Patients with metastatic renal cell carcinoma who benefit from axitinib dose titration: Analysis from a randomised, double-blind phase II study

Background: A prospective, randomised phase II study demonstrated clinical benefit of axitinib dose titration in a subset of treatment-naïve patients treated with axitinib for metastatic renal cell carcinoma. This analysis evaluated patient baseline characteristics that may impact overall survival (OS) with axitinib dose titration. Methods: Following a 4-week lead-in period during which all patients received axitinib 5 mg twice-daily (bid); patients meeting the predefined randomisation criteria were randomly assigned to receive axitinib 5 mg bid plus either axitinib or placebo titration. In exploratory analyses, patients were grouped into those who achieved OS ≥24 versus < 24 months, and compared their baseline characteristics with Fisher's exact test or Cochran-Armitage trend exact test, with a 5% significance level. Potential predictive baseline characteristics associated with effect of axitinib dose titration on OS were investigated using a Cox proportional hazard model. Results: Overall, 112 patients were randomised. Three of 56 patients receiving axitinib titration were censored; of the remaining 53, 33 (62%) achieved OS ≥24 months versus 20 (38%) with OS < 24 months. Patients with OS ≥24 vs. < 24 months, respectively, had significantly fewer metastatic sites (≤2 metastases: 52% vs. 10%; ≥3 metastases: 48% vs. 90%), fewer lymph node (45% vs. 75%) or liver (15% vs. 45%) metastases, higher haemoglobin level (i.e., ≥ lower limit of normal: 67% vs. 25%) at baseline, lower neutrophil (≤ upper limit of normal, 97% vs. 75%) and platelet (≤ upper limit of normal, 82% vs. 50%) levels at baseline and ≥ 1 year between histopathological diagnosis and treatment (64% vs. 15%). The primary reason for treatment discontinuation in both OS groups was disease progression. The frequency of toxicity-related discontinuation was comparable between the 2 groups, indicating that it was not a factor for a shorter OS. The multivariate analysis showed that ≥1 year from histopathological diagnosis to treatment and baseline haemoglobin level equal or greater than lower limit of normal were significant covariates associated with favourable OS in patients receiving axitinib titration. Conclusions: The current analyses identified potentially predictive factors that could help selecting patients who may benefit from axitinib dose titration.