Premium
Tumor thickness as a prognostic factor in extramammary Paget's disease
Author(s) -
Ito Takamichi,
Kaku Yumiko,
Nagae Konosuke,
NakanoNakamura Misa,
Nakahara Takeshi,
Oda Yoshinao,
Hagihara Akihito,
Furue Masutaka,
Uchi Hiroshi
Publication year - 2015
Publication title -
the journal of dermatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.9
H-Index - 65
eISSN - 1346-8138
pISSN - 0385-2407
DOI - 10.1111/1346-8138.12764
Subject(s) - medicine , hazard ratio , extramammary paget's disease , confidence interval , univariate analysis , multivariate analysis , carcinoembryonic antigen , univariate , lymph , lymph node , oncology , pathology , radiology , cancer , disease , multivariate statistics , statistics , mathematics
Extramammary Paget's disease ( EMPD ) is a rare tumor and a widely accepted classification system specific for the disease has not been established. To elucidate prognostic factors of EMPD , we conducted a retrospective review of 145 patients with 155 EMPD lesions and investigated clinicopathological factors using univariate and multivariate analyses. We also explored tumor thickness and metastatic lymph nodes using detection analysis to determine cut‐off points for survival. All patients were Japanese (88 men and 57 women), with EMPD in the genital (82.8%), perianal (3.4%) and axillary regions (1.4%). In the remaining cases (12.4%), there were lesions at two or more regions. Univariate analysis revealed the following prognostic factors: perianal location, presence of nodules, invasion depth, tumor thickness, number of metastatic nodes and serum carcinoembryonic antigen level. Both tumor thickness and perianal location retained statistical significance in multivariate analysis (hazard ratio, 1.39; 95% confidence interval, 1.12–1.72; P = 0.0024; hazard ratio, 50.72; 95% confidence interval, 4.20–612.63; P = 0.0020; respectively). The signal detection analysis indicated tumor thickness of more than 3 mm and three or more metastatic lymph nodes as cut‐off points for survival. In conclusion, tumor thickness and the number of metastatic lymph nodes closely correlated with patient outcome and these factors could be suitable for the tumor and node classification.