Skip to main navigation Skip to search Skip to main content

The development of optimal pathological assessment of sentinel lymph nodes for melanoma

  • Martin G. Cook
  • , Margaret A. Green
  • , Brian Anderson
  • , Alexander M.M. Eggermont
  • , Dirk J. Ruiter
  • , Alain Spatz
  • , Mark W. Kissin
  • , Barry W.E.M. Powell

Research output: Contribution to journalArticlepeer-review

176 Citations (Scopus)

Abstract

1158 sentinel lymph nodes (SLNs), excised from patients with primary cutaneous melanoma, were assessed pathologically using histology with immunohistochemistry (IHC) on all nodes, and RT-PCR for Mart-1 and tyrosinase on 55 nodes. RT-PCR was compared with the histology and IHC assessed on the same nodes. The evaluation of progressively more detailed protocols for histology and IHC modulated by the RT-PCR results led to a procedure that consistently detects metastases in 34% of patients submitted to SLN biopsy for cutaneous melanomas with a vertical growth phase and a mean thickness of 2.02 mm (range 0.25, with regression, to 19 mm). As this technique is virtually free of false positives and produces only a marginally lower detection rate than RT-PCR, which was subject to false positives of 7% in our study, it is suggested that this extended protocol should be the basis on which further evaluation of the place of RT-PCR in SLN assessment takes place. The evolved protocol described here has been adopted by the EORTC as the standard procedure for pathological handling of sentinel lymph nodes for melanoma when SLN status is a criterion in their clinical trials or studies.

Original languageEnglish
Pages (from-to)314-319
Number of pages6
JournalJournal of Pathology
Volume200
Issue number3
DOIs
Publication statusPublished - 1 Jul 2003
Externally publishedYes

Keywords

  • Immunohistochemistry
  • Melanoma
  • RT-PCR
  • Sentinel lymph nodes

Fingerprint

Dive into the research topics of 'The development of optimal pathological assessment of sentinel lymph nodes for melanoma'. Together they form a unique fingerprint.

Cite this