Background: Ductal carcinoma in situ (DCIS) accounts for over 25% of breast neoplasms diagnosed in the US. This study aims to identify factors influencing the upstaging rate of DCIS in the high risk mastectomy patient subgroup, outline which surgical interventions are optimal and whether all patients require axillary staging.
Methods: Patients undergoing mastectomy for newly diagnosed DCIS were retrospectively studied. A univariate analysis of their demographics, methods of diagnosis, complete surgical treatment, pathologic characteristics and outcomes were compiled to determine which variables impacted tumor upstaging to invasive cancer at final pathology.
Results: A total of 212 women were identified. Mean age was 59 years (range 31-88) with 74% postmenopausal. Only two patients had a genetic mutation (BRCA2). Most (95%) were diagnosed mammographically with the findings including calcifications only (77%), calcifications + mass/architectural distortion (9%), and only mass/architectural distortion (9%). Ten patients were diagnosed via MRI and one patient was diagnosed with Paget’s disease on punch biopsy. Size on imaging was stated 69% of the time with a mean of 3.1 cm (range 2-12 cm). When size was unstated, 53% of lesions were unifocal whereas others were extensive/multi-focal.
Most patients were diagnosed by core (84%) and the remainder by excisional biopsy. Thirty-four lesions (16%) were palpable. The majority of DCIS lesions were high grade (55%), with 12% low and 33% intermediate; 34% exhibited comedo-necrosis and 66% were ER-positive. Histologic suspicion of microinvasion was present among 11% of biopsies and the treating surgeon recorded clinical suspicion of invasion among 18%. Most underwent reconstruction (69%), predominantly with implant-based technique (67%) with 41% pursuing bilateral mastectomy. Axillary staging was carried out in 150 patients (71%) with SLN biopsy in all but 1 patient. Thirty-four patients (16%) were upstaged to invasive cancers with a mean tumor size of 1.3 cm. Staging of the invasive cancers was T1mic (n=8), T1a (n=13), T1b (n=5), T1c (n=3), T2 (n=4), and T3 (n=1). Only 1 patient (<1%) had a positive SLNB. Univariate analysis identified that the variables significantly associated with upstaging were a histologic suspicion of microinvasion (50% upstaged, p<0.001), ER- (p=0.026), surgeon suspicion of invasion (36% upstaged, p<0.001), palpable disease (30% upstaged, p=0.02), and pre-operative tumor size >2 cm (p=0.04). Mean tumor size was associated with upstaging with 3.8 vs 3.0 cm, p=0.066 but did not reach statistical significance. All patients that were upstaged had a SLNB. Factors not found to be associated with upstaging included type of biopsy (core vs excision), grade, comedo-necrosis, mammography findings, or use of MRI.
Conclusions: In this high risk population the upstage rate from DCIS to invasive cancer was 16%. Clinical judgement, palpable disease, suspicion of microinvasion, and large tumor size were associated with upstaging. Surgeons can use these factors in deciding to forego SLNB in patients with small, low risk DCIS undergoing mastectomy.