The effective management of patients with breast cancer need knowledge of the hormone receptor status and the HER2 overexpression.
This study is the first conducted in Togo, which determine the molecular groups of breast cancer based on the IHC expression of ER, PR and HER2.
In this retrospective study, IHC was performed in only the third (37.5%) of breast cancers diagnosed from 2016 to 2020 because of the unavaibility and the high cost of receptor testing. As in our country, access to IHC remains limited in most countries of Sub-Saharan Africa [12–15].
The mean age at diagnosis was 52.05 ± 12.38 years with age range between 30 and 85 years, with half of patients having more than 50 years old.
This age at presentation in our current study is in agreement with a previous Togolese study were the mean age at diagnosis of cancer was 50 years [10]. This is also similar to the mean age reported in Ghanaian patients at Korle bu Hospital [16] and in Indian women [17], but different from the relatively younger age of women with breast cancer reported in several studies in others countries in Africa [12, 18–22].
In this study, majority of patients (70.9%) was postmenopausal at presentation, this is contrary to findings from others African studies where majority of women were premenopausal [14, 22–26].
As observed in most breast cancer studies worldwide, invasive ductal carcinoma was the most dominant histological type of tumor in Togolese patients.
Our study revealed that less than 10% of women with breast cancer have well differentiated tumors, 42% had positive lymph nodes and the majority had T3 or T4 tumors stage. These unfavorable clinicopathological characteristics such as high grade, large tumors size, axillary lymph node involvement and advanced stage are the same as those reported in several studies [14, 15, 17, 21, 26, 27]. The advanced stage of diagnosis in our patients could be explained by the delay of consultation, the absence of national breast cancer screening program in the population, the poor health care facilities and the use of traditional medicine.
A recent systematic review and meta-analysis by Eng [28] reported most breast cancer cases in Africa as being Hormone Receptor positive. Nevertheless, the positivity of hormonal receptors in breast cancer remains varied and heterogenous in countries.
In the present study, positive ER immunostaining was found in 50.4% of cases. Our result is similar to that reported by Bangaly in Guinea [12]. However, our finding is lower than those reported in western countries [5] and Asia [ 29, 30]. The proportion of patients expressing ER (50.4%) is superior to those of PR positive (37.6%). The same observation was reported by different authors [12, 23, 29, 31].
Our finding in the current study, have implication for management of breast cancer in Togo.
Indeed, in the past, patients with breast cancer were treated blindly with tamoxifen. However, near fifty percent (50%) of our patients are ER/PR negative and may not been suitable for hormonal therapy. In addition, the high frequency of postmenopausal women in this study points out that hormonal therapy should be given according to the menopausal status for adequate care.
HER2 overexpression was seen in 15.4% of patients. This result is in agreement with the literature data which reported 15–20% of HER2 positive in invasive breast cancer [32]. Our result is similar to the finding reported in Tanzania [15] and Ivory Coast [23] but is different from the rate found in Ghanaian women [ 16, 24].
Even if our result was included in the interval rate of HER2, it could have been underestimated as FISH was not performed to ascertain the true HER2 status of tumors with an equivocal IHC score of 2+.
In our study, the distribution of molecular subtypes found a high frequency of Luminal A subtype (42.7%) followed by triple negative (41.9%), Luminal B subtype (9.4%) and HER2 enriched (6%).
This distribution was found to be in concurrence with the results obtained from others western [12, 23] and north African countries [33, 34].
Luminal A subtype which is less aggressive type of breast cancer was predominant in our study (42.7%). A similar rate has been reported by several authors from East Africa [25]. However, the proportion of Luminal A in our patients was lesser than that reported in North Africa [33, 35], in Saudi Arabia [29] and Western countries [5]. As expected, Luminal A subtype was associated with favorable clinic and biological characteristics.
Another remarkable finding in this study is the high proportion of TNBC which accounted for 41.9%. Our result is lower compared with that noted for Nigerian [36] and Senegalese women [22] but much higher as compared with those reported from Ghana [16], Ivory Coast [23], Angola [20]and Ethiopia [25].
Many studies have reported that TNBC are the dominant phenotype in native African women [37–40] and African Americans [5, 41, 42] compared with white women but in our patients, Luminal A subtype was slightly dominant. The TNBC are considered more common in younger women and associated with aggressive clinicopathologic characteristics [27, 38]. Our findings corroborate with these reports.
Luminal B subtype defined here as simultaneous expression of ER, PR and overexpression of HER2 was found in 11 patients (9.4%). Our result is lower compared with the finding reported by Hadgu in Ethiopian women where Luminal B was the second most prevalent subtype (26%) [25] but much higher as compared with those reported from Senegal (4.6%) [22] and Angola (7.9%) [20].
HER2 enriched tumors are known to be associated with particular poor breast cancer outcomes but the use of HER 2 targeted therapy improve survival among breast cancer patients whose tumors overexpress HER2.
HER2 enriched subtype was found in 6% of our patients. Similar rate has been reported by Effi in Ivorian women [23]. In comparison with the others subtypes, HER2 enriched tumors were observed to be associated with larger tumor size. The correlation between the different subtypes and tumor size was statistically significant (P = 0.026). Significant association between tumor size with breast cancer subtype has also been reported in Sudanese and Eritrean women [43]
In several study, some authors reported significant association between molecular subtype and histological grade [19], age [25, 33] and menopausal status [33] but in this current study, we did not find any correlation between molecular subtype and age, stage, menopausal status or histologic grade.
This study demonstrated interesting observations about breast cancer in Togo. However, some limitations should be mentioned such as the unavaibility of the cellular marker Ki 67 for all patients, the lack of evaluation for HER2 equivocal results using Fluorescence in situ and the absence of cytokeratin5/6 to identify the different subset of triple negative cancer.