Can Axillary Ultrasound Aid a Tricky Breast Cancer Decision?
Do all women with lymph node-positive breast cancer treated with neoadjuvant chemotherapy need to have all of their axillary nodes surgically removed?
The answer seems to depend on whom you ask.
For example, Judy Boughey, MD, a surgical oncologist from the Mayo Clinic in Rochester, Minnesota, says no.
She thinks it’s time to personalize the management of the axilla in breast cancer patients, and points to new data from the American College of Surgeons Oncology Group Z1071 trial as evidence.
The Z1071 investigators, led by Dr Boughey, report that axillary ultrasound (AUS) might be a useful tool for judging — with an “acceptable” rate of incorrect evaluations — whether or not neoadjuvant chemotherapy has eliminated cancer from the lymph nodes, and can therefore guide the use of sentinel lymph node (SLN) surgery.
“AUS is recommended post-chemotherapy to guide axillary surgery,” the Z1071 team concludes. They add that using the less invasive, less morbid alternative — SNL surgery — is now an acceptable approach in some patients with clinically involved nodes.
Their new study, the first-ever multicenter trial to look at the role of ultrasound after chemotherapy, was published online February 2 in the Journal of Clinical Oncology.
“In the old days, we always performed full axillary dissection,” Dr Boughey said in an interview with Medscape Medical News. “But surgical dictums need to be re-evaluated in the face of advancements in multidisciplinary care.”
Two other surgical oncologists contacted by Medscape Medical News disagree with Dr Boughey’s assessment that some of these patients can undergo SNL surgery, even if the decision is aided by AUS after chemotherapy.
In this setting, “axillary lymph node dissection is still the standard of care,” said Dalliah Black, MD, from the University of Texas M.D. Anderson Cancer Center in Houston. “More research is needed in node-positive patients who receive neoadjuvant chemotherapy.”
And Quyen Chu, MD, from the Feist-Weiller Cancer Center at Louisiana State University Health–Shreveport, said he believes that “surgeons should continue to offer completion axillary lymph node dissection in patients with clinically involved lymph nodes, regardless of the nodal response to chemotherapy, until we have further evidence otherwise.”
Both Dr Black and Dr Chu hope to move in the direction of personalizing this treatment decision, but want more robust evidence.
Dr Boughey acknowledged that adding AUS after neoadjuvant chemotherapy is not the “end-all, be-all” clinical solution in this setting. “But I do think it is an additional tool to help surgeons incorporate sentinel lymph node biopsy in their practice,” she said.
Furthermore, it is likely that in patients with positive nodes treated with neoadjuvant therapy, “a lot of surgeons changed their practice” after first results from Z1071 were published in 2013 (JAMA.2013;310:1455-1461), Dr Boughey said.
At that time, she and her colleagues concluded that some women with node-positive breast cancer who receive neoadjuvant chemotherapy might not need to automatically undergo axillary lymph node dissection. Instead, it might be possible for some to undergo a less invasive SLN surgery, as reportedby Medscape Medical News.
But the study results had a problem. In patients who presented with clinically node-positive disease and had two or more SLNs identified and removed, the false negative rate (FNR) for SLN surgery after neoadjuvant chemotherapy was 12.6%. This was higher than the 10.0% cutpoint defined in the trial as acceptable.
The new study is an effort to find a better way to select patients for SNL surgery to decrease the FNR.
The Z1071 investigators hypothesized that AUS, which is often used to assess the axilla at the time of breast cancer diagnosis, could be used a second time — after neoadjuvant chemotherapy — to guide patient selection for less invasive surgery.
Dr Boughey explained why: “Our chemotherapies and targeted therapies are now so effective that we are getting really good at eradicating disease in the axillary nodes.”
Pathologic nodal response rates to neoadjuvant regimens as high as 70% have been reported, the investigators point out.
“Why not repeat the ultrasound to see how things have changed?” Dr Boughey asked.
“Patients with normal-appearing lymph nodes on AUS after chemotherapy are at lower risk of residual nodal disease and may be more suitable for SLN surgery,” the investigators hypothesize.
Every Sampling Procedure Has Misses
In the current study, the team reports on a prespecified secondary end point of the Z1071 trial — namely, the results of AUS after neoadjuvant chemotherapy.
All 687 patients in the trial had T0 to T4, N1 or N2, M0 breast cancer. Postchemotherapy AUS images were available for 611 patients. The images were centrally reviewed and classified as “normal” or “suspicious.”
Suspicious cases were more likely to be node-positive at surgery than normal cases (71.8% vs 56.5%;P = .0004).
In addition, the number of positive nodes and the metastasis size were greater in suspicious than in normal cases (P < .0001).
But, notably, there was no difference in FNR for SLN between the suspicious and normal cases.
However, a strategy in which only patients with normal images undergo SLN surgery, “would potentially reduce the FNR in Z1071 patients with 2+ SLNs removed from 12.6% to 9.8% when preoperative AUS results are considered as part of SLN surgery,” the investigators report.
In other words, this strategy would result in a FNR of less than 10%, which would be acceptable.
Dr Black pointed out that the difference in FNRs between patients with suspicious and normal nodes was not statistically significant (P = .09). Still, she agrees with the conclusion of the Z1071 team.
“These findings demonstrate that including axillary ultrasound after neoadjuvant chemotherapy can identify residual abnormal lymph nodes and help with determining which patients may be candidates for SLN biopsy,” she said.
But Dr Black emphasized that this approach is not ready for everyday practice, and wants to see what other research from her institution finds out about selecting patients for SLN surgery.
“Our approach is to place a clip in the axillary lymph node when metastasis is identified on initial ultrasound-guided FNA. At surgery, the clipped metastatic node is excised by radioactive iodine seed localization, and tracers are injected to perform the SLN biopsy to determine if the clipped metastatic node is also the SLN. This will allow pathologic evaluation of these nodes in comparison to the completion axillary lymph node dissection,” she explained.
The feasibility of this targeted approach to node excision was reported last year (JAMA Surg. Published online December 17, 2014).
Dr Black, Dr Chu, and the Z1071 investigators all pointed out that other trials have evaluated the use of SLN surgery for patients who present with initially node-positive disease and undergo neoadjuvant chemotherapy.
FNRs vary in the trials. But the goal for any supplemental technique or patient selection criteria is the same — to “drive the [FNR] rate lower,” Dr Boughey said.
Perfection is unlikely. “With every sampling procedure, there is an inherent false-negative rate,” she said. For SLN biopsies, even in the absence of neoadjuvant chemotherapy, it is somewhere in the 7% to 10% range.
Both Dr Black and Dr Chu noted that findings from the SENTINA study — a multicenter trial of SNL biopsy before and after neoadjuvant chemotherapy — should be considered with caution (Lancet Oncol. 2013;14:609-618).
The trial found that the FNR for SLN biopsy in clinically involved lymph nodes (cN+) patients exceeded the 10% threshold.
Dr Chu observed that both SENTINA and Z1071 are landmark trials that were published about the same time. But in the SENTINA trial, the FNR was higher (14%) than in Z1071, he said, “even when the axilla was downstaged to cN– following neoadjuvant chemotherapy.”
Dr Boughey pointed out that patients with a single SLN resected were included in SENTINA but excluded in Z1071. “We know that the FNR decreases with the number of nodes resected, so in Z1071, we only included patients with two or more SLNs resected,” she explained.
If patients in SENTINA with a single SLN resected were excluded, the FNR would be 9.6%, she said.
The study was supported by grants from the National Cancer Institute. Dr Boughey, Dr Black, and Dr Chu have disclosed no relevant financial relationships. One of Dr Boughey’s coauthors reports financial ties with Galena Biopharma and Antigen Express.
J Clin Oncol. Published online February 2, 2015.