Epidemiological and Clinicopathological Features of Breast Cancer in Mauritania


 Background Breast cancer is the leading cause of death in African women. The aim of this retrospective study was to assess the incidence, clinico-pathological characteristics, risk factors and outcome of breast cancer in Mauritania.Methods Demographic and clinico‑pathological features of breast cancer were gathered from 11174 patient files of all cancer types referred to the Centre National d’Oncologie (CNO) between January 2009 and December 2020.ResultsBreast cancer was the most common type of cancer identified in women (30.7%). The diseaseincidence increased from 69 in 2009 to 209 in 2020 with a mean age of 49 year sat cancer detection. Grade 3 tumor was diagnosed in 31.8% patients. Stage 3 and metastasic stage were found respectively in 44.9 % and 22.6% of screened women. 40.4% of cancer patients with satisfactory immunohistochemical data were triple negative breast cancer (TNBC) but no significant variation was found in these features between TNBC and non TNBC groups. A 3‐year survival rate of 63% was observed.ConclusionsThese results support the already published studies on the likely genetic basis of breast cancer in our population.


Abstract Background
Breast cancer is the leading cause of death in African women. The aim of this retrospective study was to assess the incidence, clinico-pathological characteristics, risk factors and outcome of breast cancer in Mauritania.

Methods
Demographic and clinico-pathological features of breast cancer were gathered from 11174 patient les of all cancer types referred to the Centre National d'Oncologie (CNO) between January 2009 and December 2020.

Results
Breast cancer was the most common type of cancer identi ed in women (30.7%). The diseaseincidence increased from 69 in 2009 to 209 in 2020 with a mean age of 49 year sat cancer detection. Grade 3 tumor was diagnosed in 31.8% patients. Stage 3 and metastasic stage were found respectively in 44.9 % and 22.6% of screened women. 40.4% of cancer patients with satisfactory immunohistochemical data were triple negative breast cancer (TNBC) but no signi cant variation was found in these features between TNBC and non TNBC groups. A 3-year survival rate of 63% was observed.

Conclusions
These results support the already published studies on the likely genetic basis of breast cancer in our population.

Background
Although breast cancer (BC) survival is continually improvingin developed states [1], the disease remained a leading cause of death in women from low-and middle-income countries likely due to a late diagnosis, often at advanced stage, combined to the scarcity of adequate and personalized primary treatment [2][3].
The incidence and mortality rate from BC was also affected by other major risk factors such as age, family history and ethnic ascendance [4][5][6]. However, studies on theunderlying etiologies and the prospect of recovery from the diseaseremainedparticularly limited in sub-Sahara African women [7][8][9].
We present here data gathered from the registries of the Centre National d'Oncologie (CNO), on the frequency, demographics, clinico-pathological features and prognosis of breast cancer in Mauritania. These variables were speci cally evaluated in triple negative breast cancer (TNBC) patients and compared with data fromnon TNBC women.

Methods
Medical les of 11174 patients of all cancer types followed at the CNO, the only referring center for oncology in the country, were examined from the period of January 2009, date of opening of the center, through toDecember 2020. Cancer type, Name and date of birth of each patient were recorded.
In this study, detailed demographic and clinical characteristics, available at the center only from 01/2017 onwards,were analyzedfor breast cancer patientsand included age at diagnosis, body mass index (BMI) calculated as weight in kilograms/height in meters square (kg/m2), family history of the disease, cancer staging, histological grading, received therapy and clinical outcome. Immunohistochemical staining (IHC)was carried outon patient tissue samples embedded in para n blocks. Patients with no slides or ambiguous pattern were excluded. Grading (from To to T4) was performed according to the American joint committee on cancer /Union for international cancer control (AJCC/UICC) systems. Evaluation (from 0 to 4) of cancer stage used TNM staging. Triple negative breast cancer (TNBC) subjects were patients with slides showing no antibody staining or a tumor cells uorescence of less than 1% for, concurrently, receptors of estrogen (ER), progesterone (PR) and hormone epidermal growth factor receptor 2 (HER-2).
Local or distant recurrence was de ned by the time span from the end of primary treatment to date of return of the disease in the original site or other part of the body respectively.
Survival duration was determined as the period between the dates of BC diagnosis to the patient death if recorded or the last missed appointment and loss of follow up.

Demographic ndings
Globally, out of the 11174 cancer les of all types (5989women and 5185men) referred to CNO during the period from January 2009to December 2020, breast cancer was found in 1853(16.58 %)patients (Fig. 1).
With a proportion of 30.75%,BC was the most common type of cancer identi ed in the female group. In the male group, the disease was much less frequent with only 11 cases (0.21%)of the total cancer male carriers. Overall, 99.79 % of breast cancer cases were in females and less than 1% in males.
The incidence rate showed asteady increase of the annual number of new women breast cancer cases with 69, 160 and 209 BC diagnosed in 2009, 2015 and 2020 respectively (Fig. 2).
Adequatedemographic and clinico-pathological characteristic, available from 01/2017 to 12/2020, concerned793 women (Table 1).Median age of patients, at cancer detection, was 49 years with most breast cancers diagnosed between 35 to 55 years (52.4%) followed by the group of above 55 years (31%).The fraction in the 20s to 35s age represented 16.6%.Premenopausal status was observed in63.7%of the records. Family history ( rst-and second-degree relatives with breast cancer) was reported by 95 (14.5%) of patientsand388 women (59%) had parents with shared common ascendant. Of 785 patients, 31.3 % were overweight and 49.3 % obese.  BC did not return in most of these patients as 88% and 95% of women diagnosed respectively in 2017 and 2018 did not show any local or distant recurrence.

Discussion
In this study, we have rst addressed the main demographic and clinical characteristics of breast cancer in a cohort of patient women referred to the CNO (Centre National d'Oncology) and assessed the outcome of cancer in the context of these factors. Out of 11175 patient les of all cancer types, BC was the most common in the cohort population (16.56%), particularly in women (30.9%). This standing was also reported by a ten years study (from 2000 to 2009) which included 3305 histological samples analyzed by the department of anatomic pathology (Hopital National de Nouakchott) and showed a prevalence of 14.6% in the whole cohort and 25.2% among the female population [10].As our study was conducted in the following decade of the previous work and covered the single state referring facility for cancer, the data generated were therefore likely representative of the disease evolution in the country. Their concordance re ected an increase in the incidence of breast cancer in our population. Similar percentages of breast cancer in women were reported in neighboring populations such as in Morocco(36%) [8,11]and Senegal (26.1 %) [12].
Most women diagnosed with breast cancer (69%) in our study were ages55 or less.Registriesand community-based studies showed that 70% of women with breast cancer in Sub-Saharan Africa were in the same age group [13].The mean age of 49 years we observed was thus close to the 48 years reported globally in Africa [14]and 46 years in British black women [15].This relatively early onset of breast cancer was lower than late age of 67 years at presentation in white British women [15]. Pre-menopausal status was also predominant in our cohort (63 %) as in two-thirds of black African women with BC [16][17] while most of Europeanwomen (80 %) were postmenopausal at presentation with the disease [18].
We also observed that most patients (66%)hadmoderate to poorly differentiated tumors with widely spread stage 3 (44%) or metastasized (22%) cancer when diagnosed with BC. A similar outline was reported in a 12 sub-saharan countries study (Zimbabwe, Benin, Seychelles, Ethiopia, Mauritius, South Africa, Kenya, Mozambique, Mali, Namibia, Uganda and Cote d'Ivoire)showing that 64.9% of women patients were diagnosed in late stages, when treatment became weakly effective, of which 18.4% being metastatic at diagnosis [19].This late advanced stage at presentation, very likely accentuated by poor socioeconomic conditions and lack of access to adequate healthcare, could be therefore determinant in the low 2-and 3-year observed survival rate of 74% and 64% observed in our cohort and the overall relative survival (RS) of 61.4% (59.1-63.5) at year 3 and 52.3% (49.9-54.6) at year 5 [19]in BC patients across sub-Saharan Africa. In contrast, 79% and 89% of women with breast cancer respectively in Europe and the US had not died from their cancer 5 years later after diagnosis [20].
Despite evidences reported from many large cohort studies linking overweight to breast cancer risk [21][22], nearly 40% of women worldwide were overweight in 2020 [23].
We have shown a high cumulative prevalence of overweight and obesity among our patients. Obesity was for centuries desirable and a sign of wealth in various African countries [24]. A traditional practice of force-feeding teenage girls (known as leblouh in our country) has indeed been prevalent in Mauritania and several other African populations [25][26].
Although lifestyle choices and low provision of healthcare services in African populations may be determinant in the disease expected development, various studies have shown that other risk factors may take part in BC prognosis such as patient race or ethnic origin, parents'consanguinityand age at onset [27][28][29].
The comparableearly age atbreast canceronset observed in our cohort, globally in Sub-Saharan [13] and British black women [15]against a relatively late age of 67 years in white British women at presentation [18] was in this context relevant.African-American women havealso higher rates of grade 3 than their Caucasian counterparts [30][31].
Race related differences among BC patients have been attributed to varioushereditary grounds including breast cancer susceptibility genes and endogenous hormones [32].For instance, in the US,the frequency of samples tested negative for receptors of progesterone,estrogen and HER2 protein (TNBC) was higher in African-American women (28%) compared to Caucasian women (12%) [33]. TheMauritanian population is composed of three main groups all Muslims but of different race origin [34]: the white Maures (WM) speaking Hassaniya, a berber-arab dialect. This group ethnically and culturally self-identi es with the neighboring North Africa populations. The black Maures (BM) also speaking Hassaniya but share the same race origin with the third group, the black African Mauritanians (BAM), as both descended from native sub-Saharan Africans.
The global TNBC prevalence (40.4%)observed in this study was intermediate between the percentages of 28.5% in the white Moors and 71.51 % in the black Moors-black Africans group respectively. The frequency observed in the white Moors (28.5%) although slightly higher, was comparable to the percentages in North African populations [35][36].The frequencies in black Moors-black Africans group, is also similar to those reported in sub-Saharan African women-based studies [37][38] which is consistent with the common African ascendance above mentioned.
This ethnically associated repartition of percentages in TNBC also concords with the distribution of other biomarkers we reported previously in our population [39][40]. However, although differences of percentages between TNBC and NTNBC patients were observed, all parameters we analyzed did not reach the level of statistical signi cance set.

Conclusions
We have provided data from a representative cohort on the frequencies of BC in Mauritania, evaluated the main demographic and clinical characteristics which may affect the disease prognosis. The results, consistent with already published studies support a genetic basis of breast cancer in our population. Further studies increasing the cohort size and extending the time span of following the patients may optimize data for signi cant correlation assessment.

Declarations
Ethics approval and consent Approval to this study was given by the ethics committee of the Université de Nouakchott Al-Asriya, Mauritania. The informed consent of patients referred to the CNO was obtained. All methods were carried out in accordance with relevant guidelines and regulations.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.
Funding no speci c funding was obtained for this study carried out between the UNA and the CNO.
Authors'contributions SBR collected and organized all cancer les; She contributed to epidemiological and Immunohistochemistry data analysis; CTH contributed in pathological data analysis; EZ initiated the paper conception and contributed to the manuscript progress; MS contributed to epidemiological data analysis; FV contributed to epidemiological data analysis; MVZ contributed to patholgical data analysis; AT examined Immunohisto chemical slides (IHC) and contributed in paper conception and writing; MK contributed in paper conception and paper writing; AH were the major contributor in coordinating all data analysis and writing the manuscript. All authors read and approved the nal manuscript.