Evaluation of Nifurtimox Treatment of Chronic Chagas Disease by Means of Several Parasitological Methods

Abstract

Currently, evaluation of drug efficacy for Chagas disease remains a controversial issue with no consensus. In this work, we evaluated the parasitological efficacy of Nifurtimox treatment in 21 women with chronic Chagas disease from an area of endemicity in Chile who were treated according to current protocols. Under pre- and posttherapy conditions, blood (B) samples and xenodiagnosis (XD) samples from these patients were subjected to analysis by real-time PCR targeting the nuclear satellite DNA of Trypanosoma cruzi (Sat DNA PCR-B, Sat DNA PCR-XD) and by PCR targeting the minicircle of kinetoplast DNA of T. cruzi (kDNA PCR-B, kDNA PCR-XD) and by T. cruzi genotyping using hybridization minicircle tests in blood and fecal samples of Triatoma infestans feed by XD. In pretherapy, kDNA PCR-B and kDNA PCR-XD detected T. cruzi in 12 (57%) and 18 (86%) cases, respectively, whereas Sat DNA quantitative PCR-B (qPCR-B) and Sat DNA qPCR-XD were positive in 18 cases (86%) each. Regarding T. cruzi genotype analysis, it was possible to observe in pretherapy the combination of TcI, TcII, and TcV lineages, including mixtures of T. cruzi strains in most of the cases. At 13 months posttherapy, T. cruzi DNA was detectable in 6 cases (29.6%) and 4 cases (19.1%) by means of Sat DNA PCR-XD and kDNA PCR-XD, respectively, indicating treatment failure with recovery of live parasites refractory to chemotherapy. In 3 cases, it was possible to identify persistence of the baseline genotypes. The remaining 15 baseline PCR-positive cases gave negative results by all molecular and parasitological methods at 13 months posttreatment, suggesting parasite response. Within this follow-up period, kDNA PCR-XD and Sat DNA qPCR-XD proved to be more sensitive tools for the parasitological evaluation of the efficacy of Nifurtimox treatment than the corresponding PCR methods performed directly from blood samples.

INTRODUCTION

The protozoan Trypanosoma cruzi is the etiologic agent of Chagas disease, which affects approximately 10 million people in countries of Latin America and the Caribbean (1). Human Chagas disease presents two distinct phases: the acute phase, which appears just after infection, and the chronic phase, which may last several years. After a long asymptomatic phase, around 30% of infected individuals develop chronic disease with severe damage to the heart and digestive system (2). During the acute phase, T. cruzi is usually detected by microscopic examination of fresh or stained blood smears, as well as by xenodiagnosis (XD) and hemoculture. In contrast, during the chronic phase, diagnosis is based on the detection of circulating antibodies. However, due to the long-lasting maintenance of circulating antibodies, it is difficult to use serology as a marker for cure of the disease even after successful treatment of T. cruzi infection (3). The role of the parasite in the outcome of the disease has been demonstrated by successful chemotherapeutical treatment in early acute stages or by a decline in the progression of the disease in the chronic indeterminate period (4). To date, only two drugs have been effectively used in Chagas disease chemotherapy: Nifurtimox (NF) and Benznidazole (BZ), which presents several limitations due to secondary effects (5). The best chemotherapy results have been achieved in acute or early chronic infections in contrast to those observed in late chronic infections (6). Even in children, who are known to better tolerate treatment with these nitroheterocyclic compounds than adults, the cure rate is up to 62% at 2 years of follow-up, and it may vary according to population and geographical location (7). The susceptibility of T. cruzi lineages to different antichagasic drugs has been documented (8, 9, 10). Six different T. cruzi lineages, denominated discrete typing units (DTUs) TcI to TcIV, have been described within the taxon by means of the use of several molecular markers (11). TcI has been determined to be more resistant in vitro and in vivo to several chemotherapeutical drugs; therefore, the infective T. cruzi genotype could be of prognostic value (9, 12). Several methods have been used to monitor the efficacy of therapeutic alternatives. The results of conventional serology remain positive many years after treatment of chronic cases, except in very young individuals who develop acute or congenital Chagas disease (13). Several factors contribute to enduring positive results from conventional serological tests for patients that are parasitologically cured, including the mechanism of autoimmunity, the long-term presence of antibodies due to parasitic antigens in dendritic or cardiac cells, anti-idiotypic antibodies, antilaminine antibodies, and antiepitopes of sugar residues in T. cruzi membranes and others (14). The high sensitivity and specificity of molecular parasitological methods, such as PCR, make those methods suitable tools for the follow-up of a chemotherapeutic treatment in chagasic patients. However, it must be underlined that the validity of molecular and parasitological methods relies on the positive results they can give, and accordingly they have been proposed for earlier assessment of treatment failure of treated chronic Chagas disease patients (15). T. cruzi contains nuclear DNA and kinetoplast DNA (kDNA), both of which contain many repetitive sequences that are highly suitable for sensitive PCR detection due to their high copy numbers (16, 17). Other parasitological methods, such as the classical XD method (18), even though of much less sensitivity than PCR, are still useful in combination with PCR (19, 20). Quantitative PCR (qPCR) is the preferable alternative to determine parasitic load after treatment; at the same time, the conventional PCR directed to minicircles is useful to genotype infective T. cruzi lineages (21, 22). In this work, we aimed to compare several parasitological and molecular methods to evaluate the efficacy of treatment with NF in a group of women with chronic Chagas disease in an area of endemicity with interrupted vectorial transmission.

MATERIALS AND METHODS

Population studied.

The population studied consisted of 21 women with an average age of 38 years (range, 23 to 50) who were serologically positive for T. cruzi as determined by enzyme-linked immunosorbent assay (ELISA) and IgG immunofluorescence analysis. All were from the Province of Choapa, IV Region, which is located between 29°02′S and 32°16′S in the area of transverse valleys of Chile. Informed consent for this study was approved by the Ethics Committee of the Faculty of Medicine of the University of Chile (Resolution 046/2009).

Treatment with NF.

The study group received NF at doses of 6 mg/kg of body weight/day for 60 days in two daily fractions and was monitored under medical supervision according to the Chagas Disease Protocol of the Ministry of Health, Chile (5).

Biological samples.

The samples were collected before treatment and 1 and 13 months posttreatment (once at each time point). A sample of 2 ml of peripheral blood was mixed with an equal volume of 6 M guanidine-HCl–0.1 M EDTA buffer (pH 8), boiled for 15 min at 98°C, and maintained at 4°C until DNA extraction (23). Parallel to the sampling of blood, two XD boxes, each containing seven uninfected Triatoma infestans nymphs at the third growth stage, were administered. After application for 20 to 30 min on the outer side of each arm of each woman, the boxes were kept at 27°C and microscopic examination of the fecal sample from each insect was done in search of trypomastigote forms of T. cruzi (18). The fecal samples obtained at 30, 60, and 90 days were collected in Eppendorf tubes containing 250 μl of phosphate-buffered saline (PBS) buffer (pH 7.2), incubated at 98°C for 15 min, and centrifuged at 4,000 rpm for 3 min. The supernatants were pooled and maintained at −20°C until use. DNA extraction of blood and fecal samples was performed using a Favorgen kit according to the manufacturer's instructions (Biotech Corp., Selangor, Malaysia), and the reaction volume was maintained at −20°C until use. Both samples were collected before and 1 and 13 months after treatment (once at each time point).

Minicircle-based PCR assays.

DNA (5 μl) from blood and fecal samples of T. infestans feed by XD was used as the template for PCR. The reactions were performed thrice with oligonucleotides 121 and 122, which anneal to the four conserved regions present in minicircles of T. cruzi, and a positive control and a negative control were included in each test. The 330-bp PCR product was separated by electrophoresis in 2% agarose gels and visualized by staining with ethidium bromide, as previously described (23).

Hybridization assays.

T. cruzi DTU genotyping was performed twice by DNA blot analysis of DNA minicircle amplicons, as described previously (24). Briefly, 10 μl of each PCR product was subjected to electrophoresis, transferred onto Hybond N+ nylon membranes (Amersham, Little Chalfont, United Kingdom), and cross-linked with UV light to fix the DNA. The membranes were prehybridized for at least 2 h at 55°C and hybridized with different probes of T. cruzi minicircle ³²P-labeled DNA (1 × 10⁶ cpm/membrane). Nylon membranes were then submitted to successive washes under different conditions of stringency (24). For genotyping, different T. cruzi stocks were used to generate the DNA probes to determine the parasite lineage or mixture infecting each patient. Construction of specific probes sp104c11 (TcI, clonet 19), CBBc13 (TcII, clonet 32), NRc13 (TcV, clonet 39), and v195cl1 (TcVI, clonet 43) was performed by amplification of the variable region of T. cruzi minicircles; primers for probe generation were CV1 (5′-GATTGGGGTTGGAGTACTAT-3′) and CV2 (5′-TTGAACGGCCCTCCGAAAAC-3′), which produced a 270-bp fragment (24). The DNA probes were labeled using the random primer method with [α-³²P]dCTP, and the hybridization profiles were analyzed.

Sat qPCR assays.

PCR targeted to the tandemly repeated nuclear satellite (Sat) sequence was carried out twice using blood and XD samples and primers cruzi 1 and cruzi 2 and the specific probe cruzi 3 by real-time PCR (21). Briefly, the mixture contained Taq Platinum buffer (1×), MgCl₂ (3 mM), deoxynucleoside triphosphates (dNTPs) (0.25 mM each), oligonucleotides cruzi 1 (5′-ASTCGGCTGATCGTTTTCGA-3′) and cruzi 2 (5′-AATTCCTCCAAGCAGCGGATA-3′) (0.75 μM each), TaqMan probe cruzi 3 (5′-CACACACTGGACACCAA-3′) (0.25 μM), and Taq DNA polymerase platinum (Invitrogen) (0.5 U) in a final volume of 20 μl. Cycling conditions were 94°C for 5 min and then 40 cycles of 94°C for 10 s, 58°C for 20 s, and 72°C for 20 s in a Rotor-Gene Real-Time Thermocycler (Corbett Life Sciences, Australia). The fluorescence was read at the end of each cycle at 72°C. The parasitic load values in blood were normalized for 10⁶ human cells. These numbers were assessed by quantifying the single-copy apolipoprotein B human gene fragment (ApoB) in order to discard loss of material or carryover of PCR inhibitors (25). This was done by real-time PCR, using primers ApoB Fw (5′-TGGCAACACCAGCACAGACCATTTCAGC-3′) and ApoB Rv (5′-GTAGGAAAGCAGGTCAACCACAGAGTCAG-3′), at a final concentration of 1 μM with Sybr green (1×) (Master Mix; Qiagen). Cycling conditions were 95°C for 15 min and then 40 cycles of 94°C for 15 s, 65°C for 30 s, and 72°C for 30 s. The fluorescence was read at the end of each cycle at 72°C. Amplification was immediately followed by a melting program with initial denaturation at 95°C for 5 s and then a stepwise temperature increase of 0.1°C/s from 76 to 84°C. A dimer was amplified, giving a melting temperature peak (typically 79.5°C).

RESULTS

One panel of 21 chronic Chagas disease patients was evaluated by kDNA PCR and Sat DNA qPCR assays, with DNA taken directly from peripheral blood and from fecal samples of triatomines used for XD, before treatment (Table 1). Only 2 of 21 cases were positive by XD, both at 30 days; in contrast, kDNA PCR-B and kDNA PCR-XD were positive in 12 and 18 cases, respectively. In 18 cases, Sat DNA qPCR-XD (with a range of 92.9 parasites/ml to <1 parasite/ml) and Sat DNA qPCR-B (with a range of 18,300 parasites/10⁶ blood cells to <1 parasite/10⁶ blood cells) results were positive. One case (patient 5) was negative for all tested parasitological methods, with the Apo B assay giving a positive result. We were able to perform T. cruzi genotyping before treatment in 12 and 13 samples from kDNA PCR-B and kDNA PCR-XD amplicons, respectively. Figure 1 shows representative results of this analysis. The observed hybridization patterns represented single T. cruzi lineages or mixtures of two or three T. cruzi lineages, indicating cases with superinfections. In 8 cases, it was possible to compare the infective T. cruzi populations before treatment for each patient using the blood and XD samples. The T. cruzi lineage results were concordant in comparisons performed with blood and XD samples in 2 cases (patients 9 and 14), partially concordant with 1 of 2 identical T. cruzi lineages in 1 case (patient 15), partially concordant with 1 of 3 identical lineages in 2 cases (patients 16 and 18), partially concordant with 2 of 3 identical lineages in 2 cases (patients 6 and 7), and discordant in 1 case (patient 3) (Table 1). The most frequently represented T. cruzi lineages were TcII and TcV, whereas TcI was the least frequently represented lineage. Overall, the T. cruzi lineages detected in blood samples represented 1 single and 11 mixed infections, whereas the samples in triatomines represented 8 single and 5 mixed infections (Fisher's exact test; P value = 0.011). The patients were evaluated to determine the treatment effectiveness after 1 and 13 months using the parasitological diagnosis methods described for the pretherapy evaluation. The results showed that serology remained positive with unchanged serum titers even 13 months after treatment (not shown). Meanwhile, it was possible to evaluate the parasitological response by both kDNA PCR and Sat DNA qPCR assays. While positive kDNA PCR-B and kDNA PCR-XD values dropped after 1 month of treatment, the hybridization patterns still detected some T. cruzi lineages (Fig. 1). Sat DNA qPCR-B also detected very low parasitic loads, most of them at the detection limit (<1 parasite/10⁶ cells). Hybridization with kDNA PCR-B amplicons allowed detection of 12 cases, and hybridization from kDNA PCR-XD detected 3 cases, two of which (patients 18 and 21) gave similar T. cruzi lineage compositions. The T. cruzi lineages found in patients after 1 month of therapy were mainly TcI (12 cases), followed by TcV (4 cases) and TcII (1 case). The total distribution of cases infected after 1 month of therapy combining kDNA PCR-B and kDNA PCR-XD was 11 single infections and 4 mixed infections, while evidence found under the pretreatment conditions revealed 9 single infections and 16 mixed infections. The T. cruzi lineage composition after 13 months of treatment was identical to the one present at pretherapy, namely, two samples with TcII plus TcV and one sample with the addition of TcI to the mixture of TcII plus TcV, suggesting no special sensitivity of different T. cruzi lineages to NF.

Table 1.

Comparison of pretherapy and posttherapy conditions in samples of blood and xenodiagnosis of 21 patients with chronic Chagas disease^a

Patient	Pretherapy result						Posttherapy (1 mo) result						Posttherapy (13 mo) result
	Blood			Xenodiagnosis (XD)			Blood			Xenodiagnosis (XD)			Blood			Xenodiagnosis (XD)
	PCR-B	Hybridation-B	qPCR-B (p/106 cells)b	PCR-XD	Hybridation-XD	qPCR-XD (p/ml)c	PCR-B	Hybridation-B	qPCR-B (p/106 cells)b	PCR-XD	Hybridation-XD	qPCR-XD (p/ml)c	PCR-B	Hybridation-B	qPCR-B (p/10>6 cells)b	PCR-XD	Hybridation-XD	qPCR-XD (p/ml)c
1	+	TcII, TcV	1,340	+		12.6	−	TcI	<1	−		<1	+			+	TcI, TcII, TcV	>1,000
2	+	TcII, TcV	145	+		<1	+		<1	−			−			+	TcII, TcV	>1,000
3	+	TcII	7,140	+	TcV	<1	+			+			+			+	TcII, TcV	>1,000
4	−		<1	+	TcI, TcII, TcV	<1	−			−			−			−		<1
5	−			−			−			−		<1	−			−		<1
6	+	TcI, TcII, TcV	6,370	+	TcI, TcII	92.9	+	TcI, TcV	<1	−			−			+		<1
7	+	TcI, TcII, TcV		+	TcI, TcII	22.6	−	TcI	<1	−			−			−
8	−		<1	+	TcI	18.7	−	TcI	<1	−			−			−
9	+	TcI, TcII, TcV	<1	+	TcI, TcII, TcV		−	TcI	<1	−			−			−
10	−			+		25	−		<1	−			−			−
11	−		973	−			−	TcI	<1	−			−			−
12	−		<1	+	TcV	1.6	−	TcI	<1	−			−			−
13	−		<1	+	TcI	<1	+	TcI, TcV	<1	−			−			−
14	+	TcII, TcV	13,500	+	TcII, TcV	6.18	−			−			−			−
15	+	TcII, TcV	>108.4	+	TcII	4.4	+		<1	−			−			−
16	+	TcI, TcII, TcV	4,350	+	TcI	7.19	−		<1	−	TcI	<1	−			−
17	−		<1	+	TcI	<1	−	TcI	<1	−		<1	−			−
18	+	TcI, TcII, TcV	1,180	+	TcV	<1	−	TcI, TcV	<1	+	TcI, TcII, TcV		−			−
19	−		<1	−		5.4	−	TcI	<1	−		<1	−			−
20	+	TcII, TcV	18,300	+		2.51	+		<1	−		<1	−			−
21	+	TcII, TcV	2,490	+		<1	+	TcV	<1	−	TcV		−			−

^aValues shown as <1 represent samples that were positive for T. cruzi but not quantified.

^bData represent numbers of parasites per 10⁶ cells (p/10⁶ cells). The parasitic load values in blood were normalized to the number of human cells by amplification of an apolipoprotein B human gene fragment.

^cData represent numbers of parasites per ml fecal sample of triatomines (p/ml). Negative satellite DNA findings in samples of XD were confirmed by comparing the threshold cycle (Ct) value corresponding to the sample in study contaminated with a known amount of DNA of T. cruzi Ct obtained with the same concentration to amplify DNA in the absence of the sample.

Fig 1.

Representative results of analysis of T. cruzi genotypes by means of hybridization tests of patient blood samples and T. infestans xenodiagnosis. A, B, C, and D, pretherapy condition; E and F, 1 month posttherapy condition; a, minicircle PCR amplicons stained with ethidium bromide; b, hybridization with TcV probe; c, hybridization with TcII probe; d, hybridization with TcI probe; e, hybridization with TcVI probe. Lane M, 100-bp DNA ladder; lanes b1 to b3, b4 and b5, and b6 and b7, patient blood samples; lanes xd1 to xd3, xd4 and xd5, and xd6 and xd7, T. infestans xenodiagnosis samples.

DISCUSSION

In pursuit of establishing the real efficacy of treatment with NF in a group of adult women with an average age of 38 years, we decided to evaluate them with different parasitological methods. Thus, blood samples from these patients before and after treatment were subjected to real-time PCR analysis targeting the nuclear satellite DNA of T. cruzi, as well as PCR targeting the minicircle DNA of T. cruzi, in order to assess the parasite burden and identify the parasite lineages detectable before and after treatment. In addition, these techniques were applied to detect the parasite in T. infestans fed by XD. As a result, it was possible to agree with other authors who point out that monitoring of these patients should be done in a longitudinal manner over the long term in order to detect occasional bloodstream parasites elicited from tissues of the treated and noncured patients (6). The clinical manifestations and variations in the immune response observed during chagasic infection are not well understood but are believed to be associated with the host or parasite genetic variability. Several studies involving T. cruzi infection have confirmed that genetic diversity is correlated with intrinsic characteristics of the parasite such as virulence, drug resistance, parasitemia, tissue tropism, pathological alterations, capacity to induce host mortality, and pattern of humoral immune response (26, 27, 28, 29). A further important aspect linked to genetic diversity is the susceptibility of parasites to the two pharmacological therapies that are currently available to treat human Chagas disease, namely, BZ (Roche, São Paulo, Brazil) and NF (Bayer, Leverkusen, Germany). It has been reported that 56.0% of T. cruzi strains are susceptible to BZ, while 16.82% are partially susceptible and 27.1% are resistant to the drug, and similar results have been obtained with NF (30). Genetic variation thus appears to be an evolutionary strategy that enables parasites to survive specific chemotherapies. It is suggested that genetic variability in T. cruzi might not only drive pathological disturbances in the mammalian host but might also coordinate the intensity of specific IgGs during the acute and chronic phases of the disease (29). Specific treatment of Chagas disease has been more recently recommended for the early acute phase of infection and for all chagasic patients (4). Previous data regarding the efficacy of NF for the treatment of chronic infection are highly controversial and should be interpreted based on diverse factors, namely, numbers of participants, patient age groups (children or adults), occurrence of reinfections, variability of the follow-up time (months or years), and dose used (31, 32). PCR comparative studies of whole blood were performed in this study. These PCR methods showed that amplification of T. cruzi kDNA may complement XD in assessing parasitemia in chronic chagasic patients and may also be used as a complementary method together with serological tests in blood banks (33). Furthermore, PCR was shown to be a very useful tool for confirmation of diagnoses in patients with doubtful serology results (34). Moreover, the Sat DNA PCR and kDNA PCR test allowed detection of 0.05 to 0.5 parasite genome equivalents/ml of blood, which is above the limit of detection of conventional parasitological methods (35). Determination of the sensitivity of the molecular parasitological methods assayed before treatment indicated that Sat DNA qPCR-B, Sat DNA qPCR-XD, kDNA PCR-XD, and kDNA PCR-B were the most to least sensitive methods. Indeed, satellite DNA and minicircle DNA have been described as representing the most abundant repetitive sequences in T. cruzi (33, 36, 37, 38), and PCR methods based on these sequences were among the most sensitive ones in the context of an international comparative PCR study (35). The qPCR assay is based on satellite sequences. The organization of these sequences has been characterized in the CL Brener stock, the reference organism of T. cruzi genome project (39, 40). Elias et al. found that although satellite DNA is present in different amounts, it is distributed and organized in similar ways in the three strains that represent TcI, TcII, and TcVI DTUs (37), suggesting that this sequence has conserved an important structural role in T. cruzi chromosomes. Indeed, it is inclusive of all parasite DTUs (41), but it is represented with larger numbers of copies in TcII, TcV, and TcVI than in TcI, TcIII, and TcIV (38). Some PCR protocols have been described that have led to unequal results, probably due to differences in the volume of blood processed and in the DNA extraction procedure (38, 41, 42, 43).

The true potential of real-time PCR has been well recognized in situations such as treatment of congenital infections (41, 44), monitoring parasitemia during and after treatment (13, 38, 42, 45), early detection of relapses after heart transplantation (46), and other immunosuppressive circumstances (47). In this study, the PCR assay directed to minicircles of kDNA from triatomine XD fed from each patient allowed identification of live T. cruzi genotypes amplified in the midgut of the triatomines. Before treatment, T. cruzi genotyping allowed us to detect both single and mixed infections in 17 of 21 patients, a very good sensitivity, similar to that previously reported in patients (15, 19). At the same time, single infections are prevalent in XD; mixed infections prevail in human blood samples, suggesting extensive natural selection after triatomine amplification from the mixture of T. cruzi lineages circulating in patients' blood. T. cruzi lineages TcII, TcV, and TcI were the most to the least frequent ones. After treatment, 3 patients conserved their baseline T. cruzi lineages (TcII and TcV), the samples from another 3 patients showed negligible amounts of parasitic DNA detected only by Sat DNA qPCR and thus could not be genotyped, and the results for the remaining 15 patients with baseline PCR-positive findings became negative according to all tested PCR strategies, using blood or XD samples, suggesting favorable treatment response. It is worth noting that the above-mentioned 6 cases with treatment failure were detected by the Sat DNA qPCR-XD assay; therefore, these patients were still infected with live parasites, confirming that posttreatment PCR-positive results are indicative of active infection and not of mere naked DNA released by destroyed parasites. Treatment evaluation at 1 month posttherapy detected very low levels of parasitemia by means of kDNA PCR-XD, and hybridization tests evidenced 3 cases (patients 16, 18, and 21) that became PCR negative at 13 months of follow-up. Most of these patients gave positive results by kDNA PCR-B at 1 month and were infected with TcI as single infections. On the other hand, the negative PCR results obtained 13 months after treatment may be indicative not of cure but only of a transient reduction of parasitic loads, because parasites could still persist in target organs and circulate in blood at levels below the limits of detection of the molecular methods used. Longitudinal studies with a higher number of posttreatment samples and longer periods of follow-up would be necessary to assess a minimum number of PCR-negative posttreatment samples to allow establishing a criterion of enduring parasitological response, leading to cure. As mentioned above, at 13 months of follow-up, 3 patients conserved their T. cruzi TcII and TcV lineages after treatment. In their samples, T. cruzi lineages examined in this study showed no special resistance to treatment with NF. Our results at 1 month posttreatment with several positive cases that later converted to negative at 13 months posttreatment suggest that persistent shedding of parasite DNA in the bloodstream from infected cells had occurred in these treated chagasic patients. This observation correlates with the finding of T. cruzi DNA detected by PCR in the sera of chagasic patients (13). In contrast, studies in a murine model demonstrated that naked DNA becomes undetectable by PCR a few days after its inoculation (48) and indicated a short half-life of T. cruzi DNA in the blood, and so all PCR-positive blood samples are likely to represent detection of live or recently destroyed parasites. The treatment failures demonstrated in six patients do not seem to be associated with age, hepatic function, interrupted treatment, or T. cruzi lineage, since other patients with a parasitological response to treatment presented under the same conditions. Something similar was observed in a group of chronic patients treated with BZ and evaluated by kDNA PCR-B (49). In conclusion, Sat DNA qPCR tests based on XD triatomines from patients' bloodstreams appear to be most useful for monitoring infected subjects undergoing chemotherapy. However, these methods are of high cost and require highly qualified operators. Finally, it would be of interest to evaluate in longer follow-up studies the efficacy of treatment of chronic patients by modern parasitological methods such as the one used here together with the serological methods which certify cure.