Chapter 24 Cancer survival in Singapore, 1993−1997 Chia KS Singapore cancer registry The Singapore cancer registry is a national registry established in 1968 to obtain information on cancer patterns in the entire country. The registry has been contributing data to the quinquennial IARC publication Cancer Incidence in Five Continents since volume III [1]. Cancer notification is voluntary, and registration of cases is predominantly by passive methods with no personal contact with cases. The principal sources of information on incident cancer cases are the notification forms from all sections of the medical profession, pathology and hospital records [2]. The registry caters to a population of about 4.1 million with a sex ratio of 986 females to 1000 males in 2002, comprising major ethnic groups of Chinese, Malays and Indians. The average annual age-standardized incidence rate of all cancers and ethnic populations together is 235 per 100 000 among males and 200 per 100 000 among females in 1998− 1999 [3]. The registry contributed data on survival from 45 cancer sites or types for the first time in this volume of the IARC publication on Cancer Survival in Africa, Asia, the Caribbean and Central America. In the present volume, the main tables pertain to the period 1993−1997. The data on survival for the years 1968−1992 are also utilized to elicit the trend in cancer survival using different approaches. Data quality indices (Table 1) The proportion of cases with histologically verified cancer diagnosis in the series varied from 100% for many cancers to 27% in liver cancer. The frequency of cases registered based on a death certificate only (DCO) range between nil among many cancers to 7% in unspecified leukaemia. Cases excluded from the study, due to lack of follow-up and other basic information, are in the range of 0% for mesothelioma and 24% for bone cancers. Thus, 76−100% of the total cases registered are included in the estimation of the survival probability. Outcome of follow-up (Table 2) The follow-up of cases has been completely carried out by passive methods. Since certification of death is virtually complete, the cancer mortality information received from the death certificate is matched with the incident cancer database. The vital status of the unmatched incident case is then collected by scrutiny of hospital records, and all such cases are presumed to be alive until the end of the calendar year for which the mortality data are fully available. The closing date of follow-up was 31 st December 2001. The median follow-up ranged from 2 months for liver cancer and unspecified leukaemia to 72 months for testicular cancer. The completeness of follow-up at 5 183 Abstract The Singapore cancer registry is a national registry established in 1968. Cancer registration is done by passive methods. The registry contributed survival data on 45 cancer sites or types registered during 1993−1997. Data on 34 cancers registered during 1968−1997 were utilized for survival trend by period and cohort approaches. Follow-up was done by passive methods, with median follow-up ranging between 2−72 months for different cancers. The proportion with histologically verified diagnosis for various cancers ranged between 27−100%; death certificates only (DCOs) comprised 0−7%; 76−100% of total registered cases were included for the survival analysis. The top-ranking cancers on 5-year age-standardized relative survival rates were non- melanoma skin (96%), thyroid (90%), testis (88%), corpus uteri (77%), breast (74%), Hodgkin lymphoma (73%) and penis (70%). Five-year relative survival by age group showed either a decreasing trend with increasing age groups or was fluctuating. Localized stage of disease ranged between 18−65% for various cancers and survival decreased with increasing extent of disease. Period survival closely predicted survival experience of cancers diagnosed in that period, and an increasing trend in period survival over different periods indicated an improved prognosis for cancers diagnosed in those calendar periods. http://survcan.iarc.fr
16
Embed
Can Surv II-Chapter24 · Chapter 24 Cancer survival in Singapore, 1993−1997 Chia KS Singapore cancer registry The Singapore cancer registry is a national registry established in
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Chapter 24
Cancer survival in Singapore, 1993−1997Chia KS
Singapore cancer registry
The Singapore cancer registry is a national registryestablished in 1968 to obtain information on cancerpatterns in the entire country. The registry has beencontributing data to the quinquennial IARCpublication Cancer Incidence in Five Continents sincevolume III [1]. Cancer notification is voluntary, andregistration of cases is predominantly by passivemethods with no personal contact with cases. Theprincipal sources of information on incident cancercases are the notification forms from all sections ofthe medical profession, pathology and hospitalrecords [2]. The registry caters to a population ofabout 4.1 million with a sex ratio of 986 females to1000 males in 2002, comprising major ethnic groupsof Chinese, Malays and Indians. The average annualage-standardized incidence rate of all cancers andethnic populations together is 235 per 100 000 amongmales and 200 per 100 000 among females in 1998−1999 [3].
The registry contributed data on survival from 45cancer sites or types for the first time in this volumeof the IARC publication on Cancer Survival in Africa,Asia, the Caribbean and Central America. In thepresent volume, the main tables pertain to the period1993−1997. The data on survival for the years 1968−1992 are also utilized to elicit the trend incancer survival using different approaches.
Data quality indices (Table 1)
The proportion of cases with histologically verifiedcancer diagnosis in the series varied from 100% formany cancers to 27% in liver cancer. The frequency ofcases registered based on a death certificate only(DCO) range between nil among many cancers to 7% inunspecified leukaemia. Cases excluded from thestudy, due to lack of follow-up and other basicinformation, are in the range of 0% for mesotheliomaand 24% for bone cancers. Thus, 76−100% of the totalcases registered are included in the estimation of thesurvival probability.
Outcome of follow-up (Table 2)
The follow-up of cases has been completely carriedout by passive methods. Since certification of death isvirtually complete, the cancer mortality informationreceived from the death certificate is matched withthe incident cancer database. The vital status of theunmatched incident case is then collected by scrutinyof hospital records, and all such cases are presumedto be alive until the end of the calendar year forwhich the mortality data are fully available.
The closing date of follow-up was 31st December 2001.The median follow-up ranged from 2 months for livercancer and unspecified leukaemia to 72 months fortesticular cancer. The completeness of follow-up at 5
183
Abstract
The Singapore cancer registry is a national registry established in 1968. Cancer registration is done by passivemethods. The registry contributed survival data on 45 cancer sites or types registered during 1993−1997. Dataon 34 cancers registered during 1968−1997 were utilized for survival trend by period and cohort approaches.Follow-up was done by passive methods, with median follow-up ranging between 2−72 months for differentcancers. The proportion with histologically verified diagnosis for various cancers ranged between 27−100%;death certificates only (DCOs) comprised 0−7%; 76−100% of total registered cases were included for thesurvival analysis. The top-ranking cancers on 5-year age-standardized relative survival rates were non-melanoma skin (96%), thyroid (90%), testis (88%), corpus uteri (77%), breast (74%), Hodgkin lymphoma (73%)and penis (70%). Five-year relative survival by age group showed either a decreasing trend with increasing agegroups or was fluctuating. Localized stage of disease ranged between 18−65% for various cancers and survivaldecreased with increasing extent of disease. Period survival closely predicted survival experience of cancersdiagnosed in that period, and an increasing trend in period survival over different periods indicated animproved prognosis for cancers diagnosed in those calendar periods.
http://survcan.iarc.fr
years from the incidence date was 100% for allcancers as there are no losses to follow-up.
Survival statistics
All ages and both sexes together (Table 3)
The top-ranking cancers on 5-year relative survivalare non-melanoma skin (96%), thyroid (90%), testis(87%), corpus uteri (81%) and breast (76%). The lowestsurvival rate is encountered with cancer of thepancreas (4%), preceded by cancer of the liver (5%),oesophagus (6%) and lung (7%) and mesothelioma(9%). Salivary gland (69%) among other head and neckcancers and colon and rectum (50%) amonggastrointestinal cancers have a higher survival ratethan others in the category. Hodgkin lymphoma has abetter survival rate than non-Hodgkin. The survivalfigures for haematopoietic malignancies are asfollows: multiple myeloma (19%), lymphoid leukaemia(46%), myeloid leukaemia (18%) and unspecifiedleukaemia (9%).
The 5-year age-standardized relative survival (ASRS)probability for all ages together is observed to be lessthan or similar to the corresponding unadjusted onefor a majority of cancers. Also, the 5-year ASRS (0−74years of age) is generally higher than or similar to thecorresponding ASRS (all ages) for a majority ofcancers.
SexMale (Table 4a)
The 5-year relative survival of cancer of the testis is87%, prostate is 63% and penis is 66%. Cancers of the
hypopharynx, breast, small intestine and urinarybladder have a notably higher survival among malesthan females.
Female (Table 4a)
The 5-year relative survival from cancers of thebreast, uterine cervix, ovary and vulva are 76%, 64%,64% and 62%, respectively. Survival is markedly higheramong females than males in most cancers of thehead and neck, rectum, anus, other thoracic organs,melanoma and non-melanoma skin, renal pelvis,Hodgkin and non-Hodgkin lymphoma and lymphoidleukaemia.
Chapter 24
184
http://survcan.iarc.fr
63.7
64.0
66.2
69.2
71.9
76.2
80.6
86.7
90.3
95.8
0 20 40 60 80 100
Ovary
Urinary bladder
Penis
Salivary gland
Hodgkin lymphoma
Breast
Corpus uteri
Testis
Thyroid
Non-melanoma skin
5-year relative survival %
Figure 1a. Top ten cancers (ranked by survival), Singapore, 1993−1997
68.7
85.7
86.7
88.4
98.0
0 20 40 60 80 100
Hodgkin lymphoma
Thyroid
Testis
Breast
Non-melanoma skin
5-year relative survival %
Figure 1b. Top five cancers (ranked by survival), Male,Singapore, 1993−1997
76.9
77.1
80.6
91.6
93.8
0 20 40 60 80 100
Larynx
Hodgkin lymphoma
Corpus uteri
Thyroid
Non-melanoma skin
5-year relative survival %
Figure 1c. Top five cancers (ranked by survival), Female, Singapore, 1993−1997
Cancer survival in Singapore, 1993−1997
185
Age group (Table 4b)
The 5-year relative survival by age group reveals nodistinct pattern or trend, and fluctuates withincreasing age groups for most cancers. However, aninverse relationship between age group and survivalwas observed for cancers of the salivary gland, gallbladder, larynx, other thoracic organs, mesothelioma,corpus uteri, ovary, brain, thyroid and myeloidleukaemia.
Extent of disease (Table 5; Figure 2)
The information on the clinical extent of disease isanalysed for selected cancer sites. Most of thecancers have been diagnosed at a localized stageranging between 18−65% for various cancers.Nasopharyngeal cancer is an exception wherein amajority (36%) have regional spread of disease atdiagnosis. Distant metastasis at diagnosis vary from20% for ovarian cancer to 1% in cancer of the oralcavity. The unknown category is substantial, rangingbetween 24−50%. Survival is the highest amonglocalized cancers, followed by regional and distantmetastasis cases among the known categories.
Survival trend (Table 6)
The trend of survival data, estimated by the samemethod of semi-complete analytic approach as in theprevious tables, is available for 34 cancer sites ortypes spanning 10 years in two time periods, 1988−1992 and 1993−1997. An increasing trend with anabsolute difference of 8−10% and more between thetwo calendar periods is observed in cancers of thecolon, rectum, larynx, non-melanoma skin, prostateand urinary bladder. The survival was similar for amajority of other cancers in successive calendarperiods.
Trend of survival by period and cohort approaches (Tables 7-9; Figure 3)
The availability of data on registration and follow-uptogether for both a long (from the calendar year1968) and up to a recent period (year 1997) ofcalendar time led to the possibility of estimating up-to-date survival and trend by period approach.Survival is also estimated by cohort approach forcomparison.
The 5-, 10- and 15-year relative survival estimates bycohort and period approaches are estimated for thedifferent 5-year calendar periods from 1973−1977 to1993−1997. A distinct correspondence between thetwo approaches is forthcoming. The period survivalestimates at 5, 10 and 15 years of follow-up in acalendar period are seen to resemble the cohortsurvival estimates of the succeeding calendar periodsafter 5, 10 and 15 years respectively for mostcancers. Thus, period survival closely predicts thesurvival experience of cancer cases diagnosed in thatperiod. An increasing trend of period survivalestimates over the different calendar periods is anindicator for improved prognosis for cancersdiagnosed in those calendar periods.
References
1. Parkin DM, Whelan SL, Ferlay J and Storm H. CancerIncidence in Five Continents, Vol I to VIII: IARCCancerbase No. 7. IARCPress, Lyon, 2005.
2. Lee HP, Day NE and Shanmugaratnam K. Cancer inci-dence in Singapore 1968−1982: IARC ScientificPublications No. 91. National University of Singapore,Singapore, 1988.
3. Chia KS, Lee JJ, Wong JL, Gao W, Lee HP,Shanmugaratnam K. Cancer incidence in Singapore,1998 to 1999. Ann Acad Med Singapore. 2002; 31(6):745−750.
http://survcan.iarc.fr
Chapter 24
186
http://survcan.iarc.fr
Figure 2. Absolute survival (%) from selected cancers by extent of disease, Singapore, 1993−1997
0
20
40
60
80
100
0 1 2 3 4 5
Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
0
20
40
60
80
100
0 1 2 3 4 5Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
0
20
40
60
80
100
0 1 2 3 4 5
Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
0
20
40
60
80
100
0 1 2 3 4 5
Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
Figure 2a. Tongue
Figure 2b. Oral cavity
Figure 2c. Nasopharynx
Figure 2d. Colon
0
20
40
60
80
100
0 1 2 3 4 5Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
Figure 2e. Rectum
0
20
40
60
80
100
0 1 2 3 4 5Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
0
20
40
60
80
100
0 1 2 3 4 5Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
0
20
40
60
80
100
0 1 2 3 4 5Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
0
20
40
60
80
100
0 1 2 3 4 5
Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
Figure 2f. Larynx
Figure 2g. Breast
Figure 2h. Cervix
Figure 2i. Corpus uteri
0
20
40
60
80
100
0 1 2 3 4 5Years after diagnosis
Rat
e (%
)
Localized
Regional
Distantmetastasis
Unknown
Figure 2j. Ovary
Cancer survival in Singapore, 1993−1997
187
http://survcan.iarc.fr
Figure 3. Up-to-date 5-year relative survival of selected cancers by period and cohort approaches, Singapore
0
20
40
60
80
100
1968-72 1973-77 1978-82 1983-87 1988-92 1993-97Calendar period
5-ye
ar re
lativ
e su
rviv
al
period survival cohort survival
0
20
40
60
80
100
1968-72 1973-77 1978-82 1983-87 1988-92 1993-97Calendar period
5-ye
ar re
lativ
e su
rviv
al
period survival cohort survival
0
20
40
60
80
100
1968-72 1973-77 1978-82 1983-87 1988-92 1993-97Calendar period
5-ye
ar re
lativ
e su
rviv
al
period survival cohort survival
Figure 3a. Breast
Figure 3b. Colon
Figure 3c. Non-melanoma skin
0
20
40
60
80
100
1968-72 1973-77 1978-82 1983-87 1988-92 1993-97Calendar period
5-ye
ar re
lativ
e su
rviv
al
period survival cohort survival
0
20
40
60
80
100
1968-72 1973-77 1978-82 1983-87 1988-92 1993-97Calendar period
5-ye
ar re
lativ
e su
rviv
al
period survival cohort survival
0
20
40
60
80
100
1968-72 1973-77 1978-82 1983-87 1988-92 1993-97Calendar period
Data quality indices - Proportion of histologically verified and death certificate only cases, number andproportion of included and excluded cases by site: Singapore, 1993–1997 cases followed-up until 2001
Comparison of 1-, 3- and 5-year absolute and relative survival and 5-year age-standardized relative survivalby site: Singapore, 1993–1997 cases followed-up until 2001
Up-to-date 5-year relative survival estimates using cohort and period approaches by site and calendarperiod: Singapore, 1968–1997 cases followed-up until 1997
Table 8.
Period approach1973–77 1978–82 1983–87 1973–77 1978–82 1983–87
Up-to-date 10- and 15-year relative survival estimates using cohort and period approaches by site and calendar period: Singapore, 1968–1997 cases followed-up until 1997