Non-intervention vs. surgical interventions in (Low-Risk) Ductal Carcinoma In Situ: A DCIS multi-state model for decision analytics

s, EBCC 12 Posters A

Gifu University Hospital, Department of Breast Surgery, Gifu, Japan Background: Given the increasing number of early-stage breast cancers detected by screening mammography, we aim to establish RFA as a minimally invasive, cost-efficient, and cosmetically acceptable local treatment. In our Phase 1 study, localized tumors with a maximum diameter of 2 cm, preoperatively diagnosed by imaging and histopathology, were treated with RFA. A 90% complete ablation rate was confirmed histopathologically. Subjects andMethods: From Nov. 2009 to Nov. 2012, 58 patients with early-stage breast cancer received non-surgical RFA therapy. Patients had localized solitary N0 tumors with a maximum diameter of 1 cm. They underwent RFA and SNB under general anesthesia and adjuvant therapy and breast radiation [A1]. Follow-up evaluation for residual tumor at 3, 6, and 12 months after RFA included clinical examination, diagnostic imagings and vacuum-assisted biopsy. Surgical resection was recommended for patients with suspected residual disease or incomplete ablation. The primary endpoint was the frequency of adverse events. Secondary endpoints included the complete ablation rate and ipsilateral breast relapse-free rate.
Results: The follow-up period ranged from 15 to 109 months (median, 85 months). The 57 patients completed the non-surgical RFA procedure and underwent diagnostic imaging and needle biopsy after 3 months. Seven patients with suspected incomplete ablation underwent surgical resection; incomplete ablation was confirmed in 5 (8.6%, 2 with invasive and 3 with noninvasive ductal carcinoma). During subsequent follow-up, 1 patient each was diagnosed with contralateral breast cancer and ipsilateral breast tumor relapse. No distant recurrence was documented. Cosmetic results were excellent in 94% of patients.
Conclusions: RFA is a promising alternative to surgery for treating localized, early-stage breast cancer. Background: An active surveillance strategy has been proposed for patients with low-and intermediate-grade ductal carcinoma in situ (DCIS). Prospective trials to measure clinical outcomes are on-going, and results will not be available for >10 years. In lieu of prospective data, there is value in creating a disease model for low-risk DCIS to understand the potential impact of an active surveillance strategy.
Methods: Multi-state models were developed using patient-level data from the SEER 18 Registries database, for 4 treatment strategies (no local treatment, breast conserving surgery [BCS], BCS + radiotherapy [RT], mastectomy), and for women with low-risk features. Eligible cases included women with grade 1, 2, and 3 histologically-confirmed DCIS as first primary, diagnosed between 1992-2016, aged ≥40 years at diagnosis, and known laterality, local treatment status, survival time, and cause of death. The multistate model considers 5 mutually exclusive states: DCIS diagnosis, ipsilateral invasive breast cancer (iIBC) ≤5 years post-DCIS diagnosis, iIBC >5 years post-DCIS diagnosis, death preceded by iIBC, and death not preceded by iIBC. Transitions between each state were modelled with Cox proportional hazards models. The effects of treatment strategy, age, diagnosis year, grade, ER status, and race on each transition was assessed. Missing covariate values were imputed.
Conclusions: Baseline DCIS characteristics are predictive of iIBC events diagnosed within 5 years, and are therefore useful in selecting patients for treatment. Women with low-risk features represent 16% of the studied population, and demonstrate minimal differences by treatment strategy in the probability of surviving iIBC-free at 10 years. This suggests there is opportunity to de-escalate treatment for women aged 50-69 at diagnosis, with ER+, grade 1+2, ≤1 cm DCIS lesions. This work was supported by Cancer Research UK and by KWF Kankerbestrijding (ref. C38317/A24043).
No conflict of interest.

123
Poster Psycho-social assessment of post-surgery outcomes after breast reconstruction surgery L. Busheri