• © 2002 by American Society of Clinical Oncology

Imatinib GIST Keeps Finding New Indications: Successful Treatment of Dermatofibrosarcoma Protuberans by Targeted Inhibition of the Platelet-Derived Growth Factor Receptor

THE SUCCESS OF the tyrosine kinase inhibitor imatinib mesylate in chronic myeloid leukemia (CML) and gastrointestinal stromal tumors (GIST) has led to great optimism in the oncology community about the potential for molecularly targeted therapy of cancer. The activity of imatinib in these two diseases is explained by its activity against the Abl (for CML) and c-kit (for GIST) tyrosine kinases, which are constitutively activated in these tumors as a result of gene fusion or mutation, respectively.^1,2 Imatinib also inhibits the platelet-derived growth factor receptor (PDGFR) tyrosine kinase. Because PDGFR is expressed in a number of tumors such as small-cell lung cancer, prostate cancer, and glioblastoma as well as in the stromal and vascular compartments of many tumors, there is great interest in testing imatinib in these diseases.

The current issue of the Journal of Clinical Oncology reports a dramatic response to imatinib in a patient with recurrent, metastatic dermatofibrosarcoma protuberans (DFSP) that was not amenable to surgical resection.³ Within 2 weeks of beginning treatment, the patient had clear evidence of clinical benefit, including a dramatic reduction in uptake of fluorodeoxyglucose as measured by positron emission tomography (PET). Magnetic resonance imaging scans showed reduction in tumor size after several months of therapy, and the patient subsequently underwent surgical resection. Pathologic analysis of the surgical specimen showed no evidence of tumor cells, with only residual scar tissue.

Why did imatinib produce such a dramatic response in this patient? Patients with DFSP often contain chromosome translocations that target the PDGFB gene, which encodes the ligand for PDGFR. Most common is a t(17,22) translocation that generates a Col1/PDGF fusion gene, which is expressed by tumor cells and processed to mature platelet-derived growth factor (PDGF) ligand. Fusion to Col1 enhances PDGF action by allowing constitutive expression from the Col1 promoter and removal of negative regulatory sequences within the first exon of PDGF. Expression of Col1/PDGF leads to constitutive PDGFR kinase activation through autocrine stimulation and subsequent cellular transformation, analogous to the v-sis oncogene.^4-6 Cytogenetic analysis of tumor cells from the patient reported here failed to show the t(17,22) translocation, but fluorescence in situ hybridization studies provided evidence for a possible variant PDGFB translocation. This mechanism of PDGFR activation is distinct from that seen in a subset of patients with chronic myelomonocytic leukemia with t(5,12) translocations, who have a constitutively active Tel/PDGFR fusion, and also respond dramatically to imatinib therapy.^7,8

When applied to unique cases such as the one presented here, modern molecular biology tools can lead to critical new insights about the mechanism of drug action. Further evidence that the imatinib response in this patient truly occurred through inhibition of PDGFR might be obtained by more detailed mapping of the variant translocation breakpoint and by demonstrating enhanced activation of the PDGFR protein in the tumor cells. Although not available to community oncologists, the technologies to address these questions are readily available in cancer research laboratories at most academic medical centers. In this particular case, fluorescence in situ hybridization probes that distinguish between the 5′ and 3′ ends of the PDGFB genomic locus and one-sided polymerase chain reaction cloning could be used to define the structural details of the PDGFB fusion gene. Immunoblot and immunohistochemical analysis using antibodies directed against PDGFR and phosphorylated tyrosine residues on the receptor could demonstrate activation of the mutant receptor.

Will this single case report alter the future treatment of DFSP? Although it is obviously dangerous to speculate on the basis of results from one patient, we should recall that the successful application of imatinib in GIST began with a single case report.⁹ If we follow the current model for predicting sensitivity to imatinib (activation of the target tyrosine kinase by gene fusion or mutation), its role in treatment of DFSP will depend on the frequency of PDGF translocations in these tumors. Current estimates suggest these may occur in a significant fraction of cases, but this number could change when more precise molecular diagnostics are applied to this question.

It is also worth considering whether the success reported here might have broader implications for use of tyrosine kinase inhibitors in cancer. One potential conclusion is that fluorodeoxyglucose PET scans may be a tool for predicting clinical responses to kinase inhibitors. The PET scan responses in GIST patients whose tumors have c-kit receptor mutations are dramatic⁹ and might be explained by a specific effect of c-kit activation on glucose uptake. However, the fact that the PET scan of a DFSP tumor exposed to imatinib showed a similar response argues for a more general effect of tyrosine kinase receptor signaling on glucose uptake. If confirmed, this result could impact the testing of other kinase inhibitors in cancers, where the ability to document activation of the kinase drug target in tumor cells is more difficult. Recent studies show that the epidermal growth factor receptor inhibitor ZD-1839 can produce dramatic responses in approximately 6% of patients with advanced-stage lung cancer.¹⁰ These data indicate that a rare subset of lung cancer patients have epidermal growth factor receptor–dependent tumors, but we are unable to identify these patients in advance or early in their treatment. PET scans using fluorodeoxyglucose (or other probes) may be early markers of drug response.

A more controversial conclusion from this case report is that patients and oncologists should be encouraged to use imatinib in other off-label cancers, particularly because the risk of significant side effects is low. The logic behind the use of imatinib in the patient reported here was based on strong preclinical evidence of PDGF gene fusions in DFSP, as well as the proof of concept studies showing activity against similar tyrosine kinase gene fusions in CML and GIST. This logic is quite different from more speculative studies of imatinib in other cancers, such as small-cell lung cancer or prostate cancer, where c-kit or PDGFR or their ligands may be expressed but activating mutations in these receptors have not been observed. Although it is clear that off-label imatinib use is considered by many patients and oncologists, my personal view is that such treatment be conducted in the context of formal clinical studies. I would urge clinicians who observe clinical benefit in these trials to report their findings and collect tumor tissue (when possible) so that retrospective molecular investigations might be pursued. There may be other imatinib-sensitive tumors (or subsets of tumors) in addition to CML, GIST, chronic myelomonocytic leukemia, and DFSP where occult activation of the Abl, kit, or PDGFR tyrosine kinases have yet to be discovered. Imatinib may also have additional undiscovered kinase targets. Now that we have a powerful new class of drugs in the oncology arsenal, we must remain vigilant to fully exploit their potential.