FDG-PET Staging of the Axilla in Breast Cancer
FDG-PET Staging of the Axilla in Breast Cancer
Abstract & Commentary
Synopsis: The majority of patients with early breast cancer are free of nodal metastases at the time of axillary dissection. Sentinel node biopsy, a minimally invasive procedure, has decreased the morbidity associated with axillary dissection in instances where the sentinel node is not involved with cancer. However, PET scanning may prove to be an entirely noninvasive alternative to facilitate identification of patients who can safely be spared any axillary surgical staging. This study from the NCI Milan concluded that PET scanning was a reliable and accurate means of diagnosing the uninvolved axilla.
Source: Greco M, et al. J Natl Cancer Inst. 2001;93:
630-635.
One hundred and sixty-seven women
awaiting axillary dissection after mastectomy or quadrantectomy for T1-2 breast cancer were studied with positron emission tomography (PET) imaging at the National Cancer Institute (NCI) Milan. The outcomes were then correlated with the histologic findings following Level I-III axillary dissection, and determinations were made regarding the accuracy of PET scanning for staging of the axilla. The mean patient age was 54 years (range, 28-84), and the mean primary tumor diameter was 2.1 cm (range, 0.5-5). A mean of 23 lymph nodes was recovered from the study group (range, 9-49). The breast and the axilla were in the study field. Three nuclear-medicine physicians concurred on the status of each PET scan, ie, positive or negative for disease in the axilla. PET scanning was performed after 5 hours of fasting to minimize serum insulin levels. Lymph nodes were sectioned into 2 or 3 parts and stained with hematoxylin and eosin (H&E).
Overall, 72 patients (43%) were found to have involved axillary lymph nodes at the time of surgical dissection. Twenty-three of the 98 patients with T1 disease (23%) had positive nodes, while 49 of 69 (71%) patients with T2 primary lesions had nodal metastases. Only 4 patients with histologic evidence of disease had false-negative PET results. Two of these patients had tumor emboli in their nodes, and 2 had partial nodal involvement. Among patients with positive PET scans and confirmed nodal metastases, there was a range of patterns including micrometastases, single microembolic foci, and partial nodal replacement. No patients were noted to have activity in the supraclavicular region on PET imaging, but Greco and colleagues reported that some patients had uptake in the retrosternal area suggestive of internal mammary metastases.
For PET, the sensitivity, or the number of patients with proven nodal metastases divided by the number of patients with a true-positive PET scan + the number of patients with a false-negative PET scan, was 94.4%. The specificity, or number of patients with histologically negative nodes divided by the number of patients with a true-negative PET scan + the number with a false-positive PET scan, was 86.3%. The overall positive predictive value of PET was 84%, and the overall negative predictive value was 95.3%. There was no significant difference in PET performance between patients with palpable vs. nonpalpable lymph nodes.
Greco et al concluded that PET’s high-negative predictive value, as reflected by its low false-negative rate, was a landmark finding which reliably, accurately, and safely permits the identification of those patients who can avoid surgical staging of the axilla. The diagnostic accuracy of PET was similar across patients with all sizes of primary lesions.
COMMENT BY EDWARD J. KAPLAN, MD
Surgical management of the axilla is primarily a prognostic maneuver, and is not generally felt to be therapeutic by design. The approach to the axilla is in a state of transition based on the rising popularity of the sentinel node biopsy procedure. If a sentinel node is found to be free of tumor, the patient is spared a more comprehensive axillary dissection. However, if the sentinel node is involved with metastases, a Level I and II dissection is still considered to be the standard of care. The American College of Surgeons Oncology Group trial, where patients with a positive sentinel node are randomized to axillary dissection vs. no further treatment, is being conducted to determine whether axillary dissection offers a therapeutic benefit.
The study by Greco et al from NCI Milan presents intriguing data on the potential use of PET scanning in selecting outpatients who can forego an axillary dissection. However, important details about the histology of their patients’ primary lesions were omitted. This is of interest since it has been shown that infiltrating lobular lesions are more apt to result in false-negative scans.1 It would also have been interesting to know whether any of the false-negative PET scans of the axilla correlated with lack of 2-fluoro-2-deoxyglucose (FDG) uptake in the primary breast lesions, infiltrating lobular or otherwise. Greco et al stated that previous breast biopsies do not affect PET performance. But, unfortunately, their study design called for PET imaging postmastectomy or quadrantectomy.
Some of the claims made by Greco et al are not substantiated by their data. For example, they stated that PET provides staging information with respect to the supraclavicular and internal mammary nodal chains, as well as all 3 axillary levels, and that it can quantify metastatic involvement. In fact, none of their patients exhibited uptake in the supraclavicular region, so any comments would be speculative. Among those patients that exhibited uptake in the retrosternal area, none had a dissection there, so no conclusions can be drawn. Furthermore, no information was provided about what levels the involved axillary lymph nodes were taken from, or whether PET imaging could distinguish nodes by level or number. Greco et al also stated that quality of life is greatly improved based on the noninvasive nature of PET scanning, but no objective support for that statement was provided. Finally, it is difficult to believe that 4 patients with mere microembolic foci in their nodes had true-positive PET scans, when others write that PET scan resolution is limited to lesions greater than 1 centimeter in diameter.2,3
The role of PET-scan staging in the management of the breast cancer patient will largely depend on whether axillary dissections are ultimately determined to be diagnostic or therapeutic. Since the false-negative rate for PET-scan evaluation of the axilla seems to match that of sentinel node biopsies, PET scanning could become the diagnostic method of choice. If axillary dissection is associated with a therapeutic advantage, then it is doubtful that PET staging will become widely accepted. Meanwhile, the Health Care Financing Administration is considering whether to approve Medicare reimbursement for PET-scan staging of breast cancer.
References
1. Avril N, et al. J Clin Oncol. 2000;18:3495-3502.
2. Strasberg SM, et al. Ann Surg. 2001;233:293-299.
3. Brennan M. Ann Surg. 2001;233:320-321.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.