Imatinib

Imatinib mesylate is a tyrosine kinase inhibitor that targets c-KIT, PDGFR-A and B, and BCR-ABL. Imatinib’s first success in a solid tumor was in another connective tissue malignancy, gastrointestinal stromal tumor (GIST). Prompted by the tumor tissue expression and activation of PDGFR-B in chordoma and a lack of other standard options, Casali et al. treated six patients with advanced chordoma using imatinib at a dose of 800 mg daily. Although there were no objective responses based on RECIST, they noted changes in chordoma characteristics on imaging (liquefaction, decreased contrast enhancement and decreased glucose uptake) consistent with tumor response to treatment. Patients also reported decrease in tumor-related symptoms, predominately pain, while undergoing treatment. This study suggested that although patients derived benefit from imatinib, response was not quantifiable by standard criteria, a finding similar to that in patients with GIST treated with imatinib, which prompted development of the Choi criteria.

Studies by Tamborini and colleagues expanded upon the initial investigation of PDGFR-B in chordoma by evaluating PDGFR-A, PDGFR-B, and c-KIT protein expression in 31 chordoma tumors by immunoprecipitation, Western blot analysis and immunohistochemistry. PDGFR-B was found to be highly expressed and phosphorylated. PDGFR-A and c-KIT were also activated but were expressed at much lower levels. No activating or point mutations were identified in the tyrosine kinase receptors, supporting the concept of activation through autocrine/paracrine secretion of receptor ligand. This work led to further investigation of imatinib in patients with advanced PDGF-B and/or PDGFR-B-positive chordomas. The primary endpoint for the phase II nonrandomized study of was overall response rate (ORR) by RECIST. Patients were treated with imatinib 400 mg twice daily. Only one partial response was observed in 51 evaluable patients. A reduction in tumor uptake of FDG assessed by positron emission tomography (PET) was seen in 39% of patients at 3 months, suggesting that traditional RECIST criteria may not be the optimal considerations for detecting tumor response to small molecule inhibitors of receptor kinases. Tumor-related pain, assessed by a patient-reported pain inventory survey, improved during treatment with imatinib. The median duration of PFS was 9.2 months, and median overall survival (OS) was 34 months. While the median PFS seems clinically favorable compared to other reports of metastatic chordoma patients, the indolent nature of the disease and the lack of documented tumor progression prior to enrollment in the study may have significantly influenced the results. The true benefit of imatinib in this patient population is challenging to discern in the setting of single-arm, nonrandomized, clinical trials in which PFS or OS are primary study end-points. A follow-up retrospective analysis of 48 patients with PDGF-B/PDGFR-B positive advanced chordoma treated with imatinib 800 mg per day echoed the results of the aforementioned study. Of 46 patients evaluable for response, the best response was stable disease in 34 patients (74%) and progressive disease in 12 patients (26%). Seventy percent of the patients required reduction in dose of imatinib (600 mg per day in 19 patients and 400 mg per day in 14 patients) because of adverse drug effects. The median PFS was 9.9 months, and the median OS was 30 months. These findings add to the growing data that imatinib may help stabilize disease growth in locally advanced and metastatic chordoma but does not result in significant regression of tumor.

Imatinib Combination Strategies

Based on limited but potentially real signals of activity of imatinib in chordoma, combination strategies with imatinib have been investigated. Metronomic cyclosphosphamide and imatinib were evaluated in a phase I trial in chordoma and other targeted solid tumor types. The combination was felt to be tolerated at a dose of cyclophosphamide 50 mg twice daily and imatinib 400 mg twice daily, and among the seven chordoma patients treated, three patients were free from progression at 12 months with a median PFS that was not reached at time of study report.

Evidence of activation of the mTOR pathway in chordoma patients led to a study combining sirolimus, an mTOR inhibitor, with imatinib in imatinib-resistant advanced chordoma. Ten patients with progressive, advanced chordoma who had biochemical and/or immunohistochemical evidence of mTOR activation were begun on imatinib 400 mg daily and sirolimus 2 mg daily. Nine patients were evaluable and one discontinued treatment early. By RECIST, there was one partial response and three minor responses. Seven of the nine patients demonstrated response by both Choi criteria and FDG-PET, with 100% concordance. Eight patients had stable disease for at least 6 months and four patients had stable disease for at least 12 months. The most commonly observed adverse events during treatment were mucositis, infection, neutropenia and anemia. Analysis of pretreatment tumors revealed AKT activation in six patients and phosphylation of S6 ribosomal protein in seven patients. Biopsy of a chordoma obtained during treatment in a patient with an objective tumor response demonstrated that S6 was expressed but not phosphorylated, suggesting a direct inhibitory effect from sirolimus. These promising data led to development of a phase II study of imatinib and everolimus (an inhibitor of mTOR) in patients with advanced chordoma in Italy. To our knowledge, results of the study have not been reported or published. Aside from imatinib, other multi-targeted tyrosine kinase inhibitors, including sunitinib, sorafenib and dasatinib, have been studied in the treatment of chordoma.

Sunitinib

Targets of sunitinib include VEGFR, PDGFR and c-KIT, which are expressed in chordoma. A phase II open-label, nonrandomized study of sunitinib 37.5 mg daily in patients with sarcoma enrolled six chordoma patients. 44% of patients had stable disease for at least 16 weeks. Although not a prespecified endpoint, qualitative decreases in tumor density were demonstrated, akin to other studies with molecularly targeted agents.

Sorafenib

Sorafenib inhibits activation of VEGFR, PDGFR-B, and c-KIT, suggesting that it may have efficacy against chordoma. A nonrandomized, open-label phase II study evaluated the efficacy of sorafenib 800 mg daily in patients with advanced chordoma. The primary endpoint was 9-month PFS. 27 patients enrolled; 12 had previously been treated and 14 had metastatic disease. The objective response rate was 3.7%, the rate of 9-month PFS was 73% (confidence interval 46.1%–88%) and the rate of 12-month OS was 86.5% (confidence interval 55.8–96.5). Unlike those in the recent imatinib studies, the patients in this trial were not preselected based on expression of sorafenib targets in chordoma tissue samples.

Dasatinib

Dasatinib is a small molecule inhibitor of PDGFR, c-KIT, BCR-ABL, FAK and SRC. A recent study of dasatinib in advanced sarcoma patients enrolled 32 patients with chordoma. The primary endpoint of this study was 6-month PFS by Choi criteria, which limits direct comparison with studies using RECIST. Patients were treated with dasatinib 70 mg twice daily. One-third of the patients required reduction of dasatinib dose because of adverse events. The most common severe adverse events reported include pain, dyspnea, pleural effusion and diarrhea. The objective response rate was 19%, higher than that reported for other drugs using RECIST to assess tumor response. One of the six patients with response per Choi criteria had a response by RECIST. The rate of 6-month PFS was 54%, and the median PFS duration was 6.3 months. More than 14% of the patients were without tumor progression for more than 2 years, and six patients survived at least 5 years after enrollment in the trial.

Anti-EGFR Therapy

EGFR also has been investigated for its potential role in the pathogenesis of chordoma. In a tissue study of 173 chordoma samples from 160 patients, EGFR expression was detected in 69% of the samples and EGFR amplification was found in about 40% of the samples, although no activating mutations in EGFR were present in a subset of samples analyzed. In another study of 156 chordoma samples from 52 patients, EGFR expression was present in 83% of the samples, and higher EGFR expression correlated with higher expression of PDGFR and a worse prognosis. Deregulation of EGFR expression in chordoma may be caused by loss of specific microRNAs that have a tumor suppressor function.

Preclinical work using the EGFR inhibitor tyrphostin in a chordoma cell line demonstrated dose-dependent decrease in EGFR phosphorylation and a resultant inhibition of cell proliferation. Treatment of a human chordoma–derived cell line with the EGFR inhibitors, erlotinib and gefitinib, in vitro inhibited cell proliferation, and treatment of mouse xenografts transplanted with the cell line using erlotinib resulted in significant reduction in tumor growth. In addition, in a focused compound screen in chordoma cell lines, 21 of the 27 inhibitory compounds identified were EGFR/ERBB family inhibitors. These findings suggest that aberrant EGFR signaling may play a role in chordoma growth and that anti-EGFR therapy may be a reasonable treatment strategy. There have been several case reports of use of anti-EGFR agents including cetuximab, erlotinib and gefitinib in patients with chordoma. The one patient Hof et al. treated with cetuximab and gefitinib for an EGFR-expressing primary sacral chordoma and pulmonary metastases experienced a partial response over 9 months. Other reports suggest response to erlotinib after development of imatinib-resistance. One patient with a sacral chordoma responded for 11 months, and another had a partial and sustained response lasting longer than 28 months. A patient with a thoracic chordoma had a 70% reduction in disease with response lasting 12 months.

Erlotinib has also been used in combination with other targeted agents. Asklund et al. evaluated erlotinib in combination with the anti-VEGF therapy bevacizumab in three patients with metastatic chordoma. Patients were without disease progression for 2–4.5 years at the time of last report. Another patient treated on a phase I study of erlotinib combined with the insulin-like growth factor-1 receptor inhibitor, OSI-906, experienced prolonged stable disease for more than 5 years, suggesting that combination anti-EGFR strategies may hold promise.

Lapatinib is a tyrosine kinase inhibitor with activity against both EGFR and HER2/neu. It has been investigated in a phase II nonrandomized study of patients with EGFR-positive advanced chordoma. 18 patients were enrolled and received lapatinib 1500 mg daily. The primary endpoint was response rate by Choi criteria. Eleven patients had at least one delay in treatment, and lapatinib dose was reduced in four patients because of drug toxicity. Major adverse events during treatment were anemia, fatigue, rash, hypertension and thromboembolism. Six patients (33%) had a PR and seven had SD. There were no responses by RECIST. The median duration of PFS was 6 months by Choi criteria and 8 months by RECIST, similar to that in other trials evaluating molecularly targeted agents.

Interpretation of the benefit or efficacy of molecularly targeted agents in treatment of chordoma from currently available data is challenging. Preclinical data have demonstrated the presence and activation of different kinases in chordoma samples, but whether the kinases are the primary or relevant drivers of tumor growth and survival in patients remains to be established. It is extremely difficult to estimate the likelihood of treatment benefit in a larger population of chordoma patients based on single patient case reports and small case series. Single-arm phase II studies provide preliminary evidence of drug activity in chordoma, but when the primary effect of treatment is control of tumor growth rather than cell death and tumor regression, the natural variability in disease characteristics and tumor growth among patients precludes comparison across trials. Proof of drug efficacy in the treatment of chordoma will require randomized, comparative, controlled trials and should be pursued to establish appropriate standards of care for patients with locally advanced or metastatic chordoma.