Prognostic S ignificance o f O ccult M etastases D etected b y Sentinel L ymphadenectomy a nd R everse Transcriptase-Polymerase C hain R eaction i n E arly-Stage Melanoma P atients

Purpose: Detection of micrometastases in the regional tumor-draining lymph nodes is critical for accurate staging and prognosis in melanoma patients. We hypothesized that a multiple-mRNA marker (MM) reverse transcriptase‐polymerase chain reaction (RT-PCR) assay would improve the detection of occult metastases in the sentinel node (SN), compared with hematoxylin and eosin (H&E) staining and immunohistochemistry (IHC), and that MM expression is predictive of disease relapse. Patients and Methods: Seventy-two consecutive patients with clinical early-stage melanoma underwent sentinel lymphadenectomy (SLND). Their SNs were serially sectioned and assessed for MAGE-3, MART-1, and tyrosinase mRNA expression by RT-PCR, in parallel with H&E staining and IHC, for melanoma metastases. MM expression in the SNs was correlated with H&E and IHC assay results, standard prognostic factors, and diseasefree survival. Results: In 17 patients with H&E- and/or IHC-positive SNs, 16 (94%) expressed two or more mRNA markers. Twenty (36%) of 55 patients with histopathologically negative SNs expressed two or more mRNA markers. By multivariate analysis, patients at increased risk of metastases to the SN had thicker lesions (P 5 .03), were 60 years of age or younger (P F .05), and/or were MM-positive (P F .001). Patients with histopathologically melanoma-free SNs who were MM-positive, compared with those who were positive for one or fewer mRNA markers, were at increased risk of recurrence (P 5 .02). Patients who were MM-positive with histopathologically proven metastases in the SN were at greatest risk of disease relapse (P 5 .01). Conclusion: H&E staining and IHC underestimate the true incidence of melanoma metastases. MM expression in the SN more accurately reflects melanoma micrometastases and is also a more powerful predictor of disease relapse than are H&E staining and IHC alone. J Clin Oncol 17:3238-3244. r 1999 by American Society of Clinical Oncology.