Abstract 5314
Background
The IMDC classification scheme for OS has been validated in 2L mRCC. Recently, we showed that two new prognostic factors, Time from first to second line (<, ≥ 1 year) and tumor burden (<, ≥100 mm), are independently associated with OS in the same setting. Here, we present a new classification scheme.
Methods
mRCC patients treated in 2L between January 2005 and December 2014 after initial first line clinical trials at Gustave Roussy Cancer Campus (GRCC) formed the discovery set. Patients from 2 phase III clinical trials from Pfizer database (PFIZERDB), AXIS (NCT00678392) and INTORSECT (NCT00474786), formed the external validation set. In addition to the IMDC predictors, the 2 new prognostic factors were tested using a multivariable Cox model with a backward selection procedure. The performance of the new GRCC model and the classification scheme derived from it, measuring by R2, c-index and calibration, was evaluated on the validation set and compared to MSKCC and IMDC.
Results
Two-hundred and twenty-one patients were included in GRCC cohort and 855 patients in PFIZERDB. Median OS was similar in the two datasets (16.8 [95%CI=12.9-21.7] and 15.3 [13.6-17.2] months, respectively). Time from first to second line and tumor burden confirmed their significant effect on OS with HR = 1.68 [1.23-2.31] and 1.43 [1.03-1.99]. The new classification, derived by counting the number of factors, allows categorizing patients into 4 risk groups: median OS from the start of 2L in the validation cohort was not reached (NE) (95% CI 24.9-NE) in the favorable risk group (n = 20, 4 deaths, 0 risk factor), 21.8 months (11.6–15.8) in the intermediate risk group (n = 367, 135 deaths, 1-2 risk factors), 12.7 months (11.0–15.8) in the low-poor risk group (n = 347, 211 deaths, 3-4 risk factors), and 5.5 months (4.7–6.4) in the high-poor risk group (n = 121, 105 deaths, ≥5 risk factors). Not surprisingly, the incorporation of factors that characterize post first line features and the new stratification allowed a better performance compared to previous models.
Conclusions
The GRCC classification derived from the model is a new reliable prognostic tool that can be applied to better predict OS in previously treated mRCC patients.
Clinical trial identification
Legal entity responsible for the study
Lisa Derosa.
Funding
Has not received any funding.
Editorial Acknowledgement
Disclosure
All authors have declared no conflicts of interest.