Pathologically confirmed women's breast cancer: A descriptive study of Tunisian and Algerian series

Abstract Background Breast cancer (BC) is the most frequent malignancy among women in Tunisia and Algeria. Clinical and pathological characteristics of this cancer among these populations are not widely reported. The aim of the study was to report clinical and pathological characteristics of women's BC in a Tunisian and Algerian series. Methods Pathologically confirmed 1089 BCs were gathered in the pathology departments of three Northern Tunisian hospitals: Tunis military, Charles Nicolle and Jendouba and in the pathology department of Alger Douera hospital between January 2015 and December 2020. Clinical and pathological findings of the two series: age, tumor size, histological type, grading according to Scarff‐Bloom Richardson grading system, lymph node status at the time of diagnosis in axillary lymphadenectomy specimens and the immunohistochemical expression of estrogen and progesterone receptors (ER/PR), HER2 and Ki‐67, were collected from the pathological reports. Results The median age at diagnosis was 50 and 48 years in Tunisian and Algerian series, respectively (p = 0.016). The diagnosis of BC was made on surgical specimens (lumpectomy or mastectomy) in 373/491 (76%) cases of the Tunisian series and in 225/598 (37.6%) cases of the Algerian one. Median tumor size was 2.8 cm and 2.5 cm in Algerian and Tunisian series, respectively (p = 0.252). Invasive BCs not otherwise specified was observed in 440/481 (91.5%) BCs in Tunisian series and in 519/586 (88.6%) BCs in Algerian series. Axillary lymph node positive tumors were observed in 64.6% and 58.8% of Tunisian and Algerian women, respectively (p = 0.926). BCs were ER positive in 311/385 (80.8%) and 486/571 (85.1%) cases and HER2 positive in 86/283 (30.4%) and 60/385 (15.6%) cases of Tunisian and Algerian series, respectively. Conclusions In Tunisia and Algeria, BC has poor prognostic factors with large tumor sizes and high rates of lymph nodes involvement at diagnosis.


| INTRODUCTION
With 2.2 million new cases and 68 499 deaths in 2020, breast cancer (BC) is the most frequent malignancy in women and the first major cause of cancer death in women globally. 1 The highest incidence rates (>80 per 100 000 females) are observed in Australia and New Zealand, Western Europe, Northern America, and Northern Europe. The lowest rates (40 per 100 000 females) are observed in Central America, Eastern and Middle Africa, and South-Central Asia. The highest mortality rates (>20 per 100 000 females) were found in Melanesia, Western Africa and Micronesia/Polynesia, while rates in most other world regions range between 10 and 15 per 100 000 females. 2 These rates reflect the availability of mammography, thus, the detection of BCs at an early stage in high-income countries. BC is often discovered at a later stage in low-and middle-income countries, which partly explains the high mortality rates in these regions. 3 In Northern Africa, the Global Cancer Observatorygco database estimates that there were 57 128 new female BC cases and 21 524 related deaths in 2020. In Tunisia, BC is a major public health issue, accounting for 34.5% of all female malignancies with 3092 annual diagnosed cases in 2020. 1 In Algeria, BC is also a leading cause of cancer, accounting for 40.3% of all female malignancies, with 12 536 new cases diagnosed in 2020. Its incidence in Tunisia and Algeria has been increasing at an alarming rate for about 25 years. The data from the Tunisian and Algerian registries illustrate this real and regular increase. 4,5 According to World Health Organization (WHO) classification of breast tumors, BC has a broad spectrum of histological patterns. 6 BCs with distinct histological patterns are classified as a special tumor type, such as lobular, mucinous, or tubular carcinomas etc… Invasive BC of non-specific subtype is the term used to describe BCs lacking sufficient characteristics to achieve classification as a specific histological type. 6 With the development and implementation of genomic and expression profiling analyses, the Cancer Genome Atlas (TCGA) Network has helped establish refined subtypes of BC through extensive profiling of protein levels, mRNA level, and DNA. 7 The molecular classification changed the paradigm of BC treatment. It includes four subtypes: "luminal A," "luminal B," "HER2-positive," and "basal-like." 8 Patients with luminal A BC have an excellent prognosis and gain clinically relevant benefits from endocrine therapy but not from chemotherapy. Patients with luminal B BC, however, may have a worse prognosis if treated alone with endocrine therapy due to endocrine resistance, but they may benefit from chemotherapy. 8 HER2-positive BCs are responsive to anti-HER2 medications. 9 Finally, chemotherapy is beneficial for patients with basal-like BC. 10 The frequencies of molecular subtypes vary worldwide. In Western series, luminal A subtype is predominant followed by luminal B, HER2, and basal-like. [11][12][13] Few studies of the clinicopathological features of BC in African countries have been conducted. The results of the published studies show some degree of divergence. In Sub-Saharan Africa, BCs exhibit more aggressive features such as triple negative phenotype. 14 Other studies indicate geographical variability in the distribution of BC molecular subtypes with the risk of triple negative BCs found to be lower in East Africa. 15 In North African studies, luminal A is the predominant subtype. 16,17 Studies have shown that it is possible to reproduce this molecular classification by using immunohistochemical tests based on the expression of estrogen receptors (ER), progesterone receptors (PR), HER2, Ki-67 and other biomarkers such as high and low molecular weight cytokeratins: CK5/6. 18 Due to time and cost constraints, in the great majority of health care systems, surrogate molecular BC classification is largely based on immunohistochemical assessment of biomarkers. To discriminate between luminal A/B, HER2-positive, and triple negative tumors, a panel encompassing ER, PR, HER2 and Ki-67 might be employed. 6 Currently, little is known about the clinical and pathological characteristics of BCs in North Africa, especially in Tunisia and Algeria. 16,19,20 To our knowledge, no comparative study on BC clinicopathological characteristics among Tunisian and Algerian women has been published so far. To gain further insight into BC among the two populations, we studied the clinicopathological characteristics of BC series. This study is the first to present a large population-based study on BC among Tunisian and Algerian women.  was divided into two subgroups with a 20% cut-off according to 2013 Saint Gallen consensus. 10

| Statistical analysis
Statistical analysis was performed by SPSS (version 26.0). Categorical data were summarized by frequencies and percentages. Continuous data regarding age and tumor size were presented as groups and medians. The assessment of associations between age, tumor size, histological subtype, histological grade, lymph node status, estrogen receptors, progesterone receptors, HER2, Ki-67 cut-off levels and molecular classification was performed using the χ 2 and Fisher's tests.
p Values less or equivalent to 0.05 were considered significant.

| RESULTS
From January 2015 to December 2020, 1089 pathologically con-    in Arab countries reported a median age of 44.5 years. 27 BCs in Tunisia and Algeria occur at an earlier age compared to BCs in Western countries in which peak incidence of BC is between 60 and 70 years. 28 Young age of BC in Africa is probably due to young population and genetic and environmental factors. 27,29 In the present study, median tumor size was 2.5 and 2. series but is still greater than BC's size in Western countries which is reduced to 1.5 cm under the effect of mass screening and early detection campaigns. 32 In Algeria, a recent study reported a mean tumor size of 3.6 cm. 33 In Algeria and Tunisia, BC mainly concerns a young population, thus mammographic screening should start at the age of 40.
In addition to high tumor size, BCs in the two series of the present study were characterized by a high frequency of axillary lymph node metastases at diagnosis (64.6% and 58.8%). Similar frequencies were reported in other African countries like Morocco (60%), 34 Lybia (73.9%) 35 and Nigeria (79%). 36 These frequencies are higher to those reported in Europe (34%). 37 In Tunisia and Algeria, women still consult tardily with palpable lesions. Mammography is hardly accessible in some regions due to its high cost and inequalities in territorial imaging centers' distribution. Training of radiologists to ensure quality, validity and interpretation criteria deserve to be deployed to popularize mammography for screening purposes. In the meantime, prevention strategies should raise awareness by promoting self-examination and systematic clinical examination of women's breast. 38 The frequency of BC categories subtypes is highly variable. Luminal A BC was the predominant category in Tunisian and Algerian series (58.3% and 68.8%, respectively). The predominance of luminal A subtype was also reported in previous studies in Tunisian, 16,20 Algerian, 39 and Western series. 11 Lowest frequencies of luminal A were reported in other African countries such as Mali (29.2%) 40 and Ghana (25.6%). 41 In this study, the frequency of luminal B, in the Algerian series, is relatively low (8.4%) compared to Tunisian series (22%) and to a previous Algerian one which reported a rate of 19.7% of luminal B BCs. 39 The rates of luminal B BCs in our Tunisian series were concordant to those of previous Tunisian series. 5,26 Other studies, like a recent Tunisian one and others from Saudi Arabia and Italy found a high rate of luminal B BC. [42][43][44] Although the molecular classification is well codified, there are some technical biases such as specimen fixation, sample storage duration, and other preanalytical immunohistochemical variables which could influence the accuracy of IHC results.
In Western countries, the prevalence of HER2 positive BCs varies from 4% 45 to 21.6%. 46 In this study, HER2+ BCs were observed in 10.4% and 6.8% in the Tunisian and Algerian series, respectively.
In this study, triple negative BCs were found in 10.5% in the Tunisian series. Previous Tunisian studies reported variable triple negative BCs' frequencies: 22.5%, 20 15.5%, 42 and 14%. 47 In the Algerian series of this study, the rate of triple negative BCs was 13.7%, similar to the Tunisian one, but lower than that described in another Algerian series (20.8%). 39 Rates of triple negative BCs in both series of this study are similar to those of European countries. 48 In the literature, rates of triple negative BCs vary from 10% to 25%. 47  Guarantor of integrity of the entire study, Study concepts and design and Manuscript preparation. All authors read and approved the final version of this manuscript.

CONFLICT OF INTEREST STATEMENT
The authors have stated explicitly that there are no conflicts of interest in connection with this article.

DATA AVAILABILITY STATEMENT
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

ETHICS STATEMENT
The