2016 (txt5)
Magnetic Resonance Spectroscopy, Diffusion Weighted Imaging and Surgery in Assessment of Suspicious Breast Masses
 
 Abstract:
Background: Breast cancer is the most widely recognized type of cancer among females. Early diagnosis is one of the most essential elements influencing the disease prognosis. Several approaches have been developed for early breast cancer detection. MRI is a potential screening tool for breast cancer particularly utilizing magnetic resonance spectroscopy (MRS) and diffusion weighted imaging that reveals the pathological nature of the lesions. Surgical management of breast masses continues to be an important part of management of suspicious breast masses either in diagnosis or treatment. Objective: The aim of this work was to assess the role of diffusion-weighted imaging (DWI) and proton MRS and surgery in evaluation of breast masses and lymph node metastasis comparing the results with the histopathological findings. Patients and Methods: During the period between January 2012 and February 2014, 50 patients with suspicious breast mass were evaluated by full history, clinical examination, sonomammography (craniocaudal and mediolateral oblique) and MRI (spectroscopy and diffusion weighted imaging). Biopsy from the breast lesion; FNAC, core biopsy, frozen section or excisional biopsy were examined histopathologically. Definite surgery was done as either modified radical mastectomy or conservative breast surgery if the result of the biopsy was positive for malignancy. MRS and DWI images were analyzed and compared with the histopathological results. Results: The mean age at diagnosis of breast cancer was 46.2 years which is >10 younger than western women age at diagnosis. The frequency of diagnosis of breast cancer runs parallel with higher category of BI-RAD. MRS for choline peak detection in relation to the results of the histopathology shows sensitivity of 90%, specificity of 78.6%, accuracy of 85%, PPV of 85.7%, and NPV 84.6%. MRI/DWI showed that 90.9% of benign lesions had facilitated diffusion while 87.2% of malignant lesions had restricted diffusion. MRI/DWI showed a sensitivity of 90%, a specificity of 92.8%, an accuracy of 91.1%, a PPV of 94.7% and a NPV of 86.6%. Lymph node metastasis was found in 46.1%. In the postoperative period the commonest complication found was wound seroma in 30.7% and it was usually managed conservatively. Conclusion: MRI is a useful diagnostic modality for characterization and differentiation between benign and malignant breast lumps. Using MRS and DW-MRI with apparent diffusion coefficient (ADC) values measurement increased their accuracy and sensitivity.
Key words: Breast cancer, Breast mass, Mammography, MRI spectroscopy, Diffusion- weighted imaging, Mastectomy.
Citation: El-Hady HA, Radwan HS and Elshamy M. Magnetic Resonance Spectroscopy, Diffusion Weighted Imaging and Surgery in Assessment of Suspicious Breast Masses. AUMJ, September 1, 2016; 3 (3): 9 - 21.
Introduction
All over the globe breast cancer incidence is increasing, with a slower increase since 2000. The frequency of female breast cancer is lower in developing countries, however it is expanding quickly when compared with developed countries where rates have been steady to declining since the early 2000s. For women, cancers with the highest incidence were breast cancer (1.8 million). The top 3 causes of cancer death for women were TBL cancer (Tracheal, Bronchus, and Lung Cancer) (485000), breast cancer (464000), and colon and rectum cancer (357000). The proportion of cancerdeaths as part of all deaths has increased from 12% in 1990 to 15% in 2013(1). Increased awareness about breast cancer in the risky population leads to more frequent physical examination, breast imaging and biopsy procedures. This results in earlier detection and diagnosis of breast cancer and subsequently improved prognosis. The majority of breast lesions were found to be benign. It is imperative to identify benign lesions and recognize them from malignantlesions(2). Breast self and clinical breast exams, ultrasound and mammography can be used as diagnostic or screening tools. Mammography is currently, the best imaging technique for both clinical and screening purposes, but it is not free from complications. The harms may include unnecessary imaging, radiation exposure and biopsies in women without
cancer (3). The interpretation of mammographic images is classified according to the BI-RADS system
(Breast Imaging Reporting and Data System) developed by the American College of Radiology. The BI-RADS
scores are designed to correlate with the probability that a breast malignancy is present in a given patient
(4).
Breast MRI may be used to distinguish between benign and malignant areas, reducing the number of breast biopsies done to evaluate a suspicious breast mass . Magnetic Resonance Spectroscopy (MRS) is an MR technique that provides information on the relative concentration of biochemical components within tissue, non-invasively. Reliable biomarkers for breast cancer such as choline, phosphocholine, taurine and glycerophosphocholine can be quantitatively detected with MRS (5). Use of diffusion-weighted imaging (DWI) is another approach that may improve MR imaging and lesion characterization. DWI is an MR sequence sensitive to changes in micro-diffusion movement of water molecules in intracellular and extracellular space. It basically provides functional information based on microscopic movement of water molecules within tissues. Restricted
diffusion within the tumor appears as high signal on DWI, and as low signal on apparent diffusion coefficient (ADC) maps, and it represents high cellular density. ADC is calculated by a computer program. Studies showed that DWI is not only effective in discrimination of benign-malignant breast lesions but also indicates tumor aggressiveness (6). Surgery for breast cancer must be determined for each woman on an individual basis. The patient's desires should always be considered when choosing treatment option. For most patients, mastectomy will not impact the likelihood of survival but may affect the quality of life. Women whose breasts are preserved have a fewer episodes of depression, anxiety and insomnia(7). Treatment of the axilla in breast cancer is a subject of debate. More than 50% of patients with a metastatic SLN (sentinel lymph node) appear to
have no further LN involvement. Based on these findings and possible complications of axillary lymph node
dissection efforts are done to help the SLN examination to know the status of the axillary nodes whether it have metastatic deposits or not (8). MR imaging can help the SLN examination to know the status of
the axillary nodes whether it have metastatic deposits or not(8). MR imaging can help to differentiate metastatic from benign axillary lymph nodes(9). 

This study aimed to assess the role of diffusion-weighted imaging (DWI) and proton MRS and surgery in evaluation of suspicious breast masses and whether the axillary lymph nodes were involved by metastatic deposits or not comparing the results with the final histopathological findings. It is hoped that results would reduce future invasive procedures required for characterization of breast masses before surgery.
Patients and Methods
This study was carried out on 50 patients presented to the surgical and gynecology outpatient clinic at Al-Zahraa University Hospital and Al-Mansoura University Hospital with suspicious breast lump discovered either by clinical examination, mammography and/or breast ultrasound. This study was performed during the period from January 1, 2012 to January 1, 2014. Informed written consents were obtained from all participants enrolled voluntarily in the study after full explanation of the benefits and risks of the procedure particularly for anonymous reporting of the results. The study was approved the local ethical committees of the two hospitals.
Patients who had general contraindications for MRI examination, contrast-medium injections, refused MR examination, had noisy or non-diagnostic DWI examinations due to motion artifacts and who refused to undergo a biopsy were excluded from the study. All patients were subjected to full history taking including: personal history (patient age, occupation, marital status and menstrual history). Past history (past history of mass, previous breast surgery), history of any condition which might interfere with the MRI examination (e.g., cardiac pacemaker) were also recorded. All patients were assessed by clinical examination involving general examination and local examination of both breasts, and ipsilateral and contralateral axillary and supraclavicular lymph nodes.
Ultrasound and mammographic examination (craniocaudal and mediolateral oblique views) were done to assess and classify the mammographic appearance on mammography according to Breast Imaging-Reporting and Data System (BI-RADS). The classification is proposed by American College of Radiology is a widely accepted risk assessment tool in mammography. BI- RADS categories 4, 5 and 6 were included in this study. MRI was done using Philips Achieva 1.5T scanner with a dedicated breast coil. Patients were placed on top of specialized phased array breast coil in the prone position to allow both breasts to hang inside the loop of coil.
Starting with a three plane localizer, axial T1WI, T2WI and fat saturated T2WI or T1WI were obtained with thickness of 3 mm and 1 mm gap. The images were reviewed and evaluated for number, site, shape, margins (smooth, speculated, irregular), signal intensity in T1WI and T2WI, presence of signal internal septations, skin and nipple abnormalities, pectoralis major invasion and presence of enlarged axillary lymph nodes. In proton MRS homogeneity in the area of the field was confirmed by shimming. Also checking the effectiveness of fat and water suppression was done. PRESS sequence was performed (TR/TE 1,500/136 ms; bandwidth 1 KHz; 512 excitations); single voxel was placed upon the suspicious area. In the voxel, we wanted to study three successive selective pulses positioned in the three orthogonal intersecting planes. The acquisition period was nearly 10 minutes. The acquired data were processed including different signal filters, frequency correction, phase correction, baseline correction and curve fitting to optimize the spectral profile. Choline peaks at 3.2 ppm. According to the presence or absence of choline peak in the spectrum,MRS was analyzed, and the peak - if present – was described as tall or broad. Tall peak was interpreted as malignant and broad peak and no-peak were interpreted as benign.
DWI was performed using Philips Achieva 1.5T scanner, reduction factor, 2; 7,000/71.5; number of excitations, 2; matrix, 240 × 240; field of view, 34 cm; slice thickness, 3 mm, gap, 0; b 2 factor frown emoticon 0 and 1000 s/mm ), and the scanning time was 4 minutes. Respiratory triggering was used for better resolution. DWI is a noninvasive technique that measures the random motion of free water protons. The motion of water protons in the tissue is affected by fluid viscosity, membrane permeability, blood flow, and cellularity of the tissue. For the quantification of this motion, ADC values are used. The ADC value measures the diffusion quantitative measurement that is calculated on the basis of the attenuation in signal intensity between at least two diffusion-weighted images according to the following equation: ADC Value = -In (SDW/SSE)/b, where SDW is the attenuated spin-echo signal and is the full spin-echo signal without diffusion attenuation and b value (expressed in s/mm2) and represents the strength of diffusion weighting.
The apparent diffusion coefficient (ADC) values were automatically calculated by placing the ROI precisely within the confines of the lesion. We used fatty glandular parenchyma which shows homogeneous signal intensity on the ADC map as a reference. The provided scanner software calculates the mean value within the ROI, which equals the ADC value (multiplied by 10−3). Diffusion weighted images and ADC maps were then examined regarding signal intensity and the mean value for ADC of each breast lesion. For quantitative analysis of the data acquired from DWI, ADC maps were automatically created using software provided by the MRI system manufacturer. DWI is used to detect enlarged axillary lymph nodes and to look for finding associated with malignancy using threshold ADC value of (1.3 x 10-3 mm2/s) to distinguish metastatic from benign axillary lymph nodes
Pathological assessment of breast mass was done using fine needle aspiration cytology (FNAC), core biopsy, frozen section or resected surgically. On individual basis if the biopsy results show malignancy, definite surgery either as modified radical mastectomy (MRM) or conservative breast surgery was done. All cases undergone axillary lymph node dissection. The surgical specimen including the breast mass and the axillary nodes was examined histopathologically for presence of tumor cells. The results of the histopathology were considered the gold standard. Surgical (MRM or conservative breast surgery), neoadjuvant and adjuvant therapy were tailored per each patient condition using NCCN Guidelines Version 1.2013. Postoperative follow up to detect postoperative complications was done. Also radiotherapy and chemotherapy were additional treatment measures for the malignant lesions.
Data entry and analysis were done using the program Statistical Package for the Social Sciences (SPSS). Specificity, sensitivity, positive predictive value (PPV) and negative predictive value (NPV) were calculated by the following equations. Specificity = true negatives/(true negative + false positives). Sensitivity = true positives/(true positive + false negative). PPV = true positive/(true positive + false positive). NPV = true negatives/(true negatives +false negatives).
Results
This study included 50 female patients, their ages ranged from 20 - 70 years (46.20 ± 10.95). 43 females (86%) were multiparous and 7 were nulliparous (14%). 28 females were premenopausal (65%) and 22 were postmenopausal (45%).
Most of our patients presented complaining of breast mass except in 4 cases where breast mass was associated with bleeding per nipple in 3 of them (6% of the total) or mastalgia in one of them (2% of the total). On clinical evaluation of the breast mass, the mean size was 3.59 cm, ranging from 1 to 6 cm. The breast mass affected the right breast in 17 cases (34%), left side in 33 cases (66%). The mass was located in UOQ in 22 cases (44%), LOQ in 9 cases (18%), LIQ in 8 cases (16%), retro-areolar in 3 cases (6%) and multiple in 2 cases (4%).
The clinical assessment of the axilla revealed that the axillary lymph nodes were palpable in 29 cases (58%) and not palpable in 21 cases (42%). On mammographic assessment of the breast lesion microcalcification (Figure 2A) was seen in 22.1% of cases. According to the BI-RADS, the cases were distributed as BI-RADS 4 in 37 cases (BI-RADS 4a in 19 cases, BI-RADS 4b in 12 cases, BI- RADS 4c in 6 cases), BI-RAD 5 in 11 cases and BI-RADS 6 in 2 cases. The frequency of malignant outcomes was increased in parallel with the categories of BI-RADS. The malignancy rates according to BI-RADS category were 63.1% (12/19) for category 4a, 75% (9/12) for category 4b, 83.3% (5/6) for category 4c, and 100% (11/11) for category 5 and 100% (2/2) for category 6.
Non-contrast MRI of the breast showed that most of the benign lesions were rounded (45.5%), whereas 56.4% of malignant lesions were irregular in shape. The borders of the mass of 81.8% of benign lesions had well defined borders and 18.2% of them had ill-defined borders. On the other hand, malignant lesions borders were well-defined in 7.7% of lesions, speculated (Figure 1B, Figure 2B and D, and, Figure 3B and C)
in 53.8% of lesions and ill-defined in 38.5% of lesions.
All of the 50 lesions were iso- to hypo- intense in T1WI (Figure 1A and B, Figure 2B and C, and, Figure 3A and B) while in T2WI, 66.7% of the benign lesions were iso- to hypo-intense and 33.3% were hyperintense. 84.6% of malignant lesions were iso- to hypointense and 15.4% were hyperintense in T2WI. Low T2 internal septations were detected in 2 (22.2%) of benign lesions and one (2.6%) of malignant lesions.
On MRS examination of the breast lesions, 24% with no choline peak, 14% had broad peak, while 62% had tall peak. Regarding the benign lesions, 81.9% showed no or broad choline peak, whereas, tall choline peak was detected in 18.1% (Figure 1D) of lesions. On the other hand, 89.7% of malignant lesions (Figure 2E) showed tall choline peak and 10.3% had no or board choline peak (Figure 3D and Table 1).
Table 1. Choline peak of Magnetic Resonance Spectroscopy in the study group of breast mass patients. Data shown are frequencies; n (%).
      Statistical analysis of the MRS results in correlation to the biopsy results showed that the sensitivity of MRS was 90%, its specificity was 78.6%, accuracy was 85%, PPV was 85.7% and NPV was 84.6% as (Table 2).
The assessed breast lesions using DWI were classified as restricted or facilitated. 90.9% of benign lesions showed facilitated diffusion and 9.1% of them showed restricted diffusion (Figure 1F and E), while 87.2% of malignant lesions showed restricted diffusion (Figure 2F and Figure 3F) and 12.8% of them showed facilitated diffusion.  
Regarding the ADC values of benign lesions, the minimum ADC value was 0.9 (Figure 1G), maximum ADC value was 2.12 and their mean was 1.54. While the minimum ADC value of the malignant lesions was 0.56, their maximum value was 1.4 and their mean was 0.86 (Figure 2G, Figure 3G and Table 3).
  Statistical analysis of the DWI results in correlation to the final histopathological biopsy results showed that the sensitivity of diffusion was 90%, its specificity was 92.8%, accuracy was 91.1%, PPV was 94.7% and NPV was 86.6% (Table 4). MRI for axillary lymph nodes was found to show positive finding for lymph node involvement in 32 cases; of those 24 cases (75%) were found positive after histopathological examination, while 8 cases (25%) were found to be negative. On the other hand, as for MRI of the axillary lymph nodes, there were negative findings in 18 cases of these cases; 15 cases (83.3%) cases proved to be negative and 3 cases (16.7%) were positive after histopathological examination. These results revealed a sensitivity of 88.89%, specificity of 65.22%, PPV of 75.0%, and NPV of 83.33%.
After imaging, the histopathological results in relation to the biopsy procedure type included: 1) FNAC in 24 cases (48%) among which 18 cases were positive for malignancy, 2) Core needle biopsy was done for 20 cases among which 16 cases were positive for malignancy, and 3) Frozen section was done for 6 cases among which 5 cases were positive for malignancy. The histopathologically malignant 39 cases (78%) included 34 cases (87.2%) with IDC, ILC in 3 cases (7.7%), and 2 cases were medullary (5.1%)]. 11 cases were benign.

According to American Joint Committee on Cancer (AJCC), 7th Edition, 2010, staging of malignant cases was; stage I: 6 cases, stage II: 29 cases, and stage III: 4 cases. Modified radical mastectomy was done in 26 cases (66.7%), wide local excision in 13 cases (33.3%). Axillary lymph node dissection (ALND) was done for all cases; in 16 cases (46.1%), positive nodal deposits were confirmed. In the postoperative period following MRM or CBS, 12 patients (30.7%) developed wounds seroma and it was treated conservatively in 9 cases and by repeated aspiration in 3 cases. Only one case was complicated by wound infection and treated by antibiotics and dressing. No cases were complicated by nerve injury or arm lymphedema.
Discussion
Breast cancer is the most frequently diagnosed cancer and the leading cause of cancer death among females worldwide, with an estimated 1.7 million cases. Breast cancer alone accounts for 25% of all cancer cases and 15% of all cancer deaths among females. Developed countries account for about one-half of all
breast cancer cases and 38% of its death(10). Clinicians and researchers continue to have a major concern about early detection, diagnosis and treatment of breast cancer. Increased awareness in the affected population leads to more frequent procedures(11). Breast cancer is strongly linked to the women’s age. The frequency increments steeply correlated with age with the marked increase in rate found in post-menopausal women. In USA, breast cancer in young women is relatively rare. However, breast cancer in adult women less than 60 years of age in high-income countries is the leading cause of cancer death(12).
In this study, the median age at diagnosis was 46 years which is younger than the median age at diagnosis in Western countries. Also, most of the patients included in this study were premenopausal (65%) and multiparous (86%). These results are consistent with the results of Laraqui et al. (2015)(13) showing a median age at diagnosis in the Middle East region as 52 years, compared to 63 years for Western countries. The same results were shown by Boder et al. (2011)(14) and Missaoui et al. (2011 and 2012)(15-16). 
pre-menopausal women and characterized by unfavorable characteristics such as high histological grade and stage, large tumor size and frequent lymph node metastases(14). Moreover, comparing age at diagnosis of breast cancer in Gharbiah Cancer Registry (GCR), Egypt, to USA Surveillance, Epidemiology, and End Results (SEER) Program, showed significant differences in age, tumor grade, hormone receptor status, histology, and stage among cases included in GCR and those of SEER registry . The average age in the Egyptian GCR cases were over 10 years younger than USA SEER cases, with nearly 19% of GCR cases ≤40 years of age as compared to only 6% of USA SEER cases.

In our study, the mean breast mass size was 3.59 cm, ranging from 1 to 6 cm. The breast mass affected the right breast in 17 cases (34%), left side in 33 cases (66%). The mass was located in UOQ in 22 cases
(44%), LOQ in 9 cases (18%), LIQ in 8 cases (16%), retro-areolar in 3 cases (6 %) and multiple in 2 cases (4%). Lim et al. (2016) reported similar location within the breast where the central group constituted 14.0%, the UOQ was 57.0 %, the UIQ was 15.1%, the LOQ was 9.9% and the LIQ was 4.0% of their cases(17). Tumor location within the breast was reported to affect the survival outcomes in patients with LIQ tumor location. Other researchers reported that patients with LIQ tumor location had a significantly increased breast cancer-specific mortality(18,19). Theoretically, these results  were caused by the anatomical
accessibility of the tumor to the internal mammary lymph node than in tumors with other locations (20).
On the other hand, the histopathologic diagnosis of malignant lesions in this study included IDC (87.2 %), ILC 3 (7.7%) and medullary carcinoma (5.1%). Our results coincide with those of Pereira et al. (2009)(21) as invasive ductal carcinoma was the commonest histopathologic type comprising about 73%. In this study the shape of benign lesions were regular (in 91% of benign cases) with well-defined borders in 81.8% of them. The malignant lesions were irregular in shape in 56.4% of cases and their borders were speculated or irregular
in 53.8% and 38.5% of cases, respectively. The margins of breast mass were speculated in 100% and the shape
was irregular in 97% in a study done by Tozaki et al (2006)(22) which proved that the shape and margin of malignant lesion is usually one of the highest positive predictive factors associated with malignancy. The results were similar to Vassiou et al. (2009)(23). The incidence of malignancy increases with higher BI-
RAD grade as confirmed by other investigators(24,25).

Iso- to hypointense signal at T1WI were reported in all breast lesions, while in T2WI 66.7% of benign lesions were iso- to hypo-intense and 33.3% of them were hyperintense. 84.6% of malignant lesions were iso- to hypo-intense and 15.4% of them were hyperintense in T2WI cases. Low T2 internal septations were detected in 2 (22.2%) benign lesions and none of the malignant lesions. This is in agreement with Liberman et al. (2002)(26) who stated that T2 signal intensity was not a significant predictor of malignancy and all masses in their study with internal septa were fibroadenomas. However, Al- Khawari et al. (2009)(27) stated that internal septations, a description usually associated with fibroadenomas, is a sign that is no longer exclusive to benign lesions. In their study which included 55 lesions, internal septa were found in a lesion of well-differentiated invasive ductal carcinoma.
In our study, 81.9% of benign lesions showed no or broad choline peak whereas tall choline peak was detected in 18.1% of lesions. On the other hand, 89.7% of malignant lesions showed tall choline peak and 10.3% had no or board choline peak. Sensitivity of MRS was 90.8%, its specificity was 78.6%, accuracy was 85%, PPV was 85.7% and NPV was 84.6%. Our results were in agreement with Katz-Brull et al. (2002)(28), who reported that the overall combined sensitivity and specificity of MRS were 83% and 85%, respectively. Also, Razek et al. (2012)(29) reported that the sensitivity and specificity were 93.6% (88/94) and 77.9% (152/195), respectively. Moreover, Jacobs et al. (2004)(30) showed that the sensitivity and specificity were 87% and 85%, respectively. However, Bartella et al. (2006)(31) reported sensitivity of 70 - 100%, specificity of 88%. The sensitivity is size-dependent and increases from 72% to 90 - 100% with increase in the size of the examined lesions. This feature was supported by Katz-Brull et al. (2002)(28) and Tozaki and Fukuma (2009)(32) who included small lesions in their diagnostic study. They reported a diagnostic sensitivity of 82% in lesions larger than 15 mm in maximum length, but only 42% when considering all lesions.
In our study, facilitated diffusion was found in 90.9% of benign lesions, while 9.1% of them showed restricted diffusion. For malignant lesions, restricted diffusion was noticed in 87.2% and facilitated diffusion was noticed in 12.8%. Sensitivity of diffusion was 90%, its specificity was 92.8%, accuracy was 91.1%, PPV was 94.7% and NPV was 86.6%. Our study agrees with that of Pereira et al. (2009)(21) who reported sensitivity and specificity as high as 92% and 96%, respectively. Also, Palle and Reddy (2009)(33) reported values for the detection of malignant lesions that showed a sensitivity of 97.22% and a specificity of 100%. However, Tavassoéli and Devliee (2003)(34) report yielded a specificity of 67% (43/64 cases) and sensitivity of 97% (61/63 case). In this study, DWI could detect axillary lymph node metastasis with a sensitivity of 88%.
On the other hand it was 94.7% by Fornasa et al. (2012) , 95% in study done by He et al. (2012)(36) and the median sensitivity was 84.2% in a systematic review study (37).  This study shows 46.1% positive axillary nodal deposits after ALND which is slightly higher if compared with studies done by Dees et al. (1997)(38) and Chua et al. (2001)(39) which was 33% and 41, respectively. The commonest complication after breast cancer surgery was wound seroma. The exact etiology of seroma formation remains controversial. In this study, 30.7% patients developed wound seroma which is similar to previous report (40).
 
Conclusion 
Most of our patients developed breast cancer at younger age confirming the reported local trend. This fact must be kept in mind during screening, diagnosis and treatment. MRI could be used for characterization and differentiation between benign and malignant breast lumps. Addition of DWI and MRS can differentiate between benign and malignant breast lesions with high sensitivity and specificity. DWI and MRS are problem solving in some cases because it may be very useful avoiding unnecessary interventions as biopsy. On the other hand, it is expensive and time consuming. However, more studies with larger populations are needed for more evaluation of DWI in characterizing breast lesions.
Limitations of the Study
􏰁 Mammography in dense breasts may show false-negative results.
􏰁 MRI is costly and time-consuming compared to mammography.
􏰁 During MRI, patients must tolerate claustrophobia.
􏰁 There are certain limitations of FNAC which require final histopathological diagnosis.