In breast cancer, oligometastatic disease is a controversial concept. Generally, solitary pulmonary lesions are detected on chest imaging made for other reasons (unintentionally). The majority of these pulmonary lesions in breast cancer patients prove to be unrelated (bronchogenic carcinoma or benign lesions), 33–40% are in fact breast cancer metastases. Only a small number of breast cancer patients (0.4–1%) will have potentially resectable isolated metastasis to the lung.

Controlled studies comparing pulmonary metastasectomy to systemic therapy or best medical care are lacking. Overall survival (OS) data in patients who underwent metastasectomy are favourable despite the high incidence of recurrences. Long-term survival is seen primarily in patients with only one or two lesions and long initial disease-free interval. Radical resections with unnecessary loss of pulmonary parenchyma should be avoided, since there is no benefit of such resection in terms of survival and the incidence of complications is higher.

Foremost justification of surgical resection for lung lesions in breast cancer patients is differentiation of metastatic breast cancer from new primary lung cancer. The survival impact of pulmonary resection in ‘oligometastatic’ breast cancer is probably limited and attempts should be made to limit the amount of tissue removed.

The treatment of lymph node metastasis in human cancers has been a source of controversy for many years. Several contemporary randomized trials indicate no survival advantage to performing axillary dissection in contrast to merely observing the axilla after invasive breast cancer. This assumption is confirmed since survival advantages cannot be demonstrated when comparing greater or lesser lymph node dissections or dissection vs. observation of regional lymph nodes in melanoma and breast cancer and all other epithelial cancers.

Many recent studies indicate that axillary metastases, when they do occur in this era of mammographic screening and small invasive cancers, demonstrate metastases in the axillary sentinel nodes only in from 20% to almost 100% depending on primary cancer size. For instance, in T1a and T1b cancers, 50–100% of lymph node metastases, infrequently discovered, are contained entirely within the sentinel nodes. In T2 cancers, this proportion is roughly 25–50%, and in T3 cancers it ranges from 20% to 40%. A number of recent articles have demonstrated no difference in survival and very little nodal recurrence when patients with positive sentinel nodes are not followed by axillary dissection although the axillary disease tends to be less advanced in observed compared to treated patients with axillary dissection. Nevertheless, basic principles have been reconfirmed in these reports.

It has also been shown in animal research that clones of particular metastatic site cells demonstrate organ specificity. Transplanted human breast cancers demonstrate specific homing abilities to selective distant metastatic organ sites such as lymph nodes, bone, lung, or even adrenal gland. Genetic profiles of the different clones of cells with high metastatic specificity are each unique. Thus it is assumed that lymph node specific metastatic cells (and other organ-site-specific cells) have no ability to lodge and grow in other metastatic sites; this has been demonstrated clinically in clinical trials previously mentioned.

Therefore, there is no benefit in performing subsequent axillary dissections even when the sentinel node biopsy is positive, because of: (1) very low risk (≤2%) of recurrent clinical nodal metastases, (2) the lack of impact on disease specific or overall survival, (3) most lymph node metastases are limited to the sentinel nodes, and (4) lymph node metastases are only indicators not governors of the distant metastases. The American College of Surgeons Oncology Group (ACOSOG) Z-11 Trial was designed to specifically address this point, but unfortunately was closed early because of poor accrual; nevertheless, 900 patients were enrolled and will be analyzed over time to address this point.