A key issue for oncologists treating breast cancer patients with preoperative chemotherapy remains whether to perform sentinel lymph node (SLN) biopsy before or after administering the drugs.
BETHESDA, MarylandA key issue for oncologists treating breast cancer patients with preoperative chemotherapy remains whether to perform sentinel lymph node (SLN) biopsy before or after administering the drugs. Two questions dominate the debate: Does sentinel node biopsy work as well following preoperative therapy as it does in patients who do not receive neoadjuvant treatments? Is information important to managing patients lost when physicians delay SLN biopsy until after the drug treatments?
Two oncologists versed in this neoadjuvant therapy debate presented the pros and cons at "Preoperative Therapy in Invasive Breast Cancer," a 2-day state-of-the-science conference sponsored by the National Cancer Institute. Terry Mamounas, MD, MPH, associate professor of surgery, Northeastern Ohio University College of Medicine and medical director of the Aultman Cancer Center, Canton, Ohio, provided the case for SLN biopsy after neoadjuvant therapy. Jay Harris, MD, chief of radiation oncology, Dana-Farber Cancer Institute, and professor of radiation oncology, Harvard Medical School, presented the merits of SLN biopsy prior to preoperative treatment.
The Issue of Downstaging
The issue of SLN biopsy timing stems from the downstaging of positive axillary nodes that can occur with neoadjuvant therapy. In four studies cited by Dr. Mamounas, for example, researchers observed downstaging in 19% to 43% of patients. Dr. Harris cited results from National Surgical Adjuvant Breast and Bowel Project (NSABP) study B-18 in which the proportion of patients with positive nodes dropped from 57% to 41%, and that of patients with four or more positive nodes dropped from 27% to 16%, with neoadjuvant chemotherapy, compared with adjuvant therapy.
As long as axillary dissection remained the sole staging method, the finding had no clinical significance. "However, the advent of sentinel node biopsy produced another potential benefit from neoadjuvant chemotherapy, and that is the potential for decreasing the extent of axillary surgery with sentinel node biopsy if the axillary nodes are downstaged," Dr. Mamounas said. "The big question is, of course: Is sentinel node biopsy as feasible and accurate after neoadjuvant therapy as it is before any systemic therapy?" Dr. Mamounas argued that it is.
Data from early single-institution studies showed considerable variation in the rates of sentinel node identification after neoadjuvant chemotherapy, from 72% to 100%, and in false-negative sentinel nodes, from 0% to 33%, Dr. Mamounas noted. Later single-institution studies set more precise numbers. "If you look at all of them together, you see the overall identification rate is about 89%, with a false-negative rate of 10.8%," he said.
NSABP study B-27, a large, multicenter trial, provided researchers a look at the performance of SLN biopsy following preoperative chemotherapy. In 428 women, a sentinel node biopsy was followed by an axillary node dissection. The identification and false-negative rates varied, depending on whether surgeons relied on blue dye staining alone or also used radiocolloid. "The identification rate was 85%, higher if isotope was used," Dr. Mamounas said. "The false-negative rate was about 11%, and, again, it was higher if only blue dye was used and lower if isotope was used."
Moreover, a meta-analysis of 21 single-institution and multicenter studies with a total of 1,273 patients estimated the identification rate at 90% and put the false-negative rate at 12%. "So the conclusion from the meta-analysis was that sentinel node biopsy is a reliable tool for planning treatment after neoadjuvant therapy," Dr. Mamounas noted.
He cited several disadvantages associated with SLN biopsies performed prior to preoperative chemotherapy: They necessitate that women undergo two surgeries at separate timesthe sentinel node biopsy followed by a breast procedure after completing neoadjuvant therapy; evidence on outcomes remains unknown for either approach, pending the results of large clinical trials; and knowing that a patient had negative axillary nodes prior to neoadjuvant therapy is of limited value in planning chemotherapy following surgery.
"Patients with large operable breast cancers, who are typically candidates for neoadjuvant chemotherapy, have a high likelihood of having positive nodes, in the range of 50% to 70%," Dr. Mamounas said. "And obviously, the approach of doing sentinel node biopsy before does not take advantage of the downstaging effect of neoadjuvant chemotherapy."
Role Before Preop Chemo
Dr. Harris focused much of his discussion on the role of SLN biopsy before preoperative chemotherapy in determining postsurgery radiation treatments. Validated guidelines for local-regional radiation in breast cancer patients are based on the initial state of axillary nodes, but whether these recommendations apply to nodes examined after neoadjuvant therapy remains undetermined.
"We know that effective local-regional treatment is important. We know that preoperative treatment downstages disease, but the need for local-regional radiation therapy based on this downstaged disease is not well known. We have very limited published data on local recurrence rates with preoperative treatment and no radiation," Dr. Harris said. "So the issue is, who gets radiation in this scheme of things, and this question has added significance with the 2005 publication showing that reductions in local recurrence of 10% or greater are not only important for local control but also improve absolute 15-year survival by about 5%."
To further support his argument, he cited work by Thomas Buchholz, MD, professor and chair of radiation oncology, M.D. Anderson Cancer Center, and his colleagues. The researchers did a retrospective analysis of 150 patients, many with locally advanced cancer, who had preoperative doxorubicin or paclitaxel and did not undergo radiation after breast surgery. At 5 years, patients who were node negative after preoperative chemotherapy had a local recurrence rate of 10% (higher than that seen with de novo sampling or initial dissection); patients with one to three positive nodes had a 17% recurrence; and those with four to nine positive nodes had a 47% recurrence rate.
Among the 18 patients with a complete pathological response, the 5 year-recurrence rate was a surprisingly high 19%. "So evaluation after systemic therapy was not highly predictive for recurrence rates," Dr. Harris said.
Dr. Buchholz's team next compared the same 150 women with a group of 1,030 breast cancer patients, matched for stage who received adjuvant chemotherapy without radiation therapy. The number of positive nodes in the postoperative chemotherapy patients also proved predictive of recurrence, and 5-year local recurrence rates were not significantly different between the preop and postop patients.
Thus, the response to chemotherapy based on final nodal status at breast and axillary surgery does not appear to reduce local-regional recurrence rates to the rates seen after the same staging with initial surgery, Dr. Harris said. Both the initial and the final stage must be used to determine the local-regional recurrence risk, he said.
"The timing of sentinel node biopsy is clearly in evolution, and we clearly need more data," Dr. Harris said. "However, until we have validated prognostic data on local-regional recurrence risks, it seems prudent to do the sentinel node biopsy prior to preoperative systemic therapy."