Evaluating Access to Pediatric Cancer Care Centers of Children and Adolescents With Rare Tumors in Italy: The TREP Project
Guido Pastore, MD,1,2 Gian Luca De Salvo, MD,3 Gianni Bisogno, MD,4 Elisa Dama, MD,1 Alessandro Inserra, MD,5 Giovanni Cecchetto, MD,6 and Andrea Ferrari, MD7* on behalf of the TREP Group and the CSD of Epidemiology Biostatistics, AIEOP
Background. A national project focusing on rare malignant pediatric tumors (the TREP project) was launched in Italy in 2000. The present study compared the number of these tumors expected to be diagnosed in Italy with the number of cases actually enrolled in the TREP database in 2000-2006. Methods. The predicted number of cases was calculated from incidence data from the Italian network of cancer registries (AIRTum). Results. The TREP database included 336 patients under 18 years, that is, 261 children and 75 adolescents, as compared to 305 and 400 expected cases, respectively. For the 0- 14 years old age-group, the ratio of observed to expected cases was
1:1 for nasopharyngeal carcinoma, adrenocortical tumors, renal cell carcinoma, and gonadal non-germ-cell tumors, while for the 15-17-year old individuals there was a statistically significant under- reporting for all tumor types. Conclusions. Our study showed that the TREP project succeeded in registering and treating the vast majority of the patients under 15 years of age with rare pediatric tumors, demonstrating the feasibility of cooperative protocols even for rare diseases. Conversely, there was a large gap between those registered compared to those expected for adolescents. Pediatr Blood Cancer 2009;53:152-155. @ 2009 Wiley-Liss, Inc.
Key words: adolescents; childhood malignancies; epidemiology; rare pediatric tumors; TREP project
INTRODUCTION
Rare malignant pediatric tumors are essentially orphan diseases and make up a variety of neoplasms with particular biological and clinical characteristics. Some of these tumors are rare at any age, but most are particularly rare in childhood and adolescence [1].
In 2000, a nationwide Italian cooperative project, called the TREP (Tumori Rari in Età Pediatrica or Rare Tumours in Pediatric Age) project, was launched in Italy under the auspices of the Associazione Italiana Ematologia Oncologia Pediatrica (AIEOP), and in cooperation with the Società Italiana Chirurgia Pediatrica (SICP) to improve basic research on these tumors and their clinical management. The project aimed to establish a network of pediatric and adult oncologists, surgeons, pathologists and biologists, to pool clinical data, and develop diagnostic and therapeutic recommenda- tions for each of these rare tumors [2].
As a first step, the TREP group gave an arbitrary but pragmatic definition of “rare pediatric tumors,” defining them as solid malignancies characterized by annual incidence rates of less than 2 per million children (0-14 years) and not enrolled in specific clinical trials. The list included: nasopharyngeal carcinoma, adrenocortical tumors, pleuro-pulmonary blastoma (and other lung tumors), carcinoid tumors, cutaneous melanoma, renal carcinoma, pancreatoblastoma (and other pancreatic exocrine tumors), gonadal non-germ-cell tumors (ovary/testis), pheochromocytoma and para- ganglioma, thyroid carcinoma, salivary gland tumors, breast carcinoma, carcinoma of the gastrointestinal tract, and carcinoma of the thymus [2]. Therefore, the list excluded entities as renal rhabdoid tumors, that are registered in the national Wilms study; hepatoblastoma and malignant germ cell tumors, that have their own protocols; rare histotypes of soft part sarcomas, that are covered by the cooperative study on soft tissue sarcomas, as well as rare bone tumors; and other tumors whose incidence parallels that of rare pediatric tumors but were already included in clinical trials. Central nervous system tumors and hematological malignancies were excluded from the TREP list.
The TREP group has already published a report on its first 6 years of activity. Here, we compare the number of these rare pediatric tumors expected to be diagnosed in Italy (based on incidence data
from Italian network of well-established population-based cancer registries [AIRTum]) [3] and the number of patients enrolled in the TREP project between 2000 and 2006, with a view to appraising the nationwide diffusion and efficacy of the TREP project. The study analyzed the differences in the patterns of referral of children (0-14 years old) and adolescents less than 18 years (15-17 years old) with rare malignant tumors.
STUDY DESIGN AND METHODS
The TREP database contained 336 patients from 0 to 17 years old (261 children and 75 adolescents) diagnosed between January 1st, 2000 and December 31st, 2006, enrolled from 36 Italian centers of pediatric oncology and pediatric surgery with malignant cancers. The number of cases of rare malignant cancers in Italy for 2000- 2006 was predicted by applying the age-specific incidence rates recorded by the Italian network of cancer registries (AIRTum) between 1988 and 2002 to the general Italian population [3]. The AIRTum database contains data from 22 general registries and 3 specialized registries (2 on childhood and adolescent cancers, 1 on female breast cancer) and covers 32.9% of the Italian resident
—
1Childhood Cancer Registry of Piedmont. Cancer Epidemiology Unit, CPO Piemonte, CeRMS, S. Giovanni Hospital, University of Torino, Torino, Italy; 2Division of Pediatrics, Department of Medical Sciences, University of Eastern Piedmont, Novara, Italy; 3Clinical Trials and Biostatistic Unit, IRCCS Istituto Oncologico Veneto, Padova, Italy; 4Division of Hematology-Oncology, Pediatric Department, University of Padua, Padova, Italy; 5Pediatric Surgery, IRCCS Ospedale Pediatrico Bambino Gesù, Roma, Italy; Pediatric Surgery, University of Padova, Padova, Italy; 7Pediatric Oncology Unit, Fondazione IRCCS Istituto Nazionale Tumori, Milano, Italy
Grant sponsor: “Fondazione CARIPARO”; Grant sponsor: Italian Association for Cancer Research; Grant sponsor: Compagnia San Paolo/FIRMS.
*Correspondence to: Andrea Ferrari, Pediatric Oncology Unit, Fondazione IRCCS Istituto Nazionale Tumori, Via G. Venezian, 1- 20133 Milano (MI), Italy. E-mail: andrea.ferrari@istitutotumori.mi.it
Received 22 September 2008; Accepted 4 March 2009
@ 2009 Wiley-Liss, Inc.
DOI 10.1002/pbc.22049
population in the 0-14 years old age-group, and 26.9% of the 15-19 years old age-group. The database is updated regularly and subject to quality controls developed according to both the most recent International parameters and a specific software developed by the Association itself. All diagnoses are coded using the ICD-O-2 classification including only malignant cancer [4].
To provide more stable predictions of the expected number of cases based on AIRTum, we calculate, as a control group, the expected number of 0-14 years cases derived from the incidence rates of the United Kingdom National Childhood Tumor Registry (UK NCTR) [5], based on childhood populations approximately fivefold larger than that covered by the AIRTum. Though geographical differences in the incidence of these rare tumors might be relevant, this comparison did not show major differences (e.g., for nasopharyngeal carcinoma expected cases every year were 2 according to AIRTum and 1.9 according to the UK NCTR data, for adrenocortical tumor they were 2.4 and 1.7, for melanoma 8.6 and 10.5, or for pancreatic tumors 0.26 and 0.28, respectively), with the exception of thyroid carcinoma (13 vs. 5.5, respectively), probably due to differences in environmental iodine levels [6]. For adolescent patients, we used as control the incidence rates derived from the Automated Childhood Cancer Information System (ACCIS) project that is based on a sevenfold larger adolescent population at risk than that covered by the AIRTum [7,8]. The statistical significance of the difference between the observed (O) and expected (E) number of cases was tested by calculating confidence intervals (CI) around the O/E ratio according to the exact procedure based on the Binomial/ Incomplete Beta proposed by Silcocks [9].
RESULTS
Table I shows the incidence rates and the number of children and adolescents observed (O) in the TREP project during 2000-2006, the number of cases expected (E) according to AIRTum and the O/E ratio. The number of rare malignant tumors expected to be diagnosed in Italy between 2000 and 2006 was 305 in children and 400 in adolescents, while the numbers of cases actually observed by the TREP were 261 and 75, respectively. On average, 44 children and 57 adolescents are expected to be diagnosed with rare
pediatric tumors each year in Italy according to the AIRTum incidence rates.
Among the children enrolled in the TREP project, thyroid carcinoma (19.2%), carcinoma of appendix (18.8%), and gonadal non-germ-cell tumors (10.3%) were the larger groups, while the most common tumor types among adolescents were thyroid carcinoma (42.7%), nasopharyngeal carcinoma (17.3%), and cutaneous melanoma (10.7%).
Based on the AIRTum rates, the O/E ratio for the subset of 0- 14 year olds was around 1:1, ranging between 0.66 and 1.67, for nasopharyngeal carcinoma, adrenocortical tumors, renal cell carcinoma, gonadal non-germ-cell tumors, and carcinoma of the gastro-intestinal tract. A statistically significant deficiency in the number of cases observed in this age-group (O/E ratio lower than 1) was apparent for cutaneous melanoma, thyroid carcinoma, and salivary gland tumors (observed cases were from a quarter to half of the expected). The O/E ratio was significantly higher than 1:1 for pleuro-pulmonary blastoma, carcinoid of the appendix, pancreato- blastoma, and pheochromocytoma/paraganglioma.
In the subset of 15-17-year old patients, there was an under- reporting of all tumor types, except for adrenocortical tumors and, in particular, for nasopharyngeal. The under-reporting was statistically significant, however, for carcinomas of appendix, cutaneous melanoma, gonadal non-germ-cell tumors, thyroid carcinoma, and salivary gland tumors (P <0.05).
The incidence rate for all tumor types but malignant melanoma was lower the two cases for million among children younger that 15 years, while was higher the two cases for million for 15-17 years old adolescent for thyroid carcinoma, malignant melanoma, gonadal tumors, and carcinoma of salivary glands (Table II).
DISCUSSION
The present study focused on the epidemiology of rare tumors in childhood and adolescence and the activity of the Italian national cooperative project on these tumors. One of the possible limitations of the study is the difficulty of interpreting epidemiological evidences concerning rare diseases. This is particularly relevant for rare tumors in the 0-14 years age-group, whose incidence rate is
| Cancer types | 0-14 | 15-17 | ||||
|---|---|---|---|---|---|---|
| O | E | O/E (95% CI) | O | E | O/E (95% CI) | |
| Nasopharyngeal carcinoma | 19 | 14 | 1.36 (0.65-2.92) | 13 | 13 | 1.00 (0.43-2.34) |
| Adrenocortical tumors | 23 | 17 | 1.35 (0.69-2.70) | 2 | 4.4 | 0.45 (0.04-2.99) |
| Pleuro-pulmonary blastoma (and other lung tumors) | 13 | 1.9 | 6.84 (1.51-67.67) | 1 | 0.0 | — |
| Carcinoids of appendix | 49 | 25 | 1.96 (1.19-3.31) | 7 | 30 | 0.23 (0.09-0.54) |
| Cutaneous melanoma | 19 | 62 | 0.31 (0.17-0.52) | 8 | 107 | 0.07 (0.03-0.15) |
| Renal carcinoma | 20 | 24 | 0.83 (0.44-1.57) | 2 | 8.7 | 0.23 (0.02-1.12) |
| Pancreatoblastoma (and other pancreatic exocrine tumors) | 11 | 1.8 | 6.11 (1.26-67.16) | 1 | 4.4 | 0.23 (0.00-2.17) |
| Gonadal non-germ-cell tumors (ovary/testis) | 27 | 41 | 0.66 (0.39-1.10) | 3 | 39 | 0.08 (0.02-0.24) |
| Pheochromocytoma and paraganglioma | 18 | 3.6 | 5.00 (1.58-22.21) | 0 | 0.0 | — |
| Thyroid carcinoma | 50 | 91 | 0.55 (0.38-0.78) | 32 | 159 | 0.20 (0.13-0.30) |
| Salivary gland tumors | 5 | 22 | 0.23 (0.07-0.62) | 1 | 26 | 0.04 (0.00-0.23) |
| Breast carcinoma | 1 | 0.0 | — | 1 | 0 | — |
| Carcinoma of the gastrointestinal tract | 3 | 1.8 | 1.67 (0.18-23.40) | 3 | 8.7 | 0.34 (0.06-1.40) |
| Carcinoma of the thymus | 3 | 0.0 | — | 1 | 0 | — |
| Cancer types | 0 | 1-4 | 5-9 | 10-14 | 15-17 |
|---|---|---|---|---|---|
| Nasopharyngeal carcinoma | 0.00 | 0.00 | 0.00 | 0.73 | 1.08 |
| Adrenocortical tumors | 2.04 | 0.38 | 0.10 | 0.09 | 0.36 |
| Pleuro-pulmonary blastoma (and other lung tumors) | 0.00 | 0.13 | 0.00 | 0.00 | 0.00 |
| Carcinoids of appendix | 0.00 | 0.00 | 0.20 | 1.10 | 2.51 |
| Cutaneous melanoma | 0.00 | 0.64 | 0.30 | 2.38 | 8.78 |
| Renal carcinoma | 1.02 | 0.13 | 0.50 | 0.46 | 0.72 |
| Pancreatoblastoma (and other pancreatic exocrine tumors) | 0.00 | 0.00 | 0.00 | 0.09 | 0.36 |
| Gonadal non-germ-cell tumors (ovary/testis) | 1.02 | 0.51 | 0.40 | 1.10 | 3.23 |
| Pheochromocytoma and paraganglioma | 0.00 | 0.00 | 0.00 | 0.18 | 0.00 |
| Thyroid carcinoma | 0.00 | 0.00 | 0.89 | 3.75 | 13.08 |
| Salivary gland tumors | 0.00 | 0.00 | 0.00 | 1.10 | 2.15 |
| Breast carcinoma | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| Carcinoma of the gastrointestinal tract | 0.00 | 0.00 | 0.00 | 0.09 | 0.72 |
| Carcinoma of the thymus | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
definitely smaller than that in 15-17 years old group (for all the tumors in the TREP list, the incidence is 15.8 [95% CI 7.43; 28.92] and 32.1 [CI 22.04; 47.81]). However, the prediction of the expected number according to AIRTum was generally confirmed by using, as control group, UK NCTR data [5] for children and ACCIS data [7,8] for adolescents. Further limitations of the study came from the possible diagnostic and coding inconsistencies, and different registration policies, as probably reflected by the higher than expected O/E ratio found for some histological type, that is, pleuro- pulmonary blastoma, carcinoma of the appendix, pancreatoblas- toma, and pheochromocytoma/paraganglioma. Pleuro-pulmonary blastomas might be registered as sarcomas and it has already been emphasized that the incidence of this tumor may well be underestimated in population-based registries [5].
There may also be an overlap between benign, borderline, and malignant neoplasms for some of the entities included in the TREP list [5]. This may be particularly true for carcinoma of the appendix and pheochromocytoma, but probably also for adrenocortical tumors and gonadal non-germ-cell tumors. For carcinoma of the appendix and pheochromocytoma, it is possible that some benign or borderline cases have been registered in addition to malignant cases in the TREP database, giving rise to the high O/E ratio observed for 0-14 years cases.
Therefore, the first point of our study may be how difficult it is to establish epidemiological analysis on rare pediatric tumors. However, some results might be considered more reliable than other and should deserve consideration: our study showed that the TREP project succeeded in registering and treating the vast majority of the patients with rare pediatric tumors under 15 years of age (261 patients were observed, as opposed to 305 expected), particularly as concerns the cases of nasopharyngeal carcinoma, adrenocortical tumors, and renal cell carcinoma. Conversely, there was a large gap between the numbers registered (75 cases) and the expected (400 cases) for adolescents: in this age-group, it was noteworthy the adequate patient enrolment recorded for naso- pharyngeal carcinoma (O/E 1.00).
The results obtained for the 0-14-year old patients demonstrated the feasibility of cooperative protocols even for rare diseases, in particular when a support of an expert advisory network and a centralized histological review is offered to the Centers. These findings compare favorably with those observed for the more
common pediatric malignancies. In fact, the hospital-based registry collecting demographic and clinical information of patients with tumor diagnosed in the AIEOP Centers showed that, for the group of pediatric malignancies as a whole, the O/E ratio was 0.77 and 0.18 for the 0-14- and 15-18-year old patients, respectively [10].
However, the significance of the gap between observed and expected cases among adolescents is probably quite different for the common and the rare pediatric malignancies. If various evidences would suggest that adolescents with leukemia, Ewing sarcoma, rhabdomyosarcoma, and other pediatric tumors benefit for being treated in pediatric oncology setting [11-15], it is a reasonable course of action that a large proportion of adolescents with rare tumors were referred to adult centers. Though some findings would suggest possible biological and clinical differences in some epithelial tumors according to age [16-18], most rare pediatric tumors are relatively common in adults and oncologists or general surgeons, as well also dermatologists (for melanoma), otorhino- laryngologists (for nasopharingeal carcinoma and thyroid carci- noma), and endocrinologists (for adrenocortical tumors and thyroid carcinoma), used to dealing with adults are bound to have more experience of treating such neoplasms. Nevertheless, the TREP project demonstrated also the feasibility of a fruitful cooperation between pediatric oncologists and experts on adult-type tumors.
Another suggestion coming from the population-based registries concerns how rare pediatric tumors are defined, with a view to the validation of our practical definition, which assumes an annual incidence of less than 2 cases per million population. Judging from the AIRTum database [3], the incidence of all the TREP tumors is less than 2 per million population a year among 0-14 year olds. Differently, for the adolescents (15-17 years of age), some histotypes (i.e., carcinoid tumors, cutaneous melanoma, gonadal tumors, thyroid cancer, and salivary gland tumors) have an incidence higher than 2 per million population a year and therefore should not be considered as rare tumors (Table II). In particular, the incidence rates for cutaneous melanoma and thyroid carcinoma were 8.78 and 13.08 per million person/year, respectively. Many tumors that are considered rare by pediatric oncologists are commonly seen at adult oncology units, where the medical oncologists rarely encounter patients with tumors considered common in the pediatric setting (i.e., neuroblastoma, rhabdomyo- sarcoma, and Wilms tumors). The implications of rare tumor
management remain the same. It is difficult to perform clinical trials, few clinical and biological data are available, it is hard to establish shared treatment guidelines, dedicated financial resources are limited. However, pediatric oncologists and medical oncologists working with adults clearly mean different things when they talk about rare tumors. Although rare tumors should be defined as such on the basis of their incidence alone, it is also true that the practical definition adopted for the TREP project adequately categorizes orphan tumors in the pediatric setting, ultimately with a view to proposing a tool capable of filling a void.
ACKNOWLEDGMENT
The TREP project is supported by a grant from the “Fondazione CARIPARO”. This work was partially supported by the Italian Association for Cancer Research and the Compagnia San Paolo/ FIRMS. The Authors wish to thank Ms. Elisa Mancini for data management and the Italian Centers involved in the TREP project. The Authors wish to thank the AIRTUM network for providing incidence data, Franco Merletti, Milena M. Maule, and Charles Stiller for appreciated comments and suggestions.
REFERENCES
1. Gurney JG, Bondy ML. Epidemiology of childhood cancer. In: Pizzo PA, Poplack DG, editors. Principles and practice of pediatric oncology, 5th edition. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. pp. 1-13.
2. Ferrari A, Bisogno G, De Salvo GL, et al. The challenge of very rare tumours in childhood: The Italian TREP project. Eur J Cancer 2007;43:654-659.
3. AIRT Working Group. I tumori infantili in Italia. Epidemiol Prev 2008;32:1-112.
4. Percy C, Van Holten V, Muir C, editors. International classification of diseases for oncology, 2nd edition. Geneva: World Health Organization; 1990.
5. Stiller CA, editor. Childhood cancer in Britain: Incidence, survival and mortality. Oxford: Oxford University Press; 2007.
6. Steliarova-Foucher E, Stiller CA, Pukkala E, et al. Thyroid cancer incidence and survival among European children and adolescents (1978-1997): Report from the Automated Childhood Cancer Information System project. Eur J Cancer 2006;42:2150-2169.
7. ACCIS (Automated Childhood Information System), Pass 1.01, Developed by the IARC (Internationala Agency for Research on
Cancer), Lyon (F), November 2002, using European ACCIS database updated in November 2002. http://www-dep.iarc.fr/ accis.htm.
8. Stiller CA, Desandes E, Danon SE, et al. Cancer incidence and survival in European adolescents (1978-1997). Report from the Automated Childhood Cancer Information System project. Eur J Cancer 2006;42:2006-2018.
9. Silcocks P. Estimating confidence limits on a standardised morta- lity ratio when the expected number is not error-free. J Epidemiol Community Health 1994;48:313-331.
10. Ferrari A, Dama E, Pession A, et al. Adolescents with cancer in Italy: Entry into the national cooperative pediatric oncology group AIEOP trials. Eur J Cancer 2009;45:328- 334.
11. Boissel N, Auclerc MF, Lheritier V, et al. Should adolescents with acute lymphoblastic leukemia be treated as old children or young adults? Comparison of the French FRALLE-93 and LALA-94 trials. J Clin Oncol 2003;21:774-780.
12. de Bont JM, Van Der Holt B, Dekker AW, et al. Significant difference in outcome for adolescents with acute lymphoblastic leukemia treated on pediatric versus adult ALL protocols in The Netherlands. Leukemia 2004;18:2032-2053.
13. Ramanujachar R, Richards S, Hann I, et al. Adolescents with acute lymphoblastic leukaemia: Emerging from the shadow of paediatric and adult treatment protocols. Pediatr Blood Cancer 2006;47:748- 756.
14. Paulussen S, Ahrens S, Juergens HF. Cure rates in Ewing tumor patients aged over 15 years are better in pediatric oncology units. Results of GPOH CESS/EICESS studies. Proc Am Soc Clin Oncol 2003;22:816.
15. Ferrari A, Dileo P, Casanova M, et al. Rhabdomyosarcoma in adults. A retrospective analysis of 171 patients treated at a single institution. Cancer 2003;98:571-580.
16. Ferrari A, Bono A, Baldi M, et al. Does melanoma behave differently in younger children than in adults? A retrospective study of 33 cases of childhood melanoma from a single institution. Pediatrics 2005;115:649-654.
17. Ferrari A, Rognone A, Casanova M, et al. Colorectal carcinoma in children and adolescents: The experience of the Istituto Nazionale Tumori of Milan, Italy. Pediatr Blood Cancer 2008;50:588- 593.
18. Ramphal R, Pappo A, Zielenska M, et al. Pediatric renal cell carcinoma: Clinical, pathologic, and molecular abnormalities associated with the members of the MiT Transcrition Factor family. Am J Clin Pathol 2006;126:349-364.