Background: Intratumor heterogeneity (ITH) is a key feature in clear cell renal cell carcinomas (ccRCCs) that impacts outcomes such as aggressiveness, response to treatments, or recurrence. In particular, it may explain tumor relapse after surgery in clinically low-risk patients who did not benefit from adjuvant therapy. Recently, single-cell RNA sequencing (scRNA-seq) has emerged as a powerful tool to unravel expression ITH (eITH) and might enable better assessment of clinical outcomes in ccRCC. Objective: To explore eITH in ccRCC with a focus on malignant cells (MCs) and assess its relevance to improve prognosis for low-risk patients. Design, setting, and participants: We performed scRNA-seq on tumor samples from five untreated ccRCC patients ranging from pT1a to pT3b. Data were complemented with a published dataset composed of pairs of matched normal and ccRCC samples. Intervention: Radical or partial nephrectomy on untreated ccRCC patients. Outcome measurements and statistical analysis: Viability and cell type proportions were determined by flow cytometry. Following scRNA-seq, a functional analysis was performed and tumor progression trajectories were inferred. A deconvolution approach was applied on an external cohort, and Kaplan-Meier survival curves were estimated with respect to the prevalence of malignant clusters. Results and limitations: We analyzed 54 812 cells and identified 35 cell subpopulations. The eITH analysis revealed that each tumor contained various degrees of clonal diversity. The transcriptomic signatures of MCs in one particularly heterogeneous sample were used to design a deconvolution-based strategy that allowed the risk stratification of 310 low-risk ccRCC patients. Conclusions: We described eITH in ccRCCs, and used this information to establish significant cell population-based prognostic signatures and better discriminate ccRCC patients. This approach has the potential to improve the stratification of clinically low-risk patients and their therapeutic management. Patient summary: We sequenced the RNA content of individual cell subpopulations composed of clear cell renal cell carcinomas and identified specific malignant cells the genetic information of which can be used to predict tumor progression. Keywords: Clear cell renal cell carcinoma; Deconvolution; Low-risk progressors; Malignancy atlas; Single-cell RNA sequencing; Tumor progression.