Background: We studied the patterns of recurrence of patients with only reverse transcriptase-polymerase chain reaction (RT-PCR) evidence of regional nodal spread to see whether or not proposed treatment interventions are likely to be effective.
Study design: One hundred seventy-five patients who underwent selective lymphadenectomy for clinical stage I and II melanomas were included in this analysis. We preserved a portion of each sentinel lymph node (SLN) in liquid nitrogen in the operating room and performed RT-PCR on the specimens to detect the melanoma/melanocyte-specific marker tyrosinase. We then compared the pattern of recurrence (regional dermal metastases, regional nodal recurrence, or distant metastatic spread) of the patients with histologically positive SLNs to that of patients who had histologically negative SLNs.
Results: The mean followup time of the 175 patients was 33.83 months (SD = 15.94, median = 34.17, maximum = 62.95, minimum = 6.21). Thirty-four patients had at least one histologically positive SLN, and 17 of these patients had a recurrence (50%). Of the 141 patients that had histologically negative SLNs, 73 had SLNs that were also negative for tyrosinase by RT-PCR, and none of these patients had a recurrence. Of the 68 patients that had histologically negative but RT-PCR-positive SLNs, 14 had a recurrence (20.6%).
Conclusions: Because the pattern of recurrence of patients with only RT-PCR evidence of melanoma in SLNs was identical to that in patients who had histologically evident melanoma in the SLN and underwent subsequent completion lymphadenectomy, we conclude that completion lymphadenectomy might be ineffective in decreasing the recurrence rate of patients with only RT-PCR evidence of melanoma in SLNs.