Northwestern University Feinberg School of Medicine

Institute for Public Health and Medicine

Sentinel Lymph Node Positivity Rates for Breast Cancer and Melanoma: an Opportunity to Improve Quality and Enhance the NCCN Guidelines

May 24, 2013

Sentinel lymph node biopsy (SLNB) has become the standard of care for breast cancer and melanoma as described in the NCCN GuidelinesTM, but the Guidelines for SLNB have been changing.  SLNB provides critical prognostic information.  Importantly, SLNB allows selective use of completion lymph node dissection (CLND) as patients without evidence of metastases in their sentinel lymph nodes (SLNs) do not require a CLND.  Rather only patients with positive SLNs are considered for CLND. The information gained from SLNB guides additional surgical and adjuvant therapy.  The paradigm of CLND for SLN-positive disease has been challenged for both breast cancer and melanoma, and these changes are clearly being reflected in the NCCN Guidelines. Recently reported trials for breast cancer and ongoing trials for melanoma have suggested that CLND can be omitted in selected patients with positive SLNs. The decision to select patients for observation generally is based on tumor burden in the SLN and number of additional negative SLNs.  With the new treatment approach allowing selected patients with positive SLNs to forego CLND, the SLNB becomes even more important.  The proportion of patients with positive SLNs ranges from 15-20% for breast and melanoma, and the percentage is often used as a litmus test to assess the validity of SLNB studies.  Evidence suggests that there is a learning curve and volume-outcome association for SLNB.  Consequently, it has been suggested that some patients may not have all of their SLNs removed, thus potentially leaving disease in a regional lymph node.  Moreover, the quality of the pathologic examination is also an important factor in detecting microscopically positive lymph nodes and estimating the amount of tumor burden in the SLNs.  There is significant hospital-level variation in the quality of cancer care delivered in the U.S. for breast cancer and melanoma.  It has been suggested that a hospital’s SLN positivity rate may be a quality indicator, and when found to be lower than expected compared to other hospitals, it may be an indication of poor performance and offer an opportunity for targeted quality improvement.  Some of the variability in SLN positivity rates for melanoma comes from hospital-to-hospital differences in patient and tumor characteristics as institutions have different practices regarding which melanoma patients undergo a SLNB. NCCN guidelines frequently suggest that SLNB be "considered" or "discussed," so it is expected that providers have differences in how they recommend SLNB to patients. This could affect hospital SLN positivity rates, and must be considered in the statistical analyses of hospital SLN positivity rates.  More importantly, predicted SLNB positivity rates could be used to develop an empiric decision aide to enhance the NCCN Guidelines and provide more objective information about when to offer a SLNB in borderline candidates. 

SPECIFIC AIM 1. To examine variability in adjusted sentinel lymph node positivity rates for hospitals in order to identify hospitals with lower rates than expected for breast cancer and melanoma.  We hypothesize that some hospitals will have statistically and clinically significantly lower SLN positivity rates than expected after adjusting for patient and primary tumor factors; and provider and hospital characteristics are associated with lower SLN positivity rates.  We also hypothesize that the patients treated at hospitals with low SLN positivity rates will have worse 5-year survival rates, after adjusting for differences in patient and tumor factors. 

SPECIFIC AIM 2. To use predicted patient SLN positivity rates to develop a decision aide to facilitate patient and clinician decision making regarding when to perform a sentinel lymph node biopsy for melanoma. We hypothesize that a decision tool can enhance the NCCN Guidelines by providing an empiric method to help decide when to perform a SLNB.

The recently reported American College of Surgeons Oncology Group (ACOSOG) Z0011 prospective trial randomized breast conservation therapy patients with positive SLNs to observation vs. completion axillary dissection.  Although debated, the results suggest that SLNB may be omitted in selected breast cancer patients with microscopic nodal disease.  Similarly in melanoma, the prospective Multicenter Selective Lymphadenectomy Trial II (MSLT-2) is randomizing patients with a positive SLN to observation vs. CLND. 

SPECIFIC AIM 1.  Adherence with SLNB Guidelines for Breast Cancer and Melanoma Our prior work has focused on various aspects of national utilization of SLNB for breast cancer and melanoma.  We examined use of completion axillary lymph node dissection for breast cancer in patients with SLN positive disease. We found that there was a remarkable increase in avoiding completion axillary dissection in patients with microscopic SLN metastases from 1998 to 2005 (24.7% to 45.3%, P<0.001).  Moreover, we found that there was no difference in survival.  We have also examined national SLN biopsy utilization in melanoma in the U.S.  Only 50% of patients with Stage IB-II melanoma underwent a SLNB, and 13% of patients with Stage IA had a SLNB.  Appropriate use of SLNB was associated with treatment at an NCCN or NCI-designated cancer center.  Overall, we found that adherence with NCCN Guidelines is often less than ideal, and additional efforts are needed to improve the quality of breast cancer and melanoma nodal evaluation. Development of Cancer Quality Metrics to Improve Guideline Adherence One approach to quality improvement has been to develop quality indicators which allow assessment of provider adherence with guidelines.  We formally developed quality indicators for melanoma. We then used the National Cancer Data Base to assess the quality of melanoma care using these metrics.  Patients were more likely to receive recommended care if they were treated at a NCCN center. Hospitals can use these measures to assess the quality of melanoma care provided, but there is a clear need for more cancer quality metrics.  Subsequently, Quan and colleagues developed quality metrics in a similar fashion for SLNB for breast cancer. One of the metrics was the SLN positivity rate.  Importantly, no further evaluation of this quality metric has been performed.     Important Considerations in Measuring Hospital Performance One other quality metric involves ensuring that at least 12 lymph nodes are resected and examined to accurately stage colon cancer.  Similar to the proposal in this study, we have previously focused on the methodology of this colon cancer quality metric, and we published the first national hospital report card.  In examining nearly 1,300 hospitals, we found that compliance with the metric was highest at NCCN and NCI centers, but 60% of all hospitals failed to achieve the benchmark.

SPECIFIC AIM 2.  For melanoma, the NCCN Guidelines indicate that SLNB should be "discussed" or "considered."  Evidence to guide the Guidelines is lacking particularly for thin melanomas as the threshold for the risk of having a positive SLN is not established and varies based on surgeon practice and patient preferences.  The indications for SLNB continue to be debated.  Our collaborators at M.D. Anderson found that the overall SLN positivity rate was 17% but this varied from 4% for T1 lesions to 46% for T4 lesions.  Our prior work has involved developing patient-specific risk assessment tools to help guide patient- and physician decision making. We have used decision analysis and cost effectiveness models recently to assess pancreatic cancer surgery and neoadjuvant therapy for pancreatic cancer.

Center: Healthcare Studies

Funder: National Comprehensive Cancer Network (NCCN)

Principal Investigator:
Karl Bilimoria