PREEMPTIVE LYMPHADENECTOMY MARKEDLY IMPROVES SURVIVAL IN PATIENTS WITH CANCER OF THE PENIS WHO HARBOR OCCULT METASTASES

Abstract

Kroonet al have chan ged their approach based onthe studies published inthis issue of The Journ al. They have shownthat the 3-year disease specific survival for patien ts operated onwith a wait an d see philosophy is 35% versus 84% for those operated on with clinically nonpalpable groin nodes who have microscopic metastatic disease as determined by sentinel node biopsy. The authors further demonstrate that in a multivariate analysis the only predictor of survival is early lymphadenectomy (what we would call preemptive lymphadenectomy). They further strengthen their case by pointing out that in the delayed group at the time of dissection there was an increased number of positive nodes and an increased incidence of extracapsular spread compared to the immediate lymphadenectomy group: both extremely poor prognostic indicators for long-term survival. This well done study has sufficient power to confirm the necessity of removing regional nodes while still microscopically involved rather than waiting until they become grossly enlarged. Another issue of controversy addressed by these authors is whether canon e predict who is likely to have microscopic nodal disease when the groins have no palpable adenopathy. The authors contend that the method we have proposed in the past, namely depth of invasion and cellular differentiation,2 is too imprecise, thereby subjecting too many individuals to a needless groin dissection. On the other hand, they advocate dynamic sentinel node biopsy. As with all methods, this too is not perfect. If one looks carefully at their data in 10% of the groins explored (20 of 223) the test did not work. In17% of the patients the test falsely predicted anabsence of microscopic disease. Since 2% of the patients were lost to followup and since some were followed for as little as 5 months, it is highly likely that with more complete and longer followup the false-negative rate will be higher. However, with this test since only groins with a positive sentinel node are subjected to radical dissection, the patient is assured that he will not be subjected needlessly to an operation. This is at the expense of missing at least 20% of patients whose survival is limited by a wait and see approach. If we use depth and grade of the primary lesion as predictors of groin disease, one rarely misses a patient with microscopic metastatic disease but at the expense of operating needlessly onabout 20% of in dividuals. Fortunately there is a new modality for imaging lymph nodes using magnetic resonance imaging and ultra small particles of ironoxide, which we have found to be much more accurate than either of the aforementioned methods. In our series this methodology has 100% sensitivity, 97% specificity, 100% negative predictive value and 75% positive predictive value. These rates indicate that one is highly unlikely to miss the positive groin but will end up operating on a few patients who have no metastatic disease. However, these results need