148 Fine-Needle Aspiration and Cytologic Findings of Surgical Scar Lesions in Women With Breast Cancer Ehud Malberger, DMD, FIAC,* Yeouda Edoute, MD, PhD,t Osnaf Toledano, MD,* and Dov Sapir, MDS Benign and/or malignant lesions may occur in surgical scars after mastectomy or lumpectomy (SML) in patients with breast cancer (BC). Early diagnosis of these lesions is essential for both therapeutic and prognostic evaluation. The diagnostic value of fine-needle aspiration (FNA) was determined for these scar lesions. The findings of cytologic and histologic specimens obtained from the same lesion of SML in 83 women with BC were correlated. Twenty-five FNA yielded only acellular specimens. Of the FNA done by the cytopathologist, only 6.2% were not representative. However, 45% of those done by less experienced clinicians were not representative. Representative FNA were obtained from 58 of the women who took part in the study. Based on the histologic diagnosis, 38 patients had malignant scar lesions (MSL), and 20 had benign scar lesions (BSL). In one patient of the 38 with MSL, cytologic examination did not show the malignant lesion; in four women, the tumor was suspected cytologically; and in the remaining 33, the cytologic findings were consistent with malignancy. In 18 of the 20 patients with BSL, cytologic findings were reported as benign and in the other two, as inconclusive. The sensitivity, specificity, and positive and negative predictive values for the cytologic findings were 97.4%, loo%, loo%, and 94.7y0, respectively. The diagnostic accuracy of FNA cytology was 98.2%. No complications followed the procedure. It was concluded that FNA cytologic examination of lesions in SML is a simple, safe, highly accurate, and cost-effective method to distinguish malignant from benign lesions in women with BC. Lesions in SML should be explored routinely by FNA, rather than by the traditional From the Departments of *Diagnostic Cytology, tlnternal Medicine c, and $Oncology, Rambam Medical Center and the Faculty of Medicine, Technion, Haifa, Israel. Address for reprints: Ehud Malberger, DMD, FIAC, Department of Diagnostic Cytology, Rambam Medical Center and the Faculty of Medicine, Technion, Haifa, Israel. Accepted for publication April 1, 1991. biopsy, provided the FNA is done by an experienced operator. Cancer 1992; 69:148-152. Breast cancer (BC) is the most common malignancy in women in the western world. Local and/or regional recurrence of BC is expected in 10% to 40% of patients initially treated by radical or modified radical mastectomy, the higher incidence rate being associated with locally advanced primary lesions and extensive axillary disease.'-3 Recurrence of BC in the surgical scar after mastectomy or lumpectomy (SML) is important in therapeutic decisions and progn~sis.~ Differentiating between recurrent malignant scar lesion (MSL) and benign scar lesions (BSL) traditionally was done by surgical biopsy and histologic examination of the suspected lesion. Many such patients are, however, reluctant to undergo an additional surgical intervention. In cases where adjuvant therapy was administered, the surgeon also may be reluctant to do a biopsy for fear of impaired healing. Aspiration cytology is a method of choice for such lesions, particularly in apprehensive patients or in those who have received adjuvant therapy. We evaluated the diagnostic accuracy of aspiration cytology in SML lesions of women with BC. Patients and Methods During the years 1976 through 1986 at the Rambam Medical Center, we obtained both fine-needle aspiration (FNA) and surgical biopsy specimens from the same SML lesions, which occurred after surgical removal of BC in 83 women. This procedure was adopted either to confirm or exclude suspected recurrent malignancy. The FNA of the SML lesion was done with a 23-gauge needle. The aspirate was expelled onto glass
FNA of Scar Lesions in Breast CA/Malberger et al. 149 logic diagnosis we considered nonmalignant, inflammatory, and atypical-reactive aspirates as negative cytologic diagnosis; suspected and definite malignant findings was termed positive cytologic diagnosis. Whenever cytologic findings were such that a malignant lesion could not be excluded, it was regarded as inconclusive. The findings of FNA were categorized as true-positive (TI'), true-negative (TN), false-positive (FP), and false-negative (FN) results. The accuracy of true and false cytologic diagnoses was verified by the histologic diagnosis. The cytologic diagnosis was defined as TI' if a histologic diagnosis of malignancy based on the biopsy of the same actual SML lesion was made. The cytologic diagnosis was defined as TN if there were negative cytologic findings and a benign histologic diagnosis of the same lesion. Patients with acellular material were considered not representative and were therefore excluded from the study, as were patients with inconclusive cytologic findings, who qualified as not diagnostic. Results Figure 1. Malignant cells in aspirates from scar lesions after mastectomy. (Top) Typical dissociated cells and fragments of invasive duct carcinoma presenting minimal cellular atypia (Papanicolaou stain, original magnification X 100). (Bottom) Malignant dissociated cells of breast carcinoma with marked cellular atypia (Papanicolaou stain, original magnification X200). slides, smeared, and fixed in 95% ethanol for staining by Papanicolaou's method. The cytologic findings were reported as: 1, acellular; 2, not malignant; 3, inflammatory; 4, atypical-reactive; 5, inconclusive (malignant lesion cannot be excluded); 6, suspected malignant; and 7, malignant (Figs. 1-4). After FNA, an excisional biopsy of the same lesion was done, and the specimen underwent histologic examination. The diagnostic evaluation of the histologic specimens was made by an experienced pathologist. For correlation, we described the pathologic diagnosis as malignant or benign. All relevant data for each patient were gathered, and followup information was recorded. Data came from reviewing the Rambam Hospital's charts; the cytologic, pathologic, and oncologic findings; the discharge summaries of other hospitals; and the Israeli Cancer Registry. For calculating the sensitivity and specificity of the cyto- The patients' characteristics are shown in Table 1. There were 38 women with MSL and 20 with BSL. The median age of the 58 patients at the time of FNA was 55 years (range, 32 to 87 years). Mastectomy had been done in 54 of the women and lumpectomy, in four. The histologic types of the surgically removed BC were: invasive ductal carcinoma in 43 patients (74.1%), invasive lobular carcinoma in five patients (8.6%), anaplastic carcinoma in three patients (5.2'/0), scirrhous carcinoma in two patients (3.4%), and not otherwise specified in five patients (8.6%). Radiation to the breast area was administered to 24 women. At the time of aspiration, 25 patients had histologic or cytologic evidence of extrascar metastases or radiologic and/or radioisotopic findings suggestive of extrascar metastases. The median interval between BC surgery and FNA of the scar lesion for women with MSL was 33 months (range, 4 to 303 months); for women with BSL, it was 17 months (range, 3 to 70 months). The scar lesions were described as "a nodule" in 45 patients, a thickening or a change in the color of the scar in 16 patients, and progressive growth of the scar in 13 women; some lesions were characterized by more than one feature. Table 2 summarizes the cytologic findings. In 25 of the cases, the aspirates were acellular (Table 3). Only two of the FNA done by the cytopathologist were acellular (6.2% of his series of FNA). There were 23 FNA (45%) done by less experienced clinicians that were acellular and therefore not representative.
150 CANCER January 1,1992, Volume 69, No. 1 Figure 2. A few enlarged, atypical nuclei, suspicious for recurrence of breast carcinoma. In the background, fragments of fatty connective tissue. (Papanicolaou stain, original magnification [Left] XlOO [Right] X200). Of 38 patients with histologic evidence of MSL, cytologic examination of FNA did not diagnose the malignant lesion in one patient and suspected malignant lesions in four patients. It revealed definite malignancy in 33 patients. Among 20 patients with BSL, the cytologic findings were reported a5 benign in 18 patients and inconclusive in two patients. The sensitivity, specificity, and positive and negative predictive values for cytologic results were 97.4'10, loo%, loo%, and 94.7%, respectively. The diagnostic accuracy of FNA cytology for MSL lesions was 98.2%. No complications followed the procedure. Discussion Local and/or regional recurrence of BC is expected in 10% to 40% of patients initially treated by radical or modified radical mastectomy, the higher incidence rate being associated with locally advanced primary lesions and extensive axillary disease.'-3 Some clinicians believe that tumor recurrence in the SML is secondary to residual cancer cells spilled at the time of surgery and entrapped in the fibrotic tissues of the scar itself, therefore presenting a more "favorable" recurrence. However, it was shown that the survival of patients with Figure 3. Atypical cells that were classified as inconclusive. (Left) Numerous atypical cells exhibiting some pleomorphism, unisonucleosis, and prominent nucleoli (Papanicolaou stain, original magnification X100). (Right) An enlarged atypical dissociated cell with an eccentric hyperchromatic nucleus. Notice resemblance to malignant dissociated cells in Fig. 1 (Papanicolaou stain, original magnification X200).
FNA of Scar Lesions in Breast CA/Malberger et al. 151 Figure 4. A fragment of reactive connective (fatty) tissue exhibiting some enlarged nuclei and prominent nucleoli (Papanicolaou stain, original magnification X 100). chest wall recurrence involving the scar did not differ from that of patients whose chest wall recurrence was not associated with the scar.4 Because recurrence of BC affects therapeutic decisions and prognosis, diff erentiating BSL from recurrent MSL is essential. Traditionally, surgical biopsy, followed by histologic examination of the suspected scar lesions, was regarded as the procedure of choice. In recent years, aspiration cytology has become accepted and recognized as a diagnostic method for palpable breast rnas~es,~-~ and its diagnostic accuracy has been reported to be high.7,8 Paradoxically, FNA cytologic examination of these easily accessible, superficial, and visible lesions in surgical scars has not Table 1. Patients Characteristics - No. of women Mastectomy Radical Simple Simple modified Lumpectomy ~~ ~~ No. of patients HMSL HBSL 38 20 16 0 1 0 18 19 3 1 Clinical stage at the time of cancer diagnosis Local 15 11 Local and regional lymph nodes 20 8 Locally advanced 1 1 Not otherwise specified 2 0 Evidence for metastases at time of FNAB 19 6 -- HMSL: histologic condition of malignant scar lesion; HBSL histologic condition of benign scar lesion; FNAB fine-needle aspiration biopsy. - Table 2. Cytologic Findings of Representative Fine- Needle Aspiration Biopsy of Malignant Sear Lesions No. of patients Cvtoloeic condition HMSL HBSL Nonmalignant cells Atypical reactive cells Inflammatory cells Malignancy cannot be ruled out Suspected malignant cells Malignant cells Specified as adenocarcinoma Total 1 9 0 8 0 1 0 2 4 0 33 0 23 38 20 HMSL histologic condition of malignant scar lesion; HBSL: histologic condition of benign scar lesion. been evaluated or reported. Our cytologic findings on aspirates from 58 patients showed definite malignancy in 33 and suspected malignancy in four. In each of the latter 37 cases with positive or suspected cytologic findings, malignancy was confirmed subsequently histologically. The specificity of FNA cytologic examination in the diagnosis of MSL reached 100%. In view of this accuracy, patients can be treated on the basis of a definitive cytologic diagnosis, and there is no need for a confirmatory tissue biopsy. The high diagnostic accuracy of FNA cytologic examination is, however, conditional on obtaining an adequate and precise sample of the lesion, on the proper technical handling of the aspirate, and on accurate interpretation. Aspirations must be done by qualified and experienced personnel. In our study, 45% of FNA done by inexperienced clinicians did not yield representative samples (one third were of malignant lesions) compared with only 6.2% by an experienced operator; this emphasizes the need to restrict the use of FNA to a few experienced practitioners as is done in S~eden,~ rather than to farm it out among numerous inexperienced clinicians. There were no FP reports in our series. The only case of a FN result was caused by inadequate sampling Table 3. The Representativity of Aspirates According to Operator and Final Histologic Diagnosis Representative Acellular samples samples Operator HMSL HBSL HMSL HBSL Total Experienced cytopathologist 15 15 0 2 32 Inexperienced clinicians 24 4 8 15 51 HMSL histologic condition of malignant scar lesion; HBSL histologic condition of benign scar lesion. _-_
152 CANCER January 1, 2992, Volume 69, No. 1 not misinterpretation. Our findings indicate that FNA cytologic examination of SML lesions is an accurate and cost-effective method to provide prompt and accurate diagnosis of scar lesions. This minimally invasive procedure is safe, relatively easy to do, and suitable for multiple examinations; most patients tolerate it well. We therefore suggest that FNA be used routinely as a diagnostic procedure in scar lesions as an alternative to excisional biopsy. References 1. Haagensen LD, Stout AP. Carcinoma of the breast: I. Results of treatment. Ann Surg 1942; 116:801-815. 2. Donnegan WI, Perez-Mesa CM, Watson FR. A biostatistical study of locally recurrent breast carcinoma. Surg Gynecol Obstet 1966; 122~529-540. 3. Spratt JS. Locally recurrent cancer after radical mastectomy. Cancer 1967; 20:1051-1053. 4. Toonkel LM, Fix I, Jacobson LH, Wallach CB. The significance of local recurrence of carcinoma of the breast. Int ] Radiat Oncol Bid Phys 1983; 9~33-39. 5. Kreuzer G, Boquoi E. Aspiration biopsy cytology, mammography and clinical exploration: A modem set up in diagnosis of tumor of breast. Actn Cytol 1976; 20:319-323. 6. Barrow GH, Anderson J, Lamb L, Dixon JM. Fine-needle aspiration of breast cancer: Relationship of clinical factors to cytologic results in 689 primary malignancies. Cancer 1986; 58:1493-1498. 7. Malberger E, Toledano C, Barzilai A, Shrameck A. The decisive role of fine needle aspiration cytology in the pre-operative work-up of breast cancer. Zsr Med Sci 1981; 172399-904. 8. Palombini L, Fulciniti F, Vertani A et nl. Fine-needle aspiration biopsies of breast masses: A critical analysis of 1956 cases in 8 years (1976-1984). Cancer 1988; 61:2273-2277. 9. Zjicek J. Aspiration biopsy cytology. Monogr CIin Cytol 1971; 4:24-26.