Full text of DOI:10.1080/110241500447227

ORIGINAL ARTICLE
Survival of Women with Breast Cancer in Relation to
Smoking
Jonas Manjer,1 Ingvar Andersson,2 Go¨ran Berglund,3 Lennart Bondesson,4 Jens Peter Garne,5
Lars Janzon,1 Janne Malina4 and Sophia Matson1
2
3
4
From the Departments of 1Community Medicine, Radiology, Medicine and Pathology, Lund University, Malmo¨ University
5
˚
˚
˚
Hospital, Malmo¨, Sweden; and the Department of Surgery, Arhus University, Arhus Amtssygehus, Arhus, Denmark
Eur J Surg 2000; 166: 852–858
ABSTRACT
Objective: To compare survival of patients with breast cancer who had never smoked, were smokers, and who were ex-
smokers.
Design: Observational study.
Setting: City of Malmo¨, Sweden.
Patients: 792 patients with breast cancer diagnosed between 1977–1986 in the Malmo¨ mammographic screening trial.
Interventions: Follow-up of breast cancer cases through record-linkage with the Swedish Cause of Death Registry.
Main outcome measures: Death from breast cancer. Relative risk (RR) with 95% confidence interval (CI) of death from breast
cancer was calculated for different smoking groups using Cox’s proportional hazards analysis.
Results: During a mean follow-up of 12.1 years, 145 patients died of breast cancer. Breast cancer mortality was 1347/105
person-years in those who had never smoked, 1941/105 in smokers, and 1493/105 in ex-smokers. The crude RR for smokers and
ex-smokers, compared with those who had never smoked were 1.44 (1.01 to 2.06) and 1.13 (0.66 to 1.94), respectively. The RR
associated with smoking remained significant after adjustment for age and stage at diagnosis, 2.14 (1.47 to 3.10), and other
potential confounders.
Conclusions: Survival after breast cancer was, as expected, strongly related to stage at diagnosis. However, stage by stage there
was considerable variation between individual patients. We conclude that differences with regard to exposure to smoking
contribute to this heterogeneity.
Key words: breast cancer, survival, smoking.
INTRODUCTION
the subjects in the Malmo¨ mammographic screening
trial from 1977 to 1986 (3). During this period, 1034 of
these women were diagnosed with breast cancer of
whom 34 had been diagnosed previously. All but five
of them were treated at Malmo¨ University Hospital.
The primary objective of the trial was to find out
whether invitation to screening was associated with
reduced mortality from breast cancer (3).
A second objective was to study the clinical course
of the disease in relation to established and potential
prognostic markers, including mode of detection,
clinical stage, histological type, treatment, body height
and weight, menopause, parity, history of oophorec-
tomy, and use of hormone replacement therapy (HRT).
We used case sheets from the hospital records and a
database created by one of the authors [JPG] to retrieve
this information. Information on smoking habits was
retrieved by review of hospital records. Patients were
categorised as those who had never smoked, current
smokers, and ex-smokers. Information on smoking
habits was missing in 242 of the 1034 cases, so the
study group comprised 792 women.
Survival after breast cancer is strongly related to stage
at diagnosis (7), yet stage by stage there are
pronounced variations in the outcome between patients
even when differences with regard to treatment,
including surgery and adjuvant radiotherapy and
chemotherapy, are taken into account.
Several epidemiological studies have indicated that
smoking may increase the risk of breast cancer
(1, 10, 14). Whether smoking influences survival as
well has been given less scientific attention, but
according to at least two studies it seems that smokers
have a less favourable prognosis than non-smokers
(21, 22).
The aim of this follow-up-study of 792 women with
breast cancer diagnosed in the Malmo¨ mammographic
screening trial between 1977 and 1986 was to compare
survival rates in those who had never smoked, smokers,
and ex-smokers.
PATIENTS
In all 42283 women born between 1908 and 1932 were
Clinical stage was based on the TNM-system
 2000 Taylor & Francis. ISSN 1102–4151
Eur J Surg 166

Survival of women with breast cancer in relation to smoking 853
(tumour, nodes, metastases) (2). The histological
classification that was used was a modification of the
WHO classification proposed by Linell et al. (12). It
divides invasive ductal carcinoma into comedo and
tubuloductal carcinomas. Tubuloductal tumours are
further subdivided into two groups according to the
content of tubular structures.
Cause of death
Deaths from the time of recruitment until 31 December
1996 were retrieved by record linkage with the
Swedish Cause of Death Register. Underlying cause
of death was coded according to the 8th and 9th
versions of the ICD-code issued by WHO (20). During
a mean follow-up of 12.1 years there were 145 deaths
from breast cancer and 347 deaths from all causes.
Mean (SD) time from diagnosis to death was 5.2 years
(3.8) for those who had never smoked, 5.7 (3.8) for
smokers and 4.8 (3.2) for ex-smokers.
Statistical methods
Each woman was followed from diagnosis until death
or the end of follow up, 31 December 1996. All
statistical tests were made with the SPSS package (19).
Cox’s proportional hazards analysis was used to
calculate relative risks of dying of breast cancer for
smokers and ex-smokers (with a 95% confidence
interval (CI)), compared with those who had never
smoked, after adjustment for age and stage at diag-
nosis.
To adjust for other potential confounders, we used a
second model in which anthropometric measures,
menstrual status, parity, oophorectomy, whether they
had used HRT, and way of detection were introduced
as covariates using backward stepwise selection in the
Cox’s analysis (19). The score statistic (p < 0.05)
decided entry and removal was done by the like-
lihood-ratio statistic based on conditional parameters
estimates, (p < 0.10). Tumour histology and treatment
in relation to smoking habits were evaluated after
stratification of stage at diagnosis.
Differences in stage at diagnosis across smoking
categories were assessed by computing the odds for
stage II‡ compared with stage 0-I tumours in smokers
and ex-smokers in relation to never smokers. The odds
ratios were calculated by unconditional logistic regres-
sion analysis after adjustment for age at diagnosis,
menopausal status, BMI, and mode of detection.
The 242 cases who were excluded from the analysis
because of missing information on smoking habits,
were compared to the 792 in the study cohort in terms
of survival and prevalence of prognostic markers to
assess potential selection bias.
Eur J Surg 166

854 J. Manjer et al.
RESULTS
Comedo carcinoma and treatment by complete
mastectomy was most common among never smokers
with stage 0-I tumours, while comedo carcinoma and
treatment with adjuvant chemotherapy was most
common among ex-smokers in stage II‡ tumours,
(Table III).
Breast cancer mortality in women with stage 0-I
tumours was 443/105 person-years, in stage II: 2225/
105, in stage III: 11374/105 and in stage IV: 32927/105.
In all these groups, some women died of breast
cancer within three years of diagnosis but others
survived until the end of follow-up, more than 10
years (among women with stage IV tumours the
longest survival was 7.6 years).
The odds ratio for a stage II‡ compared with a stage
0-I tumour was 0.79 (0.52 to 1.21) in smokers
compared with 1.00 in never smokers and adjusted
for age at diagnosis, menopausal status, BMI, and
mode of detection. Corresponding comparison for ex-
smokers showed an odds ratio of 0.90 (0.56 to 1.44).
Smokers who died of breast cancer were compared
with those who did not, to assess potential modifiers,
(Tables IV and V). The influence of these factors on
survival was evaluated in a Cox’s proportional hazards
model with adjustment for age and stage. In the
analysis of the effect of adjuvant hormonal and
chemotherapy, menopausal status was added as a third
covariate. The only factor that significantly affected
survival in smokers was use of HRT, RR for ever use
compared with never use: 0.34 (0.12 to 0.97).
Breast cancer mortality was 1347/105 in those who
had never smoked, 1941/105 in smokers, and 1493/105
person-years in ex-smokers. Crude relative risk,
compared with those who had never smoked, were
1.44 (1.01 to 2.06) and 1.13 (0.66 to 1.94) for smokers
and ex-smokers, respectively, (Table I).
All-cause mortality was 3508/105 in never smokers,
4246/105 in smokers, and 2892/105 person-years in ex-
smokers. The age-adjusted relative risk was 1.46 (1.15
to 1.86) for smokers and 0.98 (0.67 to 1.44) for ex-
smokers.
The increased breast cancer mortality associated
with smoking remained significant after adjustment for
age and stage at diagnosis, RR 2.14 (1.47 to 3.10),
(Fig. 1), and other potential confounders, (Tables I and
II). The extended Cox’s analysis was limited to 555
cases because of missing information. Exclusion of 34
cases who had been diagnosed with breast cancer
before the screening trial did not change the associa-
tion. The analyses were repeated using only the 35
deaths from breast cancer that had been classified
according to ICD-9 and had been confirmed by
necropsy. In these 35 women the RR of death from
breast cancer, as compared to never smokers was 3.31
(1.57 to 6.95) for current smokers and 1.62 (0.51 to
5.07) for ex-smokers.
Breast cancer mortality was higher among the 242
cases for whom there was no information on smoking
5 person-years compared with 1517/105
habits, 2521/10
for the study group. The prevalence of stage III and IV
tumours was similarly higher in this group, (Table VI).
If all 242 women with missing information on smoking
had been never smokers, the RR of breast cancer death,
adjusted for stage and menopausal status, in current
compared with never smokers would have been 1.61
(1.15 to 2.26).
DISCUSSION
We conclude that differences in exposure to smoking
contribute to the heterogeneity in long term survival of
women with breast cancer. The appropriateness of that
conclusion should be assessed in relation to certain
methodological issues.
Differences between groups with regard to com-
pleteness of follow up and confirmation of end-points
could have confounded the results. As vital status was
updated on each patient and it has been confirmed that
cause of death in women with breast cancer is accurate
(3, 16), we consider it unlikely that the results were
confounded by biased retrieval or low validity of end-
points. In addition, when we restricted the analysis to
deaths that had been coded according to ICD-9 and
which had been confirmed by necropsy, the main
findings remained.
Misclassification with regard to exposure to smoking
is another relevant issue. Some of the women who did
not smoke were described as non-smokers in the
hospital records. These women were counted as never
Fig. 1. Survival from breast cancer according to smoking
status at the time of diagnosis.
Eur J Surg 166

Survival of women with breast cancer in relation to smoking 855
Table II. Age, mode of detection, and risk factors for breast cancer in relation to smoking habits. Figures, except
those in italics, are percentages
Factor
Never smoked (n = 491)
Current smoker (n = 216) Ex-smoker (n = 85)
Mean (SD) age at diagnosis, (years)
63.1 (7.2)
1.63 (0.06)
25.8 (4.3)
89.6
60.0 (7.0)
1.64 (0.06)
24.6 (4.1)
86.1
60.2 (7.1)
1.64 (0.06)
25.5 (4.9)
82.4
Mean (SD) body height, (meter)
Mean (SD) BMI, (kg/m2)
Postmenopausal
Nullipara
Oophorectomy
19.4
9.3
22.3
9.5
17.9
9.3
Ever taken HRT
Detected by mammography
14.7
50.0
22.0
47.7
19.0
50.6
Stage
0
I
II
III
IV
11.7
44.6
34.4
5.9
12.5
49.5
31.5
3.2
9.4
49.4
30.6
2.4
3.5
3.2
8.2
Table III. Histopathology and treatment in relation to smoking habits and stage at diagnosis. Figures are
percentages
Stage 0-I
Stage II‡
Never smoked Current smoker Ex-smoker Never smoked Current smoker Ex-smoker
Factor
(n = 275)
(n = 134)
(n = 50)
(n = 214)
(n = 82)
(n = 35)
Histological type
Tubular (‡‡‡‡, ‡‡‡)
Tubuloductal (‡‡, ‡, 0)
Comedo
18.8
26.8
24.3
4.4
4.8
21.0
25.4
23.9
16.4
9.7
4.5
20.1
20.4
30.6
12.2
14.3
6.1
11.0
34.9
31.1
12.9
10.0
–
7.3
36.6
36.6
11.0
8.5
12.1
24.2
45.5
15.2
3.0
Lobular
Invasive, varia
Carcinoma in situ
Treatment
Mastectomy
Local excision
Inoperable
16.3
–
–
72.1
27.9
0
61.9
38.1
0
62.0
38.0
0
87.7
7.5
4.7
85.0
7.5
7.5
82.9
5.7
11.4
Extent of operation
Restricted to breast
Removal of axillary lymph 86.0
nodes
Inoperable
Adjuvant hormonal therapy
Adjuvant chemotherapy
Postoperative radiotherapy
14.0
16.4
83.6
20.0
80.0
0.9
94.3
1.3
91.3
0
88.6
0
0.7
0
0
0
0
0
2.0
0
4.7
40.3
11.8
71.6
7.5
40.7
14.8
65.4
11.4
42.9
34.3
65.7
35.4
35.8
34.0
having smoked in our study, although a certain
proportion of them probably should have been labelled
ex-smokers. The prevalence of never smokers in our
cohort was however similar to that observed in health
surveys of women in corresponding age groups in the
city (11).
Changes in exposure is a problem in long-term
cohort studies. Studies on tobacco consumption in
Sweden (17) indicate that about 10% of those who
smoked in 1977 may have stopped by the end of
follow-up. Whether this figure is applicable to women
with breast cancer as well and whether breast cancer
may cause ex-smokers to take up smoking is unknown.
Use of HRT is associated with an increased
incidence of breast cancer (4). It remains controversial,
however, whether HRT influences survival (8, 18). If
HRT is associated with survival, misclassification as a
reason of changed behaviour during follow-up may
Eur J Surg 166

856 J. Manjer et al.
Table IV. Age, mode of detection, and risk factors for breast cancer in relation to death from breast cancer among
smokers. Figures, except those in italics, are percentages
Stage 0-I
Stage II‡
Current smokers
dead from breast
cancer (n = 12)
Current smokers
dead from breast
Rest of current
Rest of current
smokers (n = 46)
Factor
smokers (n = 122) cancer (n = 36)
Mean (SD) age at diagnosis, (years)
Mean (SD) body height, (meter)
Mean (SD) BMI, (kg/m2)
Postmenopausal
Nulliparous
Oophorectomy
58.7 (5.4)
1.67 (0.04)
26.9 (5.1)
91.7
20.0
9.1
59.9 (6.7)
1.63 (0.06)
24.0 (3.9)
87.7
21.7
6.3
59.9 (6.5)
1.61 (0.06)
26.0 (4.8)
88.9
25.0
15.2
60.8 (8.5)
1.66 (0.06)
24.8 (4.0)
78.3
22.2
13.6
Ever taken HRT
Detected by mammography
0.0
40.0
25.2
66.4
12.1
13.9
26.2
26.1
Table V. Histopathology and treatment in relation to death from breast cancer among smokers. Figures are
percentages
Stage 0-I
Stage II‡
Current smokers
dead from breast
cancer (n = 12)
Current smokers
dead from breast
smokers (n = 122) cancer (n = 36)
Rest of current
Rest of current
smokers (n = 46)
Factor
Histological type
Tubular (‡‡‡‡, ‡‡‡)
Tubuloductal (‡‡, ‡, 0)
Comedo
0.0
41.7
16.7
16.7
8.3
27.9
22.1
16.4
9.0
4.1
20.5
13.0
37.0
28.3
8.7
13.0
–
0.0
36.1
47.2
13.9
2.8
Lobular
Invasive, varia
Carcinoma in situ
Treatment
Mastectomy
Local excision
Inoperable
16.7
–
66.7
33.3
0.0
61.5
38.5
0.0
80.0
2.9
17.1
88.9
11.1
0.0
Extent of operation
Restricted to breast
Removal of axillary lymph nodes
Inoperable
Adjuvant hormonal therapy
Adjuvant chemotherapy
Postoperative radiotherapy
16.7
83.3
0.0
16.4
83.6
0.0
2.9
80.0
17.1
54.3
20.0
68.6
0.0
100.0
0.0
30.4
10.9
63.0
0.0
0.0
0.0
41.7
0.0
35.2
have confounded the results. As HRT is contraindi-
cated in women with breast cancer, we consider that
few non-users began HRT during follow-up and this
ought not to have confounded the results.
The 792 women in the study cohort cannot in terms
of mode of detection, stage at diagnosis, exposure to
known risk factors, and survival rate be considered
representative of all 1034 women diagnosed within the
Malmo¨ mammographic screening trial. This raises the
question of a potential selection bias associated with
smoking. However, if all cases with missing values
were assumed to have never smoked, there was still a
significantly higher mortality in current compared with
never smokers. It is our view that it is unlikely that the
results were confounded by biased selection of cases.
Differences in age, stage at diagnosis, and risk
factors were accounted for in the analysis. To what
extent the more common use of complete mastectomy
among never smokers with stage 0-I tumours may have
influenced the results is unknown.
Our results are in line with the findings in two other
studies (21, 22). It has been suggested that the lower
Eur J Surg 166

Survival of women with breast cancer in relation to smoking 857
Table VI. Age, mode of detection, and risk factors for breast cancer in cases were information on smoking was
available and missing respectively. Figures, except those in italics, are percentages
Information on smoking available
(n = 792)
Information on smoking missing
(n = 242)
Factor
Mean (SD) age at diagnosis, (years)
61.9 (7.3)
1.63 (0.06)
25.4 (4.4)
87.9
62.4 (8.3)
1.63 (0.06)
24.2 (4.2)
87.6
Mean (SD) body height, (meter)
Mean (SD) BMI, (kg/m2)
Postmenopausal
Nulliparous
20.1
26.9
Oophorectomy
Ever taken HRT
9.4
17.1
8.7
17.6
Detected by mammography
49.4
43.5
Stage
0
I
II
III
IV
11.6
46.5
33.2
4.8
21.5
38.8
23.2
6.3
3.9
10.1
survival rate of smokers may be the result of an
impairment of the immune defence system (6) or that
smoking may promote the development of more
aggressive, oestrogen-receptor negative, tumours (15).
Smokers and never smokers may differ in many
other respects that may influence survival. Breast
cancer survival correlates with several socioeconomic
circumstances (9). The increased mortality among
women of low socioeconomic status seems to be
related to stage at diagnosis (5). Whether socio-
economic deprivation may influence individual sus-
ceptibility as well is not known.
According to some studies it seems that high intake
of fat may be associated with reduced survival (23). In
nutrition surveys it has been found that smokers
consume more fat than never smokers (13).
Exposure to HRT before diagnosis was associated
with an improved prognosis among smokers. This
effect remained significant after adjustment for age and
stage at diagnosis. As use of HRT was more common
among smokers than among never smokers this may
have contributed to a more favourable prognosis in that
group. The association with smoking remained sig-
nificant, however, after adjustment for this potential
confounder.
REFERENCES
1. Ambrosone CB, Freudenheim JL, Graham S, et al.
Cigarette smoking, N-acetyltransferase 2 genetic poly-
morphisms, and breast cancer risk. JAMA 1996; 276:
1494–1501.
2. American Joint Committee on Cancer Manual for
staging of cancer. 3rd ed. Philadelphia: J. B. Lippincott,
1988: 27.
3. Andersson I, Aspegren K, Janzon L, et al. Mammo-
graphic screening and mortality from breast cancer: the
Malmo¨ mammographic screening trial. BMJ 1988; 297:
943–948.
4. Collaborative Group on Hormonal Factors in Breast
Cancer. Breast cancer and hormone replacement ther-
apy: collaborative reanalysis of data from 51 epidemio-
logical studies of 52702 women with breast cancer and
108411 women without breast cancer. Lancet 1997;
350: 1047–1059.
5. Farley TA, Flannery JT. Late-stage diagnosis of breast
cancer in women of lower socioeconomic status: public
health implications. Am J Public Health 1989; 79: 1508–
1512.
6. Ferson M, Edwards A, Lind A, Milton GW, Hersey P.
Low natural killer-cell activity and immunoglobulins
levels associated with smoking in human subjects. Int J
Cancer 1979; 23: 603–609.
7. Garne JP Invasive breast cancer in Malmo¨ 1961–1992—
an epidemiological study. Thesis. Malmo¨, Sweden:
Lund University, 1996.
8. Grodstein F, Stampfer MJ, Colditz GA, et al. Post-
menopausal hormone therapy and mortality. N Engl J
Med 1997; 336: 1769–1775.
9. Karjalainen S, Pukkala E. Social class as a prognostic
factor in breast cancer survival. Cancer 1990; 66: 819–
826.
10. Lash TL, Aschengrau A. Active and passive cigarette
smoking and the occurence of breast cancer. Am J
Epidemiol 1999; 149: 5–12.
11. Lindstro¨m M, Bexell A, Hanson BS, Isacsson S-O [The
health situation in Malmo¨: Report from a mailed
questionnaire survey, Spring 1994] in Swedish. Malmo¨,
ACKNOWLEDGEMENTS
¨
We thank the following: Mrs Inger Onnerheim for
administrative assistance and Dr Martin Lindstro¨m,
Malmo¨ University Hospital, author of “The Health
Situation in Malmo¨” from which data on smoking in
Malmo¨ were obtained. Financial support to this study
was received from the Ernhold Lundstro¨m Foundation.
Eur J Surg 166

858 J. Manjer et al.
Sweden: Department of Community Medicine, Malmo¨
University Hospital, 1995.
12. Linell F, Ljungberg O, Andersson I. Breast carcinoma:
aspects of early stages, progression and related prob-
lems. Acta Pathol Microbiol Scand 1980; (suppl 272):
120–134, 199–217.
13. Midgette AS, Baron JA, Rohan TE. Do cigarette
smokers have diets that increase their risks of coronary
heart disease and cancer. Am J Epidemiol 1993; 137:
521–529.
14. Morabia A, Bernstein M, Heritier S, Khatchatrian N.
Relation of breast cancer with passive and active
exposure to tobacco smoke. Am J Epidemiol 1996;
143: 918–928.
15. Morabia A, Bernstein M, Ruiz J, Heritier S, Berger SD,
Borisch B. Relation of smoking to breast cancer by
estrogen receptor status. Int J Cancer 1998; 75: 339–342.
16. Nystro¨m L, Rutqvist LE, Wall S, et al. Breast cancer
screening with mammography: overview of Swedish
randomised trials. Lancet 1993; 341: 973–978.
17. Pierce JP. International comparisons of trend in cigarette
smoking prevalence. Am J Public Health 1989; 79: 152–
157.
18. Schairer C, Gail M, Byrne C, et al. Estrogen replace-
ment therapy and breast cancer survival in a large
screening study. J Natl Cancer Inst 1999; 91: 264–270.
19. SPSS Inc. SPSS Advanced StatisticsTM 7.5. Chicago:
SPSS Inc, 1997.
20. The National Board of Health and Welfare, Centre for
Epidemiology. Causes of death 1996. The National
Board of Health and Welfare. Stockholm, Sweden,
1998.
21. Tominaga K, Andow J, Koyama Y, et al. Family
environment, hobbies and habits as psychosocial
predictors of survival for surgically treated patients
with breast cancer. Jpn J Clin Oncol 1998; 28: 36–41.
22. Yu G-P, Ostroff JS, Zhang Z-F, Tang J, Schantz SP.
Smoking history and cancer patient survival: a hospital
cancer registry study. Cancer Detect Prev 1997; 21:
497–509.
23. Zhang S, Folsom AR, Sellers TA, Kushi LH, Potter JD.
Better breast cancer survival for postmenopausal women
who are less overweight and eat less fat. Cancer 1995;
76: 275–283.
Submitted November 22, 1999; submitted after revision
January 31, 2000; accepted March 7, 2000
Address for correspondence:
Jonas Manjer, M.D.
Department of Community Medicine
Malmo¨ University Hospital
SE-205 02 Malmo¨
Sweden
Eur J Surg 166