Correlation between Robinson ’ s cytological grading with Modiied Bloom-Richardson histopathological grading for Breast carcinoma

E-mail: sanatchalise@hotmail.com, sanat_chalise@yahoo.com Background: Invasive breast carcinoma is the most common carcinoma in women. Fine needle aspiration cytology is the important part of triple assessment of the palpable breast lump. Robinson’s cytological grading can be done in ine needle cytology smears of breast lump. The aim of this study is to grade breast tumor on cytology according to Robinson cytological grading and to correlate with histopathological

treatment and prognosis of breast cancer in Nepal. 3ne needle aspiration cytology (FNAC) is an important of the triple assessment of palpable breast lump.It has been shown than the FNAC can reduce the number of open biopsies. 4The standard prognostic factors, recognized by the National Cancer Institute in 1990, include lymph node status, tumor size, nuclear grade, steroid receptor content, tumor type, and cellular proliferation rate and recommend that for patients who undergo preoperative chemotherapy or radiotherapy, breast ine-needle aspirates      can be used toprovide prognostic information. 5Cytological grading of breast carcinoma, which is a simple, feasible and reproducible method, can be used for selection of neoadjuvant therapy and would allow the assessment of the tumors without any surgical intervention so that the morbidity associated with overtreatment of low-grade can avoided. 6The concept of nuclear grading in breast carcinoma was introduced by Black et al 7 in 1955.

It was applied in cytological smears after modiication by
Fisher et al 8,9 or after simpliication by Cajuliset al. 10 Histopathological grading of breast carcinoma using modiied Bloom-Richardson grading system is a widely accepted tumor grading system and has been found to have good prognostic correlation.Based on the cytological features, various grading systems have evolved.Of the different cytological grading methods corresponding to modiied Bloom-Richardson grading, the Robinson's cytological grading was found to be useful in grading breast carcinoma in FNAC. 11e purpose of the present study was to compare cytological grading using Robinson's method with histological grading using modiied Bloom-Richardson grading method.

MATERIALS AND METHODS
This was a prospective study done in department of pathology, Bhaktapur cancer hospital, Nepal from 1 st April 2014 to 30 th July 2015.58 cases with cytologically conirmed breast carcinoma were included in this study and correlated with histopathological indings.FNAC of breast lump was done with 22 gauge needle which were attached to 10ml syringes.Smears made were stained with Giemsa as well as Hematoxylin and Eosin stain (H&E).The cytological grading was done by Robinson's method in which the cell dissociation, nuclear size, cell uniformity, nucleoli, nuclear margins and the chromatin patterns were studied (Table 1).The lumpectomy and mastectomy specimens of the corresponding cases were ixed in 10% formalin, routinely processed and stained with H&E stain.Histolopathological grading was done according to modiied Bloom-Richardsongrading method in which tubule formation, nuclear pleomorphism and mitotic count were evaluated (Table 2).

Correlation between Robinsons cytological and modiied
Bloom-Richardson histopathological grading was done.Data was analysed using the statistical package for social science (SPSS) and inal correlation was established using Spearman's correlation coeficient.

RESULTS
A total number of 58 cases were included in this study, aged between 27 years and 80 years.Mean age was 46.12 years.Cytological scores were analyzed using cytological grade and histological grade and the details are shown in Table 3, In this study, using Robinson's cytological grading, maximum number of carcinoma were grade II (62.1%), followed by grade I (19%) and grade III (19%) as shown in Table 4. Using histopathological grading, maximum numbers of cases were grade II (63.8%), followed by Grade III (22.4%) and Grade I (13.8%) as shown in Table 5.
The absolute concordance rate was 65.9% (Table 6).The highest concordance rate of 80% was found in Grade II tumor.

DISCUSSION
FNAC may be performed as the irst-line investigation, especially in symptomatic and screening populations.The main purpose of FNAC of breast lumps is to conirm cancer preoperatively and to avoid unnecessary surgery. 12Among the various cytological grading systems, Robinson's system is simple, more objective, takes little time and effort, is reproducible and correlates precisely with modiied Bloom-Richardson histological grade. 6tological grade and histological grade was evaluated according to the Robinson's cytological features.In the current study, the majority of the cases were Grade II (62.1%) followed by Grade I (19%) Grade (19%) cytologically.Histologically, commonest tumor was Grade II (63.8%) followed by Grade III (22.4%) and Grade I (13.8%).Sood N et al5, Gore CR et al 13  The highest level of disconcordance was found in I tumor study which is similar to the study done by Phukanet al 11 however sood N et al 5 found highest level of disconcordance in Grade III tumor.The cause of disconcordance may be because of subjective variation during assessment of cytological features that are not the features of histological grading.Dabbas DJ et al 16 and Howell LP et al 17 found dificult to assess tubule formation and mitotic index on cytology and conclude that the thedisconcordance may be due to the disparities between cytological and histological features.
In our study, using univariate analysis, all of the cytological features show strong correlation with cytological grade (p<0.005)however the multiple regression analysis show cell dissociation and nuclear margin as a most important inluencable factor.All histological features shown strong correlation and inluence in histological grading.Das et al 15 also found strong correlation between cytological features and cytological grade with cell dissociation, nuclear margin, nucleoli and uniformity as a most inluencable factor.Similarly study done by Lingegowdaet al 18 found strong correlation between cytological features and cytological grade with cell dissociation, nucleoli and nuclear margin Chalise S et al. as a most inluencable factor.Rajan J and Pai KP 14 in their study also found signiicant statistical correlation between cytological features and cytological grade however in contrary to our study they found cell size, uniformity and nucleoli as a most inluencable factor.
Cell dissociation is found to be an important cytological feature as several studies showed an association between increase in cell dissociation and incidence of lymphnode metastasis as cell cohesion status and expression of E-Cadherin are indicated by cell dissociation status thus helping in preoperative assessment of the patient. 19,20is study showed statistically signiicant correlation (p<0.005) between cytological and histological grade which was similar to the study done by Rajan J et al 14 , lingegowda et al 18 and Frias et al 21 .Table 12 showed the cytological and histological correlation between various studies in relation to our studies.All the cytological features are statistically signiicant using Univariate analysis (Table 7).Cell dissociation and nuclear margin were found to be the most inluencable factors using Multiple regression analysis (Table 8).Similarly all the histological features are statistically signiicant using Univariate analysis and all the histological features have equal inluence in histopathological grading (Table 9 and   10)

CONCLUSION
Robinson's cytological grading is a simple, feasible and easily reproducible method of assessing breast carcinoma preoperatively.This cytological grading method can be done preoperatively to determine the aggressiveness of the tumor and to plan for the neoadjuvant chemotherapy.

Table 8 : Multiple regression analysis of cytological grade
11d Phukan JP et al11also found grade II as a commonest tumor on cytological grading and histological grading, whereas Rajan J and Pai KP14in their study found grade I as a commonest tumor.Present study showed, absolute concordance rate of 65.9% which is comparable with the study done by Sood N et al5, Phuken et al11and Das et al15where they found absolute concordance rate of 68.97%, 72.2 % and 69% respectively.The highest concordance rate of 80% was found in Grade II tumor in our study which was similar to the study done by Phukenet al11where they found 83% concordance rate in grade II and grade III tumor.But in contrast to this study,