Background: Renal cell carcinoma has several subtypes, with kidney renal clear cell carcinoma (KIRC) being the most common and heterogeneous. Purine metabolism is associated with cancer progression. However, the role of purine metabolism-related long non-coding RNAs (lncRNAs) in KIRC remains unknown. Methods: KIRC were grouped into Cluster-1 and Cluster-2 based on purine genes. Limma package was used to identify differentially expressed lncRNAs between two classes of purine genes. Single-factor screening was used followed by random forest dimensionality reduction and Lasso method to screen lncRNAs. A risk score model (Purine Score) containing the 3 lncRNAs was developed using the Lasso method. Results: A total of 22 differentially expressed lncRNAs were identified. These were reduced to a final set of three (LINC01671, ARAP1-AS1 and LINC02747). Age and metastasis (M) were identified as independent prognostic factors for KIRC using univariate and multivariate Cox analysis. An abnormal immune cell response was also associated with patient survival. The Purine Score correlated with abnormal expression of immune checkpoint genes. Genetic analysis of KIRC found somatic mutations in TP53, TRIOBP, PBRM1, PKHD1, VHL, NPHP3, TLN2, CABIN1, ABCC6, XIRP2, and CHD4. In vitro cell experiments showed that knockdown of LINC01671 promoted the proliferation and migration of 786-O cells, while inhibiting apoptosis. Overexpression of LINC01671 inhibited the proliferation and migration of CAKI-1 cells, while promoting apoptosis. Gene Set Enrichment Analysis (GSEA) analysis revealed that LINC01671 was significantly enriched in the MAPK, NF-kappa B, mTOR, PI3K-Akt, and Wnt signaling pathways. Conclusions: LINC01671 may be a novel prognostic marker with important therapeutic value for KIRC.