Agenesis and Microdontia of Permanent Teeth as Late Adverse Effects after Stem Cell Transplantation in Young Children

Transcription

1 181 Agenesis and Microdontia of Permanent Teeth as Late Adverse Effects after Stem Cell Transplantation in Young Children Päivi Hölttä, D.D.S. 1 3 Satu Alaluusua, D.D.S., Ph.D. 1,2 Ulla M. Saarinen-Pihkala, M.D., Ph.D. 3 Jaakko Peltola, D.D.S., Ph.D. 4 Liisa Hovi, M.D., Ph.D. 3 1 Department of Pedodontics and Orthodontics, Institute of Dentistry, University of Helsinki, Helsinki, Finland. 2 Department of Oral and Maxillofacial Diseases, Helsinki University Central Hospital, Helsinki, Finland. 3 Hospital for Children and Adolescents, University of Helsinki, Helsinki, Finland. 4 Department of Oral Radiology, Institute of Dentistry, University of Helsinki, Helsinki, Finland. Supported by research grants from the Helsinki University Central Hospital and the Finnish Dental Society Apollonia. The authors thank Jorma Torppa, M.Sc., for statistical advice and Blackwell Publishing Ltd. for kind permission to use their material. Address for reprints: Päivi Hölttä, D.D.S., Institute of Dentistry, P.O. Box 41, University of Helsinki, Helsinki, Finland; Fax: (011) ; paivi.holtta@helsinki.fi Received August 5, 2004; revision received September 21, 2004; accepted September 21, BACKGROUND. The objective of the current study was to examine the occurrence of tooth agenesis and microdontia in pediatric stem cell transplantation (SCT) recipients. METHODS. The impact of total body irradiation (TBI) and age at SCT on agenesis and microdontia of permanent teeth was examined in 55 patients from panoramic radiographs. Assessment A1 (for tooth agenesis and microdontia) excluded the third molars, and assessment A2 (for tooth agenesis) included the third molars. Patients were grouped according to TBI status (the TBI group vs. the non-tbi group) and age at SCT (patients age 3.0 years [Group Y], patients ages years [Group M], and patients age 5.1 years [Group O]). RESULTS. From 1 to 12 teeth were missing in 77%, 40%, and 0% of patients (assessment A1) in Groups Y, M, and O, respectively (Group Y vs. Group M, P 0.055; Group Y vs. Group O, P 0.001; and Group M vs. Group O, P 0.002), increasing to 83%, 78%, and 43%, respectively, when the third molars were included (assessment A2; P values were not significant). Correspondingly, 75%, 60%, and 13%, respectively, of patients had 1 12 microdontic teeth (assessment A1: Group Y vs. Group M, P 0.306; Group Y vs. Group O, P 0.001; and Group M vs. Group O, P 0.003). Recipient age at the time of SCT was found to have a negative correlation with the number of missing teeth (P 0.001) and microdontic teeth (P 0.005). TBI appeared to have little effect on the prevalence of tooth agenesis (assessment A1: TBI group, 32%; non-tbi group, 29%; assessment A2: TBI group, 72%; non-tbi group, 46%; P values were not significant) or on the prevalence of microdontia (assessment A1: TBI, 41%; non-tbi, 50%; P value was not significant). A tendency toward an increased number of affected teeth was noticed in the group of patients who received TBI. CONCLUSIONS. Depending on their age at SCT, % of pediatric SCT recipients will later present with agenesis and/or microdontia of permanent teeth that may jeopardize occlusal development. Young age ( 5.0 years) at SCT was found to be a stronger risk factor than TBI, although TBI caused additive impairment. Cancer 2005;103: American Cancer Society. KEYWORDS: dental development, stem cell transplantation, late adverse effects, tooth agenesis, hypodontia, microdontia. Stem cell transplantation (SCT) has an established role in the treatment of selected malignant and nonmalignant diseases in children. High-dose chemotherapy (HDC) and total body irradiation (TBI) used in the preparative regimens for SCT give rise to multiple, well known, acute and long-term adverse effects, also involving teeth. Morphogenesis and calcification of teeth form a sequence of events that begins in utero and continues for years. 1 Thereafter, 2004 American Cancer Society DOI /cncr Published online 11 November 2004 in Wiley InterScience (

2 182 CANCER January 1, 2005 / Volume 103 / Number 1 FIGURE 1. The chronology of mineralization of permanent teeth. Reprinted with the permission of Blackwell Publishing Ltd. from Koch G, Thesleff I. Developmental disturbances in number and shape of teeth and their treatment. In: Koch G, Poulsen S, editors. Pediatric dentistry a clinical approach, 1st ed. Copenhagen: Munksgaard, 2001: the development of the third molars continues for several years. Abnormal events that occur during dental development have permanent sequelae that cannot be corrected later. Dental late effects of chemotherapy and/or radiotherapy in childhood may involve several kinds of developmental dental disturbances, including tooth agenesis, microdontia, and disturbed root development. It is possible to predict future dental aberrations to some extent by placing the period of therapy on the schedule of tooth mineralization (Fig. 1). Agenesis of teeth (other terms used include aplasia of teeth and hypodontia ) is a common dental anomaly in a healthy population, and its prevalence reportedly ranges from 2.8% 2 to 10%. 3 In a healthy Finnish population, a hypodontia prevalence of 8.0% (third molars excluded) has been reported. 4 Both genetic and environmental factors may result in tooth agenesis. In most individuals, hypodontia has a genetic background, as shown in family studies 5,6 or by identifying the gene mutations involved. 7 9 Some environmental factors, such as multiagent chemotherapy and radiotherapy, are known to cause tooth agenesis when used in pediatric anticancer therapy. The percentage of cancer patients with missing teeth is reported to range from 5 28% after conventional chemotherapy In several studies, the most extensive dental disturbances (agenesis, microdontia, and root anomalies) have been reported in children who were treated before the ages of 5 6 years. 11,12,15 There are fewer studies on SCT recipients. Those patients have had more dental disturbances compared with patients who were treated with conventional chemotherapy. 12,13 Information regarding the prevalence of microdontia among healthy populations is scarce, with varying criteria used in assessments. A 1.7% prevalence of small, peg-shaped, upper lateral incisors has been reported. 5 A microdontia prevalence of 1.9% was reported in healthy Japanese schoolchildren when teeth with 3.5 standard deviations below the genderspecific mean mesiodistal crown size were recorded. 16 Microdontia, registered from panoramic radiographs is more frequent after pediatric anticancer therapy, 10 13,15,17,18 ranging from 10% after conventional chemotherapy for hematologic malignancies 12 to 78% after SCT in patients with neuroblastoma. 18 In the current study, we examined 1) the prevalence of tooth agenesis and microdontia and 2) the number of missing and microdontic teeth in the permanent dentition of SCT recipients. We also analyzed the roles of TBI and age at SCT in dental adverse effects. MATERIALS AND METHODS Patients Fifty-six SCT recipients were examined for agenesis and microdontia of permanent teeth at the Institute of Dentistry, University of Helsinki (Helsinki, Finland). The patients had undergone SCT at the Hospital for Children and Adolescents, University of Helsinki, between 1980 and The eligibility criteria required that children were age 10 years at the time they underwent SCT and that the minimum follow-up was 1 year. Of the 85 consecutive survivors, 56 patients volunteered to take part in the examination. One patient was later excluded from the analysis because she had no definable teeth due to her very young age at the time of the dental examination. The final study group was comprised of 55 patients, including 28 males and 27 females, who underwent SCT at ages years (mean, 4.3 years) and who were followed after SCT for years (mean, 7.4 years). The mean patient age at the time of dental examination was 11.7 years (range, years). The underlying diagnoses of the patients are listed in Table 1. Thirteen patients underwent SCT due to recurrent disease. The treatment period between diagnosis and SCT varied from 0.2 years to 2.6 years (mean, 0.8 years). With the exception of only one patient, all patients were in continuous disease remission at the time of the dental examination. Informed consent was obtained from the patients

3 Tooth Agenesis and Microdontia after SCT/Hölttä et al. 183 TABLE 1 Key Characteristics of 55 Pediatric Stem Cell Transplantation Recipients No. of patients Diagnosis Total TBI Yes No Ara-C Cy Mel VMP ECT HDC Mean age in yrs at SCT (range) NBL a ( ) ALL ( ) AML ( ) Wilms tumor ( ) CML ( ) Lymphoma ( ) MDS ( ) SAA b 2 ( ) RMS Yolk sac tumor Total ( ) TBI: total body irradiation; HDC: high-dose chemotherapy; SCT: stem cell transplantation; Ara-C: cytosine arabinoside (3 g/m 2 12); Cy: cyclophosphamide ( mg/kg); Mel: melphalan ( mg/m 2 ); VMP: etoposide (300 mg/m 2 ), melphalan ( mg/m 2 ), and cisplatin (90 mg/m 2 ); ECT: etoposide (250 mg/m 2 3), carboplatin (500 mg/m 2 3), and thiotepa (300 mg/m 2 3); NBL: neuroblastoma; ALL: acute lymphoblastic leukemia; AML: acute myeloid leukemia; CML: chronic myeloid leukemia; MDS: myelodysplastic syndrome; SAA: severe aplastic anemia; RMS: rhabdomyosarcoma. a One patient in the non-total body irradiation-group received TBI that excluded the head. b The patient in the non-total body irradiation-group received total lymph node irradiation. and/or their guardians. The Institutional Review Board of the Hospital for Children and Adolescents, University of Helsinki, approved the study protocol. Treatment of the Patients Prior to SCT Conventional chemotherapy was given to patients with acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), and non-hodgkin lymphoma according to Nordic protocols 19,20 : For most patients, this included prednisolone, vincristine, doxorubicin, methotrexate, L-asparaginase, cyclophosphamide, cytosine arabinoside, and 6-mercaptopurine. Patients with neuroblastoma (NBL) received multiagent chemotherapy with vincristine, cyclophosphamide, dacarbazine, cisplatin, and doxorubicin with the addition of ifosfamide and etoposide in some patients. 21 The five patients with Wilms tumor were treated according to National Wilms Tumor Study-like protocols with vincristine, actinomycin D, cyclophosphamide, and doxorubicin. The one patient with rhabdomyosarcoma received vincristine, actinomycin D, cyclophosphamide, doxorubicin, ifosfamide, and etoposide. The one patient with yolk sac tumor received etoposide, bleomycin, and cisplatin. The only chemotherapeutic agent given to the two patients with chronic myeloid leukemia was hydroxyurea, whereas the patients with myelodysplastic syndrome and severe aplastic anemia did not receive any cytostatic chemotherapy prior to SCT. Radiotherapy was administered to 17 patients, either to the tumor bed or to local metastases. Two patients with AML received cranial irradiation of 12 grays (Gy), and 1 patient with ALL received 24 Gy. Local irradiation was given to 3 patients with NBL for metastases in the skull area. One patient received 20 Gy to the left frontal bone, the second patient received 20 Gy to the right orbital area, and the third patient received 6 Gy to the left temporal bone. Eleven patients received radiation either to the tumor bed or to metastases outside the head. None of those radiation fields covered the jaws, and this radiotherapy was not considered significant with regard to dental development. Conditioning for SCT The preparative regimen for SCT included TBI at a dose of Gy either in a single fraction (n 1 patient) or in 5 6 fractions (n 38 patients; the TBI group). The remaining 16 patients belonged to the non-tbi group (Table 1). The mean age at SCT did not differ between the groups (in the TBI group, the mean age was 4.4 years [range, years] and in the non-tbi group, the mean age was 4.1 years [range, years]; P 0.657). The chemotherapeutic agents with dosages used for HDC are provided in Table 1. Dental Examinations The clinical and radiologic examinations of all but two patients were performed at the Institute of Dentistry,

4 184 CANCER January 1, 2005 / Volume 103 / Number 1 FIGURE 2. Example of tooth agenesis and microdontia recording. Microdontia was recorded when the size of a tooth crown was approximately 50% of the size considered normal. (A) The crowns of the first premolar (short arrow), the second premolar (arrowhead), and the permanent first molar (long arrow) are considered normal. (B) The crown of the first premolar (short arrow) is microdontic, the second premolar (arrowhead) is missing, and the crown of the permanent first molar (long arrow) is normal. University of Helsinki. Panoramic radiographs (PRGs) from two patients were obtained from local health centers. One author (P.H.) performed all clinical examinations and radiographic assessments. PRGs were used to study agenesis and microdontia of permanent teeth. The prevalence of tooth agenesis and the number of missing teeth were recorded in 2 assessments, A1 and A2, using the following criteria. For assessment A1, third molars were excluded (52 patients with a mean age of 4.5 years at SCT were analyzed; the mean follow-up after SCT was 7.6 years). For assessment A2, third molars were included (29 patients with a mean age of 4.8 years at SCT were analyzed; the mean follow-up after SCT was 9.4 years). Recordings were performed taking into account the different calcification schedules of teeth. Because calcification in permanent incisors, canines, and first molars begins at birth or soon after birth, these teeth always were recorded. Agenesis of the first premolars was registered if no sign of tooth development was observed at age 5 years. The corresponding age for the second premolars and the permanent second molars was 6 years. Because the third molars develop late, and the mineralization schedule shows wide variation, 22 their agenesis was not recorded until age 12 years. All 55 patients were included in the microdontia study. Commonly used criteria for the microdontia assessment of PRGs are not available; therefore, recording was based on subjective visual judgment when the size of a tooth crown was 50% of the size considered normal (Fig. 2A,B). Of the two assessments, A1 and A2, only assessment A1 results are presented because only three third molars in one patient were found to be microdontic, and the results of the assessments were nearly identical. The underlying factors behind tooth agenesis and microdontia also were analyzed by dividing the patients into a TBI group and a non-tbi group and into 3 categories according to age at SCT: age 3.0 years (Group Y), ages years (Group M), and age 5.1 years (Group O). The numbers and percentages of patients who had received TBI in Group Y, Group M, and Group O were 10 of 16 patients (63%), 11 of 15 patients (73%), and 18 of 24 patients (75%), respectively (P 0.675). Statistical Analysis The Statistical Package for the Social Sciences (SPSS for Windows), version 10.0 (SPSS, Inc., Chicago, IL) was used in statistical analyses. The statistical significance of categoric variables between the groups were studied with the Pearson chi-square test or the Fisher exact test, and continuous variables were studied with the Mann Whitney U test. Linear regression analyses and Pearson correlation coefficients were used to study associations of TBI and age at SCT with the dental outcome. P values 0.05 were considered significant. RESULTS Tooth Agenesis Fifty-two SCT recipients were analyzed for agenesis of permanent teeth, which occurred in 16 of 52 patients (31%) in assessment A1 (third molars excluded). The most frequently missing teeth were second premolars (58%; 45 of 78 missing teeth), followed by second molars (28%), first premolars (10%), and upper lateral incisors (4%). Other permanent incisors as well as all canines and first molars were present. When all teeth were included (assessment A2; n 29 patients), tooth agenesis occurred in 19 of 29 SCT recipients (62%). The agenesis prevalence in third molars alone was 52% (15 of 29 patients). The prevalence of tooth agenesis did not differ statistically between the TBI group and the non-tbi group, although, in assessment A2, the prevalence among patients in the TBI group (72%) tended to be higher compared with the prevalence among patients

5 Tooth Agenesis and Microdontia after SCT/Hölttä et al. 185 TABLE 2 The Prevalence of Tooth Agenesis and the Mean Number of Missing Teeth in Pediatric Stem Cell Transplantation Recipients According to Total Body Irradiation and Age at Stem Cell Transplantation No. of patients with tooth agenesis/ no. of patients studied (%) Mean no. of missing teeth per patient (range) Variable A1 (n 52) A2 (n 29) A1 (n 52) A2 (n 29) Irradiation group TBI 12/38 (32) 13/18 (72) 1.7 (0 11) 4.1 (0 12) Non-TBI 4/14 (29) 5/11 (46) 0.8 (0 4) 1.1 (0 4) P value a b Age at SCT 3.0 yrs (Group Y) 10/13 (77) 5/6 (83) 3.9 (0 11) 4.7 (0 8) yrs (Group M) 6/15 (40) 7/9 (78) 1.8 (0 8) 4.9 (0 12) 5.1 yrs (Group O) 0/24 (0) 6/14 (43) 0 (0) 0.9 (0 4) P values a Group Y vs. Group M Group M vs. Group O c b b Group Y vs. Group O d d b A1: assessment with third molars excluded; A2: assessment with third molars included; TBI: total body irradiation; SCT: stem cell transplantation. a The Fisher exact test was used for categoric variables, and the Mann Whitney U test was used for continuous variables. b P c P d P in the non-tbi group (46%) (Table 2). Tooth agenesis was most prevalent in the youngest group (Group Y) (Table 2). In assessment A1, the difference was highly significant compared with Group O, in which no teeth were missing (P 0.001). The frequency of tooth agenesis in Group M also exceeded the frequency in Group O (P 0.002) (Table 2). In assessment A2, the agenesis prevalence was 83%, 78%, and 43%, respectively, in Group Y, Group M, and Group O. In Group O, the percentage indicated the prevalence of third molar agenesis, because no other teeth were missing (Table 2). The Number of Missing Teeth The mean number and range of missing teeth per SCT recipient was 1.5 teeth (range, 0 11 teeth) and 2.9 teeth (range, 0 12 teeth), respectively, in assessments A1 and A2. If only the patients with tooth agenesis were included, then the mean number of missing teeth was 4.8 teeth in assessment A1 (16 patients) and 4.7 teeth in assessment A2 (18 patients). A mean of 5.0 teeth (range, 2 11) were missing in all patients age 2 years at the time of SCT (n 7 patients). Patients in the TBI group were affected more severely than patients in the non-tbi group with regard to the number of missing teeth in assessment A2 (P 0.031) (Table 2, Fig. 3). The mean number of missing teeth was highest in Group Y: increasing from 3.9 teeth in assessment A1 to 4.7 teeth in assessment A2. In Group M, the increase was from 1.8 to 4.9 teeth. Both younger groups (Groups Y and M) differed significantly from Group O in both assessments (Table 2). Patient age at the time of SCT was found to have a negative correlation with the number of missing teeth (R 0.580; P 0.001). Age at SCT alone explained approximately 34% of the variation, irrespective of whether the third molars were included or excluded (assessment A1 or assessment A2). TBI was correlated negatively with the number of missing teeth in assessment A2 (R 0.439; P 0.017) but not in assessment A1 (R 0.164; P 0.246). Microdontia Microdontia was present in 44% of the 55 patients analyzed, including 41% of patients in the TBI group and in 50% of patients in the non-tbi group (assessment A1, third molars excluded; P 0.377) (Table 3). The most frequently microdontic permanent teeth were first premolars (46%; 28 of 64 microdontic teeth), followed by second premolars (26%), and second molars (23%). Other teeth seldom were involved. Microdontia prevalence was high in the younger SCT recipients, with rates of 75% in Group Y and 60% in Group M, percentages that were significantly higher compared with Group O (13%) (Table 3).

6 186 CANCER January 1, 2005 / Volume 103 / Number 1 FIGURE 3. The number of missing teeth (third molars included; assessment A2) is shown in individual patients (n 29 patients) according to their age at the time they underwent stem cell transplantation (SCT). Patients in the total body irradiation (TBI) group are indicated by dark triangles, and patients in the non-tbi group are indicated by white triangles. TABLE 3 The Prevalence of Microdontia and the Mean Number of Microdontic Teeth in Pediatric Stem Cell Transplantation Recipients According to Total Body Irradiation and Age at Stem Cell Transplantation Variable Assessment A1 (n 55 patients) No. of patients with microdontia/no. of patients studied (%) Irradiation group TBI 16/39 (41) 1.5 (0 12) Non-TBI 8/16 (50) 1.1 (0 3) P value a Age at SCT 3.0 yrs (Group Y) 12/16 (75) 1.9 (0 4) yrs (Group M) 9/15 (60) 2.8 (0 12) 5.1 yrs (Group O) 3/24 (13) 0.2 (0 2) P values a Group Y vs. Group M Group M vs. Group O b b Group Y vs. Group O c c Mean no. of microdontic teeth per patient (range) A1: assessment with third molars excluded; TBI: total body irradiation; SCT: stem cell transplantation. a The Fisher exact test was used for categoric variables, and the Mann Whitney U test was used for continuous variables. b P c P The Number of Microdontic Teeth The mean number of microdontic teeth was 1.4, with 1.5 teeth found in the TBI group and 1.1 teeth in the non-tbi group (P 0.821) (Table 3, Fig. 4). The lowest mean number of microdontic teeth (0.2) was recorded in Group O (P and P compared with Group Y and Group M, respectively) (Table 3). A negative correlation was found between the number of microdontic teeth and patient age at SCT (R 0.375; P 0.005). Patient age at SCT explained only 14% of the variation in the dependent variable. There was no correlation found between TBI and the number of microdontic teeth (R 0.077; P 0.578). Summarized Tooth Agenesis and/or Microdontia Summarized agenesis and/or microdontia were observed in 51% and 69%, respectively, of the SCT recipients for assessments A1 and A2. In assessment A1, the TBI and non-tbi groups were affected equally (51% vs. 50%, respectively). The difference was not significant in assessment A2 either (72% vs. 64%, respectively; P 0.694). A very high prevalence (94%) of summarized agenesis and/or microdontia was found in assessment A1 among the patients age 3.0 years at SCT (Group Y). All age groups (Groups Y, M, and O) differed significantly from one another (assessment A1: Group Y vs. Group M, P 0.033; Group Y vs. Group O, P 0.001; Group M vs. Group O, P 0.007) (Fig. 5). All patients in Group Y were affected in assessment A2; whereas, in the least affected group (Group O), only 50% of patients presented with agenesis and/or microdontia (Group Y vs. Group O, P 0.044). Group M with the angensis and/or microdontia prevalence of 78% did not differ from Groups Y and O. The mean number of affected teeth per patient appeared greater in the TBI group than in the non-tbi group (3.3 teeth vs. 1.8 teeth in assessment A1 and 5.2 teeth vs. 2.1 teeth in assessment A2), but the differences were not significant (assessment A1, P 0.447; assessment A2, P ). With regard to the mean number of affected teeth per patient in the 3 age groups, Groups Y and M scored significantly higher than Group O (in assessment A1: Group Y vs. Group O, P and Group M vs. Group O, P 0.005; in

7 Tooth Agenesis and Microdontia after SCT/Hölttä et al. 187 FIGURE 4. The number of microdontic teeth (third molars excluded; assessment A1) is shown in individual patients (n 55 patients) according to their age at the time they underwent stem cell transplantation (SCT). Patients in the total body irradiation (TBI) group are indicated by black triangles, and patients in the non-tbi group are indicated by white triangles. FIGURE 5. The percentages of patients with tooth agenesis (solid bars), microdontia (open bars), and summarized agenesis and/or microdontia (shaded bars) are shown for the youngest age group (Group Y) (age 3 years at the time of stem cell transplantation [SCT]; n 16 patients), the middle age group (Group M) (ages years at the time of SCT; n 15 patients), and the oldest age group (Group O) (age 5.1 years at the time of SCT; n 24 patients). Third molars were excluded (assessment A1). The prevalence of agenesis and microdontia in Group O differed significantly from that in Groups Y and M (agenesis: Group Y vs. Group O, P and Group M vs. Group O, P 0.002; and microdontia: Group Y vs. Group O, P and Group M vs. Group O, P 0.003). Each group differed from the others when agenesis and/or microdontia were summarized (Group Y vs. Group M, P 0.033; Group Y vs. Group O, P 0.001; and Group M vs. Group O, P 0.007). assessment A2: Group Y vs. Group O, P and Group M vs. Group O, P 0.016) (Fig. 6). DISCUSSION The results of the current study concerning pediatric SCT recipients emphasize the role of young age at SCT FIGURE 6. The mean number of missing (solid bars), microdontic (open bars), and summarized mean number of missing and/or microdontic (shaded bars) teeth per patient is shown in the youngest age group (Group Y) (age 3 years at the time of stem cell transplantation [SCT]; n 16 patients), the middle age group (Group M) (ages years at SCT; n 15 patients), and the oldest age group (Group O) (age 5.1 years at SCT; n 24 patients). Third molars were excluded (assessment A1). The mean number of affected teeth in Group O differed significantly from the other groups in terms of agenesis (Group Y vs. Group O, P and Group M vs. Group O, P 0.038), microdontia (Group Y vs. Group O, P and Group M vs. Group O, P 0.005), and summarized agenesis and/or microdontia (Group Y vs. Group O, P and Group M vs. Group O, P 0.005). as a risk factor for late dental adverse effects, which we defined as tooth agenesis and microdontia. Age seemed to be a stronger risk factor than TBI, although TBI caused additive impairment. Numerous signal molecules that are used repeatedly in complicated interactions between surface epithelium and the underlying mesenchyme mediate the

8 188 CANCER January 1, 2005 / Volume 103 / Number 1 developmental stages of teeth from initiation to morphogenesis and, furthermore, to the differentiation of tooth-specific cells, mineralization, and root formation. 23,24 Pediatric anticancer therapy may affect tooth development either by a direct toxicity toward the odontogenic cells, or by interfering with the delicate signaling network between ectoderm and mesenchyme or within one tissue layer. To our knowledge, the exact molecular mechanisms of the anticancer treatment that result in dental aberrations are not known. Proliferating preodontoblasts and their precursors usually are more sensitive than mature, functioning cells, as shown in biochemical and histologic studies of developing rat or hamster teeth. Disturbed odontogenesis has been reported after the administration of several chemotherapeutic agents, such as cyclophosphamide, vincristine, 28,29 actinomycin D, 30,31 doxorubicin, 32,33 and daunorubicin. 34 Typical, dosedependent findings include, for example, dentin niches (reduced thickness of the dentin wall); constrictions (reduction in the tooth width); and, at high doses, interrupted odontogenesis and degeneration of tooth-forming cells. The same agent that kills a cell at high doses by directly inhibiting a vital metabolic process, resulting in necrosis (straight toxicity), at lower doses, may trigger the apoptotic pathway. 31,34 The dental effects of radiation therapy are known well. although to our knowledge the minimum radiation dose that is harmful to developing teeth remains uncertain. Teeth exposed to a dose as low as 4 Gy have shown some abnormality in patients treated for soft tissue sarcomas of the head and neck. 35 The typical clinical consequences of irradiation on developing human teeth include microdontia, agenesis, and dwarfing of the roots. 36 Dentine niches identical to those observed after chemotherapy also have been described. 37,38 The dental consequences depend on the cell sensitivity and the radiation dose. The 31% prevalence of permanent tooth agenesis in the SCT group (third molars excluded) was 4-fold greater compared with the prevalence of 8% reported in a population of healthy Finns, 4 indicating a considerable effect of chemotherapy and/or radiotherapy on tooth development in children. According to the same study, the most frequently missing teeth were mandibular second premolars (42%), followed by maxillary second premolars (29%) and maxillary lateral incisors (19%). Missing of permanent second molars was very rare (1.4%). Compared with these percentages, there were striking differences in the distribution of missing permanent teeth among patients in the SCT group. For instance, only 4% of the missing teeth were maxillary lateral incisors, and a high percentage (28%) was recorded for second molars. These percentages, which clearly differ from the rates of genetic hypodontia, further stress that agenesis in the SCT group resulted mainly from the toxicity of anticancer therapy. A major difference between the healthy population and the SCT recipients also was seen in the third molar agenesis rate. Haavikko 4 reported that at least 1 of the third molars was missing in 21% of healthy Finns, compared with 52% among the SCT recipients in our study. The preparative regimens for SCT have caused tooth agenesis at a frequency that reportedly ranges from 56 58% 12,13 to 80%. 18 To our knowledge, the current series of SCT patients is the largest reported to date with an agenesis prevalence of 31%, which is lower than the prevalence reported previously. 12,13,18 The present results are not easy to compare with earlier studies due to variations in methodology, age of the patients, and treatment protocols. In the study by Näsman et al., 12 all patients who underwent SCT received TBI in a single fraction, whereas only 71% of patients in the current study received TBI that was fractionated in all but 1 patient. The underlying diseases also have been different (e.g., poor-risk NBL patients versus patients with mainly hematologic malignancies). 12,13,18 The current study included several diagnoses (Table 1). According to our study, TBI alone was not found to increase the prevalence of tooth agenesis significantly. Quantitatively more serious consequences were observed, especially when the third molars were included: The maximum number of missing teeth per patient was 4 teeth in the non-tbi group and 12 teeth in the TBI group (Table 2; Fig. 3). Young age at the time of SCT predisposes a tooth germ to permanent destruction prior to its mineralization due to the chemotherapy or chemoradiation therapy. A high prevalence of patients with tooth agenesis and high numbers of missing teeth per patient in the youngest age groups (Groups Y and M; age 5 years) indicate a high risk of developmental dental disturbances at young age. At age 5 years, the permanent teeth already are in the mineralization phase, except for the third molars, which begin their mineralization as late as at ages 9 10 years. 22 Consequently, children age 5 years at the time of SCT had an agenesis prevalence of 43% for the third molars, whereas all other teeth were present. Furthermore, the clinical significance of tooth agenesis was remarkable in the youngest children. The resulting gaps were situated more anteriorly on the alveolar ridge than the gaps that resulted from third molar agenesis. Primary molars often are able to substitute the missing premolars for some time; however,

9 Tooth Agenesis and Microdontia after SCT/Hölttä et al. 189 after their exfoliation, the mastication ability is decreased. The third molar agenesis alone did not appear to jeopardize occlusion in the oldest patients (Group O). Information regarding microdontia prevalence is scarce, and there are no solid diagnostic criteria for its assessment on PRGs. Measurements of horizontal distances on PRGs, such as the mesiodistal crown size affected in microdontia, are not considered as reliable as the vertical measurements. 39,40 Thus, radiographic microdontia assessments after pediatric anticancer treatment have been based on clinical judgment ,15,17,18 In the current study, only distinctly small teeth, maximally half of a normal size, were considered microdontic. The prevalence would have been higher than the rate of 44% found in our study if dental plaster casts could have been used for accurate crown size measurements. However, in young children such as those in the current study, many unerupted teeth, already seen in radiographs, are nonmeasurable from dental casts. Nevertheless, the microdontia prevalence among SCT recipients clearly exceeded the microdontia prevalence of 1.9% among healthy Japanese schoolchildren as measured from dental casts. 16 Näsman et al. 12 reported a microdontia prevalence of 68% in SCT recipients (n 19 patients) who had a mean age of 6.5 years at diagnosis and 75% in 16 SCT recipients who had a mean age of 6.3 years at the start of treatment. 13 These percentages are higher than what was found in the current study (44% among recipients with a mean age of 4.3 years at SCT), despite the fact that our patients were younger and more susceptible to dental aberrations. The subjective criteria for the microdontia assessment may explain the difference. Calculated from the material of 16 SCT recipients with a mean age of 7.1 years at SCT, the microdontia prevalence was 25%, 15 which is more in accordance with the prevalence rate determined according to our data. Although the prevalence of microdontia was found to be greatest in patients age 3.0 (Group Y) at the time of SCT, the most severely affected patients, with 4 microdontic teeth, were ages years (Group M) at the time of SCT and belonged to the TBI group. This is understandable because fewer teeth were missing in Group M compared with Group Y, leaving more teeth vulnerable to microdontia. Tooth agenesis and microdontia often occurred simultaneously after SCT. With the third molars included, 100% of patients age 3 years at the time of SCT, 78% of patients ages years, and 50% of the patients age 5.0 years were affected, often resulting in clinically significant occlusal disturbances. Proper dental care and rehabilitation, from the beginning of the treatment until adulthood, play one part in achieving a good quality of life for this increasingly larger group of children. REFERENCES 1. Schour I, Massler M. Studies in tooth development: the growth pattern of human teeth. Part II. J Am Dent Assoc. 1940;27: Byrd E. Incidence of supernumerary and congenitally missing teeth. J Dent Child. 1943;10: Hunstadbraten K. Hypodonti i det permanente tannsett. Nor Tannlaegeforen Tid. 1964;74: Haavikko K. Hypodontia of permanent teeth. An orthopantomographic study. Suom Hammaslaak Toim. 1971;67: Grahnen H. Hypodontia in the permanent dentition. A clinical and genetical investigation. Odont Revy. 1956;7: Pirinen S, Kentala A, Nieminen P, Varilo T, Thesleff I, Arte S. Recessively inherited lower incisor hypodontia. J Med Genet. 2001;38: Vastardis H, Karimbux N, Guthua SW, Seidman JG, Seidman CE. A human MSX1 homeodomain missense mutation causes selective tooth agenesis. Nat Genet. 1996;13: Stockton DW, Das P, Goldenberg M, D Souza RN, Patel PI. Mutation of PAX9 is associated with oligodontia. Nat Genet. 2000;24: Nieminen P, Arte S, Tanner D, et al. Identification of a nonsense mutation in the PAX9 gene in molar oligodontia. Eur J Hum Genet. 2001;9: Maguire A, Craft AW, Evans RG, et al. The long-term effects of treatment on the dental condition of children surviving malignant disease. Cancer. 1987;60: Sonis AL, Tarbell N, Valachovic RW, Gelber R, Schwenn M, Sallan S. Dentofacial development in long-term survivors of acute lymphoblastic leukemia. A comparison of three treatment modalities. Cancer. 1990;66: Näsman M, Bjork O, Soderhall S, Ringden O, Dahllof G. Disturbances in the oral cavity in pediatric long-term survivors after different forms of antineoplastic therapy. Pediatr Dent. 1994;16: Näsman M, Forsberg CM, Dahllof G. Long-term dental development in children after treatment for malignant disease. Eur J Orthod. 1997;19: Kaste SC, Hopkins KP, Jones D, Crom D, Greenwald CA, Santana VM. Dental abnormalities in children treated for acute lymphoblastic leukemia. Leukemia. 1997;11: Dahllof G, Barr M, Bolme P, et al. Disturbances in dental development after total body irradiation in bone marrow transplant recipients. Oral Surg Oral Med Oral Pathol. 1988; 65: Ooshima T, Ishida R, Mishima K, Sobue S. The prevalence of developmental anomalies of teeth and their association with tooth size in the primary and permanent dentitions of 1650 Japanese children. Int J Paediatr Dent. 1996;6: Kaste SC, Hopkins KP, Bowman LC, Santana VM. Dental abnormalities in children treated for neuroblastoma. Med Pediatr Oncol. 1998;30: Hölttä P, Alaluusua S, Saarinen-Pihkala UM, Wolf J, Nystrom M, Hovi L. Long-term adverse effects on dentition in children with poor-risk neuroblastoma treated with high-dose chemotherapy and autologous stem cell transplantation with or without total body irradiation. Bone Marrow Transplant. 2002;29:

10 190 CANCER January 1, 2005 / Volume 103 / Number Gustafsson G, Schmiegelow K, Forestier E, et al. Improving outcome through two decades in childhood ALL in the Nordic countries: the impact of high-dose methotrexate in the reduction of CNS irradiation. Nordic Society of Pediatric Haematology and Oncology (NOPHO). Leukemia. 2000;14: Lie SO, Abrahamsson J, Clausen N, et al. Treatment stratification based on initial in vivo response in acute myeloid leukaemia in children without Down s syndrome: results of NOPHO-AML trials. Br J Haematol. 2003;122: Saarinen UM, Wikstrom S, Makipernaa A, et al. In vivo purging of bone marrow in children with poor-risk neuroblastoma for marrow collection and autologous bone marrow transplantation. J Clin Oncol. 1996;14: Haavikko K. The formation and the alveolar and clinical eruption of the permanent teeth. An orthopantomographic study. Suom Hammaslaak Toim. 1970;66: Jernvall J, Thesleff I. Reiterative signaling and patterning during mammalian tooth morphogenesis. Mech Dev. 2000; 92: Thesleff I. Epithelial-mesenchymal signalling regulating tooth morphogenesis. J Cell Sci. 2003;116: Koppang HS. Histomorphologic investigations on the effect of cyclophosphamide on dentinogenesis of the rat incisor. Scand J Dent Res. 1973;81: Vahlsing HL, Kim SK, Feringa ER. Cyclophosphamide-induced abnormalities in the incisors of the rat. J Dent Res. 977;56: Näsman M, Hammarstrom L. Influence of the antineoplastic agent cyclophosphamide on dental development in rat molars. Acta Odontol Scand. 1996;54: Stene T. Effect of vincristine on odontoblasts in rat incisor. Scand J Dent Res. 1978;86: Lyaruu DM, van Duin MA, Bervoets TJ, Woltgens JH, Bronckers AL. Effects of vincristine on the developing hamster tooth germ in vitro. Connect Tissue Res. 1995;32: Adkins KF. The effect of actinomycin D on differentiation of odontoblasts in the rat. Arch Oral Biol. 1972;17: Lyaruu DM, van Duin MA, Bervoets TJ, Woltgens JH, Bronckers AL. Effects of actinomycin D on developing hamster molar tooth germs in vitro. Eur J Oral Sci. 1997; 105: Dahl JE. Immediate and delayed effects of repeated doxorubicin injections on rat incisor mesenchymal cells. Acta Odontol Scand. 1985;43: Karim AC, Woltgens JH, Bervoets TJ, Lyaruu DM, Bronckers AL. Effect of Adriamycin on hamster molar tooth development in vitro: 1. Morphological changes. Anat Rec. 1989;225: Lyaruu DM, van Duin MA, Bervoets TJ, Bronckers AL, Woltgens JH. Daunorubicin-induced pathology in the developing hamster molar tooth germ in vitro. Cancer Detect Prev. 1999;23: Fromm M, Littman P, Raney RB, et al. Late effects after treatment of twenty children with soft tissue sarcomas of the head and neck. Experience at a single institution with a review of the literature. Cancer. 1986;57: Bruce KW, Stafne EC. The effect of irradiation on the dental system as demonstrated by the roentgenogram. J Am Dent Assoc. 1950;41: Collet WK, Thonard JC. The effect of fractional radiation on dentinogenesis in the rat. J Dent Res. 1965;44: Adkins KF. The effect of 1200 R of x-radiation on dentinogenesis in the mandibular teeth of rats. Arch Oral Biol. 1967;12: Larheim TA, Svanaes DB. Reproducibility of rotational panoramic radiography: mandibular linear dimensions and angles. Am J Orthod Dentofacial Orthop. 1986;90: Welander U, Tronje G, McDavid W. Theory of rotational panoramic radiography. In: Langland O, Langlais R, Mc- David W, DelBalso A, editors. Panoramic radiology, 2nd ed. Philadelphia: Lea and Febiger, 1989:38 75.