The study aimed to explore the prognostic value of platelet distribution width (PDW) in patients with nonmetastatic renal cell carcinoma (RCC).
We retrospective analyzed 706 patents with nonmetastatic RCC from January 2015 to December 2017. Clinicopathologic data and platelet indices were collected and analyzed by univariable and multivariable cox proportional hazard model. Progression-free survival (PFS) was analyzed using the Kaplan–Meier curve. Net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were performed to evaluate the improvement of predictive accuracy.
Patients were divided into low PDW (N = 241, PDW ≤11.7%), intermediate PDW (N = 232, 11.7%< PDW ≤15.6%), and high PDW (N = 233, PDW >15.6%) groups according to the tertiles. Patients with low PDW were associated with more symptoms at presentation, larger tumor size, higher AJCC tumor stage, and more sarcomatoid differentiation. Besides, patients with low PDW had significantly shorter PFS compared to intermediate PDW and high PDW groups. On the multivariable model, AJCC tumor stage, nuclear grade, and PDW (either continuous or categorical variables) were independent factors correlated with PFS. The NRI and IDI showed adding PDW to SSIGN score improves its predictive accuracy related to 2-, 3-, and 4-year PFS.
Low PDW was related to advanced clinicopathologic features and worse prognosis in patients with nonmetastatic RCC. Thus, PDW could serve as a novel biomarker for risk stratification in these patients when used pre-or postoperatively.