Spindle cell mesenchymal neoplasms represent a diverse group of benign and malignant tumors, the diagnosis of which relies on histomorphological features supported by ancillary investigations which include immunohistochemistry and, increasingly, molecular analysis. Microscopically solitary fibrous tumors are characterised by hypocellular collagen rich areas alternating with a proliferation of uniform elongated spindled cells in a haphazard distribution. Imunohistochemistry is extremely useful in establishing the diagnosis, no more so than CD34. CD34 is a myeloid progenitor cell antigen which is also positive in endothelial cells and some mesenchymal cells, including subsets of fibroblasts (3). It is no coincidence that since the description of CD34 expression in solitary fibrous tumors there has been a flurry of case reports in a wide range of sites. Cytogenetic and fluorescence in situ hybridization has shown no specific chromosomal abnormality (4) and, unlike a growing number of sarcomas, molecular tests are not utilised in confirming the diagnosis.

The pathological enigma surrounding solitary fibrous tumor is twofold. The first is identifying those tumors which have malignant potential and the second is the histological diagnosis of dedifferentiated solitary fibrous tumors. Most solitary fibrous tumors behave in a benign fashion. When arising from the pleura, 13-23% are classified as malignant in contrast to most extrapleural tumors which, with the exception of those of mediastinal origin, have a benign outcome (5). England et al used high cellularity, mitotic activity (more than four mitotic figures per 10 high-power fields), pleomorphism, hemorrhage and necrosis as criteria for distinguishing tumors with a favourable course from those that have the propensity for recurrence, local invasion and metastatic spread (6). Unfortunately biological behaviour does not always correlate with atypical histological features. De Perrot et al stratified the risk of recurrence based on histologic and morphologic indicators among 185 reported solitary fibrous tumors of the pleura (7). Recurrence was observed in 63% of all patients presenting with a malignant sessile lesion but 2% of the patients with a benign pedunculated tumor recurred. The proliferation marker Ki67 has been used to stratify lesions as to their clinical outcome. Positive staining is greater in malignant versus benign tumors but the overlap limits its usefulness (8). Clearly whilst the site, growth pattern and histological features correctly identify the malignant potential in the majority of cases, there still remains a small subset which behaves in an unpredictable fashion.

As sarcomas progress they may acquire additional molecular alterations which aids tumor progression. This is accompanied by transformation of the typical histological appearance to an anaplastic component, a process referred to as dedifferentiation, with frequent loss of CD34 expression (9). Frankly sarcomatous solitary fibrous tumors can only be recognized as such if they are associated with typical solitary fibrous tumors or recur at the site of a previous documented benign-looking solitary fibrous tumor (5). The largely favourable outcome of extrathoracic solitary fibrous tumors may be that only the typical ones are recognized and those showing cytomorphological atypia are diagnosed as something else.

In this issue of the Journal of Thoracic Disease, Milano et al investigate the survival of malignant solitary fibrous tumours of the pleura, lung and mediastinum in 82 patients identified from a population based dataset (10). Given the rarity of this tumor and the difficulties in identifying the malignant form, this approach has generated a large cohort of patients and provides the basis for understanding the progression of the disease and factors affecting outcome. In line with other studies, this paper confirms the positive impact of surgical excision on outcome. Univariate analysis did not support a role for adjuvant radiotherapy. The effect of chemotherapy was not analyzed but this is not surprising given that it is not effective and therefore rarely used (11).

The clinical management of malignant solitary fibrous tumors remains a dilemma, especially in cases were surgical excision is not feasible. Oncologist marvel in the success of gastrointestinal stromal tumors, a neoplasm, which like solitary fibrous tumor, is a spindle cell lesion which may behave in a benign or malignant fashion. However that is where the similarities stop. There are well defined morphological and histological parameters for defining risk of progression in gastrointestinal stromal tumors and specific molecular events have been identified which are exploited for therapy and prognosis. Indeed imatinib, a tyrosine kinase inhibitor, has revolutionised the management of patients with gastrointestinal stromal tumors. In an attempt to mirror this success, investigators have used imatinib in malignant solitary fibrous tumors overexpressing platelet-derived growth factor, both in vitro (12) and in vivo (13), with promising results.