Current Status of Sentinel Node Surgery in Breast Cancer

Krag and Harlow believe that sentinel lymph node biopsy for breast cancer is still an experimental procedure because "longterm randomized trials comparing the survival outcome of this procedure with that of conventional axillary node resection have not been completed." Specifically, they are awaiting the results of the American College of Surgeons Oncology Group (ACOSOG) Z0011 trial and the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-32 trial before they make a determination about the "safety" of this procedure.[1] This is a common misconception, because these studies will not prove whether sentinel node biopsy results in equivalent survival compared to standard level I/II axillary dissection in patients with clinically node-negative breast cancer. The time has come for clear thinking about the design of these trials and precise reasoning about the interpretation of their ultimate results. Contrasting Trial Designs
The ACOSOG Z0011 trial randomizes patients with positive sentinel lymph nodes to completion axillary dissection or no axillary dissection. It will determine whether axillary dissection in patients with positive nodes improves survival-a long-standing and worthwhile question regarding the therapeutic value of axillary dissection. It will not determine whether the sentinel node biopsy procedure results in reduced survival compared to standard axillary dissection. Implicit in the design of this trial is the notion that patients with negative sentinel nodes do not require any further axillary surgery. In contrast, the NSABP B-32 trial is based on the premise that axillary dissection in patients with negative nodes could improve survival. Krag and Harlow believe that the B-32 trial will determine whether "sentinel node surgery results in a survival rate as good as that of axillary node resection." However, the trial is designed as follows: Patients are randomized to sentinel node biopsy followed by completion axillary dissection (regardless of the sentinel node result) vs sentinel node biopsy with completion axillary dissection only in patients with positive sentinel nodes (Figure 1). It is important to note that NSABP B-32 patients with nodal metastases are treated similarly in both groups- ie, they both undergo sentinel node biopsy and axillary dissection. Accordingly, a survival difference between treatment arms among node-positive patients is impossible. Nonetheless, they are included in the randomization. Clearly then, any difference in survival must occur among the nodenegative patients. Axillary Dissection in Node-Negative Patients
One might reasonably ask, why would it be beneficial to perform axillary dissection in patients with negative nodes? Can there be any advantage to removing normal lymph nodes, and is it really possible that sufficient hypothesis-generating evidence exists to support the contradictory notions that axillary dissection can be avoided safely in patients with positive nodes (ACOSOG Z0011) but improves survival in patients with negative nodes (NSABP B-32)? Of course not. In fact, there are no convincing data to suggest that removing normal lymph nodes will improve survival.[2] Krag and Harlow suggest that the 10-year follow-up data from the landmark NSABP B-04 trial showed a 4% difference in survival favoring axillary dissection. Conveniently, they omit the 25-year follow-up data, which demonstrated that this supposed survival advantage completely disappears with longer follow-up.[3] Based on a meta-analysis of six studies, including the 10-year follow-up data from B-04, it has been suggested that there is a survival advantage of as much as 5.4% favoring axillary dissection.[4] This meta-analysis suffers from the significant limitation that B-04 is the only study that directly compared axillary dissection to no axillary dissection. In all of the other studies, the control group (no axillary dissection) comprised a hodgepodge of radiation treatments, including radiation therapy to the axillary, supraclavicular, and, in all cases, internal mammary nodes. It is not difficult to believe that radiation therapy, especially internal mammary treatment using techniques from the 1950s and 1960s, could result in cardiac toxicity significant enough to negatively affect survival.[5-7] Keep in mind that these studies were all performed in an era prior to modern systemic adjuvant therapy, which likely would diminish any potential therapeutic benefit of axillary dissection. Investigational Ironies
I believe that axillary dissection is an excellent procedure for regional disease control, and I do not discount the possibility that it imparts a small survival advantage to patients with positive nodes. But even if one believes that axillary dissection improves survival, does anyone truly believe that there would be any survival advantage if node-positive patients were excluded? How ironic that the NSABP- which has for years taught us that axillary lymph node dissection is a staging procedure with no impact on survival- is now sponsoring a trial based on the premise that axillary dissection not only improves survival, but improves survival in patients with negative lymph nodes. In truth, the only way there could be a legitimate survival difference between treatment arms in the NSABP B-32 study is for the sentinel node false-negative rate to be so high that it affects survival, but even that is impossible. Suppose that, of 100 patients, 33 have positive axillary nodes. Even with a false-negative rate of 15% (the unacceptably high end of the reported literature), that would mean that 15% of 33 patients, or 5 of 100 patients would be inaccurately staged as having negative axillary nodes. Even if we assume the 5.4% survival advantage for axillary dissection from the meta-analysis, that would equate to 5.4% of 5 patients, or an absolute increase in survival of 0.27% of our 100 patients. And even if axillary dissection in patients with positive nodes resulted in a survival advantage of as much as 25% (unlikely), this would only increase survival by 1.25%. NSABP B-32 is designed to detect a 2% survival difference. Appropriate Trial Design
If one wanted to prove that sentinel lymph node biopsy is equivalent to axillary node dissection in terms of survival, the trial design is obvious:
Randomize patients to standard level I/II axillary dissection vs sentinel node biopsy (with completion axillary dissection for patients with sentinel node metastases). In this way, the gold standard procedure-axillary dissection- would be compared to the procedure that seeks to replace it. Such a study design would determine, for example, whether increased mortality, recurrence, or other unknown adverse outcomes are related to the sentinel node procedure (eg, caused by tumor manipulation, radioactive tracers, allergic reaction to blue dye, or other factors beyond our present comprehension). However, the safety of sentinel node biopsy cannot be addressed by a trial (ie, NSABP B-32), in which all patients undergo sentinel node biopsy. It is equally likely that sentinel node biopsy could improve survival by detection of early nodal metastases that would otherwise be missed by routine pathologic analysis of the axillary dissection specimen. This more appropriate randomized study design has been employed in the ongoing Axillary Lymphatic Mapping Against Nodal Axillary Clearance (ALMANAC) trial.[8] However, the primary end points of that study are axillary morbidity, health economics, and quality of life. A secondary goal is to evaluate axillary recurrence. Specifically, this study is not designed or powered to evaluate overall survival. Impact of NSABP B-32
When the predictably negative results of NSABP B-32 are released, many who have not carefully considered these issues will proclaim that this study provides the final proof that sentinel node biopsy is safe and can be accepted as a standard procedure for axillary nodal staging. The study will provide much important information about the prognostic significance of nodal micrometastases and many other factors, and has been an excellent mechanism for surgical training in the context of a cooperative group trial. Contrary to what Drs. Krag and Harlow profess that NSABP B-32 will show, however, it will not tell us whether sentinel node biopsy reduces overall survival compared to standard axillary dissection. It will prove only what we already know-that there is no advantage to removing normal lymph nodes. Conclusions
After all, sentinel node biopsy is not designed to be a therapeutic procedure or to improve survival-it is a diagnostic staging test to determine the status of the axillary nodes. It is a less invasive alternative to axillary dissection. As with any other diagnostic test, the validity of sentinel node biopsy is not accomplished by randomization, but by performing nonrandomized validation studies to determine the sensitivity, specificity, positive- and negative- predictive values, overall accuracy, and false-negative rate.[9] The results of such studies have been reported for thousands of patients internationally.[10] Once surgeons have established that the procedure can be performed accurately, there is no reason not to offer sentinel node biopsy to women who are informed that they are trading a small chance of a false-negative result for a less invasive test. That is why sentinel node biopsy is no longer an experimental procedure in the hands of experienced surgeons and is currently practiced as a standard of care in thousands of centers around the world.[11]