Source: www.oncologystat.com
Author: K Wachter, Elsevier Global Medical News

Surgical excision of breast cancer prior to sentinel lymph node dissection can displace isolated tumor cells to the sentinel lymph node, but these tumor cells appear to have little clinical significance, according to an analysis of more than 17,000 patients in a large database.

“Earlier surgical excision leads to a nearly fourfold increase in the risk of having isolated tumor cells in the sentinel node indicating iatrogenic displacement,” Dr. Tove F. Tvedskov said at the annual San Antonio Breast Cancer Symposium. She advised, however, that these cells are probably without clinical significance, and that the omission of axillary lymph node dissection should be considered.

The study was based on data from the Danish Breast Cancer Cooperative Group database, which includes more than 80,000 women with breast cancer. Approximately 3,000 new sentinel lymph node dissections (SLNDs) are included in the database each year, with clinical and histopathologic data prospectively collected for these cases. Data from this database were combined with data from the Danish National Health Registry, which includes all surgical procedures performed in Danish hospitals.

The researchers identified 414 breast cancer patients who underwent surgical excision up to 2 months before SLND and compared them with 16,960 breast cancer patients who underwent SLND without prior surgical tumor excision.

“The proportion of patients with isolated tumor cells was almost three times higher in the group with earlier surgical excision, compared to the group without earlier surgical excision,” said Dr. Tvedskov of Copenhagen University Hospital. “One obvious explanation for these differences between groups is, of course, that the group with earlier surgical excision is a highly selected group of patients, where the cancer diagnosis was not obvious from the beginning.”

Therefore, the researchers used a multivariate model to take into account these differences by adjusting for tumor size, histologic type and malignancy grade.

“We found that the risk of having isolated tumor cells in the sentinel node was nearly fourfold increased when surgical excision was performed, compared to patients without any earlier surgical excision,” Dr. Tvedskov said. “Surgical excision leads to some degree of iatrogenic displacement.”

In all, 64% of patients who underwent earlier surgical excision and 59% of those without earlier surgical excision were negative for sentinel lymph node metastases. Among the prior-excision group, 9% had isolated tumor cells in the sentinel lymph node, compared with 3% of the control group (odds ratio, 3.90; P less than .001).

Likewise, micrometastases in the sentinel lymph node were more common (15%) among the prior-excision group, compared with 12% in the control group (OR, 1.48; P = .006). However, macrometastases in the sentinel lymph node were more common in the control group (26% vs. 12%).

The researchers next looked at whether the histologic type of tumor mattered. After adjustment, “we found that the risk of having isolated tumor cells was three times higher in patients with lobular carcinomas, compared to patients with ductal carcinomas,” said Dr. Tvedskov. The odds ratio was 3.02 (P less than .0001), after adjusting for tumor size, malignancy grade, and earlier surgical excision.

“Finding these results, we expected that the extra isolated tumor cells after earlier surgical excision would especially come from lobular carcinomas,” she said. However, in a subanalysis, “we found that in the group of patients with ductal carcinomas, the risk of isolated tumor cells after surgical excision was nearly fivefold increased, compared to patients without surgical excision, whereas, in the group of patients with lobular carcinomas, there was no significant increased odds ratio.”

Among patients with ductal carcinoma, isolated tumor cells were significantly more likely to be present in patients who underwent prior surgical excision compared with control patients (OR, 4.91; P less than .0001), after adjusting for tumor size and malignancy grade. Among lobular carcinoma patients though, isolated tumor cells were not significantly more likely to be present in those who underwent prior surgical excision (OR, 1.27; P = .69), after adjusting for tumor size and malignancy grade.

“So, despite the fact that the lobular carcinomas are, in general, more likely to present with isolated tumor cells in the sentinel node, the extra isolated tumor cells caused by the earlier surgical excision seems to come from ductal carcinomas,” said Dr. Tvedskov.

The investigators also considered the question of clinical significance: Were the isolated tumor cells related to nonsentinel node metastases, or could an axillary lymph node dissection be omitted in these patients?

Looking at the distribution of nonsentinel node metastases in patients with either isolated tumor cells or micrometastases in the sentinel node, they found that isolated tumor cells were not present in nonsentinel lymph nodes among patients who underwent prior surgical excision. These cells, however, were found in nonsentinel lymph nodes in 12% of control patients.

Dr. Tvedskov pointed out that this difference did not achieve significance because of the small number of patients. There was no significant difference in the likelihood of micrometastases in nonsentinel lymph nodes between the two groups.

“This indicates that isolated tumor cells in patients with earlier surgical excision are not related to nonsentinel node metastases, and are therefore without clinical significance,” she said.

The investigators reported that they have no relevant financial relationships.