
| 20–29 | F | Rash, itching, headache, dyspepsia | 12 | BZ; completed | |
| 30–39 | F | Headache, dizziness | 8 | BZ; completed | |
| 30–39 | F | Psychomotor agitation | 10 | BZ; discontinued because of ADR | |
NF, nifurtimox; BZ, benznidazole. All patients had a good response to symptomatic treatment.
A total of 20/320 (6.2%) subjects temporarily interrupted NF (Table 4), without differences among adults (9.5%) and children (4.7%) (OR = 2.1, 95% CI = 0.77 to 5.9; Padjusted = 0.14). Temporary interruption was due to ADRs in 7 adults and 5 children, and that in the remaining 8 patients was a result of personal decision. The median temporary interruption length was 7 days (IQR25–75: 2 to 9 days), with no differences between adults and children.
ADRs leading to temporary treatment interruptions were digestive (5 reactions), CNS (3 reactions), skin (2 reactions), cardiovascular (2 reactions), and nutritional (2 reactions) among adults, and digestive (2 reactions), skin (2 reactions) and hematologic (2 reactions) among children. Treatment discontinuation occurred in 4 of these 12 patients (2 adults, 2 children). The remaining patients completely recovered after symptomatic treatment and/or transient interruption of NF.
For the 289 patients who completed treatment, the mean dose, number of tablets, and length of treatment are described in Table 6.
TABLE 6.
Treatment descriptiona
| Parameter | Children | Adults | Total patients |
|---|---|---|---|
| Dose | |||
| Median (Q1, Q3) (mg/kg) | 11.0 (10.0, 12.0) | 9.00 (8.20, 9.70) | 10.0 (9.19, 12.0) |
| Missing [no. (%) of patients] | 5 (2.5) | 5 (1.7) | |
| No. of doses | |||
| Median (Q1, Q3) | 2.00 (2.00, 3.00) | 2.00 (2.00, 3.00) | 2.00 (2.00, 3.00) |
| Missing [no. (%) of patients] | 6 (3.0) | 6 (2.1) | |
| Days of treatment | |||
| Median (Q1, Q3) | 62.0 (60.5, 73.0) | 30.0 (29.0, 32.0) | 61.0 (33.0, 69.0) |
| Concomitant medication [no. (%) of patients] | |||
| Yes | 3 (1.5) | 3 (1.0) | |
| No | 196 (98.5) | 90 (100) | 286 (99.0) |
| Compliance [no. (%) of patients] | |||
| Yes | 166 (83.4) | 75 (83.3) | 241 (83.4) |
| No | 33 (16.6) | 15 (16.7) | 48 (16.6) |
| Temporary interruption [no. (%) of patients] | |||
| Yes | 7 (3.5) | 8 (8.9) | 15 (5.2) |
| No | 192 (96.5) | 82 (91.1) | 274 (94.8) |
| Total [no. (%) of patients] | 199 (92.6) | 90 (100) | 289 (90.3) |
A total of 289 patients completed NF treatment. Q1 and Q3 denote the first and third quartiles, respectively.
Pediatric cohort analysis.
A subanalysis by age group of the pediatric cohort was carried out to elucidate whether there was any trend in the number, frequency, or type of ADRs. A high rate of treatment completion (92.6%) was observed without differences within pediatric age groups (Table 7). No significant differences among pediatric subgroups were observed in the rates of temporary interruption or in the rate of ADRs (25 to 38%). In addition, a high rate of compliance (greater than 70%) was observed in all groups and, notably, was >90% in patients under 2 years old (Table 7). The most frequently observed pediatric ADRs were nutritional, followed by CNS adverse reactions, without clear differences among age groups. Headache was most frequent in children older than 7 years of age. Digestive and hematological events were mainly observed in children younger than 2 years old (Table 8).
TABLE 7.
Treatment details for the pediatric cohort
| Treatment parameter | Result for patients in age groupa |
|||||
|---|---|---|---|---|---|---|
| 0–7 mo (n = 56) | 8 mo–1 yr (n = 43) | 2–6 yr (n = 44) | 7–11 yr (n = 37) | 12–17 yr (n = 35) | Total (n = 215) | |
| Treatment complete | ||||||
| Yes | 53 (94.6) | 41 (95.3) | 41 (93.2) | 35 (94.6) | 29 (82.9) | 199 (92.6) |
| No | 3 (5.4) | 2 (4.7) | 3 (6.8) | 2 (5.4) | 6 (17.1) | 16 (7.4) |
| Treatment discontinued | ||||||
| Patient decision | 1 (2.3) | 2 (4.5) | 1 (2.7) | 2 (5.7) | 6 (2.8) | |
| Adverse event | 1 (2.3) | 1 (2.3) | 0 (0) | 2 (5.7) | 4 (1.9) | |
| Death | 1 (1.8) | 1 (0.5) | ||||
| Lost to follow-up | 2 (3.6) | 1 (2.7) | 2 (5.7) | 5 (2.3) | ||
| Temporary interruption | ||||||
| Yes | 2 (3.6) | 1 (2.3) | 2 (4.5) | 1 (2.7) | 4 (11.4) | 10 (4.7) |
| No | 54 (96.4) | 42 (97.7) | 42 (95.5) | 36 (97.3) | 31 (88.6) | 205 (95.3) |
| Compliance | ||||||
| Yes | 48 (90.6) | 37 (90.2) | 29 (70.7) | 27 (73.0) | 25 (71.4) | 166 (83.4) |
| No | 5 (9.4) | 4 (9.8 | 12 (29.3) | 12 (27.0) | 10 (28.6) | 33 (16.6) |
| Adverse events | ||||||
| Yes | 14 (25.0) | 14 (32.6) | 17 (38.6) | 11 (29.7) | 7 (20.0) | 63 (29.3) |
| No | 42 (75.0) | 29 (67.4) | 27 (61.4) | 26 (70.3) | 28 (80.0) | 152 (70.7) |
| No. of events per patient [median (Q1, Q3)] | 1.0 (1.0, 1.00) | 1.0 (1.0, 1.0) | 1.0 (1.0, 2.0) | 1.0 (1.00, 1.0) | 2.0 (1.0, 2.0) | 1.0 (1.0, 1.5) |
All results are numbers (percentages) of patients, except for the last row, which is number of events.
TABLE 8.
ADR occurrence and patient incidence by organ system in the pediatric cohorta
| Site or system and ADR | No. (%) in age group |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 0–7 mo |
7 mo–1 yr |
2–6 yr |
7–11 yr |
12–17 yr |
||||||
| Patients with ADRs (n = 14) | ADRs (n = 17) | Patients with ADRs (n = 14) | ADRs (n = 16) | Patients with ADRs (n = 17) | ADRs (n = 24) | Patients with ADRs (n = 11) | ADRs (n = 13) | Patients with ADRs (n = 7) | ADRs (n = 14) | |
| Body as a whole | 1 (5.9) | 1 (4.2) | ||||||||
| Fever | 1 (5.9) | 1 (4.2) | ||||||||
| Digestive | 4 (28.6) | 4 (23.5) | 2 (14.3) | 2 (12.5) | 2 (11.8) | 2 (8.3) | 3 (42.9) | 4 (28.6) | ||
| Vomiting | 4 (28.6) | 4 (23.5) | 2 (14.3) | 2 (12.5) | 1 (5.9) | 1 (4.2) | 1 (14.3) | 1 (7.1) | ||
| Nausea | 2 (28.6) | 2 (14.3) | ||||||||
| Dyspepsia | 1 (14.3) | 1 (7.1) | ||||||||
| Abdominal pain | 1 (5.9) | 1 (4.2) | ||||||||
| Hematological | 2 (14.3) | 2 (11.8) | 2 (14.3) | 3 (18.7) | 1 (5.9) | 1 (4.2) | 1 (9.1) | 1 (7.7) | ||
| Eosinophilia | 1 (7.1) | 1 (6.2) | 1 (5.9) | 1 (4.2) | 1 (9.1) | 1 (7.7) | ||||
| Leukopenia | 2 (14.3) | 2 (11.8) | 1 (7.1) | 1 (6.2) | ||||||
| Thrombocytopenia | 1 (7.1) | 1 (6.2) | ||||||||
| Nutritional | 6 (42.9) | 6 (35.3) | 8 (57.1) | 8 (50) | 8 (47.1) | 8 (33.3) | 5 (45.5) | 5 (38.5) | 3 (42.9) | 4 (28.6) |
| Wt loss | 1 (7.1) | 1 (5.9) | 4 (28.6) | 4 (25) | 3 (17.6) | 3 (12.5) | 3 (27.3) | 3 (23.1) | 2 (28.6) | 2 (14.3) |
| Hyporexia | 5 (35.7) | 5 (29.4) | 4 (28.6) | 4 (25) | 5 (29.4) | 5 (20.8) | 2 (18.2) | 2 (15.4) | 2 (28.6) | 2 (14.3) |
| Musculoskeletal | 1 (14.3) | 1 (7.1) | ||||||||
| Chest pain | 1 (14.3) | 1 (7.1) | ||||||||
| CNS | 3 (21.4) | 3 (17.6) | 3 (21.4) | 3 (18.7) | 8 (47.1) | 10 (41.7) | 5 (45.5) | 6 (46.2) | ||
| Headache | 2 (11.8) | 2 (8.3) | 4 (36.4) | 4 (30.8) | ||||||
| Irritability | 3 (21.4) | 3 (17.6) | 3 (21.4) | 3 (18.7) | 7 (41.2) | 7 (29.2) | 2 (18.2) | 2 (15.4) | ||
| Nightmare | 1 (5.9) | 1 (4.2) | ||||||||
| Respiratory | 1 (14.3) | 3 (21.4) | ||||||||
| Acute bronchitis | 1 (5.6) | 1 (3.7) | ||||||||
| Rhinorrhea | 1 (5.6) | 1 (3.7) | ||||||||
| Influenza syndrome | 1 (14.3) | 1 (7.1) | ||||||||
| Skin | 2 (14.3) | 2 (11.8) | 2 (11.8) | 2 (8.3) | 1 (9.1) | 1 (7.7) | 1 (14.3) | 2 (14.3) | ||
| Rash | 2 (14.3) | 2 (11.8) | 2 (11.8) | 2 (8.3) | 1 (9.1) | 1 (7.7) | 1 (14.3) | 2 (14.3) | ||
| Total | 14 (100) | 17 (100) | 14 (100) | 16 (100) | 17 (100) | 24 (100) | 11 (100) | 13 (100) | 7 (100) | 14 (100) |
A total of 84 ADRs occurred in the 64 pediatric patients.
DISCUSSION
We present a large retrospective study of ChD patients (children and adults) treated with NF. We observed a low rate of loss to follow-up, similar to other prospective pediatric and adult ChD studies (7, 8, 10, 11). Our institution followed the standard-of-care guidelines for ChD patients. However, as with any retrospective study conducted over a long period of time, possible sources of bias must be considered, due to insufficient detailed information about the incidence or severity of ADRs.
ADR incidence in our study was strongly associated with patient age, since adults had a higher incidence of ADRs and related treatment discontinuations than children. Moreover, NF-related ADRs were significantly more frequent in adults. Although pediatric pharmacologic studies on NF are still lacking, it is possible that observed differences in incidence of ADRs among the pediatric and adult subpopulations could be due to age-related differences in drug metabolism. NF is metabolized in the liver, and, similar to many other drugs (12–14), it would be expected to undergo faster liver clearance in children than adults, leading to shorter half-lives and steady-state plasma concentrations. Pharmacological NF studies are under way to clarify this issue (www.clinicaltrials.gov, no. NCT01927224 and NCT02625974).
NF-related ADRs had a lower incidence in our study than in previous studies in children (6, 8, 10) and in adults (4, 6, 11). Moreover, the rate of ADRs per patient in our adult cohort was 4-fold lower than that reported in previous studies (11). These differences could be explained by the larger size of our cohort, which would yield a more accurate estimate of the incidence of ADRs, and by the shorter time of treatment (30 days) prescribed for our patients. However, some alternative explanations are possible, such as socioeconomic and ethnic differences among the studied populations, detection bias (our data were retrospectively collected for this analysis), or other, yet-unknown issues.
A high incidence of ADRs in patients with acute oral T. cruzi infection was reported (6). In our study, we did not observe a higher incidence of ADRs in patients during the acute phase of infection. This difference could be related to the different route of infection, mainly congenital, and to the fact that the majority of our cases were asymptomatic or had only mild symptoms of infection.
Regarding severity, the ADRs observed in our cohort were mostly mild. Severe events were infrequent. and all patients recovered with no sequelae. This is comparable to previous studies in children (8, 10). However, we report a lower incidence of severe and serious adverse events in adult patients than previous studies (4, 11).
ADRs appeared earlier in adults than in children but most of them occurred within the first month of treatment, suggesting that most NF ADRs are not dependent on cumulative doses. ADR profiles were similar between pediatric and adult populations, except for hematological ADRs, which appeared only in children, mainly those under 2 years of age. The most frequent ADRs were nutritional, mainly hyporexia and weight loss, in line with the observations of many other ChD researchers (4, 7, 8, 11).
CNS ADRs are a major concern. Seizures and psychiatric ADRs related to NF have been described previously (15). NF has a high level of fat solubility and is well distributed throughout the tissues, including the CNS. Even though NF is a substrate of breast cancer resistance protein (BCRP), which may be responsible for the active transfer of NF out of the CNS, it is possible that some patients may have BCRP polymorphisms that decrease this transfer, thus exposing them to a high level of NF in the CNS for longer periods of time (i.e., exposing them to CNS ADRs) (16, 17). However, this remains a hypothesis to be tested. We observed mostly headaches and CNS irritability as NF ADRs (16). An adult patient presented tremors, dysarthria, and panic attacks, which required hospitalization, but recovered without consequences after temporary treatment interruption. Evaluation of CNS ADRs such as headaches, in children and particularly in infants was difficult, because only older children can accurately express these symptoms. However, associated signs, such as unexplained irritability, food refusal, or vomiting, were not reported by caregivers or observed by our pediatricians, who have significant experience in evaluating ChD pediatric patients.
The profile of digestive ADRs was similar to that reported in other NF studies (4, 5, 10, 11, 18). In children, digestive ADRs could be related to the lack of an appropriate pediatric formulation of NF which requires pill fractioning. As pill fragments are not easily (or willingly) swallowed by small children, this sometimes results in vomiting and other problems that may not be specifically related to the active drug. A new pediatric NF formulation in the late stages of clinical development (www.clinicaltrials.gov, no. NCT02625974), over time, would eventually be helpful to address the pediatric formulation gap and possibly decrease the incidence of digestive ADRs.
Skin reactions are the main ADRs observed during treatment with the alternative drug BZ (19), but they are much less frequently described with NF. Accordingly, we observed few skin manifestations in our cohort, and only 3 children developed skin reactions that led to treatment discontinuation. The differences in ADR profiles between BZ and NF are not clearly explained to date, particularly given that they are both nitro drugs. Unfortunately, NF metabolism and metabolite profiles for both drugs remain poorly studied, which hampers any speculation on the pharmacological reasons behind these ADR differences (20, 21).
Pediatric treatment discontinuation rates due to NF ADRs in our study were comparable to those in other studies (8, 10). Even though we observed higher discontinuation rates in adults than in children, these adult rates (i.e., 14.3%) were lower than those reported in previous studies, which ranged from 19.8% to 43.8% (4, 11). This difference could be due to the fact that most of the adult patients in our study were relatives of previously treated and cured children, to whom we offered treatment as part of our ChD family screening and treatment protocol. This population is highly motivated to persist and complete treatment.
In summary, our results suggest that NF is a safe drug to use in both pediatric and adult ChD patients. Considering the retrospective nature of the study, these results are not conclusive, and further prospective studies would be required in order to confirm our results.
Since more primary infections of ChD occur during childhood, early diagnosis and treatment of children is vital to prevent long-term ChD sequelae. In the light of our findings, which strongly suggest, that NF is safe in children, we believe that treatment should not be delayed.
MATERIALS AND METHODS
Study design and population.
This was a retrospective age-stratified study to assess the safety and tolerability of oral NF in subjects with ChD. The study protocol was approved by the research and teaching committee and the bioethics committee of the Buenos Aires Children’s Hospital Dr Ricardo Gutierrez.
All patients were treated and followed up at the Parasitology and Chagas Service, Hospital de Niños Ricardo Gutiérrez. Buenos Aires, Argentina, from January 1980 to July 2019. Patients were stratified according to age. Subanalysis among children was done considering the following age groups: 0 to 7 months, 8 months to 1 year, 2 to 6 years, 7 to 11 years, and 12 to 17 years.
Chagas disease diagnostic criteria.
For infants younger than 8 months, ChD diagnostic criteria included direct observation of T. cruzi using the parasitological concentration method (microhematocrit test [MH]) or xenodiagnosis (XD). For older patients, the criteria included 2 reactive serological tests: enzyme-linked immunosorbent assay (ELISA), indirect hemagglutination (IHA), or direct agglutination (DA).
Exclusion criteria included cases where nifurtimox was prescribed but not taken (patients did not come back; n = 52) and cases where medication data were not properly documented. (n = 10).
For the safety analysis, the study population included all patients who started treatment, regardless of whether they completed the treatment.
Treatment.
NF treatment (120-mg tablets; Bayer) was prescribed in doses of 10 to 15 mg per kg of body weight per day, divided into two or three daily doses for 60 to 90 days for infants and children and 8 to 10 mg per kg for 30 days for adults, according to national guidelines at the time of diagnosis. Note that regimens were modified for shorter treatment in more recent years. Enrollment of children started in January 1980, and enrollment of adults started in July 2008. Infant NF doses were provided as fractionated tablets prepared by a pharmacist and administered with water or mother's milk. Medication was provided to patients or their guardians in monthly batches, and compliance was assessed by counting remaining tablets at each visit. Treatment was considered complete when patients took the medication for at least 60 days for children and 30 days for adults.
Data collection.
Data were collected from medical records of treated patients and entered into an Access clinical database (ACD) designed for this study. All individual data sets were anonymized.
Demographic data, clinical and biochemical assessments, and complementary studies were collected during follow-up. Baseline data values were obtained at the beginning of the treatment. Following the standard of care of our service for treated ChD patients, visits were carried out at 7 days, at 30 days, at the end of treatment, every 3 months during the first year posttreatment, and every 6 to 12 months thereafter.
ADRs were evaluated through laboratory tests, clinical interviews, and physical examinations and classified according to World Health Organization (WHO) definitions (22, 23). Causality assessment was performed using the WHO criteria for causality assessment.
Information on treatment duration and dosage, temporary interruptions and concomitant medications was systematically collected from medical records and documented in the clinical database.
Statistical analysis.
Continuous variables were expressed with means and medians, as applicable, with the corresponding standard deviation or interquartile range. Categorical variables were expressed in percentages. To test for significance, as appropriate, a t test or Wilcoxon unpaired rank test was carried out for continuous variables and Fisher exact tests for categorical ones. P values were adjusted by false discovery rate (Benjamini-Hochberg procedure). Adjusted P values (Padjusted) of <0.05 were considered statistically significant. The statistical package R was used (24).
ACKNOWLEDGMENTS
This work was supported by an Institution initiative research by Bayer (grant PR5071885). A.J.B., F.G.-B., G.M., and J.A., are researchers in the National Scientific and Technical Research Council of Argentina (CONICET).
J.A. is a consultant for Bayer. All other authors report no potential conflicts.
REFERENCES
- 1.Schmunis GA, Yadon ZE. 2010. Chagas disease: a Latin American health problem becoming a world health problem. Acta Trop 115:14–21. doi: 10.1016/j.actatropica.2009.11.003. [DOI] [PubMed] [Google Scholar]
- 2.Altcheh JM, Freilij H. 2019. Chagas disease: a clinical approach. Springer Nature, Cham, Switzerland. [Google Scholar]
- 3.Pereiro AC 2019. Guidelines for the diagnosis and treatment of Chagas disease. Lancet 393:1486–1487. doi: 10.1016/S0140-6736(19)30288-0. [DOI] [PubMed] [Google Scholar]
- 4.Jackson Y, Alirol E, Getaz L, Wolff H, Combescure C, Chappuis F. 2010. Tolerance and safety of nifurtimox in patients with chronic chagas disease. Clin Infect Dis 51:e69–e75. doi: 10.1086/656917. [DOI] [PubMed] [Google Scholar]
- 5.Olivera MJ, Cucunubá ZM, Álvarez CA, Nicholls RS. 2015. Safety profile of nifurtimox and treatment interruption for chronic Chagas disease in Colombian adults. Am J Trop Med Hyg 93:1224–1230. doi: 10.4269/ajtmh.15-0256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Alarcón de Noya B, Ruiz-Guevara R, Noya O, Castro J, Ossenkopp J, Díaz-Bello Z, Colmenares C, Suárez JA, Noya-Alarcón O, Naranjo L, Gutiérrez H, Quinci G, Torres J. 2017. Long-term comparative pharmacovigilance of orally transmitted Chagas disease: first report. Expert Rev Anti Infect Ther 15:319–325. doi: 10.1080/14787210.2017.1286979. [DOI] [PubMed] [Google Scholar]
- 7.Freilij H, Altcheh J. 1995. Congenital Chagas’ disease: diagnostic and clinical aspects. Clin Infect Dis 21:551–555. doi: 10.1093/clinids/21.3.551. [DOI] [PubMed] [Google Scholar]
- 8.Moya PR, Paolasso RD, Blanco S, Lapasset M, Sanmartino C, Basso B, Moretti E, Cura D. 1985. [Treatment of Chagas’ disease with nifurtimox during the first months of life]. Medicina (B Aires) 45:553–558. [PubMed] [Google Scholar]
- 9.Garcia-Bournissen F, Altcheh J, Giglio N, Mastrantonio G, Della Védova CO, Koren G. 2009. Pediatric clinical pharmacology studies in Chagas disease. Paediatr Drugs 11:33–37. doi: 10.2165/0148581-200911010-00012. [DOI] [PubMed] [Google Scholar]
- 10.Bianchi F, Cucunubá Z, Guhl F, González NL, Freilij H, Nicholls RS, Ramírez JD, Montilla M, Flórez AC, Rosas F, Saavedra V, Silva N. 2015. Follow-up of an asymptomatic Chagas disease population of children after treatment with nifurtimox (Lampit) in a sylvatic endemic transmission area of Colombia. PLoS Negl Trop Dis 9:e0003465. doi: 10.1371/journal.pntd.0003465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Forsyth CJ, Hernandez S, Olmedo W, Abuhamidah A, Traina MI, Sanchez DR, Soverow J, Meymandi SK. 2016. Safety profile of nifurtimox for treatment of Chagas disease in the United States. Clin Infect Dis 63:1056–1062. doi: 10.1093/cid/ciw477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hines RN 2007. Ontogeny of human hepatic cytochromes P450. J Biochem Mol Toxicol 21:169–175. doi: 10.1002/jbt.20179. [DOI] [PubMed] [Google Scholar]
- 13.Hines RN 2008. The ontogeny of drug metabolism enzymes and implications for adverse drug events. Pharmacol Ther 118:250–267. doi: 10.1016/j.pharmthera.2008.02.005. [DOI] [PubMed] [Google Scholar]
- 14.Kearns GL, Abdel-Rahman SM, Alander SW, Blowey DL, Steven Leeder J, Kauffman RE. 2003. Developmental pharmacology—drug disposition, action, and therapy in infants and children. N Engl J Med 349:1157–1167. doi: 10.1056/NEJMra035092. [DOI] [PubMed] [Google Scholar]
- 15.Wegner DH, Rohwedder RW. 1972. The effect of nifurtimox in acute Chagas’ infection. Arzneimittelforschung 22:1624–1635. [PubMed] [Google Scholar]
- 16.Watson CP, Dogruel M, Mihoreanu L, Begley DJ, Weksler BB, Couraud PO, Romero IA, Thomas SA. 2012. The transport of nifurtimox, an anti-trypanosomal drug, in an in vitro model of the human blood–brain barrier: evidence for involvement of breast cancer resistance protein. Brain Res 1436:111–121. doi: 10.1016/j.brainres.2011.11.053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Garcia-Bournissen F, Altcheh J, Panchaud A, Ito S. 2010. Is use of nifurtimox for the treatment of Chagas disease compatible with breast feeding? A population pharmacokinetics analysis. Arch Dis Child 95:224–228. doi: 10.1136/adc.2008.157297. [DOI] [PubMed] [Google Scholar]
- 18.Valencia NC, Mancilla M, Ra months D, Zulantay I, Molina M, Torres A, Corral G, Apt W. 2012. Tratamiento de la enfermedad de Chagas crónica en Chile: efectos adversos de nifurtimox. Rev Ibero Latinoam Parasitol 71:97–108. [Google Scholar]
- 19.Altcheh J, Moscatelli G, Moroni S, Garcia-Bournissen F, Freilij H. 2011. Adverse events after the use of benznidazole in infants and children with Chagas disease. Pediatrics 127:e212–e218. doi: 10.1542/peds.2010-1172. [DOI] [PubMed] [Google Scholar]
- 20.Hall BS, Bot C, Wilkinson SR. 2011. Nifurtimox activation by trypanosomal type I nitroreductases generates cytotoxic nitrile metabolites. J Biol Chem 286:13088–13095. doi: 10.1074/jbc.M111.230847. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hall BS, Wilkinson SR. 2012. Activation of benznidazole by trypanosomal type I nitroreductases results in glyoxal formation. Antimicrob Agents Chemother 56:115–123. doi: 10.1128/AAC.05135-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Edwards IR, Aronson JK. 2000. Adverse drug reactions: definitions, diagnosis, and management. Lancet 356:1255–1259. doi: 10.1016/S0140-6736(00)02799-9. [DOI] [PubMed] [Google Scholar]
- 23.Edwards IR, Biriell C. 1994. Harmonisation in pharmacovigilance. Drug Saf 10:93–102. doi: 10.2165/00002018-199410020-00001. [DOI] [PubMed] [Google Scholar]
- 24.R Core Team. 2019. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria: https://www.R-project.org/. [Google Scholar]