[39] Utility of p16 in Distinguishing Lipomas and Well-Differentiated Liposarcomas Compared to Fluorescence In-Situ Hybridization

Raul S Gonzalez, Colt M McClain, Ben K Chamberlain, Cheryl M Coffin, Justin M Cates. Vanderbilt University Medical Center, Nashville, TN

Background: Fluorescence In-Situ Hybridization (FISH) for amplification of MDM2 is considered the gold standard assay for the diagnosis of well-differentiated liposarcoma (WDL). It is especially useful in lipoma-like WDL, in which cytologic atypia may be focal. Immunohistochemistry (IHC) for p16 has recently been suggested as a sensitive and specific method for distinguishing these two entities, but it has not been validated on cases in which the diagnosis was confirmed by FISH analysis.
Design: A series of 19 lipoma-like WDL and 21 lipomas in which the diagnosis had been previously confirmed by FISH analysis were stained for p16 by IHC. Nuclear p16 staining was scored according to previously published methods: 0% of cells staining was recorded as negative, 1-10% as focal, 11%-50% as multifocal, and >50% as diffuse. The presence of strong, diffuse cytoplasmic p16 staining was also recorded. Regions of fat necrosis were excluded from evaluation.
Results: Nuclear p16 staining was significantly increased in WDL compared to lipoma (Mann-Whitney, P=<0.0001). Of the 19 WDL, 17 (89%) showed at least multifocal strong staining. In contrast, 18 of 21 lipomas (86%) showed no more than focal strong staining for p16; 3 cases showed multifocal strong staining. ROC analysis indicated maximum diagnostic accuracy (88%) was achieved using multifocal strong staining as the threshold value for defining a positive IHC result. Diagnostic performance parameters using this criterion for nuclear p16 in the diagnosis of WDL are presented in the table below.

Sensitivity (95% CI)0.90 (0.67-0.99)
Specificity (95% CI)0.86 (0.64-0.97)
Positive likelihood ratio (95% CI)6.26 (2.17-18.10)
Negative likelihood ratio (95% CI)0.12 (0.03-0.46)


Although cytoplasmic staining was specific for WDL (95%), it lacked diagnostic sensitivity (53%).
Conclusions: Multifocal or diffuse strong nuclear staining for p16 in a well-differentiated lipomatous neoplasm is highly suggestive of malignancy. Strong nuclear staining limited to ≤10% of adipocytes may be seen routinely in lipoma. Although nuclear staining for p16 has been described in lipomas with fat necrosis, we report for the first time multifocal strong p16 staining of adipocyte nuclei within viable regions of lipoma (3/21 cases). While there was not complete concordance (0.88; 95% CI 0.73-0.96) between FISH results and nuclear p16 staining in WDL, the IHC assay performs well. For situations where turn-around time and/or cost are priorities, IHC for p16 may offer an alternative to FISH in the differential diagnosis of well-differentiated lipomatous neoplasms.
Category: Bone & Soft Tissue

Tuesday, March 5, 2013 1:00 PM

Poster Session IV # 22, Tuesday Afternoon

 

Close Window