SAN FRANCISCO, CA USA (UroToday.com) - Dr. Jenny J. Ko presented her study utilizing the IMDC model to predict prognosis in patients undergoing second-line targeted therapy for metastatic RCC.
The IMDC model utilizes 6 prognostic factors: Karnofsky performance score (KPS) < 80% , diagnosis to treatment interval of less than 1 year, anemia, hypercalcemia, neutrophilia, and thrombocytosis, in assigning patients to favorable, intermediate or poor prognostic groups.
This study looked at 1 021 consecutive patients treated between 2004 and 2013 at 19 different centers who received second-line targeted therapy after discontinuing first-line targeted therapy. The primary endpoint was prediction of overall survival (OS) from the initiation of second-line therapy using the IMDC model. The cohort of patients was predominantly male (75%), with more than one metastatic site (82%), and with clear cell histology (88%). Eighty-five percent of the cohort underwent previous nephrectomy and 22% underwent prior immunotherapy. Sunitinib was the failed first-line therapy in 67%. Second-line therapies included sorafenib (29%), sunitinib (22%), everolimus (22%), and temsirolimus (12%), as well as pazopanib, interferon, bevacizumab and axitinib to lesser degrees.
Median time on second-line therapy was reported to be 3.9 months and OS from the time of initiation of second-line therapy was 12.5 months. Five of the 6 prognostic factors (excluding hypercalcemia) were shown to be independent predictors of worse OS, and Dr. Ko explained that hypercalcemia was not shown to be a significant predictor primarily because of the low number of patients (only 9% of the cohort) exhibiting hypercalcemia at time of initiation of second-line therapy. When combining the factors, patients in favorable-, intermediate-, and poor-risk categories demonstrated a median OS of 35.8 months, 16.6 months and 5.4 months, respectively.
Dr. Ko then sought to compare the IMDC model to the MSKCC nomogram in this setting. The MSKCC nomogram utilizes 3 of the prognostic factors included in the IMDC model (KPS, hypercalcemia, anemia). The IMDC model demonstrated a higher predictive ability, in comparison to the MSKCC nomogram, by concordance index (0.70 vs 0.66), likelihood ratio test, and reclassification calibration.
Dr. Ko stated that the strengths of the study included the multinational nature of the cohort, as well its improved applicability to the targeted therapy era. The limitations discussed included the retrospective nature of the study, with the potential for selection bias as well as the heterogenous data included, with some patients on- or off-trial, and patients treated in community and academic centers. She concluded that the IMDC model could be used in the second-line targeted therapy setting to predict prognosis, as well as the first-line setting, and in non-clear cell setting as has been previously been shown.
Highlights of a presentation by Jenny J. Ko, MD at the 2014 Genitourinary Cancers Symposium - January 30 - February 1, 2014 - San Francisco Marriott Marquis - San Francisco, California USA