Despite stricter criteria and important changes for stage III melanoma criteria, the new American Joint Committee on Cancer (AJCC) 8th Edition Melanoma Staging System compares well overall to the older 7th Edition in terms of prognostic and discriminatory ability for predicting patients’ melanoma-specific survival, according to findings from a prospective database analysis of AJCC stage III melanoma patients (abstract 9500) presented at the 2018 American Society of Clinical Oncology (ASCO) Annual Meeting, held June 1–5 in Chicago.
However, there are new complexities to bear in mind, cautioned lead study coauthor Max F. Madu, MD, of the Netherlands Cancer Institute.
“The 8th Edition holds up during external validation,” Madu reported. “There is similar prognostic accuracy in the 7th and 8th editions, and survival differentiation is comparable—though slightly worse in the 8th edition for stage IIIA vs IIIB.”
The AJCC is the most widely used and validated melanoma staging system.
“The introduction of effective adjuvant treatment options for patients with stage III and IV melanoma has created the need for more accurate prognostication and risk stratification," said Dr. Madu. “Now that effective adjuvant therapy has arrived in melanoma, accurate staging and patient selection to optimize a risk/benefit ratio is crucial.”
The 8th Edition is intended to improve risk stratification. Dr. Madu and colleagues conducted an external-validation analysis of the prognostic and discriminatory abilities of melanoma criteria in the 8th Edition, comparing its distributions of stage-specific melanoma-specific survival rates over time with those yielded by 7th edition criteria using data from the same 640 patients.
The new edition has four stage III subgroups, up from three in the previous edition, and new tumor-node-metastasis (TNM) combinations, he noted.
“There have been a lot of changes,” Madu said. “IIIA in the old edition can be IIIC in the 8th, which makes it hard to translate.” The new edition is also “slightly worse” at differentiating prognosis, he added.
Unlike the 7th edition, for which melanoma-specific survival rates were well-differentiated throughout follow-up for patients with stage IIIA, IIIB, and IIIC disease, Madu’s team found that using 8th Edition criteria, survival for patients with IIIA and IIIB melanoma overlap for the first 4 years after diagnosis.
Survival in stage IIIA melanoma “remains heterogeneous” in the 8th Edition, Madu noted. Stage III can be more precisely stratified using Rotterdam/ European Organisation for Research and Treatment of Cancer (EORTC) criteria for sentinel node tumor burden, he said. That system uses the maximum diameter of the largest metastatic deposit in the sentinel node, with a cutoff of 1 mm, which is strongly correlated with survival.
Dr. Madu emphasized that using both the AJCC (8th Edition) and EORTC SN tumor burden criteria is more prognostic than AJCC classification based on complete lymph node dissection, AJCC classification based on sentinel lymph node biopsy only, or ulceration and sentinel node tumor burden.
“Age, Breslow thickness, ulceration of the primary tumor, and number of positive lymph nodes were significant prognostic parameters,” Madu reported. “The 8th Edition performed similarly to the 7th in terms of survival discrimination, but failed to differentiate melanoma-specific survival between stage IIIA and IIIB after correction for sex and age,” he said.
Upgrading the 8th Edition for stage III disease is a “necessity” in the rapidly evolving treatment landscape for melanoma but translating some categories of disease between the 7th and 8th editions will be complicated at times, according to Madu.
The Rotterdam/EORTC tumor burden criteria can be “a powerful adjunct to AJCC staging, especially in the post-CLND [complete lymph node dissection] era,” to aid clinical decision making about adjuvant treatment, said Dr. Madu.