1 Introduction

Cancer is often considered a metabolic disorder in which energy is mainly supplied by anaerobic glycolysis and lipid metabolism, and thus, more nutrients are available, eventually resulting in cachexia [1,2,3]. In addition, metabolic products can also partly reflect tumor malignancy [4]. Prostate cancer (PCa) is a main malignancy that ranks first in urology and is believed to be a “cold” malignancy with a good prognosis until the development of disseminated lesions. Unlike other tumors, PCa is highly responsive to endocrine therapy, and surgery has the same effect on PCa patients with the lowest risk [5]. Given the influence of hormones, we assumed that the metabolic process in PCa differs from that in “non-inert” tumors. There have been several Mendelian randomized studies and clinical studies concerning the correlation between lipids and prostate cancer. To conclude, total cholesterol and low-density lipoprotein cholesterol were found to be positively related to a higher incidence, while some adjusted analyses eliminated this effect [4, 6, 7]. However, real-world studies on lipids and PCa in East Asia are lacking. Therefore, we performed a single-center study to investigate the correlation between metabolic indices and prostate cancer incidence.

2 Methods

2.1 Study design and participants

A retrospective, single-center cohort study was conducted and approved by the Ethics Review Committee of Qilu Hospital of Shandong University (Approval No. 202306–068), and exemption approval for informed consent was obtained.

Figure 1 shows a flowchart describing the selection of patients. From January 2013 to December 2022, 1944 patients diagnosed with PCa were recruited via prostate puncture. The indications for prostate biopsy were as follows: (1) total prostate-specific antigen (tPSA) > 10.0 ng/ml. (2) tPSA > 4.0 ng/ml and < 10.0 ng/ml with suspicious free PSA (fPSA)/tPSA < 0.16. (3) The presence of suspicious tubercles on digital rectal examination (DRE) at any level of the tPSA. (4) Abnormal signal areas on imaging techniques at any level of the tPSA. The exclusion criteria were as follows: (1) 855 patients had incomplete lipid data. (2) 319 patients had diabetes mellitus (DM). (3) 218 patients had unreachable PSA before receiving endocrine therapy. Eventually, 552 PCa patients were retained. Moreover, without DM, 80 patients with benign prostatic hyperplasia (BPH) were recruited to compare metabolic differences between BPH and PCa patients.

Fig. 1
figure 1

Flow diagram of the study design. PCa: prostate cancer; DM: diabetes mellitus; BPH: benign prostatic hyperplasia

Full size image

2.2 Data collection and definitions

Clinical data were collected from the medical records system. All laboratory examinations were performed under fasting conditions. We collected basic characteristics and metabolic indices. The basic characteristics included age at diagnosis, previous medical history, smoking and drinking habits, height and body mass index (BMI), inflammation indices, various blood cell counts and complex calculations, such as the neutrophil–lymphocyte ratio (NLR), monocyte–lymphocyte ratio (MLR), platelet–lymphocyte ratio (PLR), prognostic nutritional index (PNI), pan-immune–inflammation value (PIV) and systemic immune–inflammatory index (SII). Nutrient indices were measured to determine protein synthesis; thus, prealbumin, total protein and total albumin were measured. The metabolic indices included total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglyceride (TG), glucose and triglyceride-glucose (TyG) indices. The NLR, MLR, and PLR are defined as the ratio of the lymphocyte count divided by the neutrophil, monocyte and platelet counts in the blood, respectively. The PNI was calculated as the albumin concentration (g/L) + 5 × total lymphocyte count (10^9/L). The PIV was calculated as (neutrophil count × monocyte count × platelet count)/lymphocyte count, while the SII was calculated as the platelet count × neutrophil count/lymphocyte count. The TyG index affects tumor progression and is calculated as Ln [fasting TG (mg/dL) × fasting glucose (mg/dL)/2].

First, 80 patients with benign prostatic hyperplasia (BPH) were compared between BPH and malignant BPH. Then, the Gleason score after radical prostatectomy was collected to distinguish different degrees of risk of prostate cancer. The Gleason score consisted of primary and secondary components and was represented by ten scores and five degrees according to the International Society of Urology Pathology in 2015 (ISUP/WHO), with each component having a score no less than 3. A Gleason score of 3 + 3 was classified as degree 1, 3 + 4 was classified as degree 2, 4 + 3 was classified as degree 3, a total score of up to 8 was classified as degree 4, and the other was classified as degree 5. We defined patients with ISUP/WHO < 4, = 4, and > 4 as the low-, medium- and high-risk groups, respectively.

Localized PCa was defined as a neoplasm limited to the prostate capsule. If at least one of the following conditions were met, the patient was classified as having advanced PCa: a tumor invading adjacent tissues of the prostate and penetrating the capsule, a pelvic lymph node diameter greater than 1 cm on radiological examination, or distant organ or bone metastasis.

2.3 Statistical analysis

The collected data were analyzed with SPSS software (version 26.0). Statistics were separated into continuous variables and discrete variables according to continuity. Continuous variables are presented as the means ± standard deviations (SD), and Shapiro‒Wilk’s test was performed for normality. Student’s t test and ANOVA were used for normally distributed data; otherwise, the Mann–Whitney U test or Kruskal–Wallis H test was used. Discrete variables are presented as frequencies (percentages) and were compared with the chi-square test; if the expected number of units was < 5, Fisher’s exact test was used for adjustment. P < 0.05 was considered to indicate statistical significance.

3 Results

3.1 Baseline characteristics

The characteristics of 552 PCa patients were collected, and the average age of onset was 69.08 ± 7.52 years, while the average age of BPH patients was 68.06 ± 9.31 years. The basic characteristics and metabolic-associated indices of the two conditions are shown in Table 1. PCa patients had shorter heights and greater BMI, and examinations revealed significantly greater prealbumin, TC, LDL, TG and glucose levels. It seemed that PCa patients had a more severe inflammatory status, but this difference was not significant.

Table 1 Comparison of characteristics and metabolic indices between benign prostatic hyperplasia and prostate cancer patients
Full size table

3.2 Correlations between metabolic indices and different risks of prostate cancer

The average ages of patients with low-, medium- and high-risk prostate cancer were 69.23 ± 7.58, 68.63 ± 7.69 and 69.30 ± 7.30 years, respectively. A comparison of low-, medium- and high-risk PCa patients is shown in Table 2. No significant differences were detected, although height, HDL-C and LDL-C tended to increase with increasing risk of prostate cancer, while BMI tended to decrease.

Table 2 Comparison of metabolic indices and low-, medium- and high-risk prostate cancer patients
Full size table

3.3 Correlation between localized and advanced PCa in metabolic-associated indices

There was no significant difference between localized PCa and advanced PCa in terms of basic characteristics or lipids, as shown in Table 3.

Table 3 Comparison between localized prostate cancer and advanced prostate cancer
Full size table

4 Discussion

Inner metabolism can be influenced by many factors, such as diet, genetic alterations and inflammation, and further alters tumor progression; sometimes these factors mutually influence each other. A Western pattern diet with more fat and meat products tends to trigger tumor growth, and a case‒control study demonstrated a tight correlation between high-energy diets and prostate cancer [8]. With increasing energy intake and fat accumulation, organisms eventually develop obesity. However, there are multiple correlations between obesity and tumors. For example, hyperinsulinemia is considered to be the main alteration triggering high inflammatory escape and dysregulation of growth-associated signaling pathways. The metabolic pathway seemed to impact organisms via lipids. Under hyperinsulinemia, lipid anabolism is upregulated, and cholesterols and TG are produced. LDL-C is considered a “bad cholesterol” that promotes the formation of arterial plaques and participates in tumor energetic activity, while HDL-C acts as a “good cholesterol” to transform cholesterol into liver cholesterol. Conversely, high cholesterol levels result in growth and obesity via rapamycin signaling, which is mediated by increased insulin levels [9].

A greater BMI has an adverse impact on the development and prognosis of prostate cancer. BMI could reflect the metabolic status of redundant cholesterols stored in adipose tissue hypodermically, and BMI seems to correlate directly with glutamate-rich diets [10]. Obese individuals decline to develop aggressive and advanced disease, and they easily experience recurrence accompanied by complications [11]. However, a study in Africa reported a negative correlation between central obesity and aggressive PCa but a positive correlation between central obesity and medium-risk PCa [12].

Cholesterols nourish prostate cancer cells, and the process might be aggressive. CYP27A1, whose products regulate cholesterol metabolism, is expressed at lower levels in PCa, especially in advanced PCa [13]. Transcriptomic profiling revealed that anoctamin 1 (ANO1), which is enriched in primary malignant tumors, interacts with JUN, decreases CYP27A1 regulation, and promotes tumor invasion and metastasis [14]. Inhibition of cholesterol metabolism and storage can help decrease the invasiveness of tumors [15]. However, Xu and Li et al. discovered that increased cholesterols on the membrane enforced fatal T cells by inhibiting the esterification of cholesterols [16]. Lipid-associated signaling pathways are correlated with vital nutrient processes and are enriched in the PCa microenvironment [17]. As lipid-rich cells, macrophages transform cholesterols to targets that modulate hormone synthesis and immunity [17]. However, the impact of exogenous cholesterol intake on the immune system is controversial.

A Mendelian study revealed that higher TG levels increased the risk of PCa, while another study reported a positive correlation between high-grade PCa and TG levels [6, 18].

The exact correlation between lipids and PCa is still unclear. Two initial Mendelian studies concluded that high LDL-C increased the incidence of prostate cancer, but Ioannidou subsequently performed a matched analysis to eliminate this difference [4, 7]. A retrospective study involving 237 PCa patients revealed that patients with lower LDL-C and TC levels less than 130 mg/dl and 200 mg/dl, respectively, were less likely to suffer from PCa, but the tPSA level was controlled between 2 and 10 ng/ml. A meta-analysis of data from Western Europe and the United States revealed that the incidence of aggressive PCa was slightly greater than that of aggressive PCa, as indicated by higher LDL-C and TC [4, 19].

Our study confirmed a correlation between higher lipid levels and PCa, but no significant correlation was detected for HDL-C. In addition, increased glucose was also detected in PCa. Higher cholesterol and TG levels were detected in PCa tissues than in BPH tissues. However, during development, lipid components had no significant effect on either the Gleason score or extension. Several studies confirmed higher LDL-C in aggressive PCa patients, but none were from East Asia; thus, we assumed that intrinsic race and genetic factors played more important roles in PCa. Few studies have investigated the mechanism of TG dysfunction in PCa, and only Awad et al. reported that site changes in adipose triglyceride lipase (ATGL) decreased the aggressiveness and metastasis of castration-resistant prostate cancer (CRPC) but did not impact enzyme activity [20].

There are several limitations in our study. The data were collected retrospectively in a single center with a strong locality and could not reflect the real epidemic of PCa in China. However, there is still a need for other lipids, especially Lp(a), and it is believed that Lp(a) has a greater effect on arteriosclerosis than does LDL-C; thus, reducing the combination of LDL-C and Lp(a) might be more effective in both arteriosclerosis and tumor development. Our study did not support the use of lipid-lowering drugs, so there is still a conflict regarding whether the cholesterol effect of tumor maintenance is initiating.

5 Conclusion

PCa is less common in shorter and heavier individuals and seems to result in higher prealbumin, TC, LDL, TG and glucose levels but not in other individuals. Keeping a proper weight and a low-fat diet may decrease the probability of suffering PCa.