Renal cell carcinoma (RCC) accounts for approximately 2%-3% of all malignant tumors worldwide. Its incidence is increasing in developed countries (1). RCC is a histopathologically heterogeneous group of diseases with major subtypes, including clear cell renal cell carcinoma (KIRC), papillary renal cell carcinoma (KIRP), and chromophobe renal cell carcinoma (KICH). The most common histological subtype is KIRC, accounting for approximately 80% of all RCC cases (2). KIRC exhibits a characteristically clear cytoplasm due to the accumulation of abundant intracellular lipids and glycogen (3). This phenotype is thought to result from inactivation of the von Hippel–Lindau (VHL) gene, a central driver of KIRC, and subsequent constitutive activation of the hypoxia-inducible factor (HIF) pathway, leading to a characteristic metabolic profile (4). Consequently, the expression of angiogenic factors (VEGF and PDGF) and glycolytic enzymes (GLUT1 and HK2) is increased, which is associated with enhanced cell proliferation. Accordingly, increased angiogenesis, the Warburg effect, and enhanced cell proliferation have been observed (5). Such constitutive activation of the HIF pathway is a fundamental mechanism underlying the development and progression of KIRC. This has provided a rationale for therapies targeting VEGF and mTOR pathways. Nevertheless, therapeutic responses vary considerably, and most patients eventually experience disease progression (6). Additionally, KIRC has a poorer prognosis than other RCC subtypes, underscoring the need for further elucidation of its pathogenic mechanisms (7).

Liang et al. reported that statin use may be associated with improved survival in patients with RCC (8). In contrast, McKay et al. reported that an association between statin use and survival outcomes in patients with RCC was not definitively established (9). Although previous studies have reported an association between statin use and prognosis in patients with RCC, the impact of statins on different RCC subtypes is still under discussion, and the differences among subtypes remain unclear. Statins target HMG-CoA reductase (HMGCR), the rate-limiting enzyme in cholesterol biosynthesis. In a comprehensive analysis of ferroptosis-related genes, Wu et al. reported that high HMGCR expression was associated with improved overall survival in patients with KIRC (10). However, the association between HMGCR expression and disease-free survival has not yet been reported. Furthermore, analyses specifically focusing on HMGCR have not been performed in other RCC subtypes. Therefore, we employed an independent online bioinformatics platform to evaluate the expression and prognostic significance of HMGCR in patients with KIRC, KIRP, and KICH using data from The Cancer Genome Atlas (TCGA).

Assessment of HMGCR expression in tumor tissues of patients with renal cell carcinoma by histological subtype.HMGCR expression and its prognostic significance were analyzed using three publicly available bioinformatics platforms: GEPIA2 (http://gepia2.cancer-pku.cn/), UALCAN (http://ualcan.path.uab.edu/), and the UCSC Xena Browser (https://xenabrowser.net/). GEPIA2 and UALCAN were accessed on March 2, 2026, and the UCSC Xena Browser was accessed on July 7, 2026.

The gene name “HMGCR” was queried in The Cancer Genome Atlas (TCGA) database. HMGCR mRNA expression levels in kidney renal clear cell carcinoma (KIRC), kidney renal papillary cell carcinoma (KIRP), and kidney chromophobe carcinoma (KICH) were analyzed using the UALCAN platform based on the TCGA RNA-sequencing dataset. UALCAN compares tumor tissues with normal kidney tissues derived from the TCGA dataset.

Survival analysis according to the mRNA expression level of HMGCR in renal cell carcinoma by histological subtype. To investigate the prognostic significance of HMGCR expression in RCC, Kaplan–Meier survival analyses were performed using the GEPIA2 and UALCAN platforms. GEPIA2 integrates RNA-sequencing data from both the TCGA and GTEx projects, whereas UALCAN uses TCGA-derived data only. In both platforms, patients were divided into high- and low-HMGCR expression groups using the default median expression cutoff. Overall survival and disease-free survival were evaluated using GEPIA2, whereas overall survival was additionally assessed using UALCAN to compare the consistency of the findings obtained from the two analytical platforms.

To further evaluate whether HMGCR expression was independently associated with patient prognosis, a multivariable Cox proportional hazards regression analysis was performed using patient-level RNA-sequencing expression and clinical data downloaded from the UCSC Xena Browser. Only primary tumor samples (sample type code 01) were included in the analysis. Normal tissue samples (sample type code 11), discrepant pathological stage records, and cases with missing histological grade were excluded. After excluding cases with missing HMGCR expression, 523 patients, including 173 death events, were included in the multivariable Cox proportional hazards analysis.

Expression differences between tumor and normal tissues were evaluated using the statistical methods implemented in the UALCAN platform. Kaplan–Meier survival curves were generated using the Kaplan–Meier method and compared using the log-rank test. Multivariable Cox proportional hazards regression analysis was performed using the survival package in R (version 4.2.2), with adjustment for age at diagnosis, sex, pathological stage, and histological grade. HMGCR expression was analyzed as a continuous variable. All statistical tests were two-sided, and p<0.01 was considered statistically significant. Because this study evaluated a single predefined candidate gene (HMGCR), multiple-testing correction was not applied.

HMGCR mRNA expression is decreased in KIRC and KIRP tissues. To determine HMGCR mRNA expression in KIRC, KIRP, and KICH tissues, TCGA datasets were analyzed using the UALCAN platform. We analyzed whether HMGCR expression was associated with the development of KIRC, KIRP, and KICH. Figure 1 shows the expression of HMGCR in tumor tissues from patients with KIRC, KIRP, and KICH compared with normal kidney tissues. HMGCR expression was significantly decreased in the tumor tissues of patients with KIRC (p<0.01) and KIRP (p<0.01) (Figure 1a, b). In contrast, there was no significant difference in HMGCR expression between KICH and normal kidney tissues (Figure 1c).

Low expression of HMGCR is associated with shorter patient survival in KIRC. We used the GEPIA2 platform to generate Kaplan–Meier survival plots to analyze overall survival and disease-free survival in KIRC, KIRP, and KICH. Decreased expression of HMGCR was significantly associated with shorter overall survival (p<0.01) (Figure 2a) and disease-free survival (p<0.01) (Figure 2d) in patients with KIRC. In contrast, no significant differences in overall survival or disease-free survival were observed in patients with KIRP or KICH (Figure 2b, c, e, f).

Kaplan–Meier survival plots were generated using the UALCAN platform to compare high- and low-HMGCR expression groups in patients with each RCC subtype. The results showed that decreased HMGCR expression was significantly associated with shorter survival in patients with KIRC (p<0.01) (Figure 3a). In UALCAN, no significant association with overall survival was observed for KIRP. In KICH, a nominal association was observed (p=0.04), but this did not meet the pre-specified significance threshold of p<0.01 and was not supported by GEPIA2 (Figure 3b, c).

Multivariable Cox proportional hazards analysis identifies HMGCR as an independent prognostic factor in KIRC. To determine whether HMGCR expression was independently associated with overall survival, a multivariable Cox proportional hazards regression analysis was performed using TCGA-KIRC RNA-sequencing expression and clinical data obtained from the UCSC Xena Browser. After adjustment for age at diagnosis, sex, pathological stage, and histological grade, higher HMGCR expression remained independently associated with improved overall survival (hazard ratio=0.632, 95% confidence interval=0.468-0.855, p=0.003) (Table I).

In this study, we used the bioinformatics platforms GEPIA2 and UALCAN to analyze HMGCR mRNA expression and Kaplan–Meier survival plots in patients with KIRC, KIRP, and KICH. HMGCR mRNA expression levels in tumor tissues from patients with KIRC and KIRP were significantly lower than those in normal kidney tissues (Figure 1a, b). Furthermore, a reduction in HMGCR expression among patients with KIRC correlated with shorter overall survival and disease-free survival (Figure 2a, d, and Figure 3a). Importantly, multivariable Cox proportional hazards analysis demonstrated that HMGCR expression remained independently associated with overall survival after adjustment for age at diagnosis, sex, pathological stage, and histological grade (Table I). These findings indicate that the prognostic significance of HMGCR expression is not solely explained by established clinicopathological factors, suggesting that HMGCR may serve as an independent prognostic biomarker in patients with KIRC.

KIRC is primarily driven by inactivation of the tumor suppressor gene VHL, which results in constitutive activation of the HIF signaling pathway (11, 12). HIF signaling upregulates the expression of lipid uptake–related receptors, including CD36 (13). In KIRC tumor tissues, activation of the HIF pathway promotes the uptake of circulating lipid particles, including cholesterol, leading to the accumulation of cytoplasmic lipid droplets. Indeed, cytoplasmic lipid droplets have been observed together with increased expression of the lipid-droplet–associated protein PLIN2 (14). Under these pathological conditions, excessive intracellular cholesterol levels may suppress HMGCR mRNA expression as part of a regulatory feedback mechanism (15, 16). In KIRC, the association between low HMGCR expression and shorter overall survival and disease-free survival may be attributed to excessive lipid droplet accumulation. Furthermore, Saulsbury et al. reported that reduced SYNJ2BP expression was associated with poor survival outcomes in KIRC, suggesting that alterations in mitochondrial homeostasis may contribute to disease progression (17). In addition, metabolic alterations in tumor cells may influence the tumor microenvironment, and interactions between tumor cells and surrounding stromal components have been recognized as important determinants of disease progression in KIRC (18). Previous studies have demonstrated impaired lipid droplet turnover in KIRC, and improvements in lipid droplet metabolism have been reported to be associated with improved prognosis (19). Therefore, in KIRC, excessive lipid influx and delayed lipid droplet turnover may contribute to poor prognosis. Collectively, these findings suggest that decreased HMGCR expression reflects a lipid-accumulated state and is consequently associated with poor survival.

In contrast, although HMGCR mRNA expression was lower in KIRP tissues than in normal tissues (Figure 1b), no significant association with prognosis was observed (Figure 2b, e, and Figure 3b). These observations suggest that, in KIRP, enhanced lipid metabolic activity could suppress HMGCR expression via feedback mechanisms, similar to KIRC. However, unlike KIRC, pathological intracellular lipid accumulation is less prominent in KIRP (20). The imported lipids may be rapidly utilized for membrane biosynthesis or taken up by stromal macrophages as foam cells. Therefore, lipid accumulation in the renal parenchyma is suppressed. Collectively, these observations raise the possibility that pathological intracellular cholesterol accumulation does not reach the same threshold of accumulation in KIRP as in KIRC, thereby limiting its prognostic impact. In KICH, HMGCR mRNA expression was not significantly altered, nor was it associated with patient prognosis (Figures 1c, 2c, f, and 3c). KICH is defined by widespread chromosomal loss and is thought to originate from the distal nephrons (21, 22). Morphologically, tumor cells display distinctly thickened cell membranes, resulting in a plant-cell-like appearance (23). Furthermore, compared with KIRC, which is characterized by abundant intracellular lipid accumulation, KICH exhibits little to no pathological lipid accumulation (24). Tumorigenesis in KICH is thought to be driven primarily by mitochondrial dysfunction and activation of the mTOR signaling pathway. Unlike other RCC subtypes, lipid metabolism, including cholesterol biosynthesis, does not appear to play a central role in the pathogenesis of KICH.

Overall, similar findings were obtained using both GEPIA2 and UALCAN; both analyses were based primarily on TCGA-derived datasets and therefore should not be regarded as independent validation cohorts. HMGCR expression was not significantly associated with the prognosis of KIRP or KICH. In contrast, decreased HMGCR expression remained independently associated with poorer overall survival in patients with KIRC, indicating that HMGCR is an independent prognostic biomarker for this RCC subtype.

In conclusion, our findings indicate that decreased HMGCR expression is a potential prognostic marker for poor survival in KIRC, likely reflecting pathological lipid accumulation and impaired lipid droplet turnover. By contrast, reduced HMGCR expression in KIRP and unchanged levels in KICH do not significantly impact prognosis, underscoring the distinct pathogenic mechanisms across RCC subtypes.

The Authors declare that they have no potential conflicts of interest with respect to the research, authorship, or publication of this article.

All authors contributed to the conception and design of this study. Etsushi Akitaya drafted the manuscript. Yuzuki Nakamura, Shiho Kato, and Hinayo Kikuta were involved in data collection and analysis. Teruo Endoh, Yasuhiro Kuramitsu, and Yuji Takahashi provided comments on the initial draft of the manuscript. All authors reviewed and edited the manuscript and approved the final version.

We would like to thank Editage for English language editing.

During the preparation of this manuscript, a large language model (ChatGPT, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning–based image enhancement tools.