p53, p16 and ki67 as immunohistochemical prognostic markers in uterine smooth muscle tumors of uncertain malignant potential (STUMP)

The risk stratification in gynecologic smooth muscle tumors of uncertain malignant potential (STUMP) is a crucial issue, but at present there are no validated prognostic markers. We aimed to assess p53, p16 and ki67 as immunohistochemical prognostic markers in STUMP through a systematic review and meta-analysis. Electronic databases were searched from their inception to July 2020. All studies assessing p53, p16 and/or ki67 immunohistochemistry in gynecologic STUMP series were included. Immunohistochemical patterns were categorized as "abnormal" vs "wild-type" for p53, "diffuse" vs "focal/negative" for p16, >= 10% vs 10% for ki67. The prognostic value for recurrence was assessed through Cox regression analysis; a p-value 0.05 was considered significant. Markers that resulted significant were assessed for prognostic accuracy with calculation of area under the curve (AUC) and post-test probability of recurrence. Seven studies with 171 patients were included. Significant association with disease-free survival was found for p53 (p 0.0001) and p16 (p 0.0001), but not for ki67 (p = 0.911). p53 showed sensitivity= 83%, specificity= 86%, AUC= 0.89, and post-test recurrence probabilities of 54% and 7% in the case of abnormal and wild-type expression, respectively. p16 showed sensitivity= 84%, specificity= 88%, AUC= 0.91 and post-test recurrence probabilities of 56% and 7% in the case of diffuse and focal/negative expression, respectively. In conclusion, p53 and p16 might be useful in the risk assessment of STUMP, despite not being suitable as stand-alone prognostic markers.