A multicenter study has found that the experimental targeted therapy everolimus (Cetican, RAD001) delays cancer progression in patients with metastatic kidney cancer that has progressed despite treatment with other targeted therapies. Lead author Robert J. Motzer, md, attending physician at Memorial Sloan-Kettering Cancer Center, presented the results at the ASCO meeting (abstract LBA5026).
A multicenter study has found that the experimental targeted therapy everolimus (Cetican, RAD001) delays cancer progression in patients with metastatic kidney cancer that has progressed despite treatment with other targeted therapies. Lead author Robert J. Motzer, md, attending physician at Memorial Sloan-Kettering Cancer Center, presented the results at the ASCO meeting (abstract LBA5026).
Everolimus is administered orally and targets the mTOR protein, which regulates cell division and blood vessel growth in cancer cells. It is being evaluated for the treatment of several other cancers and is currently approved as an immunosuppressant to prevent the rejection of organs after transplant.
“This study has given us a new and clearly useful tool for treating renal cell tumors. It’s an important step forward for patients living with this disease,” said Dr. Motzer. “In the future, kidney cancer is likely to be managed as a chronic disease with treatments including this one.”
Study Specifics
All patients in this study had the most common type of kidney cancer, renal cell carcinoma with a clear cell component. They also had been previously treated with sunitinib (Sutent) and/or sorafenib (Nexavar), which target a different receptor commonly found in kidney tumors (vascular endothelial growth factor, or VEGF), but had stopped responding.
Patients were randomized to receive everolimus (272 patients) or placebo (138 patients), plus best supportive care (which can include measures such as palliative radiation therapy). After 6 months, 26% of patients in the everolimus group had not progressed, vs only 2% in the placebo group. The difference in median progression-free survival (the average time it took for the cancer to get worse) was 4 months for everolimus, vs 1.9 months for placebo.
The most common side effects for patients receiving everolimus were mouth ulcers (36%, vs 7% in the placebo group), anemia (28% vs 15%) and weakness (28% vs 20%). The instances of severe (grade 3 or 4) toxicity were 5% or lower for each side effect.