Antibacterial potential of saliva in children with leukemia

Antibacterial potential of saliva in children with leukemia Ewa Karolewska, DMD, PhD,a Tomasz Konopka, Prof. DMD, PhD,a Małgorzata Pupek, PhD,b Alicja Chybicka, Prof. MD, PhD,c Magdalena Mendak, DMD,a Wroclaw, Poland WROCLAW MEDICAL UNIVERSITY

Objectives: The objectives of this study were to evaluate the local oral defense mechanisms during the course of leukemia, and to define the correlation between the activity of salivary antibacterial factors and the oral clinical findings. Study design: A total of 44 children with newly diagnosed acute leukemia participated in the study. The control group consisted of 23 healthy children. The examination took place at the time of the diagnosis, and during and at the end of the chemotherapy treatment course. During the collection of resting mixed saliva samples the salivary flow rate was measured. In the saliva’s supernatant the following parameters were determined: total protein, peroxidase, myeloperoxidase, lysozyme, lactoferrin, and secretory immunoglobulin A. Results: The introduction of chemotherapy caused a slight decrease of salivary secretion rate (P ⬍ .05), as well as the decrease of S-IgA concentration (P ⬍ .01), which remained at the same level after the end of chemotherapy (P ⬍ .001). Patients with aplasia had decreased levels of peroxidase (P ⫽ .014) and myeloperoxidase (P ⫽ .013). Patients with oral mucositis presented with lower myeloperoxidase (P ⫽ .026) and peroxidase (P ⫽ .003) activity levels as well as the drop of S-IgA (P ⫽ .000) concentration compared with subjects with no mucositis. Conclusions: Antileukemic treatment contributes to the compromise of salivary defense mechanisms, therefore it is reasonable to support pharmacologically the saliva’s antibacterial potential of leukemic patients to impede the development of local infection. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;105:739-44)

Childhood leukemia accounts for about one third of all malignancies among children younger than 15 years. Each year in Poland about 300 children younger than 18 fall ill with leukemia.1 Considering, in spite of great progress in the treatment of leukemia, the high mortality rate (the most frequent cause of death in children, second only to injuries) and treatment costs, childhood leukemia remains a serious clinical and social problem. The oral manifestations of leukemia comprise of a series of symptoms resulting from the disease itself and treatment complications. The most frequently observed is chemotherapy and radiotherapy-induced inflammation of the oral mucosa termed mucositis. Mucositis typically manifests as an erythematous, burnlike lesion or as random, focal to diffuse, ulcerative lesions. It is associated with significant morbidity, pain, odynodysphagia, dysgeusia, and subsequent dehydration and malnutrition reducing the quality of life of affected patients.2 The second most common group of oral complications are opportunistic infections that often dea

Department of Oral Pathology, Wroclaw Medical University, Poland Department of Chemistry and Immunochemistry, Wroclaw Medical University, Poland c Department of Paediatric Bone Marrow Transplantation, Oncology and Haematology, Wroclaw Medical University, Poland. Received for publication Dec 20, 2006; returned for revision Sep 28, 2007; accepted for publication Oct 15, 2007. 1079-2104/$ - see front matter © 2008 Mosby, Inc. All rights reserved. doi:10.1016/j.tripleo.2007.10.010 b

velop under the condition of suppressed immunity in leukemia patients. Over 30% of patients suffer from oral infectious lesions,3 with fungal infections being the most frequent. The presence of a local infection in the oral cavity of an immunocompromised patient can lead to life-threatening systemic infection. Saliva plays a significant role in maintaining oral homeostasis; it appears to be one of the most important host factors influencing the evolution of oral lesions. In the oral cavity, saliva serves as the first barrier defending against microbial invasion by mechanical, nonimmunologic (nonspecific), and immunologic (specific) means. The main defending element is its continuous flow and the “flushing effect,” i.e., eliminating food debris and exogenous harmful factors. Salivary immunoglobulins, along with high-molecular-weight glycoproteins, restrain bacteria metabolism, its adhesion, and cause bacterial aggregation. A group of salivary proteins (e.g., lysozyme, peroxidase, myeloperoxidase, lactoferrin) working together with other salivary components (thiocyanate, chlorine, hydrogen peroxide) can affect oral bacteria and fungi by interfering with their growth or killing them. The high prevalence of oral lesions in the course of leukemia, their participation in the reduction of the quality of life, and the risk of systemic complications in children became the starting point for the conducted study. In the available literature concerning hematological malignancies, there was not a lot of research in which saliva served as a biological material,4-9 and the results are sometimes 739

740

OOOOE June 2008

Karolewska et al.

Table I. Descriptive data of experimental subjects n Boys/girls Age range (mean ⫾ SD) Risk group Standard High Relapse

ALL pre-B

ALL T

AML

31 13/18 3-17 (9.8 ⫾ 5.2)

7 7/0 6-17 (9.5 ⫾ 3.8)

6 5/1 9-17 (15.7 ⫾ 2.4)

20 8 3

2 3 2

2 3 1

ALL, acute lymphoblastic leukemia; AML, acute myelogenous leukemia.

contradictory. The aim of the present study was to evaluate several parameters in saliva in charge of the local host immune system, from the time of the diagnosis of leukemia until the end of the chemotherapy course, and to define the condition of the salivary defense system in children with mucositis and aplasia. MATERIAL AND METHODS A total of 44 children with newly diagnosed acute leukemia, admitted to the Department of Paediatric Bone Marrow Transplantation, Oncology, and Haematology, Wroclaw Medical University, between November 2001 and April 2004, participated in the study. The mean age was 10.26 (range, 3 to 17 years) (Table I). All children treated at this time in the aforementioned hospital who were able to participate in saliva collection (children younger than 3 were disqualified) and whose parents, after becoming aware of the planned examination, signed the informed consent approved by the Bioethical Committee of Wroclaw Medical University, participated in the study. The control group consisted of 23 generally healthy children who attended the Department of Pedodontic Dentistry, Medical University of Wroclaw, for dental checkups. The mean age was 8.7 ⫾ 2.6 (range, 5 to 14 years). Children were treated according to protocols accepted by the Polish Pediatric Group of Leukemia and Lymphoma Treatment: ALLIC BFM 2002, for subjects with acute lymphoblastic leukemia, and ANLL-98, for subjects with acute non-lymphoblastic leukemia. The treatment in acute leukemia is long (lasting for up to 2 years), and according to the aforementioned protocol, is adapted individually to each patient. The cytostatics used in the ANLL-98 protocol are etoposide, vincristine, idarubicin, cytarabine, 6-thioguanine, methotrexate, and cyclophosfamide, whereas in ALLIC BFM 2002, methotrexate, vincristine, daunorubicin, L-asparaginase, cyclophosfamide, 6-mercaptopurine, and cytarabine are used. All children received a 0.2% chlorhexidine gluconate mouthwash and nystatin suspension (100,000 UI/mL) 4

times a day for the duration of their treatment. The mouth rinses were administered to the younger children by their parents or by older patients themselves. Patients with any oral lesions (e.g., oral mucositis, oral candidiasis) additionally received amphotericin B suspension (Amphomoronal suspension, Squibb-Heyden, Munich, Germany) to be applied topically. The oral examination took place at the time of the diagnosis of leukemia (as soon as possible after admission to hospital), each week during the course of chemotherapy, and directly after the end of the course of chemotherapy. An oral examination was performed at the patient’s bed. The presence of mucositis was coded according to the World Health Organization’s (WHO) mucositis rating scale. Resting mixed saliva samples were collected in the morning hours (before or 2 hours after breakfast) in plastic test tubes placed in a flask of ice. During saliva collection, the time of collection was noted in order to measure the salivary secretion rate (SSR) (mL/min). Saliva samples were centrifuged at 3000g for 15 minutes. The obtained supernatant was frozen at –30°C and stored at this temperature until analyzed. In the supernatant, the total protein (TP) was determined by means of bicynchonic micromethod and expressed in mg/mL.10 Peroxidase activity (SPO) was determined by calculating the frequency of oxidation of acid 5-tio-2-nitrobenzoesic acid (Nbs) to 5.5-ditiobis2-nitrobenzoesic acid (Nbs2) by OSCN– formed during oxidation of SCN– by peroxidase, according to the Mansson-Rahemtulla et al. method.11 The specific activity was expressed in IU per mg of total protein in saliva. The determination of myeloperoxidase activity (MPO) was performed in the same way as the determination of peroxidase except that the saliva was dialyzed to remove the naturally occurring thiocyanate, whose presence increases the value of obtained results. The method of the determination of MPO is based on the time of oxidation of Cl– to OCl– present in the substrate Nbs-Cl. The specific activity was expressed in IU per mg of saliva total protein. The determination of the lysozyme level was based on the lytic activity of lysozyme (LZ) on the bacteria Micrococcus lysodeikticus (Micrococcus luteus). Because of the formation in saliva of nonsoluble lysozyme complexes with mucins and proteins, the lysozyme determination was performed in parallel with and without the mucin precipitation of 0.2% cethyltrimethylammonium bromide (CETAB) according to the modified method of Smolelis and Hartsell.12 The results were expressed in ␮g per mg of total protein. Lactoferrin and S-immunoglobulin A (IgA) were determined with the immunoenzymatic method (enzymelinked immunosorbent assay [ELISA]). The determined values of lactoferrin were expressed in ␮g per mg of total protein and S-IgA in ng per mg of total protein.

OOOOE Volume 105, Number 6

Karolewska et al. 741

Table II. Mean values of examined parameters in sick and healthy children Children with leukemia

Variable

Before treatment Mean ⫾ SD

During CHT course Mean ⫾ SD

At the end of CHT course Mean ⫾ SD

Conrol group Mean ⫾ SD

V [mL/min] TP [mg/mL] LZ-CETAB [␮g/mgTP] LZ-H2O [␮g/mgTP] SPO [IU/mgTP] MPO [IU/mgTP] LF [␮g /mgTP] S-IgA [ng/mgTP]

0.39 ⫾ 0.15 1.84 ⫾ 0.92 41.58 ⫾ 37.03 42.38 ⫾ 57.2 0.38 ⫾ 0.24 0.34 ⫾ 0.22 1.21 ⫾ 0.68 74.89 ⫾ 50.5

0.39 ⫾ 0.2* 2.07 ⫾ 2.04 43.01 ⫾ 29.09 32.41 ⫾ 25.07 0.51 ⫾ 0.52 0.26 ⫾ 0.21 2.16 ⫾ 5.88 64.28 ⫾ 66.19†

0.41 ⫾ 0.17 2.07 ⫾ 1.77 44.53 ⫾ 25.05 38.35 ⫾ 30.31 0.49 ⫾ 0.38 0.26 ⫾ 0.19 1.44 ⫾ 0.98 54.8 ⫾ 40.96‡

0.56 ⫾ 0.29 1.71 ⫾ 1.33 40.67 ⫾ 36.06 14.59 ⫾ 5.25 0.45 ⫾ 0.4 0.48 ⫾ 0.76 1.71 ⫾ 0.82 98.31 ⫾ 48.56

CHT, chemotherapy; V, salivary secretion rate; TP, total protein; LZ-CETAB, lysozyme determined after mucin precipitation; LZ-H2O, lysozyme determined without mucin precipitation; SPO, salivary peroxidase; MPO, myeloperoxidase; LF, lactoferrin; S-IgA, secretory immunoglobulin A. *Statistically significantly different from controls at P ⬍ .05. †Statistically significantly different from controls at P ⬍ .01. ‡Sstatistically significantly different from controls at P ⬍ .001.

Because of poor general health, not all patients could participate each week in the oral examination and saliva collection. The obtained data therefore could not be analyzed weekly and for each patient an average of all measures during chemotherapy for each variable was calculated. For the purpose of defining the changes in saliva composition during mucositis, all of the weekly results of patients receiving chemotherapy were taken into consideration. Similarly, when assessing the saliva antimicrobial potential in children with aplasia, all weekly results of children with total neutrophil count less than 0.5 ⫻ 109 cells/L were compared with the results obtained from children who did not have aplasia. In the statistical analysis, the significance was set at the 95% confidence level (P ⬍ .05 for hypothesis testing) to compare the mean values between 2 groups. Because of the lack of normal distribution of analyzed variables, the Mann-Whitney U test was used. RESULTS The mean values for the concentrations of all analyzed salivary parameters in the experimental and control groups are presented in Table II. At the time of diagnosis, no significant differences in the salivary secretion rate or the composition of saliva were observed when compared with the control group. The introduction of chemotherapy induced a slight decrease in the salivary secretion rate (P ⫽ .012), as well as a decrease in the concentration of secreted IgA (P ⬍ .001). At the end of the course of chemotherapy, no changes in the secretion rate or the composition of saliva were observed apart from the lower level of salivary S-IgA (P ⬍ .002). Despite the lack of significant difference in salivary lysozyme level during the course of leukemia, a rising trend in its activity could be observed (Fig. 1).

Patients with aplasia had a significantly lower activity level of peroxidase (P ⫽ .014) and myeloperoxidase (P ⫽ .013) (Table III) in comparison with patients who did not have aplasia. While analyzing the salivary parameters, depending on the occurrence of mucositis, it was observed that subjects with oral cytotoxic complications had significantly decreased levels of myeloperoxidase (P ⫽ .026), salivary peroxidase (P ⫽ .003), and S-IgA (P ⫽ .000) as well as almost twice the amount of total protein found in saliva compared with patients with no oral complications (P ⫽ .001) (Table IV). DISCUSSION Leukemia itself did not seem to cause any changes in the saliva’s antibacterial potential. Changes affecting the salivary defense mechanisms appeared after the introduction of chemotherapy. Cytostatic treatment led to a significant decrease in salivary secretion rates (SSR) and in S-IgA concentrations in sick children compared with healthy subjects. The lack of significant difference in SSR at the end of the course of chemotherapy excludes the direct influence of anticancer treatment on salivary glands. These results are consistent with the results of Mansson-Rahemtulla et al.,6 who noted the decrease in SSR only during the first 2 weeks after the introduction of treatment. Wahlin5 noted similar observations. She observed the reduction of SSR in the first days of chemotherapy. At the end of the chemotherapy (CHT) course, the secretion rate reached the normal values. Patients are known to be very anxious when the cytotoxic therapy starts.13 Emotional stress can lead to the decrease of the SSR14 and this can be explained as having a sympathoadrenal effect on the salivary glands. The examined patients

742

OOOOE June 2008

Karolewska et al. 45 40 35

µg/mg TP

30 25

LZ-CETAB (median) LZ-H2O (median)

20 15 10 5 0 before treatment

during CHT

after CHT

Fig 1. The distribution of salivary lysozyme levels in children with leukemia.

Table III. Mean values of biochemical salivary parameters in patients with aplasia and without aplasia Subjects with aplasia n ⫽ 47

Subjects without aplasia n ⫽ 68

Variable

mean ⫾ SD

mean ⫾ SD

P

V [mL/min] LZ-CETAB [␮g/mgTP] LZ-H2O [␮g/mgTP] SPO [IU/mgTP] MPO [IU/mgTP] LF [␮g /mgTP] S-IgA [ng/mgTP]

0.38 ⫾ 0.20 37.28 ⫾ 25.01 32.98 ⫾ 30.37 0.37 ⫾ 0.26 0.21 ⫾ 0.16 1.39 ⫾ 1.26 68.05 ⫾ 57.51

0.40 ⫾ 0.16 47.54 ⫾ 39.03 33.78 ⫾ 30.26 0.53 ⫾ 0.37 0.27 ⫾ 0.16 1.27 ⫾ 0.97 51.02 ⫾ 43.54

ns ns ns .014 .013 ns ns

Table IV. Mean values of determined variables of patients with leukemia with and without mucositis Children with mucositis n ⫽ 20

Children without mucositis n ⫽ 95

Variable

mean ⫾ SD

mean ⫾ SD

P

V [mL/min] LZ-CETAB [␮g/mgTP] LZ-H2O [␮g/mgTP] SPO [IU/mgTP] MPO [IU/mgTP] LF [␮g/mgTP] S-IgA [ng/mgTP]

0.43 ⫾ 0.13 45.77 ⫾ 41.25 32.13 ⫾ 19.60 0.27 ⫾ 0.18 0.17 ⫾ 0.10 1.67 ⫾ 1.51 42.23 ⫾ 72.29

0.42 ⫾ 0.21 41.99 ⫾ 32.88 33.37 ⫾ 34.03 0.49 ⫾ 0.42 0.31 ⫾ 0.37 1.62 ⫾ 3.34 71.86 ⫾ 54.01

ns ns ns .003 .026 ns .000

ns, not significant. Other abbreviations as in Table II.

ns, not significant. Other abbreviations as in Table II.

were receiving their first course of chemotherapy. If it was the cytotoxic treatment that had a destructive effect on salivary glands, the decrease of SSR would last for a longer period of time. The drop in the salivary secretion rate could be indirectly influenced by the administration of cytostatics. Nausea and vomiting in the first days of treatment is a frequent side effect of CHT administration. For the purpose of elimination of these ailments, patients take antiemetic drugs, which can limit the secretion of saliva. Similar results of chemotherapy-induced decrease of S-IgA salivary levels were also observed by other authors.9,15,16 The study by Meurman et al.,9 on the effect of anticancer treatment in patients with lymphomas lasting 5 years, showed the decrease of the salivary S-IgA concentration in resting saliva during chemotherapy compared with the initial concentration. Moreover, S-IgA concentration remained at this low level for 5 years after the introduction of chemotherapy, indicating

the lasting immunosuppression. Furthermore, Meurman et al.9 observed that patients who died during the study had a lower concentration of IgA initially compared with patients who survived. The progressive decrease of the IgA concentration in patients receiving CHT can be explained by the CHT influence on the population of T and B cells in Payer’s patches. The study carried out on animals revealed a drastic decrease of lymphocyte B in the mucous lamina propria of the jejunum and a significant temporary IgA deficiency as the effect of the administration of only 1 dose of cyclophosphamide.17 Other studies also report the presence of acquired IgA deficiencies in patients with acute leukemia.7,18-20 Chemotherapy can cause IgA disturbances, while at the same time not affecting the values of other immunoglobulins.20 It has been proven that cyclophosfamide at a larger dose and taken for a much longer period than methotrexate or cisplatin causes the reduction of lymphocyte B in the marginal

OOOOE Volume 105, Number 6

zone of rat spleen, at the same time reducing the number of lymphocytes producing antibodies.21 The reduction of the spleen marginal zone and the delayed recovery process suggests that despite the increased risk of infection due to neutropenia, patients treated with CHT are prone to infections from the encapsulated bacteria long after the end of therapy. It is not known how the CHT-induced suppression of lymphocyte B, which causes the decrease of serum IgA concentration, affects the level of S-IgA in saliva. Until now, there is no research evaluating the level of IgA in serum and S-IgA in saliva at the same time after the administration of cytostatics. However, the concentration of IgA in both environments was evaluated in patients with hypoglobulinemia.22 The concentration of S-IgA in saliva was slightly higher when compared with serum level, although the level of antibodies in saliva was significantly lower when compared with the control group. The low S-IgA level was accompanied by the slight increase of nonspecific defense factors— lysozyme and lactoferrin, whose levels were slightly higher compared with the control group. Although no other significant changes were observed in the saliva composition in children with leukemia, some differences appeared within the group of children who received cytostatic treatment. Patients presenting with oral mucositis had significantly lower activity of sialo- and myeloperoxidase. Large ulcerative lesions seem to favor the formation of granulocytic infiltration especially if these lesions are settled with bacteria. However, the lack of peripheral blood granulocytes caused by cytostatic treatment leads to an extremely low number of neutrophils in saliva and subsequently low MPO activity. Dreizen et al.23 showed that oral granulocytes were no longer present in saliva of patients 13 days after the start of chemotherapy. The decrease of oral granulocytes accompanied the decrease of serum granulocytes. The granulocytes reach normal limits in saliva even before they reach a normal count in peripheral blood.24,25 Failure of the peroxidase system may result in an accumulation of excessive amounts of H2O2, leading to further tissue damage. At the time of complete depletion of the immunologic system, the salivary defense function also fails. Patients with aplasia had significantly lower concentrations of SIgA and decreased peroxidase activity. Antileukemic treatment did not significantly affect the lysozyme activity level, although a rising trend during therapy could be observed. The immunosuppressive therapy being the risk factor for developing fungal infections could explain this rising trend. A few research studies indicate the aggravation of salivary lysozyme levels resulting from cytotoxic treatment.4,8 Distinct results were published by Main et al.15 A number of observations demonstrate a higher con-

Karolewska et al. 743

centration of salivary lysozyme in patients prone to oral fungal infections.26-29 These studies concerned HIV-positive patients, in whom the increase of lysozyme was detected in stimulated and resting saliva produced by large salivary glands.26-29 The unchanged lysozyme activity level observed in children with leukemia does not necessarily prove its effectiveness in the antifungal defense. Samaranayake et al.30 report that, while analyzing the antifungal effect of lysozyme against genetically similar Candida albicans species that have been isolated from HIV patients, the presence of fungal resistance to lysozyme increases over the duration of the disease. Because of the small amount of collected saliva from sick children, it was impossible to determine the level of thiocyanate, as it is its presence that determines the efficacy of peroxidases. Referring to the study by ManssonRahemtulla et al.,6 we can assume that in our study chemotherapy had a significant impact on the decrease of the thiocyanate concentrations in saliva. This would mean that the actual antibacterial activity of the enzyme may be disturbed despite the unchanged salivary peroxidase levels due to decreased levels of cyanate. The biological function of IgA in the secretions is manifold. Apart from agglutinating bacteria and limiting the mechanisms of microbial adhesion to mucous membranes, it plays an important role in the regulation of the inflammatory response.31-33 IgA binding with the Fc␣R receptor present on phagocytic cells leads to the decrease of chemotaxis, phagocytosis, and ROS production.32 Human serous IgA induces a significant increase of interleukin (IL)-1Ra secretion in peripheral blood mononuclear cells and adherent monocytes. By the induction of IL-1Ra and the decrease of proinflammatory cytokine release such as IL-1␤, tumor necrosis factor (TNF)-␣, and IL-6, IgA contributes to the inflammatory response regulation. The current mucositis pathobiology according to Sonis34 describes the influence of proinflammatory cytokines (IL1␤, TNF-␣, and IL-6) generated by radio- and chemotherapy on epithelium and submucosa. The low concentration of S-IgA in patients undergoing antineoplastic treatment could be partly responsible for a higher risk of developing mucositis. The decrease of salivary peroxidase activity and secreted IgA not only affects the salivary antimicrobial activity, but also the antioxidant capacity of saliva, making oral mucosa prone to cytotoxic injury and infection. Considering a high prevalence of oral infections and mucositis in children treated for leukemia, as well as a partial failure of the saliva’s protective function, it is necessary to enhance the local host defense. Apart from a whole range of antiseptics and more specific antimicrobial therapeutics commonly used to prevent and manage oral infections and mucositis, it seems reasonable to activate the host defense in human

744

OOOOE June 2008

Karolewska et al.

saliva by using products (dentifrices, chewing gums, mouth rinses, and moisturizing gels) containing peroxidase, lactoferrin, and lysozyme. REFERENCES 1. Kowalczyk JR, Dudkiewicz E, Balwierz W, Bogusławska-Jaworska J, Rokicka-Milewska R. Incidence of childhood cancers in Poland in 1995-1999. Med Sci Monit 2002;8:87-90. 2. McGuire DB, Yeager KA, Dudley WN, Peterson DE, Owen DC, Lin LS, Wingard JR. Acute oral pain and mucositis in bone marrow transplant and leukemia patients: data from a pilot study. Cancer Nurs 1998;21:385-93. 3. Dreizen S, McCredie KB, Bodey GP, Keating MJ. Quantitive analysis of the oral complications of antileukemia chemotherapy. Oral Surg Oral Med Oral Pathol 1986;62:650-3. 4. Pajari U, Poikonen K, Larmas M, Lanning M. Salivary immunoglobulins, lysozyme, pH, and microbial counts in children receiving anti-neoplastic therapy. Scand J Dent Res 1989;97:171-7. 5. Wahlin YB. Salivary secretion rate, yeast cells and oral candidiasis in patients with acute leukemia. Oral Surg Oral Med Oral Pathol 1991;71:689-95. 6. Mansson-Rahemtulla B, Techanitiswad T, Rahemtulla F, McMillan TO, Bradley EL, Wahlin YB. Analyses of salivary components in leukemia patients receiving chemotherapy. Oral Surg Oral Med Oral Pathol 1992;73:35-46. 7. Abrahamsson J, Marky I, Mellander L. Immunoglobulin levels and lymphocyte response to mitotic stimulation in children with malignant disease during treatment and follow-up. Acta Paediatr 1995;84:177-82. 8. Laine P, Meurman JH, Tenovuo J, Murtomaa H, Lindqvist C, Pyrhonen S, et al. Salivary flow and composition in lymphoma patients before, during and after treatment with cytostatic drugs. Eur J Cancer B Oral Oncol 1992;28B:125-8. 9. Meurman JH, Laine P, Keinanen S, Pyrhonen S, Teerenhovi L, Linqvist C. Five-year follow-up of saliva in patients treated for lymphomas. Oral Surg Oral Med Oral Pathol 1997;83:447-52. 10. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, et al. Measurement of protein using bicinchonic acid. Anal Biochem 1985;150:76-85. 11. Mansson-Rahemtulla B, Baldone DC, Pruitt KM, Rahemtulla F. Specific assay for peroxidases in human saliva. Arch Oral Biol 1986;31:661-8. 12. Majewska A, Kasiak M, Sozan´ska Z. Aktywnos´c´ lizozymu a próchnica ze¸bów u dzieci. Streszczenie IX Zjazdu Sekcji Stomatologii Dziecie¸cej PTS. Poste¸py w Stomatologii Wieku Rozwojowego, Łódz´, 1997. p. 25. 13. Rhodes VA, Watson PM, Johnson MH. Association of chemotherapy-related nausea and vomiting with pretreatment anxiety. Oncol Nurs Forum 1986;13:41-7. 14. Bates JF, Adams D. The influence of mental stress on the flow of saliva in man. Arch Oral Biol 1986;13:593-6. 15. Main BE, Calman KC, Ferguson MM, Kaye SB, MacFarlanr TW, Mairs RJ, et al. The effect of cytotoxic therapy on saliva and oral flora. Oral Surg 1984;58:545-8. 16. Harrison T, Bigler L, Tucci M, Pratt L, Malamud F, Thigpen JT, et al. Salivary sIgA concentrations and stimulated whole saliva flow rates among women undergoing chemotherapy for breast cancer: an exploratory study. Spec Care Dentist 1998;18:109-12. 17. Cozon G, Cannella D, Perriat-Langevin A, Jeannin M, Trublereau P, Ecochard R, et al. Transient secretory IgA deficiency in mice after cyclophosfamide treatment. Clin Immunol Immunopathol 1991;61:93-102. 18. Khalifa AS, Take H, Cejka J, Zuelzer WW. Immunoglobulins in acute leukemia in children. J Pediatr 1974;85:788-91.

19. Kristinsson VH, Kristinsson JR, Jonmundsson GK, Jonsson OG, Thorsson AV, Haraldsson A. Immunoglobulin class and subclass concentrations after treatment of childhood leukemia. Pediatr Hematol Oncol 2001;18:167-72. 20. Uram R, Rosoff PM. Isolated IgA deficiency after chemotherapy for acute myelogenous leukemia in an infant. Pediatr Hematol Oncol 2003;20:487-92. 21. Zandvoort A, Lodewijk ME, Klok PA, Dammers PM, Kroese FG, Timens W. Slow recovery of follicular B cells and marginal zone B cells after chemotherapy: implications for humoral immunity. Clin Exp Immunol 2001;124:172-9. 22. Kirstilä V, Tenovuo J, Ruuskanen O, Nikoskelainen J, Irjala K, Vilja P. Salivary defense factors and oral health in patients with common variable immunodeficiency. J Clin Immunol 1994;14:229-36. 23. Dreizen S, Menkin DJ, Keating MJ, McCredie KB, O’Neill PA. Effect of antileukemia chemotherapy on marrow, blood, and oral granulocyte counts. Oral Surg Oral Med Oral Pathol 1991;71:45-9. 24. Wright DG, Meierovics AI, Foxley JM. Assessing the delivery of neutrophils to tissues in neutropenia. Blood 1986;67:1023-30. 25. Lieschke GJ, Ramenghi U, O’Connor MP, Sheridan W, Szer J, Morstyn G. Studies of oral neutrophil levels in patients receiving G-CSF after autologous marrow transplantation. Br J Haematol 1992;82:589-95. 26. Yeh CK, Fox PC, Ship JA, Busch KA, Bermudez DK, Wilder A, et al. Oral defense mechanisms are impaired early in HIV-1 infected patients. J Acq Immune Defic Syndr 1988;1:361-6. 27. Atkinson JC, Yeh C, Oppenheim FG, Bermudez D, Baum BJ, Fox PC. Elevation of salivary antimicrobial proteins following HIV-1 infection. J Acq Immune Defic Syndr 1990;3:41-8. 28. Mandel ID, Barr CE, Turgeon L. Longitudinal study of parotid saliva in HIV-1 infection. J Oral Pathol Med 1992;21:209-13. 29. Muller F, Holberg-Petersen M, Rollag H, Degre M, Brandtzaeg P, Froland SS. Nonspecific oral immunity in individuals with HIV infection. J Acq Immune Defic Syndr 1992;5:46-51. 30. Samaranayake YH, Samaranayake LP, Pow EH, Beena VT, Yeung KW. Antifungal effects of lysozyme and lactoferrin against similar, sequential Candida albicans isolates from a human immunodeficiency virus-infected southern Chinese cohort. J Clin Microbiol 2001;39:3296-302. 31. Wolf HM, Fischer MB, Puhringer H, Samstag A, Vogel E, Eibl MM. Human serum IgA downregulates the release of inflammatory cytokines (tumor necrosis factor-alpha, interleukin-6) in human monocytes. Blood 1994;83:1278-88. 32. Wolf HM, Vogel E, Fischer MB, Rengs H, Schwarz H-P, Eibl MM. Inhibition of receptor-dependent and receptor-independent generation of the respiratory burst in human neutrophils and monocytes by human serum IgA. Pediatr Res 1994;36:235-43. 33. Wolf HM, Hauber I, Gulle H, Samstag A, Fischer MB, Ahmad RU, et al. Antiinflammatory properties of human serum IgA: induction of IL-1 receptor antagonist and Fc alpha R (CD89)-mediated downregulation of tumour necrosis factor-alpha (TNFalpha) and IL-6 in human monocytes. Clin Exp Immunol 1996;105:537-43. 34. Sonis ST. The biologic role for nuclear factor-kappaB in disease and its potential involvement in mucosal injury associated with anti-neoplastic therapy. Crit Rev Oral Biol Med 2002;13:380-9.

Reprint requests: Ewa Karolewska, DMD, PhD Wroclaw Medical University Oral Pathology ul. Wybrzeze Wyspianskiego 14/8 50-370 Wrocław, Poland [email protected]