Identification and Correlation of Cytologic Criteria to Histologic Grade in Canine Cutaneous Mast Cell Tumors
Mast cell tumors are the most common cutaneous tumors recognized in dogs. Histologic grade has been considered one of the most valuable factors in determining prognosis. Cytologic evaluation of cutaneous tumors is relatively inexpensive, requires no specialized equipment, is minimally invasive and associated with little to no patient morbidity. The aim of this study was to identify cytologic criteria of mast cell tumors that correlate to their histologic grade using the Patnaik scale.
This retrospective study identified cases of mast cell tumors (n=52) from computerized searches medical records and pathology reports. All cases were required to have at least one cytology slide of the tumor available for review. All samples were stained by clinical pathology laboratory personnel with Wright-Giemsa stain using a standard protocol. Cytologic criteria evaluated (n=14) included: granule staining properties, predominant nuclear shape, presence and prominence of nucleoli, nuclear to cytoplasmic ratio, cytoplasmic border, cellular granule percentage, multinucleated cells, binucleated cells, mitotic figures, nuclear size, red blood cells, neutrophils, eosinophils, and overall cell size. Cytologic evaluation was performed without knowledge of histologic grade. Kendal-Tau non-parametric analysis was performed on the initial data set and the value for statistical significance was set at p < 0.1. Logistic regression was performed on those cytologic criteria identified as significant from the initial statistical analysis.
Multiple slides were available in 39/52 cases, so 132 total slides were reviewed. Direct smear (n=80) and cytospin (n=52) samples were included. Histopathology results were available for 26/52 cases. Twenty of the 26 cases had multiple cytology slides, giving a total of 64 histologic grades.
Four cytologic factors were determined to be significantly related to histologic grade in the initial analysis: presence of multinucleated cells [Kendal-Tau value = 0.42, p-value = 0.04], mitotic index [Kendal-Tau value = 0.39, p-value = 0.06], large cell size [Kendal-Tau value = 0.31, p-value = 0.09], and nuclear shape [Kendal-Tau value = 0.45, p-value = 0.03]. Logistic regression was performed and 3 cytologic factors were determined to be statistically significant: presence of multinucleated cells (p-value = 0.02), large cell size (p-value = 0.07), and nuclear shape (p-value = 0.06). All statistically significant results were obtained from cytospin samples.
Prognostic information obtained from cytology samples could significantly decrease patient morbidity since surgical planning and adjunctive therapy could be more appropriately tailored to the patient.