Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Bone Metastasis in Renal Cell Carcinoma Patients: Risk and Prognostic Factors and Nomograms

Bone Metastasis in Renal Cell Carcinoma Patients: Risk and Prognostic Factors and Nomograms Hindawi Journal of Oncology Volume 2021, Article ID 5575295, 17 pages https://doi.org/10.1155/2021/5575295 Research Article Bone Metastasis in Renal Cell Carcinoma Patients: Risk and Prognostic Factors and Nomograms 1,2 2 1 Zhiyi Fan, Zhangheng Huang, and Xiaohui Huang Hangzhou Medical College, Hangzhou, Zhejiang Province, China Department of Spine Surgery, Affiliated Hospital of Chengde Medical University, Chengde, Hebei Province, China Correspondence should be addressed to Xiaohui Huang; 373644723@qq.com Received 10 February 2021; Revised 7 April 2021; Accepted 27 April 2021; Published 12 May 2021 Academic Editor: Liren Qian Copyright © 2021 Zhiyi Fan et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Bone metastasis (BM) is one of the common sites of renal cell carcinoma (RCC), and patients with BM have a poorer prognosis. We aimed to develop two nomograms to quantify the risk of BM and predict the prognosis of RCC patients with BM. Methods. We reviewed patients with diagnosed RCC with BM in the Surveillance, Epidemiology, and End Results (SEER) database from 2010 to 2015. Multivariate logistic regression analysis was used to determine independent factors to predict BM in RCC patients. Univariate and multivariate Cox proportional hazards regression analyses were used to determine independent prognostic factors for BM in RCC patients. Two nomograms were established and evaluated by calibration curve, receiver operating characteristic (ROC) curve, and decision curve analysis (DCA). Results. +e study included 37,554 patients diagnosed with RCC in the SEER database, 537 of whom were BM patients. BM’s risk factors included sex, tumor size, liver metastasis, lung metastasis, brain metastasis, N stage, T stage, histologic type, and grade in RCC patients. Currently, independent prognostic factors for RCC with BM included grade, histologic type, N stage, surgery, brain metastasis, and lung metastasis. +e calibration curve, ROC curve, and DCA showed good performance for diagnostic and prognostic nomograms. Conclusions. Nomograms were established to predict the risk of BM in RCC and the prognosis of RCC with BM, separately. +ese nomograms strengthen each patient’s prognosis-based decision making, which is critical in improving the prognosis of patients. treatment plans [10]. Studies have shown that race, sex, age, 1. Introduction and tumor size may also affect the prognosis of patients with Renal cell carcinoma (RCC) is one of the most common RCC [11–13]. +e TNM staging system relies on three cancers worldwide, with approximately 403,262 new cases pathological indicators and ignores other prognostic factors, and 17,598 deaths in 2018 [1]. Approximately 15–30% of thereby reducing the accuracy of prognostic prediction for RCC patients have metastases at the initial diagnosis, and RCC patients. +erefore, it is necessary to combine clin- bone is a common site of metastasis [2, 3]. Bone metastasis icopathology and other prognosis-related variables to con- (BM) from RCC is predominantly osteolytic and can lead to struct a tool to accurately predict the prognosis and overcome the limitations of the traditional TNM staging skeletal-related diseases, which can reduce the quality of life and prognosis of the patients [4, 5]. +e median overall system. survival (OS) of RCC patients with BM has been reported to Nomogram is a tool that combines multiple biological be only 12–28 months [6, 7]. In contrast, patients with and clinical variables to predict specific endpoints and has metastatic RCC without BM had a more prolonged median been widely used to predict the prognosis of cancer patients OS to 31 months [8, 9]. +erefore, understanding the BM in [14–16]. By combining these important variables, the no- RCC patients is an unmet need. mograms can individually estimate the probability of events +e TNM staging system is widely used to assess the over time, such as the OS of cancer patients. In addition, prognosis of cancer patients, and clinicians use it to develop nomograms can be used to estimate the survival rate of 2 Journal of Oncology tumor size in terms of OS were determined by X-tile soft- cancer patients with higher accuracy than the TNM staging system [17]. ware, and patients were divided into three groups (<4, 4–7, and>7 cm). +e histologic type was defined by the following Risk factors and prognosis-related factors for BM in RCC have been reported in several previous studies [18–20]. ICD-O-3 codes: clear cell (8310/3, 8313/3), papillary (8260/ However, no studies have focused on constructing predictive 3), chromophobe (8317/3, 8270/3), and collecting duct models for the risk and prognosis of BM in RCC, which (8319/3). Regarding marital status, we excluded misleading means that the probability of outcome cannot be quantified. data on unmarried or domestic partners and then included +erefore, based on the data from the Surveillance, Epi- “unmarried,” “separated,” “single,” and “widowed” all in the demiology, and End Results (SEER) database, we developed unmarried group. Insurance status is divided into insured two nomograms for predicting the risk of BM with RCC and and uninsured, with both “insured” and “insured/unspe- the OS of RCC patients with BM, separately. cific” included in the insured group. All cases in this study were staged using version 7 of the American Joint Com- mittee on Cancer TNM staging system. In the survival 2. Methods analysis, the primary endpoint of our study was OS, which was defined as the date from diagnosis to death (for any 2.1. Study Population Selection. +e SEER database covers reason) or the date of the last follow-up. approximately 28% of cancer registries in the United States [21]. +e data contained in this study were downloaded from the SEER Stat software version 8.3.6. Analysis of anony- mous data from the SEER database is exempt from medical 2.3. Statistical Analysis. +is study used SPSS 25.0 and R ethics review and does not require informed consent. +e software (version 3.6.1) for statistical analysis. +e chi- SEER database provides clinical information on cancer square test was used for categorical data. Variables with P patients that greatly facilitate clinical research. Patients di- values <0.05 in univariate analysis were incorporated into a agnosed before 2010 were excluded because the SEER da- multivariate logistic regression analysis to identify inde- tabase did not record information on distant metastases pendent risk factors for BM in RCC patients. At the same until 2010. In addition, to ensure adequate follow-up time, time, univariate Cox proportional hazards regression patients diagnosed after 2015 are not included. +erefore, analysis was used to determine OS-related variables. Sig- only patients diagnosed with RCC between 2010 and 2015 nificant variables in the univariate Cox proportional hazards were considered in this study. regression analysis were then included in the multivariate Inclusion criteria were as follows: (1) RCC as the first Cox proportional hazards regression analysis to identify primary tumor, (2) patients with a histologic diagnosis of independent prognostic factors in RCC patients with BM. RCC, and (3) patients with complete clinicopathological Nomograms were developed separately based on in- features, demographic information, and follow-up infor- dependent BM-related predictors and prognostic factors mation. In addition, patients who were certified by autopsy using the “rms” package in R software. In the nomograms, or death were excluded from this study. Finally, a total of values for the individual patient were located along the 37,554 patients with RCC were enrolled to study the risk variable axes, and a line was drawn upward to the points factors for BM in patients with RCC and to establish a axis to determine the number of points assigned for each diagnostic nomogram. Subsequently, for RCC patients with variable. +ere was a total points line at the bottom of the BM with survival time≥ one month, specific treatment in- nomogram, and each variable score was summed to give formation, including surgery, radiotherapy, and chemo- the total points. Receiver operating characteristic (ROC) therapy, were used to form a new cohort to explore the curves for two nomograms were generated, and the cor- prognostic factors for RCC patients with BM and develop a responding area under the curve (AUC) was used to prognostic nomogram. Ultimately, 537 patients were used to evaluate the discrimination of nomograms. +e clinical study prognostic factors in patients with BM from RCC. application value of the nomogram model was evaluated by Patients in each cohort were randomized into training and calibration curve and decision curve analysis (DCA). Fi- validation cohorts in a 7 : 3 ratio. In this study, patients in the nally, all patients were divided into high-risk and low-risk training cohort were used to construct the predicted no- groups according to the median of risk score, and survival mogram, while patients in the validation cohort were used to curves were used to verify the prognostic value of the validate the constructed nomogram. nomogram [22]. 2.2. Data Collection. Based on patient-specific information 3. Results from the SEER database, we selected 14 variables to identify risk factors for BM in RCC, including age, sex, race, tumor 3.1. )e Characteristics of the Study Population. +e work- size, histologic type, grade, laterality, Tstage, N stage, distant flow of our study is illustrated in Figure 1. A total of 37,554 metastatic site (lung, brain, liver), insurance status, and RCC patients from the SEER database were included. marital status. In addition to the aforementioned variables, Furthermore, 26,290 and 11,264 patients were included in information on surgery, radiotherapy, and chemotherapy the training and validation cohorts, respectively. Clinico- are included to study the factors that influence the prognosis pathological information of 26,290 RCC patients is given in of RCC patients with BM. +e optimal cutoff values for Table 1. Journal of Oncology 3 RCC patients (N = 89795) Diagnostic cohort RCC patients with BM (N = 37554) (N = 1852) Survival study cohort (N = 537) Validation cohort Training cohort Training cohort Validation cohort (N = 11264) (N = 26290) (N = 377) (N = 160) Univariate analysis Univariate cox Multivariate analysis Multivariate cox Predictor of BM Prognostic variables Diagnostic nomogram Prognostic nomogram ROC curve Calibration curve DCA Survival curve Figure 1: +e workflow describing the schematic overview of the project. 3.2. Risk Factors of BM in RCC Patients. To identify BM- As shown in Figure 5, the AUC of the nomogram is higher related variables in RCC patients, 14 factors were analyzed. than the AUCs of all independent variables in both training +e results showed that ten factors were related to the BM in and validation cohorts, indicating a significant advantage in RCC patients, including race, sex, grade, histologic type, T the accuracy of predictions using the nomogram compared to predictions using individual independent predictors. stage, N stage, brain metastasis, liver metastasis, lung me- tastasis, and tumor size (Table 1). Subsequently, the above variables were included in the multivariate logistic regres- 3.4. Prognostic Factors for RCC Patients with BM. sion analysis, which showed that tumor size, liver metastasis, lung metastasis, brain metastasis, N stage, T stage, histologic According to the selection process, a total of 537 patients with BM were included in our research. Meanwhile, 377 patients type, and grade were independent predictors of RCC with BM (Table 2). were incorporated into the training cohort, and the remaining 160 patients were incorporated into the validation cohort. Univariate and multivariate Cox proportional hazards regres- 3.3. Development and Validation of a Nomogram for BM in sion analyses were performed to screen for prognostic factors. Univariate Cox proportional hazards regression analysis Newly Diagnosed RCC Patients. Based on eight independent BM-related variables, a nomogram was constructed to assess showed that grade, T stage, histologic type, N stage, surgery, the risk of BM in RCC patients (Figure 2). +e AUCs of the chemotherapy, brain metastasis, liver metastasis, and lung nomogram were 0.865 and 0.859 in the training and validation metastasis are OS-related factors (Table 3). After controlling for cohorts, respectively, showing good discrimination confounding variables using multivariate Cox proportional (Figure 3(a) and Figure 4(a)). +e calibration curve showed hazards regression analysis, grade, histologic type, N stage, that the observations are highly consistent with the predicted surgery, brain metastasis, and lung metastasis were identified as results (Figure 3(b) and Figure 4(b)). Moreover, DCA indi- independent prognostic factors in RCC patients with BM cated that the diagnostic nomogram performs well in clinical (Table 3). As shown in Figure 6, the survival curve analysis practice (Figure 3(c) and Figure 4(c)). Importantly, ROC further demonstrated the impact of screened independent curves were generated for each independent predictor variable. prognostic factors on the OS of RCC patients with BM. Validation Validation 4 Journal of Oncology Table 1: Demographic and clinical characteristics of RCC patients. Table 1: Continued. 2 2 Without BM With BM χ P Without BM With BM χ P Age 1.463 0.226 25089 Yes 437 (96.9%) 14388 (97.1%) <65 264 (58.5%) (55.7%) Marital status 3.481 0.062 No 8109 (31.4%) 123 (27.3%) ≥65 187 (41.5%) (44.3%) Yes 328 (72.7%) Sex 6.982 0.008 (68.6%) Female 9111 (35.3%) 132 (29.3%) BM, bone metastasis; RCC, renal cell carcinoma; CP, chromophobe; CC, clear cell; CD, collecting duct; PL, papillary. Male 319 (70.7%) (64.7%) Race 6.289 0.043 Table 2: Multivariate logistic regression analysis of BM in RCC Black 2648 (10.2%) 30 (6.7%) patients. Others 1585 (6.1%) 28 (6.2%) Variables OR (95% CI) P value White 393 (87.1%) (83.6%) Grade Grade 236.193 ≤0.001 G1-2 Reference I-II 16818 (65.1%) 136 (30.2%) G3-4 1.749 (1.388–2.204) ≤0.001 III-IV 9021 (34.9%) 315 (69.8%) T stage T Stage 437.434 ≤0.001 T1-2 Reference T3-4 1.748 (1.379–2.216) ≤0.001 T1-2 182 (40.4%) (80.3%) Histologic type T3-4 5079 (19.7%) 269 (59.6%) CP Reference Laterality 1.692 0.193 CC 3.300 (1.459–7.466) 0.004 CD 4.216 (1.245–14.277) 0.021 Left 236 (52.3%) (49.2%) PL 1.850 (0.762–4.490) 0.174 Tumor size, cm Right 215 (47.7%) (50.8%) <4 Reference Histologic type 79.734 ≤0.001 4–7 3.937 (2.745–5.648) ≤0.001 CP 1416 (5.5%) 6 (1.3%) >7 3.510 (2.374–5.189) ≤0.001 N stage CC 407 (90.2%) (78.3%) N0 Reference CD 54 (0.2%) 7 (1.6%) N1 2.654 (2.005–3.513) ≤0.001 PL 4125 (16.0%) 31 (6.9%) Brain metastasis Tumor size, cm 392.797 ≤0.001 No Reference Yes 4.283 (2.780–6.598) ≤0.001 <4 39 (8.6%) (46.8%) Liver metastasis 4–7 8559 (33.1%) 167 (37.0%) No Reference >7 5193 (20.1%) Yes 3.309 (2.211–4.952) ≤0.001 (54.3.0%) Lung metastasis N stage 733.940 ≤0.001 No Reference N0 345 (76.5%) Yes 5.351 (4.123–6.946) ≤0.001 (97.6%) BM, bone metastasis; RCC, renal cell carcinoma; CP, chromophobe; CC, N1 621 (2.4%) 106 (23.5%) clear cell; CD, collecting duct; PL, papillary. Brain 679.401 ≤0.001 metastasis 3.5. Prognostic Nomogram for RCC Patients with BM. A No 408 (90.5%) (99.6%) prognostic nomogram of RCC patients with BM based on six Yes 98 (0.4%) 43 (9.5%) independent prognostic factors was established (Figure 7). Liver metastasis 629.229 ≤0.001 +e ROC curve showed that the AUCs at 1, 2, and 3 years were 0.711, 0.772, and 0.766 in the training cohort and 0.684, No 404 (89.6%) (99.5%) 0.663, and 0.691 in the validation cohort (Figure 8(a) and Yes 132 (0.5%) 47 (10.4%) 8(c)). +e optimal cutoff point for the total score was de- Lung metastasis 1835.076 ≤0.001 termined by X-tile software and was 285. +erefore, we specified less than 285 as the low-risk group and greater than No 274 (60.8%) (97.3%) 285 as the high-risk group. By depicting the Kaplan–Meier Yes 700 (2.7%) 177 (39.2%) survival curve, we can find that patients in the high-risk Insurance 0.064 0.800 group showed a worse prognosis than patients in the low- status risk group (Figure 8(b) and 8(d)). In addition, we further No 750 (2.9%) 14 (3.1%) compared the discrimination between the nomogram and Journal of Oncology 5 Glm regression Points 0 10 20 30 40 50 60 70 80 90 100 ∗∗∗ >7 Tumor size <4 4–7 ∗∗∗ Lung metastasis Yes No ∗∗∗ Liver metastasis Yes No ∗∗∗ Brain metastasis Yes No ∗∗∗ N stage N1–2 No ∗∗∗ T stage T3–4 T1-2 ∗∗ Papillary adenocardnoma Collecting duct cardnoma Histologic type Chromophobe cardnoma Clear cell adenocardnoma T1–2 ∗∗∗ III-IV Grade I-II Total-points-to-outcome nomogram: Total points 0 50 100 150 200 250 300 350 400 450 500 0.00988 Odds (bone) 5e – 04 0.001 0.0015 0.0025 0.004 0.006 0.01 0.015 0.03 0.06 0.1 0.15 0.3 0.6 1 1.5 2.5 Figure 2: Nomogram to estimate the risk of BM in patients with RCC. 0.7 AUC = 0.865 0.02 1.0 0.6 0.01 0.5 0.8 0.4 –0.01 0.6 0.3 0.4 0.2 –0.03 0.1 0.2 0.0 –0.05 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Nomogram-predicted reshold probability False positive rate probability of BM Nomogram Apparent All Bias-corrected None Ideal (a) (b) (c) Figure 3: ROC curves (a), calibration curves (b), and DCA (c) of the training cohort. AUC = 0.859 0.02 0.6 1.0 0.01 0.5 0.8 0.4 –0.01 0.6 0.3 0.2 0.4 –0.03 0.1 0.2 0.0 –0.05 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Nomogram-predicted reshold probability False positive rate probability of BM Nomogram Apparent All Bias-corrected None Ideal (a) (b) (c) Figure 4: ROC curves (a), calibration curves (b), and DCA (c) of the validation cohort. True positive rate True positive rate Actual BM Actual BM Net benefit Net benefit 6 Journal of Oncology Statistical comparison Statistical comparison 100 100 80 80 60 60 40 40 20 20 0 0 100 80 60 40 20 0 100 80 60 40 20 0 Specificity (%) Specificity (%) Nomogram Brain metastasis Nomogram Brain metastasis Grade Liver metastasis Grade Liver metastasis Histologic type Lung metastasis Histologic type Lung metastasis T stage Tumor size T stage Tumor size N stage N stage (a) (b) Figure 5: Comparison of AUC between diagnostic nomogram and each independent predictor in the training cohort (a) and the validation cohort (b). Table 3: Univariate and multivariate Cox analyses in RCC patients with BM. Univariate Cox analysis Multivariate Cox analysis HR 95% CI P HR 95% CI P Age <65 ≥65 1.026 0.798 1.319 0.844 Race Black Others 1.660 0.913 3.020 0.097 White 1.293 0.827 2.023 0.260 Sex Female Male 1.109 0.847 1.452 0.451 Grade I-II III-IV 1.411 1.074 1.853 0.013 1.669 1.235 2.257 ≤0.001 T stage T1-2 T3-4 1.388 1.080 1.785 0.010 Laterality Left Right 1.099 0.864 1.397 0.443 Histologic type CP CC 0.848 0.537 1.339 0.479 0.706 0.440 1.132 0.148 CD 1.157 0.487 2.746 0.742 0.723 0.300 1.743 0.471 PL 2.767 1.167 6.557 0.021 2.492 1.036 5.994 0.041 Sensitivity (%) Sensitivity (%) Journal of Oncology 7 Table 3: Continued. Univariate Cox analysis Multivariate Cox analysis HR 95% CI P HR 95% CI P Tumor size, cm <4 4–7 1.125 0.683 1.851 0.644 >7 1.323 0.821 2.132 0.250 N stage N0 N1 1.791 1.378 2.328 ≤0.001 1.388 1.049 1.838 0.022 Surgery No Yes 0.416 0.313 0.552 ≤0.001 0.394 0.284 0.546 ≤0.001 Radiotherapy No Yes 1.103 0.862 1.411 0.438 Chemotherapy No Yes 1.463 1.128 1.898 0.004 Brain metastasis No Yes 2.315 1.575 3.403 ≤0.001 1.801 1.201 2.700 0.004 Liver metastasis No Yes 1.960 1.349 2.847 ≤0.001 Lung metastasis No Yes 2.261 1.771 2.887 ≤0.001 1.745 1.342 2.269 ≤0.001 Insurance status No Yes 1.227 0.628 2.396 0.549 Marital status No Yes 0.994 0.768 1.286 0.961 BM, bone metastasis; RCC, renal cell carcinoma; CP, chromophobe; CC, clear cell; CD, collecting duct; PL, papillary. the independent prognostic factors, and the results showed analyzing massive data, respectively. We believe that two that the AUC of the nomogram was higher than the AUCs of nomograms representing OS and distant metastasis are complementary and can increase their clinical value in all independent factors at 1, 2, and 3 years, both in the training cohort and in the validation cohort (Figure 9). patients with RCC. +e total score can be calculated by Calibration curves of predicting 1, 2, and 3-year OS prob- obtaining data for each RCC patient’s corresponding abilities also show good agreement between the OS predicted variable on the nomogram. +e risk of BM can then be by the prognostic nomogram and the actual results easily identified on the diagnostic nomogram, identifying (Figure 10(a) and 10(b)). +e DCA was used to evaluate the patients in the high-risk group and guiding clinical practice clinical utility of a nomogram. As shown in Figure 10, the in early intervention. Similarly, the prognosis of RCC prognostic nomogram shows a significant positive net patients with BM can be determined from the prognostic benefit over a wide range of mortality risks, suggesting its nomogram. In the validation of the two nomograms, the high clinical utility in predicting OS in RCC patients with two nomograms showed excellent performance in BM risk BM. assessment and OS prediction in RCC patients, respec- tively, which will enable more accurate personalized clinical decision making and monitoring. 4. Discussion Despite the poor prognosis of RCC patients with BM, RCC accounts for 3% of all malignancies and 80%–85% of early detection of BM may be critical for patients with RCC primary renal cancer [23]. Bone is the second most to receive appropriate treatment. +erefore, exploring the common site of metastasis in RCC patients, following the risk factors for BM in RCC patients is important for clinical lung [24, 25]. In the present study, we constructed diag- decision making. At the molecular level, cadherin-11, nostic and prognostic nomograms to predict the risk of BM transforming growth factor-β, insulin-like growth factor, in RCC patients and the OS of RCC patients with BM by and the fibroblast growth factor have been associated with 8 Journal of Oncology 1.00 0.75 0.50 0.25 p < 0.0001 0.00 020 40 60 80 Time Brain Brain = no Brain = yes (a) 1.00 0.75 0.50 0.25 p = 0.012 0.00 020 40 60 80 Time Grade Grade = I-II Grade = III-IV (b) Figure 6: Continued. Survival probability Survival probability Journal of Oncology 9 1.00 0.75 0.50 0.25 p = 0.0084 0.00 020 40 60 80 Time Hist CP CO CC PL (c) 1.00 0.75 0.50 0.25 p < 0.0001 0.00 020 40 60 80 Time Lung Lung = no Lung = yes (d) Figure 6: Continued. Survival probability Survival probability 10 Journal of Oncology 1.00 0.75 0.50 0.25 p < 0.0001 0.00 020 40 60 80 Time N = N0 N = N1 (e) 1.00 0.75 0.50 0.25 p < 0.0001 0.00 020 40 60 80 Time Surgery Surgery = no Surgery = yes (f) Figure 6: Survival curves for each independent prognostic factor in the training cohort. CP, chromophobe; CC, clear cell; CD, collecting duct; PL, papillary. BM in RCC patients [26, 27]. Nevertheless, these biomarkers for patients to afford. Of course, if molecular level indicators are difficult and impractical to apply immediately to clinical could be included in the nomogram, this would undoubtedly decision making. In our daily clinical work, it is difficult for increase the predictive accuracy of the nomogram, which us to examine every patient at the molecular level because it could lead to better survival for patients. In addition, re- requires a lot of human and material resources. At the same garding some practical clinical features, sex, T stage, N stage, time, the high cost of testing at the molecular level is difficult grade, liver metastasis, lung metastasis, brain metastasis, and Survival probability Survival probability Journal of Oncology 11 Coxph regression Points 010 20 30 40 50 60 70 80 90 100 N stage N1-2 No ∗∗ Brain metastasis Yes No Chromophobe type Histologic type Collecting duct cardnoma Clear cell adenocardnoma Papillary adenocardnoma ∗∗∗ Grade III-IV I-II Lung metastasis Yes No ∗∗∗ Surgery No Yes Total-points-to-outcome nomogram: Total points 220 240 260 280 300 320 340 360 380 400 420 0.247 Pr (futime < 12) 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.85 0.9 0.94 0.97 0.985 0.995 0.403 Pr (futime < 24) 0.2 0.3 0.4 0.5 0.6 0.7 0.85 0.9 0.94 0.97 0.985 0.994 0.998 0.543 Pr (futime < 36) 0.3 0.4 0.5 0.6 0.7 0.85 0.9 0.94 0.97 0.985 0.994 0.998 Figure 7: Nomogram to predict the OS of RCC patients with BM. 1.00 1.00 0.75 0.50 0.25 0.75 0.00 020 40 60 80 Time in months 0.50 Strata Risk = high Risk = low 0.25 Number at risk Risk = high 180 45 11 3 0 Risk = low 197 114 54 16 1 0.00 0 20 40 60 80 Time in months 0.00 0.25 0.50 0.75 1.00 FP Number at censoring AUC 4 12 months AUC: 0.711 24 months AUC: 0.772 0 20 40 60 80 36 months AUC: 0.766 Time in months (a) (b) Figure 8: Continued. TP Strata Survival probability n. censor 12 Journal of Oncology 1.00 1.00 0.75 0.50 0.25 0.75 0.00 020 40 60 80 Time in months 0.50 Strata Risk = high Risk = low 0.25 Number at risk Risk = high 68 21 8 1 0 Risk = low 92 44 21 5 1 0.00 020 40 60 80 Time in months 0.00 0.25 0.50 0.75 1.00 FP Number at censoring AUC 2 12 months AUC: 0.684 24 months AUC: 0.663 0 20 40 60 80 36 months AUC: 0.691 Time in months (c) (d) Figure 8: (a) Receiver operating characteristic curves of 1, 2, and 3 years in the training cohort. (b) +e Kaplan–Meier survival curve of the training cohort. (c) Receiver operating characteristic curves of 1, 2, and 3 years in the validation cohort. (d) +e Kaplan–Meier survival curve of the validation cohort. histologic type have been reported as relevant risk factors for effective tool for identifying high-risk patients. +e impact of BM in RCC [20]. However, to date, no predictive model has histologic type on metastatic potential and prognosis of been developed, which means that it is impossible to identify metastatic patients is often overlooked when discussing an individual’s risk of BM by combining all independent treatment options. In this study, collecting duct RCC had a predictors associated with BM. +e present study showed higher incidence of BM and a worse prognosis compared to that tumor size, liver metastasis, lung metastasis, brain other renal cancer subtypes. Collecting duct RCC is reported metastasis, N stage, T stage, histologic type, and the grade to be a rare entity that occurs in <2% of patients with kidney cancer, often resulting in a poor prognosis [28]. In addition, were significant predictors of BM in RCC. +e association between these factors and BM in RCC patients has been the above correlation has been confirmed in previous studies reported in previous studies. Although metastasis to mul- [29, 30]. +e relationship between lung metastasis, brain tiple organs is a risk factor for BM in patients with RCC, metastasis, surgery, and prognosis in patients with RCC has unfortunately, we were unable to obtain the sequence of also been widely reported in previous studies. Lin et al. organ metastasis due to the shortcomings of the SEER reported a better prognosis in patients with only BM than in database itself. Previous studies have confirmed the rela- patients with concomitant pulmonary metastases and a tionship between tumor grade, TNM staging, and BM in significantly better prognosis for patients with single BM RCC patients [20]. TNM staging is widely used in the as- than in patients with multiple bones and/or visceral me- tastasis [31]. Similarly, Toyoda et al. reported a shorter sessment of prognosis in cancer patients. Notably, a more significant contribution of TNM staging was shown in both median survival in patients with extra-BM compared to those without (8 vs. 33 months, P � 0.0084) [32]. Surpris- the diagnostic nomogram and the prognostic nomogram. With increasing tumor size, an increasing number of lymph ingly, contrary to previous reports, the presence of liver node metastases, and distant organ metastases, the risk of metastasis was not an independent prognostic factor in our BM in RCC and the risk of death in RCC patients with BM study [33, 34]. However, this is consistent with what has are significantly increased. been reported by Santoni et al. [19]. Previous studies have In addition, our study found a poor prognosis of patients reported age as a factor associated with patient prognosis with lymph node metastasis, brain metastasis, lung metas- regarding RCC, but other studies show no difference in tasis, without surgery, poor tumor differentiation, and prognosis between younger and older patients with RCC histologic type of the collecting duct. A prognostic nomo- [35, 36]. Some of these studies included only a restricted age group of patients or limited sample size or follow-up time. gram was established based on six independent prognostic factors. +e results suggested that a nomogram can be an +us, until now, the role of age as a prognostic factor in TP Strata Survival probability n. censor Journal of Oncology 13 1.0 1.0 1.0 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 False positive rate False positive rate False positive rate Nomogram (AUC = 0.711) Nomogram (AUC = 0.772) Nomogram (AUC = 0.755) Grade (AUC = 0.541) Grade (AUC = 0.538) Grade (AUC = 0.549) Hist (AUC = 0.526) Hist (AUC = 0.532) Hist (AUC = 0.528) N (AUC = 0.592) N (AUC = 0.608) N (AUC = 0.604) Surgery (AUC = 0.596) Surgery (AUC = 0.611) Surgery (AUC = 0.608) Brain (AUC = 0.539) Brain (AUC = 0.560) Brain (AUC = 0.551) Lung (AUC = 0.637) Lung (AUC = 0.675) Lung (AUC = 0.670) (a) (b) (c) 1.0 1.0 1.0 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 False positive rate False positive rate False positive rate Nomogram (AUC = 0.684) Nomogram (AUC = 0.663) Nomogram (AUC = 0.691) Grade (AUC = 0.540) Grade (AUC = 0.508) Grade (AUC = 0.529) Hist (AUC = 0.509) Hist (AUC = 0.500) Hist (AUC = 0.549) N (AUC = 0.526) N (AUC = 0.517) N (AUC = 0.564) Surgery (AUC = 0.555) Surgery (AUC = 0.577) Surgery (AUC = 0.562) Brain (AUC = 0.596) Brain (AUC = 0.543) Brain (AUC = 0.543) Lung (AUC = 0.628) Lung (AUC = 0.634) Lung (AUC = 0.641) (d) (e) (f) Figure 9: +e receiver operating characteristic curves of nomogram and all independent predictors at 1 (a), 2 (b), and 3 years (c) in the training cohort and at 1 (d), 2 (e), and 3 years (f) in the validation cohort. patients with RCC has been controversial. +e study in- but that patients who had surgery showed a better prognosis. cluded as many factors as possible that may be associated As reported in several studies, surgical removal of isolated or with the prognosis of patients with RCC and identified the minimally metastatic lesions can improve the prognosis of relevant prognostic factors by rigorous statistical methods, patients with BM, thus providing a multidisciplinary team to so the results are trustworthy. However, due to the retro- support the treatment plan for these patients [38–40]. Al- though renal cancer is usually not sensitive to radiotherapy spective nature of the study, selection bias is inevitable. For the treatment of RCC patients with BM, recent consensus and chemotherapy, palliative radiotherapy can significantly suggests using a multimodal treatment strategy that includes relieve local symptoms and improve quality of life [41, 42]. extensive resection of the lesion, radiotherapy, systemic Tyrosine kinase inhibitors (TKIs) and antivascular endo- therapy, and other local treatment options [37]. Of RCC thelial growth factor antibodies are now widely used as first- patients with BM, surgical treatment aims to improve the and second-line therapy for advanced RCC. Direct evidence prognosis, local tumor control, pain relief, and preservation on the effects of targeted drugs on BM is currently limited to or reconstruction of function. Based on the results, we found a few studies that have shown that TKIs can prolong the that surgery was not only an independent prognostic factor mean time to progression of existing bone lesions and True positive rate True positive rate True positive rate True positive rate True positive rate True positive rate 14 Journal of Oncology 1-year 2-years 3-years 1.0 1.0 1.0 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Nomogram-predicted OS Nomogram-predicted OS Nomogram-predicted OS (a) 1-year 2-years 3-years 1.0 1.0 1.0 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Nomogram-predicted OS Nomogram-predicted OS Nomogram-predicted OS (b) 1-year 2-years 3-years 0.4 0.5 0.6 0.3 0.4 0.4 0.3 0.2 0.2 0.1 0.2 0.1 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 reshold probability reshold probability reshold probability None None None All All All Nomogram Nomogram Nomogram (c) Figure 10: Continued. Net benefit Actual OS Net benefit Actual OS Net benefit Actual OS Journal of Oncology 15 1-year 2-years 3-years 0.6 0.5 0.3 0.5 0.4 0.4 0.2 0.3 0.3 0.2 0.1 0.2 0.1 0.1 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 reshold probability reshold probability reshold probability None None None All All All Nomogram Nomogram Nomogram (d) Figure 10: (a) +e calibration curves of the prognostic nomogram in the training cohort. (b) +e calibration curves of the prognostic nomogram in the validation cohort. (c) +e decision curve analysis of the prognostic nomogram in the training cohort. (d) +e decision curve analysis of the prognostic nomogram in the validation cohort. reduce the formation of new bone lesions [2, 43]. Unfor- Abbreviations tunately, the SEER database does not contain specific ana- AUC: Area under the curve lyses of targeted therapies, chemotherapy, and radiotherapy, BM: Bone metastasis and we are unable to analyze their influence on prognosis in DCA: Decision curve analysis further detail. In addition, further research on important OS: Overall survival prognostic factors for OS with BM in RCC is necessary. ROC: Receiver operating characteristic However, some limitations of our study should be noted. RCC: Renal cell carcinoma First, information collected in the SEER database is about the SEER: Surveillance, Epidemiology, and End Results disease at the first diagnosis and does not record BM that TKIs: Tyrosine kinase inhibitors. occurred later. Second, the prognostic impact of the amount of BM should not be overlooked, but there is no record of this in the SEER database. +ird, we did not have access to some Data Availability biomarkers from the SEER database, such as transforming +e dataset from the SEER database that was generated and/ growth factor-β, insulin-like growth factor, and fibroblast growth factor. Fourth, this was a retrospective study in which or analyzed during the current study is available in the SEER selection bias was inevitable, and detailed treatment was not dataset repository (https://seer.cancer.gov/). available in the SEER database. Immunotherapy was rec- ommended for patients with RCC because of its OS benefit, Ethical Approval but, unfortunately, the SEER database does not contain this information. As such, the validity of this data is not any more Approval was waived by the local ethics committee, as SEER known. In addition, since the construction and validation data are publicly available and deidentified. cohorts are from the same database, it is still necessary to validate the accuracy of the nomograms in other databases. Conflicts of Interest +e authors declare that they have no conflicts of interest. 5. Conclusions Two nomograms we created could be used as a supportive Authors’ Contributions graphic tool in RCC patients to help clinicians distinguish, assess, and evaluate the risk and prognosis of RCC with BM. ZY F, ZH H, and XH H conceived and designed the study. At the same time, when faced with individualized condition ZH H performed the literature search. ZY F generated the consultation, these nomograms are valuable methods to figures and tables. ZY F and ZH H analyzed the data. ZY F provide prognostic information to clinical patients and and ZH H wrote the article, and XH H critically reviewed the strengthen each patient’s prognosis-based decision making, article. ZY F and XH H supervised the research. All authors which is of great significance in improving the prognosis of have read and approved the final draft of the article. Zhiyi patients. Fan and Zhangheng Huang contributed equally to this work. Net benefit Net benefit Net benefit 16 Journal of Oncology [16] M. W. Kattan, V. Reuter, R. J. Motzer, J. Katz, and P. Russo, “A References postoperative prognostic nomogram for renal cell carci- [1] F. Bray, J. Ferlay, I. Soerjomataram, R. L. Siegel, L. A. Torre, noma,” )e Journal of Urology, vol. 166, no. 1, pp. 63–67, 2001. and A. Jemal, “Global cancer statistics 2018: GLOBOCAN [17] N. Pu, J. Li, Y. Xu et al., “Comparison of prognostic prediction estimates of incidence and mortality worldwide for 36 cancers between nomogram based on lymph node ratio and AJCC 8th in 185 countries,” CA: A Cancer Journal for Clinicians, vol. 68, staging system for patients with resected pancreatic head no. 6, pp. 394–424, 2018. carcinoma: a SEER analysis,” Cancer Management and Re- [2] S. L. Wood and J. E. Brown, “Skeletal metastasis in renal cell search, vol. 10, pp. 227–238, 2018. carcinoma: current and future management options,” Cancer [18] H. Kume, S. Kakutani, Y. Yamada et al., “Prognostic factors Treatment Reviews, vol. 38, no. 4, pp. 284–291, 2012. for renal cell carcinoma with bone metastasis: who are the [3] K. Gupta, J. D. Miller, J. Z. Li, M. W. Russell, and long-term survivors?” Journal of Urology, vol. 185, no. 5, C. Charbonneau, “Epidemiologic and socioeconomic burden pp. 1611–1614, 2011. of metastatic renal cell carcinoma (mRCC): a literature re- [19] M. Santoni, A. Conti, G. Procopio et al., “Bone metastases in view,” Cancer Treatment Reviews, vol. 34, no. 3, pp. 193–205, patients with metastatic renal cell carcinoma: are they always associated with poor prognosis?” Journal of Experimental & [4] J. Zekri, N. Ahmed, R. E. Coleman, and B. W. Hancock, “+e Clinical Cancer Research, vol. 34, no. 1, p. 10, 2015. skeletal metastatic complications of renal cell carcinoma,” [20] Q. Guo, C. Zhang, X. Guo et al., “Incidence of bone metastasis International Journal of Oncology, vol. 19, no. 2, pp. 379–382, and factors contributing to its development and prognosis in newly diagnosed renal cell carcinoma: a population-based [5] D. Y. C. Heng, W. Xie, G. A. Bjarnason et al., “Progression- study,” Cancer Management and Research, vol. 10, free survival as a predictor of overall survival in metastatic pp. 2935–2944, 2018. renal cell carcinoma treated with contemporary targeted [21] K. A. Cronin, L. A. G. Ries, and B. K. Edwards, “+e sur- therapy,” Cancer, vol. 117, no. 12, pp. 2637–2642, 2011. veillance, epidemiology, and end results (SEER) program of [6] T. Yuasa, S. Urakami, S. Yamamoto et al., “Treatment out- the national cancer institute,” Cancer, vol. 120, pp. 3755–3757, come and prognostic factors in renal cell cancer patients with bone metastasis,” Clinical & Experimental Metastasis, vol. 28, [22] J. Ranstam and J. A. Cook, “Kaplan-Meier curve,” )e British no. 4, pp. 405–411, 2011. Journal of Surgery, vol. 104, no. 4, p. 442, 2017. [7] Y. Toyoda, N. Shinohara, T. Harabayashi et al., “Survival and [23] J. Rueckert, K. Devitt, and J. A. Gardner, “Renal cell carci- prognostic classification of patients with metastatic renal cell noma with monosomy 8: a case series and review of the carcinoma of bone,” European Urology, vol. 52, no. 1, literature,” Journal of the Association of Genetic Technologists, pp. 163–169, 2007. vol. 44, no. 1, pp. 5–9, 2018. [8] F. Ruatta, L. Derosa, B. Escudier et al., “Prognosis of renal cell [24] B. Paule, E. Deslandes, S. E. Le Mouel et al., “Identification of a carcinoma with bone metastases: experience from a large novel biomarker signature associated with risk for bone cancer centre,” European Journal of Cancer, vol. 107, metastasis in patients with renal cell carcinoma,” )e Inter- pp. 79–85, 1990. national Journal of Biological Markers, vol. 25, no. 2, [9] M. H. Younis, L. Fuentes-Rivera, S. Summers, and J. Pretell- pp. 112–115, 2010. Mazzini, “Survival in patients with carcinomas presenting [25] T. Chandrasekar, Z. Klaassen, H. Goldberg, G. S. Kulkarni, with bone metastasis at diagnosis: a SEER population-based R. J. Hamilton, and N. E. Fleshner, “Metastatic renal cell cohort study,” Archives of Orthopaedic and Trauma Surgery, carcinoma: patterns and predictors of metastases-A con- vol. 141, no. 3, pp. 367–373, 2021. temporary population-based series,” Urologic Oncology, [10] H. B. Burke, “Outcome prediction and the future of the TNM vol. 35, no. 11, p. 661, 2017. staging system,” JNCI Journal of the National Cancer Institute, [26] P. V. Hauschka, A. E. Mavrakos, M. D. Iafrati, S. E. Doleman, vol. 96, no. 19, pp. 1408-1409, 2004. and M. Klagsbrun, “Growth factors in bone matrix. Isolation [11] E.-J. Jung, H. J. Lee, C. Kwak, J. H. Ku, and K. C. Moon, of multiple types by affinity chromatography on heparin- “Young age is independent prognostic factor for cancer- Sepharose,” Journal of Biological Chemistry, vol. 261, no. 27, specific survival of low-stage clear cell renal cell carcinoma,” pp. 12665–12674, 1986. Urology, vol. 73, no. 1, pp. 137–141, 2009. [27] S.-C. Chen and P.-L. Kuo, “Bone metastasis from renal cell [12] J. Wu, P. Zhang, G. Zhang et al., “Renal cell carcinoma carcinoma,” International Journal of Molecular Sciences, histological subtype distribution differs by age, gender, and vol. 17, no. 6, p. 987, 2016. tumor size in coastal Chinese patients,” Oncotarget, vol. 8, [28] J. L. Wright, M. C. Risk, J. Hotaling, and D. W. Lin, “Effect of no. 42, pp. 71797–71804, 2017. collecting duct histology on renal cell cancer outcome,” [13] C. T. Lee, J. Katz, P. A. Fearn, and P. Russo, “Mode of Journal of Urology, vol. 182, no. 6, pp. 2595–2600, 2009. presentation of renal cell carcinoma provides prognostic [29] N. Tokuda, S. Naito, O. Matsuzaki, Y. Nagashima, S. Ozono, information,” Urologic Oncology: Seminars and Original In- and T. Igarashi, “Collecting duct (Bellini duct) renal cell vestigations, vol. 7, no. 4, pp. 135–140, 2002. carcinoma: a nationwide survey in Japan,” Journal of Urology, [14] W. Zheng, Y. Huang, H. Chen et al., “Nomogram application vol. 176, no. 1, pp. 40–43, 2006. to predict overall and cancer-specific survival in osteosar- [30] P. I. Karakiewicz, Q.-D. Trinh, N. Rioux-Leclercq et al., coma,” Cancer Management and Research, vol. 10, “Collecting duct renal cell carcinoma: a matched analysis of 41 pp. 5439–5450, 2018. cases,” European Urology, vol. 52, no. 4, pp. 1140–1146, 2007. [15] S. Y. Kim, M. J. Yoon, Y. I. Park, M. J. Kim, B.-H. Nam, and S. R. Park, “Nomograms predicting survival of patients with [31] P. P. Lin, A. N. Mirza, V. O. Lewis et al., “Patient survival after surgery for osseous metastases from renal cell carcinoma,” )e unresectable or metastatic gastric cancer who receive com- bination cytotoxic chemotherapy as first-line treatment,” Journal of Bone and Joint Surgery-American Volume, vol. 89, Gastric Cancer, vol. 21, no. 3, pp. 453–463, 2018. no. 8, pp. 1794–1801, 2007. Journal of Oncology 17 [32] Y. Toyoda, N. Shinohara, T. Harabayashi et al., “Survival and prognostic classification of patients with metastatic renal cell carcinoma of bone,” European Urology, vol. 52, no. 1, pp. 163–169, 2007. [33] R. J. Motzer, B. Escudier, P. Tomczak et al., “Axitinib versus sorafenib as second-line treatment for advanced renal cell carcinoma: overall survival analysis and updated results from a randomised phase 3 trial,” )e Lancet Oncology, vol. 14, no. 6, pp. 552–562, 2013. [34] R. R. McKay, N. Kroeger, W. Xie et al., “Impact of bone and liver metastases on patients with renal cell carcinoma treated with targeted therapy,” European Urology, vol. 65, no. 3, pp. 577–584, 2014. [35] H. W. Kang, S. P. Seo, W. T. Kim et al., “Impact of young age at diagnosis on survival in patients with surgically treated renal cell carcinoma: a multicenter study,” Journal of Korean Medical Science, vol. 31, no. 12, pp. 1976–1982, 2016. [36] R. H. +ompson, M. A. Ordonez, A. Iasonos et al., “Renal cell carcinoma in young and old patients-is there a difference?” Journal of Urology, vol. 180, no. 4, pp. 1262–1266, 2008. [37] V. Grunwald, ¨ B. Eberhardt, A. Bex et al., “An interdisciplinary consensus on the management of bone metastases from renal cell carcinoma,” Nature Reviews Urology, vol. 15, no. 8, pp. 511–521, 2018. [38] A. Fottner, M. Szalantzy, L. Wirthmann et al., “Bone me- tastases from renal cell carcinoma: patient survival after surgical treatment,” BMC Musculoskeletal Disorders, vol. 11, p. 145, 2010. [39] B. Fuchs, R. T. Trousdale, and M. G. Rock, “Solitary bony metastasis from renal cell carcinoma,” Clinical Orthopaedics & Related Research, vol. 431, no. 431, pp. 187–192, 2005. [40] A. L. Alt, S. A. Boorjian, C. M. Lohse, B. A. Costello, B. C. Leibovich, and M. L. Blute, “Survival after complete surgical resection of multiple metastases from renal cell carcinoma,” Cancer, vol. 117, no. 13, pp. 2873–2882, 2011. [41] J. Lee, D. Hodgson, E. Chow et al., “A phase II trial of palliative radiotherapy for metastatic renal cell carcinoma,” Cancer, vol. 104, no. 9, pp. 1894–1900, 2005. [42] D. Wilson, L. Hiller, L. Gray, M. Grainger, A. Stirling, and N. James, “+e effect of biological effective dose on time to symptom progression in metastatic renal cell carcinoma,” Clinical Oncology, vol. 15, no. 7, pp. 400–407, 2003. [43] J. Zołnierek, P. Nurzynski, ´ P. Langiewicz et al., “Efficacy of targeted therapy in patients with renal cell carcinoma with pre-existing or new bone metastases,” Journal of Cancer Research and Clinical Oncology, vol. 136, no. 3, pp. 371–378, http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Oncology Hindawi Publishing Corporation

Bone Metastasis in Renal Cell Carcinoma Patients: Risk and Prognostic Factors and Nomograms

Download PDF
17 pages

Loading next page...
 
/lp/hindawi-publishing-corporation/bone-metastasis-in-renal-cell-carcinoma-patients-risk-and-prognostic-0ZnselpQRi

References (43)

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2021 Zhiyi Fan et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
1687-8450
eISSN
1687-8469
DOI
10.1155/2021/5575295
Publisher site
See Article on Publisher Site

Abstract

Hindawi Journal of Oncology Volume 2021, Article ID 5575295, 17 pages https://doi.org/10.1155/2021/5575295 Research Article Bone Metastasis in Renal Cell Carcinoma Patients: Risk and Prognostic Factors and Nomograms 1,2 2 1 Zhiyi Fan, Zhangheng Huang, and Xiaohui Huang Hangzhou Medical College, Hangzhou, Zhejiang Province, China Department of Spine Surgery, Affiliated Hospital of Chengde Medical University, Chengde, Hebei Province, China Correspondence should be addressed to Xiaohui Huang; 373644723@qq.com Received 10 February 2021; Revised 7 April 2021; Accepted 27 April 2021; Published 12 May 2021 Academic Editor: Liren Qian Copyright © 2021 Zhiyi Fan et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Bone metastasis (BM) is one of the common sites of renal cell carcinoma (RCC), and patients with BM have a poorer prognosis. We aimed to develop two nomograms to quantify the risk of BM and predict the prognosis of RCC patients with BM. Methods. We reviewed patients with diagnosed RCC with BM in the Surveillance, Epidemiology, and End Results (SEER) database from 2010 to 2015. Multivariate logistic regression analysis was used to determine independent factors to predict BM in RCC patients. Univariate and multivariate Cox proportional hazards regression analyses were used to determine independent prognostic factors for BM in RCC patients. Two nomograms were established and evaluated by calibration curve, receiver operating characteristic (ROC) curve, and decision curve analysis (DCA). Results. +e study included 37,554 patients diagnosed with RCC in the SEER database, 537 of whom were BM patients. BM’s risk factors included sex, tumor size, liver metastasis, lung metastasis, brain metastasis, N stage, T stage, histologic type, and grade in RCC patients. Currently, independent prognostic factors for RCC with BM included grade, histologic type, N stage, surgery, brain metastasis, and lung metastasis. +e calibration curve, ROC curve, and DCA showed good performance for diagnostic and prognostic nomograms. Conclusions. Nomograms were established to predict the risk of BM in RCC and the prognosis of RCC with BM, separately. +ese nomograms strengthen each patient’s prognosis-based decision making, which is critical in improving the prognosis of patients. treatment plans [10]. Studies have shown that race, sex, age, 1. Introduction and tumor size may also affect the prognosis of patients with Renal cell carcinoma (RCC) is one of the most common RCC [11–13]. +e TNM staging system relies on three cancers worldwide, with approximately 403,262 new cases pathological indicators and ignores other prognostic factors, and 17,598 deaths in 2018 [1]. Approximately 15–30% of thereby reducing the accuracy of prognostic prediction for RCC patients have metastases at the initial diagnosis, and RCC patients. +erefore, it is necessary to combine clin- bone is a common site of metastasis [2, 3]. Bone metastasis icopathology and other prognosis-related variables to con- (BM) from RCC is predominantly osteolytic and can lead to struct a tool to accurately predict the prognosis and overcome the limitations of the traditional TNM staging skeletal-related diseases, which can reduce the quality of life and prognosis of the patients [4, 5]. +e median overall system. survival (OS) of RCC patients with BM has been reported to Nomogram is a tool that combines multiple biological be only 12–28 months [6, 7]. In contrast, patients with and clinical variables to predict specific endpoints and has metastatic RCC without BM had a more prolonged median been widely used to predict the prognosis of cancer patients OS to 31 months [8, 9]. +erefore, understanding the BM in [14–16]. By combining these important variables, the no- RCC patients is an unmet need. mograms can individually estimate the probability of events +e TNM staging system is widely used to assess the over time, such as the OS of cancer patients. In addition, prognosis of cancer patients, and clinicians use it to develop nomograms can be used to estimate the survival rate of 2 Journal of Oncology tumor size in terms of OS were determined by X-tile soft- cancer patients with higher accuracy than the TNM staging system [17]. ware, and patients were divided into three groups (<4, 4–7, and>7 cm). +e histologic type was defined by the following Risk factors and prognosis-related factors for BM in RCC have been reported in several previous studies [18–20]. ICD-O-3 codes: clear cell (8310/3, 8313/3), papillary (8260/ However, no studies have focused on constructing predictive 3), chromophobe (8317/3, 8270/3), and collecting duct models for the risk and prognosis of BM in RCC, which (8319/3). Regarding marital status, we excluded misleading means that the probability of outcome cannot be quantified. data on unmarried or domestic partners and then included +erefore, based on the data from the Surveillance, Epi- “unmarried,” “separated,” “single,” and “widowed” all in the demiology, and End Results (SEER) database, we developed unmarried group. Insurance status is divided into insured two nomograms for predicting the risk of BM with RCC and and uninsured, with both “insured” and “insured/unspe- the OS of RCC patients with BM, separately. cific” included in the insured group. All cases in this study were staged using version 7 of the American Joint Com- mittee on Cancer TNM staging system. In the survival 2. Methods analysis, the primary endpoint of our study was OS, which was defined as the date from diagnosis to death (for any 2.1. Study Population Selection. +e SEER database covers reason) or the date of the last follow-up. approximately 28% of cancer registries in the United States [21]. +e data contained in this study were downloaded from the SEER Stat software version 8.3.6. Analysis of anony- mous data from the SEER database is exempt from medical 2.3. Statistical Analysis. +is study used SPSS 25.0 and R ethics review and does not require informed consent. +e software (version 3.6.1) for statistical analysis. +e chi- SEER database provides clinical information on cancer square test was used for categorical data. Variables with P patients that greatly facilitate clinical research. Patients di- values <0.05 in univariate analysis were incorporated into a agnosed before 2010 were excluded because the SEER da- multivariate logistic regression analysis to identify inde- tabase did not record information on distant metastases pendent risk factors for BM in RCC patients. At the same until 2010. In addition, to ensure adequate follow-up time, time, univariate Cox proportional hazards regression patients diagnosed after 2015 are not included. +erefore, analysis was used to determine OS-related variables. Sig- only patients diagnosed with RCC between 2010 and 2015 nificant variables in the univariate Cox proportional hazards were considered in this study. regression analysis were then included in the multivariate Inclusion criteria were as follows: (1) RCC as the first Cox proportional hazards regression analysis to identify primary tumor, (2) patients with a histologic diagnosis of independent prognostic factors in RCC patients with BM. RCC, and (3) patients with complete clinicopathological Nomograms were developed separately based on in- features, demographic information, and follow-up infor- dependent BM-related predictors and prognostic factors mation. In addition, patients who were certified by autopsy using the “rms” package in R software. In the nomograms, or death were excluded from this study. Finally, a total of values for the individual patient were located along the 37,554 patients with RCC were enrolled to study the risk variable axes, and a line was drawn upward to the points factors for BM in patients with RCC and to establish a axis to determine the number of points assigned for each diagnostic nomogram. Subsequently, for RCC patients with variable. +ere was a total points line at the bottom of the BM with survival time≥ one month, specific treatment in- nomogram, and each variable score was summed to give formation, including surgery, radiotherapy, and chemo- the total points. Receiver operating characteristic (ROC) therapy, were used to form a new cohort to explore the curves for two nomograms were generated, and the cor- prognostic factors for RCC patients with BM and develop a responding area under the curve (AUC) was used to prognostic nomogram. Ultimately, 537 patients were used to evaluate the discrimination of nomograms. +e clinical study prognostic factors in patients with BM from RCC. application value of the nomogram model was evaluated by Patients in each cohort were randomized into training and calibration curve and decision curve analysis (DCA). Fi- validation cohorts in a 7 : 3 ratio. In this study, patients in the nally, all patients were divided into high-risk and low-risk training cohort were used to construct the predicted no- groups according to the median of risk score, and survival mogram, while patients in the validation cohort were used to curves were used to verify the prognostic value of the validate the constructed nomogram. nomogram [22]. 2.2. Data Collection. Based on patient-specific information 3. Results from the SEER database, we selected 14 variables to identify risk factors for BM in RCC, including age, sex, race, tumor 3.1. )e Characteristics of the Study Population. +e work- size, histologic type, grade, laterality, Tstage, N stage, distant flow of our study is illustrated in Figure 1. A total of 37,554 metastatic site (lung, brain, liver), insurance status, and RCC patients from the SEER database were included. marital status. In addition to the aforementioned variables, Furthermore, 26,290 and 11,264 patients were included in information on surgery, radiotherapy, and chemotherapy the training and validation cohorts, respectively. Clinico- are included to study the factors that influence the prognosis pathological information of 26,290 RCC patients is given in of RCC patients with BM. +e optimal cutoff values for Table 1. Journal of Oncology 3 RCC patients (N = 89795) Diagnostic cohort RCC patients with BM (N = 37554) (N = 1852) Survival study cohort (N = 537) Validation cohort Training cohort Training cohort Validation cohort (N = 11264) (N = 26290) (N = 377) (N = 160) Univariate analysis Univariate cox Multivariate analysis Multivariate cox Predictor of BM Prognostic variables Diagnostic nomogram Prognostic nomogram ROC curve Calibration curve DCA Survival curve Figure 1: +e workflow describing the schematic overview of the project. 3.2. Risk Factors of BM in RCC Patients. To identify BM- As shown in Figure 5, the AUC of the nomogram is higher related variables in RCC patients, 14 factors were analyzed. than the AUCs of all independent variables in both training +e results showed that ten factors were related to the BM in and validation cohorts, indicating a significant advantage in RCC patients, including race, sex, grade, histologic type, T the accuracy of predictions using the nomogram compared to predictions using individual independent predictors. stage, N stage, brain metastasis, liver metastasis, lung me- tastasis, and tumor size (Table 1). Subsequently, the above variables were included in the multivariate logistic regres- 3.4. Prognostic Factors for RCC Patients with BM. sion analysis, which showed that tumor size, liver metastasis, lung metastasis, brain metastasis, N stage, T stage, histologic According to the selection process, a total of 537 patients with BM were included in our research. Meanwhile, 377 patients type, and grade were independent predictors of RCC with BM (Table 2). were incorporated into the training cohort, and the remaining 160 patients were incorporated into the validation cohort. Univariate and multivariate Cox proportional hazards regres- 3.3. Development and Validation of a Nomogram for BM in sion analyses were performed to screen for prognostic factors. Univariate Cox proportional hazards regression analysis Newly Diagnosed RCC Patients. Based on eight independent BM-related variables, a nomogram was constructed to assess showed that grade, T stage, histologic type, N stage, surgery, the risk of BM in RCC patients (Figure 2). +e AUCs of the chemotherapy, brain metastasis, liver metastasis, and lung nomogram were 0.865 and 0.859 in the training and validation metastasis are OS-related factors (Table 3). After controlling for cohorts, respectively, showing good discrimination confounding variables using multivariate Cox proportional (Figure 3(a) and Figure 4(a)). +e calibration curve showed hazards regression analysis, grade, histologic type, N stage, that the observations are highly consistent with the predicted surgery, brain metastasis, and lung metastasis were identified as results (Figure 3(b) and Figure 4(b)). Moreover, DCA indi- independent prognostic factors in RCC patients with BM cated that the diagnostic nomogram performs well in clinical (Table 3). As shown in Figure 6, the survival curve analysis practice (Figure 3(c) and Figure 4(c)). Importantly, ROC further demonstrated the impact of screened independent curves were generated for each independent predictor variable. prognostic factors on the OS of RCC patients with BM. Validation Validation 4 Journal of Oncology Table 1: Demographic and clinical characteristics of RCC patients. Table 1: Continued. 2 2 Without BM With BM χ P Without BM With BM χ P Age 1.463 0.226 25089 Yes 437 (96.9%) 14388 (97.1%) <65 264 (58.5%) (55.7%) Marital status 3.481 0.062 No 8109 (31.4%) 123 (27.3%) ≥65 187 (41.5%) (44.3%) Yes 328 (72.7%) Sex 6.982 0.008 (68.6%) Female 9111 (35.3%) 132 (29.3%) BM, bone metastasis; RCC, renal cell carcinoma; CP, chromophobe; CC, clear cell; CD, collecting duct; PL, papillary. Male 319 (70.7%) (64.7%) Race 6.289 0.043 Table 2: Multivariate logistic regression analysis of BM in RCC Black 2648 (10.2%) 30 (6.7%) patients. Others 1585 (6.1%) 28 (6.2%) Variables OR (95% CI) P value White 393 (87.1%) (83.6%) Grade Grade 236.193 ≤0.001 G1-2 Reference I-II 16818 (65.1%) 136 (30.2%) G3-4 1.749 (1.388–2.204) ≤0.001 III-IV 9021 (34.9%) 315 (69.8%) T stage T Stage 437.434 ≤0.001 T1-2 Reference T3-4 1.748 (1.379–2.216) ≤0.001 T1-2 182 (40.4%) (80.3%) Histologic type T3-4 5079 (19.7%) 269 (59.6%) CP Reference Laterality 1.692 0.193 CC 3.300 (1.459–7.466) 0.004 CD 4.216 (1.245–14.277) 0.021 Left 236 (52.3%) (49.2%) PL 1.850 (0.762–4.490) 0.174 Tumor size, cm Right 215 (47.7%) (50.8%) <4 Reference Histologic type 79.734 ≤0.001 4–7 3.937 (2.745–5.648) ≤0.001 CP 1416 (5.5%) 6 (1.3%) >7 3.510 (2.374–5.189) ≤0.001 N stage CC 407 (90.2%) (78.3%) N0 Reference CD 54 (0.2%) 7 (1.6%) N1 2.654 (2.005–3.513) ≤0.001 PL 4125 (16.0%) 31 (6.9%) Brain metastasis Tumor size, cm 392.797 ≤0.001 No Reference Yes 4.283 (2.780–6.598) ≤0.001 <4 39 (8.6%) (46.8%) Liver metastasis 4–7 8559 (33.1%) 167 (37.0%) No Reference >7 5193 (20.1%) Yes 3.309 (2.211–4.952) ≤0.001 (54.3.0%) Lung metastasis N stage 733.940 ≤0.001 No Reference N0 345 (76.5%) Yes 5.351 (4.123–6.946) ≤0.001 (97.6%) BM, bone metastasis; RCC, renal cell carcinoma; CP, chromophobe; CC, N1 621 (2.4%) 106 (23.5%) clear cell; CD, collecting duct; PL, papillary. Brain 679.401 ≤0.001 metastasis 3.5. Prognostic Nomogram for RCC Patients with BM. A No 408 (90.5%) (99.6%) prognostic nomogram of RCC patients with BM based on six Yes 98 (0.4%) 43 (9.5%) independent prognostic factors was established (Figure 7). Liver metastasis 629.229 ≤0.001 +e ROC curve showed that the AUCs at 1, 2, and 3 years were 0.711, 0.772, and 0.766 in the training cohort and 0.684, No 404 (89.6%) (99.5%) 0.663, and 0.691 in the validation cohort (Figure 8(a) and Yes 132 (0.5%) 47 (10.4%) 8(c)). +e optimal cutoff point for the total score was de- Lung metastasis 1835.076 ≤0.001 termined by X-tile software and was 285. +erefore, we specified less than 285 as the low-risk group and greater than No 274 (60.8%) (97.3%) 285 as the high-risk group. By depicting the Kaplan–Meier Yes 700 (2.7%) 177 (39.2%) survival curve, we can find that patients in the high-risk Insurance 0.064 0.800 group showed a worse prognosis than patients in the low- status risk group (Figure 8(b) and 8(d)). In addition, we further No 750 (2.9%) 14 (3.1%) compared the discrimination between the nomogram and Journal of Oncology 5 Glm regression Points 0 10 20 30 40 50 60 70 80 90 100 ∗∗∗ >7 Tumor size <4 4–7 ∗∗∗ Lung metastasis Yes No ∗∗∗ Liver metastasis Yes No ∗∗∗ Brain metastasis Yes No ∗∗∗ N stage N1–2 No ∗∗∗ T stage T3–4 T1-2 ∗∗ Papillary adenocardnoma Collecting duct cardnoma Histologic type Chromophobe cardnoma Clear cell adenocardnoma T1–2 ∗∗∗ III-IV Grade I-II Total-points-to-outcome nomogram: Total points 0 50 100 150 200 250 300 350 400 450 500 0.00988 Odds (bone) 5e – 04 0.001 0.0015 0.0025 0.004 0.006 0.01 0.015 0.03 0.06 0.1 0.15 0.3 0.6 1 1.5 2.5 Figure 2: Nomogram to estimate the risk of BM in patients with RCC. 0.7 AUC = 0.865 0.02 1.0 0.6 0.01 0.5 0.8 0.4 –0.01 0.6 0.3 0.4 0.2 –0.03 0.1 0.2 0.0 –0.05 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Nomogram-predicted reshold probability False positive rate probability of BM Nomogram Apparent All Bias-corrected None Ideal (a) (b) (c) Figure 3: ROC curves (a), calibration curves (b), and DCA (c) of the training cohort. AUC = 0.859 0.02 0.6 1.0 0.01 0.5 0.8 0.4 –0.01 0.6 0.3 0.2 0.4 –0.03 0.1 0.2 0.0 –0.05 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Nomogram-predicted reshold probability False positive rate probability of BM Nomogram Apparent All Bias-corrected None Ideal (a) (b) (c) Figure 4: ROC curves (a), calibration curves (b), and DCA (c) of the validation cohort. True positive rate True positive rate Actual BM Actual BM Net benefit Net benefit 6 Journal of Oncology Statistical comparison Statistical comparison 100 100 80 80 60 60 40 40 20 20 0 0 100 80 60 40 20 0 100 80 60 40 20 0 Specificity (%) Specificity (%) Nomogram Brain metastasis Nomogram Brain metastasis Grade Liver metastasis Grade Liver metastasis Histologic type Lung metastasis Histologic type Lung metastasis T stage Tumor size T stage Tumor size N stage N stage (a) (b) Figure 5: Comparison of AUC between diagnostic nomogram and each independent predictor in the training cohort (a) and the validation cohort (b). Table 3: Univariate and multivariate Cox analyses in RCC patients with BM. Univariate Cox analysis Multivariate Cox analysis HR 95% CI P HR 95% CI P Age <65 ≥65 1.026 0.798 1.319 0.844 Race Black Others 1.660 0.913 3.020 0.097 White 1.293 0.827 2.023 0.260 Sex Female Male 1.109 0.847 1.452 0.451 Grade I-II III-IV 1.411 1.074 1.853 0.013 1.669 1.235 2.257 ≤0.001 T stage T1-2 T3-4 1.388 1.080 1.785 0.010 Laterality Left Right 1.099 0.864 1.397 0.443 Histologic type CP CC 0.848 0.537 1.339 0.479 0.706 0.440 1.132 0.148 CD 1.157 0.487 2.746 0.742 0.723 0.300 1.743 0.471 PL 2.767 1.167 6.557 0.021 2.492 1.036 5.994 0.041 Sensitivity (%) Sensitivity (%) Journal of Oncology 7 Table 3: Continued. Univariate Cox analysis Multivariate Cox analysis HR 95% CI P HR 95% CI P Tumor size, cm <4 4–7 1.125 0.683 1.851 0.644 >7 1.323 0.821 2.132 0.250 N stage N0 N1 1.791 1.378 2.328 ≤0.001 1.388 1.049 1.838 0.022 Surgery No Yes 0.416 0.313 0.552 ≤0.001 0.394 0.284 0.546 ≤0.001 Radiotherapy No Yes 1.103 0.862 1.411 0.438 Chemotherapy No Yes 1.463 1.128 1.898 0.004 Brain metastasis No Yes 2.315 1.575 3.403 ≤0.001 1.801 1.201 2.700 0.004 Liver metastasis No Yes 1.960 1.349 2.847 ≤0.001 Lung metastasis No Yes 2.261 1.771 2.887 ≤0.001 1.745 1.342 2.269 ≤0.001 Insurance status No Yes 1.227 0.628 2.396 0.549 Marital status No Yes 0.994 0.768 1.286 0.961 BM, bone metastasis; RCC, renal cell carcinoma; CP, chromophobe; CC, clear cell; CD, collecting duct; PL, papillary. the independent prognostic factors, and the results showed analyzing massive data, respectively. We believe that two that the AUC of the nomogram was higher than the AUCs of nomograms representing OS and distant metastasis are complementary and can increase their clinical value in all independent factors at 1, 2, and 3 years, both in the training cohort and in the validation cohort (Figure 9). patients with RCC. +e total score can be calculated by Calibration curves of predicting 1, 2, and 3-year OS prob- obtaining data for each RCC patient’s corresponding abilities also show good agreement between the OS predicted variable on the nomogram. +e risk of BM can then be by the prognostic nomogram and the actual results easily identified on the diagnostic nomogram, identifying (Figure 10(a) and 10(b)). +e DCA was used to evaluate the patients in the high-risk group and guiding clinical practice clinical utility of a nomogram. As shown in Figure 10, the in early intervention. Similarly, the prognosis of RCC prognostic nomogram shows a significant positive net patients with BM can be determined from the prognostic benefit over a wide range of mortality risks, suggesting its nomogram. In the validation of the two nomograms, the high clinical utility in predicting OS in RCC patients with two nomograms showed excellent performance in BM risk BM. assessment and OS prediction in RCC patients, respec- tively, which will enable more accurate personalized clinical decision making and monitoring. 4. Discussion Despite the poor prognosis of RCC patients with BM, RCC accounts for 3% of all malignancies and 80%–85% of early detection of BM may be critical for patients with RCC primary renal cancer [23]. Bone is the second most to receive appropriate treatment. +erefore, exploring the common site of metastasis in RCC patients, following the risk factors for BM in RCC patients is important for clinical lung [24, 25]. In the present study, we constructed diag- decision making. At the molecular level, cadherin-11, nostic and prognostic nomograms to predict the risk of BM transforming growth factor-β, insulin-like growth factor, in RCC patients and the OS of RCC patients with BM by and the fibroblast growth factor have been associated with 8 Journal of Oncology 1.00 0.75 0.50 0.25 p < 0.0001 0.00 020 40 60 80 Time Brain Brain = no Brain = yes (a) 1.00 0.75 0.50 0.25 p = 0.012 0.00 020 40 60 80 Time Grade Grade = I-II Grade = III-IV (b) Figure 6: Continued. Survival probability Survival probability Journal of Oncology 9 1.00 0.75 0.50 0.25 p = 0.0084 0.00 020 40 60 80 Time Hist CP CO CC PL (c) 1.00 0.75 0.50 0.25 p < 0.0001 0.00 020 40 60 80 Time Lung Lung = no Lung = yes (d) Figure 6: Continued. Survival probability Survival probability 10 Journal of Oncology 1.00 0.75 0.50 0.25 p < 0.0001 0.00 020 40 60 80 Time N = N0 N = N1 (e) 1.00 0.75 0.50 0.25 p < 0.0001 0.00 020 40 60 80 Time Surgery Surgery = no Surgery = yes (f) Figure 6: Survival curves for each independent prognostic factor in the training cohort. CP, chromophobe; CC, clear cell; CD, collecting duct; PL, papillary. BM in RCC patients [26, 27]. Nevertheless, these biomarkers for patients to afford. Of course, if molecular level indicators are difficult and impractical to apply immediately to clinical could be included in the nomogram, this would undoubtedly decision making. In our daily clinical work, it is difficult for increase the predictive accuracy of the nomogram, which us to examine every patient at the molecular level because it could lead to better survival for patients. In addition, re- requires a lot of human and material resources. At the same garding some practical clinical features, sex, T stage, N stage, time, the high cost of testing at the molecular level is difficult grade, liver metastasis, lung metastasis, brain metastasis, and Survival probability Survival probability Journal of Oncology 11 Coxph regression Points 010 20 30 40 50 60 70 80 90 100 N stage N1-2 No ∗∗ Brain metastasis Yes No Chromophobe type Histologic type Collecting duct cardnoma Clear cell adenocardnoma Papillary adenocardnoma ∗∗∗ Grade III-IV I-II Lung metastasis Yes No ∗∗∗ Surgery No Yes Total-points-to-outcome nomogram: Total points 220 240 260 280 300 320 340 360 380 400 420 0.247 Pr (futime < 12) 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.85 0.9 0.94 0.97 0.985 0.995 0.403 Pr (futime < 24) 0.2 0.3 0.4 0.5 0.6 0.7 0.85 0.9 0.94 0.97 0.985 0.994 0.998 0.543 Pr (futime < 36) 0.3 0.4 0.5 0.6 0.7 0.85 0.9 0.94 0.97 0.985 0.994 0.998 Figure 7: Nomogram to predict the OS of RCC patients with BM. 1.00 1.00 0.75 0.50 0.25 0.75 0.00 020 40 60 80 Time in months 0.50 Strata Risk = high Risk = low 0.25 Number at risk Risk = high 180 45 11 3 0 Risk = low 197 114 54 16 1 0.00 0 20 40 60 80 Time in months 0.00 0.25 0.50 0.75 1.00 FP Number at censoring AUC 4 12 months AUC: 0.711 24 months AUC: 0.772 0 20 40 60 80 36 months AUC: 0.766 Time in months (a) (b) Figure 8: Continued. TP Strata Survival probability n. censor 12 Journal of Oncology 1.00 1.00 0.75 0.50 0.25 0.75 0.00 020 40 60 80 Time in months 0.50 Strata Risk = high Risk = low 0.25 Number at risk Risk = high 68 21 8 1 0 Risk = low 92 44 21 5 1 0.00 020 40 60 80 Time in months 0.00 0.25 0.50 0.75 1.00 FP Number at censoring AUC 2 12 months AUC: 0.684 24 months AUC: 0.663 0 20 40 60 80 36 months AUC: 0.691 Time in months (c) (d) Figure 8: (a) Receiver operating characteristic curves of 1, 2, and 3 years in the training cohort. (b) +e Kaplan–Meier survival curve of the training cohort. (c) Receiver operating characteristic curves of 1, 2, and 3 years in the validation cohort. (d) +e Kaplan–Meier survival curve of the validation cohort. histologic type have been reported as relevant risk factors for effective tool for identifying high-risk patients. +e impact of BM in RCC [20]. However, to date, no predictive model has histologic type on metastatic potential and prognosis of been developed, which means that it is impossible to identify metastatic patients is often overlooked when discussing an individual’s risk of BM by combining all independent treatment options. In this study, collecting duct RCC had a predictors associated with BM. +e present study showed higher incidence of BM and a worse prognosis compared to that tumor size, liver metastasis, lung metastasis, brain other renal cancer subtypes. Collecting duct RCC is reported metastasis, N stage, T stage, histologic type, and the grade to be a rare entity that occurs in <2% of patients with kidney cancer, often resulting in a poor prognosis [28]. In addition, were significant predictors of BM in RCC. +e association between these factors and BM in RCC patients has been the above correlation has been confirmed in previous studies reported in previous studies. Although metastasis to mul- [29, 30]. +e relationship between lung metastasis, brain tiple organs is a risk factor for BM in patients with RCC, metastasis, surgery, and prognosis in patients with RCC has unfortunately, we were unable to obtain the sequence of also been widely reported in previous studies. Lin et al. organ metastasis due to the shortcomings of the SEER reported a better prognosis in patients with only BM than in database itself. Previous studies have confirmed the rela- patients with concomitant pulmonary metastases and a tionship between tumor grade, TNM staging, and BM in significantly better prognosis for patients with single BM RCC patients [20]. TNM staging is widely used in the as- than in patients with multiple bones and/or visceral me- tastasis [31]. Similarly, Toyoda et al. reported a shorter sessment of prognosis in cancer patients. Notably, a more significant contribution of TNM staging was shown in both median survival in patients with extra-BM compared to those without (8 vs. 33 months, P � 0.0084) [32]. Surpris- the diagnostic nomogram and the prognostic nomogram. With increasing tumor size, an increasing number of lymph ingly, contrary to previous reports, the presence of liver node metastases, and distant organ metastases, the risk of metastasis was not an independent prognostic factor in our BM in RCC and the risk of death in RCC patients with BM study [33, 34]. However, this is consistent with what has are significantly increased. been reported by Santoni et al. [19]. Previous studies have In addition, our study found a poor prognosis of patients reported age as a factor associated with patient prognosis with lymph node metastasis, brain metastasis, lung metas- regarding RCC, but other studies show no difference in tasis, without surgery, poor tumor differentiation, and prognosis between younger and older patients with RCC histologic type of the collecting duct. A prognostic nomo- [35, 36]. Some of these studies included only a restricted age group of patients or limited sample size or follow-up time. gram was established based on six independent prognostic factors. +e results suggested that a nomogram can be an +us, until now, the role of age as a prognostic factor in TP Strata Survival probability n. censor Journal of Oncology 13 1.0 1.0 1.0 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 False positive rate False positive rate False positive rate Nomogram (AUC = 0.711) Nomogram (AUC = 0.772) Nomogram (AUC = 0.755) Grade (AUC = 0.541) Grade (AUC = 0.538) Grade (AUC = 0.549) Hist (AUC = 0.526) Hist (AUC = 0.532) Hist (AUC = 0.528) N (AUC = 0.592) N (AUC = 0.608) N (AUC = 0.604) Surgery (AUC = 0.596) Surgery (AUC = 0.611) Surgery (AUC = 0.608) Brain (AUC = 0.539) Brain (AUC = 0.560) Brain (AUC = 0.551) Lung (AUC = 0.637) Lung (AUC = 0.675) Lung (AUC = 0.670) (a) (b) (c) 1.0 1.0 1.0 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 False positive rate False positive rate False positive rate Nomogram (AUC = 0.684) Nomogram (AUC = 0.663) Nomogram (AUC = 0.691) Grade (AUC = 0.540) Grade (AUC = 0.508) Grade (AUC = 0.529) Hist (AUC = 0.509) Hist (AUC = 0.500) Hist (AUC = 0.549) N (AUC = 0.526) N (AUC = 0.517) N (AUC = 0.564) Surgery (AUC = 0.555) Surgery (AUC = 0.577) Surgery (AUC = 0.562) Brain (AUC = 0.596) Brain (AUC = 0.543) Brain (AUC = 0.543) Lung (AUC = 0.628) Lung (AUC = 0.634) Lung (AUC = 0.641) (d) (e) (f) Figure 9: +e receiver operating characteristic curves of nomogram and all independent predictors at 1 (a), 2 (b), and 3 years (c) in the training cohort and at 1 (d), 2 (e), and 3 years (f) in the validation cohort. patients with RCC has been controversial. +e study in- but that patients who had surgery showed a better prognosis. cluded as many factors as possible that may be associated As reported in several studies, surgical removal of isolated or with the prognosis of patients with RCC and identified the minimally metastatic lesions can improve the prognosis of relevant prognostic factors by rigorous statistical methods, patients with BM, thus providing a multidisciplinary team to so the results are trustworthy. However, due to the retro- support the treatment plan for these patients [38–40]. Al- though renal cancer is usually not sensitive to radiotherapy spective nature of the study, selection bias is inevitable. For the treatment of RCC patients with BM, recent consensus and chemotherapy, palliative radiotherapy can significantly suggests using a multimodal treatment strategy that includes relieve local symptoms and improve quality of life [41, 42]. extensive resection of the lesion, radiotherapy, systemic Tyrosine kinase inhibitors (TKIs) and antivascular endo- therapy, and other local treatment options [37]. Of RCC thelial growth factor antibodies are now widely used as first- patients with BM, surgical treatment aims to improve the and second-line therapy for advanced RCC. Direct evidence prognosis, local tumor control, pain relief, and preservation on the effects of targeted drugs on BM is currently limited to or reconstruction of function. Based on the results, we found a few studies that have shown that TKIs can prolong the that surgery was not only an independent prognostic factor mean time to progression of existing bone lesions and True positive rate True positive rate True positive rate True positive rate True positive rate True positive rate 14 Journal of Oncology 1-year 2-years 3-years 1.0 1.0 1.0 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Nomogram-predicted OS Nomogram-predicted OS Nomogram-predicted OS (a) 1-year 2-years 3-years 1.0 1.0 1.0 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Nomogram-predicted OS Nomogram-predicted OS Nomogram-predicted OS (b) 1-year 2-years 3-years 0.4 0.5 0.6 0.3 0.4 0.4 0.3 0.2 0.2 0.1 0.2 0.1 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 reshold probability reshold probability reshold probability None None None All All All Nomogram Nomogram Nomogram (c) Figure 10: Continued. Net benefit Actual OS Net benefit Actual OS Net benefit Actual OS Journal of Oncology 15 1-year 2-years 3-years 0.6 0.5 0.3 0.5 0.4 0.4 0.2 0.3 0.3 0.2 0.1 0.2 0.1 0.1 0.0 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 reshold probability reshold probability reshold probability None None None All All All Nomogram Nomogram Nomogram (d) Figure 10: (a) +e calibration curves of the prognostic nomogram in the training cohort. (b) +e calibration curves of the prognostic nomogram in the validation cohort. (c) +e decision curve analysis of the prognostic nomogram in the training cohort. (d) +e decision curve analysis of the prognostic nomogram in the validation cohort. reduce the formation of new bone lesions [2, 43]. Unfor- Abbreviations tunately, the SEER database does not contain specific ana- AUC: Area under the curve lyses of targeted therapies, chemotherapy, and radiotherapy, BM: Bone metastasis and we are unable to analyze their influence on prognosis in DCA: Decision curve analysis further detail. In addition, further research on important OS: Overall survival prognostic factors for OS with BM in RCC is necessary. ROC: Receiver operating characteristic However, some limitations of our study should be noted. RCC: Renal cell carcinoma First, information collected in the SEER database is about the SEER: Surveillance, Epidemiology, and End Results disease at the first diagnosis and does not record BM that TKIs: Tyrosine kinase inhibitors. occurred later. Second, the prognostic impact of the amount of BM should not be overlooked, but there is no record of this in the SEER database. +ird, we did not have access to some Data Availability biomarkers from the SEER database, such as transforming +e dataset from the SEER database that was generated and/ growth factor-β, insulin-like growth factor, and fibroblast growth factor. Fourth, this was a retrospective study in which or analyzed during the current study is available in the SEER selection bias was inevitable, and detailed treatment was not dataset repository (https://seer.cancer.gov/). available in the SEER database. Immunotherapy was rec- ommended for patients with RCC because of its OS benefit, Ethical Approval but, unfortunately, the SEER database does not contain this information. As such, the validity of this data is not any more Approval was waived by the local ethics committee, as SEER known. In addition, since the construction and validation data are publicly available and deidentified. cohorts are from the same database, it is still necessary to validate the accuracy of the nomograms in other databases. Conflicts of Interest +e authors declare that they have no conflicts of interest. 5. Conclusions Two nomograms we created could be used as a supportive Authors’ Contributions graphic tool in RCC patients to help clinicians distinguish, assess, and evaluate the risk and prognosis of RCC with BM. ZY F, ZH H, and XH H conceived and designed the study. At the same time, when faced with individualized condition ZH H performed the literature search. ZY F generated the consultation, these nomograms are valuable methods to figures and tables. ZY F and ZH H analyzed the data. ZY F provide prognostic information to clinical patients and and ZH H wrote the article, and XH H critically reviewed the strengthen each patient’s prognosis-based decision making, article. ZY F and XH H supervised the research. All authors which is of great significance in improving the prognosis of have read and approved the final draft of the article. Zhiyi patients. Fan and Zhangheng Huang contributed equally to this work. Net benefit Net benefit Net benefit 16 Journal of Oncology [16] M. W. Kattan, V. Reuter, R. J. Motzer, J. Katz, and P. Russo, “A References postoperative prognostic nomogram for renal cell carci- [1] F. Bray, J. Ferlay, I. Soerjomataram, R. L. Siegel, L. A. Torre, noma,” )e Journal of Urology, vol. 166, no. 1, pp. 63–67, 2001. and A. Jemal, “Global cancer statistics 2018: GLOBOCAN [17] N. Pu, J. Li, Y. Xu et al., “Comparison of prognostic prediction estimates of incidence and mortality worldwide for 36 cancers between nomogram based on lymph node ratio and AJCC 8th in 185 countries,” CA: A Cancer Journal for Clinicians, vol. 68, staging system for patients with resected pancreatic head no. 6, pp. 394–424, 2018. carcinoma: a SEER analysis,” Cancer Management and Re- [2] S. L. Wood and J. E. Brown, “Skeletal metastasis in renal cell search, vol. 10, pp. 227–238, 2018. carcinoma: current and future management options,” Cancer [18] H. Kume, S. Kakutani, Y. Yamada et al., “Prognostic factors Treatment Reviews, vol. 38, no. 4, pp. 284–291, 2012. for renal cell carcinoma with bone metastasis: who are the [3] K. Gupta, J. D. Miller, J. Z. Li, M. W. Russell, and long-term survivors?” Journal of Urology, vol. 185, no. 5, C. Charbonneau, “Epidemiologic and socioeconomic burden pp. 1611–1614, 2011. of metastatic renal cell carcinoma (mRCC): a literature re- [19] M. Santoni, A. Conti, G. Procopio et al., “Bone metastases in view,” Cancer Treatment Reviews, vol. 34, no. 3, pp. 193–205, patients with metastatic renal cell carcinoma: are they always associated with poor prognosis?” Journal of Experimental & [4] J. Zekri, N. Ahmed, R. E. Coleman, and B. W. Hancock, “+e Clinical Cancer Research, vol. 34, no. 1, p. 10, 2015. skeletal metastatic complications of renal cell carcinoma,” [20] Q. Guo, C. Zhang, X. Guo et al., “Incidence of bone metastasis International Journal of Oncology, vol. 19, no. 2, pp. 379–382, and factors contributing to its development and prognosis in newly diagnosed renal cell carcinoma: a population-based [5] D. Y. C. Heng, W. Xie, G. A. Bjarnason et al., “Progression- study,” Cancer Management and Research, vol. 10, free survival as a predictor of overall survival in metastatic pp. 2935–2944, 2018. renal cell carcinoma treated with contemporary targeted [21] K. A. Cronin, L. A. G. Ries, and B. K. Edwards, “+e sur- therapy,” Cancer, vol. 117, no. 12, pp. 2637–2642, 2011. veillance, epidemiology, and end results (SEER) program of [6] T. Yuasa, S. Urakami, S. Yamamoto et al., “Treatment out- the national cancer institute,” Cancer, vol. 120, pp. 3755–3757, come and prognostic factors in renal cell cancer patients with bone metastasis,” Clinical & Experimental Metastasis, vol. 28, [22] J. Ranstam and J. A. Cook, “Kaplan-Meier curve,” )e British no. 4, pp. 405–411, 2011. Journal of Surgery, vol. 104, no. 4, p. 442, 2017. [7] Y. Toyoda, N. Shinohara, T. Harabayashi et al., “Survival and [23] J. Rueckert, K. Devitt, and J. A. Gardner, “Renal cell carci- prognostic classification of patients with metastatic renal cell noma with monosomy 8: a case series and review of the carcinoma of bone,” European Urology, vol. 52, no. 1, literature,” Journal of the Association of Genetic Technologists, pp. 163–169, 2007. vol. 44, no. 1, pp. 5–9, 2018. [8] F. Ruatta, L. Derosa, B. Escudier et al., “Prognosis of renal cell [24] B. Paule, E. Deslandes, S. E. Le Mouel et al., “Identification of a carcinoma with bone metastases: experience from a large novel biomarker signature associated with risk for bone cancer centre,” European Journal of Cancer, vol. 107, metastasis in patients with renal cell carcinoma,” )e Inter- pp. 79–85, 1990. national Journal of Biological Markers, vol. 25, no. 2, [9] M. H. Younis, L. Fuentes-Rivera, S. Summers, and J. Pretell- pp. 112–115, 2010. Mazzini, “Survival in patients with carcinomas presenting [25] T. Chandrasekar, Z. Klaassen, H. Goldberg, G. S. Kulkarni, with bone metastasis at diagnosis: a SEER population-based R. J. Hamilton, and N. E. Fleshner, “Metastatic renal cell cohort study,” Archives of Orthopaedic and Trauma Surgery, carcinoma: patterns and predictors of metastases-A con- vol. 141, no. 3, pp. 367–373, 2021. temporary population-based series,” Urologic Oncology, [10] H. B. Burke, “Outcome prediction and the future of the TNM vol. 35, no. 11, p. 661, 2017. staging system,” JNCI Journal of the National Cancer Institute, [26] P. V. Hauschka, A. E. Mavrakos, M. D. Iafrati, S. E. Doleman, vol. 96, no. 19, pp. 1408-1409, 2004. and M. Klagsbrun, “Growth factors in bone matrix. Isolation [11] E.-J. Jung, H. J. Lee, C. Kwak, J. H. Ku, and K. C. Moon, of multiple types by affinity chromatography on heparin- “Young age is independent prognostic factor for cancer- Sepharose,” Journal of Biological Chemistry, vol. 261, no. 27, specific survival of low-stage clear cell renal cell carcinoma,” pp. 12665–12674, 1986. Urology, vol. 73, no. 1, pp. 137–141, 2009. [27] S.-C. Chen and P.-L. Kuo, “Bone metastasis from renal cell [12] J. Wu, P. Zhang, G. Zhang et al., “Renal cell carcinoma carcinoma,” International Journal of Molecular Sciences, histological subtype distribution differs by age, gender, and vol. 17, no. 6, p. 987, 2016. tumor size in coastal Chinese patients,” Oncotarget, vol. 8, [28] J. L. Wright, M. C. Risk, J. Hotaling, and D. W. Lin, “Effect of no. 42, pp. 71797–71804, 2017. collecting duct histology on renal cell cancer outcome,” [13] C. T. Lee, J. Katz, P. A. Fearn, and P. Russo, “Mode of Journal of Urology, vol. 182, no. 6, pp. 2595–2600, 2009. presentation of renal cell carcinoma provides prognostic [29] N. Tokuda, S. Naito, O. Matsuzaki, Y. Nagashima, S. Ozono, information,” Urologic Oncology: Seminars and Original In- and T. Igarashi, “Collecting duct (Bellini duct) renal cell vestigations, vol. 7, no. 4, pp. 135–140, 2002. carcinoma: a nationwide survey in Japan,” Journal of Urology, [14] W. Zheng, Y. Huang, H. Chen et al., “Nomogram application vol. 176, no. 1, pp. 40–43, 2006. to predict overall and cancer-specific survival in osteosar- [30] P. I. Karakiewicz, Q.-D. Trinh, N. Rioux-Leclercq et al., coma,” Cancer Management and Research, vol. 10, “Collecting duct renal cell carcinoma: a matched analysis of 41 pp. 5439–5450, 2018. cases,” European Urology, vol. 52, no. 4, pp. 1140–1146, 2007. [15] S. Y. Kim, M. J. Yoon, Y. I. Park, M. J. Kim, B.-H. Nam, and S. R. Park, “Nomograms predicting survival of patients with [31] P. P. Lin, A. N. Mirza, V. O. Lewis et al., “Patient survival after surgery for osseous metastases from renal cell carcinoma,” )e unresectable or metastatic gastric cancer who receive com- bination cytotoxic chemotherapy as first-line treatment,” Journal of Bone and Joint Surgery-American Volume, vol. 89, Gastric Cancer, vol. 21, no. 3, pp. 453–463, 2018. no. 8, pp. 1794–1801, 2007. Journal of Oncology 17 [32] Y. Toyoda, N. Shinohara, T. Harabayashi et al., “Survival and prognostic classification of patients with metastatic renal cell carcinoma of bone,” European Urology, vol. 52, no. 1, pp. 163–169, 2007. [33] R. J. Motzer, B. Escudier, P. Tomczak et al., “Axitinib versus sorafenib as second-line treatment for advanced renal cell carcinoma: overall survival analysis and updated results from a randomised phase 3 trial,” )e Lancet Oncology, vol. 14, no. 6, pp. 552–562, 2013. [34] R. R. McKay, N. Kroeger, W. Xie et al., “Impact of bone and liver metastases on patients with renal cell carcinoma treated with targeted therapy,” European Urology, vol. 65, no. 3, pp. 577–584, 2014. [35] H. W. Kang, S. P. Seo, W. T. Kim et al., “Impact of young age at diagnosis on survival in patients with surgically treated renal cell carcinoma: a multicenter study,” Journal of Korean Medical Science, vol. 31, no. 12, pp. 1976–1982, 2016. [36] R. H. +ompson, M. A. Ordonez, A. Iasonos et al., “Renal cell carcinoma in young and old patients-is there a difference?” Journal of Urology, vol. 180, no. 4, pp. 1262–1266, 2008. [37] V. Grunwald, ¨ B. Eberhardt, A. Bex et al., “An interdisciplinary consensus on the management of bone metastases from renal cell carcinoma,” Nature Reviews Urology, vol. 15, no. 8, pp. 511–521, 2018. [38] A. Fottner, M. Szalantzy, L. Wirthmann et al., “Bone me- tastases from renal cell carcinoma: patient survival after surgical treatment,” BMC Musculoskeletal Disorders, vol. 11, p. 145, 2010. [39] B. Fuchs, R. T. Trousdale, and M. G. Rock, “Solitary bony metastasis from renal cell carcinoma,” Clinical Orthopaedics & Related Research, vol. 431, no. 431, pp. 187–192, 2005. [40] A. L. Alt, S. A. Boorjian, C. M. Lohse, B. A. Costello, B. C. Leibovich, and M. L. Blute, “Survival after complete surgical resection of multiple metastases from renal cell carcinoma,” Cancer, vol. 117, no. 13, pp. 2873–2882, 2011. [41] J. Lee, D. Hodgson, E. Chow et al., “A phase II trial of palliative radiotherapy for metastatic renal cell carcinoma,” Cancer, vol. 104, no. 9, pp. 1894–1900, 2005. [42] D. Wilson, L. Hiller, L. Gray, M. Grainger, A. Stirling, and N. James, “+e effect of biological effective dose on time to symptom progression in metastatic renal cell carcinoma,” Clinical Oncology, vol. 15, no. 7, pp. 400–407, 2003. [43] J. Zołnierek, P. Nurzynski, ´ P. Langiewicz et al., “Efficacy of targeted therapy in patients with renal cell carcinoma with pre-existing or new bone metastases,” Journal of Cancer Research and Clinical Oncology, vol. 136, no. 3, pp. 371–378,

Journal

Journal of Oncology – Hindawi Publishing Corporation

Published: May 12, 2021

There are no references for this article.