Introduction
Breast cancer is the commonest cancer among women in urban registries of Delhi, Mumbai, Ahmedabad, Kolkata and Trivandrum according to ICMR-PBCR data.1 Carcinoma breast is associated with various changes in organ and at the tissue level. Microscopic changes such as histological grade, presence of inflammatory infiltrate, lymphovascular invasion, adipocytic invasion, type of borders and margins, presence or absence of necrosis, degree of nuclear pleomorphism, nuclear and cellular atypia, hormone receptor status and various other well-established factors have been known to affect the prognosis of patient.2, 3 In our present study we attempted to establish a correlation between relatively less established parameter i, e stromal fibrosis and other well known prognostic indicators of carcinoma breast. In our study, we studied the topographical distribution, histological characterization of fibrosis and correlated it with various other histopathological parameters.
Materials and Methods
The study was a cross sectional descriptive study conducted in the Departments of Pathology and Surgery, Maulana Azad Medical College and associated Lok Nayak Hospital, New Delhi-110002 over a period of 1.5 years from October 2012 to April 2014.
50 patients of carcinoma breast presenting to our surgery department who had not undergone any preoperative chemotherapy were included in the study. All subjects were selected from the patients visiting the surgery OPD, LokNayak Hospital after taking their consent
Gross examination was done with special reference to tumor size, quadrant of breast involved, involvement of nipple areola, skin, deep resected plane and resected margin (Figure 1). Sections were routinely processed in histokinette, paraffin embedded and 4µ sections were stained with H & E. Immunohistochemical staining was done for ER, PR and HER2Neu and the percentage of immunopositive cells was determined under light microscopy at 40X.
Expression of ER and PR was scored4 between 0 and 2 as follows: 0: less than 5% of nuclei staining; 1: 5%-19% of nuclei staining; 2: >20% of nuclei staining and interpreted as positive when more than 10% of tumor cells showed positive nuclear staining. Expression of HER2/neu was from scored 0 to 3. It was considered positive when a weak to strong complete membranous staining in at least 10% of tumor cells was seen4 (score 2 and 3).
Fibrotic focus is defined in literature as a scar-like area replacing necrosis in the center of a carcinoma.5 It appears as a radially expanding fibrosclerotic core and consists of loose, dense or hyalinized collagen bundles and a variable number of fibroblasts. FF (fibrotic focus) smaller than 3mm in diameter do not contain carcinoma cells, while larger FF sometimes do.5, 6, 7 For absolute assessment of the size of fibrotic focus, its largest diameter was measured on H& E slides and special stains under microscope and distribution of fibrosis was recorded. Grading of fibrosis was don-e+-+ as follows.8
No fibrosis
Grade 1-large number of fibroblasts with very few collagen fibres
Grade 2- admixed fibroblasts and collagen in various ratios
Grade 3-mainly collagen fibres, mostly hyalinized.
Fibrosis was correlated with clincopathological profile in these cases.
Statistical analysis
Association between the presence of fibrosis with other clinicopathological parameters and any statistical significant association among various histopathological characterstics were examined by Chi-square test or Fischer’s exact probability test. A p-value <0.05 was taken significant.
Observations
Of these 50 cases, 48 were of infiltrating duct carcinoma and 2 cases were of mucinous carcinoma. The patient’s age ranged from 30 to 80 years. Mean age of the patients was 52 years with majority being in the age group of 41-50 years. The most common presenting complaint was a lump, seen in 48 patients, followed by pain (30%) and nipple discharge (6%). Other symptoms like ulceration of skin, weight loss, and fatigue were also seen. Besides a palpable lump, other important clinical findings were palpable axillary lymph nodes (24), nipple/skin involvement (10) and nipple discharge (3). Among 50 cases, various risk factors which were present were history of oral contraceptive pill intake in 5 cases(10%), history of no breast feeding in 4 cases(8%), family history of breast cancer in 2(4%) cases and history of concurrent malignancy in 1 case which was well differentiated type of endometrial carcinoma in a patient aged 75 years. 52% patients had left side while 48% had right sided tumor. Most common quadrant involved in carcinoma breast was upper outer quadrant (46%) followed by central quadrant (20%). The cases were diagnosed by aspiration cytology, mammography or trucut biopsy before surgery Mammography was performed in 10 (20%) cases. Majority of the cases belonged to BIRADS 4/5. Besides mammography, the other common modality included FNAC (76%) and trucut biopsy (44%). For 1 case, diagnosis was made on excision biopsy. Modified radical mastectomy was performed in 49(98%) of the cases while lumpectomy was done in 1 case. Majority of the cases were of clinical stage T2 N0 M0 (48%) followed by T2N1M0 (18%).
The tumor size varied from 0-10 cm, with the mean of 4 cm. 8 cases (16%) had a tumor size ≤ 2cm while 5 cases (10%) were >5 cm in size. Majority 37 (74%) of tumors were in the size range of 2 to 5cm. There was a strong and significant association between lymph node involvement and stage of tumor (p value =0.001).
Breast cancer staging
Staging was done according to the AJCC guidelines9. Majority of the cases belonged to stage II (72%), followed by stage III (26% cases). The cases in stage III increased from 24% to 26% with a decline in stage II cases from 74% in clinical stage to 72% in pathological staging. This is explained by the fact that not all metastatic lymph nodes are clinically palpable
Table 1
Presence of various histopathological features
The intensity of pattern of inflammatory infiltrate was graded as - absent (0), mild (1), moderate (2) marked (3) and with germinal centre (4). The intensity of inflammatory infiltrate was further correlated with necrosis and grade of the tumor. On evaluating the inflammatory infiltrate with clinicopathological parameters, association was found with tumors showing adipocytic invasion (p value =0.006). Also higher number of cases with necrosis and lymphovascular invasion exhibited inflammatory component though the difference was not significant statistically (Figure 2).
ER/PR/HER2Neu status
Out of 50 patients,25 were positive for ER(50%)and 23 cases were positive for PR expression(46%), 31 cases showed ER/PR positivity, 17 cases were both ER and PR positive and 19 cases were both ER and PR negative. Out of 50 cases, HER2Neu was positive in 17 cases (34%). Out of total 50 cases, 12 were triple negative (Figure 3, Figure 4).
Fibrosis
All the specimens were assessed thoroughly for presence of fibrosis and sections were taken from the centre of tumor and special stains Masson’s trichrome and Von Gieson Elastica were performed in addition to routine Hematoxylin Eosin sections. A focus of >1 mm in size was called as a fibrotic focus (FF).5, 6, 7, 8 Cases were screened for the number of fibrotic foci and they were graded from 0 to3. In cases where there were more than one fibrotic foci, the highest grade of FF was taken up for statistical analysis.
Out of 50 cases, 30 cases showed presence of fibrotic foci. Out of these 30 cases, 7 had single foci, 12 had 2 foci and 11 cases showed >2 foci. Out of 30 cases, 7 had single foci, 12 had 2 foci and 11 cases showed >2 foci. Fibrotic foci were measured under microscope in mm. Results showed most of the foci were between 1-2 mm.4 cases had fibrosis but less than 1 mm so were not counted as fibrotic foci, 9 cases showed a fibrotic focus >8 mm in diameter (Figure 5, Figure 6, Figure 7, Figure 8).
Figure 6
Fibrotic focus grade 2 with loose oedematous areas and scattered lymphocytes. (Masson's trichrome; 100x)
Correlation between size of tumor size and size of fibrotic focus was not significant statistically.
Table 2
Correlation of fibrosis with clinicopathological parameters
Only statically significant correlation was obtained between presence of fibrotic foci and presence of lymphovascular invasion (p value < 0.05).
Discussion
Tumor hypoxia may result in desmoplastic response in the form of dense or sparse, homogenous or focally hetrogenous fibrosis of the breast tissue. 2 This desmoplstic response mainly consists of collagen fibres in varying amonts and arrangements. It is termed as fibrotic focus when it is more than 1 mm in diameter. Peritumoral and stromal fibrosis has been assessed as an marker of intratumoral hypoxia and is being studied as histological surrogate prognostic parameter in breast cancer. 9, 10, 11, 12, 13, 14, 15, 16
The study of stroma adjacent to tumor cells comprising of fibroblasts, collagen fibres,elastic fibres, and inflammatory cells may help in understanding the difference in prognosis of similar grade and stage patients. 17
34out of 50 patients in our study showed presence of intratumoral fibrosis. Out of those 34, 30 had a presence of fibrotic focus defined as an area more than 1 mm in largest diameter.so fibrosis was present in 68 % cases and fibrotic focus was demonstrable in 60% of total cases which is in accordance with Hasebe et al(1996) who demonstrated fibrotic focus in 80 out of 152 patients(52.63%). 18
Tumor cells and infiltrating macrophages around the tumor cells initiate the release of profibrotic and proangiogenic growth factors. VEGF has been proposed to play a central role. 19 We obtained a statistically significant correlation of presence of fibrosis with presence of lymphovascular invasion (p value =0.043), while we could not obtain any significant difference with age, menopause, tumor size, tumor grade, pathological stage ,lymph node involvement, presence of necrosis, adipocytic invasion, duration of tumor, laterality, presence of inflammatory infiltrate and with ER, PR, HER2Neu status. Contrary to our results, Hasebe et al(1996) reported a statistical significant association of presence of FF with increasing grades of tumor (p value< 0.05) and axillary lymph node involvement( p value <0.05). He also failed to obtain any significant correlation of fibrosis with age,menopausal status and hormonal receptor status contrary. 18 In his further study, Hasebe et al (2002) stated the parameters significantly associated with the presence of FF were an invasive tumor size of < 20 mm, histological Grade 3, the presence of lymphatic invasion, the presence of vascular invasion, the presence of adipose tissue invasion ,the presence of skin invasion , and the presence of nodal metastasis. 20 Among these parameters, those that still significantly increased the relative risks of the presence of FF were the presence of adipose tissue invasion, invasive tumor size, the presence of vascular invasion, and the presence of nodal metastasis. The significant associations of the presence of FF with lymphovascular invasion probably suggest that IDCs with FF have a greater aggressive invasive potential than do IDCs without FF. The significant associations of the presence of FF with lymphovascular invasion most likely reflect IDCs with FF having a higher metastatic potential than IDCs without FF.
Out of 30 cases showing fibrotic focus, 19 cases showed grade 3 fibrosis, 11showed grade 2 fibrosis. Hasebe et al 2001 demonstrated a significant correlation between grade of fibrosis and lymph node metastasis with a p value < 0.05, whereas in our case p value was 0.346(>0.05) and not significant. 21
Hasebe20 et al, 2002 stated in IDCs with FF, FF dimension of >8 mm significantly increased the risk of initial distant organ metastasis. His prospective study clearly demonstrated FF dimension to be a very useful prognostic parameter for patients with IDCs with FF
Conclusions and Recommendations
This is a limited study in a small number of patients focussing on various stromal changes in breast cancer. This study has further characterized the heterogeneous nature of desmoplasia in breast carcinoma using special staining which may facilitate the physician in making patient diagnosis, prognostication and outcome. The results of our study add to the growing literature that attempts to classify and categorize various stromal changes and tries to find out their prognostic significance. The study of stroma adjacent to tumor consisting of fibroblasts, collagen bundles, elastic fibres, microvessels and inflammatory cells, in addition to study of tumor cells may help in prognostication of breast cancer patients.
The presence of fibrotic focus as a marker of intratumoral hypoxia and as a surrogate of angiogenesis is easily identifiable pathological feature, which can be easily added to pathology report. Its independent prognostic value merits confirmation by future trials.
Article DOI
