Evaluation of p53 in breast cancer and its correlation with various histological prognostic factors

Background: Carcinoma of the breast is the most common non-skin malignancy in women. The p53 gene is one of the most commonly mutated genes described in human neoplasia. Mutant p53 protein has greater stability and longer half time than the wild type protein that can be detected by Immunohistochemistry. This study was conducted to evaluate the expression of p53 in breast carcinoma and its correlation with various histological prognostic markers to determine its significance as a prognostic marker in breast cancer. Materials and Methods: The prospective study included 52 modified radical mastectomy specimens diagnosed as breast cancer in the Department of Pathology at SRMSIMS, Bareilly between November 2016 and April 2018. Routine H&E staining and immunohistochemical analysis for ER, PR, Her2/neu and P53 was carried out in all the cases. Results: Majority of the cases belonged to the age group 30-39 years and the most common cancer type was infiltrating ductal carcinoma. Significant correlation of p53 was observed with ER and PR expression, however no significant correlation could be found with tumor type, grade, size, type of margin, necrosis, stromal response, lymphovascular invasion, lymph node status and Her2/neu status. Conclusion: P53 was found to have a significant inverse correlation with ER and PR expression only. The results do not resolve whether detectable p53 protein expression represents a random product of dedifferentiation or an important feature of the malignant phenotype, playing a key role in tumor behavior. © 2020 Published by Innovative Publication. This is an open access article under the CC BY-NC license (https://creativecommons.org/licenses/by-nc/4.0/)


Introduction
Carcinoma of the breast is the most common non-skin malignancy in women and is second only to lung cancer as a cause of cancer death. A women who lives to age of 90 has one in eight chance of developing breast cancer. 1 The location of breast carcinoma is usually indicated in relation to the breast quadrants. Approximately 50% are in upper outer quadrant. Majority of the breast malignancies are adenocarcinoma other types are less than 5%. 2 The major risk factors for the development of breast cancer are Age, Age at menarche, Age at first live birth, First -degree relatives with breast cancer, Atypical hyperplasia, Race/Ethnicity, Estrogen exposure, Breast density, radiation exposure, Carcinoma of the contralateral decision. A wide range of potential prognostic factors have now been studied, some well established, eg. ER/PR, HER2/neu expression while some at the developmental stage eg. p53, bcl2, Ki-67, Cyclin D1, Cyclin E, ERβ , etc. 4 Nearly one third of breast cancers have mutation in p53, a tumor suppressor gene. Literature suggests that overexpression of HER-2 and p53 may have an adverse effect in breast cancer. 5 The p53 gene is one of the most commonly mutated genes thus far described in human neoplasia with mutations estimated to occur in upto 50% of all cancers. 6 First described in 1979, and initially believed to be an oncogene, p53 was the first tumor suppressor gene to be identified. 7 It is located on short (p) arm of chromosome 17 and is another proved breast cancer progression gene that regulates cell cycle and DNA repair. Unlike normal p53, nonfunctional mutated p53 accumulates in the nucleus of tumor cells, and therefore, it can be detected by immunohistochemical analysis. 8 The association between p53 alterations and clinical outcome in breast cancer has been the subject of numerous investigations. P53 immunoreactivity could bear some prognostic significance. 9 Therefore this study was conducted to evaluate the expression of p53 in breast carcinoma and its correlation with various histological prognostic markers to determine its significance as a prognostic marker in breast cancer.

Material and Methods
The present study was conducted in the Department of Pathology in our institute. Study duration was from November 2016 to April 2018. A total of 52 cases presenting with breast lumps in surgical OPD and diagnosed as suffering from breast cancer on FNAC/Trucut needle biopsy and who further underwent modified radical mastectomy in the department of surgery were enrolled for the study. The mastectomy specimens received in the Department of Pathology were examined for gross details. Representative sections were taken and processed in a tissue proccesor to make paraffin embedded blocks. 3 to 5 µm thick sections were cut and stained with Haematoxylin and Eosin (H & E) and tumor sections were also stained with ER, PR, Her2/neu and p53 antibodies (IHC). The stained sections were studied in detail for the presence of tumor, type of tumor (WHO classification 2012) and grade of tumor (Nottinghams grading system), type of tumor margin, presence and absence of necrosis, lymphovascular invasion, stromal reaction and lymph node status.
P53 scoring was done as per semi quantitative method as described by Wee et al 10 which takes into account the intensity of staining and the percentage of nuclei stained. Total score was then obtained by the sum of intensity score and percentage score, ranging from 0-6 (Table 1).

Scoring system for ER and PR4
Scoring for ER/PR was done using the Allred score which takes into account proportion and intensity of positive tumors cells which are scored from 0-5 and 0-3 respectively. Total score is obtained by the sum of proportion score and intensity score, ranging from 0-8 (Table 2).

Scoring system for HER2/neu4
The scoring method recommended is a semiquantitative system based on the intensity of reaction product and percentage of membrane positive cells, giving a score range of 0 to 3+ (Table 3). Association/correlation of p53 expression with independent variables such as histological type, grades of tumor, tumor size, tumor margin, necrosis, lymphovascular invasion, stromal reaction, lymph node status, ER, PR, and Her2/neu expression was analysed by applying Pearson's Chi Square test, Fisher's exact test. Statistical analysis was performed using SPSS version 23 and p-value of less than 0.05 was considered statistically significant at 95% level of significance.

Result
A prospective study was undertaken on 52 mastectomy specimens diagnosed as breast cancer in the Department Majority of the cases, 17 (32.7%) belonged to the age group, 30-39 years followed by 14 (26.9%) cases in the age group 40-49 years. Male to Female ratio was 51:1. The most common type of tumor was found to be infiltrating ductal carcinoma constituting 43 (84.6%) cases, followed by 6 (9.6%) cases of medullary carcinoma, 2 (3.8%) cases of infiltrating ductal carcinoma with medullary features and 1 (1.9%) case of lobular carcinoma. Most of the cases of infiltrating ductal carcinoma belonged to age group of 30-39 years. Medullary carcinoma was found to be most common in the age group 40-49 years. Nottingham histological grade 2 was the most common grade in the present study comprising 30 (57%) cases followed by grade 3 with 13 (25%) cases and grade 1 with 9 (18%) cases. In the present study the tumor size was classified according to the TNM classification. In 31 (59.6%) cases the tumor size was between 2-5 cm followed by 19 (34.6%) cases of tumor size > 5 cm and 2 (5.85%) cases of tumor size <2 cm. In the present study 38 (75%) cases had infiltrative type of tumor margins while only 14 (25%) cases had pushing type of tumor margins. Tumor necrosis is a less commonly studied prognostic factor. In this study necrosis was present in only 23 (44%) cases. Inflammmatory stromal response was found in 29 (56%) cases while 23 (44%) cases either showed only desmoplasia or did not have any stromal response. lymphovascular invasion which was present only in 22 (42.3%) cases. In the present study 22 (42.3%) cases had 4 or more lymph nodes positive for tumor metastasis, followed by 17 (32.7%) cases which had <4 lymph nodes positive and 13 (25%) cases with no lymph node metastasis. Only 12 (23.1%) cases were positive and 40, (76.9%) were negative for estrogen receptors. Only 19 (36.5%) cases were positive and 33 (63.5%) cases were negative for progesterone receptors. 23 (44%) cases were positive and 29 (56%) cases were negative for Her2/neu expression. 32 (61.5%) cases were positive for p53 expression and only 20 (38.5%) cases were negative.
25 out of 43 cases of infiltrating ductal carcinoma were positive for p53, both the cases of infiltrating ductal carcinoma with medullary features were positive for p53. Of the 6 cases of medullary carcinoma 4 were positive for p53. Only one case of lobular carcinoma was included in the study which was negative for p53. No significant association was seen between p53 and tumor type (p value = 0.52). 5 (9.6%) cases of grade 1, 19 (36.5%) cases of grade 2 and 8 (15.4%) cases of grade 3 were positive for p53 expression. No significant correlation was seen between the tumor grade and p53 expression (p value = 0.9). 1 (1.9%) case of tumor size < 2 cm, 17 (32.7%) cases of tumor size between 2-5 cm and 14 (26.9%) cases of tumor size ≥5 cm were positive for p53 expression. No significant correlation was seen between tumor size and p53 expression. Majority of tumors in this study had infiltrative type of tumor margin and 25 (48.1%) cases with infiltrative type of tumor margin were positive for p53. No significant association was found between type of margin and p53 expression. No significant correlation of p53 expression was observed with necrosis, stromal response, lymphovascular invasion. 14 (26.9%) cases with ≥ 4 lymph nodes were positive for p53 expression followed by 11 (21.1%) cases with 1-3 lymph nodes and 7 (13.5%) cases with no positive lymph nodes. No significant correlation was found between lymph node status and p53 expression (p value = 0.8). In our study 28 (53.8%) ER negative cases were positive for p53 expression and significant inverse correlation was seen between ER and P53 expression (p value = 0.02). 24 (46.2%) PR negative cases were positive for p53 and a significant inverse correlation was seen between PR and p53 expression. P53 expression was positive in 14 (26.9%) Her2/neu positive cases and 18 (34.6%) Her2/neu negative cases. No significant correlation was found between Her2/neu and p53 expression.

Discussion
Breast cancer is no longer seen as a single disease but rather a multifaceted disease comprising of distinct biological subtypes with diverse natural history, presenting a varied spectrum of clinical, pathologic and molecular features with different prognostic and therapeutic implications. 11 Pathological variables such as tumor size, histological type, histological grade, lymph node metastasis, vascular space invasion, tumor cell proliferation, extent of ductal carcinoma in situ are the predictors of prognosis and for the need of adjuvant therapy. Biomarkers such as ER, PR and HER-2neu expression represent the most acceptable ones for predicting prognosis, response/resistance to treatment and in deciding the use of newer drugs such as transtuzumab in the case of HER-2over expression. 8 Mutations in the p53 gene or increased expression of the p53 protein as an indirect marker of a mutation have been described as prognostic factors for a long list of human malignant tumours, including breast cancer. Recent studies have suggested that p53 status might have a different predictive value for the efficacy of anthracycline/alkylating agent based chemotherapy regimen between triple negative & non triple negative breast cancers. 12 In the present study age of the patients ranged from 25 to 76 years and maximum number of cases were between 30-39 years (32.7%) followed by 40-49 years (26.9%). The mean age of presentation was 45.31 ± 10.81years. The findings are almost similar to that reported by Sheikhpour et al (44.75±9.5 years) 13 and Kanna et al (58% cases between 36 to 50 years of age), 14 while the patients were slightly younger than cases reported by Patnayak et.al (50.7 ± 11.5 years), 5 Abdollahi et.al (50.2±12.3 years) 15 and Gupta et.al (50 years). 8 It is a documented fact that advancement of age increases the risk of breast cancer and most women are over the age of 60 years when diagnosed. 13 Although there is evidence that Indian women are more likely to develop breast cancer at earlier ages than their Western counterparts. 16 In this study 84.6% cases were of infiltrating ductal carcinoma which was in concordance with the findings of Sheikhpour et.al (84.6% cases), 13 Goel et.al (85% cases), 17 Patnayak et.al 5 19 Neharika et al. 20 and Goel et al. 17 reported a higher frequency of grade 3 tumors.
One of the most important and well established prognostic factor in carcinoma breast is tumor size. 3 In this study tumor size was divided into three categories as per the TNM classification system. 31(59.6%) cases had tumor size between 2-5 cm constituting the largest group. 19 (36.5%) cases had tumor size >5 cm. This is in concurrence with the results obtained by Patnayak et al, 5 Piplani et al, 19 Gupta et al 8 and Neharika et al 20 who found 79.6%, 67.7%, 63.9%, 53.9% cases of tumor size between 2-5 cm in their studies respectively, whereas in the study done by Taucher et al the tumors were predominantly less than 2 cm in size which could be due to more awareness and early detection programs prevalent in the western countries. 21 Type of tumor margin, presence and absence of necrosis, inflammatory type of stromal response and lympovascular invasion are described as factors associated with poor prognosis in the literature, 4 but these are less frequently studied. 23 (44%) cases had necrosis associated with the tumor. Carter and colleagues gave a figure of 40%, compared with 60% estimated by Fisher and coworkers. 4 22 (42.3%) cases showed lymphovascular invasion. The reported range for lymphovascular invasion extends from 10% to 54%. 4 In this study 22 (42.3%) cases had ≥ 4 lymph nodes positive for tumor deposits, followed by 17(32.7%) cases having 1-3 lymph nodes positive and 13(25%) cases with no lymph positive for tumor deposits. Similar observations were made by Goel et al. 17 and Neharika et al. 20 Friedrichs et al. observed that risk of recurrence increases with the number of lymph nodes involved. 22 The prevalence of hormone receptor positive breast cancer in Asian countries has been found to be lower than those in the western world. 23 Indian literature reports estrogen receptor positivity varying between 30-50%. 24 In the present study we found only 12 (23.1%) cases positive for estrogen receptors. Desai et al., 25 Abdollahi et al., 15 Neharika et al., 20 Piplani et al. 19 and Patnayak et al. 5 reported prevalence of 32.6%, 36.4%,38.16%, 41.5% and 47.6% ER psitive cases respectively.
In the present study 32 (61.5%) cases were positive for p53 expression. This was quiet close to observations of Patnayak et al, 5 Song et al 31 and Sekar et al 32 who reported 69.2%, 51.6% and 71.6% p53 positive expression respectively. Many authors have reported low positivity of 22%, 14 28.8%, 13 29.6%, 33 34%, 34 36%, 35 37.1%, 5 37.4%, 15 38.5%, 22 45.5%, 36 45.7%, 17 47.4% 20 and 48.9%. 23 In contrast to this Gupta et al 8 reported 88.9% p53 positive expression. This variability in the results may be attributed to the diversity of procedures and reagents used by different investigators. Significant differences have been reported 36 in p53 immunoreactivity depending upon the antibody used. Another important difference among the reported studies was the lack of a standardized immunohistochemical criteria to classify a carcinoma as p53 positive.
In our study we did not observe any significant association between p53 expression and type of tumor. This is in concordance with Sheikhpour et al 13 and Goel et al. 17 In contrast to this Jacquemier et al 37 observed a significant correlation. This could be because we did not have significant number of different subtypes of breast carcinoma unlike their study.
In our study the p53 expression in grade 1, 2 and 3 tumors was found to be 55.6%, 63.3% and 61.5% respectively, unlike the study done by Ranade et al 38 in which the expression of p53 increased with the tumor grade and was 33%, 52% and 67% respectively in grade 1, 2 and 3. This study showed no significant association between grade of tumor and p53 expression. This is in accordance with Sheikhpour et al 13 and Abdollahi et al. 15 However in contrast to our study many studies have reported a significant positive association (Cattoretti et al, 36 Thor et al, 39 Rosen et al, 40 Friedrichs et al, Jacquemier et al, 37 Neharika et al, 20 Shoukouh et al, 18 Piplani et al, 19 Al-Joudi et al, 33 Sekar et al, 32 Kanna et al, 14 Goel et al. 17 and Gupta et al). 8 This variation could be because of low sample size in our study.
In this study a higher frequency of cases with tumor size ≥5 cm showed p53 positive expression as compared to the number of cases with tumor size between 2-5 cm but as the majority of the cases in our study had tumor size between 2-5 cm therefore higher overall p53 positivity was observed in this group i.e in 17 (32.7%) cases. No significant correlation was observed between tumor size and p53 expression. Our findings are similar to the findings of Friedrichs et al, 22 Shoukouh et al, 18 Al-Joudi et al, 33 Abdollahi et al, 17 Goel et al 17  48.1% cases with infiltrative tumor margin and 11.5% cases with pushing margins were positive for p53 expression and no significant correlation was seen between p53 expression and type of tumor margin. We could not find similar correlation being done by any other study.
Similarly no correlation was seen between p53 expression and presence and absence of necrosis whereas Kanna et al. 14 reported that cases with p53 positivity demonstrated mild necrosis in the tumors. 29 (56%) cases in this study had lymphocytic stromal response and 20 of these 29 cases were positive for p53 expression. However, no significant correlation was seen between p53 expression and lymphocytic type of stromal reaction. Kanna et al. 14 reported p53 overexpression in tumors with severe lymphocytic reaction. 38.5% cases with lymphovascular invasion were positive for p53 expression but no significant correlation was seen between lymphovascular invasion and p53 expression.
In this study no significant correlation of p53 expression was observed with type of tumor, tumor grade, tumor size, type of tumor margin, presence and absence of necrosis, lymphocytic stromal reaction and lymphovascular invasion. Hence p53 does not seem to be associated with tumor aggressiveness.
Estrogen receptors were negative in 40 cases and of these 28 cases were positive for p53 expression. A significant correlation was found between ER and p53 expression. We observed that p53 is inversely associated with ER. Our results are in concordance with Cattoretti et al, 36 Thor et al, 39 Rosen et al, 40 Friedrichs et al, 22 Jacquemier et al, 37 Neharika et al, 20 Shoukouh et al, 18  A statistically significant negative correlation was also found between PR and p53 expression which is in concordance with the studies done by Friedrichs et al, 22 Jacquemier et al., 37 Neharika et al 20 Shoukouh et al 8 and Ranade et al. 38 However, Piplani et al 19 and Al-Joudi et al 33 did not find any significant association between the two. However, no significant correlation was seen between Her2/neu and p53 expression, which is similar to Abdollahi et al 15 and S Han Jeong et al. 41 In contrast Patnayak et al, 5 Rashed et al., 34 L Ding et al 44 and Neharika et al 20 reported a significant positive association between Her2/neu expression and p53 expression. Coexistence of Her2/neu overexpression and p53 protein accumulation has been suggested to be a strong prognostic molecular marker in breast cancer. 45 The results of this study do not resolve whether detectable p53 protein expression represents a random product of dedifferentiation or plays any significant role in tumor behavior.

Conclusion
In present study highest incidence of infiltrating ductal carcinomas was noted in age group 30 -39 years and of medullary carcinomas in 40-49 years. No statistically significant association was observed between p53 expression and type of tumor, grade of tumor, size of tumor, lymphovascular invasion, necrosis, stromal reaction, lymph node status and Her2/neu expression. Thus p53 does not seem to be related to tumor aggressiveness. The only significant correlation was observed with ER and PR expression. P53 was found to be inversely correlated with ER and PR. The results do not resolve whether detectable p53 protein expression represents a random product of dedifferentiation or an important feature of the malignant phenotype, playing a key role in tumor behavior. Although no significant correlation was seen between p53 expression and type of tumor, grade of tumor, size of tumor, lymphovascular invasion, necrosis, stromal reaction, LN status and Her2/neu expression, p53 expression increased with tumor size, infiltrative tumor margins, presence of necrosis, lymphocytic stromal response and lymphovascular invasion. Thus, the present study cannot definitely comment on their probability of association due to small individual sample size; and therefore this necessitates follow-up studies with a larger sample size.

Source of Funding
None.

Conflict of Interest
None. .