A genetic model of melanoma tumorigenesis based on allelic losses

Authors

  • Graeme J. Walker,

    1. Queensland Cancer Fund Research Unit, Joint Experimental Oncology Program, Queensland Institute of Medical Research, Herston, Australia
    Search for more papers by this author
  • Jane M. Palmer,

    1. Queensland Cancer Fund Research Unit, Joint Experimental Oncology Program, Queensland Institute of Medical Research, Herston, Australia
    Search for more papers by this author
  • Marilyn K. Walters,

    1. Queensland Cancer Fund Research Unit, Joint Experimental Oncology Program, Queensland Institute of Medical Research, Herston, Australia
    Search for more papers by this author
  • Dr. Nicholas K. Hayward

    Corresponding author
    1. Queensland Cancer Fund Research Unit, Joint Experimental Oncology Program, Queensland Institute of Medical Research, Herston, Australia
    • Queensland Institute of Medical Research, Herston, 4029, Australia
    Search for more papers by this author

Abstract

Previous karyotypic studies have indicated a possible series of non-random chromosomal events involved in progression of melanoma. We sought to verify and augment this model of melanocyte tumorigenesis by studying allelic deletions of markers mapping to these regions in 30 matched pairs of melanoma and constitutional DNA samples. Polymorphic loci on chromosomes 1, 7, 10, 11, 17, and 21 were analyzed and data combined with those previously obtained for chromosome arms 6q and 9p in the same series of tumours. The most frequent and earliest deletions were found on 9p (57%) and 10q (32%). With the exception of one case, no sample had loss of markers on another chromosome without concomitant loss of markers on 9p or 10q. Losses on 6q were also a frequent (31%) and early event whereas losses of loci on distal 1p (22%) or 11q (26%) occurred only in metastatic melanomas. A “background” rate (0–17%) of allele loss was seen on chromosomes 7, 17, and 21. These data strongly support the previous model based on karyotypic findings in melanocytic lesions. However, we have been able to further augment that model by delimiting the regions of loss on 10q, to that distal to D10S254, and on 1p, to between D1S243 and D1S160.

Ancillary