Project Title:

Comparative study of targeted sentinel node biopsy versus axillary sampling in clinically node negative breast cancer after neo-adjuvant chemotherapy

1.    OBJECTIVES:

1. To compare low axillary sampling with targeted sentinel node biopsy (Blue dye with or without radio colloid) in a clinical trial setting to assess its ability to predict the rest of axillary nodal status in women with a clinically node negative breast cancer after neo-adjuvant chemotherapy.

2. Comparison of respective false negative rates and sensitivity as the immediate end points.

Inclusion criteria:

1. Confirmed locally advanced or large operable breast cancer

2. No prior excision biopsy

3. Clinically node negative breast cancer after neo-adjuvant chemotherapy

Sample Size (Recalculated):

Results from 200 patients: At presentation, the tumors were large (median 5.0 cm) with 70% clinically palpable nodes. SNB was defined as blue &/or hot node plus palpable node(s). A blue or hot node (median 2 nodes) was identified in 93.8%, and median of 5 sentinel nodes including palpable nodes were removed. The false negative rate of SNB was 15.3% (95% CI 8.7-25.3). The LAS technique comparatively had nodal yield in 98.5% with median 8 nodes removed; and FNR 8.5% (95% CI, 4.2-16.6, p=0.19). Comparative NPV for LAS and SNB were 94.6% and 91.8% respectively

The confidence intervals (CI) for false negative rate (FNR) for SNB and LAS overlap suggesting that the two procedures are comparable, although the CI is very wide indicating a small sample size. A larger and adequate sample size will therefore be necessary to prove beyond doubt the equivalence of the two procedures. The observed FNR of SNB in early breast cancer (reference standard) was 9.8% in the large randomized trial NASBP 32*. Keeping FNR of 10% as acceptable and comparable, the sample size was recalculated for LAS/SNB in post NACT status. We would then require 750 patients for a FNR of 10% with an upper limit of 95%CI of 12 (95% CI 0-12) with adequate power to prove equivalence beyond doubt.

End Points:

1.    Number of nodes identified by sentinel node biopsy and axillary sampling

2. Number of times the sentinel node is found within the sampled nodes
3. Sensitivity in predicting status of rest of axillary nodes

4.    False negative rate by both methods

5.    Negative predictive value for both methods

Methodology

Eligible women will be recruited in the study after obtaining an informed consent.

The radio labeled Tc-99 labeled nanocolloid colloid (particle size 200nm) will be injected over the primary tumor (subdermal and intraparenchymal) 2 hours prior to surgery. A localized scintiscan will then be performed to confirm the radio-labeling of the sentinel node before surgery and for documentation. Isosulphan blue dye will be injected subdermal (0.5ml) over the tumor 10-15 min before incision. Axillary sampling will be performed first. A 2 cm incision will be made in the middle third of the proposed axillary clearance incision below the axillary hairline. All axillary fat and tissue in an area of 2cm diameter will be dissected out. The medial limit of dissection is the posterior border of pectoralis major, the lateral limit of dissection is the anterior border of latissimus dorsi muscle, upper limit of dissection is the intercostobrachial nerve, and base is serratus anterior muscle on lateral chest wall. In those cases that undergo a mastectomy, the tissue in the axillary tail will be identified first and then removed as specified above. After completion of sampling procedure, the remaining axillary tissue will be checked for any other node in remaining axilla (to be labeled separately) showing a blue discoloration or radioactivity and the same will be documented as found outside of axillary sampling. Axillary clearance will then be separately completed in all patients by standard technique after extending the incision without waiting for the frozen section report.

Documentation of nodes:

All nodes will be sent to the frozen section laboratory. The sampled nodes will be grossed and all lymph nodes found will be serially labeled starting from ‘a’, ‘b’…e.t.c. Each node will also be individually documented if blue stained (as “blue node”, and radioactivity (as “radioactive node”), and other nodes. Blue node and or radioactive node found outside the axillary sampling specimen will also be serially numbered and documented.

Nodes more than 1cm in diameter will be sliced into two halves. One half will be kept aside for paraffin section and immunohistochemistry. The other half will be processed for frozen sectioning, and sections stained by toluidene blue and haemotoxylin-eosin. The half node kept aside for paraffin section will be embedded sections will be cut after making blocks. These sections will be stained routinely and if negative immunohistochemistry will be done to detect micrometastasis. Nodes less than 1cm size will be submitted for analysis of single sections each in frozen and after embedding for paraffin sections.

Statistical analysis

Analysis will be carried out using the chi-square test and sensitivity, specificity, accuracy statistical tests to compare the mean number of nodes identified. The false negative rates and sensitivity will be compared. Nodes found by sentinel node biopsy using blue dye and radio-guided biopsy will be compared with axillary sampling for predictability of rest of axilla. Also the chance of finding the sentinel node within the sampled nodes will also be calculated.