Predictive factors of recurrence free survival of patients with luminal breast cancer in southern Sri Lanka

Lakmini K B Mudduwa; Harshini H Peiris; Thushari Liyanage

doi:10.4103/bjoc.bjoc_2_22

ORIGINAL ARTICLES

Year : 2021 | Volume : 1 | Issue : 2 | Page : 78-85

Predictive factors of recurrence free survival of patients with luminal breast cancer in southern Sri Lanka

Lakmini K B Mudduwa¹, Harshini H Peiris², Thushari Liyanage¹
¹ Department of Pathology, Faculty of Medicine, University of Ruhuna, Galle, Sri Lanka
² Department of Medical Laboratory Science, Faculty of Allied Health Sciences, University of Ruhuna, Galle, Sri Lanka

Date of Submission	18-Mar-2022
Date of Acceptance	05-Jul-2022
Date of Web Publication	24-Aug-2022

Correspondence Address:
Lakmini K B Mudduwa
Department of Pathology, Faculty of Medicine, University of Ruhuna, Galle (Postal code: 80000)
Sri Lanka

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/bjoc.bjoc_2_22

Abstract

Purpose: To assess the predictive factors of recurrence free survival (RFS) of luminal breast cancers in Southern Sri Lanka to bridge the existing gap in published literature on luminal breast cancers in our country. Materials and Methods: This was a bidirectional; retro prospective, cohort study. All breast cancer patients who sought the services of our immunohistochemistry (IHC) laboratory from 2006 to 2015 were included. Tissue micro arrays were constructed using the archives for the IHC analysis of Ki67, CK5/6 and EGFR. Details of ER, PR and Her2 expression were extracted from the laboratory records. Clinico-pathological data were retrieved from the records available in the laboratory, Oncology clinic and from patients. Pearson chi-square test, Kaplan-Meier model, log-rank test and Cox-regression model were used with SPSS for data analysis. Results: A total of 595 luminal breast cancers were included. Majority, were between 36–60 years of age (65.5%), Nottingham grade 2 (57.4%), TNM Stage II/III (82.0%) and had T2 tumours (53.6%). Recurrences were noted in 15.9% (n = 395, local-16.9%; distant metastasis-76.9%; second cancer-6.2%). Median follow up time was 35 months. Only 443 luminal breast cancers could be subtyped (luminal A-58.7%); luminal B Her2-negative- 22.3%; luminal B Her2-positive-19.0%). Five-year RFS of the cohort was 74.1% (luminal A-83.6%; luminal B(Her2-negative)-73.9%; luminal B(Her2-positive)-55.3%; P = 0.001). Over-expression of Her2 (P = 0.018) and CK5/6 expression (P = 0.003) had an independent effect on the RFS of the cohort. Conclusion: The three subtypes of luminal breast cancers are distinct in their recurrence free survival and tumor characteristics. Over-expression of Her2 and CK5/6 expression had an independent negative effect on the RFS of the luminal breast cancer cohort.

Keywords: Immunohistochemical markers, luminal breast cancer, predictive factor, recurrence free survival

How to cite this article:
Mudduwa LK, Peiris HH, Liyanage T. Predictive factors of recurrence free survival of patients with luminal breast cancer in southern Sri Lanka. Bengal J Cancer 2021;1:78-85

How to cite this URL:
Mudduwa LK, Peiris HH, Liyanage T. Predictive factors of recurrence free survival of patients with luminal breast cancer in southern Sri Lanka. Bengal J Cancer [serial online] 2021 [cited 2023 Jun 3];1:78-85. Available from: http://www.bengaljcancer.org/text.asp?2021/1/2/78/354407

Introduction

Breast cancer is the most common cancer among females worldwide and it is the most common cancer in Sri Lanka since year 2000. The incidence of breast cancer has risen from age standardized incidence rate of 9.4 to 28.1 per 100,000 population over a 30-year period from 1985 to 2015.^[1] Over these decades, the understanding on the biology of breast cancer and treatment has evolved remarkably. The heterogeneity of breast cancer has been unravelled in gene sequencing studies. The current enormous literature on clinical management and disease outcome supports this concept. The seminal article published by Perou et al. in the year 2000 revealed that breast cancers could be classified by the differences in their gene expression patterns.^[2] Based on molecular-genetic analysis, five molecular subtypes of breast cancer were described; luminal A and B, Her2 positive, basal-like and normal-like.^[2] Subsequently, luminal B was divided into two subtypes based on Her2 over expression, progesterone receptor (PR) and Ki67. The normal-like lost its identity.^[3],[4]

The luminal breast cancers express hormone receptors, reflecting the expression profile of the luminal epithelium of the breast. Luminal breast cancers are a heterogenous group of tumours, the common feature being estrogen receptor (ER) and/or PR expression which is considered a good prognostic feature as it allows estrogen modulating therapy.^[5] The heterogeneity of luminal breast cancers is evident at many levels; molecular signature, phenotype, relapse patterns and therapeutic response to endocrine and chemotherapy.^[5] The identification of the two intrinsic luminal subtypes; luminal A and B, has proved this heterogeneity at genetic level.^[2]

The clinicopathological surrogate definition recommended by the St Gallen International Expert Consensus working group in 2013 is now widely used in identifying luminal subtypes. There are three luminal subtypes of breast carcinoma. The luminal A includes all ER and/or PR positive, Her2 negative and Ki67 low breast cancers. It is claimed to have a low risk of recurrence.^[3] When luminal breast cancers over-express Her2, with any PR and any Ki67 expression, they are classified as luminal B (Her2-positive). To be included in the luminal B (Her2-negative) category, the breast cancers should be ER positive, Her2 negative and either PR low/negative or expressing high Ki67.^[3] These markers are now routinely done for breast cancers in Sri Lanka too, and identification of luminal subtypes is possible even in the absence of multigene analysis which is recommended by the St Gallen Consensus guideline to be done only if facilities are available. These molecular subtypes are identified to have significant prognostic differences. Luminal A has the best prognosis out of all subtypes but luminal B has an aggressive clinical behavior similar to that of non-luminal breast cancers.^[6]

The current study was designed in the wake of worldwide emerging literature and when there is no published literature on luminal breast cancers from Sri Lanka.

The aim of this study was to determine the recurrence free survival (RFS) and factors affecting RFS of luminal breast cancers in Southern Sri Lanka.

Materials and Methods

This cohort study included both retrospectively and prospectively collected data. It included all breast cancer patients who sought the services of the Diagnostic Immunohistochemistry (IHC) Laboratory of our department from 2006 to 2015. Tissue micro arrays (TMA) were constructed using the archival breast cancer tissue blocks retrieved from the department archives for the IHC analysis of Ki67, CK5/6 and EGFR. The procedure followed in the preparation of TMA has already been published by the research team.^[7] Details of ER, PR and Her2 expression of each tumour were extracted from the laboratory records. Other clinico-pathological data were retrieved from the laboratory records, records available at the oncology clinic and records available with patients for which informed written consent was obtained. The approval of the Ethical Review Committee of our institution was obtained before commencing the study.

This study included all breast cancer patients referred to the laboratory from 2006 to 2015 from the Southern province, covering a major proportion of breast cancer patients in southern Sri Lanka as our laboratory was the single IHC laboratory in the Southern province till 2012 and continued to provide the service. Patients from the oncology unit of the Teaching Hospital Karapitiya which is the tertiary care center for the province, were recruited prospectively for this study, from 2013 to 2015.

IHC staining

IHC staining for Ki67 (Dako M7240) in 1/75 dilution, CK 5/6 (Dako M7237) in 1/50 dilution and EGFR (Dako M3563) in 1/100 dilution were used for the IHC staining of TMAs of breast cancer tissue. These optimum dilutions were achieved by repeated testing of control material. EGFR antigen retrieval was done using proteinase. Microwave antigen retrieval with pH 9 Tris buffered saline was done at 1100 W for 20 min following preheating for CK5/6. Citrate buffer at pH 6 was used for antigen retrieval by pressure cooking for 7 min after pre-heating, for Ki67. IHC staining was done manually with a positive control for each primary antibody. Our IHC laboratory has been using same protocol for staining of breast cancers for ER, PR and Her2 which had been validated and optimized in the laboratory.

The Allred score for ER and PR and UK recommendations for Her2 were used in assessing IHC staining, on the original whole sections.^[8] A score of >2 for ER and PR and a score of +3 for Her2 were considered the criterion to identify positive staining with the respective antibodies. Patients with Her2 equivocal expression (2+) were excluded when in situ hybridization results were not available. If >5% of tumour cells expressed EGFR and CK5/6, the tumour was considered positive for the corresponding antigen. The cut off for Ki67 was 14% tumour cell nuclear staining.

Follow-up and outcomes

Mean follow-up time was 35.94 (±25.39) months. The actual minimum follow-up period was 28 months. RFS time was calculated from the date of first therapeutic intervention to first loco-regional and/or first distant recurrence. Deaths and second cancers were not considered as events for the estimation of RFS. However, second cancers were considered for calculating the prevalence of recurrences. Radiological and histopathological evidence were used to confirm the recurrence. The date of the relevant investigation was taken as the date of recurrence. Censoring was done at the last follow up for patients who did not experience a recurrence or death.^[9]

Statistical analysis

The Pearson chi-square test was used to determine the association between the three luminal subtypes and their clinico-pathological features. Kaplan-Meier model was used to estimate the RFS and the log-rank test was used to compare the survival of luminal breast cancer patients and their subtypes. The Predictors of RFS was determined using Kaplan-Meier model for univariate and Cox-regression model for multivariate analyses.

Results

Clinicopathological profile of the luminal breast cancer cohort

From a total of 1122 breast cancer patients, luminal breast cancers comprised 595 (53.0%) and included in the study. [Table 1] includes the clinicopathological profile of the luminal breast cancer cohort. Majority of the luminal breast cancer patients, 65.5%, were between 36–60 years and 3.4% were 35 years old or less while 31.1% were >60 years. T2 tumours (53.6%) were noted in the majority (2–5 cm in maximum diameter). Tumours <2 cm (T1) was seen in 38.8%. Most of the patients had Nottingham grade 2 tumours (57.4%) while 21.7% were grade 3. Axillary lymph node metastasis has been detected at the time of diagnosis in 50.9% (lymph node stage 0–49.1%, 1–25.7%, stage 2–14.6%, stage 3–10.6%). TNM Stage group details were available for 529 (Stage I=17.2%, Stage II=46.3%, stage III=35.7%, Stage IV=0.8%). Lympho-vascular invasion was present in 45.6% (n = 551).

Table 1: Clinico-pathological features of the luminal breast cancers

Click here to view

ER was positive in 88.2% while PR positive expression was seen in 86.2%. (ER+PR+: 74.4%; ER+PR-: 13.8% and ER-PR+: 11.7%). Her2 was positive in 15.6% and equivocal in 7.9% whose FISH results were not available. CK5/6 was expressed in 12.9% (n = 464) and EGFR was positive in 4.0% (n = 481). Out of the two basal markers assessed, either CK5/6 and/or EGFR was positive in 17.0% (n = 442). Follow up details were available for 395 patients and 15.9% of them had recurrences (local-16.9%, distant-76.9%, second cancer-6.2%).

The majority had surgical treatment (n = 438, mastectomy-98.6%, wide local excision-0.9%, data not available-0.2%). Data on adjuvant therapy was available for 439 patients and majority had at least two forms of adjuvant treatment (hormone therapy-89.5%, chemotherapy- 88.4%, radiotherapy- 68.6.3%, Anti Her2 therapy-42.3% (36/85)).

Out of all luminal breast cancers only 443 could be classified into the three luminal subtypes due to the absence of Ki67 and /or PR expression results due to tissue loss in TMAs. Breast cancers with equivocal expression of Her2 for which FISH results were not available also could not be classified beyond identifying them as luminal based on hormone receptor expression. Luminal A comprised the majority (58.7%) while luminal B (Her2-negative) comprised 22.3% and luminal B (Her2-positive) accounted for 19.0% of patients [Figure 1].

Figure 1: RFS of luminal breast cancers according to the CK 5/6 expression (p=0.011)

Click here to view

The three subtypes of luminal breast cancers

The descriptive statistics of the three subtypes of luminal breast cancers had a significant difference with regard to the Nottingham grade (P < 0.001), lymph node stage (P < 0.001), lympho-vascular invasion (P = 0.004) and the proportion of patients who developed recurrences (P = 0.008, luminal A-9.9%; luminal B Her2 negative-12.5%; luminal B Her2 positive-27.1%).

A higher percentage of luminal B (Her2-positive) breast cancers (63.3%) had lympho-vascular invasion compared to the luminal A (44.8%) and luminal B (Her2-negative) (39.6.6%) sub types. Grade 3 tumours were more prevalent among the luminal B (Her2-positive) (38.8%) than luminal A (16.7%) and luminal B (Her2-negative) (28.1%) (P = 0.001). There was no significant difference in the distribution of the four stages among the three luminal subtypes (P = 0.67). The proportions of tumours expressing either EGFR or CK5/6 in the three subgroups of luminal breast cancers had no significant difference.

Luminal A vs luminal B

Luminal A and luminal B breast cancers had no significant difference with regard to the age at presentation, tumour size, tumour stage, lymphovascular invasion and expression of EGFR and CK5/6. However, the two luminal subtypes had a significant difference in respect to Nottingham grade (P < 0.001) and lymph node stage (P < 0.001). Grade 3 tumours comprised 33.1% of luminal B while it was 16.7% of luminal A. Lymph node stage 3 was more prevalent among luminal B (17.5%) compared to luminal A (6.5%). TNM stage had no significant difference between the luminal A (stage I-19.3%, II-47.0%, III-32.5, IV-1.2%) and B tumours (stage I-14.7%, II-45.3%, III-39.4%, IV-0.6%)

The prevalence of recurrences was significantly different between the luminal A (9.4%) and B (18.0%) (P = 0.033). Type of recurrence was similar between luminal A (local=15.8%; Metastasis=73.7%; secondary cancers=10.5%) and luminal B (local=4.5%; Metastasis=86.4; secondary cancers=9.1%) with no statistically significant difference.

Analysis of survival data

Data on recurrences were not available for 200 patients. For the cox proportional hazard model, all features listed in the [Table 1] except ER and PR expression were included. None of the clinicopathological features included in the multivariate model 1 had a significant effect on RFS; age, grade, lymph node stage, TNM stage and lympho-vascular invasion, P < 0.005) [Table 2]. However, an independent effect on RFS of the cohort was detected with CK5/6 expression (P = 0.003) and over-expression of Her2 (P = 0.018) in the model 2 [Table 2]. Since CK5/6 had an independent negative effect on the survival [Figure 1] the characteristic features of luminal breast cancers with CK5/6 expression was compared with the rest of the cohort. Over expression of Her2 (P = 0.039) and lympho-vascular invasion (P = 0.031) were more prevalent among CK5/6 positive luminal breast cancers compared to CK5/6 negative luminal breast cancers.

Table 2: The effect of clinico-pathological features on the RFS by Cox proportional hazards model

Click here to view

Five-year RFS of the overall luminal breast cancers was 74.1%. Five-year RFS was significantly different between the three luminal subtypes (P = 0.001) [Figure 2]. The RFS of luminal A, luminal B (Her2-negative) and luminal B (Her2-positive) breast cancer patients were 83.6%, 73.9% and 55.3% respectively. The total luminal B breast cancers had a five-year RFS of 64.8% [Figure 3].

Figure 2: Recurrence free survival of the 3 subtypes of luminal breast cancers (p=0.001)

Click here to view

Figure 3: RFS of luminal A and luminal B breast cancers (p=0.006)

Click here to view

Discussion

Luminal cancers are said to comprise more than 50% of breast cancers,^[10] but in some reports it reaches even 70%.^[11] However, it is less prevalent in our country.^[12],[13] The current study reveals that the luminal cancers in the cohort have many factors which indicate higher risk of developing recurrences. Most of the tumours were >2 cm in size (61.2%) and of Nottingham grade 2/3 (79.1%), in this luminal breast cancer cohort. Patients with grade 3 tumours comprising 21.7% of the cohort is also very significant, requiring chemotherapy in addition to hormone therapy as adjuvant therapy. Lymph node metastasis was present at the time of diagnosis in 50.9%. Lympho-vascular invasion had been identified in 45.6%(n = 551) of the study cohort. These poor prognostic factors would have had a significant effect on the RFS of the study group which was 74.1%.

Majority of the luminal B (Her2-positive) group has not received anti Her2 therapy which may have resulted in 5-year RFS of only 55% in the luminal B (Her2-positive) group contributing to the comparatively lower RFS of the luminal breast cancers in the cohort [Figure 4]. Though endocrine therapy was recommended, 10.5% patients refused endocrine therapy. Hence, it may have contributed to the lower RFS rate of the cohort.

Figure 4: RFS of luminal breast cancers according to the over expression of Her 2 (p=0.014)

Click here to view

A large population-based study done in Netherlands on 8002 breast cancer patients diagnosed in 2005 revealed that the percentages of recurrence free patients per breast cancer subtypes, luminal A and luminal B were 87.5%, 76.6%, respectively at a median follow up of 9.5 years.^[14] Three fourth of luminal cancers in this study sample were either grade 1 or 2. Only a 2.7% of luminal A and 4.7% of luminal B were either stage III or IV.^[14]

The 10-year overall survival was found to be 94.1% in luminal-like subtype of a cohort of 1595 breast cancer patients in stage I who were diagnosed from 1990 to 2008 in Taiwan.^[15] In both these studies a large percentage of study sample comprise low grade low stage tumours compared to the current study. A study on breast cancer cases diagnosed and reported to the SEER-18 during 2010–2013 including a total of 196,094 invasive breast cancer cases revealed the best survival was among the women with hormone receptor positive, Her2 negative subtype (survival rate 92.5% at 4 years), followed by hormone receptor positive, Her2 positive (survival rate 90.3%).^[16]

In a retrospective review of data from 2004 to 2011 at a hospital in China, the five-year disease-free survival for luminal A, luminal B (Her2-negative) and luminal B (Her2-positive) subtypes reported by Lian et al. were 91.8%, 79.6% and 65.0% respectively. They also found that nodal status to have an independent effect on the relapse free survival.^[17]

The estimated five-year RFS rate in our cohort is considerably lower compared to these studies, most likely due to the aggressive features seen in the cohort which include comparatively higher proportion of higher stage at presentation and high grade tumours among the current cohort. Only a 17.2% of patients presented at the stage I [Figure 5]. As we do not have a national level mammographic screening programme for breast cancer, like many developed countries, breast cancer patients present late even though not as late as stage IV. Breast screening by self-examination and clinical examination of women at Well Women Clinics may not be satisfactory with the current trend in the incidence of breast cancer. Lympho-vascular invasion being present in 45.3% also would have contributed to the relatively lower survival.

Figure 5: RFS of luminal breast cancers according to the pathological stage (p=0.01)

Click here to view

Adjuvant endocrine therapy is the standard of care in patients with luminal tumors which express ER. However, the indication for adjuvant chemotherapy is less clear-cut for luminal breast cancers.^[5] The St. Gallen International Breast Cancer Conference 2013 – Zurich Consensus statement revealed that the luminal A subtype is less sensitive to chemotherapy and thus should rather be treated by endocrine therapy alone.^[18] However, there is clearly a subset of patients with ER-positive tumors that are chemo-sensitive.^[5] When to use chemotherapy and whether to use a regime with full efficacy has been in debate.^[18] Patients with luminal B (Her2-negative) breast cancers have an increased risk of relapse and are sensitive to chemotherapy because of the increased proliferation rate.^[3],[18] Therefore, the St. Gallen International Breast Cancer Conference 2013 Consensus statement recommends adding chemotherapy to endocrine therapy for luminal A breast cancer when the tumour grade is 3 or lymph node stage is 2 or more. The luminal B (Her2-positive) cancers are recommended to have endocrine therapy, anti Her2 therapy and chemotherapy. For the luminal B (Her2-negative) breast cancers, endocrine therapy and cytotoxic therapy are recommended.^[3]

A large percentage of the cohort had received chemotherapy as it included 16.7% of luminal A with grade 3 tumours, 25.2% with lymph node stage ≥2 and chemotherapy indicated all luminal B tumours comprising 41.3% of the cohort. Those who received chemotherapy would have had one or more indications to be treated with chemotherapy.

In addition to the routinely assessed prognostic factors we have assessed how the expression of EGFR and CK5/6, basal markers, affect the RFS. A significant percentage (16.4%) comprises a subset of patients with basal marker expression substantiating the use of chemotherapy. The expression of CK5/6 has imparted a survival disadvantage independent of the other markers which is statistically significant [Figure 1]. Expression of these basal markers was not significantly different between the two luminal subtypes A and B. These markers are generally not recommended to be done routinely on luminal breast cancers. But the results of the current study welcome sufficient consideration on adding basal markers to the routine panel.

Haughian et al. reported that established luminal cell lines retain the ability to yield extensive cellular heterogeneity without genetic manipulation. They described on cell lines, ER and PR negative CK5 positive lumino-basal subpopulation in more than half of luminal breast tumors.^[19] The results of the current study also support this heterogeneity of luminal breast cancer cells in displaying basal features.

Basal-like and claudin-low cancers are hormone-independent, characterized by brief disease-free survival, a high proliferative index and poor histologic grade, and require aggressive chemotherapy.^[19] In one of our previous publications, we have described the survival disadvantage added by low claudine3 to the molecular subtypes of breast cancer which is most significant in the luminal A (both breast cancer specific survival and RFS) and luminal B (Her2−) (RFS) subtypes.^[20]

It is obvious in the IHC assessment of ER positivity that most luminal breast cancers contain a subset of ER negative tumour cells in the primary tumour itself as their ER expression vary from Allred score 3 to 8. If this sub-population of tumour cells become prominent during endocrine therapy, recurrences can be expected. This tumour heterogeneity has been investigated by many research teams. Haughiana et al. established a previously unrecognized plasticity of ER and PR positive luminal breast cancers that, without genetic manipulation, mobilizes outgrowth of hormone-resistant basal-like disease in response to common endocrine therapies. They proposed that this undesirable outcome can be avoided by combining γ-secretase inhibitors with endocrine therapies.^[19]

Limitations in our study include inherent problems in retrospective studies. Missing data was one issue and therefore, a large cohort of a subtype was selected to study minimising its effect. Tissue loss in TMAs was another issue which resulted in missing data and hence different totals obtained for some features. The study group included patients who were still on adjuvant treatment and those who have completed. Therefore, effectiveness of treatment cannot be commented based on the current results.

The major strength of this study is that it includes patients treated at a single tertiary care center homogenizing the effects of treatment on the survival. To ensure validity, patient data were collected from three sources enabling cross checking for accuracy of collected data.

In conclusion, this study confirms that the three luminal subtypes of breast cancers are distinct in their recurrence free survival and tumor characteristics. The luminal B includes breast cancers with more aggressive features. Her2 over expression and CK5/6 expression has an independent negative effect on the RFS. The independent negative effect of CK5/6 proved in this study highlights its prognostic role in luminal breast cancers.

The RFS of the luminal breast cancers which is expected to be the best out of all subtypes is less in this cohort compared to countries of high-resource settings. The findings of this study warrant re-visiting of healthcare policies on screening for breast cancers in Sri Lanka to ensure early detection increasing the opportunity for better survival.

Acknowledgement

The authors wish to acknowledge the Ministry of Science, Technology and Research, Sri Lanka for funding the project under the Indo-Sri Lanka Joint Research Programme 2012, the technical support provided by Mrs. GGDD Gunawardhane and the staff of the Department of Pathology, Faculty of Medicine, University of Ruhuna and the Histopathologists, Onco-surgeons, Oncologists and the staff of the Histopathology and Oncology Units of the Teaching Hospital Karapitiya, Galle, Sri Lanka, for the support extended. All research Assistants are acknowledged.

Financial support and sponsorship

Ministry of Science, Technology and Research, Sri Lanka for funding the project under the Indo-Sri Lanka Joint Research Programme 2012.

Conflicts of interest

There are no conflicts of interest.

References

1.	Cancer Incidence Data: Sri Lanka Cancer Registry. Sri Lanka: National Cancer Control Programme Ministry of Health; 2015.
2.	Perou CM, Sørlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, et al. Molecular portraits of human breast tumours. Nature 2000;406:747-52.
3.	Goldhirsch A, Winer EP, Coates AS, Gelber RD, Piccart-Gebhart M, Thürlimann B, et al; Panel members. Personalizing the treatment of women with early breast cancer: Highlights of the st gallen international expert consensus on the primary therapy of early breast cancer 2013. Ann Oncol 2013;24:2206-23.
4.	Eliyatkın N, Yalçın E, Zengel B, Aktaş S, Vardar E Molecular classification of breast carcinoma: From traditional, old-fashioned way to A new age, and A new way. J Breast Health 2015;11: 59-66.
5.	Lim E, Winer EP Adjuvant chemotherapy in luminal breast cancers. Breast 2011;20 Suppl 3:S128-31.
6.	Li ZH, Hu PH, Tu JH, Yu NS Luminal B breast cancer: Patterns of recurrence and clinical outcome. Oncotarget 2016;7:65024-33.
7.	Mudduwa LKB, Gunasekara SN, Gunawardhana GGDD Tissue micro arrays; establishing a cost effective tool for cancer biomarker research in Sri Lanka. Galle Medical Journal 2018;23:15-8.
8.	Ellis IO, Bartlett J, Dowsett M, Humphreys S, Jasani B, Miller K, et al. Best practice no 176: Updated recommendations for Her2 testing in the Uk. J Clin Pathol 2004;57:233-7.
9.	Rakha EA Pitfalls in outcome prediction of breast cancer. J Clin Pathol 2013;66:458-64.
10.	Gao JJ, Swain SM Luminal A breast cancer and molecular assays: A review. Oncologist 2018;23:556-65.
11.	Dunnwald LK, Rossing MA, Li CI Hormone receptor status, tumor characteristics, and prognosis: A prospective cohort of breast cancer patients. Breast Cancer Res 2007;9:R6.
12.	Mudduwa LKB Quick score of hormone receptor status of breast carcinoma: Correlation with the other clinicopathological prognostic parameters. Indian J Pathol Microbiol 2009;52:166-70.
13.	Lokuhetty MD, Ranaweera GG, Wijeratne MD, Wickramasinghe KH, Sheriffdeen AH Profile of breast cancer in a group of women in a developing country in South Asia: is there a difference? World J Surg 2009;33:455-9.
14.	van Maaren MC, de Munck L, Strobbe LJA, Sonke GS, Westenend PJ, Smidt ML, et al. Ten-year recurrence rates for breast cancer subtypes in the Netherlands: A large population-based study. Int J Cancer 2019;144:263-72.
15.	Cheng SHC, Yu BL, Horng CF, Tsai SY, Chen CM, Chu NM, et al. Long-term survival and stage I breast cancer subtypes. J Cancer Res Pract 2016;3:1-8.
16.	Howlader N, Cronin KA, Kurian AW, Andridge R Differences in breast cancer survival by molecular subtypes in the united states. Cancer Epidemiol Biomarkers Prev 2018;27:619-26.
17.	Lian W, Fu F, Chen D, Wang C Effect of node status on breast cancer survival by subtype: A single-center retrospective cohort study. Transl Cancer Res 2020;9:5900-8.
18.	Untch M, Gerber B, Herbeck N, Jackisch C, Mobus N, Minckwitz GV, et al. 13th St. Gallen International Breast Cancer Conference 2013: Primary Therapy of Early Breast Cancer Evidence, Controversies, Consensus – Opinion of a German Team of Experts (Zurich 2013). Breast Care 2013;8:221-9.
19.	Haughian JM, Pinto MP, Harrell JC, Bliesner BS, Joensuu KM, Dye WW, et al. Maintenance of hormone responsiveness in luminal breast cancers by suppression of Notch. PNAS 2012;109:2742-47. https://doi.org/10.1073/pnas.1106509108
20.	Mudduwa L, Peiris H, Gunasekara S, Abeysiriwardhana D, Liyanage T Prognostic significance of low claudin3 expression in luminal breast cancer. Breast Cancer: Basic and Clinical Research 2017;11:1-11.