Breast Cancer in Sabah, Malaysia

RESEARCH COMMUNICATION

Breast Cancer in Sabah, Malaysia: A Two Year Prospective

Study
Benjamin Dak Keung Leong1, Jitt Aun Chuah2, Vinod Mutyala Kumar2, Cheng
Har Yip1

Abstract
Introduction: Malaysian women have a 1 in 20 chance of developing breast cancer in their lifetime. Sabah,
formerly known as North Borneo, is part of East Malaysia with a population of 3.39 million and more than 30
ethnic groups. We conducted a 2 year prospective epidemiological study to provide unreported data of breast
cancer from this part of the world and to recognise which particular group of patients are more likely to present
with advanced disease. Methods: All newly diagnosed breast cancers seen at the Queen Elizabeth Hospital,
Kota Kinabalu, from January 2005 to December 2006 were included in the study. Patient and tumour
characteristics, including age, race, education, socioeconomic background, parity, practice of breast feeding,
hormonal medication intake, menopausal status, family history, mode of presentation, histology, grade, stage of
disease and hormonal receptors status were collected and analysed. Results: A total of 186 patients were seen.
The commonest age group was 40 to 49 years old (32.3%). Chinese was the commonest race (30.6%) followed by
Kadazan-Dusun (24.2%). The commonest histology was invasive ductal carcinoma (88.4%). Stages at presentation
were Stage 0- 4.8%, Stage I- 12.9%, Stage II- 30.1%, Stage III- 36.6% and Stage IV- 15.6%. The estrogen and
progesterone receptor status was positive in 59.1% and 54.8% of cases, respectively. 73.7% of Chinese patients
presented with early cancer compared to 36.4% of the other races. Patients who presented with advanced disease
were also poor, non-educated and from rural areas. 20.4% of patients defaulted treatment; most of them opted
for traditional alternatives. Conclusions: Sabahan women with breast cancer present late. Great efforts are
needed to improve public awareness of breast cancer, especially among those who have higher risk of presenting
with advanced disease.
Key Words: Breast cancer - late presentation - advanced stage - poor, rural - ethnic groups

Asian Pacific J Cancer Prev, 8, 525-529

Introduction largest state in Malaysia. Sabah has an estimated

population of 3.39 million with more than 30 ethnic
Breast cancer is the commonest cancer among women groups. The largest indigenous ethnic group is Kadazan-
in Malaysia as well as worldwide. Globally, it was Dusun (17.8%), followed by Bajau and Murut. The largest
estimated there were 1,150,000 new cases with 410,000 non-indigenous ethnic group is Chinese (9.6%). To date,
deaths in 2002 (Ferlay et al., 2002). Breast cancer has data on breast cancer from Sabah has been poorly reported
high incidence rates in USA, Europe, Australia, New and published. National data on breast cancer published
Zealand and in some parts of South America, especially in the National Cancer Registry of Malaysia 2002 did not
Uruguay and Argentina and is relatively less common in include data from Sabah due to inadequate reporting of
Africa and Asia (Stewart and Kleihues, 2003). Overall cases (Lim and Yakaya, 2003).
worldwide Age Standardised Rate (ASR) of breast cancer Queen Elizabeth Hospital, situated in the capital of
was estimated to be 37.4 per 100,000 (Ferlay et al., 2002). Sabah, Kota Kinabalu, is the main public tertiary referral
In Malaysia, there were 3,738 new cases reported in centre in the state. We conducted a prospective
20033. Breast cancer made up 31% of all newly diagnosed epidemiological study in our hospital on breast cancer to
cancer among Malaysian women with an overall ASR of provide unreported data on breast cancer in Sabah and,
46.2 per 100 000 (Lim and Yakaya, 2003). Each Malaysian with statistical analysis, to recognise which particular
woman has a 1 in 20 chance to develop breast cancer in groups of patients who are more likely to present with
her lifetime. Sabah, formerly a British Colony known as advanced disease. We hope that a clearer epidemiological
North Borneo, is part of East Malaysia and the second picture of breast cancer in Sabah will lead to appropriate

1
Department of Surgery, University Malaya Medical Centre, Kuala Lumpur 2Department of Surgery, Queen Elizabeth Hospital,
Kota Kinabalu, Malaysia *For Correspondence: Fax: +60-3-79586360 Email: [email protected]

Asian Pacific Journal of Cancer Prevention, Vol 8, 2007 525

Benjamin Dak Keung Leong et al
measures to improve the outcome of management of the square test was used for associations between categorical
disease in Sabah and Malaysia as a whole. variables and the Student’s t test for associations between
continuous and categorical variables. Logistic regression
Materials and Methods analysis was used to identify independent covariates
predicting patients with advanced breast cancer. The level
The subjects of our study were all newly diagnosed of statistical significance was set at P < 0.05.
breast cancer patients seen in our hospital from January
2005 to December 2006. We included patients who were Results
referred from private institutions for continuation of
treatment at different stages of management of the disease. In 2005 and 2006, we saw a total of 186 cases of newly
We excluded patients who presented with recurrent breast diagnosed breast cancer. 99.5% of them were female and
cancer and those who refused to participate in the study. 0.5%, male. The most common age group was 40 to 49
Patients’ data including age, race, place of residence, level years old (32.3%) (Figure 1), with a mean age of 51.0
of education, religion, income, age of menarche, marital (standard deviation, 11.0). The youngest and oldest
status, parity, age of first child birth, practice of breast patients were 24 and 83 year old respectively. 54.8% of
feeding, menopausal status, hormonal medication intake, them were post-menopausal. The most common race
family history, mode of presentation and as well as tumour affected was Chinese (30.6%), followed by Kadazan-
characteristics including tumour site, type, grade, stage Dusun (24.2%) and Bajau (13.4%) (Figure 2).
and hormonal receptors status were collected and analysed. Interestingly, our subjects included 5 (2.7%) illegal
Data were collected at the dedicated breast clinic, general immigrants (Filipinos and Indonesians). 46.8% of patients
surgical clinic and wards of the hospital prospectively for were from the rural area (villages) and 30.1% of them
2 years. Staging of the disease was in accordance to the had never received any formal education. 56.5% of
American Joint Committee on Cancer (AJCC) Cancer patients earned less than 1,000 Ringgit Malaysia (RM)
Staging Manual, sixth edition. Advanced breast cancer is per month (330 Australian Dollars). 15.6% of patients
defined as breast cancer at either stage 3 or 4 of the disease. were referred from private institutions for continuation
Statistical Package for the Social Sciences (SPSS of treatment, majority due to financial constraint on
version 15.0) was used for statistical analysis. The Chi chemotherapy and radiotherapy.
The proportion of patients with risk factors are as
follows: early menarche (before 13 year old) in 31.7%,
unmarried in 9.1%, nulliparity in 16.7%, late first child
birth (after 30 year old) in 11.3%, late menopause (after
55 year old) in 4.3%, non-breast feeding in 29.6%,
hormonal medication (oral contraceptive pill and
hormonal replacement therapy) intake in 32.3% and
Numbers

positive family history in 11.3%.

Table 1 presents the proportion of mode of clinical
presentations and tumour characteristics. The commonest
mode of presentation was with a breast lump, 78.0%. 9.1%
of patients presented to us with ulcerative mass and only
2.2% presented with suspicious lesion on mammography.
51.6% of the tumours were on the right side and 46.2%
on the left. 2.2% had bilateral disease. The commonest
Figure 1. Age Distribution of Breast Cancer Cases
Numbers

Numbers

Figure 2. Race Distribution of Breast Cancer. KadDus:

Kadazan-Dusun Figure 3. Stage at Presentation of Breast Cancer
526 Asian Pacific Journal of Cancer Prevention, Vol 8, 2007
 Breast Cancer in Sabah, Malaysia
with PR-negative and PR-positive tumours where 66.7%
and 41.0% respectively presented as advanced disease
(P=0.0001).
The commonest stage of breast cancer at presentation
in our institution was stage III- 36.6%, followed by stage
II- 30.1%, stage IV- 15.6%, stage I- 12.9% and only 4.8%
presented at stage 0 (Fig. 3). Further sub-staging are as
follows: stage 0- 4.8%, stage I- 12.9%, stage IIA- 20.4%,
stage IIB- 9.7%, stage IIIA- 17.2%, stage IIIB- 11.3%,
stage IIIC- 8.1%, stage IV- 15.65%. Majority of patients
(52.2%) presented with advanced disease (Figure 4).
Patient characteristics associated with advanced
disease presentation which were statistically significant
in univariate analysis were non-Chinese race, patients
Figure 4. A Patient Presenting with an Ulcerative from rural area, patients with income of less than 1000
Lesion RM per month and the non-educated (Table 2). 63.6% of
site was the upper outer quadrant with 53.8%. Infiltrating non-Chinese and only 26.3% of Chinese presented with
ductal carcinoma was the commonest histological type advanced disease (P = 0.0001). 70.1% of patients from
(88.4%) with grade II the commonest grade (59.4%) rural area (P = 0.0001), 60.0% of patients who earn less
according to Bloom and Richardson grading. 59.1% of
tumours were estrogen receptor (ER) positive and 54.8% Table 2. Correlation between Patient Characteristics
were progesterone receptor (PR) positive. 65.9% of ER- and Stage at Presention
negative tumours presented at advanced stage compared Patient Stages Stages P Value
to ER-positive, 42.7% (P=0.02). Similar pattern was seen Characteristics 0, I & II III & IV

Table 1. Clinical Presentations and Tumour Mean age(year) 51.2 50.8 0.822
Race
Characteristics
Chinese 42 (73.7%) 15 (26.3%) 0.0001
Characteristics Number (%) Non-Chinese 47 (36.4%) 82 (63.6%)
Residence
Presentation
Urban 63 (63.6%) 36 (36.4%) 0.0001
Breast Lump 145 (78.0%)
Rural 26 (29.9%) 61 (70.1%)
Ulcerative Mass 17 (9.1%)
Income
Skin and Nipple Changes 10 (5.3%)
< 1000 RM 42 (40%) 63 (60%) 0.015
Metastasis 6 (3.2%)
≥1000 RM 47 (58%) 34 (42%)
Suspicious Mammography 4 (2.2%)
Education
Nipple Discharge 2 (1.1%)
None 12 (21.4%) 44 (78.6%) 0.0001
Others 2 (1.1%)
Yes 77 (59.2%) 53 (40.8%)
Tumour Histology
Religion
Invasive Ductal Carcinoma 157 (84.4%)
Muslim 28 (42.4%) 38 (57.6%) 0.272
Invasive Lobular Carcinoma 6 (3.2%)
Non-Muslim 61 (50.8%) 59 (49.2%)
Medullary Carcinoma 4 (2.2%)
Marital status
Mucinious Carcinoma 5 (2.7%)
Single 7 (41.2%) 10 (58.8%) 0.563
Ductal Carcinoma In Situ 8 (4.3%)
Married 82 (48.5%) 87 (51.5%)
Colloid Carcinoma 2 (1.1%)
Age of first child birth
Papillary Carcinoma 3 (1.6%)
≤30 63 (46.7%) 72 (53.3%) 0.035
Apocrine Carcinoma 1 (0.5%)
>30 15 (71.4%) 6 (28.6%)
Tumour Grade
Parity
I 24 (13.3%)
Nulliparous 11 (35.5%) 20 (64.5%) 0.131
II 107 (59.4%)
Non-nulliparous 78 (50.3%) 77 (49.7%)
III 49 (27.2%)
Breast feeding
Tumour Side
Yes 63 (48.1%) 68 (51.9%) 0.919
Right 96 (51.6%)
No 26 (47.3%) 29 (52.7%)
Left 86 (46.2%)
Hormonal Medication
Both 4 (2.2%)
Yes 34 (56.7%) 26 (43.3%) 0.097
Tumour Site
No 55 (43.7%) 71 (56.3%)
Upper Outer 100 (53.8%)
Menarche
Upper inner 26 (14.0%)
<13 year old 33 (55.9%) 26 (44.1%) 0.4
Lower outer 26 (14.0%)
≥13 year old 45 (48.9%) 47 (51.1%)
Lower inner 10 (5.4%)
Menopause
Areolar 20 (10.8%)
Yes 38 (45.2%) 46 (54.8%) 0.518
More than 1 4 (2.2%)
No 51 (50.0%) 51 (50.0%)
Hormonal Receptor Status Family History
ER Positive 110 (59.1%) Yes 9 (42.9%) 12 (57.1%) 0.627
PR Positive 102 (54.8%) No 80 (48.5%) 85 (51.5%)

Asian Pacific Journal of Cancer Prevention, Vol 8, 2007 527

Benjamin Dak Keung Leong et al
Table 3. Logistic Regression Analysis Predicting helped to explain the similar high proportion of advanced
Advanced Breast Cancer disease seen here (Hisham and Yip, 2004). Proportion of
Patient P Value Odds Ratio patients presented with advanced disease seen in large
Characteristics (95% confidence interval) medical centres in the region such as Tan Tock Seng
Hospital in Singapore and Asan Medical Centre in Korea
Non-Chinese 0.043 2.35 (1.03 to 5.39)
Rural residence 0.043 2.18 (1.03 to 4.64) were 21.6% and 12.3-17.4% respectively (Tan et al., 2005;
Non-educated 0.020 3.70 (1.62 to 8.45) Son et al., 2006). In the USA, only 7% of patients had
Salary < 1000 RM 0.592 0.82 (0.40 to 1.70) stage III disease at the time of diagnosis and the Western
world in general had 25% of breast cancer patients
than 1000 RM per month (P = 0.015) and 78.6% of patients presented at advanced stage (National Cancer Institute,
who had never been to school (P = 0.0001) presented with 2001; Lim and Yahaya, 2004).
advanced disease. A patient who is non-Chinese, from Patients with advanced disease seen here were more
rural area, has an income of less than RM 1000 and non- likely to be non-Chinese, poor- with salary of less than
educated had a 83.3% likelihood to present with advanced RM 1000 per month, from rural area and non-educated.
disease in our study (P< 0.0001). In multivariate analysis These four parameters were found to be all statistically
where these four relevant coefficients were put in logistics significant in univariate analysis and, except for income
regression analysis, non-Chinese race (P= 0.043), rural less than RM 1000, in multivariate analysis (Figures 2
residence (P= 0.043) and non-educated (P= 0.02) were and 3). It is also important to highlight that 9.1% of patients
recognized as independent factors predicting advanced here presented with ulcerative mass and only 2.2% had a
breast cancer at presentation (Table 3). suspicious lesion on mammography. This is in great
In our study too, 20.4% of patients refused proper contrast with USA, where 30% to 40% of breast cancers
recommended treatment or defaulted follow up altogether are detected with screening mammography. Besides,
with majority of them opted for traditional or alternative 20.4% of patients defaulted proper treatment and follow
therapy. up, most opted for alternative or traditional therapy. This
further delayed appropriate treatment. Most of them came
Discussion back to us with more advanced disease after a period of
time. This figure is in huge contrast even if compared to
Breast cancer is the commonest cancer among women Kuala Lumpur Hospital where only 5% of patients
in Malaysia and despite being the second largest state in defaulted treatment (Lim and Yahaya, 2004).
the country, data on breast cancer from Sabah has been This picture of late presentation and high default rate
poorly reported and published. From our study, the can be attributed to multiple factors. Lack of awareness
commonest affected race was Chinese but they present at of breast cancer among women in Sabah and strong
earlier stage of the disease compared to other races. Non- influence of traditional and cultural belief are two main
Chinese were twice more likely to present with advanced factors. Lack of awareness with wrong social and cultural
disease (Table 3). A similar picture was also seen in perception of breast cancer have been associated with
Peninsular Malaysia and Singapore (Greene et al., 2002; advanced disease at presentation (Sandelin et al., 2002;
Hisham and Yip, 2003; 2004; Lim and Yahaya, 2004; Tan Lim and Yahaya, 2004). Women feel that their role as a
et al., 2005). In our study too, Chinese had less ER- wife, mother and female as a whole will be seriously
negative tumours compared to non-Chinese (33.3%% vs. threatened and affected if they have breast cancer. Thus,
44.2%). It has been reported that ER-negative tumours a strong sense of denial normally developed as a protective
are more aggressive and carry a poorer prognosis mechanism against such threat. Due to strong traditional
compared to ER-positive tumours (Rochefort et al., 2003). influence, many women will initially seek traditional or
In line with the report, our study has shown ER-negative alternative treatment such as ‘bomoh’ or faith healing
tumours were more likely to present as advanced disease before they present to hospital when the initial treatment
compared to ER-positive tumours (65.9% vs. 42.7%). has failed.
Commonest age group was 40 to 49 years of age which is Most parts of Sabah are still very much under-
similar with Peninsular Malaysia (Lim and Yahaya, 2004). developed with limitation of quality education,
The western world has the peak prevalence in the 50 to communication system and health care. An ill patient may
59 year old age group (Greene et al., 2002). There was take up more than a full day of travelling on foot, by river
no significant age difference in patients who presented in or gravel road to reach the nearest health centre. Sabah
early and late disease in our study (Table 2). This is in has the poorest health care service in the country with a
contrast with several reports that younger patients were doctor to patient ratio of 1: 4000, in comparison with Kuala
more likely to present with advanced and aggressive Lumpur, 1: 500 and Malaysia in general, 1: 150,014.
disease9. Efforts to improve education, communication and health
52.2% of our patients presented with either stage III care system have remained a great challenge. There is
or IV disease. This is compared to within Malaysia- Kuala currently no national mammography screening of breast
Lumpur Hospital, 50-60%; University Malaya Medical cancer in Malaysia due to lack of financial funding and
Centre, Kuala Lumpur, 30-40% and Penang State, 37.7% resources (Lim and Yahaya, 2004).
(Penang Cancer Registry, 1999; Hisham and Yip, 2004). Much effort is needed to improve the awareness of
Kuala Lumpur Hospital caters mainly to the lower breast cancer among women in Sabah. The month of
socioeconomic group of patients in Kuala Lumpur which October is dedicated as breast cancer awareness month in
528 Asian Pacific Journal of Cancer Prevention, Vol 8, 2007
 Breast Cancer in Sabah, Malaysia
Malaysia where talks, seminars and related activities on Rochefort H, Glondu M, Sahla ME, et al (2003). How to target
breast cancer awareness, detection and treatment are held estrogen receptor-negative breast cancer? Endocr Relat
all over the country. These activities are seen to be effective Cancer, 10, 261-6.
and beneficial but the effect of which will mainly be Sandelin K, Apffelstaedft JP, Abdullah H, et al (2002). Breast
Surgery International-breast cancer in developing countries.
among women living in urban areas where such activities
Scand J Surg, 91, 222-6.
are normally held. Effort to involve more rural areas must Son BH, Kwak BK, Kim JK, et al (2006). Changing patterns in
be multiplied as most patients with advanced cancer in the clinical characteristics of Korean patients with breast
our report were from rural areas. The media and non- cancer during the last 15 years. Arch Surg, 141, 155-60.
government organisations can play an invaluable role in Stewart BW, Kleihues P (2003). Breast Cancer. World Cancer
such areas. Hospitals and health clinics in rural areas Report. Lyon: International Agency for Research on Cancer
should also improve efforts to educate and promote breast Press, 188-93.
cancer awareness including breast self examination among Tan EY, Wong HB, Ang BK, et al (2005). Locally advanced and
the villagers at all times. metastatic breast cancer in a tertiary hospital. Ann Acad Med
Singapore, 34, 595-601.
The public should also be made aware that alternative
and traditional therapy have no role in the management
of breast cancer. The practice of alternative and traditional
medicine should be regulated and monitored strictly by
the authorities. Besides, poor road conditions and
communication network to rural areas in the state should
be improved to ease referrals and delivery of more
effective health care. The discrepancy of doctor-patient
ratio with the rest of Malaysia should be addressed to and
setting up a national screening programme for breast
cancer will be desirable. There should also be a fast tract
referral system for patients suspected of breast cancer to
be seen earlier at referral hospitals.
In conclusion, the majority of women with breast cancer
in the state of Sabah present with advanced disease. Those
who presented late in our study were more likely to be
non-Chinese, poor, non-educated and from rural areas. A
significant portion of them defaulted proper treatment and
had a strong belief in alternative and traditional therapies
More effort is needed to increase awareness of breast
cancer among women in Sabah, especially among those
who are more likely to present with advanced disease.

References
Agnese DM, Yusuf F, Wilson JL, et al (2004). Trends in breast
cancer presentation and care according to age in a single
institution. Am J Surg, 188, 437-9.
Ferlay J, Bray F, Pisani P, et al (2002). Globocan 2002: Cancer
Incidence, Mortality and Prevalence Worldwide. Lyon:
International Agency for Research on Cancer Press.
Greene FL, Page DL, Fleming ID, et al (2002). AJCC Cancer
Staging Manual. 6th Ed. New York: Springer-Verlag.
Hisham AN, Yip CH (2004). Overview of breast cancer in
Malaysian Women: A problem with late diagnosis. Asian J
Surg, 27, 130-3.
Hisham AN, Yip CH (2003). Spectrum of breast cancer in
Malaysian women: overview. World J Surg, 27, 921-3.
Lim GCC, Yahaya H (2004). Second Report of the National
Cancer Incidence in Malaysia. Kuala Lumpur: National
Cancer Registry, 26.
Malaysia Medical Association Berita Malaysia Medical
Association Kuala Lumpur: Malaysia Medical Association,
February 2006.
National Cancer Institute, DCCPS, Surveillance Research
Program, Cancer Statistics Branch. SEER Program Public
Use Data Tapes 1973-1998, November 2000 Submission.
Issued April 2001.
Penang Cancer Registry (1999). 1996 Statistical Report. Penang:
State Health Department.

Asian Pacific Journal of Cancer Prevention, Vol 8, 2007 529