Multifactor nomogram predicts bone metastasis in patients initially diagnosed with prostate cancer

doi:10.21203/rs.3.rs-4898299/v1

Download PDF

Article

Multifactor nomogram predicts bone metastasis in patients initially diagnosed with prostate cancer

https://doi.org/10.21203/rs.3.rs-4898299/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objectives

To explore the correlation between prostate-specific antigen density (PSAD), Cystatin C (CysC), systemic immune-inflammatory index (SII), and bone metastasis in initially diagnosed prostate cancer patients, and to develop a predictive nomogram model.

Methods

A retrospective analysis was conducted on 208 newly diagnosed prostate cancer patients, divided into a modeling group (146 cases) and a validation group (62 cases). Logistic regression analyses identified independent risk factors for bone metastasis, which were used to construct a nomogram. Model accuracy was assessed using AUC and calibration curves.

Results

Gleason score (GLS), PSAD, CysC, fibrinogen (FIB), SII, and pelvic lymph node metastasis were identified as independent risk factors for bone metastasis (OR > 1, P < 0.05). The model showed an AUC of 0.897 (95% CI: 0.845–0.948) in the modeling group and 0.840 (95% CI: 0.741–0.940) in the validation group.

Conclusion

PSAD, CysC, SII, FIB, GLS, and pelvic lymph node metastasis are significant risk factors for bone metastasis in newly diagnosed prostate cancer patients. The nomogram model can assist in clinical diagnosis, especially in hospitals lacking bone scanning equipment. This study aimed to explore the clinical correlation among PSAD, CysC, SII, and bone metastasis in initial prostate cancer patients. Additionally, a nomogram incorporating relevant risk factors was developed.

Prostate cancer

nomogram

bone metastasis

Cystatin C

systemic immune inflammatory index

Based on the 2020 GLOBOCAN data from the World Health Organization, the prevalence of prostate cancer in men is rapidly rising, ranking second only to lung cancer. Additionally, the bone is the primary site of metastasis¹. The fatality rate among individuals with prostate cancer and bone metastasis is considerably higher than that among those without bone metastasis². Patients with bone metastasis in prostate cancer typically experience intense pain in the bones, pathological fractures, hypercalcaemia, and other events related to the bones (such as skeletal-related events). The quality of life of these patients is seriously negatively impacted, and the mortality rate is greatly increased^2,3. The most frequently employed technique for detecting bone metastasis in patients with prostate cancer is whole-body bone scanning, which, however, poses certain risks associated with radiation and imposes a significant financial burden on patients^4,5. A large number of these patients with bone metastases are not timely diagnosed, which delays their subsequent treatment and significantly reduces their survival rate. The rate of bone metastasis in patients with metastatic prostate cancer is as high as 90%^6–8. Hence, combining relevant risk factors for bone metastasis is crucial to construct a reliable forecasting framework, which would efficiently and economically guide the conduct of whole-body bone scans for patients with prostate cancer.

A potential correlation exists between age, alkaline phosphatase levels, Gleason score (GLS), prostate-specific antigen (PSA), and various other factors associated with bone metastasis in prostate cancer; however, a consensus has not yet been reached^6,9–12. Therefore, this study retrospectively examined the medical information of 208 individuals recently diagnosed with prostate cancer and investigated the prognostic significance of patient age, PSA density (PSAD), Cystatin C (CysC), systemic immune-inflammatory index (SII), GLS, lymphocyte-to-monocyte ratio (LMR), and positive pelvic lymph node metastasis for bone metastasis in these patients. Additionally, a novel nomogram prediction model was developed.

Inclusion and exclusion criteria

We retrospectively analysed the clinical information of 208 patients who were newly diagnosed with prostate cancer at our hospital between February 2019 and September 2023. The inclusion criteria were as follows: prostate adenocarcinoma diagnosed via pathology after transrectal ultrasound-guided prostate biopsy or surgery, first confirmed prostate cancer case, no other tumour history, and all examinations and assays related to blood-related indicators completed within a week before the patient underwent prostate biopsy. The exclusion criteria included individuals without prostate cancer or those with other types of tumour diseases, bone metastasis caused by congenital bone disease or other unknown factors, severe infection, active liver disease, haematological diseases, prolonged use of hormones or medications that impact blood clotting, and incomplete clinical data. Following the Declaration of Helsinki, this study was authorised by the Ethics Committee of the General Hospital of the Northern Theatre of the People's Liberation Army. Informed consent was obtained from all study participants.

Diagnostic criteria of bone metastasis

Currently, the prevailing approach for detecting bone metastasis in prostate cancer is whole-body bone scanning, which has a sensitivity ranging from 62–89%¹³. Moreover, this scan can identify bone metastasis 3–6 months prior to X-ray examinations^10,13–15. Diagnostic criteria involve comparing the radioactive distribution in the lesion to that of the control by using adjacent normal bone tissue or the contralateral part as the normal control. Additionally, the accumulation of imaging agents in the lesion should be compared to the contralateral corresponding part or adjacent healthy tissue. A positive result is indicated by higher or lower accumulation than that of the control or healthy tissue^10,13–15.

Clinical data collection

Data regarding patient age, total PSA (TPSA) level, prostate volume, PSAD level, CysC level, GLS, pelvic lymph node metastasis, FIB, lymphocytes, neutrophils, and platelets were collected for statistical analysis. TPSA was detected using a fully automatic time-resolved fluorescence immunoassay analyser (Wallac Oy, Finland). Serum CysC levels were determined utilising a 7170 automatic biochemical analyser (Hitachi, Japan). The prostate volume was measured employing PHILIPS HD-11 GE-VOLUSION 730 EXPERT colour Doppler ultrasound diagnostic instrument (Philips Healthcare, Inc.) to measure the anterior-posterior, transverse, and head-tail diameters of the prostate, and the prostate volume was calculated using the following formula: Prostate volume = (anteroposterior diameter × transverse diameter × cephalocaudal diameter) × π/6. The PSAD value was calculated as PSAD = PSA/prostate volume. According to the guidelines of the European Association of Urology, the GLS is divided into three categories (① GLS ≤ 6, ② GLS = 7, and ③ GLS ≥ 8).

Statistical analysis

R software (4.3.1, SPSS 25.0) and GraphPad Prism 10.0.3 were used to analyse the indicators of potential risk factors included in this study. The measurement data that followed a normal distribution are presented as mean ± standard deviation (x ± s). To compare the groups, t-tests were conducted on two independent samples. For non-normally distributed data, the median (quartile) was used. M (q25-q75) and the rank-sum test (Mann–Whitney U test) was used for intergroup comparisons. Count data are expressed as rates (%), and comparisons between groups were conducted using the chi-square test (X² Inspection). Following univariate analysis, additional tests were conducted, such as those for multi-collinearity, interaction analyses, and logistic regression analysis of the variables demonstrating statistically significant (P < 0.05), to determine independent risk factors. The R software was used to test the interaction of the study factors, and the results indicated no interaction among any of the factors (P > 0.05). SPSS software was used to perform multi-collinearity analysis of the independent risk factors, while excluding collinearity between the factors of interest, in conjunction with multivariate logistic regression analysis. Differences were considered statistically significant at P < 0.05.

Finally, after considering the partial regression coefficient of each individual risk factor, we developed a prognostic model for bone metastasis in patients initially diagnosed with prostate cancer, created a visual representation called a nomogram, and evaluated the prediction efficiency of the model by adopting receiver operating characteristic (ROC) and calibration curves.

General characteristics of the included participants

According to the results of radionuclide bone scanning combined with imaging, 86 patients (41.3%) had bone metastases, and 122 patients (58.7%) had no bone metastases. Table 1 displays that of the 146 patients in the modelling group, 58 (39.7%) had bone metastases. Similarly, in the validation group, 28 (45.2%) of the 62 patients had bone metastasis. Notably, no notable differences were observed in bone metastasis, age, PSAD, FIB, CysC, SII, LMR, pelvic lymph node metastasis, and GLS between the two groups (P > 0.05). Based on the bone scan findings, individuals in the modelling group were categorised into groups with and without bone metastasis, and the variables of both groups were subjected to statistical analysis. Statistically significant differences were observed in PSAD, FIB, CysC, SII, LMR, pelvic lymph node metastasis, and GLS (P < 0.05), whereas no significant difference was detected in age between the two groups (P > 0.05) (Table 2).

Table 1

**Comparison of clinical characteristics between modelling and validation groups**
Parameter	Modelling group (n = 146)	Validation group (n = 62)	P value
Bone metastasis			0.566
(-)	88 (60.3%)	34 (54.8%)
(+)	58 (39.7%)	28 (45.2%)
Age	70.53 ± 7.17	72.4 ± 7.69	0.104
PSAD	4.18 ± 7.57	3.15 ± 5.59	0.328
FIB (g/L)	4.1 ± 1.55	3.93 ± 1.14	0.989
CysC (mg/L)	0.99 ± 0.31	1 ± 0.28	0.781
SII	584.24 ± 445.07	559.39 ± 559.9	0.789
LMR	3.96 ± 1.68	4.26 ± 1.37	0.089
Pelvic lymph node metastasis			0.77
(-)	87 (59.6%)	39 (62.9%)
(+)	59 (40.4%)	23 (37.1%)
GLS			0.505
≤6分	21 (14.4%)	6 (9.7%)
7	51 (34.9%)	26 (41.9%)
≥8分	74 (50.7%)	30 (48.4%)
PSAD, prostate-specific antigen density; FIB, fibrinogen; CysC: Cystatin C; SII, systemic immune inflammatory index; LMR, lymphocyte-to-monocyte ratio; GLS, Gleason score.

Table 2

Comparison of clinical characteristics between non-bone metastasis and bone metastasis groups
Parameter	N-BMT (n = 88)	BMT (n = 58)	t/Z/χ 2	P value
Age	69.59 ± 6.607	71.95 ± 7.783	t = -1.964	0.051
PSAD	0.73 (0.35, 1.65)	3.42 (1.67, 9.25)	Z = -6.579	< 0.001
CysC(mg/L)	0.89 (0.77, 1.05)	1.03 (0.91, 1.21)	Z = -3.878	< 0.001
FIB (g/L)	3.28 (2.80, 4.24)	4.28 (3.30, 6.11)	Z = -4.291	< 0.001
LMR	4.18 (3.12, 5.15)	3.56 (2.29, 4.33)	Z = -2.492	0.013
SII	377.57 (283.90, 525.28)	564.56 (409.39, 1021.18)	Z = -4.403	< 0.001
GLS			χ 2 = 27.073	< 0.001
≤ 6分	20 (22.7%)	1 (1.7%)
7分	38 (43.2%)	13 (22.4%)
≥ 8分	30 (34.1%)	44 (75.9%)
Pelvic lymph node metastasis	22 (25%)	37 (63.8%)	χ 2 = 21.847	< 0.001
BMT, bone metastasis; N-BMT, non-bone metastasis; PSAD, prostate-specific antigen density; FIB, fibrinogen; CysC: Cystatin C; SII, systemic immune inflammatory index; LMR, lymphocyte-to-monocyte ratio; GLS, Gleason score.

Examination of bone metastasis in individuals initially diagnosed with prostate cancer using logistic regression analysis

Among patients newly diagnosed with prostate cancer, univariate logistic analysis revealed that PSAD, FIB, CysC, SII, LMR, pelvic lymph node metastasis, and GLS were risk factors for bone metastasis. However, multivariate logistic analysis indicated that PSAD, FIB, CysC, SII, pelvic lymph node metastasis, and GLS were independent risk factors for bone metastasis in patients with prostate cancer (P < 0.05), whereas LMR was not significant (P > 0.05) (Table 3).

Table 3

Univariate and multivariate logistic regression analysis
Parameter	Univariate logistic regression analysis				Multivariate logistic regression analysis
Parameter	B	OR	95%CI	P	B	OR	95%CI	P
PSAD	0.219	1.244	1.108–1.397	< 0.001	0.138	1.148	1.029–1.282	0.013
CysC (mg/L)	2.770	15.962	3.551–71.746	< 0.001	2.491	12.07	1.950–74.700	0.007
FIB (g/L)	0.565	1.760	1.355–2.286	< 0.001	0.412	1.510	1.049–2.175	0.027
GLS				< 0.001				0.032
≤ 6分		1.000				1.000
7分(VS GLS ≤ 6)	1.923	6.842	0.834–56.142	0.073	1.556	4.739	0.438–51.287	0.200
≥ 8分(VS GLS ≤ 6)	3.379	29.333	3.734-230.444	0.001	2.523	12.465	1.222-127.145	0.033
LMR	-0.268	0.765	0.608–0.962	0.022	0.085	1.089	0.767–1.546	0.633
SII	0.002	1.002	1.001–1.003	0.001	0.001	1.001	1.000-1.003	0.045
Pelvic lymph node metastasis	1.665	5.286	2.571–10.868	< 0.001	1.211	3.358	1.277–8.830	0.014
B, partial regression coefficient; P, P value; OR, Odds ratio; CI, confidence interval; PSAD, prostate-specific antigen density; FIB, fibrinogen; CysC: Cystatin C; SII, systemic immune inflammatory index; LMR, lymphocyte-to-monocyte ratio; GLS, Gleason score.

Interaction analysis and multicollinearity test among the study factors

R software was used to test the interaction between the above factors, and no significant difference (P > 0.05) was noted. Combined with multivariate logistic regression analysis, six independent risk factors were analysed via multicollinearity analysis: PSAD (0.786), FIB (0.773), CysC (0.951), GLS (0.820), SII(0.867), and pelvic lymph node metastasis tolerance (0.854) (all greater than 0.1; P < 0.05). The variance inflation factors were all < 3 (1.273, 1.294, 1.052, 1.219, 1.153, and 1.172, respectively) (P < 0.05).

Development and verification of a predictive model and chart for bone spread in individuals initially diagnosed with prostate cancer

Based on the independent risk factors acquired from multivariate logistic analysis, The prediction model was: ‘logistic (P) =-8.241 + 0.137 × PSAD + 0.421 × FIB + 2.440 × CysC + 0.001 × SII + 1.454 × GLS + 2.426 × GLS + 1.185 × pelvic lymph node metastasis (yes = 1, no = 0)’. A nomogram model was created to predict bone metastasis in initially diagnosed individuals with prostate cancer (Fig. 1). Model calibration and decision curves were constructed. Figure 2 illustrates the alignment of the model calibration curve with the ideal curve. In conjunction with the ROC curve area, this indicates a high level of accuracy of the constructed model. The decision curve indicated that the prediction model had a high prediction efficiency (Fig. 3). When comparing the impact index of the forecast model with each individual risk factor, the forecast model exhibited the highest sensitivity and specificity, along with the highest AUC for the Youden index and ROC (sensitivity: 81.0%, specificity: 87.5%, Youden index: 0.685, AUC: 0.897, 95%CI: 0.845–0.948) (Table 4). The ROC curves of PSAD, CysC, SII, and each individual risk factor (Fig. 4) were generated using GraphPad Prism software. Additionally, AUCs were calculated for PSAD (AUC, 0.822; 95%CI: 0.753–0.892), CysC (AUC, 0.690; 95%CI: 0.602–0.778), and SII (AUC, 0.716; 95%CI: 0.630–0.802). Furthermore, all P values were less than 0.05. Simultaneously, the model was verified using data from the validation group, and the ROC curves of the modelling and validation groups were compared (Fig. 5). The AUC for the validation group was 0.840 (95%CI: 0.741–0.940,P < 0.05), and the sensitivity and specificity rates were 82.1% and 73.5%, respectively. The maximum Youden index was 0.556.

Table 4

Comparison of effect indexes of independent risk factors and prediction model
Parameter	Cut off value	Sensitivity	specificity	Youden index	AUC	95%CI	P value
PSAD	1.693	0.759	0.773	0.531	0.822	0.753–0.892	< 0.001
CysC(mg/L)	0.945	0.690	0.636	0.326	0.690	0.602–0.778	< 0.001
SII	507.187	0.621	0.727	0.348	0.716	0.630–0.802	< 0.001
FIB(g/L)	3.790	0.655	0.705	0.360	0.710	0.623–0.797	< 0.001
GLS	≧ 8	0.759	0.659	0.418	0.731	0.649–0.812	< 0.001
Pelvic lymph node metastasis	+	0.638	0.750	0.388	0.694	0.605–0.783	< 0.001
prediction model	0.471	0.810	0.875	0.685	0.897	0.845–0.948	< 0.001
AUC, area under the receiver characteristic curve; CI, confidence interval; PSAD, prostate-specific antigen density; FIB, fibrinogen; CysC: Cystatin C; SII, systemic immune inflammatory index; GLS, Gleason score.

Metastasis is a critical stage of prostate cancer, seriously impairing patient prognosis. In China, the percentage of newly diagnosed prostate cancer patients with metastatic prostate cancer is significantly higher at 51.4–54%, compared to only 7–8% in Europe and the United States^16,17. Prostate cancer frequently spreads to the bones, with a bone metastasis rate of up to 90% in patients with metastatic prostate cancer^1,6–8. Individuals with bone metastasis from prostate cancer experience a notable decline in their quality of life, and the mortality rate among these patients is considerably higher than that among those without bone metastasis³. The primary cause of death in such cases is predominantly attributed to severe harm caused by the spread of cancer to the bones^2–4. In recent years, an increasing number of patients with prostate cancer have exhibited symptoms of bone metastasis upon visiting the hospital. Of the 208 individuals newly diagnosed with prostate cancer in this study, 41.3% (86 patients) exhibited bone metastasis. Hence, the timely identification and treatment of bone spread in these individuals can significantly enhance their prognosis and quality of life.

The occurrence and death rates of prostate cancer are influenced by age^18,19, Nevertheless, there has been a debate regarding the connection between age and bone metastasis in individuals diagnosed with prostate cancer^6,20. Some studies suggest that age is associated with bone metastasis in prostate cancer but is not an independent risk factor^21,22. However, our study found no notable disparity in age between recently diagnosed prostate cancer patients with and without bone metastasis on univariate analysis. The inconsistent conclusions of the above studies may have resulted from differences in sample sizes, regions, and ethnicities.

PSAD can differentiate between benign prostatic hyperplasia and prostate cancer. Furthermore, it can assess the clinical extent of prostate cancer and exhibit a strong association with bone metastasis in individuals recently diagnosed with prostate cancer^22–25. Similar to the findings of the aforementioned studies, this study revealed that PSAD was associated with bone metastasis in patients with prostate cancer. Moreover, multivariate regression analysis demonstrated that PSAD independently posed a risk for bone metastasis of prostate cancer (AUC: 0.822, 95%CI: 0.602–0.778, P < 0.001), with a sensitivity of 75.9%, specificity of 77.3%, and corresponding cutoff value of 1.693. This finding suggests that PSAD had a specific predictive value for bone metastasis in individuals initially diagnosed with prostate cancer. This study included a greater number of variables that were not influenced by external factors, thereby enhancing the predictive efficacy of PSAD.

In the last few years, there has been an increasing body of evidence from clinical and experimental data that CysC might play a role in the development of various diseases, such as cancer^26–30. CysC controls the expression of androgen receptors in prostate cancer through the MAPK/ERK 1/2 signalling pathway, consequently hindering the invasion of tumour cells. The MAPK/ERK 1/2 signalling pathway is linked to tumour advancement, reliance on androgens, and unfavourable prognosis³¹. Serum CysC may have a major impact on the development of bone metastasis in individuals diagnosed with prostate cancer. This is achieved by suppressing the function of cysteine proteases and their natural ability to control cell growth, specialisation, movement, and bone restructuring. Consequently, elevated serum CysC levels in patients with prostate cancer may be strongly correlated with the severity of the disease and its invasive nature. In this study, CysC and other potentially relevant risk factors were analysed using univariate and multivariate analyses. These findings indicated that CysC exhibited a significant correlation with bone metastasis in individuals recently diagnosed with prostate cancer and was independently identified as a risk factor for bone metastasis in patients newly diagnosed with prostate cancer. CysC had an AUC of 0.690 (95%CI: 0.753–0.892, P < 0.001), sensitivity and specificity of 69.0% and 63.6%, respectively, and the corresponding cutoff value of 0.945. Using CysC to predict prostate cancer bone metastasis, we recommend incorporating additional indicators to exclude confounders. This suggestion stems from the susceptibility of CysC levels to intervention^26,32–36.

SII integrates the lymphocyte, neutrophil, and platelet counts (SII = P × N/L). The spread of prostate cancer to the bones is associated with inflammatory reactions. Additionally, the SII has been identified as a separate factor that increases the risk of bone metastasis in individuals diagnosed with prostate cancer^37,38. Moreover, as a marker of inflammation, the influence of LMR on the prognosis of urinary system tumours has been reported for many years ^39,40. Under the action of certain cytokines (such as CSF-1 or IL-10) secreted by tumour cells, M2-like macrophages proliferate and promote tumour invasion and metastasis via angiogenesis^39,40. Herein, univariate analysis revealed a correlation between bone metastasis in patients initially diagnosed with prostate cancer and both SII and LMR (P < 0.05). In the multivariate logistic regression analysis, SII emerged as a separate risk factor for bone metastasis in individuals initially diagnosed with prostate cancer. The AUC for SII was 0.716 (95%CI: 0.630–0.802, P < 0.001), with a sensitivity of 62.1%, specificity of 72.7%, and a cutoff value of 507.187. However, LMR was not an independent risk factor (P > 0.05).

There is growing evidence regarding the poor prognosis of tumour patients with elevated FIB levels, including those with prostate cancer^41–43. In a study by Xie et al., FIB increased the risk of bone metastasis in patients newly diagnosed with prostate cancer, and the cutoff value for FIB was 3.08 g/l, with a sensitivity of 0.684 and a specificity of 0.760 (AUC, 0.739; 95%CI: 0.644–0.833, P < 0.001)⁴⁴. In this study, the results of FIB analysis included in our study are consistent with those of the abovementioned studies. The AUC (95%CI: 0.623–0.797, P < 0.001) was 0.710, with a sensitivity of 65.5%, specificity of 70.5%, and cutoff value of 3.790.

If the GLS is > 7, it is crucial to consider the potential for bone metastasis in individuals diagnosed with prostate cancer, as stated in the European Association of Urology^45,46. According to the guidelines of the American Urological Association, if GLS exceeds 8, individuals diagnosed with prostate cancer have a higher probability of experiencing bone metastasis⁴⁶. In order to determine the significance of the GLS in predicting bone metastasis in patients recently diagnosed with prostate cancer, we transformed the quantitative data of the GLS into qualitative data for statistical analysis. The grading standard was: GLS ≤ 6 points, defined as GLS ① group; GLS = 7 points, defined as GLS ② group; and GLS ≥ 8 points, defined as GLS ③ group. The findings demonstrated a significant correlation between bone metastasis and GLS in individuals initially diagnosed with prostate cancer, with a GLS threshold of 8. The AUC value for the ROC curve was 0.731 (95%CI: 0.649–0.812; P < 0.001), indicating a sensitivity of 75.9% and a specificity of 65.9%.

Lymph node metastasis is closely associated with bone metastasis in patients initially diagnosed with prostate cancer⁴⁷. Guo et al. obtained the clinical information of 249331 individuals diagnosed with prostate cancer from the SEER database, and statistical analysis revealed that lymph node metastasis was an independent risk factor for bone metastasis in patients initially diagnosed with prostate cancer⁴⁸. In the present study, pelvic lymph node metastasis was significantly associated with bone metastasis in patients newly diagnosed with prostate cancer. It was an independent risk factor, as evidenced by an AUC of 0.694 (95%CI: 0.605–0.783; P < 0.001), with a sensitivity and specificity of 63.8% and 75.0%, respectively.

This nomogram model was constructed according to the weight of the influence coefficient of each risk factor, and the internal influence between the research factors was excluded through the interaction between the research factors and the multicollinearity test. External validation was carried out using the validation group data. After excluding the interaction between different factors and multicollinearity, our constructed nomogram prediction model accurately assessed the risk of bone metastasis with a high level of sensitivity (81.0%) and specificity (87.5%), as indicated by the findings of both univariate and multivariate logistic regression analyses. This study validated and incorporated new reliable indicators, namely, PSAD, CysC, and SII, in combination with several other related factors. In clinical practice, it would be convenient to obtain the variable indicators of the aforementioned model. Clinicians can easily comprehend and apply this nomogram, thereby enhancing their ability to assess the risk of bone metastasis in patients with newly diagnosed prostate cancer, especially under constrained conditions in actual clinical practice. The nomogram’s image was intuitive and illustrative, while the prediction results were particularly objective and reliable.

The primary objective of this study was to assess the risk of bone metastases in patients newly diagnosed with prostate cancer. The study is currently limited by its single-centre sample size; however, future efforts will involve collaboration with multiple centres to expand the sample size and incorporate additional research variables, including the clinical stage of patients diagnosed for the first time. Notably, our study introduced three novel variables—SII, CysC level, and PSAD—that have not been previously explored in this area. Our aim was to encourage more researchers to investigate and validate the risk factors associated with bone metastasis in patients newly diagnosed with prostate cancer and establish a more precise and reliable predictive model. Ultimately, this will enhance the diagnostic capabilities of frontline medical personnel in regions with limited healthcare resources, aiding in the early detection of prostate cancer bone metastasis.

In patients newly diagnosed with prostate cancer, PSAD, CysC, SII, FIB, GLS, and pelvic lymph node metastasis were identified as autonomous factors that contributed to bone metastasis. The nomogram model developed in this study demonstrated excellent precision and effectiveness in predicting bone metastasis in these individuals, indicating its potential for practical medical applications.

Additional Information

Competing interests: The authors declare that they have no competing interests.

Conflict of Interest

The authors declare that they have no conflicts of interest.

Authors’ Contribution

TRL conceived the study and designed the data collection tool. WM, GS, LSL, LJ, and YJH analyzed the data. WM and LHT drafted the manuscript, which was reviewed and edited by all authors.

Acknowledgments

This study was financially supported by the Natural Science Foundation of the Liaoning Province (Contract No. 2022-MS-047).

Ethics Declarations

This study was reviewed and approved by the General Hospital of the Northern Theatre Command PLA Ethics Committee (approval number: Y [2023] 209).

Data availability statement

All data generated or analyzed in the course of this study are encompassed within this published article, along with its supplementary information files.

Sung, H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J. Clin. 71, 209-249, doi:10.3322/caac.21660 (2021).
Gartrell, B. & Saad, F. Managing bone metastases and reducing skeletal related events in prostate cancer. Nature reviews. Clinical oncology 11, 335-345, doi:10.1038/nrclinonc.2014.70 (2014).
Hong, S., Youk, T., Lee, S., Kim, K. & Vajdic, C. Bone metastasis and skeletal-related events in patients with solid cancer: A Korean nationwide health insurance database study. PLoS One 15, e0234927, doi:10.1371/journal.pone.0234927 (2020).
Langsteger, W., Rezaee, A., Pirich, C. & Beheshti, M. F-NaF-PET/CT and Tc-MDP Bone Scintigraphy in the Detection of Bone Metastases in Prostate Cancer. Semin. Nucl. Med. 46, 491-501, doi:10.1053/j.semnuclmed.2016.07.003 (2016).
Kabunda, J. et al. Comparing 99mTc-PSMA to 99mTc-MDP in Prostate Cancer Staging of the Skeletal System. Clin. Nucl. Med. 46, 562-568, doi:10.1097/rlu.0000000000003702 (2021).
Briganti, A. et al. Predicting the risk of bone metastasis in prostate cancer. Cancer Treat. Rev. 40, 3-11, doi:10.1016/j.ctrv.2013.07.001 (2014).
Perez-Lopez, R. et al. Volume of Bone Metastasis Assessed with Whole-Body Diffusion-weighted Imaging Is Associated with Overall Survival in Metastatic Castration-resistant Prostate Cancer. Radiology 280, 151-160, doi:10.1148/radiol.2015150799 (2016).
Klaassen, Z. et al. Factors predicting skeletal-related events in patients with bone metastatic castration-resistant prostate cancer. Cancer 123, 1528-1535, doi:10.1002/cncr.30505 (2017).
Yamada, Y. et al. Investigation of risk factors for prostate cancer patients with bone metastasis based on clinical data. Exp. Ther. Med. 1, 635-639, doi:10.3892/etm_00000099 (2010).
Tanaka, N. et al. Bone scan can be spared in asymptomatic prostate cancer patients with PSA of <=20 ng/ml and Gleason score of <=6 at the initial stage of diagnosis. Jpn. J. Clin. Oncol. 41, 1209-1213, doi:10.1093/jjco/hyr118 (2011).
Izumi, K. et al. The relationship between prostate-specific antigen and TNM classification or Gleason score in prostate cancer patients with low prostate-specific antigen levels. The Prostate 75, 1034-1042, doi:10.1002/pros.22985 (2015).
Koo, K. et al. Predictors of survival in prostate cancer patients with bone metastasis and extremely high prostate-specific antigen levels. Prostate international 3, 10-15, doi:10.1016/j.prnil.2015.02.006 (2015).
Hofman, M. et al. Prostate-specific membrane antigen PET-CT in patients with high-risk prostate cancer before curative-intent surgery or radiotherapy (proPSMA): a prospective, randomised, multicentre study. Lancet (London, England) 395, 1208-1216, doi:10.1016/s0140-6736(20)30314-7 (2020).
Hricak, H., Choyke, P., Eberhardt, S., Leibel, S. & Scardino, P. Imaging prostate cancer: a multidisciplinary perspective. Radiology 243, 28-53, doi:10.1148/radiol.2431030580 (2007).
Tabotta, F. et al. Quantitative bone SPECT/CT: high specificity for identification of prostate cancer bone metastases. BMC Musculoskelet. Disord. 20, 619, doi:10.1186/s12891-019-3001-6 (2019).
Chunguang, M. et al. Epidemiology of prostate cancer from three centers and analysis of the first-fine hormonal therapy for the advanced disease. Chin J Surg 46, 921-925 (2008).
Bo, F. et al. Baseline characteristics and survival analysis of patients with prostate cancer in the Fourth Hospital of Hebei Medical University. Chin Oncol 30, 620-625, doi:10.19401/j.cnki.1007-3639.2020.08.009 (2020).
de Camargo Cancela, M., Comber, H. & Sharp, L. Age remains the major predictor of curative treatment non-receipt for localised prostate cancer: a population-based study. Br. J. Cancer 109, 272-279, doi:10.1038/bjc.2013.268 (2013).
Liu, D. et al. Prognosis of prostate cancer and bone metastasis pattern of patients: a SEER-based study and a local hospital based study from China. Sci. Rep. 10, 9104, doi:10.1038/s41598-020-64073-6 (2020).
Merdan, S. et al. Toward better use of bone scans among men with early-stage prostate cancer. Urology 84, 793-798, doi:10.1016/j.urology.2014.06.010 (2014).
Chen, Y. et al. Associations of Prostate-Specific Antigen, Prostate Carcinoma Tissue Gleason Score, and Androgen Receptor Expression with Bone Metastasis in Patients with Prostate Carcinoma. Medical science monitor : international medical journal of experimental and clinical research 23, 1768-1774, doi:10.12659/msm.900977 (2017).
Chaoying, L. et al. Risk factors of bone metastasis in patients with newly diagnosed prostate cancer. Eur. Rev. Med. Pharmacol. Sci. 26, 391-398, doi:10.26355/eurrev_202201_27863 (2022).
Amirrasouli, H. et al. Accurate cut-off point for free to total prostate-specific antigen ratio used to improve differentiation of prostate cancer from benign prostate hyperplasia in Iranian population. Urology journal 7, 99-104 (2010).
Lucarelli, G. et al. Serum sarcosine increases the accuracy of prostate cancer detection in patients with total serum PSA less than 4.0 ng/ml. The Prostate 72, 1611-1621, doi:10.1002/pros.22514 (2012).
Zhu, B., Guo, Z., Lin, L. & Liu, Q. Serum BSP, PSADT, and Spondin-2 levels in prostate cancer and the diagnostic significance of their ROC curves in bone metastasis. Eur. Rev. Med. Pharmacol. Sci. 21, 61-67 (2017).
Xu, Y., Ding, Y., Li, X. & Wu, X. Cystatin C is a disease-associated protein subject to multiple regulation. Immunol. Cell Biol. 93, 442-451, doi:10.1038/icb.2014.121 (2015).
Mathews, P. & Levy, E. Cystatin C in aging and in Alzheimer's disease. Ageing research reviews 32, 38-50, doi:10.1016/j.arr.2016.06.003 (2016).
Cox, J. Cystatins and cancer. Frontiers in bioscience (Landmark edition) 14, 463-474, doi:10.2741/3255 (2009).
Xie, L. et al. Cystatin C increases in cardiac injury: a role in extracellular matrix protein modulation. Cardiovasc. Res. 87, 628-635, doi:10.1093/cvr/cvq138 (2010).
Luo, J. et al. Cystatin C and cardiovascular or all-cause mortality risk in the general population: A meta-analysis. Clinica chimica acta; international journal of clinical chemistry 450, 39-45, doi:10.1016/j.cca.2015.07.016 (2015).
Wegiel, B. et al. Cystatin C is downregulated in prostate cancer and modulates invasion of prostate cancer cells via MAPK/Erk and androgen receptor pathways. PLoS One 4, e7953, doi:10.1371/journal.pone.0007953 (2009).
Hasegawa, A. et al. Regulation of glial development by cystatin C. J. Neurochem. 100, 12-22, doi:10.1111/j.1471-4159.2006.04169.x (2007).
Frendéus, K., Wallin, H., Janciauskiene, S. & Abrahamson, M. Macrophage responses to interferon-gamma are dependent on cystatin C levels. The international journal of biochemistry & cell biology 41, 2262-2269, doi:10.1016/j.biocel.2009.05.005 (2009).
Hu, Y. et al. Role of cystatin C in amyloid precursor protein-induced proliferation of neural stem/progenitor cells. The Journal of biological chemistry 288, 18853-18862, doi:10.1074/jbc.M112.443671 (2013).
Sun, Q. Growth stimulation of 3T3 fibroblasts by cystatin. Exp. Cell Res. 180, 150-160, doi:10.1016/0014-4827(89)90219-x (1989).
Magister, S. & Kos, J. Cystatins in immune system. J. Cancer 4, 45-56, doi:10.7150/jca.5044 (2013).
Lolli, C. et al. Systemic Immune-Inflammation Index Predicts the Clinical Outcome in Patients with mCRPC Treated with Abiraterone. Front. Pharmacol. 7, 376, doi:10.3389/fphar.2016.00376 (2016).
Jun, G. et al. The predictive value of systemic immune-inflammation index for bone metastasis in patients newly diagnosed with prostate cancer. Chin J Urol 42, 752-757, doi:10.3760/cma.j.cn112330-20200403-00259 (2021).
Grivennikov, S., Greten, F. & Karin, M. Immunity, inflammation, and cancer. Cell 140, 883-899, doi:10.1016/j.cell.2010.01.025 (2010).
Singh, R., Mishra, M. & Aggarwal, H. Inflammation, Immunity, and Cancer. Mediators Inflamm. 2017, 6027305, doi:10.1155/2017/6027305 (2017).
Unsal, E., Atalay, F., Atikcan, S. & Yilmaz, A. Prognostic significance of hemostatic parameters in patients with lung cancer. Respir. Med. 98, 93-98, doi:10.1016/j.rmed.2003.07.001 (2004).
Lin, Y. et al. Clinical significance of plasma D-dimer and fibrinogen in digestive cancer: A systematic review and meta-analysis. European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology 44, 1494-1503, doi:10.1016/j.ejso.2018.07.052 (2018).
Fan, S., Guan, Y., Zhao, G. & An, G. Association between plasma fibrinogen and survival in patients with small-cell lung carcinoma. Thoracic cancer 9, 146-151, doi:10.1111/1759-7714.12556 (2018).
Xie, G. et al. Clinical association between pre-treatment levels of plasma fibrinogen and bone metastatic burden in newly diagnosed prostate cancer patients. Chin. Med. J. 132, 2684-2689, doi:10.1097/cm9.0000000000000506 (2019).
Heidenreich, A. et al. EAU guidelines on prostate cancer. part 1: screening, diagnosis, and local treatment with curative intent-update 2013. Eur. Urol. 65, 124-137, doi:10.1016/j.eururo.2013.09.046 (2014).
Lowrance, W. et al. Advanced Prostate Cancer: AUA/ASTRO/SUO Guideline PART I. The Journal of urology 205, 14-21, doi:10.1097/ju.0000000000001375 (2021).
Ho, C. et al. Retrospective study of predictors of bone metastasis in prostate cancer cases. Asian Pacific journal of cancer prevention : APJCP 14, 3289-3292, doi:10.7314/apjcp.2013.14.5.3289 (2013).
Guo, X. et al. The homogeneous and heterogeneous risk factors for the morbidity and prognosis of bone metastasis in patients with prostate cancer. Cancer Manag. Res. 10, 1639-1646, doi:10.2147/cmar.S168579 (2018).

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Multifactor nomogram predicts bone metastasis in patients initially diagnosed with prostate cancer

Status:

Version 1

Abstract

Objectives

Methods

Results

Conclusion

Figures

Introduction

Materials and methods

Inclusion and exclusion criteria

Diagnostic criteria of bone metastasis

Clinical data collection

Statistical analysis

Result

General characteristics of the included participants

Examination of bone metastasis in individuals initially diagnosed with prostate cancer using logistic regression analysis

Interaction analysis and multicollinearity test among the study factors

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1