Using patient and survival data from the IMvigor210 trial, the authors performed univariate and multivariate Cox regression analyzes to identify prognostic factors for OS. A stepwise selection approach was used to identify an optimal model. The model was then validated using patient and survival data obtained from the PCD4869g phase I clinic trial, consisting of 95 pretreated metastatic bladder cancer patients who also received atezolizumab. The model consisted of 6-factors: Neutrophil to Lymphocyte Ratio (NLR) ≥ 5, ECOG-PS ≥ 1, Platelet level ≥ 400 109/L, Hgh < 10 g/dL, LDH level ≥ 280 U/L and presence of liver metastases. The C-Index for the training dataset was 0.690 and 0.759 for the validation set. Interestingly, tumor PD-L1 level was not associated with OS in either the training or validation sets. On survival analysis, increasing number of factors was associated with worsening overall survival in both the discovery (0-1: 19.6; 2-3: 5.9; 4+: 2.8 months) and training sets (0-1: 19.4; 2-3: 7.2; 4+: 2.6 months).
In summary, a prognostic model for OS has been introduced for patients with advanced urothelial carcinoma receiving post-platinum atezolizumab. The model consists of 6-factors which include a measure of performance status, the presence of liver metastases along with other hematological factors. The model appears promising in predicting OS in both the training and validation sets. Further validation in larger cohorts and in those treated with other checkpoint inhibitors will be required prior to its implementation into clinical practice.
Presented by: Speaker: Gregory R. Pond, Ph.D., McMaster University, Hamilton, Ontario Canada
Written by: Andres F. Correa, MD, Fox Chase Cancer Center-TempleHealth, Philadelphia, PA, at the 2018 American Society of Clinical Oncology Genitourinary (ASCO GU) Cancers Symposium, February 8-10, 2018 - San Francisco, CA