Occult Tumor Cells in Marrow Predict Breast Cancer Survival

NEW ORLEANS—A considerable proportion of patients with primary breast cancer have minimal residual disease in the bone marrow that is prognostically relevant and detectable by bone marrow aspiration, according to German investigators who reported their findings at the 92nd Annual Meeting of the American Association for Cancer Research (AACR).

The results of the studies do not have immediate therapeutic implications, but they add to a growing concern that patients who appear to have a favorable prognosis may actually be headed for metastatic disease.

"The message is that we may be treating some patients in error or some not at all who are at increased risk," said Wolfgang Janni, MD, of Ludwig-Maximillians University, I. Frauenklinik, Munich, Germany. The director of the I. Frauenklinik is Professor Dr. G. Kindermann.

Dr. Janni’s trial (abstract 204) examined the fate of occult metastatic cells detected in the bone marrow at the primary diagnosis of breast cancer and at a second aspiration at a later time. The aim was to determine whether outcome was affected by the presence of these occult metastatic cells on one aspiration or both (persistently positive).

Cytokeratin Positivity

The study included 89 patients with stage I-III breast cancer who were free of recurrence. Bone marrow aspirates taken at diagnosis and after a median interval of 19 months (range, 7 to 67 months) were analyzed for cytokeratin positivity. The patients then were followed prospectively for a median of 41 months (range, 12 to 78 months) after the first bone marrow aspiration.

At the time of primary breast cancer diagnosis, 24 of 89 patients presented with occult metastatic cells in the bone marrow. At the second aspiration, 50 patients remained negative for occult tumor cells; 10 patients remained positive; 15 became newly positive; and 14 who were initially positive became negative (see Table), Dr. Janni reported, along with Christian Schindlbeck, MD, and their colleagues.

The 50 patients with a persistently negative bone marrow status had significantly improved survival at the median 41-month follow-up, both by univariate analysis (P = .045) and multivariate analysis (P = .034).

Their overall survival was 90% at this time point, compared with 60% for the 25 patients who had positive findings at the second aspiration. Patients who were persistently positive for occult tumor cells had a relative risk of death of 4.48, the study found.

An Interesting Question

Dr. Janni could only speculate as to why some patients with an initially negative finding would have occult tumor cells detected on the second analysis. "Maybe there are too few tumor cells initially for detection," he said. Furthermore, he noted, tumor cells might be released into the circulation by the primary treatment (surgery or chemotherapy). "It would be interesting to examine this question by doing an aspirate before surgery and then another one soon after surgery," he said.

At this time, the finding of occult tumor cells does not influence treatment decisions. "We know from other studies that chemotherapy really has no influence on the presence of tumor cells in the bone marrow, because they are dormant and not proliferating," he noted.

This research group currently has access to over 1,300 bone marrow aspirates in their database. Additional research based on this sample has found that the direct identification of these occult metastatic cells may be a more accurate prognostic predictor than HER-2/neu status and angiogenesis.

In this ongoing study (abstract 279) led by Stephan Braun, MD, of the Technical University, Munich, almost 400 tumor samples and bone marrow aspirates were analyzed and their characteristics correlated with patient survival.

The investigators looked at HER-2/neu overexpression (by CB11 monoclonal antibody), CD31 expression, and presence of occult tumor cells in the bone marrow. Bone marrow micrometastases were detected in 30% of patients.

At 4-year follow-up, both univariate and multivariate analyses confirmed that occult metastatic cells are an independent prognostic factor for cancer-specific survival. Hormone receptor status and lymph node status were also prognostic.

At the time of analysis, neither HER-2 overexpression (2+ and 3+) nor angiogenesis (Chalkley counts of greater than 4 microvessels) gave significant information regarding prognosis, Dr. Braun said.

For the 235 patients for whom all factors were known, multivariate analysis yielded the following P values: lymph node status, P = .0054; hormone receptor status, P = .014; bone marrow cells, P = .028.

"The presence of occult metastatic cells in the bone marrow defined the high-risk population," Dr. Braun said, noting that these are potentially precursor cells of subsequent metastasis that chemotherapy does not entirely eliminate.

"In addition, the presence of occult metastatic cells provided a clear separation of high-risk patients in each biologically defined tumor subgroup," he said. "Allowing for all factors, the presence of occult tumor cells was an independent prognostic factor for death in node-negative breast cancer. Tumor cell dissemination is, in itself, the risk factor we are looking for."