The current trend in analysing the clinical outcome of a patient with breast cancer is to examine predictive and prognostic factors related to the patient and her tumour. The former is related to the degree to which the patient could respond to a specific therapy, while the latter is related to the metastatic potential of the tumour. Several studies have examined predictive and prognostic factors, such as the age of the patient, tumour size, grade, proliferation, hormone status, histological type of the tumour, and lymph node involvement, to name a few [6
]. With the advancement in science and technology, molecular markers have been added to the above list, in an attempt to help the clinicians to better monitor the course of the disease and predict its outcome [10
]. In this study, we conducted a comprehensive analysis of breast carcinomas taken from 166 patients in Kuwait. We limited our study to a sample size of 166 because these were the only cases for which we had complete information about the patient and the tumour. Also, these were the only cases whose paraffin blocks had enough tissue in them so that taking extra sections from these blocks for our study did not jeopardize the amount of tissue remaining in them in case of future examination.
In contrast to what is commonly known about a rising incidence of breast cancer with age, our results showed that 70.5% of the patients we examined were young with an age not exceeding 55 years. The mean age of these patients was 40, and the majority of them were between the ages of 30 and 55. This age distribution is significantly younger than what is currently seen in Western and Arab countries [1
], and requires further careful examination to determine the nature of the predisposing factor(s). One possible explanation is that traditional marriages among first-degree relatives in Kuwait are very common, and, accordingly, hereditary factors could play a major role. Another factor could be the degree of obesity associated with a diet high in fat, carbohydrate, and protein, and lack of exercise, which have been prevalent in Kuwait for the past 20 years. We are in the process of conducting a comprehensive study at the Mubarak Al-Kabeer Hospital, in order to examine the above and some other predisposing factors. An interesting finding in this study is that the carcinomas from the patients in the above age category predominantly over express Her-2. These results confirm those obtained from studies where an association between age of breast cancer patients and their tumour over expression of Her-2 was found [12
]. Other studies, however, did not find any significant association between the former and the latter [16
]. We also found that there is a significant association between the nature of the tumours' expression of ER and the age of the patients. Patients with ER negative tumours were mostly young (< 30 years and between 30–55 years), as compared to positive ER expression in patients aged above 55 years.
Studies where an association between the nature of breast carcinomas' expression of ER and age of the patients was found have been documented. A recent study conducted by Jalava et al. [18
] reported a significant association between the former and the latter. Ferno et al. [19
] reported a lower expression of ER in patients below 50 years old, and Quong et al. [20
] found that the expression of ER by breast carcinomas increases with age. Similar findings to Quong were observed by Holdaway and Mountjoy [21
], Clark et al. [22
], Wilking et al. [23
], Gaskell et al. [24
], Rhodes et al. [25
], and Tominaga et al. [26
]. Other studies have, however, found no association between the age of the patient and the degree of expression of ER by the tumour [27
In our study, we did not find any significant association between the age of the patients and their tumour expression of PgR. Similar findings were reported by Holdaway and Mountjoy [21
], Clark et al. [22
], and Wilking et al. [23
]. Other studies, however, have reported a higher tumour expression of PgR in patients older than 59 years, as compared to those between 50 and 59 years [19
]. The over expression of Her-2 and lack of ER expression by the tumours of the patients aged below 55 years in our study might explain the high mortality rate reported earlier [1
] among these patients, since such tumours often become resistant to adjuvant and hormone therapies.
Other characteristics that we examined were the margins of the tumours, laterality (right versus left breast), and the type of surgical management. Tumour margins often represent a reliable source of positive or negative disease outcome. In our study, the margins of the tumours were mostly irregular (stellate). Having an irregular (stellate) margin means that the tumour is not confined and there is a potential for metastasis. This is further confirmed by the degree of malignancy of the tumours, which was significantly associated in our study with bad prognostic markers such as over expression of Her-2 and negative expression of ER. Irregular (stellate) margins did not associate with PgR expression. An association between the tumour margin status and Her-2, ER, or PgR expression has been investigated in several studies [28
]. Putti et al. [29
] recently demonstrated that breast tumours with a pushing margin (another terminology for irregular or stellate margin) were found to be ER negative and over expressing Her-2. A similar association with Her-2 over expression was reported by Miller et al. [38
]. The circumscription of the tumour margin was significantly associated with negative PgR expression in a study conducted on 281 women with breast cancer in Finland [30
]. In another study conducted on 980 patients with breast cancer, and in which the patients were divided into three age categories (≤ 35 yrs, 36–50 yrs, and > 50 yrs), Fowble et al. [42
] reported that young patients had significantly more association between tumour margin status and negative ER expression. Unlike our results, Kim et al. [37
] reported an association between tumour margin and positive rather than negative ER expression: moreover, the authors found an association with positive PgR expression. Lack of an association between the tumour margin and hormone receptor status was reported in a study conducted on 254 patients undergoing partial mastectomy [35
]. Similarly, Horiguchi et al. [31
] found no association with tumour expression of ER.
In our study, we also took breast laterality into consideration. The number of carcinomas present in the right breast was slightly more than in the left one (53.6% versus 42.2%). Such tumour laterality was significantly associated with Her-2 over expression, but not with ER or PgR expression. However, since the difference in the location of the carcinomas (right versus left breast) was not significant, we prefer not to deduce, at this stage, any conclusions in relation to the expression of the above markers. The small marginal difference in relation to the location of the tumour in the right versus left breast in our study was similar to the one reported by Largent et al. [43
]. When analysing the demographic and tumour characteristics of early breast cancer patients, the authors found that 52% of the carcinomas were present in the right breast, as compared to 48% in the left one. The only studies that we found in the literature, which attempted to find an association between breast laterality and hormone receptor expression were the ones conducted by Tominaga et al. [26
] and by Borisenkov and Bazhenov [44
]. Borisenkov and Bazhenov reported that the degree of expression of hormone receptors by breast carcinomas taken from Russian patients significantly depended on whether the tumour was present in the right or left breast [44
]. On the other hand, such an association was lacking in the study conducted by Tominaga et al. [26
] on Japanese women with breast cancer. In our study, the percentage of patients who underwent total mastectomy with axillary clearance was the highest. Interestingly, there was a significant association between this type of surgical management and the tumour expression of Her-2, ER, and PgR. This association was also seen in the patients who underwent total lumpectomy with axillary clearance. The carcinomas of the patients in both surgical categories predominantly over expressed Her-2, and were mostly ER and PgR negative. According to our knowledge, this is the first time where such an association has been reported.
The histological characteristics of breast carcinomas have been investigated in several studies trying to correlate the histological type/subtype of the tumour with the disease outcome such as local recurrence, site of recurrence (ipsilateral versus bilateral), metastasis (regional versus distant), and response to therapy. For instance, even earlier reports have shown that the 30-year survival rate of women with certain histological types of breast cancer such as tubular or lobular is greater than 60% as compared with less than 20% for women with breast cancer of no special type [45
]. Chen et al. [46
] have demonstrated that breast cancer of lobular histological type is more often bilateral when compared to other types.
In our study, we conducted a comprehensive histological analysis of the breast tumours. Our results showed that 83.1% of the carcinomas were invasive. Eighty two percent of the in situ
carcinomas were ductal, and they were predominantly comedo or cribriform. The invasive carcinomas were mostly ductal-not otherwise specified, lobular, or tubular/cribriform, and the breast tissue type surrounding the carcinoma was predominantly adenosis or fibrocystic. Our results are similar to those reported by Andersson et al. [47
], where the incidence of in situ
breast carcinoma was 16%, in contrast to a higher incidence (26%) reported by May et al. [48
]. When we analysed the above parameters in relation to the tumour expression of Her-2, ER, and PgR, we found that the invasive carcinomas predominantly over expressed Her-2 and were mostly ER and PgR negative. On the other hand, the in situ
tumours were mostly Her-2 negative, and ER and PgR positive. This provides more evidence to the hypothesis that aggressive tumours seem to lack hormone receptors and to over express Her-2. Our results are similar to those reported by Zafrani et al. [49
], where 81% of the in situ
tumours the authors examined were ER positive and 73% were PgR positive. We also found an association between some histological subtypes of the in situ
ductal tumours and Her-2, ER, and PgR expression, whereby the comedo and cribriform subtypes significantly over expressed Her-2 and were ER negative. This is similar to the findings reported by Janssens et al. [50
] and by Provenzano et al. [51
], and confirms previous reports that showed lack of ER expression in comedo histological subtype of ductal carcinoma in situ
]. As far as the histological subtypes of invasive carcinoma are concerned, we found that the most common subtypes were ductal-not otherwise specified, lobular, and tubular/cribriform. This confirms the WHO classification of invasive breast carcinomas in relation to the percentage occurrence of these subtypes [3
]. The ductal-not otherwise specified carcinomas over expressed Her-2 and were predominantly ER negative. This again shows the association between the nature of the biological expression of Her-2 and ER by the tumour and its degree of malignancy, since it has been argued that ductal-not otherwise specified carcinomas are the most aggressive type of breast cancer. This is based on the fact that its tumour cells are often seen infiltrating into the surrounding tissue, including perivascular and perineural spaces, as well as lymphatic and blood vessels. We also found a significant association between Her-2 over expression and lack of ER expression and the lobular histological subtype. This could imply that, although this histological subtype of invasive breast carcinoma is less common than the ductal-not otherwise specified one, lobular carcinomas could be equally aggressive. Various and sometimes contradictory reports have been published in the literature regarding the association between the expression of Her-2 and hormone receptors and the various histological subtypes of invasive breast carcinomas. In a recent study conducted by Jalava et al. [18
], the breast carcinomas' ER expression was found to be greater in lobular than in ductal tumours. Other authors have reported that the mucinous type is associated with an increase in the expression of ER and with a decrease in the expression of Her-2 [52
]. Similarly, Her-2 was found to be inversely associated with ER status based on the histological type in a study conducted by Coradini et al. [53
]. Still, other studies confirmed a lack of an association between the histological type of breast tumour and its hormone receptor status [54
As far as other tumour characteristics are concerned, we have noticed that tumours with a surrounding breast tissue that is fibrocystic in nature commonly over expressed Her-2. Whether there is a direct biological association between the former and the latter remains to be investigated.