Rosen's Breast Pathology, 4e

391

Ductal Carcinoma In Situ

comedocarcinomas. One of the 28 patients (4%) with micro- invasive carcinoma had ALN metastases. Silver and Tavassoli 308 defined microinvasion as “a single focus of invasive carcinoma less than or equal to 2 mm or up to three foci of invasion, each less than or equal to 1 m in greatest dimension” in a study of 38 patients. “Comedo” DCIS was present in 31 (82%), and papillary or other types of DCIS were present in 7 (18%). All patients were treated by mastectomy with axillary dissection, and no lymph node metastases were found. After a mean follow-up of 7.5 years, no patient had developed recurrent breast carcinoma. de Mascarel et al. 309 subclassified microinvasive duct car- cinoma into type 1 (single tumor cells) and type 2 (clusters of tumor cells). Type 1 cases would qualify for classification as T1mic , as would some type 2 cases. Among a subset of 20 type 2 cases that weremeasured, 6were T1 mic and 14 had invasive foci of between 2 and 10 mm. None of the 59 type 1 patients who had ALNs removed had nodal metastases. On the other hand, there were nodal metastases in 14 (10%) of the 139 patients with type 2 microinvasion who had ALNs examined. Distant me- tastases were reported in 2 (3%) of the 72 patients with type 1 microinvasion and in 12 (7%) of 171 with type 2microinvasion. The survival of patients with type 1 microinvasive carcinoma was similar to that of patients with pure DCIS and significantly better than that of patients with type 2 microinvasion. Information about patients with microinvasion defined as T1 mic (less than 1 mm) are becoming increasingly avail- able. Jimenez and Visscher 310 described 75 patients with mi- croinvasion, defined as one focus less than 5 mm or multiple foci with an aggregate diameter of less than 10 mm. Two or more histologically separate foci of invasion were present in 59% of the cases. Microinvasion consisting of isolated cell clusters less than 1 mm was present in 25 cases (33%). ALN dissection performed in 69 cases revealed metastatic carci- noma in five (7%). Two of these patients had invasive foci measuring less than 1 mm (T1 mic ), and in a third case the invasive lesion measured 1.1 mm. Walker et al. 311 compared the clinical and pathologic fea- tures of DCIS detected by mammography to patients who had symptoms, usually a mass or nipple discharge. Micro- invasion (T1 mic ) was found in 5 (5%) of 92 mammographi- cally detected and in 10 (13.5%) of 74 symptomatic cases. All but 1 of the 15 DCIS lesions with microinvasion were larger than 2 cm. Most DCIS with microinvasion had a comedo growth pattern or necrosis and high nuclear grade. By using a double immunostaining procedure for actin and cytokeratin, Prasad et al. 312 were able to confirm mi- croinvasion (T1 mic ) in 21 of 109 cases originally diagnosed as microinvasion or in which microinvasion was suspected. Eighteen lesions were ductal and three were lobular. The car- cinoma had high nuclear grade and necrosis in 16 of the 18 (89%) ductal lesions, including 13 (72%) described as comedo type. Axillary dissection performed in 15 patients revealed metastatic carcinoma in two cases, each with one lymph node involved. One of the 18 patients had recurrent carcinoma in the same breast after conservation surgery and radiotherapy, and another developed a chest wall recurrence of invasive duct carcinoma after a mastectomy. There were no systemic metastases after a median follow-up of 28 months.

Microinvasion associated with florid and pleomorphic LCIS may be mistaken for microinvasive ductal carcinoma. In the only series of 16 cases reported to date, ALN biopsies (including nine SLN samplings) were performed in 13 cases, with negative results in each, and all patients were alive without evidence of recurrence or of metastases in a mean follow-up of 24 months. 313 The foregoing data indicate that microinvasion is more likely to be found associated with high-grade, clinically evident DCIS than with low-grade, mammographically de- tected DCIS. When microinvasion is detected, the frequency of nodal metastases is 10% or less in patients who undergo axillary dissection, and the probability of systemic metasta- ses is 5% or less. Axillary dissection can be avoided in mi- croinvasive duct carcinoma if SLN biopsy is performed. This procedure adds greater precision with less morbidity to the detection of axillary nodal metastases. SENTINEL LYMPH NODE BIOPSY SLN biopsy is used to assess axillary nodal status in patients with invasive and in situ carcinoma. Routine use of SLN is generally not employed in DCIS, except when there is exten- sive high-grade DCIS, when invasive carcinoma is suspected, or when mastectomy is performed. 314 The incidence of nodal involvement in patients with DCIS has been reported to range from 0.5% to 1.5%, respectively 253 ; however, SLN posi- tivity can be higher in higher grade and more widespread DCIS. In a set of 854 DCIS cases treated at the European Institute of Oncology over more than a decade-long period ending 2006, SLN involvement was detected in 12 (1.4%), consisting of seven cases with micrometastases (less than 0.2 mm) and four patients with isolated tumor cells (ITC). 315 None of the 11 patients who underwent subsequent axillary dissection had additional nodal involvement. 316 Zavotsky et al. 317 found metastatic carcinoma in the SLN from 2 (14.3%) of 14 patients with DCIS. Completion axil- lary dissection revealed no other nodal metastases. Dauway et al. 318 cited nine patients with microinvasive carcinoma as- sociated with DCIS (T1 mic ). Three (33%) of these patients had micrometastases detected in an SLN by cytokeratin IHC and no other metastases in a completion axillary dissection. These investigators also reported that 5 (6%) of 86 patients with lesions classified as DCIS had metastases in an SLN. Four of the nodal metastases were detected only by cytokera- tin IHC. Four of the five patients had “comedo” DCIS, and the fifth had a 9.5-cm low-grade micropapillary and cribri- form lesions. Completion axillary dissection in four cases yielded no additional metastases. Several additional studies have examined the yield of SLN mapping in intraductal and microinvasive ductal carci- noma. Wilkie et al. 319 found a positive SLN in 27 (5%) of 559 patients with DCIS. Nineteen (70%) of the 27 positive SLNs were detected by IHC. Among 51 women with microinva- sive ductal carcinoma, 7 (14%) had a positive SLN, 5 (71%) of which were immunohistochemical findings. Katz et al. 320 reported finding a positive SLN associated with 8 (7.2%) of 110 breasts with DCIS. In four of these