| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Department of Breast Surgery, Joyce Eisenberg Keefer Breast Center, John Wayne Cancer Institute, 2200 Santa Monica Boulevard, Santa Monica, California 90404
Correspondence: Address correspondence and reprint requests to: Armando E. Giuliano, MD; E-mail: giulianoa{at}jwci.org.
The proposed utility of a nomogram is to assist decision making for complete axillary lymph node dissection (CLND) in breast cancer patients who have a positive sentinel lymph node (SLN). According to the American Joint Committee on Cancer Cancer Staging Manual (sixth edition, 2002), a positive SLN is defined as tumor deposits >.2 mm. Isolated tumor cells, clusters smaller than .2 mm, are classified as N0. Of all patients with a positive SLN, about 30% to 50% will have additional non-SLNs that contain tumor cells detectable by hematoxylin and eosin.1 The size of these metastases is not well documented since only in the most recent update of the American Joint Committee on Cancer staging system has size been relevant. The standard of care is to perform CLND on all patients who have positive SLNs. Accurately predicting which patients with a positive SLN are the 50% to 70% with no additional nodal disease might spare them the morbidity of a CLND. However, the therapeutic impact of such management has not been determined. Small studies2 suggest that further dissection may not be necessary. Indeed, extrapolation from National Surgical Adjuvant Breast and Bowel Project B04 data might even suggest that axillary surgery itself may not impact survival.3 Van Zee et al.4 published an excellent nomogram for predicting the likelihood of additional nodal metastases in 2003. The Memorial Sloan-Kettering Cancer Center (MSKCC) nomogram used eight clinical and tumor characteristics that are readily available to most clinicians to predict the likelihood that additional positive nodes will be found if the patient undergoes CLND. These variables include tumor size, histological type, nuclear grade, presence of lymphovascular invasion, multifocality, estrogen receptor status, number of positive and negative SLNs, and the method of detection of the SLN metastases.
Lambert et al.5 applied the nomogram to 200 consecutive patients at the University of Texas M. D. Anderson Cancer Center to assess its performance. They found that the area under the receiver operating characteristic (ROC) curve was .71. The area under the ROC curve is an overall assessment of the test’s ability to predict accurately. ROC curves were developed in the 1950s in the study of how to interpret radio signals contaminated by noise. An ROC curve plots a test’s performance with varying thresholds of sensitivity and specificity. As sensitivity improves, specificity suffers. For fewer missed diagnoses, there are more false positives. A value of .5 represents no ability to predict, and a value of 1.0 represents perfect accuracy. ROC curves are being used more frequently to describe the accuracy of radiological tests. Surgeons are often less familiar with this kind of tool because they are trained to manage uncertainty through a less quantitative process known as surgical judgment. ROC curves are compared by measuring the area under the curve as a quantification of accuracy. A rough correlation of surgical judgment and the area under the ROC curve values in terms of reliability might coincide like this: .9 to 1.0, intuitive genius professor; .8 to .9, dedicated midcareer surgeon; .7 to .8, average chief resident; .6 to .7, disinterested intern; and .5 to .6, motivated passer-by. In their original description of the MSKCC nomogram, Van Zee et al.4 explained the concept of a value of .77 of the ROC as follows:
If we randomly select two women, of whom one has at least one positive non-SLN and the other has negative non-SLNs, there is a 77% chance that the nomogram will predict a higher probability for the positive woman. This is a scale that ranges from .5, which would be achieved by tossing a coin, to 1.0, which would require perfect ability to tell the positive woman from the negative one.4
The verified area under the ROC curve obtained by Lambert et al.5 was .71.
Lambert et al.5 also assessed whether performance could be improved by substituting touch imprint cytology (TIC) for frozen section. A total of 94 patients had TIC performed. The results of this substitution improved the area under the ROC curve to .74. Both frozen section and TIC are acceptable methods of intraoperative analysis. TIC requires special experience in cytopathology, which may not be available in all institutions.
The MSKCC nomogram was also studied by Kocsis et al.6 in 2004. They examined the results of 140 patients who had both positive SLNs and axillary dissection. The correlation between the predicted and observed proportions of patients was only .84, compared with the .97 reported by Van Zee et al.4 They could not validate the nomogram and recommended against using it unless it could be validated. Replicating the environment and surgical technique from one institution to another is difficult. A nomogram that may work under identical circumstances but cannot be translated to other settings has very limited utility.
The MSKCC nomogram was studied by Smidt et al.7 in the Netherlands. They studied 222 consecutive breast cancer patients with positive SLNs who then underwent CLND. The area under the ROC curve was .77. This correlated well with Van Zee’s findings, even though the populations differed substantially.
The most useful information that a nomogram could offer is identification of a group of patients with a less than 5% to 10% chance of positive non-SLNs. This group of patients would likely be the best candidates to forgo additional axillary procedures. Although many patients might not accept a 5% to 10% chance of incomplete resection, a 3% to 5% risk category is close to the range of reported incidence of false-negative SLN biopsy results. This is a risk level that the patient has demonstrated a willingness to accept because she has already proceeded with SLN biopsy, which, in large series, has nearly a 10% false-negative rate. Forty patients were predicted to have a 
10% incidence of non-SLN involvement, and these patients did have a 10% incidence of non-SLN metastases. The sensitivity in this group was 97%, and the specificity was 20%. The MSKCC nomogram may be useful for these patients provided that they are comfortable with the potential inaccuracy. Among the 14 patients who had TIC intraoperative analysis and a 10% predicted incidence of non-SLN metastases, the observed incidence was <10%.
Specht et al.8 recently published a comparison of nomogram versus expert clinician prediction. They presented 33 patients to 17 breast cancer specialists using the pathologic features of the tumor and the SLN characteristics. The area under the curve of the ROC of the nomogram was .72, and the clinicians achieved an area under the curve of only .54. They concluded that the nomogram outperforms the clinical experts but that clinicians were unlikely to change their surgical management on the basis of the nomogram results.
In the practice of breast oncology, surgical plans are based on a consensus achieved between the patient, the surgeon, and members of the multidisciplinary team. Despite surgeons’ recommendations and careful patient education, some patients and some surgeons have strongly held beliefs that will prevail over factors such as nomogram-generated risk calculations. In addition, patients have individual risk tolerance that varies widely, as do individual surgeons. Some patients drive 80 mph on the freeway, whereas others avoid cracks in the sidewalk. To some patients, a 3% to 5% risk is too high, and CLND should be performed. The trust and communication between a surgeon and her patient is central to devising a plan that delivers optimal outcome and satisfaction. A nomogram is unlikely in practice to supplement this decision-making process, but it forces surgeons to examine the value of an operation with significant morbidity and enables patients to contemplate risks and rewards.
Received for publication November 10, 2005. Accepted for publication November 16, 2005.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
