Inter-observer variation in the histopathology reports of head and neck melanoma; a comparison between the seventh and eighth edition of the AJCC staging system.

BACKGROUND

TNM staging of melanoma has recently been altered by the introduction of the 8th edition of the AJCC Cancer Staging manual. The purpose of this study is to analyze the inter-observer variation of histopathology reports and its effect on recommended treatment policy.

METHODS

We retrospectively analyzed 296 cases, diagnosed as primary cutaneous head and neck melanoma (2005-2016), referred to the Netherlands Cancer Institute (NCI) for treatment after prior diagnosis in another hospital (non-NCI). All reports were analyzed for patients demographics, tumor characteristics and histopathologic features.

RESULTS

In 53% and 40% of the cases, the histopathologic parameters were discordant, according to AJCC 7th and 8th edition, respectively. This indicated a perfect inter-observer agreement for the measurement of Breslow thickness (Intraclass correlation coefficient (ICC) = 0.981) and a substantial agreement for subtype (kappa statistic (κ) = 0.648) and ulceration (κ = 0.802), while only moderate for dermal mitotic activity (κ = 0.472). After NCI review, recommended treatment policies were changed in 13% and 11% of the patients when applying TNM 7 and TNM 8, respectively. Scheduling sentinel lymph node biopsy (SLNB) changed in 14 (5%) and 10 (3%) cases when using TNM 7 and TNM 8, respectively.

CONCLUSION

Review by a NCI pathologist of histopathologic parameters of primary cutaneous head and neck melanoma led to significant changes in treatment decision. Introduction of the AJCC 8th edition led to slightly less discordances between NCI and non-NCI reports and consequently smaller impact on treatment planning. Expert review remains indicated when a SLNB is considered for additional staging in selected cases.