IJC International Journal of Cancer
Second malignancies after childhood noncentral nervous system solid cancer: results from 13 cancer registries Milena Maule1*, Ghislaine Sce´lo2*, Guido Pastore1, Paul Brennan2, Kari Hemminki3,4, Jorgen H. Olsen5, Elizabeth Tracey6, Eero Pukkala7,8, Elisabete Weiderpass9,10,11,12, David H. Brewster13, Sharon Tamaro14, Kee-Seng Chia15,16, Vera Pompe-Kirn17, Erich V. Kliewer14,18,19, Jon M. Tonita20, Carmen Martos21, Jon G. Jonasson22,23, Franco Merletti1 and Paolo Boffetta2,24,25 1
Childhood Cancer Registry of Piedmont, Cancer Epidemiology Unit, CPO Piemonte, CeRMS, University of Turin, Turin, Italy International Agency for Research on Cancer (IARC), Lyon, France 3 Division of Molecular Genetic Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany 4 Center for Primary Health Care Research, University of Lund, Malmo¨, Sweden 5 Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark 6 New South Wales Cancer Registry, Eveleigh, New South Wales, Australia 7 Finnish Cancer Registry, Institute for Statistical and Epidemiological Cancer Research, Helsinki, Finland 8 School of Public Health, University of Tampere, Tampere, Finland 9 The Cancer Registry of Norway, Oslo, Norway 10 Department of Community Medicine, Tromso, Norway 11 Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden 12 Samfunded Folkhalsan, Helsinki, Finland 13 Scottish Cancer Registry, Information Services Division, NHS National Services Scotland, Edinburgh, Scotland, UK 14 British Columbia Cancer Agency, Vancouver, British Columbia, Canada 15 Center for Molecular Epidemiology, Singapore 16 Singapore Cancer Registry, Singapore 17 Cancer Registry of Slovenia, Institute of Oncology, Ljubljana, Slovenia 18 Epidemiology and Cancer Registry, CancerCare Manitoba, Winnipeg, Canada 19 Community Health Sciences, University of Manitoba, Winnipeg, Canada 20 Saskatchewan Cancer Agency, Regina, Saskatchewan, Canada 21 Cancer Registry of Zaragoza, Health Department of Aragon Government, Zaragoza, Spain 22 Icelandic Cancer Registry, Icelandic Cancer Society, Reykjavik, Iceland 23 Faculty of Medicine, University of Iceland, Reykjavik, Iceland 24 The Tisch Cancer Institute, Mount Sinai School of Medicine, New York, NY 25 International Prevention Research Institute, Lyon, France
Epidemiology
2
Children diagnosed with noncentral nervous system solid cancers (NCNSSC) experience several adverse late effects, including second malignant neoplasm. The aim of our study was to assess the risk of specific second malignancies after a childhood NCNSSC. Diagnosis and follow-up data on 10,988 cases of NCNSSC in children (0–14 years) were obtained from 13 registries. Standardized incidence ratios (SIRs) with 95% confidence intervals (CI) and cumulative incidence of second malignancies were computed. We observed 175 second malignant neoplasms, yielding a SIR of 4.6, 95% CI: 3.9–5.3. When considering second
Key words: second malignant neoplasm, childhood solid cancer Abbreviations: AER: absolute excess risk; CI: confidence interval; IARC: International Agency for Research on Cancer; IACR: International Association of Cancer Registries; ICCC: International Classification of Childhood Cancer; ICD-O-2: International Classification of Diseases for Oncology, 2nd edition; ICD-9: International Classification of Diseases, Version 9; IQR: interquartile range; (131)I-MIBG: (131)Imetaiodobenzylguanidine; NCNSSC: noncentral nervous system solid cancers; SEER: surveillance, epidemiology and end results; SIR: standardized incidence ratio; US: United States Additional Supporting Information may be found in the online version of this article. *M.M. and G.S. contributed equally to this work Grant sponsors: US National Cancer Institute; Grant number: CA101442-02; Grant sponsors: Compagnia di San Paolo, Fondazione Internationale di Ricerca in Medicina Sperimentale, Italian Association for Cancer Research DOI: 10.1002/ijc.26135 History: Received 19 Nov 2010; Accepted 28 Mar 2011; Online 25 Apr 2011 Correspondence to: Milena Maule, Cancer Epidemiology Unit, University of Turin, Via Santena 7, 10126, Turin, Italy, Tel.: +390116334628, Fax: þ390116334664, E-mail:
[email protected]
C 2011 UICC Int. J. Cancer: 129, 1940–1952 (2011) V
Maule et al.
1941
cancers with at least 10 occurrences, highest relative risks were found for second malignant bone tumors (SIR 5 26.4, 16.6– 40.0), soft tissue sarcomas (SIR 5 14.1, 6.7–25.8) and myeloid leukemia (SIR 5 12.7, 6.3–22.8). Significant increased risks for all malignancies combined were observed after sympathetic nervous system tumors (SIR 5 11.4, 5.2–21.6), retinoblastomas (SIR 5 7.3, 5.4–9.8), renal tumors (SIR 5 5.7, 3.8–8.0), malignant bone tumors (SIR 5 5.6, 3.7–8.2), soft tissue sarcomas (SIR 5 4.7, 3.2–6.8), germ-cell, trophoblastic and other gonadal neoplasms (SIR 5 2.5, 1.1–4.9), carcinomas and other malignant epithelial neoplasms (SIR 5 2.2, 1.4–3.3). The highest risk of a second malignancy of any type occurred 5 to 9 years after NCNSSC (SIR 5 9.9, 6.8–13.9). The cumulative incidence of second malignancies 10 years after the first neoplasm was eight times higher among NCNSSC survivors than in the general population, with the absolute difference between observed and expected cumulative incidence still increasing after 50 years of follow-up. Children who survived a NCNSSC experience a large increased risk of developing a new malignancy, even many years after their initial diagnosis.
C 2011 UICC Int. J. Cancer: 129, 1940–1952 (2011) V
the type of childhood NCNSSC, calendar period of NCNSSC diagnosis and time elapsed since NCNSSC diagnosis. Particularly, by means of an extensive follow-up that for some cancer types exceeds 50 years, our study provides insight into the long-term occurrence of second malignancies.
Methods To conduct a systematic analysis of second primary cancers, an international multicentric study was set up by the International Agency for Research on Cancer (IARC) in 2002 involving large cancer registries that had been in operation for at least 20 years. Nineteen registries whose data had been included consistently in volumes of Cancer Incidence in Five Continents12 were invited to participate. A similar analysis of data from US cancer registries was in progress in the US at that time, and these were therefore not included in the analysis. Inclusion in consecutive versions of Cancer Incidence in Five Continents12 was taken as a measure of quality of registration, including a high level of morphological verification and low proportion of cancers identified only through death certificates. Of the initial group of 19 registries, 15 confirmed that the project was feasible and provided all necessary data. Two registries were subsequently excluded because of discrepancies in the observed rates of second primaries, leaving 13 registries in the current analysis. These registries are located in British Columbia, Manitoba and Saskatchewan (Canada), Denmark, Finland, Iceland, New South Wales (Australia), Norway, Scotland (United Kingdom), Singapore, Slovenia, Sweden and Zaragoza (Spain). All the registries were able to provide high-quality data, with no relevant differences in data collection and with similar and high proportion of completeness. Follow-up for vital status uses both active and passive methods in eight registries (Saskatchewan, New South Wales, Denmark, Finland, Iceland, Slovenia, Zaragoza and Sweden) and relies exclusively on passive methods (linkage with mortality registries) in five registries (British Columbia, Manitoba, Norway, Singapore and Scotland). The population covered by these cancer registries was 47 million in the 1990s. The registries contributed data for different time periods in the years 1943 to 2000, with a median observation period of 32 years. This dataset partially overlaps with the populations included in previous studies from the Nordic countries9,11,13 and Slovenia.14
Epidemiology
Noncentral nervous system solid cancers (NCNSSC) account for 30% of all childhood and adolescent cancers.1 In the United States (US), in the period 2002 to 2006, their overall incidence rate was 47.8 per million person-year among 0 to 14 year-old children but varied considerably with age and histologic type. The highest incidence rates were observed among children in their first years of life for most types, including neuroblastoma, retinoblastoma, renal tumors, hepatic tumors, soft tissue sarcomas and germ cell tumors, whereas incidence rates were highest among 10- to 14-yearold children for malignant bone tumors and carcinomas.2 Five-year survival rates have significantly improved in the last three decades, reaching 97% in the US for children diagnosed with retinoblastoma (the type with the best prognosis) in the period 1999–2005 and 64.7% for those with hepatic tumors (the type with the worst prognosis).2 The improvement is due to several advances both in diagnostic techniques, leading to earlier diagnosis and treatment strategies, that involve an integrated approach of surgery, radiotherapy and chemotherapy. The increasingly large and ageing population of long-term childhood cancer survivors are at increased risk for several adverse late effects, that manifest months to many years after completion of cancer treatment, and include the development of a second malignancy. The main known risk factors for second cancers are the type and dose of the therapy received for the treatment of the childhood cancer, host factors such as genetic predispositions and the interaction between these two. The incidence of second cancers after a malignant neoplasm in the pediatric ages has been estimated in several studies,3–10 especially for the most frequent malignancies (lymphohaematopoietic and brain neoplasms) and for short or intermediate follow-up times. On the other hand, information on the longterm risk of second malignancies after the rarest types of childhood cancer is more limited because large groups of patients have to be followed up for more than three or four decades.3,4,9,11 We have estimated the risk of second malignancies after childhood NCNSSC by taking advantage of a unique source of data consisting of a large cohort of survivors from 13 population-based registries located in Europe, North America, Asia and Oceania. The aim of our study is to investigate the magnitude of risk and type of second cancer in relation to
1942
Second malignant neoplasms after childhood solid cancer
Table 1. Distribution of childhood noncentral nervous system solid cancers (NCNSSC) and number of second malignant neoplasms, by selected characteristics Second cancer status1
Characteristics
Free from 2nd cancer
Have a 2nd cancer
N
N
%
%
Total N
%
Sex Females
5,271
48.7
98
56.0
5,369
48.9
Males
5,542
51.3
77
44.0
5,619
51.1
825
7.6
14
8.0
839
7.6
Age at first cancer registration, years
