# A Pragmatic Grouping Model for Bone-Only De Novo Metastatic Breast Cancer (MetS Protocol MF22-03)

## TL;DR

This study proposes a new staging model for bone-only metastatic breast cancer to identify patients who benefit from surgery and other local treatments.

## Contribution

A pragmatic grouping model to guide locoregional therapy decisions in de novo bone-only metastatic breast cancer.

## Key findings

- Group A patients had a 43% lower hazard of death compared to Group B.
- Primary tumor surgery significantly improved survival in Group A but not in Group B.
- Median overall survival was 65 months for Group A and 44 months for Group B.

## Abstract

A refined staging model incorporating biological and anatomical tumor characteristics can better identify de novo bone-only metastatic breast cancer (dnBOMBC) patients who benefit from multimodal therapy, potentially guiding future treatment guidelines. A subgroup of dnBOMBC patients demonstrating overall survival (OS) comparable to Stage III breast cancer may be reclassified as Stage III D and thus directed toward locoregional therapy (LRT). Our study offers a practical model to identify those benefiting most from LRT, providing a more applicable framework than current nomograms.

De novo metastatic breast cancer (dnMBC) accounts for 3–10% of newly diagnosed cases, with 20–40% presenting as a bone-only metastatic disease, which can achieve survival outcomes exceeding 10 years with multimodal therapy. However, the role of multimodal therapy remains controversial in the guidelines. Objective: This study aims to identify dnBOMBC subgroups to develop a pragmatic staging system for guiding locoregional therapy decisions. Materials and Methods: Data from the MF07-01 phase III randomized trial (2021, median follow-up time (mFT): 40 months (range 1–131)) and the BOMET prospective multi-institutional registry trial (2021, mFT: 34 months (range 25–45)) were combined for analysis, including only patients who presented with bone-only metastases. Exclusion criteria were patients under 18 and those with a history of prior cancer or cancer metastases. Patients with missing data and positive surgical margins were excluded. Out of 770 patients, 589 were included. Survival analyses were first conducted according to molecular subgroups, after which patients were further stratified by hormone receptor status, human epidermal human epidermal growth factor receptor 2 (HER2) status, tumor grade, and clinical T (cT) stage. Group A (GrA) included hormone receptor (HR)-positive, low- or intermediate-grade tumors at any cT; HR-positive, high-grade tumors with cT0–3; or any HER2-positive tumors. Group B (GrB) included HR-positive, high-grade tumors with cT4 disease or any triple-negative (TN) tumors. Results: The hazard of death (HoD) was 43% lower in GrA than in GrB. Median OS was 65 months (39–104) for GrA patients and 44 months (28–72) for GrB patients (HR 0.57, 95% CI 0.41–0.78, p = 0.0003). Primary tumor surgery (PTS) significantly improved OS in GrA patients, regardless of the number of metastases (solitary: HR, 0.375, 95% CI 0.259–0.543, p < 0.001; multiple: HR 0.435, 95% CI 0.334–0.615, p < 0.001). Conversely, GrB patients did not experience a significant benefit from PTS. Conclusions: This study demonstrates that GrA patients have better OS than GrB patients, and PTS reduces the HoD in GrA patients compared to systemic therapy alone. These findings support using a modified staging system in dnBOBMC to identify patients who may benefit from multimodal therapy including PTS.