Tumor Cells in Marrow May Better Predict Metastases Than Axillary Node Dissection

HEIDELBERG, Germany—Evaluation of bone marrow for breast cancer cellsproved superior to axillary lymph node dissection in predicting subsequentmetastases in a German study of more than 1,000 patients, Ingo J. Diel,MD, said at a general session of the San Antonio Breast Cancer Symposium.

Tumor cell detection in the marrow had the strongest independent associationwith disease-free survival during a median follow-up of 40 months. In particular,bone marrow tumor cell status proved more accurate than lymph node statusas a prognostic factor associated with small tumors. Notably, a third ofpatients with negative lymph nodes had tumor cells in aspirated marrowsamples.

At the very least, the findings warrant consideration of tumor celldetection as a replacement for node dissection in patients with T1 disease,said Dr. Diel, a surgeon at Women's Hospital, University of Heidelberg.

"At present, axillary node status is the best prognostic factorin operable primary breast cancer," he said. "But 30% of node-negativepatients will relapse at distant sites, and 10% of all patients with metastaticdisease are node negative at first diagnosis."

More than 1,000 Samples

Since 1985, Dr. Diel and his colleagues have performed immunocytologicassays on bone marrow samples from 1,026 breast cancer patients prior tosurgery. The samples are collected from both anterior iliac crests, andstaining is performed using a monoclonal antibody that reacts with thecore protein of TAG-12, a tumor associated glycoprotein expressed by morethan 95% of breast cancers.

Bone marrow samples tested positive in 42% (428 cases), and 546 women(53%) were node negative. Tumor cells appeared in 131 of 408 women withstage T1 disease and in 177 of 436 with T2 disease. Almost a third of node-negativewomen (172) had tumor cells detected by bone marrow assay.

Thus far, 176 patients have developed metastases, and 144 of them (82%)initially had positive bone marrow assays. Similarly, 53 of 73 patients(81%) who have died during follow-up had tumor cells in their bone marrowat the time of surgery.

Women with positive bone marrow assays have had a significantly brieferdisease-free survival, a median of 31 months, compared with 43 months forwomen whose marrow samples were free of tumor cells at the time of surgery.Multivariate analysis showed tumor cell status of bone marrow to be themost important predictor of disease-free survival:

Prognostic Factors for Disease-Free Survival in Breast Cancer: Heidelberg

A separate analysis of disease-free survival by disease stage showedthat tumor cell detection was significantly related to the risk of metastaticrecurrence in women with small (stage T1) tumors (RR, 13.9). In contrast,node status did not predict metastasis in women who had T1 disease (RR,1.9). Both factors were predictive for T2 disease, but tumor cell detectionstill proved to be a stronger prognostic factor (RR, 3.9 versus 2.4).

"In patients with T2 disease, node status provides additional information,but in T1 patients, tumor cell detection is the most important predictorof disease-free survival. Node status adds no additional prognostic information,"Dr. Diel said.

Axillary node dissection is associated with significant morbidity, henoted. Up to 30% of patients develop seromas, the incidence of dysesthesiaexceeds 20%, restriction-associated morbidity occurs in 15% to 20% of cases,and lymphedema, infection, and hemorrhage occur in smaller numbers of patients.

In contrast, only one major hemorrhage occurred in association withbone marrow aspiration in the Heidelberg series. The most common adverseeffect was hematoma, which occurred in 39 patients. Three others had postoperativepain attributed to the marrow aspiration.

The Question

"The question is, Why don't we replace axillary dissection withbone marrow aspiration tumor cell detection in patients with small tumors?"Dr. Diel said. The Heidelberg team has begun a clinical trial to answerthis question. The trial involves women who have small tumors and who arenode negative by clinical examination and ultrasound. The patients arerandomly assigned to axillary lymph node dissection or tumor cell detection.

In the tumor cell detection cohort, those who test positive will receiveadjuvant chemotherapy, and those who are tumor cell negative will receiveno chemotherapy after surgery. All patients in the cohort will have follow-upexaminations of axillary lymph nodes, and women who have regional relapseswill have secondary lymphadenectomies.

"We hope we can diminish the morbidity and cost associated withaxillary lymph node dissection, while at the same time maintaining thelevel of safety for our patients," Dr. Diel said.