Atypical fibroxanthoma is a controversial entity with a disputed histogenesis. It has recently been suggested that most atypical fibroxanthoma are actually variants of squamous cell carcinoma. We reviewed 100 purported cases of atypical fibroxanthoma received over 4 years to perform clinical follow up and immunohistochemical markers. In particular, we focused on the detection of any recurrence or metastasis. Ten cases were subsequently excluded on the basis of either incorrect coding, or insufficient or absent paraffin blocks on file. A further case was interpreted as a malignant fibrous histiocytoma. Additional new markers, such as CD10 and CD99, were employed in a proportion of cases. Our cases were typical of the usual clinical presentation of atypical fibroxanthoma on the skin of the sun-damaged elderly. We found no cases of recurrent or metastatic atypical fibroxanthoma. Two patients developed a second primary atypical fibroxanthoma. CD10 proved to be a useful marker for atypical fibroxanthoma when used on 20 cases in the present study, as was CD99 in seven cases. The only case demonstrating positive staining for keratin also stained for CD10. It had dual features of atypical fibroxanthoma and squamous cell carcinoma. However, as the majority of atypical fibroxanthoma had no adjacent solar keratosis, our data suggest it is unlikely that atypical fibroxanthoma is a variant of squamous cell carcinoma.