Japanese Journal of Clinical Oncology Advance Access originally published online on February 1, 2008
Japanese Journal of Clinical Oncology 2008 38(2):146-157; doi:10.1093/jjco/hym156
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Authors (2008). Published by Oxford University Press. All rights reserved
Partial Cancer Prevalence in Japan up to 2020: Estimates Based on Incidence and Survival Data from Population-based Cancer Registries
1 Department of Mathematical Health Science, School of Health Science, Graduate School of Medicine, Osaka University, Osaka, Japan
2 Health Services Section, Nishi-Shinjuku Public Health Center, Tokyo, Japan
3 Department of Epidemiology, Radiation Effects Research Foundation, Hiroshima, Japan
4 Department of Cancer Control and Statistics, Osaka, Japan
5 Cancer Information Service, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan
For reprints and all correspondence: Nana Tabata, Department of Mathematical Health Science (Ohno Research Laboratory), Course of Health Science, Graduate School of Medicine, Osaka University, 1-7 Yamadaoka, Suita, Osaka 565-0871, Japan. E-mail: nana-t{at}sahs.med.osaka-u.ac.jp
Received June 25, 2007; accepted November 4, 2007
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONMETHODRESULTSDISCUSSIONFundingReferences |
|---|
Measuring cancer prevalence in Japan has been difficult because population-based cancer registries have been conducted in limited areas. The purpose of this study was to estimate cancer prevalence in Japan from 1995 to 2020 for 5-year periods based on selected population-based cancer registry data. 1-, 2–3-, 4–5- and 5-year partial prevalence were estimated using incidence and survival data. Incidence and survival were calculated using data from selected cancer registries. We estimated the cancer survival by age group, primary site, and sex using the mixture cure fraction model. Kaplan–Meier estimates were applied to subgroups for which the survival did not converge to the estimated model. We projected that 1-year cancer prevalence for all sites would increase from 209 971 to 367 354 for men and from 164 622 to 275 776 for women, that 2–3-year prevalence would increase from 288 284 to 508 731 for men and from 255 684 to 418 630 for women, that 4–5-year prevalence would increase from 216 834 to 379 461 in men and from 211 764 to 342 031 in women, and that 5-year prevalence would increase from 715 089 to 1 255 546 in men and from 632 070 to 1 036 437 in women. This study is the first estimate of cancer prevalence in the future in Japan.
Key Words: cancer • prevalence • cancer registry
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONFundingReferences |
|---|
Cancer is still the first cause of death in both men and women in Japan, although early diagnosis by screening and the efficacy of cancer treatment have improved the prognosis for cancer patients in recent years. Therefore, monitoring cancer prevalence enables policymakers to grasp the numbers of cancer patients who need care and resources. In this paper, prevalence is defined as the number of patients diagnosed at the beginning of a fixed period (1, 2–3, 4–5 or 5 years) and surviving at the end of that period. The length of the period is significant. Prevalence at 1, 2–3 and 4–5 years are applicable to the effect of initial treatment for cancer, clinical follow-up and point of cure (1). And 5-year prevalence indicated the total uncured patients (2). In Japan, we do not have a national cancer registration system. Therefore, in this study, 1-, 2–3-, 4–5- and 5-year age-specific cancer prevalence by primary site (for 13 sites) and sex were estimated from 1995 to 2020 for 5-year periods based on selected population-based cancer registries.
|
METHOD |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONFundingReferences |
|---|
Method of Estimation
First, we calculated survival of cancer patients by age (15–44, 45–54, 55–64, 65–74 and 75+ years old) and sex for 13 primary sites using the Kaplan–Meier method. Then the survival was applied to the mixture cure fraction model (3,4). The Kaplan–Meier estimates were applied to subgroups for which the survival did not converge to estimates of the mixture cure fraction model. Prevalence was estimated by incidence and year-specific survival using
|
| (2) |
Source of Survival
We used data from seven population-based cancer registries (for Miyagi, Yamagata, Niigata, Fukui, Osaka, Tottori and Nagasaki prefectures) that met the required standards (5) for quality of registration and prognosis investigation that were constructed by the Collaborative Study of Population-Based Cancer Registries in Japan.
We analyzed cancer prevalence for 13 sites: esophagus (C15), stomach (C16), colon (C18), rectum (C19–C21), liver (C22), gallbladder (C23–C24), pancreas (C25), lung (C33–C34), breast (C50, D05), uterus (C53–C55, D06), prostate (C61), bladder (C67) and all sites (C00–C96, D05–D06). We excluded breast cancer cases in men and cases for patients under 15 years old from the analysis.
Incidence Data
We applied the incidence data estimated from 11 cancer registries provided by the Center for Cancer Control and Information Services, National Cancer Center, Japan, to estimate prevalence in 1995 and 2000 (6). To project future prevalence (from 2005 to 20), we applied the incidence estimated by Ohno et al. using the Bayesian age-period-cohort model based on the incidence data in Japan (7).
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONFundingReferences |
|---|
Tables 1
–4 show 1-, 2–3-, 4–5- and 5-year age-specific cancer prevalence by sites and sex from 1995 to 2020 for 5-year periods. We estimated that 1-year cancer prevalence for all sites would increase from 209 971 to 367 354 for men and from 164 622 to 275 776 for women. For 2–3-year prevalence, it would increase from 288 284 to 508 731 for men and from 255 684 to 418 630 for women. For 4–5-year prevalence, it would increase from 216 834 to 379 461 in men and from 211 764 to 342 031 in women. For 5-year prevalence, it would increase from 715 089 to 1 255 546 in men and from 632 070 to 1 036,437 in women. For all sites and each primary site in both sexes, prevalence increased from 1995 to 2020.
|
|
|
|
In our projections, the highest prevalence was shown in stomach for both sexes in 1995. In case of men, the second highest prevalence was shown in lung and the third was in colon and the forth was shown in liver. In case of women, the second highest prevalence was shown in breast, the third was in colon and the forth was shown in uterus. Meantime, in 2020, the highest prevalence for men was shown in prostate, the second was in stomach, the third was in lung and the forth was shown in colon. For women, the highest prevalence was shown in breast, the second was in colon and the third was in uterus (include CIS) and the forth was shown in stomach.
During the quarter century, the prevalence of prostate, lung, colon and rectal cancer increased rapidly, however, that of stomach and liver cancer showed slightly increase in men. For women, prevalence of colon, breast and uterus increased prominently, on the other hand, the prevalence of stomach and gallbladder showed rather stable.
Prevalence for patients over 75 years old increased the most remarkably among all age groups except uterus. In addition, prevalence in 15–44-year-old patients declined for all sites except breast and uterus.
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONFundingReferences |
|---|
Partial cancer prevalence is usually proposed as an index of actual cancer revelation instead of total prevalence for the reason that it is difficult to count up all the cancer patient through several decades even in a country providing fairly well-managed cancer registry system. Using the estimated incidence and survival function, in this study we calculated less than 5-year prevalence that is regarded as a reference index of cancer treatment, though there exist apparent differences of prognosis among sites (8).
Relating to the Survival Function Used in the Future Projection
In order to estimate the future partial cancer prevalence, survival and incidence are necessary to calculate. Survival for some sites is improving due to the progress of medical technology and the increase of early-detected cases; however, we employed the survival of cancer patients diagnosed between 1993 and 1996 to estimate the future prevalence. For the up-to-date calculation of partial prevalence, sequential revision of survival using with period analysis or other mathematical model and more long-term survival would be imperative (9,10).
Relating to the Incidence in the Future Projection
We applied the estimated cancer incidence presented by Ohno et al. for the projection of prevalence (7). Applying the Bayesian age-period-cohort model to the past age-specific cancer incidence data from 1975 to 1994 by sites and sex, there obtained longitudinal changes of the three effects, namely age, period and cohort. Using the value of the three effects, the age-specific incidence at any year by site and sex could be calculated. Furthermore, the future trends in three effects would be projected assuming some adaptive scenarios, for details, age effect was assumed the same as those estimate and the period effect was extrapolating using an adequate function chosen among constant, linear, and quadratic functions and the cohort effect was set at the latest birth cohort. The future incidence reflecting the change of the age distribution, period effect and cohort effect was applied to the future prevalence estimation.
Further Discussions
This is the first estimate of cancer prevalence in the future in Japan. We will need to evaluate the accuracy of this estimate in a few years. The quality of cancer registration in Japan has not been reliable, but concerted efforts are being made to improve the quality of registration in the future. Thus, the incidence and survival data will be changed, so we will need to consider them when we analyze future data. In addition, we need to construct the system and estimate cancer prevalence regularly, because it is important data for cancer control program.
|
Funding |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONFundingReferences |
|---|
This study was supported by a Grant-in-Aid for Cancer Research from the Japanese Ministry of Health, Labour and Welfare: (14-2) and was carried out under the ISM Cooperative Research Program (2004-ISM CRP-2037).
Conflict of interest statement
None declared.
|
References |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONFundingReferences |
|---|
1 Pisani P, Freddie B, Parkin DM. Estimates of the world-wide prevalence of cancer for 25 sites in the adult population. Int J Cancer (2002) 97:72–81.[CrossRef][Web of Science][Medline]
2 Ferlay J, Bray F, Sankila R, Parkin DM. EUCAN: Cancer incidence, mortality and prevalence in the European Union 1995, version 2.0. (1999) Lyon: IARC Press. IARC Cancer Base No. 4.
3 De Angelis R, Capocaccia R, Hakulinen T, Soderman B, Verdecchia A. Mixture models for cancer survival analysis: application to population-based data with covariates. Statist Med (1999) 18:441–54.[CrossRef]
4 Phillips N, Coldman A, McBride ML. Estimating cancer prevalence using mixture models for cancer survival. Stat Med (2002) 21:1257–70.[CrossRef][Web of Science][Medline]
5 Tsukuma H, Ajiki W, Ioka A, Oshima A. Research Group of Population-Based Cancer Registries of Japan. Survival of cancer patients diagnosed between 1993 and 1996: collaborative study of population-based cancer registries in Japan. Jpn J Clin Oncol (2006) 36(9):602–7.
6 The Japan Cancer Surveillance Research Group. Cancer incidence and incidence rates in Japan in 2000: estimates based on data from 11 Population-based cancer registries. Jpn J Clin Oncol (2006) 36(10):668–75.
7 Ohno Y, Nakamura T, Murata K, Tsukuma H, Ajiki W, Oshima A. Cancer Statistics-2004. Oshima A, Kuroishi T, Tajima K, eds. (2004) Tokyo: Shinohara shuppan. 202–17. (in Japanese).
8 Tai P, Yu E, Csemi G, Vlastos G, Royce M, Kunkler I, et al. Minimum follow-up time required for the estimation of statistical. BMC Cancer (2005) 5:48.[CrossRef][Medline]
9 Brenner H, Gefeller O. Deriving more up-to-date estimates of long-term patient survival. J Clin Epidemiol (1997) 50:211–6.[CrossRef][Web of Science][Medline]
10 Ito Y, Ono Y, Soda M, Oshima A. Up-to-date estimates of cancer survival by period analysis-Women Lung Cancer patients in Nagasaki, Japan, as an instance. Jpn J Cancer Clin (2006) 52:97–102. (in Japanese).
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Shimbo, S. Tomioka, M. Sasaki, T. Shima, N. Suzuki, S. Murakami, H. Nakatsu, and J. Shimazaki PSA Doubling Time as a Predictive Factor on Repeat Biopsy for Detection of Prostate Cancer Jpn. J. Clin. Oncol., November 1, 2009; 39(11): 727 - 731. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||