Annals of Oncology Advance Access published online on January 30, 2008
Annals of Oncology, doi:10.1093/annonc/mdn003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The incidence and risk of second primary cancers in patients with nasopharyngeal carcinoma: a population-based study in Taiwan over a 25-year period (1979–2003)
1 Department of Public Health and Biostatistics Consulting Center, School of Medicine, Chang Gung University, Tao-Yuan, Taiwan
2 Department of Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, Ohio, USA
3 Department of Hematology and Oncology, Chang Gung Memorial Hospital, Chiayi
4 Department of Nursing, Chang Gung Institute of Technology, Chiayi, Taiwan
* Correspondence to: Dr. K.-D. Lee, Department of Hematology and Oncology, Chang Gung Memorial Hospital, 6 West Sec, Chia-Pu Road, Pu-Tz city, Chiayi 613, Taiwan. Tel: +886-5-3621000 ext 2772; Fax: +886-5-3623781; E-mail: kdlee{at}adm.cgmh.org.tw
 |
Abstract |
|---|
 Top Abstract introductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Background: Very few reports are available on the incidence and risk of second primary cancers in nasopharyngeal carcinoma (NPC) cases, and most of these are single-institution reviews with relatively small case numbers and short follow-up.
Patients and methods: A population-based study was conducted. We quantified standardized incidence ratios (SIRs) and cumulative incidence of second cancers among 23 639 individuals with initial diagnoses of NPC.
Results: We found a 24% increased risk of second cancers in NPC patients compared with the general population [SIR = 1.24, 95% confidence interval 1.15–1.33]. Elevated SIRs were observed in the following second primary cancers: oral/pharyngeal, salivary gland, sarcoma, skin and leukemia/lymphoma. The cumulative incidence >10 years was 3.26%. The risk was higher in younger patients, especially those <40 years old. After diagnosis of second cancers, the median survival time was 1.7 years.
Conclusions: This is the largest population-based study to date from a high-incidence area. We found that NPC is associated with an increased risk of second malignancies, which had a negative impact on the survival of patients who survived NPC.
cumulative incidence, nasopharyngeal carcinoma, second primary cancer
|
introduction |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Nasopharyngeal carcinoma (NPC) is rare in most parts of the world but is uniquely prevalent in southeast Asia. The reported rates per 100 000 persons are 20–30 males and 7–13 females in southeastern China (1998–2002) [1] and 20.2 males and 7.8 females in Hong Kong (1995–1999) [2]. Irrespective of geographical distribution, Epstein–Barr virus (EBV) is the most important etiological factor. Recent studies in Taiwan have shown that circulating EBV DNA levels are positively correlated with disease stage and prognosis [3, 4].
Environmental and genetic factors may also play a role in the genesis of NPC. Studies from Hong Kong [5], Thailand [6], Malaysia [7] and China [8] have all indicated that early exposure to Chinese-style salted fish is strongly related to the risk. Familial aggregation and susceptibility gene loci for NPC have been proposed [9, 10].
NPC is ranked the ninth most common cancer in Taiwan, with 1200 new cases annually. The crude incidences are 10.05 and 3.17 per 100 000 in males and females, respectively. Most affected patients are aged 40–50 years and the 5-year survival rate is 
59.14% [11]. With disease onset at a young age and an increased survival time, second primary cancers after NPC are becoming more prevalent, either as a consequence of therapy for the first cancer or because of the effect of shared risk factors. While the risk for second cancers in patients with oral and pharyngeal cancers has been extensively studied [12–14], very few papers have concerned NPC patients [15–17], and most of these describe either relatively few cases from single institutions or short follow-up durations, so they are not informative. In 2007, a large-scale analysis was published that included a total of 8947 first NPC patients, 291 of whom subsequently developed second primary cancers [18]. However, this study was weakened because the dataset used was pooled from 13 cancer registries located in different countries; of these, only Singapore was in the endemic area while the rest were in low-incidence areas. The results were conflicting as they showed an increased risk of second cancers in Australia, Europe and Canada, but a decreased risk in Singapore. This discrepancy could be attributed to geographical and ethnic heterogeneity. Thus, there is a clear need for another large-scale epidemiological study to define the incidence and risk of second primary cancers in NPC survivors. To achieve this objective, we conducted a retrospective population-based study using a database from the Taiwan National Cancer Registry (TNCR) that included a total of 29 861 subjects with initial diagnosis of NPC from 1979 to 2003. This study is, to our knowledge, the largest population-based study to date on this specific disease.
|
patients and methods |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
data sources
We quantified second cancer incidences among 29 861 patients with initial diagnoses of NPC [International Classification of Diseases, 9th Revision (ICD-9): 147] who were reported to the TNCR (http://crs.cph.ntu.edu.tw/) from 1 January 1979 to 31 December 2003. TNCR was founded in 1979 and financially supported by the National Department of Health with the aim of estimating the cancer incidence in Taiwan. It is a population-based cancer registry. The population covered by TNCR was 22 million in 2003. Hospitals with >50 beds are obligated to submit the information on newly diagnosed cancer patients to TNCR. Registration fees are paid to the reporting hospitals on the basis of case numbers reported. All cancer registry database in TNCR have been systemically converted to ICD-9 codes [19] and linked with death certificates from the National Death Database. Persons not identified by this process were therefore considered to be alive for the purpose of the current study (passive follow-up). The incidence rates of lymphoma and leukemia were combined because they are both classified as cancers of the hematopoietic and reticuloendothelial systems in TNCR. Informed consent was not required because all registry records were anonymous and open to the public.
To assess the age of onset accurately, estimate person-year follow-up and minimize potentially unconfirmed cancer diagnosis in this study cohort, 6222 patients were excluded from analysis once they met at least one of the following criteria: (i) missing birth dates (37 cases), (ii) missing follow-up date or death status (4691 cases) or (iii) second cancer diagnosis or death occurred <2 months after NPC diagnosis (1563 cases). There were 51 cases recorded as NPC twice due to different histological types. For these cases with two diagnoses of NPC, we analyzed them as 51 single cases. As a result, a total of 23 639 patients (17 143 males and 6496 females) were included in the analysis.
statistical analysis
To quantify the excess of second cancers after NPC, we calculated the standardized incidence ratios (SIRs) [20] and the corresponding 95% of confidence intervals (CIs) for all types of second primary cancers. SIRs were taken as the ratio of the observed number (O) of second cancers to the expected number (E), which was obtained by assuming that these persons experienced the same cancer incidence as the corresponding general population. The number of person-years at risk was defined as the number of years from the date of NPC diagnosis to the date of death, date of last follow-up, date of diagnosis of second primary cancer or the end of the study period (31 December 2003), whichever came first. The person-years of observation for each gender, 5-year age group, 5-year period (1979–1983, 1984–1988, 1989–1993, 1994–1998, 1999–2003) and time since entry to the cohort (
1, 2–5, 6–10 and >10 years) were multiplied by the incidence rates of cancers for the Taiwanese population. The corresponding products were summed over all ages and calendar years to yield the expected number of second cancer at each site. CIs of SIR were on the basis of the assumption of a Poisson distribution of second cancer cases.
Cumulative incidence rates for occurrence of second cancers were calculated in the survivors' cohort, with death treated as a competing risk according to the method described by Kalbfleisch and Prentice [21]. Briefly, this method allows for the fact that patients who die are no longer at risk for second cancers, so it differs from the cumulative incidence estimated by the Kaplan–Meier method, which treats competing events as censored at the time they occurred. Gray's test [22] was used to assess the statistical significance of difference in cumulative incidence between genders. The survival curves of NPC patients with and without second cancers were calculated by the Kaplan–Meier method and the differences between these two groups were examined by a log-rank test. All statistical tests were two sided and P < 0.05 was considered statistically significant.
|
results |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
patient characteristics
The TNCR database provided a total of 23 639 cases (17 143 males and 6496 females) for whom NPC was diagnosed as the first malignancy and complete data were available for analysis. The characteristics of the patient population are listed in Table 1. Approximately three times as many males as females had NPC diagnosed at mean age 49.35 years. The mean age at diagnosis of second primaries was 56.96 years. The average follow-up time was 5.38 years, including 18 925 cases (84%) followed up for at least 1 year, 4677 cases (20%) for 5–10 years and 4403 cases (19%) for >10 years. Within this cohort, 712 cases (501 males and 211 females) developed second primary cancers (3.01%) during 127 074 person-years of follow-up. The average interval between first and second cancers was 5.33 years with standard deviation 4.68 years.
|
risk of second primary cancers by gender and site
SIRs and corresponding 95% CIs for all sites of second primary cancers were calculated by gender and site. Irrespective of site, the risk of developing a new cancer was significantly greater in both genders of NPC patients than in the general population (Table 2). A total of 712 subjects developed a subsequent cancer (SIR = 1.24, 95% CI 1.15–1.33); 501 males (SIR = 1.17, 95% CI 1.07–1.27) and 211 females (SIR = 1.46, 95% CI 1.27–1.67). The site-specific SIRs of the second cancers were calculated. In general, elevated risks with statistical significance were observed in five second cancers: oral/pharyngeal (SIR = 2.67, 95% CI 2.18–3.23), major salivary gland (SIR = 3.82, 95% CI 1.40–8.32), sarcoma (SIR = 4.05, 95% CI 2.09–7.08), skin (SIR = 2.43, 95% CI 1.73–3.32) and leukemia/lymphoma (SIR = 1.68, 95% CI 1.16–2.36). The risk of oral/pharyngeal cancer was much higher in females (SIR = 9.36, 95% CI 5.63–14.62) than males (SIR = 2.31, 95% CI 1.85–2.84). The SIRs were significantly decreased for prostate cancer (SIR = 0.53, 95% CI 0.27–0.92) and rectal cancer (SIR = 0.67, 95% CI 0.43–1.0), suggesting that these occurred less frequently than expected. Although the absolute numbers of lung cancers (n = 92) and liver cancers (n = 84) were large, the risk for them did not exceed that of the general population. There were 129 second cancer cases classified in the category ‘others’, including larynx, bone, testis, ovary, brain, gall bladder, thyroid gland and thymus.
|
risk of second cancers with significance stratified by follow-up intervals
Elevated SIRs were observed with statistical significance in the following cancers: oral/pharyngeal, salivary gland, sarcoma, skin and leukemia/lymphoma. To explore the latency of development of these cancers, we stratified their SIRs by the length of follow-up time after diagnosis of NPC (Table 3). The entire follow-up period was divided into consecutive 5-year intervals. As shown, the risk of oral/pharyngeal cancer increased at a constant rate throughout the follow-up period. The risk of sarcoma increased significantly after a latency of 5 years and continued increasing after 10 years. Skin cancer had a bimodal risk period of ‘2–5 years’ and ‘>10 years’. The risk of salivary gland cancer was increased only in men during the first year of follow-up and after a latency of 10 years. Females had a higher risk of oral/pharyngeal cancer than males in every interval (Table 4), consistent with the observation in Table 2.
|
|
risk of second cancers by age at first diagnosis of NPC
NPC peaks in the age range 40–49 years in Taiwan [23]. To study the effect of age at diagnosis on second cancer risk, we stratified the SIRs of all second primary cancers by age at initial NPC diagnosis (Table 5). The SIRs were significantly higher for patients diagnosed while young (onset age <40 years: SIR = 7.59, 95% CI 6.27–9.11) than for age >40 years (SIR = 1.07, 95% CI 0.98–1.16). For those diagnosed at age >40, a moderately increased risk remained at age 40–49 (SIR = 2.27, 95% CI 2.39–3.20) and was attenuated at age 50–59 (SIR = 1.63, 95% CI 1.42–1.86), whereas a decreased risk was found for NPC diagnosed at age >60 (SIR = 0.52, 95% CI 0.45–0.60). In general, SIRs were higher for younger patients diagnosed before 40 years old than for those diagnosed when older.
|
cumulative incidence rates of second cancers
The estimated overall risk of developing a second cancer after NPC in the survivors' cohort was calculated, with death treated as a competing risk. The cumulative incidence and the point-wise CIs are shown (Figure 1A). The overall cumulative risks of second primaries >5, 10, 15 and 20 years after NPC diagnosis were estimated to be 1.89%, 3.26%, 4.23% and 5.37%, respectively. There was no risk plateau and the cumulative incidences over time did not differ between male and female patients, indicating that both genders were at equivalent risk for all second cancers following NPC (P = 0.20) (Figure 1B).
|
second primary cancers and the overall survival of NPC patients
Many second cancers occur after a latency following the first cancer, so the longer the patient survives, the higher the risk for a second cancer becomes. We calculated the Kaplan–Meier survival curves for 22 927 patients without second cancers (median survival 6.99 years, 95% CI 6.71–7.28, 5-year survival 56%) and 712 patients with second cancers (median survival 8.42 years, 95% CI 7.54–9.31, 5-year survival 67%). Overall, these two curves are different (P = 0.031) (Table 6). However, they crossover around the ninth year of survival (Figure 2A). Before the ninth year, these two groups have similar probabilities of survival (P = 0.34), but the patients with second cancers have a significantly inferior survival after the ninth year (P < 0.001). The second cancer group had a worse 15-year survival rate than the nonsecond cancer group (24% versus 37%) (Table 6).
|
|
survival time after second cancers
The overall survival time for 712 patients after the diagnosis of second cancers was calculated by the Kaplan–Meier method (Figure 2B). The median survival was 1.70 years (95% CI 1.38–2.03). The 1-year, 5-year and 10-year survival rate were 62%, 30%, and 18%, respectively.
|
discussion |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
In this population-based study of 23 639 initial NPC cases, including 712 who developed a subsequent cancer, we found a 24% increased risk of all second cancers after NPC compared with the general population (SIR = 1.24, 95% CI 1.15–1.33). Our dataset showed a male to female ratio of 2.64 : 1, similar to other native populations of southern China [1] and Hong Kong [2]. This predicted gender distribution reflected the large size of our study population. Population-based studies have the strength of minimizing the selection bias inherent in single-institution reviews. The data analyzed in the present study came from a TNCR dataset that included cancer registries from 1979 to 2003 in Taiwan. TNCR is a nationwide cancer registration system that provides fees to reimburse reporting hospitals, thus reducing the chance of underreporting of second cancers among patients. However, our results should still be viewed in the context of cancer registry-based data, which have some limitations and potentially confounding factors that could have led to erroneous interpretations. For example, to eliminate the bias of unconfirmed initial cancer diagnosis or misclassification of recurrent disease as a new primary, the cases were excluded if their first and second cancer diagnoses differed by <2 months. This may eliminate some synchronous multiple primary cancers and thus led to an underestimate of the number of second cancers. On the other hand, the false classification of a recurrent or metastatic site as second primary cancer may have overestimated the SIR, especially if the histology was the same. Most NPC patients in Taiwan are World Health Organization type III (undifferentiated carcinoma). In contrast to other squamous cell carcinoma of the head and neck, it is easier to distinguish NPC from second primary cancers because of the distinct histology.
Second primary cancers may occur because of shared environment or genetic risks or may be therapy related. NPC is an EBV-related cancer. None of the second cancers in NPC patients were EBV associated [16]. Field cancerization theory cannot provide an explanation either, because none of the second tumors were undifferentiated carcinomas. Therefore, the pathogenesis of second cancers after NPC remains obscure, although radiotherapy can probably explain part of the observed excess. We speculate that the risk pattern that emerges from our study is in general consistent with radiation-related second cancers. In a cohort of 326 patients with undifferentiated NPC, Kong et al. [24] reported that the incidence of second primary cancers was 5.2% and tended to increase >5 years after radiation. In our larger study population, the incidence of second primary cancers was 3.01% among the 23 639 NPC patients. With respect to oral/pharyngeal cancer and leukemia/lymphoma, Wang et al. [16] and Scelo et al. [18] have shown the same risk in NPC patients. Our database corroborated their findings. The reduction of risk for the prostate and rectal cancers was intriguing. More work is needed to identify their mechanistic associations with NPC.
Our study has shown the incidence and risk of second primary cancers increase in NPC patients. The mean follow-up of this cohort is relatively short (5.38 years) and thus, the magnitude of risk (as measured by SIR and cumulative incidence) could increase with longer follow-up. Overall, the SIRs for second primary cancers were strikingly higher for patients diagnosed before 40 years of age (SIR = 7.59, 95% CI 6.27–9.11) than for those diagnosed later (SIR = 1.07, 95% CI 0.98–1.16). This finding was not spurious because Scelo et al. [18] reported the same pattern of risk in NPC. Shared genetic susceptibility might contribute to this association. Moreover, the lower baseline risk in the younger population could also contribute in part to the higher magnitude of excess risk. Our finding could have important implications regarding future cancer surveillance especially for young patients.
A major finding of our study is the negative impact of second primary cancers on survival, which has never been examined previously because the numbers of second cancer patients have been insufficient. The survival curves of NPC patients with and without second cancers did not differ statistically before the crossover in the ninth year of survival (P = 0.34) but became statistically different thereafter (P < 0.001). A possible explanation for the crossing-over of the curves could be that the longer the patients survive the first cancer, the greater their risk of developing a second cancer; but once the second cancer occurs, prognosis is usually poor [25]. We have shown that among 712 patients after diagnosis of second cancers, the median survival was only 1.7 years; 48% died within the first year and only 30% survived >5 years. Similarly, in a study of head and neck cancer, 41% of second cancer patients died within 1 year and only 29% survived >5 years [13]. Our data emphasize that the longevity of NPC survivors is prejudiced by the emergence of second primary cancers.
In conclusion, our data indicate that NPC is associated with an increased risk of second malignancies. In contrast to single institution or pooled data which are hampered by small numbers or geographical heterogeneity of registries, the major strength of our study is the size of the cohort and a uniform population-based dataset from a high-incidence area. Identifying the incidence and risk of second cancers may help develop surveillance and early prevention strategies targeted at individuals who survive NPC.
|
funding |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
National Science Council (95-2118-M-182-001); Chang Gung Molecular Medicine Research Center, Chang Gung University (CMRPD140041).
|
Acknowledgements |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
We are grateful to the staff in the Taiwan National Cancer Registry, National Department of Health, Taiwan, R.O.C.
Received for publication November 30, 2007. Revision received December 25, 2007. Accepted for publication December 31, 2007.
|
References |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
1. Jia WH, Huang QH, Liao J, et al. Trends in incidence and mortality of nasopharyngeal carcinoma over a 20–25 year period (1978/1983–2002) in Sihui and Cangwu counties in southern China. BMC Cancer (2006) 6:178.[Medline]
2. Lee AW, Foo W, Mang O, et al. Changing epidemiology of nasopharyngeal carcinoma in Hong Kong over a 20-year period (1980–99): an encouraging reduction in both incidence and mortality. Int J Cancer (2003) 103:680–685.[CrossRef][Web of Science][Medline]
3. Lin JC, Chen KY, Wang WY, et al. Detection of Epstein-Barr virus DNA the peripheral-blood cells of patients with nasopharyngeal carcinoma: relationship to distant metastasis and survival. J Clin Oncol (2001) 19:2607–2615.
4. Lin JC, Wang WY, Chen KY, et al. Quantification of plasma Epstein-Barr virus DNA in patients with advanced nasopharyngeal carcinoma. N Engl J Med (2004) 350:2461–2470.
5. Yu MC, Ho JH, Lai SH, Henderson BE. Cantonese-style salted fish as a cause of nasopharyngeal carcinoma: report of a case-control study in Hong Kong. Cancer Res (1986) 46:956–961.
6. Sriamporn S, Vatanasapt V, Pisani P, et al. Environmental risk factors for nasopharyngeal carcinoma: a case-control study in northeastern Thailand. Cancer Epidemiol Biomarkers Prev (1992) 1:345–348.[Abstract]
7. Armstrong RW, Armstrong MJ, Yu MC, Henderson BE. Salted fish and inhalants as risk factors for nasopharyngeal carcinoma in Malaysian Chinese. Cancer Res (1983) 43:2967–2970.
8. Ning JP, Yu MC, Wang QS, Henderson BE. Consumption of salted fish and other risk factors for nasopharyngeal carcinoma (NPC) in Tianjin, a low-risk region for NPC in the People's Republic of China. J Natl Cancer Inst (1990) 82:291–296.
9. Feng BJ, Huang W, Shugart YY, et al. Genome-wide scan for familial nasopharyngeal carcinoma reveals evidence of linkage to chromosome 4. Nat Genet (2002) 31:395–399.[Web of Science][Medline]
10. Xiong W, Zeng ZY, Xia JH, et al. A susceptibility locus at chromosome 3p21 linked to familial nasopharyngeal carcinoma. Cancer Res (2004) 64:1972–1974.
11. Cancer Registry Annual Report. (2001) Taiwan, R.O.C: Department of Health.
12. Jones AS, Morar P, Phillips DE, et al. Second primary tumors in patients with head and neck squamous cell carcinoma. Cancer (1995) 75:1343–1353.[CrossRef][Web of Science][Medline]
13. Day GL, Blot WJ, Shore RE, et al. Second cancers following oral and pharyngeal cancer: patients' characteristics and survival patterns. Eur J Cancer B Oral Oncol (1994) 30B:381–386.[CrossRef]
14. Choy AT, van Hasselt CA, Chisholm EM, et al. Multiple primary cancers in Hong Kong Chinese patients with squamous cell cancer of the head or neck. Cancer (1992) 70:815–820.[CrossRef][Medline]
15. Teshima T, Inoue T, Chatani M, et al. Incidence of other primary cancers in 1,569 patients with pharyngolaryngeal cancer and treated with radiation therapy. Strahlenther Onkol (1992) 168:213–218.[Medline]
16. Wang CC, Chen ML, Hsu KH, et al. Second malignant tumors in patients with nasopharyngeal carcinoma and their association with Epstein-Barr virus. Int J Cancer (2000) 87:228–231.[CrossRef][Web of Science][Medline]
17. Cooper JS, Scott C, Marcial V, et al. The relationship of nasopharyngeal carcinomas and second independent malignancies based on the Radiation Therapy Oncology Group experience. Cancer (1991) 67:1673–1677.[Medline]
18. Scelo G, Boffetta P, Corbex M, et al. Second primary cancers in patients with nasopharyngeal carcinoma: a pooled analysis of 13 cancer registries. Cancer Causes Control (2007) 18:269–278.[CrossRef][Web of Science][Medline]
19. Slee VN. The International Classification of Diseases: ninth revision (ICD-9). Ann Intern Med (1978) 88:424–426.
20. Breslow NE, Day NE. Statistical methods in cancer research. Volume II—the design and analysis of cohort studies. IARC Sci Publ (1987) 1–406.
21. Kalbfleisch JD, Prentice R. The Analysis of Failure Time Data (1980) New York, NY: John Wiley.
22. Gary R. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat (1988) 16:1141–1154.[CrossRef]
23. Yu MC, Yuan JM. Epidemiology of nasopharyngeal carcinoma. Semin Cancer Biol (2002) 12:421–429.[CrossRef][Web of Science][Medline]
24. Kong L, Lu JJ, Hu C, et al. The risk of second primary tumors in patients with nasopharyngeal carcinoma after definitive radiotherapy. Cancer (2006) 107:1287–1293.[Medline]
25. Alberts DS. Second cancers are killing us! Cancer Epidemiol Biomarkers Prev (2006) 15:2019.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||