Candida on oral cavity of pediatric individuals with ALL and its susceptibility to nystatin and amphotericin B Candida na cavidade oral de indivíduos pediátricos com LLA e sua susceptibilidade à nistatina e à anfotericina B

Objective: The aim of this study was to evaluate the prevalence of Candida colonizationon oral cavity of pediatric individuals with acute lymphocytic leukemia (ALL) and its susceptibility/resistance to nystatin and amphotericin B. Methods: This was a cross sectional study with observational, descriptive and analytic approach. Saliva was collected from40 individuals diagnosed with ALL and from40 healthy subjects, as a comparative group, matched by age and gender with ALL group. The mean age for both groups were 8 years-old. The isolation and identification of the Candidaspecies were performed using the CHROMagarCandidaTM and confirmed by polymerase chain reaction. The samples were subjected to antifungal susceptibility by microdilution assay for nystatin and amphotericin B. Salivary alterations and chemotherapy-induced oralmucositis were evaluated using modifiedOral Assessment Guide. Results: The positivity to Candida was higher inALL individuals (32.5%,13/40)than in a comparative group(2.5%, 1/40) (p<0.001). Candida albicans was the most prevalent strain (86.6%). The mucositis was directly associated with positive Candidacolonization (p=0.017) in the ALL group but not related with salivary alterations (p= 0.479). Six strains of C. albicans (54.5%), on ALL group, were resistant to nystatin

and all strains were not susceptible to amphotericin B. Conclusion: Candida colonization was associated with ALL condition and with

INTRODUCTION
The acute lymphocytic leukemia (ALL) represents 75% of all diagnosed leukemia in oncopediatric individuals and 25% of all malignant childhood illnesses [1,2]. The immunosuppression favoring related to disease, disease profile and myeloablative treatment could be associated with bacterial and fungal infections, including candidosis [2][3][4] In ALL patients, it could be raised in some part of chemotherapic treatment and be associated with immunosuppresionfavoring nonalbicans species colonization [4]. Other concern could be recurrent infections caused by resistant Candida species [5]. Even with the use of drugs such as nystatin and amphotericin B, it is highly desirable to select better treatment options defined by sensibility tests [6,7]. It is important because candidemia could be a critical problem in ALL patients with bad outcome and worsening of overall prognosis [3].
Highlighting the importance of knowing the profile of Candida colonization in ALL pediatric individuals and its resistance to usual treatment, the purpose of this study was to evaluate the prevalence of Candida colonizationon oral cavity of pediatric individuals with acute lymphocytic leukemia (ALL) and its susceptibility/resistance to nystatin and amphotericin B.

METHODS
The study was cross-sectional, with dual observational and descriptive characteristics approved by ethic committee of Health Centre of Federal University of Paraiba. For this study, we studied two groups. The first was ALL group, formed by 40 individuals, enrolled by census approach during 2014 and 2015. Individuals with ages ranged from 1 to 19 years old were diagnosed with ALL, at a reference oncopediatric unit located in João Pessoa, Paraíba, Brazil. The inclusion criteria were determined as ALL diagnosis and being submitted to chemotherapy or with recent history of chemotherapic treatment.
Data about hemogram with differential count, chemotherapy employed, antimicrobial drugs used, oral mucositis, xerostomia and DMFT/dmft (Decayed/Missing/ Filled Teeth) were collected from medical and dental records. Last hematological exams performed close to the salivary collection date, during the same week, were considered.
The indexes of mucositis and xerostomia were obtained from OAG (Oral Assessment Guide), modified by Cheng et al. [8] and made by calibrated examiners being dichotomized in presence/absence. Values above 9 points indicated presence of mucositis and scores 2 and 3 on saliva item, salivary alterations. The table 1 demonstrated this modified guide used in this research.
Clinical candidosis was evaluated by calibrated examiners seeking for symptomatic white removable plaques (pseudomembranous form) over a bleeding or erythematous layer or reddish areas on mucosa associated with pain/sensibility complains (erythematous form). In cases of clinical doubt between erythematous candidosis and oral mucositis, salivary smears could be used to confirm or dismiss Candida presence in culture.
Of 40 ALL, 19 were in induction therapy, 3 in reinduction, 3 in consolidation and 15 in maintenance phases. These phases were reclassified in induction (induction/reinduction/consolidation) and maintenance phases for statistical analysis due to similarities among them. The standard chemotherapeutical protocol was BMF (Berlin-Frankfurt-Munich), 2002 [9].
A comparative group was formed by 40 individuals matched by age and gender with the ALL group. This group was formed by healthy individuals, without systemic alterations and other diseases/drugs associated to bacterial/ fungal infections or salivary complaints. Clinical candidosis was evaluated by calibrated examiners following the same criteria for ALL group. The Kappa results for the calibration of oral diagnosis were 0.877 to 1.0 (individually) and 0.844 between the evaluators [10].

Isolation, Culture and Identification of Candida species
Salivary collection was made with a sterile swab on the buccal mucosa (right and left) and tongue, in order After the collection transportation, the swabs were immediately seeded into plates with CHROMagarCandida TM (BD TM CHROMAGAR TM Candida, France). After seeding, plates were placed on incubator at 370C for 48 hours. The mean of CFU/mL was obtained by the counting of positive growth of Candida. The CFUs were presumptively identified according to color and texture patterns. These colonies were stored at -2 °C into Sabouraud Dextrose Broth (SDB) (KASVI ® , Curitiba, Brazil) supplemented by glycerol (40% v/v), into cryotubes (2 mL).
Thermocycling phase consisted of an initial denaturation cycle at 95°C for 2 minutes followed by 40 cycles of denaturation at 95°C for 15 seconds, primer annealing at 58 °C for 30 seconds, and primer extension at 72°C for 30 seconds. A final primer extension at 72°C for 2 minutes was performed, followed by cooling to 4°C. The samples ran on agarose gel prepared with 2% Trisbase buffer + Boric acid + EDTA (TBE) diluted in water with adjusted pH. The run was carried out for 50 minutes at 80W.

Antifungal susceptibility
The samples of each identified species on ALL group were activated in Falcon sterile tubes containing 5mL of SBS and cultivated at 37°C for 48 hours. The concentration of Candida employed was 5 x 10 [6] UFC/ mL [12]. Candida inoculums were diluted to a final concentration of 5.0X10 [3] UFC/mL and placed into 96 well microplates (Global Plast®, Monte Alto-SP, Brazil). Amphotericin B (Sigma Aldrich ® , Saint Louis, USA) (320μg/ mL) was diluted in filtered DMSO and nystatin (Dilecta, João Pessoa-PB, Brazil) (256μg/mL) on distilled water to achieve the initial concentrations.
In each well microplate, 100μL of Sabouraud Dextrose Broth (SDB) (KASVI ® , Curitiba, Brazil) were placed. In the next step, 100μL of evaluated antifungal drugs were added to the first well microplate and serially diluted. The microplates were incubated at 37°C for 48 hours. Focusing cellular viability, 50μL of TCT (Triphenyl Tetrazolium Chloride) dye (Sigma-Aldrich ® , Saint Louis, USA) were dropped in each well microplate [13]. The reference values to MIC for nystatin were ≤ 4 μg/mL for sensitive isolates, and ≥ 64μg/mL for resistant isolates Fornari et al. [14]. For amphotericin B, susceptibility was found to be < 1μg/mL, and resistance > 2μg/mL [14]. C. albicans (ATCC 10221), C. tropicalis (ATCC 750), C. glabrata (ATCC 2001) and C. krusei (ATCC 34135) were used as reference strains. The results of the MIC obtained from the reference samples was MIC < 4g/mL for nystatin, and CIM ≤ 1mg/mL, for amphotericin B.

Data analysis
Proportion tests (Chi-Square Pearson and Fisher Exacttests) for evaluation of distribution differences were made to Candida colonization and variables of interest on ALL group. The analysis was performed using IBM SPSS (22.0) software, at significance level of 5%.

Ethical aspects
This research was approved by the Ethics and Research Committee of the Health Sciences Center of the Federal University of Paraíba with the number of opinion 706.409.

RESULTS
The mean age of ALL group was 8.1 (± 5.11) and the comparative group was 8.2 (±4.58). The mean DMFT of both groups was 2.04 and dmft was 1.25. The matched and other variables evaluated in both groups are placed on table 2. The positivity to Candida was higher in the ALL group compared with the comparative one (p=0.001). In both groups, clinical lesions suggesting oral candidosis were not found and apparently none clinical overlaying with oral mucositis on ALL group.
The descriptive analysis of Candida colonization of groups is detached on table 3. CHROMagarCandida suggested the isolation of 8 strains as being C. albicans (61.5%), being one collected from the comparative group, followed by C. glabratain five strains (35.7%), C. krusei on one ALL individual and on one comparative subject (14.3%) and one strain of C. tropicalis (7.1%). According to PCR identification, C. albicans colonies were detected in 13 samples (86.6%), while only on 2 (13.4%) samples was identified as being non albicans. Of these positive individuals to Candida, 7 (53.8%) were in induction/reinduction/ consolidation therapy and 6 (46.2%) in maintenance phase. Eight (61.5%) ALL individuals with positive Candida detection were under the use of systemic steroids at the moment of sampling. Moreover, 2 (15.0%) individuals were under the use of nystatin and/or amphotericin B to systemic candidosis but without any oral signs of infection. All individuals who had 10³ CFU/mL counts were on the induction phase of chemotherapy.   Considering age strata, gender, treatment phase, hematological variables and OAG indicators, the association of presence of oral mucositis and positivity to Candida was evident (p=0.017). Positive individuals colonized to Candida presented mucositis indexes mainly ranging from 10 to 14 points. The results of these associations or the lack of them were placed on table 4. The table 5 describes the results of microdilution assay and resistance/susceptibility protocol to the employed antifungal drugs. It was found that six out of 11 strains of C. albicans (54.5%) on ALL

DISCUSSION
The ALL affects children of all ages, with peak incidence normally happening between two and five years old, with a slight predominance of males corroborating our data with previous distribution found in the literature [15,16]. It is extremely important to undergo the dental treatment in ALL individuals and to be screened to oral infections [17,18]. Curiously,ALL individuals could present good dental condition compared with other populations as mentioned in our study [19]. In our study both groups had similar access to dental care without any considerable sociodemographic differences [20].
Candidemia in ALL patients could be associated with relapses, prolonged neutropenia and antibiotic administration [3,21]. Curiously, our study showed a lack of association between Candida colonization and hematological parameters. However, considering different types of variables including hematological counts, supportive and treatment conditions could be more relevant [21].CHROMagarCandidacould be used to presumptively identify Candida species. Good compliance between CHROMagarCandidaand other methods of identification was shown but without perfect concordance [22]. In fact, the use of nonmolecular identification methods could be markedly influenced by personal assessment [23]. In fact, molecular identification is an important addition to the conventional identification of Candida species [24].
C. Albicans was the most prevalent strain in the ALL group. A study which evaluated the oral cavity of 111 positive HIV individuals found C. Albicans was the most isolated species (83.5%) whereas non albicans species were isolated from 16.5% of colonized individuals [25]. Previously, our group showed increasing of nonalbicans species but on elderly irradiated individuals on head and neck [26].
Other interesting result was the higher counting of CFU of Candida associated with the induction phase of chemotherapy. Probably, induction phase reduces inflammatory response, even on salivary level, favoring Candida colonization. As previously showed, the induction phase had more association of oral complications [19] but them could remain on maintenance phase. The association between the absence of candidosis and mucositis was evident. Previously, it was observed correlation between mucositis and the presence of Candida on ALL patients [27,28].
It is important to carry out antifungal susceptibility testing because of the resistance of Candida strains to certain drugs. Curiously, a previous study found 95% of their isolates in HIV patients were inhibited by nystatin and amphotericin B [29]. However, emerging resistance of Candida species to antifungal drugs is a real problem. For example, a study with six out of nine children with ALL, fungal infection was progressive despite intravenous antifungals. The high percentage (21%) of death from invasive fungal infection among lethal infections in pediatric ALL individuals illustrates the relevance of fungi in this group of individuals [30].

CONCLUSION
In conclusion, Candida colonization was associated with ALL probably due to its relation to mucositis events being the higher colony counts found during the induction phase of chemotherapy. Candida Albicans was the prevalent strain and resistance to nystatin and amphotericin B was found. PCR fingerprinting could be used as definitive method to identify Candida species in addition to presumptive identification Collaborators LC MONTEIRO, conceptualization; data curation; investigation; methodology; visualization; writing-original draft. ILA RIBEIRO, RFB BONAN and AMG VALENÇA, data curation; formal analysis. PP MACIEL, data curation; formal analysis; investigation; methodology; visualization. ACB DULGHEROFF, JR SOUZA and LRC CASTELLANO, data curation; investigation; methodology; visualization; supervision; validation; writing-review & editing. Y WANDERLEY, conceptualization; data curation; visualization; supervision; validation. PRF BONAN, conceptualization; data curation; funding acquisition; project administration; resources; software; visualization; supervision; validation; writing-review & editing.