Survival among Chinese women with lung cancer in Singapore: a comparison by stage, histology and smoking status

Lung Cancer 40 (2003) 237
/246 www.elsevier.com/locate/lungcan

Survival among Chinese women with lung cancer in Singapore: a comparison by stage, histology and smoking status Y.K. Tan a, T.C. Wee a, W.P. Koh a,*, Y.T. Wang b, P. Eng c, W.C. Tan d, A. Seow a a

Department of Community, Occupational and Family Medicine, National University of Singapore, 16 Medical Drive, MD3, Singapore 117597, Singapore b Department of Respiratory Medicine, Tan Tock Seng Hospital, Singapore c Department of Respiratory and Critical Care Medicine, Singapore General Hospital, Singapore d Department of Medicine, National University of Singapore, Singapore Received 11 October 2002; received in revised form 16 December 2002; accepted 30 December 2002

Abstract Lung cancer generally carries a poor prognosis and the determinants of survival have been of interest. However, survival estimates in Asian populations are scarce. This study describes survival rates and their determinants in Singapore Chinese women, a primarily non-smoking population. Three hundred and twenty-six Chinese women, diagnosed with primary lung carcinoma in three major hospitals in Singapore between April 1996 and December 1998, were followed up till 31 December 2000. The Kaplan
/Meier method was used for survival analysis. Two hundred and eighty (85.7%) died from the disease during follow-up. The median survival time was 0.7 years and the three-year survival was 15.8%. These survival rates are similar to those of Western populations, and they provide a basis for examining trends over time. Age at diagnosis was an independent prognostic factor [adjusted hazard ratio (relative risk) 1.4, 95% confidence intervals (CI) 1.1
/1.9 for women above 65 years relative to younger women]. Most (70.5%) tumours were stage III/IV at diagnosis. Three-year survival ranged from 72% among patients with stage I tumours to 7% for stage IV tumours. Overall, there was no survival difference among different histological types in all stages combined. When limited to stages I and II cancers, adenocarcinomas were associated with a better outcome relative to other histological subtypes combined (adjusted relative risk 0.4, 95% CI 0.1
/1.0). Smoking was an independent risk factor (adjusted relative risk 1.3, 95% CI 1.0
/1.8). Nevertheless, non-smokers comprised 57.4% of this series, highlighting the importance of increased awareness among health professionals and the public that lung cancer is not only a disease of smokers. The high proportion of late-stage tumours in this study and the impact of disease stage on outcome underline the importance of early detection in improving survival of lung cancer. # 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Lung cancer; Chinese; Women; Survival; Histology; Cancer staging; Smoking; Singapore

1. Introduction Lung cancer is a major health problem, accounting for 12.8% of cancer cases and 17.8% of cancer deaths worldwide [1]. Since 1990, the rate of lung cancer deaths in men has been on the decline while it has continued to rise in women, with the relative male-to-female lung cancer mortality ratio approaching unity [2,3]. Today it is one of the leading causes of cancer deaths in women worldwide [4]. In the United States, the age-adjusted

* Corresponding author. Tel.: /65-6874-4989; fax: /65-6779-1489. E-mail address: [email protected] (W.P. Koh).

incidence rate for lung cancer in females in 1973 was 18.3 per 100,000 and it grew steadily to 43.4 per 100,000 in 1998 [5]. In Singapore, an island republic with a population of 4 million, of which 75% are ethnic Chinese [6], lung cancer is now the third most common cancer amongst Chinese women and constitutes 9.8% of all cancers in this group. The age-standardised rate was 19.9 per 100,000 per year in the period 1993 to 1997 [7]. The prevalence of smoking among Chinese women in this country is low, and among those aged 18
/64 years the prevalence rate has remained about 3% in the last decade [6]. As a result, a significant proportion of lung cancers among Chinese women here occur in lifetime non-smokers.

0169-5002/03/$ - see front matter # 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0169-5002(03)00038-2

238

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

Although prognosis is generally poor and closely related to stage at diagnosis, several studies in Western populations have highlighted the role of histology and other factors in determining survival, and patterns which may aid in the search for molecular markers and therapeutic targets [8
/12]. In the present study, we sought to describe survival rates and their determinants in Singapore Chinese women, a primarily non-smoking Asian population.

up. All surviving cases after their last contact with their specialists as documented in the medical records were matched by computer linkage to the records of the Singapore Cancer Registry from 1 January 1996 to 31 December 2000. This allowed us to update the survival status and dates of death for these cases. All surviving cases as on 31 December 2000 were censored on that date. For all subjects, the length of follow-up for survival was calculated from the date of diagnosis to the date of death or censor.

2. Materials and methods

2.3. Data analysis

2.1. Selection of study population

The Kaplan
/Meier method was used for survival function estimates and the log-rank test for statistical testing of survival differences by subgroups. Hazard ratios (relative risks) and their 95% CI for risk of mortality were calculated using Cox’s proportional hazards regression model with multivariate adjustment. The variables included in the full model were stage, age group, histology and smoking status. Likelihood ratio x2 tests were used to test for differences between strata, and a P value of B/0.05 was considered statistically significant. All analyses were performed using the SPSS 10.0 statistical package (Chicago, Illinois).

Chinese women, with newly diagnosed lung cancer from three major hospitals in Singapore, were recruited consecutively for this study between April 1996 and September 1998. They were participants in a hospitalbased lung cancer case-control study, details of which are given elsewhere [13]. Only 5% of patients declined to participate when approached. Written consent was obtained for participation in the study and access to their medical records. Both pathologically and radiologically diagnosed cases of lung cancer were included in this study. Three hundred and thirty-three patients entered the study, and after excluding patients who were subsequently found to have secondary lung cancers (n/6) and recurrence of lung cancer (n/1), the final study population was 326. 2.2. Personal and clinical variables Demographic and information on smoking were obtained by standardized in-person interview by a trained nurse. All interviews took place within 3 months of diagnosis of cancer. In the questionnaire, a smoker was defined as someone who had ever smoked at least one cigarette a day for a year or more. The clinical staging of lung cancer in this study was based on the International System for staging lung cancer that has been adopted by the American Joint Committee on Cancer and the International Union Against Cancer [14]. To obtain information on histological subtype, tissue specimens of cases with pathological diagnoses were reviewed and classified independently by two pathologists. Stage of disease at diagnosis, based on the TNM staging system, and follow-up information regarding survival status, date and cause of death were obtained from the medical records. Cause of death was defined according to the underlying cause recorded in the death certificate, which was entered as such into the medical records. Only deaths from primary lung carcinoma were considered as events, subjects who died from other causes were considered as censored at the date of death. To our knowledge, no cases were lost to follow-

3. Results Of the 326 women with newly diagnosed lung cancer included in this study, follow-up information was obtained on all (100%), with a total of 369.6 years of follow-up. By the end of the follow-up period, 280 (85.9%) had died from lung cancer and two (0.6%) had died from causes other than lung cancer (one from perforated gastric ulcer and another from carcinoma of the stomach), while 44 subjects (13.5%) were still alive on 31 December 2000. Information on stage of disease at diagnosis was insufficient or unavailable for 53 (16.3%) women, and 33 (10.1%) did not have histological typing as they were diagnosed on radiological grounds. Subjects with missing values were excluded from the relevant analyses, except where indicated. The proportional distribution of histology, tumour stage, age and smoking status is shown in Table 1. The mean age at diagnosis was 65.3 years (range from 23 to 93 years). 3.1. Overall survival The overall median survival, and the 1- and 3-year survival rates, and the respective figures by stage, histology, age group and smoking status are shown in Table 1. The overall survival curve using the Kaplan
/ Meier method is shown in Fig. 1, while the survival curves based on age category, stage at diagnosis,

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

239

Table 1 Distribution of lung cancer cases by histology, tumour stage, age and smoking status (n /326) and their respective survival experience Survival rates (Standard error)

Overall survival of all patients Age 5/65 years /65 Tumour stage I II III IV Missing Histology Squamous cell carcinoma Small cell carcinoma Adenocarcinoma Large cell carcinoma/others Unknowna Smoking status Ever smoker Never smoker a

Number (%)

1-year

3-year

Median survival (years)

326 (100.0)

0.36 (0.03)

0.16 (0.02)

0.70

147 (45.1) 179 (54.9)

0.47 (0.04) 0.27 (0.03)

0.21 (0.03) 0.11 (0.02)

0.90 0.46

27 (8.3) 16 (4.9) 106 (32.5) 124 (38.0) 53 (16.3)

0.89 0.75 0.32 0.24 0.36

(0.06) (0.11) (0.05) (0.04) (0.07)

0.72 0.43 0.10 0.07 0.11

(0.09) (0.13) (0.03) (0.02) (0.04)


/ 1.96 0.70 0.44 0.63

54 (16.6) 25 (7.7) 164 (50.3) 50 (15.3) 33 (10.1)

0.31 0.32 0.44 0.24 0.30

(0.06) (0.09) (0.04) (0.06) (0.08)

0.13 0.07 0.20 0.16 0.09

(0.05) (0.05) (0.03) (0.05) (0.05)

0.56 0.50 0.89 0.44 0.48

139 (42.6) 187 (57.4)

0.25 (0.04) 0.45 (0.04)

0.08 (0.02) 0.21 (0.03)

0.50 0.89

These cases were radiologically diagnosed.

histological type, and smoking status are shown in Figs. 2
/5, respectively. 3.2. Age Among older women (above 65 years) in our study, there was a higher proportion of smokers than among

younger women (55.9 vs. 26.5%) and correspondingly more squamous and small cell carcinomas (33.0 vs. 13.6%). However, the proportion with late stage cancers at diagnosis (84
/85%) did not differ between the two age groups. The difference in survival between these two age-groups of patients was statistically significant (P B/ 0.001). The higher risk of death among older women

Fig. 1. Overall cumulative survival of all patients with lung cancer.

240

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

Fig. 2. Cumulative survival of patients with lung cancer based on age category.

remained significant after adjustment for tumour stage, smoking and histology (Table 2). 3.3. Tumour stage As expected, tumour stage was an important prognostic indicator. The 72% 3-year survival rate of subjects with stage I tumours contrasted greatly with the dismal 7% 3-year survival rate for patients with stage IV tumours. The differences in survival rates among subjects in stages I, II and III/IV were statistically significant. However, the difference between subjects with stages III and IV tumours did not reach statistical significance, hence these two groups of patients were combined as one in the multivariate analysis. On multivariate analysis, tumour stage remained the variable with the greatest impact on prognosis. Patients with stages II and III/IV tumours had three and eight times the risk of death compared with those with stage I tumours, respectively (Table 2). 3.4. Histology There were 33 (10.1%) of patients with unknown histology. The mean age of diagnosis for this group was 70.8 years and was significantly older than the group with known histology which had a mean age of diagnosis at 64.7 years (P /0.009). From the Kaplan
/Meier analysis, subjects with adenocarcinomas had the longest median survival time (0.89 years) compared with small cell (0.50 years) and

squamous cell (0.56 years) carcinomas. The 1-year survival rate for small cell lung tumours was similar to the other known histological types but its 3-year survival rate was distinctly the lowest among the known histological types. However, the difference in median survival time between different histological subtypes was not statistically significant using the log-rank test. The crude relative risks calculated using the Cox model (Table 2) indicated that relative risk of death for patients with squamous/small cell carcinoma was between 30 and 40% higher than adenocarcinoma, with risk for both squamous and small cell being very similar. We compared the stage distribution among adenocarcinomas and tumours of other histological types, to see if this could have accounted for the differences in survival. Among patients with adenocarcinomas, 19.7% had early (stages I and II) and 80.3% had late (stages III and IV) tumours. The corresponding figures for the other histological types combined, were 11.7 and 88.3%, respectively. Patients with adenocarcinoma tended to be younger, and a smaller proportion were ever smokers, than those with other histological types. On adjusting for stage and these variables, there was no significant difference in risk of death between the various histological types (Table 2). Among the early stage (I and II) cancers alone (n/ 43), 3-year survival rate for adenocarcinomas was 71.5%, compared with 40.0% for other histological types combined. When stratified by stage at diagnosis, adenocarcinomas had a more favourable prognosis than other known histological types in stages I and II cancers

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

241

Fig. 3. Cumulative survival of patients with lung cancer based on tumour stage at diagnosis.

but not in stages III and IV cancers. Among early stage cancers, the age- and smoking-adjusted relative risk was 0.4 (95% CI 0.1
/1.0) for adenocarcinoma compared with other histological types combined. On the other hand, among late stage cancers, the corresponding figure was 1.1 (0.8
/1.5).

3.5. Smoking The survival of ever smokers was significantly poorer than the lifetime non-smokers (P B/0.001). Smokers tended to be older at diagnosis (71.9% older than 65 years vs. 42.2% of non-smokers) and a slightly higher

Fig. 4. Cumulative survival of patients with lung cancer based on histology.

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

242

Table 2 Crude and adjusted relative risk (and 95% CI) of mortality by relevant prognostic factors

Age 65 years and below More than 65 years Stage at diagnosis Stage I Stage II Stages III and IV Histology Adenocarcinoma Squamous cell carcinoma Small cell carcinoma Large cell carcinoma /other/unknown Smoking status Never smokers Ever smokers a

Crude relative risk

95% CI

Adjusteda relative risk

95% CI

1.0 1.5

1.2
/2.0

1.0 1.4

1.1
/1.9

1.0 3.5 8.2

1.4
/8.4 4.0
/16.8

1.0 2.9 7.8

1.2
/6.9 3.8
/16.0

1.0 1.3 1.4 1.4

0.9
/1.7 0.9
/2.1 1.0
/1.8

1.0 0.9 0.7 1.3

0.6
/1.3 0.4
/1.2 0.9
/1.8

1.0 1.7

1.3
/2.2

1.0 1.3

1.0
/1.8

Adjusted for all other variables in the table using Cox’s proportional hazards model.

proportion had late stage tumours (89.6% stages III and IV vs. 80.4% among non-smokers). As expected, the proportion of squamous and small cell carcinomas was markedly higher among smokers than non-smokers (47.3 vs. 8.6%). However, on multivariate analysis, a significant difference in risk between smokers and nonsmokers persisted after adjusting for stage, age and histology, although it was attenuated. There was also a statistically significant effect from the interaction between age of diagnosis and smoking (Wald value /4.72, P /0.030). When stratified by age and adjusted for stage and histology, there was an increased risk of 2.0

(95% CI 1.3
/3.2) among smokers aged 65 and below. However, this increased risk was no longer present in the age group 66 years and older (adjusted relative risk 0.9, CI 0.6
/1.4).

4. Discussion In summary, we show that survival among female Chinese lung cancer patients in this study is poor, with a median survival time of 0.7 years, and only one-third (36%) of patients being alive one year after diagnosis.

Fig. 5. Cumulative survival of patients with lung cancer based on smoking status.

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

Not unexpectedly, older patients above 65 years of age did worse than the younger patients. Tumour stage is the most important predictor of survival, and since about 70% of patients were diagnosed at stages III or IV of the disease, this would largely account for the poor prognosis observed in this study. In addition, we find that while overall, there is no significant difference in survival between the various histological subtypes, early stage adenocarcinomas appear to have a better prognosis than other histological types at the same stage, although it should be noted that the numbers supporting this observation are small. Since a large proportion (57%) of our subjects were non-smokers, we were able to examine differences in survival by smoking status and find that a history of smoking confers a 30% higher risk of death after adjustment for other relevant variables. When stratified by age, this difference was particularly marked among those below 65 years. While this study is limited by its small sample size, its strength lies in the availability of accurate data on smoking and histological subtype. It was carried out in an ethnically and gender-wise homogenous population, in a small city-state where there is uniform and good access to specialized medical care [15]. Follow-up was complete and date and cause of death information would have a high degree of accuracy because of the nature of the vital registration system in this country. Although the current study has the limitation of being hospital-based, we were able to recruit consecutive cases over thirty months in the hospitals in our study, which together accounted for approximately 75% of all cases of lung cancer diagnosed among Singapore Chinese females (Singapore Cancer Registry, unpublished data). No specific catchment boundaries exist for the hospitals in this country, and due to the small size of the country and ready accessibility to hospital care, our study population is likely to be representative of patients from the entire country. There remains a slight possibility that patients who choose to be admitted to a private hospital may differ from those in this series in ways which would affect their survival, and this should be borne in mind when interpreting the results. Generally, our results are comparable with survival rates from studies in Europe and the United States (Table 3). Janssen-Heijnen et al. looked at the variation in survival of patients with lung cancer in Europe among 173,448 lung cancer cases from 1985 to 1989 and found that with the exception of a slight improvement in short term survival for small cell lung cancer, survival has been poor since 1978. The age-standardised 1-year survival rates varied greatly amongst European countries, ranging between 40 and 24% [16]. Some studies have demonstrated that, for reasons that are not currently clear, women tend to have a better survival than men [11,12]. Although this study involves only women, recent data from Chia et al. suggests that rates

243

between men and women in Singapore are not dissimilar [17]. Relative to other Asian countries, the survival rates of lung cancer in Singapore in Chia et al.’s study appear to be higher than those in Qidong (China), Madras (India) and Chiang Mai (Thailand), but lower than in Japan and Shanghai (China) [18]. Our findings are consistent with other reports of poorer survival among older patients [2,16]. In Europe, the 1-year survival for the age group 15
/44 years was 42% compared with 21% for those 75 years and older, a finding that was attributed to less access to specialised care and the presence of comorbidity at the time of diagnosis [16]. Another reason for better survival in the younger age group could be the fact that these patients tend to receive more aggressive treatment [19], probably due to the relative absence of comorbidity. In the United States, Fry et al. noted out that the survival differences among histology group to some extent may have accounted for the demographic differences by age, as adenocarcinomas also formed a greater proportion of cases among younger patients in that series [2]. That tumour stage is an important prognostic indicator is also a consistent finding across studies. Fry et al. showed that 5-year relative survival rates for stages I and II were 42 and 23%, respectively, while the figures for stage IIIA, IIIB and stage IV were 11, 5 and 1% [2]. Merrill et al. also illustrated the benefit of early stage diagnosis and showed that even among cases with an initial survival of 2 years after diagnosis, the 5-year survival probability continues to be influenced by their stage of disease at diagnosis [20]. In addition, our study highlights the importance of early detection of lung cancer, and its impact on prognosis. However, screening for lung cancer remains controversial, and its effectiveness unproven despite several large trials [21
/24]. Adenocarcinomas form the largest proportion (50.3%) of lung tumours in this series, and a similar preponderance or increase has been reported in other studies [2,25,26]. Some of these have also demonstrated a difference in survival according to histology. They suggest that patients with squamous cell carcinomas or adenocarcinomas (or more specifically bronchioalveolar carcinoma in some studies [20]), have significantly better overall survival compared with small cell or undifferentiated carcinoma [2,16,20,26]. Although no clear explanation has emerged to date, it has been suggested that the later stage at which small cell carcinomas are commonly diagnosed may explain part of this [16], and possibly the more efficient radiographic detection of peripheral tumours has contributed to better survival among patients with adenocarcinomas [27]. Among early stage tumours, adenocarcinomas were associated with better prognosis, suggesting that differences in survival by histology may not be entirely due to differences in distribution by stage, and possibly pointing to inherent biological behaviour as a factor. Some

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

244

Table 3 Comparison of survival among lung cancer patients in other studies

Survival (%) 1-year 3-year 5-year

Malmberg et al. [26], Sweden, (n/3285)a

Fry et al. [2], United States, (n/713,043)

Janssen-Heijnen et al. [16], Europe, (n/173,448)b

Current study, (n/326)

27.9 11.4 8.3

40 17 13

29 13 11

36 16
/

a

In the study by Malmberg et al. figures here represent crude survival while those by Fry et al. and Janssen-Heijnen et al. represent relative survival. b In the study by Janssen-Heijnen et al., n/173,448 represents combined male and females cases whereas the figures quoted here are for the subset of female patients only.

[28,29], but not all [30,31], studies have demonstrated possible roles and prognostic value of molecular markers involved in apoptosis (e.g. p53), angiogenesis (Factor VIII) or adhesion (CD44). Among histological types, and within early stage adenocarcinomas, genetic markers such as nuclear DNA content have also been used as predictors of tumour aggressiveness [32
/34]. The biological basis for these differences, and the use of such markers for substaging of lung tumours, will require further study. In the current series, there is no overall survival advantage for adenocarcinomas when stage is taken into account. The small cell tumours had a comparable 1-year survival rate but a more dismal 3year rate compared to the other histological subtypes. However, the number of small cell cancers in the current series is too small for the difference in long-term prognosis (3 years or more) to reach statistical significance. Such data are similar to observations in Western populations which have also shown a comparable 1-year survival rate but lower 3- and 5-year rates in small cell lung cancers compared to the other histology subtypes [2]. This has been attributed to the aggressive use of chemotherapy making a difference in short-term survival of small cell lung cancers [35]. It is estimated that smoking is responsible for about 80% of lung cancer cases in Western populations [36]. Few studies are able to adequately document survival differences between smokers and non-smokers because of lack of reliable information on smoking status, or due to the small numbers of non-smokers among lung cancer patients. In the current study, data on smoking was obtained by in-person interview, and while it was not possible to use biochemical means to validate selfreported smoking status, we conducted a separate validation study to assess the accuracy of the information by cross-checking with the patients’ historical and current hospital records. From the 244 cases reviewed, we obtained a concordance rate of 96.3% and a kappa statistic of 0.93, indicating a very high degree of reproducibility. In addition, every effort was made to minimize recall bias with regard to smoking status, such as using the general term ‘women’s health’ rather than

specifically ‘lung cancer’ in introducing the study to the interviewee. However, this does not rule out the possibility of recall bias which is a limitation inherent in a retrospective study of this nature. The better survival among non-smokers was best seen in patients below the median age of our study population (65 years old and younger). A previous study among women with lung cancer in US showed that smokers had a 1.74-fold increased risk of dying compared to non-smokers [37]. Another recent study in Finland demonstrated a 13% higher risk of death among heavy, compared with light, smokers; although the effect was not statistically significant [38]. A worse prognosis for lung cancer in smokers may be attributable to the presence of other cardiorespiratory comorbidities associated with smoking [39], or possibly to other reasons yet to be understood. In this study, there were differences in histological type between smokers and nonsmokers, the latter having a higher proportion of adenocarcinomas. However, this did not account entirely for survival differences, as the relationship between smoking and poorer outcome remained significant after adjusting for this variable.

5. Conclusion The results of this study form a baseline by which trends in survival over time may be compared, Smoking is an independent prognostic factor, and smoking cessation cannot be overly emphasized especially in the younger patients with lung cancer. However, in this population of women, among which 57% of lung cancer cases occurred among non-smokers, there is also a need for increased awareness among health professionals and the public that lung cancer is not only a disease of smokers. Earlier diagnosis remains a key to better outcome, and while risk factor modification remains the target of public health interventions against lung cancer, which among non-smoking women include exposure to environmental tobacco smoke, indoor air pollution exposure, as well as occupational exposures

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

[40], there is still an urgent need for sustained effort and research into means by which these tumours may be detected earlier.

Acknowledgements The authors are grateful to Drs Wee-Teng Poh and Ming Teh for reviewing the pathological specimens, Esther Quek for valuable assistance in the data collection, the Medical Boards of the National University Hospital, Singapore General Hospital and Tan Tock Seng Hospital for permission to carry out this study in their institutions, and the Singapore Cancer Registry.

References [1] Parkin DM, Pisani P, Ferlay J. Global cancer statistics. CA Cancer J Clin 1999;49:33
/64. [2] Fry WA, Phillips JL, Menck HR. Ten-year survey of lung cancer treatment and survival in hospitals in the United States: a national cancer data base report. Cancer 1999;86:1867
/76. [3] Baldini EH, Strauss GM. Women and lung cancer: waiting to exhale. Chest 1997;112:229S
/34S. [4] Coleman MP, Esteve J, Damiecki P, Arslan A, Renard H. Time trends in cancer incidence and mortality. IARC Scientific Publications, Lyon, 1993:1
/806. [5] National Cancer Institute D, Surveillance Research Program, Cancer Statistics Branch, Surveillance, Epidemiology, and End Results (SEER) Program Public-Use Data (1973
/1998), 2001. [6] Epidemiology and Disease Control Department, Ministry of Health, Singapore, National Health Survey 1998, 1999. [7] Chia KS, Seow A, Lee HP. Shanmugaratnam, Cancer incidence in Singapore 1993
/1997, Singapore Cancer Registry, Singapore, No. 5., 2000. [8] Muers MF, Shevlin P, Brown J. Prognosis in lung cancer: physicians’ opinions compared with outcome and a predictive model. Thorax 1996;51:894
/902. [9] Gregor A, Thomson CS, Brewster DH, Stroner PL, Davidson J, Fergusson RJ, Milroy R, Scottish Cancer Trials Lung Group, Scottish Cancer Therapy Network. Management and survival of patients with lung cancer in Scotland diagnosed in 1995: results of a national population based study, Thorax, 2001;56:212
/217. [10] Storm HH, Dickman PW, Engeland A, Haldorsen T, Hakulinen T. Do morphology and stage explain the inferior lung cancer survival in Denmark? Eur Respir J 1999;13:430
/5. [11] Ries LA. Influence of extent of disease, histology, and demographic factors on lung cancer survival in the SEER populationbased data. Semin Surg Oncol 1994;10:21
/30. [12] Bouchardy C, Fioretta G, De Perrot M, Obradovic M, Spiliopoulos A. Determinants of long term survival after surgery for cancer of the lung: a population-based study. Cancer 1999;86:2229
/37. [13] Seow A, Poh WT, Teh M, Eng P, Wang YT, Tan WC, Yu MC, Lee HP. Fumes from meat cooking and lung cancer risk in Chinese women. Cancer Epidemiol Biomarkers Prev 2000;9:1215
/21. [14] Mountain CF. Revisions in the International System for Staging Lung Cancer. Chest 1997;111:1710
/7. [15] Lim MK. Health care systems in transition. II. Singapore, Part I. An overview of health care systems in Singapore. J Public Health Med 1998;20:16
/22.

245

[16] Janssen-Heijnen ML, Gatta G, Forman D, Capocaccia R, Coebergh JW. Variation in survival of patients with lung cancer in Europe, 1985
/1989. EUROCARE Working Group. Eur J Cancer 1998;34:2191
/6. [17] Chia KS, Du WB, Sankaranarayanan R, Sankila R, Seow A, Lee HP. Population-based cancer survival in Singapore, 1968 to 1992: an overview. Int J Cancer 2001;93:142
/7. [18] Sankaranarayanan R, Black RJ, Parkin DM. Cancer survival in developing countries. IARC Scientific Publication no. 145. 1998. [19] Kuo CW, Chen YM, Chao JY, Tsai CM, Perng RP. Non-small cell lung cancer in very young and very old patients. Chest 2000;117:354
/7. [20] Merrill RM, Henson DE, Barnes M. Conditional survival among patients with carcinoma of the lung. Chest 1999;116:697
/703. [21] Fontana RS, Sanderson DR, Woolner LB, Taylor WF, Miller WE, Muhm JR, Bernatz PE, Payne WS, Pairolero PC, Bergstralh EJ. Screening for lung cancer. A critique of the Mayo Lung Project. Cancer 1991;67:1155
/64. [22] Diederich S, Wormanns D, Heindel W. Low-dose CT: new tool for screening lung cancer? Eur Radiol 2001;11:1916
/24. [23] Tsukada H, Kurita Y, Yokoyama A, Wakai S, Nakayama T, Sagawa M, Misawa H. An evaluation of screening for lung cancer in Niigata Prefecture, Japan: a population-based case-control study. Br J Cancer 2001;85:1326
/31. [24] Strauss GM. The Mayo Lung Cohort: a regression analysis focusing on lung cancer incidence and mortality. J Clin Oncol 2002;20:1973
/83. [25] Seow A, Duffy SW, Ng TP, McGee MA, Lee HP. Lung cancer among Chinese females in Singapore 1968
/1992: time trends, dialect group differences and implications for aetiology. Int J Epidemiol 1998;27:167
/72. [26] Malmberg R, Bergman B, Branehog I, Larsson S, Olling S, Wernstedt L. Lung cancer in West Sweden 1976
/1985. A study of trends and survival with special reference to surgical treatment. Acta Oncol 1996;35:185
/92. [27] Melamed MR, Flehinger BJ, Zaman MB. Impact of early detection on the clinical course of lung cancer. Surg Clin North Am 1987;67:909
/24. [28] D’Amico TA, Massey M, Herndon JE, 2nd, Moore MB, Harpole DH, Jr.. A biologic risk model for stage I lung cancer: immunohistochemical analysis of 408 patients with the use of ten molecular markers. J Thorac Cardiovasc Surg 1999;117:736
/ 43. [29] D’Amico TA, Aloia TA, Moore MB, Herndon JE, 2nd, Brooks KR, Lau CL, Harpole DH, Jr.. Molecular biologic substaging of stage I lung cancer according to gender and histology. Ann Thorac Surg 2000;69:882
/6. [30] Greatens TM, Niehans GA, Rubins JB, Jessurun J, Kratzke RA, Maddaus MA, Niewoehner DE. Do molecular markers predict survival in non-small-cell lung cancer? Am J Respir Crit Care Med 1998;157:1093
/7. [31] Pastorino U, Andreola S, Tagliabue E, Pezzella F, Incarbone M, Sozzi G, Buyse M, Menard S, Pierotti M, Rilke F. Immunocytochemical markers in stage I lung cancer: relevance to prognosis. J Clin Oncol 1997;15:2858
/65. [32] Asamura H, Nakajima T, Mukai K, Shimosato Y. Nuclear DNA content by cytofluorometry of stage I adenocarcinoma of the lung in relation to postoperative recurrence. Chest 1989;96:312
/8. [33] Tanaka I, Masuda R, Furuhata Y, Inoue M, Fujiwara M, Takemura T. Flow cytometric analysis of the DNA content of adenocarcinoma of the lung, especially for patients with stage 1 disease with long term follow-up. Cancer 1995;75:2461
/5. [34] Hofmann HS, Knolle J, Bahn H, Klapperstuck T, Lautenschlager C, Neef H. Flow cytometric analysis of DNA content and proliferation and immunohistochemical staining of Ki-67 in non-small cell lung cancer. J Cardiovasc Surg 2001;42:555
/60.

246

Y.K. Tan et al. / Lung Cancer 40 (2003) 237
/246

[35] Janssen-Heijnen ML, Schipper RM, Klinkhamer PJ, Crommelin MA, Coebergh JW. Improvement and plateau in survival of small-cell lung cancer since 1975: a population-based study. Ann Oncol 1998;9:543
/7. [36] Hoffman PC, Mauer AM, Vokes EE. Lung cancer. Lancet 2000;355:479
/85. [37] Hinds MW, Yang HY, Stemmermann G, Lee J, Kolonel LN. Smoking history and lung cancer survival in women. J Natl Cancer Inst 1982;68:395
/9.

[38] Holli K, Visakorpi T, Hakama M. Smoking and survival from lung cancer. Acta Oncol 1999;38:989
/92. [39] Doyle JT. Risk factors in arteriosclerosis and cardiovascular disease with special emphasis on cigarette smoking. Prev Med 1979;8:264
/70. [40] Brownson RC, Alavanja MC, Caporaso N, Simoes EJ, Chang JC. Epidemiology and prevention of lung cancer in nonsmokers. Epidemiol Rev 1998;20:218
/36.