Molecular diagnosis of toxoplasmosis in immunocompromised patients

REVIEW C URRENT OPINION Molecular diagnosis of toxoplasmosis in immunocompromised patients Florence Robert-Gangneux a,b and Sorya Belaz a,b Purpose of review Toxoplasmosis in immunocompromised patients is associated with a high mortality rate. Molecular techniques are important tools to diagnose acute disease in immunocompromised patients, but there are various methods with variable efficiency. Some of them have been validated for the diagnosis of congenital toxoplasmosis, but the impact of their use has not been evaluated in immunocompromised patients. Recent findings Toxoplasmosis is of increasing importance in non-hiv immunocompromised patients. In addition, the picture of disease shows greater severity in South America, both in immunocompetent study participants and in congenitally infected infants. These epidemiological differences could influence the sensitivity of diagnostic methods. This review analyzes recent data on molecular diagnosis and compares them with older ones, in light of progress gained in molecular techniques and of recent epidemiological findings. Most recent studies were conducted in South America and used PCR targeting the B1 gene. PCR on blood could allow diagnosing a significant proportion of patients with ocular toxoplasmosis in Brazil. Summary Quantitative PCR methods with specific probes should be used to improve sensitivity and warrant specificity. Performance of quantitative PCR targeting the repeated 529 bp sequence for the diagnosis of toxoplasmosis in immunocompromised patients needs evaluation in field studies in South America and in western countries. Keywords blood, immunocompromised patients, molecular diagnosis, PCR, toxoplasmosis INTRODUCTION Toxoplasma gondii has been early recognized as a protozoon parasite responsible for severe opportunistic infection in immunocompromised patients [1]. First cases of toxoplasmosis were described in transplant patients in the early 1980s, and in HIV-infected patients a few years later, enlightening the concept of reactivation of past infection and the mechanisms of parasite control. It appeared that primary infection and parasite reactivation in patients with previous immunity were equally life-threatening in patients with low CD4 þ T-cell counts. Initially, diagnosis was made by microscopic techniques, showing Toxoplasma tachyzoites in biopsy samples or fluids, such as broncho-alveolar lavage (BAL), cerebrospinal fluid (CSF), or blood. However, such observation is infrequent, and requires trained readers and high parasite loads. The few published cases were mainly pneumonias, as tachyzoites are more easily seen in BAL [2 &&,3]. However, immunocompromised patients soon benefited from molecular diagnosis, which was developed in the early 1990s to diagnose congenital toxoplasmosis [4,5]. Since the last 25 years, various gene targets have been described to detect Toxoplasma DNA, and the PCR techniques have greatly evolved to gain in sensitivity and to meet quality guidelines to avoid false positive results. These techniques have been mostly evaluated in the setting of congenital toxoplasmosis, a situation where it is easy to confirm the diagnosis by the serologic a Laboratoire de Parasitologie et Mycologie, Centre Hospitalier Universitaire de Rennes and b INSERM IRSET U1085, Université Rennes 1, Rennes, France Correspondence to Florence Robert-Gangneux, Service de Parasitologie, Faculté de Médecine, Université Rennes 1, 2 rue Prof. Léon Bernard 35043, Rennes cedex, France. Tel: +33 2 23 23 44 99; fax: +33 2 99 28 41 21; e-mail: Florence.robert-gangneux@univ-rennes1.fr Curr Opin Infect Dis 2016, 29:330 339 DOI:10.1097/QCO.0000000000000275 www.co-infectiousdiseases.com Volume 29 Number 4 August 2016

Molecular diagnosis of toxoplasmosis Robert-Gangneux and Belaz KEY POINTS In immunocompromised patients, rapid and reliable diagnosis of acute toxoplasmosis is needed, and DNA detection is an essential tool, whatever the sample type and the clinical presentation. Real-time qpcr methods targeting the REP-529 sequence have proved very sensitive and specific in the diagnosis of congenital toxoplasmosis, but it is not known whether they would improve the diagnosis in immunocompromised patients, as data regarding their use in these patients are lacking. PCR has a better sensitivity than serological methods in ocular fluids for the diagnosis of ocular toxoplasmosis in immunocompromised patients. Patients from South America should benefit from PCR on blood prior to aqueous humor puncture for the diagnosis of ocular toxoplasmosis. PCR on blood can be performed systematically during several months following engraftment in allo-hsct patients to guide preemptive treatment if positive. screening of the newborn or the examination of fetal biopsies in case of termination of pregnancy. By contrast, the assessment of PCR methods is far more difficult in immunocompromised patients, as deep sites of infection may be difficult to reach by biopsy, thus definite diagnosis usually relies on the association of imaging findings, molecular diagnosis, serology, and response to specific therapy. In this context, field evaluation of molecular methods is difficult, but this review will describe the state of the art of PCR methods and what may be expected from their use. Published literature, restricted to human studies, was searched in the PubMed database, using the terms pcr toxoplasmosis or pcr toxoplasma combined or not with diagnosis, toxoplasma b1, toxoplasma 529, toxoplasma rdna, TGR1E. The authors focused on the most recent articles but also included a few articles published several decades ago which were thought to be of technical or historical interest. Recommendations provided in this article are based upon critical analysis of the literature when available and upon the authors own experience in other cases. When needed, comparison of results is expressed as mean SEM. THE EVOLUTION OF MOLECULAR DIAGNOSIS The ability of PCR methods to detect low amounts of parasites in fluids or tissues is a key issue, as Toxoplasma can circulate at low concentrations, or inconstantly. In amniotic fluids from congenitally infected infants, concentrations as low as 10 tachyzoites/ml were observed in about 40% of cases in a French study [6]. Therefore, continuous efforts are undertaken to improve the sensitivity of molecular diagnosis, and immunocompromised patients benefit from advances in the diagnosis of congenital toxoplasmosis. Gene targets The p30 protein surface single-copy gene has been the first gene target described for the detection of T. gondii DNA by PCR [7]. Other sequence targets have been proposed thereafter which were given to be repeated from 30 to 300-fold in the parasite genome [8 13] (Table 1), but the true number of repetitions is still a matter of debate and could be lower than expected, or differ among parasite strains [14]. However, it has become clear that PCR methods targeting the repeated 529 bp (REP-529) sequence [11] have a better sensitivity, as compared with B1 PCR, with a mean gain of three amplification cycles [15 &,16 19]. In a multicenter study comparing the efficiency of B1 and REP-529 targets on spiked samples, it appeared that the difference in sensitivity was perceptible for parasite loads below 2 Toxoplasma per reaction tube [20]. Although some doubt has been raised on the conservatism of the REP-529 sequence, particularly in atypical genotype strains [21], those threats have not yet been confirmed [22], and this target is now used by all reference laboratories in France Table 1. Gene targets used in molecular diagnosis of toxoplasmosis Target Number of repetitions Main use First description for each application (reference) P30 Single Diagnosis 1988 [7] B1 30 35 Diagnosis 1989 [8] rdna 110 Diagnosis, genotyping 1991 [9], 1987 [12] TGR1E 100 Diagnosis, genotyping 1992 [13], 1991 [10] REP-529 200 300 Diagnosis 2000 [11] 0951-7375 Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-infectiousdiseases.com 331

Infections of the immunocompromised host (www.cnr-toxoplasma.fr), as well as in the few commercial assays available on the market. PCR methods In the industrialized countries, real-time quantitative PCR (qpcr) was introduced in 2000 [23] and has gradually replaced conventional end point PCR. It has the advantage to minimize contaminations and to amplify and detect PCR products in a single step, by the use of specific fluorescent probes. In a multicenter evaluation, it has been shown that qpcr had a similar sensitivity than nested PCR, and a better specificity [24]. However, in developing countries, nested PCR and conventional PCR (cnpcr) methods are still widely used, whether associated to hybridization, digestion with restriction enzymes, or none of both. As qpcr is usually reputed to be more sensitive than cnpcr [24 26], this may introduce considerable bias in the comparison of data between series from various continents. On the other hand, nested PCR carries a higher risk of contamination [24,26], and thus has a lower specificity, which can also bias interpretation. The lack of sensitivity of cnpcr is perceptible for parasite concentrations below 20/ml [24]. However, the sole PCR method does not account for all gaps in sensitivity, and the optimization of the whole PCR method (extraction and amplification) is a key point [20]. As underlined in a French multicenter study, the extraction method can also influence the PCR sensitivity for low parasite concentrations [27]. Although qpcr easily offers the possibility to quantify parasite loads in fluid samples, no clear correlation has been made to date between parasite concentration and severity of infection. In addition, interlaboratory comparison of parasite quantification points to huge discrepancies between results, which motivated the validation and diffusion of a calibrated Toxoplasma suspension among centers of the National Reference Center for Toxoplasmosis in France [28]. An alternative method to PCR, the loop-mediated isothermal amplification (LAMP) has been described in recent years. LAMP relies on autocycling strand displacement DNA synthesis by the Bst DNA polymerase, under isothermal conditions, usually at 638C. High specificity of amplification results from the use of four to six different primers. Amplification proceeds by displacing and releasing a single-strand DNA. It is initiated by an inner primer, followed by strand displacement by an outer primer. LAMP products can be detected by direct fluorescence by the use of fluorescent dyes such as SYBR Green, and quantified by spectrophotometry. LAMP assays targeting the SAG1 gene, REP-529, B1 gene, or SAG2 gene of T. gondii were developed for human or animal diagnosis [29 31]. Although described as a robust and specific technique in veterinary studies, LAMP has been used in only one human study in Malaysia, where it showed a better sensitivity than nested PCR to detect Toxoplasma DNA in blood from immunocompromised patients [32]. Its place among other molecular tools still needs to be clarified, but this technique could be of interest in developing countries. To date, several qpcr assays for molecular detection of Toxoplasma are commercialized, but few of them have been evaluated on clinical samples. The assay by Clonit (Milan, Italy) has been evaluated on spiked samples, but showed a higher detection threshold (i.e., lower sensitivity) than the in-house PCR method. This lack of sensitivity was confirmed on amniotic fluids, as only 50% of samples from congenitally infected fetuses were detected, whereas all tested positive using the in-house method [33]. Another qpcr assay (bio-evolution) was evaluated in a recent study which showed a good sensitivity and specificity on amniotic fluid samples [34 & ]. Although this assay is validated for parasite DNA detection in blood, BAL and CSF, no data are available on its use in field studies, thus its diagnostic performance in immunocompromised patients still needs evaluation. Other commercial PCR assays by Roche Diagnostics (Meylan, France), Elitech (Puteaux, France), as well as a LAMP commercial assay by Diasorin (Antony, France) are currently under evaluation by the French National Reference Center for Toxoplasmosis, and results should be released soon. Another important point to be underlined is the diversity of the preanalytical treatments applied to samples prior to DNA extraction, which may influence the sensitivity of the PCR assay in clinical studies. The centrifugation of liquids (aqueous humor, CSF, and BAL) and DNA extraction from the pellet must be the rule to improve sensitivity. In addition, the results of PCR on blood may greatly vary according to procedures, that is, prior isolation of leukocyte cells from 5 or 10 ml of whole blood, or direct DNA extraction of whole blood. In a murine model of disseminated toxoplasmosis, Brenier- Pinchart et al. [35 && ] nicely showed that parasite loads detected in buffy coats obtained from 1200 ml of blood were higher than those detected from 200 ml. Additionally, buffy coat isolated from 1200 ml of blood yielded better results than 200 mlof whole blood, to detect circulating Toxoplasma. This observation can be explained by the fact that T. gondii replicates in leukocyte cells, and whole blood is rich in PCR inhibitors, thus can decrease the sensitivity of parasite DNA detection, as shown in a study comparing spiked samples of amniotic fluid 332 www.co-infectiousdiseases.com Volume 29 Number 4 August 2016

Molecular diagnosis of toxoplasmosis Robert-Gangneux and Belaz and blood [36]. These interesting data pave the way for future recommendations regarding molecular diagnosis of toxoplasmosis in immunocompromised patients. VALUE OF MOLECULAR DIAGNOSIS ACCORDING TO THE CLINICAL SETTING The performance of molecular diagnosis has been mostly evaluated in cerebral toxoplasmosis and ocular toxoplasmosis, two clinical settings where final diagnosis can be confirmed by clinical response, at least partial, to specific therapy. Cerebral toxoplasmosis Most studies evaluating molecular diagnosis of toxoplasmosis were conducted in HIV-infected patients in Europe during the 1990s [37 46]. Consequently, all accumulated data are based on cnpcr techniques, and on small series of cerebral toxoplasmosis cases (mean 18 2 patients). These studies reported a sensitivity of PCR in CSF and blood samples ranging from 17 to 65% (mean 42 5.7%), and from 13 to 69% (mean 38.4 8.8), respectively (Table 2). These techniques have been adopted by developing countries, which now provide data about diagnostic performances of molecular tools [47 53 & ]. Much attention is paid to these results, particularly from Southern America, where the pattern of infection is different [54,55], because of the predominance of atypical and type I parasite strains, displaying higher virulence [56]. In this context, it might be expected that virulent strains, which were shown to produce higher parasite loads than avirulent genotype II strains [57], could be detected more easily in circulating fluids. However, most recent data do not support this hypothesis, as they show similar or slightly higher levels of sensitivity, ranging from 35 to 72% (mean 55.6 4.7), and from 15 to 80% (mean 41 39), in CSF and blood, respectively, using comparable methods [47 53 & ] (Table 2). In fact, only two studies are available regarding blood samples; they reported very contrasting results which explain the huge SEM. Studies reporting data obtained with qpcr methods are still scarce, and do not provide significantly different results than those using nested PCR, excepting a better specificity (100%) [48,49]. As expected, qpcr targeting the REP-529 sequence yielded a better sensitivity (69%) [48] than that targeting the B1 gene (35%) [49]. The sensitivity of molecular diagnosis on CSF samples was reported to be higher when cellularity exceeds 4 cells/ml (P < 0.05) or for patients with at least four brain lesions seen on MRI (P < 0.05) [49]. A high specificity of a positive PCR in CSF suggests that the parasite circulates in this compartment only in case of reactivation, thus a positive PCR must be taken into account, even in case of low parasite counts (i.e., high cerebral toxoplasmosis value). The time between CSF collection and beginning of specific treatment was pointed out to influence the sensitivity in several studies. For Anselmo et al. [53 & ] 13 of 15 and zero of 14 patients had a positive CSF, when puncture was performed less than 1 week or more than 1 week of treatment, respectively. Such a trend had been previously reported in a study, where 11 of 11 and seven of 24 patients sampled less than 1 week and more than 1 week of treatment, respectively, were diagnosed with a positive PCR [40]. Therefore, PCR can be useful to monitor treatment efficacy. Ocular toxoplasmosis Diagnosis of ocular toxoplasmosis mainly relies on ophthalmological examination. However, immunocompromised patients may present with atypical ocular fundus, which can mislead the ophthalmologist, as seen in clinical studies scoring the pretest diagnosis by physicians and the final diagnosis retained [58,59 & ]. Biological diagnosis of ocular toxoplasmosis mostly relies on immunologic and molecular methods on ocular fluids (aqueous humor or vitreous fluid), but Toxoplasma PCR on peripheral blood has also been proposed as a noninvasive diagnostic tool. Immunological diagnosis consists of the comparison of antibody synthesis in two compartments, that is, eye and blood, using quantitative or qualitative approaches. The quantitative one relies on the calculation of the Goldmann Witmer coefficient, whereas the qualitative one relies on the use of western-blot to demonstrate the production of neosynthesized antibody (IgG, IgM, or IgA) in the eye compartment [60,61]. Contrasting with cerebral toxoplasmosis, a greater amount of recent data evaluating the molecular diagnosis is available in ocular toxoplasmosis, most of them using qpcr methods. However, most studies included both immunocompetent and immunocompromised patients, making it sometimes uneasy to analyze the performance of each technique according to the patient immune background. Taken together, it appears that the sensitivity of molecular techniques depends on both the immunological status of the patients and the delay between onset of ocular symptoms and sampling. Indeed, in several studies, molecular diagnosis was the most sensitive method in the first weeks after the onset of symptoms [62,63], and it performed better than immunologic methods in immunocompromised patients [64 68 && ]. As seen in Table 3, the 0951-7375 Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-infectiousdiseases.com 333

Infections of the immunocompromised host Table 2. Main studies evaluating the sensitivity of molecular techniques for the diagnosis of cerebral toxoplasmosis Number of patients Number of confirmed CT Country Samples (number) PCR Target PCR technique Sensibility n/n (%) Specificity (%) HIV status Reference Year 14 9 USA CSF B1 cnpcr þ hyb 4/9 (44.5) 100 HIVþ [37] 1992 38 13 France Blood (71) P30 cnpcr þ hyb 9/13 (69) 76.5 HIVþ [38] 1993 33 20 The Netherlands CSF (49) B1 cnpcr þ hyb 13/20 (65) 100 HIVþ [39] 1993 82 35 Italy CSF (82) B1 Nested PCR 18/35 (51) 100 HIVþ [40] 1994 52 15 Germany CSF (8) B1 cnpcr þ hyb 3/8 (37.5) 100 HIVþ [41] 1994 Blood (15) 2/15 (13) 72 13 France Blood (72) 18SrDNA cnpcr þ hyb 5/13 (38.5) 100 HIVþ [42] 1995 CSF (48) 6/13 (46) 100 88 18 Italy CSF B1 Nested PCR 6/18 (33) 100 HIVþ [43] 1996 100 19 France Blood (19) B1 cnpcr þ hyb 5/19 (26) 95 HIVþ [44] 1996 TGR1E 83 18 France CSF (33) B1 cnpcr þ hyb 1/6 (17) 86 HIVþ [45] 1996 Blood (93) TGR1E 10/17 (59) 81 59 20 Germany Blood (89) B1 cnpcr þ hyb 5/20 (25) 100 HIVþ [46] 1997 TGR1E 192 64 Brazil Blood (192) B1 cnpcr 51/64 (80) 98 HIVþ [47] 2005 51 16 Brazil CSF REP-529 qpcr 11/16 (69) 100 39 HIVþ12 HIV [48] 2009 135 85 Brazil CSF (120) B1 qpcr 30/85 (35) 100 HIVþ [49] 2010 Blood (120) 1/67 (1.5) 100 207 59 Cuba CSF (207) T1-T4 (B1) Nested PCR 30/59 (51) 98 147 HIVþ60 HIV [50] 2013 S1-AS1 (B1) Nested PCR 40/58 (69) 97 B22-B23 (B1) cnpcr 42/58 (72.5) 95 28 16 Japan CSF (16) 18SrDNA Nested PCR 8/16 (50) 100 HIVþ [51] 2013 B1 LAMP ND (lower) LAMP ND (lower) 13 13 Japan CSF (13) 18SrDNA Nested PCR 3/7 (43) NA HIVþ [52] 2014 79 27 Brazil CSF (79) B1 Nested PCR 15/27 (56) 100 78 HIVþ1 HIV [53 & ] 2014 cnpcr, conventional PCR; CSF, cerebrospinal fluid; CT, cerebral toxoplasmosis; hyb, hybridization; LAMP, loop-mediated isothermal amplification; ND, not determined; qpcr, real-time quantitative PCR. 334 www.co-infectiousdiseases.com Volume 29 Number 4 August 2016

Molecular diagnosis of toxoplasmosis Robert-Gangneux and Belaz Table 3. Main studies reporting the evaluation of molecular methods for the diagnosis of ocular toxoplasmosis in immunocompromised patients Number of patients Number of confirmed OT Country Samples PCR target PCR technique Sensitivity n/n (%) Specificity (%) Immune background a Reference 15 7 USA VF B1 cnpcr 7/7 (100) 100 OT: 3 IC, 4 non IC [69] IC 3/3 (100) 67 43 France AH B1 cnpcr 9/43 (21) 100 All: 14 IC [64] IC 8/9 (89) OT: 9 IC, 25 non IC 22 16 USA 5 AH, 13 VF, 3 B1 qpcr All: 7/16 (44) ND All: 17 IC (6 HIVþ), 5 non IC [65] retinal biopsies Argentina IC: 5/7 (71) Puerto Rico OT: 7 IC 56 11 The Netherlands 27 AH B1 cnpcr 4/10 (40) 100 All: 56 IC (27 HIVþ) OT: 11 IC (1 HIVþ) [66] 110 34 France 102 AH, 18 VF B1 qpcr 9/34 (26.5) 100 OT: 4 HIVþ; 30 non IC [70] REP-529 qpcr 13/34 (38) 100 IC: 4/13 (31) 54 40 France 51 AH, 3 VF REP-529 qpcr 22/40 (55) 100 OT: 11 IC (6 HIVþ); 29 non IC [66] IC : 8/11 (73) 93 133 18 USA 105 AH þ 38 VF B1 qpcr 12/18 (67) 100 79 IC (61 HIVþ), 54 non IC [58] 132 61 France 133 AH, 7 VF? qpcr 15/55 (27) 100 All: 16 IC [63] IC: 4/5 (80) 100 OT: 5 IC 159 16 South Africa 150 AH þ VF B1 Nested PCR 11/16 (69) 100 All: 142 HIVþ [59 & ] 405 114 France Blood (200 ml) B1 and REP-529 qpcr ND/ND (4) 99.6 All: 157 non HIV, 57 IC (31 HIVþ) [68 && ] Blood (2 ml) ND/ND (25) 89.5 OT: 66 non IC, 18 IC (15HIVþ) 368 AH, 82 VF ND/ND (36) 100 IC: ND/ND (61.5) AH, aqueous humor; cnpcr, conventional PCR; IC, immunocompromised; ND, not determined; OT, ocular toxoplasmosis; qpcr, quantitative real-time PCR; VF, vitreous fluid. a The number of immunocompromised patients in the group of confirmed OT is given, and when available, the number of immunocompromised patients in the whole population study ( all ). 0951-7375 Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-infectiousdiseases.com 335

Infections of the immunocompromised host sensitivity in immunocompromised patients ranged from 61.5 to 100% in 8/10 studies, whereas it ranged from 21 to 100% in the whole study population, and from 25 to 53% in studies including only immunocompetent patients [62,69 73]. In addition, the size of the lesions was another factor which was reported to influence the positivity of PCR in aqueous humor, as lesions with disk area, more than three, were more likely associated with positive PCR results [64,67,74,75]. Finally, in spite of the limitation because of small series, it appears that PCR on vitreous fluid is more often positive than on aqueous humor [59 &,69]. The performance of PCR on aqueous humor is usually better than on peripheral blood, whatever the immunological background of the patients. Bourdin et al. [68 && ] showed in a large series that molecular diagnosis on peripheral blood is less sensitive than in aqueous humor, especially in immunocompetent patients (1 vs. 24%, respectively). They also confirmed that molecular diagnosis can be improved by increasing the volume of whole blood extracted. Overall, combining the results of B1 and REP-529 qpcr allowed them to confirm the diagnosis of ocular toxoplasmosis in 61.5%, and only 24% of immunocompromised patients and of immunocompetent patients, respectively (P < 0.001). Interestingly, the sensitivity of Goldmann Witmer coefficient (72%) was higher than that of western-blot (48%). Combining the three methods allowed them to reach a sensitivity of 89%, as seen by others [67]. As for congenital and neurotoxoplasmosis, the targeted sequence REP-529 is more sensitive than B1 [68 &&,70] and qpcr methods seem the most sensitive and specific PCR techniques. However, it must be kept in mind that the best results are obtained by combining molecular and immunological methods on ocular samples. Despite the constraint regarding the small volume available, this is made possible by the completion of the PCR on the pellets and western-blot on the supernatant, as this latter technique is sample-sparing. The role of the parasite genotype on the sensitivity of PCR in ocular toxoplasmosis is difficult to assess, as very few studies are available from South America. Some Brazilian authors recently reported the positivity of PCR result in blood samples from four of five patients with ocular toxoplasmosis [76]; all were immunocompetent, thus this rate seems unusually high. Mattos et al. [77] previously described a sensitivity of 41%, in Brazilian patients of unknown HIV status. If such results are confirmed, the diagnosis of ocular toxoplasmosis in Brazilian immunocompromised patients might rely at first on blood PCR, thus avoiding invasive procedures in a significant amount of cases. Pulmonary toxoplasmosis and disseminated toxoplasmosis The value of molecular diagnosis in these settings is difficult to establish, as the evolution of infection in immunocompromised patients may be rapidly lethal, and the diagnosis may remain undone, or unrecognized because of multiple opportunistic infections. Case reports are regularly published, but few series meet the required criteria to evaluate diagnostic efficiency, or even disease prevalence in immunocompromised patients. A recent study reported Toxoplasma screening by qpcr and direct examination in 336 BAL fluids from immunocompromised patients over a 2-year period, and found two patients with positive microscopic examination and PCR [2 && ]. Although the authors did not find a gain in sensitivity in favor of PCR, it must be underlined that microscopic diagnosis may be difficult, and requires specific skills, thus the diagnosis of pulmonary toxoplasmosis cannot rely only on direct examination. This incidence seems low, but is in agreement with a recent study recording the number of patients diagnosed with a positive PCR result in 15 French University Hospitals [78 & ]. INTEREST OF MOLECULAR TOOLS TO MONITOR TREATMENT qpcr can be of interest for monitoring treatment efficacy, by showing decreasing parasite loads in samples (blood, BAL, and CSF), as shown in transplant patients [79,80]. Circulating parasites should become undetectable within 3 weeks of treatment. On the other hand, several surveys have reported the use of PCR screening to monitor allo- HSCT patients in the months following engraftment, by the completion of weekly PCR on blood samples [81 83]. Although they found circulating parasite DNA in patients without Toxoplasma disease in 6 13% of patients, the treatment of such patients probably allowed avoiding full-blown disease. In a recent study, the survival of allo-hsct patients with circulating Toxoplasma DNA was compared among 11 French University Hospitals, and it appeared that better survival was observed in centers who had implemented a weekly or 2-monthly PCR screening, than in those who had not [78 & ]. Whether this screening is cost-effective and should be systematically included in the patient workup remains to be determined. The prevention of toxoplasmosis has 336 www.co-infectiousdiseases.com Volume 29 Number 4 August 2016

Molecular diagnosis of toxoplasmosis Robert-Gangneux and Belaz been addressed in a recent review examining in detail recommendations for HSCT patients [84 && ]. CONCLUSION Despite a long history of molecular diagnosis of toxoplasmosis and numerous evaluations, a new sheet remains to be written, as the picture of the disease in developing countries, in particular South America, has been shown to be different from Europe and Northern America. Therefore, it might have diagnostic implications and modify the use of diagnostic tools, in particular in ocular toxoplasmosis. Regarding cerebral toxoplasmosis, large series using qpcr methods are awaited, as most of our data of sensitivity are based on old methods which were subjected to technical pitfalls. New studies should precise the contribution of qpcr, and its use could bring new insights into the comprehension of the pathophysiology of the disease, which may differ among continents [85]. Acknowledgements None. Financial support and sponsorship None. Conflicts of interest There are no conflicts of interest. REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest && of outstanding interest 1. Weiss LM, Dubey JP. Toxoplasmosis: a history of clinical observations. Int J Parasitol 2009; 39:895 901. 2. Desoubeaux G, Cabanne E, Franck-Martel C, et al. Pulmonary toxoplasmosis && in immunocompromised patients with interstitial pneumonia: a single-centre prospective study assessing PCR-based diagnosis. J Clin Pathol 2016. [Epub ahead of print] The only prospective study evaluating the diagnostic methods and incidence of pulmonary toxoplasmosis in immunocompromised patients. 3. Wheeler RR, Bardales RH, North PE, et al. Toxoplasma pneumonia: cytologic diagnosis by bronchoalveolar lavage. Diagn Cytopathol 1994; 11:52 55. 4. Dupouy-Camet J, Bougnoux ME, Lavareda de Souza S, et al. Comparative value of polymerase chain reaction and conventional biological tests for the prenatal diagnosis of congenital toxoplasmosis. Ann Biol Clin (Paris) 1992; 50:315 319. 5. Grover CM, Thulliez P, Remington JS, Boothroyd JC. Rapid prenatal diagnosis of congenital Toxoplasma infection by using polymerase chain reaction and amniotic fluid. J Clin Microbiol 1990; 28:2297 2301. 6. Romand S, Chosson M, Franck J, et al. Usefulness of quantitative polymerase chain reaction in amniotic fluid as early prognostic marker of fetal infection with Toxoplasma gondii. Am J Obstet Gynecol 2004; 190:797 802. 7. Burg JL, Perelman D, Kasper LH, et al. Molecular analysis of the gene encoding the major surface antigen of Toxoplasma gondii. J Immunol 1988; 141:3584 3591. 8. Burg JL, Grover CM, Pouletty P, Boothroyd JC. Direct and sensitive detection of a pathogenic protozoan, Toxoplasma gondii, by polymerase chain reaction. J Clin Microbiol 1989; 27:1787 1792. 9. Cazenave J, Cheyrou A, Blouin P, et al. Use of polymerase chain reaction to detect Toxoplasma. J Clin Pathol 1991; 44:1037. 10. Cristina N, Liaud MF, Santoro F, et al. A family of repeated DNA sequences in Toxoplasma gondii: cloning, sequence analysis, and use in strain characterization. Exp Parasitol 1991; 73:73 81. 11. Homan WL, Vercammen M, De Braekeleer J, Verschueren H. Identification of a 200- to 300-fold repetitive 529 bp DNA fragment in Toxoplasma gondii, and its use for diagnostic and quantitative PCR. Int J Parasitol 2000; 30:69 75. 12. Johnson AM, Murray PJ, Illana S, Baverstock PJ. Rapid nucleotide sequence analysis of the small subunit ribosomal RNA of Toxoplasma gondii: evolutionary implications for the Apicomplexa. Mol Biochem Parasitol 1987; 25:239 246. 13. Cristina N, Derouin F, Pelloux H, et al. [Detection of Toxoplasma gondii by Polymerase Chain Reaction (PCR) technique in AIDS infected patients using the repetitive sequence TGR1E]. Pathol Biol (Paris) 1992; 40:52 55. 14. Costa JM, Bretagne S. Variation of B1 gene and AF146527 repeat element copy numbers according to Toxoplasma gondii strains assessed using realtime quantitative PCR. J Clin Microbiol 2012; 50:1452 1454. 15. & Belaz S, Gangneux JP, Dupretz P, et al. A 10-year retrospective comparison of two target sequences, REP-529 and B1, for Toxoplasma gondii detection by quantitative PCR. J Clin Microbiol 2015; 53:1294 1300. The study compared the performances of qpcrs targeting B1 and REP-529 sequences in routine use, and showed a significant lack of sensitivity of the B1 qpcr, particularly in samples rich in PCR inhibitors (placenta), which could impact detection of parasite DNA in blood of immunocompromised patients. 16. Cassaing S, Bessieres MH, Berry A, et al. Comparison between two amplification sets for molecular diagnosis of toxoplasmosis by real-time PCR. J Clin Microbiol 2006; 44:720 724. 17. Kasper DC, Sadeghi K, Prusa AR, et al. Quantitative real-time polymerase chain reaction for the accurate detection of Toxoplasma gondii in amniotic fluid. Diagn Microbiol Infect Dis 2009; 63:10 15. 18. Delhaes L, Yera H, Ache S, et al. Contribution of molecular diagnosis to congenital toxoplasmosis. Diagn Microbiol Infect Dis 2013; 76:244 247. 19. Reischl U, Bretagne S, Kruger D, et al. Comparison of two DNA targets for the diagnosis of Toxoplasmosis by real-time PCR using fluorescence resonance energy transfer hybridization probes. BMC Infect Dis 2003; 3:7. 20. Sterkers Y, Varlet-Marie E, Cassaing S, et al. Multicentric comparative analytical performancestudyformolecular detection oflow amountsoftoxoplasma gondii from simulated specimens. J Clin Microbiol 2010; 48:3216 3222. 21. Okay TS, Yamamoto L, Oliveira LC, et al. Significant performance variation among PCR systems in diagnosing congenital toxoplasmosis in Sao Paulo, Brazil: analysis of 467 amniotic fluid samples. Clinics (Sao Paulo) 2009; 64:171 176. 22. Torres E, Rivera R, Cardona N, et al. Evaluation of IgG antitoxoplasma avidity and polymerase chain reaction in the postnatal diagnosis of congenital toxoplasmosis. Pediatr Infect Dis J 2013; 32:693 695. 23. Costa JM, Pautas C, Ernault P, et al. Real-time PCR for diagnosis and followup of Toxoplasma reactivation after allogeneic stem cell transplantation using fluorescence resonance energy transfer hybridization probes. J Clin Microbiol 2000; 38:2929 2932. 24. Kaiser K, Van Loon AM, Pelloux H, et al. Multicenter proficiency study for detection of Toxoplasma gondii in amniotic fluid by nucleic acid amplification methods. Clin Chim Acta 2007; 375:99 103. 25. Nagy B, Ban Z, Beke A, et al. Detection of Toxoplasma gondii from amniotic fluid, a comparison of four different molecular biological methods. Clin Chim Acta 2006; 368:131 137. 26. Teixeira LE, Kanunfre KA, Shimokawa PT, et al. The performance of four molecular methods for the laboratory diagnosis of congenital toxoplasmosis in amniotic fluid samples. Rev Soc Bras Med Trop 2013; 46:584 588. 27. Yera H, Filisetti D, Bastien P, et al. Multicenter comparative evaluation of five commercial methods for toxoplasma DNA extraction from amniotic fluid. J Clin Microbiol 2009; 47:3881 3886. 28. Varlet-Marie E, Sterkers Y, Brenier-Pinchart MP, et al. Characterization and multicentric validation of a common standard for Toxoplasma gondii detection using nucleic acid amplification assays. J Clin Microbiol 2014; 52:3952 3959. 29. Krasteva D, Toubiana M, Hartati S, et al. Development of loop-mediated isothermal amplification (LAMP) as a diagnostic tool of toxoplasmosis. Vet Parasitol 2009; 162:327 331. 30. Zhang H, Thekisoe OM, Aboge GO, et al. Toxoplasma gondii: sensitive and rapid detection of infection by loop-mediated isothermal amplification (LAMP) method. Exp Parasitol 2009; 122:47 50. 31. Kong QM, Lu SH, Tong QB, et al. Loop-mediated isothermal amplification (LAMP): early detection of Toxoplasma gondii infection in mice. Parasit Vectors 2012; 5:2. 32. Lau YL, Meganathan P, Sonaimuthu P, et al. Specific, sensitive, and rapid diagnosis of active toxoplasmosis by a loop-mediated isothermal amplification method using blood samples from patients. J Clin Microbiol 2010; 48:3698 3702. 33. Morelle C, Varlet-Marie E, Brenier-Pinchart MP, et al. Comparative assessment of a commercial kit and two laboratory-developed PCR assays for molecular diagnosis of congenital toxoplasmosis. J Clin Microbiol 2012; 50:3977 3982. 34. & Filisetti D, Sterkers Y, Brenier-Pinchart MP, et al. Multicentric comparative assessment of the bio-evolution Toxoplasma gondii detection kit with eight laboratory-developed PCR assays for molecular diagnosis of congenital toxoplasmosis. J Clin Microbiol 2015; 53:29 34. The first evaluation of a commercial assay on a significant amount of clinical samples. However, the performance of the assay in samples from immunocompromised patients (particularly blood samples) remains to be assessed. 0951-7375 Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-infectiousdiseases.com 337

Infections of the immunocompromised host 35. Brenier-Pinchart MP, Capderou E, Bertini RL, et al. Molecular diagnosis of && toxoplasmosis: value of the buffy coat for the detection of circulating Toxoplasma gondii. Diagn Microbiol Infect Dis 2015; 82:289 291. The work nicely shows that blood PCR has a greater sensitivity when leucocyte pellets are separated from whole blood. The murine model is tricky, but is the only way to compare different sample proceedings. 36. Chabbert E, Lachaud L, Crobu L, Bastien P. Comparison of two widely used PCR primer systems for detection of toxoplasma in amniotic fluid, blood, and tissues. J Clin Microbiol 2004; 42:1719 1722. 37. Parmley SF, Goebel FD, Remington JS. Detection of Toxoplasma gondii in cerebrospinal fluid from AIDS patients by polymerase chain reaction. J Clin Microbiol 1992; 30:3000 3002. 38. Dupouy-Camet J, de Souza SL, Maslo C, et al. Detection of Toxoplasma gondii in venous blood from AIDS patients by polymerase chain reaction. J Clin Microbiol 1993; 31:1866 1869. 39. Schoondermark-van de Ven E, Galama J, Kraaijeveld C, et al. Value of the polymerase chain reaction for the detection of Toxoplasma gondii in cerebrospinal fluid from patients with AIDS. Clin Infect Dis 1993; 16:661 666. 40. Novati R, Castagna A, Morsica G, et al. Polymerase chain reaction for Toxoplasma gondii DNA in the cerebrospinal fluid of AIDS patients with focal brain lesions. AIDS 1994; 8:1691 1694. 41. Khalifa Ke-S. Roth A, Roth B, et al. Value of PCR for evaluating occurrence of parasitemia in immunocompromised patients with cerebral and extracerebral toxoplasmosis. J Clin Microbiol 1994; 32:2813 2819. 42. Dupon M, Cazenave J, Pellegrin JL, et al. Detection of Toxoplasma gondii by PCR and tissue culture in cerebrospinal fluid and blood of human immunodeficiency virus-seropositive patients. J Clin Microbiol 1995; 33:2421 2426. 43. Cingolani A, De Luca A, Ammassari A, et al. PCR detection of Toxoplasma gondii DNA in CSF for the differential diagnosis of AIDS-related focal brain lesions. J Med Microbiol 1996; 45:472 476. 44. Lamoril J, Molina JM, de Gouvello A, et al. Detection by PCR of Toxoplasma gondii in blood in the diagnosis of cerebral toxoplasmosis in patients with AIDS. J Clin Pathol 1996; 49:89 92. 45. Robert F, Ouatas T, Blanche P, et al. [Retrospective evaluation of the detection of Toxoplasma gondii by polymerase chain reaction in AIDS patients]. Presse Med 1996; 25:541 545. 46. Franzen C, Altfeld M, Hegener P, et al. Limited value of PCR for detection of Toxoplasma gondii in blood from human immunodeficiency virus-infected patients. J Clin Microbiol 1997; 35:2639 2641. 47. Colombo FA, Vidal JE, Penalva de Oliveira AC, et al. Diagnosis of cerebral toxoplasmosis in AIDS patients in Brazil: importance of molecular and immunological methods using peripheral blood samples. J Clin Microbiol 2005; 43:5044 5047. 48. Nogui FL, Mattas S, Turcato Junior G, Lewi DS. Neurotoxoplasmosis diagnosis for HIV-1 patients by real-time PCR of cerebrospinal fluid. Braz J Infect Dis 2009; 13:18 23. 49. Correia CC, Melo HR, Costa VM. Influence of neurotoxoplasmosis characteristics on real-time PCR sensitivity among AIDS patients in Brazil. Trans R Soc Trop Med Hyg 2010; 104:24 28. 50. Alfonso Y, Fraga J, Cox R, et al. Conventional polymerase chain reaction for the diagnosisof neurotoxoplasmosis: comparison of three sets of primers for theb1 gene using CSF samples. Diagn Microbiol Infect Dis 2013; 75:150 154. 51. Mikita K, Maeda T, Ono T, et al. The utility of cerebrospinal fluid for the molecular diagnosis of toxoplasmic encephalitis. Diagn Microbiol Infect Dis 2013; 75:155 159. 52. Sakamoto N, Maeda T, Mikita K, et al. Clinical presentation and diagnosis of toxoplasmic encephalitis in Japan. Parasitol Int 2014; 63:701 704. 53. Anselmo LM, Vilar FC, Lima JE, et al. Usefulness and limitations of polymerase & chain reaction in the etiologic diagnosis of neurotoxoplasmosis in immunocompromised patients. J Neurol Sci 2014; 346:231 234. The work included a homogeneous series of immunocompromised patients, and points out the necessity of early samples of CSF after the treatment beginning for the diagnosis of cerebral toxoplasmosis. 54. Carme B, Bissuel F, Ajzenberg D, et al. Severe acquired toxoplasmosis in immunocompetent adult patients in French Guiana. J Clin Microbiol 2002; 40:4037 4044. 55. Gilbert RE, Freeman K, Lago EG, et al. Ocular sequelae of congenital toxoplasmosis in Brazil compared with Europe. PLoS Negl Trop Dis 2008; 2:e277. 56. Darde ML. Toxoplasma gondii, new genotypes and virulence. Parasite 2008; 15:366 371. 57. de-la-torre A, Sauer A, Pfaff AW, et al. Severe South American ocular toxoplasmosis is associated with decreased Ifn-gamma/Il-17a and increased Il-6/Il-13 intraocular levels. PLoS Negl Trop Dis 2013; 7:e2541. 58. Harper TW, Miller D, Schiffman JC, Davis JL. Polymerase chain reaction analysis of aqueous and vitreous specimens in the diagnosis of posterior segment infectious uveitis. Am J Ophthalmol 2009; 147:140 147. 59. & Scheepers MA, Lecuona KA, Rogers G, et al. The value of routine polymerase chain reaction analysis of intraocular fluid specimens in the diagnosis of infectious posterior uveitis. ScientificWorldJournal 2013; 2013:545149. The authors included a large series of HIV-infected patients to evaluate PCR on ocular fluids, and provide the first molecular data from South Africa. Interestingly, they rate the probable diagnosis before biological testing, and final diagnosis in view of clinical evolution and biological results. 60. Robert-Gangneux F, Binisti P, Antonetti D, et al. Usefulness of immunoblotting and Goldmann-Witmer coefficient for biological diagnosis of toxoplasmic retinochoroiditis. Eur J Clin Microbiol Infect Dis 2004; 23: 34 38. 61. Garweg JG, Garweg SD, Flueckiger F, et al. Aqueous humor and serum immunoblotting for immunoglobulin types G, A, M, and E in cases of human ocular toxoplasmosis. J Clin Microbiol 2004; 42:4593 4598. 62. De Groot-Mijnes JD, Rothova A, Van Loon AM, et al. Polymerase chain reaction and Goldmann-Witmer coefficient analysis are complimentary for the diagnosis of infectious uveitis. Am J Ophthalmol 2006; 141:313 318. 63. Errera MH, Goldschmidt P, Batellier L, et al. Real-time polymerase chain reaction and intraocular antibody production for the diagnosis of viral versus toxoplasmic infectious posterior uveitis. Graefes Arch Clin Exp Ophthalmol 2011; 249:1837 1846. 64. Fardeau C, Romand S, Rao NA, et al. Diagnosis of toxoplasmic retinochoroiditis with atypical clinical features. Am J Ophthalmol 2002; 134:196 203. 65. Moshfeghi DM, Dodds EM, Couto CA, et al. Diagnostic approaches to severe, atypical toxoplasmosis mimicking acute retinal necrosis. Ophthalmology 2004; 111:716 725. 66. Westeneng AC, Rothova A, de Boer JH, de Groot-Mijnes JD. Infectious uveitis in immunocompromised patients and the diagnostic value of polymerase chain reaction and Goldmann-Witmer coefficient in aqueous analysis. Am J Ophthalmol 2007; 144:781 785. 67. Talabani H, Asseraf M, Yera H, et al. Contributions of immunoblotting, real-time PCR, and the Goldmann-Witmer coefficient to diagnosis of atypical toxoplasmic retinochoroiditis. J Clin Microbiol 2009; 47:2131 2135. 68. && Bourdin C, Busse A, Kouamou E, et al. PCR-based detection of Toxoplasma gondii DNA in blood and ocular samples for diagnosis of ocular toxoplasmosis. J Clin Microbiol 2014; 52:3987 3991. The article describes a large series of ocular toxoplasmosis in immunocompromised and nonimmunocompromised patients, and compares different sample types and diagnostic methods. 69. Montoya JG, Parmley S, Liesenfeld O, et al. Use of the polymerase chain reaction for diagnosis of ocular toxoplasmosis. Ophthalmology 1999; 106:1554 1563. 70. Fekkar A, Bodaghi B, Touafek F, et al. Comparison of immunoblotting, calculation of the Goldmann-Witmer coefficient, and real-time PCR using aqueous humor samples for diagnosis of ocular toxoplasmosis. J Clin Microbiol 2008; 46:1965 1967. 71. Bou G, Figueroa MS, Marti-Belda P, et al. Value of PCR for detection of Toxoplasma gondii in aqueous humor and blood samples from immunocompetent patients with ocular toxoplasmosis. J Clin Microbiol 1999; 37:3465 3468. 72. Figueroa MS, Bou G, Marti-Belda P, et al. Diagnostic value of polymerase chain reaction in blood and aqueous humor in immunocompetent patients with ocular toxoplasmosis. Retina 2000; 20:614 619. 73. Lee SE, Hong SH, Lee SH, et al. Detection of ocular Toxoplasma gondii infection in chronic irregular recurrent uveitis by PCR. Korean J Parasitol 2012; 50:229 231. 74. Sugita S, Ogawa M, Inoue S, et al. Diagnosis of ocular toxoplasmosis by two polymerase chain reaction (PCR) examinations: qualitative multiplex and quantitative real-time. Jpn J Ophthalmol 2011; 55:495 501. 75. Labalette P, Delhaes L, Margaron F, et al. Ocular toxoplasmosis after the fifth decade. Am J Ophthalmol 2002; 133:506 515. 76. Previato M, Frederico FB, Murata FH, et al. A Brazilian report using serological and molecular diagnosis to monitoring acute ocular toxoplasmosis. BMC Res Notes 2015; 8:746. 77. Mattos CC, Meira CS, Ferreira AI, et al. Contribution of laboratory methods in diagnosing clinically suspected ocular toxoplasmosis in Brazilian patients. Diagn Microbiol Infect Dis 2011; 70:362 366. 78. & Robert-Gangneux F, Sterkers Y, Yera H, et al. Molecular diagnosis of toxoplasmosis in immunocompromised patients: a 3-year multicenter retrospective study. J Clin Microbiol 2015; 53:1677 1684. The study reports the practices of 15 French University Hospitals for the molecular diagnosis of toxoplasmosis. The crude mortality of allo-hsct patients diagnosed with a positive blood PCR was compared among centers according to systematic PCR screening practices. 79. Patrat-Delon S, Gangneux JP, Lavoue S, et al. Correlation of parasite load determined by quantitative PCR to clinical outcome in a heart transplant patient with disseminated toxoplasmosis. J Clin Microbiol 2010; 48:2541 2545. 80. Menotti J, Vilela G, Romand S, et al. Comparison of PCR-enzyme-linked immunosorbent assay and real-time PCR assay for diagnosis of an unusual case of cerebral toxoplasmosis in a stem cell transplant recipient. J Clin Microbiol 2003; 41:5313 5316. 81. Martino R, Bretagne S, Einsele H, et al. Early detection of Toxoplasma infection by molecular monitoring of Toxoplasma gondii in peripheral blood samples after allogeneic stem cell transplantation. Clin Infect Dis 2005; 40:67 78. 338 www.co-infectiousdiseases.com Volume 29 Number 4 August 2016

Molecular diagnosis of toxoplasmosis Robert-Gangneux and Belaz 82. Meers S, Lagrou K, Theunissen K, et al. Myeloablative conditioning predisposes patients for Toxoplasma gondii reactivation after allogeneic stem cell transplantation. Clin Infect Dis 2010; 50:1127 1134. 83. Fricker-Hidalgo H, Bulabois CE, Brenier-Pinchart MP, et al. Diagnosis of toxoplasmosis after allogeneic stem cell transplantation: results of DNA detection and serological techniques. Clin Infect Dis 2009; 48:e9 e15. 84. Gajurel K, Dhakal R, Montoya JG. Toxoplasma prophylaxis in haematopoietic && cell transplant recipients: a review of the literature and recommendations. Curr Opin Infect Dis 2015; 28:283 292. An excellent review analyzing the prevention measures of toxoplasmosis in allo- HSCT patients. 85. Ma H, Wang Z, Wang C, et al. Fatal Toxoplasma gondii infection in the giant panda. Parasite 2015; 22:30 32. 0951-7375 Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-infectiousdiseases.com 339