Lack of efficacy of lymphoscintigraphy in detecting axillary lymph node metastases from breast cancer

Preoperative lymphoscintigraphy during lymphatic mapping for breast cancer: improved sentinel node imaging using subareolar injection of technetium 99m sulfur colloid

BACKGROUND

Preoperative lymphoscintigraphy has been recommended to confirm the successful uptake and direction of migration of radiotracer into sentinel nodes during lymphatic mapping for breast cancer. In addition, preoperative lymphatic mapping may provide a visually useful aid to the relative location of sentinel nodes within a nodal basin. One common method of breast lymphoscintigraphy involves injections of unfiltered technetium 99m sulfur colloid (Tc-99m-SC) directly into parenchymal tissues surrounding a tumor or biopsy cavity (IP injection). Because of the many imaging failures and prolonged imaging times of IP lymphoscintigraphy, the procedure has fallen into disfavor by oncologic surgeons. The purpose of this study is to document the increased success rate of preoperative breast lymphoscintigraphy using a new anatomic site of injection, the subareolar lymphatic plexus (SA injection).

STUDY DESIGN

In the 12 months between December 1, 1998, and December 29, 1999, 42 women with stage I and II breast cancer underwent preoperative lymphoscintigraphy by either the IP (n = 12, December 1998 to May 1999) or SA (n = 30, May 1999 to December 1999) route of injection. Both groups were injected with 1 mCi (37 MBq) of unfiltered Tc-99m-SC followed immediately by external gamma-camera imaging. The success rate for preoperative sentinel node imaging and the total imaging time were recorded in both groups.

RESULTS

The success rate of identifying a sentinel node by SA lymphoscintigraphy was 90% (n = 27 of 30 patients), compared with 50% (n = 6 of 12 patients) for IP lymphoscintigraphy (p = 0.009). The imaging time in the SA injection group was 34 +/- 16 minutes, which was 59% shorter than the imaging time in the IP injection group of 82 +/- 48 minutes (p < 0.001). No uptake into internal mammary nodes was seen in either group.

CONCLUSIONS

Moving the site of injection ofunfiltered Tc-99m-SC to the subareolar lymphatic plexus (SA injection) increased the success rate of preoperative lymphoscintigraphy to 90%, compared with 50% using IP injections. Preoperative SA lymphoscintigraphy resulted in the rapid visualization of axillary sentinel nodes within 30 minutes of SA injection, enabling a visual determination of the approximate number of sentinel nodes and their relative locations within the axilla. We conclude SA injection of unfiltered Tc-99m-SC is superior to IP injections when performing preoperative breast lymphoscintigraphy and is a visually useful aid to lymphatic mapping for breast cancer.

The detection of axillary lymph node metastases from breast cancer by radiolabelled monoclonal antibodies: a prospective study

In a prospective study to assess the accuracy of monoclonal immunoscintigraphy for the detection of axillary lymph node metastases in breast cancer, two murine monoclonal antibodies that react with human breast cancer (3E1.2 and RCC-1) were labelled with 131iodine, and the radiolabelled antibody was injected subcutaneously into the interdigital spaces of both hands of 40 patients, 36 of whom had breast cancer and the remaining four of whom had fibroadenoma (the normal, contralateral axilla was used as a control). Of the patients with breast cancer, the findings from the scintigraphy images were correlated with histopathology or cytology of the axillary lymph nodes; images were regarded as positive and hence indicative of lymph node metastases if the amount of background-subtracted radioactive count in axilla on the side of breast cancer exceeded the contralateral normal side by a ratio greater than or equal to 1.5:1.0 as assessed by computer analysis. Using this method, immunoscintigraphy had an overall sensitivity of 33% (23% with 131I-3E1.2 and 50% with 131I-RCC-1) for the detection of lymph node metastases and a specificity of 63% (67% with 131I-3E1.2 and 60% with 131I-RCC-1) with problems of non-specific uptake by presumably normal lymph nodes. The results of immunoscintigraphy obtained with 131I-RCC-1 (IgG) were superior to 131I-3E1.2 (IgM) although the accuracy of immunoscintigraphy using 131I-RCC-1 (56%) was not much better than preoperative clinical assessment (50%). However, there were cases when immunoscintigraphy using radiolabelled antibody (IgM or IgG) detected axillary lymph node metastases not suspected by clinical examination. Thus it appears that while immunoscintigraphy may be a useful adjunct to preoperative clinical assessment and is simple and safe, a major improvement in its accuracy is needed before it can replace axillary dissection and histological examination in the accurate staging of axilla in breast cancer.