Researchers seek improved system to more accurately identify melanoma risk levels

May 1, 2007

Philadelphia - According to a new study, the use of the Surveillance, Epidemiology and End-Results (SEER)-based classification system can help identify high-risk patients with thin cutaneous melanomas and thus improve clinical decision-making, Reuters Health reports.

Philadelphia - According to a new study, the use of the Surveillance, Epidemiology and End-Results (SEER)-based classification system can help identify high-risk patients with thin cutaneous melanomas and thus improve clinical decision-making, Reuters Health reports.

The study, conducted by researchers at the University of Pennsylvania School of Medicine and published in the March issue of the Journal of Clinical Oncology, sought to establish a more accurate classification scheme than the American Joint Commission on Cancer’s current system that doctors use to identify patients with high- and low-risk melanomas.

The researchers studied 26,291 patients with thin melanomas from the SEER cancer registry (1988 to 2001) and 2,389 patients seen by the University of Pennsylvania’s Pigmented Lesion Group (PLG; 1972 to 2001). AJCC prognostic factors included thickness, level, ulceration, anatomic site, sex and age. The PLG prognostic factors included biologically based prognostic factors such as radial growth phase (RGP) and vertical growth phase (VGP), RGP regression, VGP mitogenicity, and the presence or absence of tumor-infiltrating lymphocytes (TIL).

The authors used a recursively applied algorithm to develop a SEER-based classification tree that was validated using PLG data. They then used the expanded set of prognostic factors to develop a new PLG-based classification “tree.”

The SEER-based classification tree identified additional criteria to explain survival heterogeneity among melanoma patents with thin, non-ulcerated lesions. Ten-year survival rates ranged from 89.1 percent to 99 percent. The new PLG-based classification identified groups using level, tumor-cell mitotic rate and sex. Survival rates ranged from 83.4 percent to 100 percent. The PLG-based tree had better discrimination.

The study concludes that the expanded classification using standard patient and tumor characteristics can be employed to help doctors offer therapy options tailored to individual patients’ risk profiles.

“When verified, the ‘new’ classification will further improve decisions about therapy and follow-up,” the study’s authors write. “We think that measures of dividing tumor cells should become part of the next version of the AJCC’s classification system.”