SEROLOGIC REBOUNDS AFTER ONE-YEAR-LONG TREATMENT FOR CONGENITAL TOXOPLASMOSIS

Abstract

Rises in specific IgG antibody titers to Toxoplasma gondii, also known as serologic rebounds, have been reported in children with congenital toxoplasmosis during the first months of life1-4 but their frequency, pathophysiology and clinical significance remain unclear. In addition a possible relation between their occurrence and nature and the duration of specific treatment against T. gondii has been considered but not clearly demonstrated.1, 4 In France monthly serologic monitoring of nonimmune pregnant women allows for early diagnosis of maternal infection and detection of fetal infection by antenatal diagnosis.5 Consequently specific treatment combining pyrimethamine (PYR) and sulfadiazine (SDZ) is often administered early after diagnosis of congenital toxoplasmosis, either during pregnancy or shortly after birth. Since 1994 postnatal treatment with PYR plus SDZ has been standardized and administered continuously throughout the first year of life according to the recommendations of the Chicago Collaborative Treatment Trial (CCTT).3 Since the introduction of this 1-year-long treatment in our center, serologic rebounds have been observed in most infected children within the months after discontinuation of therapy. These findings prompted us to study further the frequency, timing of appearance and immunologic profile of the serologic rebound as well as its clinical significance. Patients and methods. Charts of all congenitally infected infants followed up in our center who received treatment according to the CCTT recommendations were reviewed. All children were treated either antenatally or immediately after birth and had sequential clinical and ophthalmologic evaluations in addition to serologic monitoring. A series of 84 children who were followed up during the first year of life while receiving treatment was selected on the basis that an additional clinical, ophthalmologic and serologic evaluation was performed at least 2 months after discontinuation of 1-year therapy. Treatment was initiated antenatally in 32 children (after positive prenatal diagnosis) or immediately after birth in remainder. Five infants had specific signs associated with congenital infection, i.e. retinal scars associated with cerebral calcifications in 2 cases. These symptoms were diagnosed during early infancy, either at birth (3 cases) or at 3 and 4 months of life (2 cases). Among these 84 cases 50 children were selected for whom the follow-up serum sample drawn at least 2 months after discontinuation of therapy and kept frozen at −20°C was still available for a retrospective serologic study. For the 84 patients both serum samples collected at the time of discontinuation of treatment and in the following months were previously examined in parallel with the use of the HS agglutination assay for specific IgG.6 A serologic rebound was defined as at least a 4-fold increase in antibody titer. To characterize the anti-Toxoplasma IgG antibody response, the HS/AC differential agglutination test7 and an avidity assay8 were performed on each posttreatment serum sample from the 50 children. The HS/AC test is currently used in adult serum samples to differentiate recently acquired infections from those acquired in a more distant past.7 AC antigen detects IgG antibodies produced against tachyzoites formed only during the acute stage of infection. The test was performed as previously described,7 and results were interpreted as follows: an HS/AC ratio ≥4 was consistent with a chronic pattern, whereas a ratio below 4 was interpreted as an acute pattern. The avidity assay, which is commonly used in adult specimens, measures the antigen-binding affinity of specific IgG, dependent on the stage of the infection.8 The Vidas Toxo IgG Avidity (bioMérieux, Marcy-l'Etoile, France) was used according to the manufacturer's instructions. In this commercially available enzyme-linked immunosorbent assay, washing buffers are replaced by a solution containing a dissociating agent (urea 6 M) for determination of avidity.9 Sera with an IgG antibody avidity ≥30% were considered as samples exhibiting high avidity (chronic pattern, infection acquired >4 months in adults), whereas sera with an antibody avidity index <30% were considered intermediate/low avidity (acute pattern). Detection of specific IgM and IgA antibody was performed by using the respective class-specific immunosorbent agglutination assays (bioMérieux).10 The results were expressed as an agglutination index on a scale from 0 to 12. All indices ≥3 were considered positive. Results. Retrospective analysis of follow-up serologic data revealed a serologic rebound after discontinuation of 1-year treatment in 82 (97.6%) of the 84 congenitally infected children. A continuous and progressive decrease in the specific IgG antibodies was shown throughout the treatment period. Thus the geometric mean of 167.3 IU/ml of specific IgG (measured by the direct HS agglutination test) at birth declined to 6.3 IU/ml at the time of treatment discontinuation (median, 0.99 year), i.e. to a 3.76% fraction of the initial value. In contrast the serum sample collected 2 to 9 months after cessation of treatment (mean, 4.6 ± 0.13 months) showed an almost 50-fold increase in the specific IgG titer, amounting to a geometric mean of 308.1 IU/ml. In all 50 children for whom a posttreatment sample was still available, a serologic rebound was observed. The differential agglutination test performed in 45 of 50 posttreatment samples (5 patients had an HS agglutination test titer below 100 IU/ml, which is the minimum required for the HS/AC test) revealed a chronic pattern in 41 cases (91.1%) and an acute pattern in 4. Of the 48 sera tested in the IgG antibody avidity assay (2 serum samples were exhausted), 41 (85.4%) had high avidity (mean index, 0.552 ± 0.116), and intermediate/low avidity was present in 7 patients (mean, 0.132 ± 0.088). Correlation of the results obtained by the 2 tests showed 39 patients exhibiting a chronic and 3 an acute profile, with discrepant findings in 2 patients (whereas 6 remained undetermined), thus amounting to an agreement of 95.4% (42 of 44). Specific IgM and IgA antibodies were detected in 22 (44%) and 31 (62%) rebound samples, respectively, of which the simultaneous presence of both isotypes was noted in 16 (32%) rebound samples (Table 1). In contrast specific IgM and IgA were detected in the immediate prerebound sample in only 3 and 2 patients, respectively.TABLE 1: Toxoplasma serology in 50 congenitally infected infants with IgG rebound after discontinuation of 1-year-long therapy Clinical and ophthalmologic evaluations, performed at the time of the rebound (median age, 16.6 months), did not reveal signs of infectious relapse in any of the children. Furthermore all 13 children available for longer follow-up (mean, 24.6 months; range, 1.5 to 3.5 years) were free of clinical signs throughout this period, i.e. 8 months after the rebound. On the other hand all 5 children who had Toxoplasma-specific signs were diagnosed months before the serologic rebound, and no recurrences or new active lesions were noted at the time of the rebound. A single one of these, who had cerebral calcifications at birth, developed hydrocephalus 2 months after assessment of rebound. Discussion. In our experience treatment for congenital toxoplasmosis combining pyrimethamine and sulfadiazine continuously during the first year of life, according to CCTT recommendations, is regularly associated with serologic rebounds after discontinuation of therapy. Other treatment regimens previously administered, particularly those including shorter periods of administration of PYR plus SDZ alternating with spiramycin, were also associated with such IgG rebounds, but they occurred less frequently and at a moderate level.11 In their initial study with the use of a continuous course of PYR plus SDZ, McAuley et al.3 reported a high rate of rebounds with no apparent clinical significance, occurring in 58% of children after discontinuation of therapy, although initiation of therapy was delayed after birth in severely impaired children. In our study serologic rebounds were clearly not predictive of clinical recurrences inasmuch as ophthalmologic and clinical examinations were normal or stable in all 82 children evaluated at the time a rebound was assessed. Moreover children followed up for a longer period had no subsequent signs of infectious relapses, with the exception of one child who developed progressive hydrocephalus 2 months after the serologic rebound was noted. These findings are consistent with a few recent reports that failed to show any relation between clinical recurrences and IgG antibody rises during and after treatment in congenitally infected children.1, 2, 4 As demonstrated by the avidity and HS/AC tests, serologic profiles of rebound samples suggested a chronic pattern in 39 of 44 (88.6%) evaluable cases, thus consistent with a secondary immune response, despite a high rate of detection of specific IgM or IgA, similar to that found in a recent report.1 Furthermore in the 16 rebound samples with the simultaneous presence of both specific IgM and IgA antibody, 13 had both a chronic differential agglutination pattern and a high avidity index. Thus given the absence of clinical signs of relapse, such secondary responses are unlikely to be related to renewed tachyzoite proliferation after cessation of therapy, as suggested by others.1, 2 On the other hand a rather chronic serologic pattern of the rebound suggests either reexpression of Toxoplasma antigens after cessation of drug pressure or completion of some delayed specific immune responses in congenitally infected children. This latter view is consistent with frequent observations of a suppressed immune response to T. gondii in the congenitally infected newborn,12 as well as with experimental data demonstrating a delayed immune response to T. gondii after a long term tolerant state in infected newborn mice.13 Finally, irrespective of the underlying mechanism, the systematic occurrence of a serologic rebound can be considered as an event of no short term clinical prognostic significance. Consequently these results do not support reinstitution of treatment after the serologic rebound. This conclusion is also supported by the fact that in the few patients treated after 1 year of age (data not shown), the serologic rebound occurred several months after its discontinuation. However, further studies are needed to clarify the immunologic mechanisms underlying the serologic rebounds as well as their possible long term clinical significance. Acknowledgments. ODD is a recipient of an Open Society Fund research fellowship. Olgica Djurkovic-Djakovic, M.D., Ph.D. Stéphane Romand, M.D. Roberto Nobrè, M.D. Jacques Couvreur, M.D. Philippe Thulliez, M.D. Toxoplasmosis Research Laboratory; Institute for Medical Research; Belgrade, Yugoslavia (ODD) Laboratoire de la Toxoplasmose (SR, JC, PT) and Service de Pédiatrie (RN); Institut de Puériculture de Paris; Paris, France