Sarcomatoid renal cell carcinoma: Biologic behavior, prognosis, and response to combined surgical resection and immunotherapy

Purpose: Sarcomatoid variants of renal cell carcinoma (RCC) are aggressive tumors that respond poorly to immunotherapy. We report the outcomes of 31 p atients with sarcomatoid RCC treated with a combination of surgical resecti on and immunotherapy. Patients and Methods: Patients were identified from the database of the Uni versity of California Los Angeles Kidney Cancer Program. We retrospectively reviewed the cases of 31 consecutive patients in whom sarcomatoid RCC was diagnosed between 1990 and 1997. Clinical stage, sites of metastasis, patho logic stage, and type of immunotherapy were abstracted from the medical rec ords. The primary end point analyzed was overall survival, and a multivaria te analysis was performed to distinguish any factors conferring an improved survivorship. Results: Twenty-six percent of patients were male and 74% were female, and the median age was 59 years (range, 34 to 73 years). Length of follow-up ra nged from 2 to 77 months (mean, 21.4 months). Twenty-eight patients (84%) h ad known metastases at the time of radical nephrectomy (67% had lung metast ases and 40% had bone, 21% had liver, 33% had lymphatic, and 15% had brain metastases). Twenty-five patients (81%) received immunotherapy, including l ow-dose interleukin(IL)-2-based therapy (five patients), tumor-infiltrating lymphocyte-based therapy plus IL-2 (nine patients), high-dose IL-2-based t herapy (nine patients), dendritic cell vaccine-based therapy (one patient), and interferon alpha-based therapy alone (one patient). Two patients (6%) achieved complete responses (median duration, 46+ months) and five patients (15%) achieved partial responses (median duration, 36 months). One- and 2- year overall survival rates were 48% and 37%, respectively. Using a multiva riate analysis, age, sex, and percentage of sarcomatoid tumor (< or > 50%) did not significantly correlate with survival. Improved survival was found in patients receiving high-dose IL-2 therapy compared with patients treated with surgery alone or any other form of immunotherapy (P =.025). Adjusting for age, sex, and percentage of sarcomatoid tumor, the relative risk of de ath was 10.4 times higher in patients not receiving high-dose IL-2 therapy. Final pathologic T stage did not correlate significantly with outcome, but node-positive patients had a higher death rate per year of follow-up than did the rest of the population (1.26 v 0.76, Cox regression analysis). Conclusion: Surgical resection and high-dose IL-2-based immunotherapy may p lay a role in the treatment of sarcomatoid RCCs in select patients. (C) 199 9 by American Society of Clinical Oncology.