Liver　cancer　manifests　as　a　profoundly　heterogeneous　malignancy,　posing　significant　challenges　in　terms　of　both　therapeutic　intervention　and　prognostic　evaluation.　Given　that　the　liver　is　the　largest　metabolic　organ,　a　prognostic　risk　model　grounded　in　single-cell　transcriptome　analysis　and　a　metabolic　perspective　can　facilitate　precise　prevention　and　treatment　strategies　for　liver　cancer.　Hence,　we　identified　11　cell　types　in　a　scRNA-seq　profile　comprising　105,829　cells　and　found　that　the　metabolic　activity　of　malignant　cells　increased　significantly.　Subsequently,　a　prognostic　risk　model　incorporating　tumor　heterogeneity,　cell　interactions,　tumor　cell　metabolism,　and　differentially　expressed　genes　was　established　based　on　eight　genes;　this　model　can　accurately　distinguish　the　survival　outcomes　of　liver　cancer　patients　and　predict　the　response　to　immunotherapy.　Analyzing　the　immune　status　and　drug　sensitivity　of　the　high-　and　low-risk　groups　identified　by　the　model　revealed　that　the　high-risk　group　had　more　active　immune　cell　status　and　greater　expression　of　immune　checkpoints,　indicating　potential　risks　associated　with　liver　cancer-targeted　drugs.　In　summary,　this　study　provides　direct　evidence　for　the　stratification　and　precise　treatment　of　liver　cancer　patients,　and　is　an　important　step　in　establishing　reliable　predictors　of　treatment　efficacy　in　liver　cancer　patients.