Development of a prognostic model for long-term survival of young patients with bladder cancer: a retrospective analysis of the SEER Database

Abstract

Objectives

This study aims to present the clinical characteristics of young patients with bladder cancer (YBCa), evaluate related risk factors and construct a nomogram based on data acquired from the Surveillance, Epidemiology, and End Results (SEER) Database.

Design

Retrospective analysis of the SEER Database (2004–2015) for primary YBCa.

Setting and participants

Data for YBCa (defined as those aged 40 years or younger) were extracted from the SEER Database, which covers approximately 28% of the US population, using the SEER*Stat software (V.8.4.0.1). A total of 1233 YBCa were identified. Patients were randomly assigned to the training and validation sets. The database included clinicopathological features, demographic information and survival outcomes, such as age, gender, race, year of diagnosis, marital status at diagnosis, primary tumour site, histological type, tumour grade, tumour, node, metastases (TNM) staging, treatment regimen for the primary tumour, cause of death and survival time. A nomogram model was developed using univariate and multivariate analyses. The prediction model was validated using the consistency index (C-index), calibration curve and receiver operating characteristic curve.

Primary outcome measures

3-year, 5-year and 10-year overall survival (OS).

Results

1233 YBCa from 2004 to 2015 were randomly assigned to the training set (n=865) and validation set (n=368). Age, marital status, tumour grade, histological type and TNM staging were included in the nomogram. The C-index of the model was 0.876. The 3-year, 5-year and 10-year OS area under the curve values for the training and validation sets were 0.949, 0.923 and 0.856, and 0.919, 0.890 and 0.904, respectively. Calibration plots showed that the nomogram had a robust predictive accuracy.

Conclusions

To our knowledge, this is the first study to establish a precise nomogram predicting the 3-year, 5-year and 10-year OS in YBCa based on multivariate analyses. Our nomogram may serve as a valuable reference for future diagnostics and individualised treatments for YBCa. However, external validation is warranted to assess the accuracy and generalisability of our prognostic model.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The Surveillance, Epidemiology, and End Results (SEER) Database offers one of the largest sample sizes for patient analysis.

• We established and internally validated a robust nomogram for predicting overall survival in young patients with bladder cancer.

• Some key variables, such as smoking history, immunotherapy and targeted therapy administered, are known to be associated with the prognosis of young patients with bladder cancer but were unavailable in the SEER Database.

• The prognostic model has not been externally validated.

Introduction

Bladder cancer (BCa) stands as one of the most prevalent urological cancers. According to the WHO, BCa ranks 10th among all malignancies. Globally, 573 278 cases of BCa and 9007 cases of young patients with BCa (YBCa) are diagnosed every year. In 2020, an estimated 212 536 BCa-related deaths occurred worldwide (online supplemental figure 1).¹ Urothelial carcinoma is the most common among urinary bladder neoplasms, typically affecting elderly men, with a median initial diagnosis age of 65–70 years.^2–4

In contrast to the high incidence of BCa among the elderly population, BCa is rare in patients aged 40 years or younger, accounting for 15.8% of all BCa cases, according to the WHO.¹ At present, the clinicopathological features and prognosis information for YBCa are lacking.

Currently, physicians mostly rely on the American Joint Committee on Cancer (AJCC) tumour, node, metastases (TNM) staging system to determine prognosis and guide BCa treatment. However, this system is not specifically tailored for YBCa. Tumour invasion (T), regional lymph nodes (N) and distant metastases (M) are considered predictors of outcome according to this system.⁵ Nonetheless, TNM staging still has limitations in terms of prognosis and cannot be relied upon to provide accurate predictions. Several studies⁶ have shown that the neutrophils-to-albumin ratio in patients with cancer can predict both overall survival (OS) and cancer-specific survival (CSS). Despite this, several individualised characteristics, such as tumour size, lymph node metastasis, histological grade, lymph node removal, surgery and chemoradiotherapy, which may be predictive, have not been included in previous studies involving YBCa.^{7 8} In recent years, numerous prognostic factors have been integrated into nomograms based on correlation coefficients, enhancing their accuracy as survival predictors.⁹ Nomograms have proven effective in predicting outcomes in various cancers, including breast cancer, endometrial cancer and thyroid cancer, aiding physicians in making informed clinical decisions and promoting personalised medicine through statistical presentation and result visualisation.^10–12

To address this gap, we investigated the clinicopathological characteristics of patients aged 40 years or younger and identified associated prognostic factors. Subsequently, we developed a nomogram to predict OS at 3, 5 and 10 years for YBCa, aiming to facilitate the establishment of individualised treatment plans and improve patient outcomes.

Methods

Data source and patients

Data from the National Cancer Institute-funded Surveillance, Epidemiology, and End Results (SEER) Database (http://seer.cancer.gov/seerstat), which covers approximately 28% of the US population, were used for this study. The database included clinicopathological features, demographic information and survival outcomes, such as age, gender, race, year of diagnosis, marital status at diagnosis, primary tumour site, histological type, tumour grade, TNM staging based on the AJCC seventh edition, treatment regimen for the primary tumour, cause of death and survival time.¹³ The variables C67.0–C67.9, positive histology and first malignant primary indicator were used in the SEER*Stat software (V.8.4.0.1) to identify patients with BCa enrolled in the database between 1 January 2004 and 31 December 2015. The exclusion criteria were as follows: (1) age over 40 years; (2) survival time unknown or 0; (3) TNM stage unknown or T0; (4) unknown lymph node resection; (5) unknown tumour staging, marital status or race; and (6) unknown surgery.

The SEER Database encompasses a wide range of clinical variables, including sociodemographic information (such as age at diagnosis, gender, race and marital status), clinical characteristics (such as histological type, grade stage, TNM staging and distant metastasis), treatment characteristics (such as surgery, lymph node resection and chemotherapy regimen used) and prognostic indicators (such as survival time and survival status). Next, YBCa were categorised into three groups based on their age (<32, 32–37 and >37 years; online supplemental figure 2) following the optimal cut-off value that was calculated using the X-tile software (V.3.6.1, Yale University School of Medicine, USA). OS was defined as the time from diagnosis to death due to any cause.

Statistical analysis

The study cases were divided into training and validation sets in a 7:3 ratio, with all cases included in the study set. Among the three datasets, categorical data were expressed as numbers and percentages, and intergroup comparisons were performed using the Χ² test. Continuous variables were expressed as mean±SD, and the Student’s t-test was used to compare the baseline characteristics of the patients in the two groups.

Univariate and multivariate Cox regression hazard models were used to identify prognostic factors in YBCa. The nomogram construction and validation were based on the results from the Cox proportional hazard estimation. Additionally, the consistency index (C-index), receiver operating characteristic (ROC) curve and calibration curve were used to evaluate the performance and accuracy of the nomogram. A C-index and area under the curve (AUC) of >0.70 are considered acceptable on the ROC graph. All statistical analyses were performed using SPSS (V.25.0) and R (http://www.rproject.org/). P values of <0.05 were considered of statistical significance.

Patient and public involvement

None.

Results

Demographic and clinicopathological characteristics

A total of 1233 YBCa from the SEER Database were eligible for enrolment in our study. The training set comprised of 865 YBCa, while the validation set included 368 patients following randomisation. As depicted in table 1, the basic characteristics of YBCa and Χ² test results varied between the sets. Of the total set, 403 individuals (32.7%) were younger than 32 years, 530 (43.0%) were between 32 and 37 years old, and 300 (24.3%) were above 37 years old. In terms of race and gender, the majority of patients were white, and there was a predominance of men. Transitional cell carcinoma (TCC) was the predominant histological type affecting 93.8% of patients, followed by squamous cell carcinoma and adenocarcinoma in 2.3% and 3.2% of the total patients, respectively. Regarding the TNM stage, there were 1065 cases classified as stages Ta, Tis and T1, and 168 cases as stages T2–4, accounting for 86.4% and 13.6%, respectively. Regional and distant lymph nodes were not observed in the majority of YBCa. There was no difference in the baseline data between the training and validation sets (all p>0.05, table 1).

Table 1.

Baseline demographic and clinicopathological characteristics for YBCa, n (%)

Variable	Overall set	Training set	Validation set	P value
Age, years				0.518
<32	403 (32.7%)	286 (33.1%)	117 (31.8%)
32~37	530 (43.0%)	363 (42.0%)	167 (45.4%)
>37	300 (24.3%)	216 (25.0%)	84 (22.8%)
Sex				0.554
Female	331 (26.8%)	228 (26.4%)	103 (28.0%)
Male	902 (73.2%)	637 (73.6%)	265 (72.0%)
Race				0.333
White	1070 (86.8%)	757 (87.5%)	313 (85.1%)
Black	82 (6.7%)	57 (6.6%)	25 (6.8%)
Others	81 (6.6%)	51 (5.9%)	30 (8.2%)
Marital status				0.419
Single	510 (41.4%)	351 (40.6%)	159 (43.2%)
Married	659 (53.4%)	465 (53.8%)	194 (52.7%)
SDW	64 (5.2%)	49 (5.7%)	15 (4.1%)
Grade				0.201
I	301 (24.4%)	210 (24.3%)	91 (24.7%)
II	593 (48.1%)	429 (49.6%)	164 (44.6%)
III	136 (11.0%)	86 (9.9%)	50 (13.6%)
IV	203 (16.5%)	140 (16.2%)	63 (17.1%)
Histological type				0.718
TCC	1156 (93.8%)	809 (93.5%)	343 (94.3%)
SCC	28 (2.3%)	22 (2.5%)	6 (1.6%)
Adenocarcinoma	40 (3.2%)	27 (3.1%)	13 (3.5%)
Others	9 (0.7%)	7 (0.8%)	2 (0.5%)
Stage T				0.893
Ta+Tis	867 (70.3%)	609 (70.4%)	258 (70.1%)
T1	198 (16.1%)	138 (16.0%)	60 (16.3%)
T2	79 (6.4%)	54 (6.2%)	25 (6.8%)
T3	50 (4.1%)	34 (3.9%)	16 (4.3%)
T4	39 (3.2%)	30 (3.5%)	9 (2.4%)
Stage N				0.103
N0	1178 (95.5%)	821 (94.9%)	357 (97.0%)
N1–N3	55 (4.5%)	44 (5.1%)	11 (3.0%)
Stage M				0.177
M0	1214 (98.5%)	849 (98.2%)	365 (99.2%)
M1	19 (1.5%)	16 (1.8%)	3 (0.8%)
Surgery				0.216
None	44 (3.6%)	34 (3.9%)	10 (2.7%)
TURB	853 (69.2%)	606 (70.1%)	247 (67.1%)
Radical cystectomy	68 (5.5%)	45 (5.2%)	23 (6.3%)
Pelvic exenteration	41 (3.3%)	32 (3.7%)	9 (2.4%)
Others	227 (18.4%)	148 (17.1%)	79 (21.5%)
Lymph nodes removed				0.835
None	1109 (89.9%)	777 (89.8%)	332 (90.2%)
Removed	124 (10.1%)	88 (10.2%)	36 (9.8%)
Chemotherapy				0.776
None/unknown	989 (80.2%)	692 (80.0%)	297 (80.7%)
Yes	244 (19.8%)	173 (20.0%)	71 (19.3%)
Bone metastasis				0.557
No	532 (99.4%)	370 (99.2%)	162 (100.0%)
Yes	3 (0.6%)	3 (0.8%)	0 (0.0%)
Brain metastasis				1.000
No	534 (99.8%)	372 (99.7%)	162 (100.0%)
Yes	1 (0.2%)	1 (0.3%)	0 (0.0%)
Liver metastasis				1.000
No	534 (99.8%)	372 (99.7%)	162 (100.0%)
Yes	1 (0.2%)	1 (0.3%)	0 (0.0%)
Lung metastasis				0.989
No	530 (99.1%)	369 (98.9%)	161 (99.4%)
Yes	5 (0.9%)	4 (1.1%)	1 (0.6%)
Status				0.990
Alive	1059 (85.9%)	743 (85.9%)	316 (85.9%)
Dead	174 (14.1%)	122 (14.1%)	52 (14.1%)
Survival month	109 (65–151)	112 (66–151)	106 (62.3–150)

M, metastases; N, node; SCC, squamous cell carcinoma; SDW, separated/divorced/widowed; T, tumour; TCC, transitional cell carcinoma; TURB, transurethral resection of the bladder; YBCa, young patients with bladder cancer.

Identification of OS prognostic factors in the training set

Univariate and multivariate Cox regression analyses were performed to identify independent prognostic factors for OS. Univariate Cox regression analysis showed that age at diagnosis, marital status, grade stage, histological type, TNM stage, surgery of the primary tumour, lymph node resection and chemotherapy used were factors related to OS. In the subsequent multivariate Cox regression, variables with a p value less than 0.05 in the univariate analyses were included. Notably, the TNM staging system, widely employed for predicting cancer prognosis, retained its significance as an independent prognostic risk factor, consistent with our study findings. The results of the multivariate Cox regression model further identified tumour grade, histological type and age at diagnosis as prognostic risk factors. Interestingly, marital status emerged as a prognostic protective factor for OS in these patients (table 2). Kaplan-Meier curves for the relevant variables are illustrated in figure 1 .

Table 2.

Univariate and multivariable analyses of prognostic factors for YBCa

Variables	Univariate analysis			Multivariate analysis
	HR	95% CI	P value	HR	95% CI	P value
Age, years
<32	Reference			Reference
32~37	1.869	1.146 to 3.050	0.012*	1.752	1.020 to 3.010	0.042*
>37	2.725	1.652 to 4.495	<0.001†	2.791	1.586 to 4.912	<0.001†
Sex
Female	Reference
Male	0.729	0.498 to 1.067	0.104
Race
White	Reference
Black	1.781	1.000 to 3.172	0.050
Others	1.015	0.472 to 2.183	0.970
Marital status
Single	Reference			Reference
Married	0.633	0.435 to 0.920	0.017*	0.533	0.348 to 0.816	0.004*
SDW	1.383	0.743 to 2.575	0.307	1.403	0.720 to 2.738	0.320
Grade
I	Reference			Reference
II	1.387	0.716 to 2.685	0.322	0.868	0.432 to 1.744	0.690
III	9.355	4.853 to 18.032	<0.001†	3.108	1.486 to 6.502	0.003*
IV	6.814	3.582 to 12.963	<0.001†	2.067	0.973 to 4.392	0.059
Histological type
TCC	Reference			Reference
SCC	12.788	7.360 to 22.220	<0.001†	3.419	1.597 to 7.323	0.002*
Adenocarcinoma	12.406	7.361 to 20.910	<0.001†	3.776	1.604 to 8.886	0.002*
Others	5.300	1.674 to 16.780	0.005*	2.779	0.839 to 9.204	0.094
Stage T
Ta+Tis	Reference			Reference
T1	3.230	1.877 to 5.560	<0.001†	1.997	1.110 to 3.595	0.021*
T2	11.740	6.860 to 20.090	<0.001†	6.991	3.592 to 13.604	<0.001†
T3	22.361	12.911 to 38.730	<0.001†	5.810	2.317 to 14.566	<0.001†
T4	33.419	19.446 to 57.430	<0.001†	8.051	3.305 to 19.612	<0.001†
Stage N
N0	Reference			Reference
N1–N3	16.887	11.100 to 25.690	<0.001†	2.007	1.070 to 3.767	0.030*
Stage M
M0	Reference			Reference
M1	27.678	15.670 to 48.900	<0.001†	5.239	2.594 to 10.579	<0.001†
Surgery
None	Reference			Reference
TURB	0.357	0.162 to 0.784	0.010*	0.490	0.212 to 1.134	0.095
Radical cystectomy	2.790	1.200 to 6.507	0.018*	1.036	0.323 to 3.322	0.953
Pelvic exenteration	3.317	1.362 to 8.082	0.008*	0.727	0.209 to 2.524	0.615
Others	0.633	0.270 to 1.483	0.292	0.678	0.272 to 1.688	0.404
Lymph nodes removed
None	Reference			Reference
Removed	7.843	5.398 to 11.400	<0.001†	0.457	0.184 to 1.137	0.092
Chemotherapy
None/unknown	Reference			Reference
Yes	3.618	2.515 to 5.206	<0.001†	0.601	0.359 to 1.008	0.053

*P<0.05.

†P<0.001.

M, metastases; N, node; SCC, squamous cell carcinoma; SDW, separated/divorced/widowed; T, tumour; TCC, transitional cell carcinoma; TURB, transurethral resection of the bladder; YBCa, young patients with bladder cancer.

Figure 1.

Kaplan-Meier-estimated OS in YBCa stratified by age (A), marital status (B), grade (C), histological type (D), T stage (E), N stage (F), M stage (G), lymph node resection (H) and chemotherapy (I). M, metastases; N, node; OS, overall survival; T, tumour; YBCa, young patients with bladder cancer.

Development of a prognostic nomogram for OS in the training set

Nomograms were constructed to develop a diagnostic model based on the results of multivariate Cox regression models. Age at diagnosis, marital status, grade stage, histological type and TNM staging (figure 2) were independent prognostic factors for the 3-year, 5-year and 10-year OS. For each variable, a vertical line was drawn along the top row of the nomogram. The total point axis represented the sum of the scores, and on the survival axis for 3, 5 and 10 years, a downward line was drawn to indicate the likelihood of survival.

Figure 2.

The nomogram contains independent prognostic factors for the 3-year, 5-year and 10-year OS prediction of YBCa. M, metastases; N, node; OS, overall survival; SCC, squamous cell carcinoma; SDW, separated/divorced/widowed; T, tumour; TCC, transitional cell carcinoma; YBCa, young patients with bladder cancer.

Validation and calibration of the nomogram for OS in the training set

Using the C-index and ROC curves, we determined the accuracy of the nomograms in predicting survival in months. Model performance showed a positive correlation with C-index values ranging from 0.50 to 1.00. In both the training and validation sets, the C-index value was consistent at 0.876. In general, AUC value usually ranged from 0.5 and 1, and a higher AUC value indicates better discrimination ability for the nomogram. The AUC is a measure of how well the model distinguishes between different outcomes; the closer the AUC is to 1, the better the predictive value. In both the training and validation sets, the ROC curve estimation of the nomogram demonstrated acceptable accuracy. The 3-year, 5-year and 10-year AUC values for the training set were 0.949, 0.923 and 0.856, respectively (figure 3A–C). The 3-year, 5-year and 10-year AUC values for the validation set were 0.919, 0.890 and 0.904, respectively (figure 3D–F). Based on these results, our model exhibited relatively high accuracy.

Figure 3.

The ROC curve for predicting YBCa survival at 3 years (A and D), 5 years (B and E) and 10 years (C and F) in the training and validation sets. The calibration curve for predicting YBCa survival at 3 years (G and J), 5 years (H and K) and 10 years (I and L) in the training and validation sets. AUC, area under the curve; ROC, receiver operating characteristic; YBCa, young patients with bladder cancer.

The 3-year, 5-year and 10-year nomograms were calibrated simultaneously for both the training and validation sets. The prediction falls on a diagonal of 45° based on a fully calibrated model. Based on the calibration plots, the nomogram exhibited a favourable predictive accuracy (figure 3G–L), demonstrating good agreement between the nomogram predictions and observed results.

Discussion

Prognostic nomograms serve as widely used tools in medical oncology, integrating multiple relevant clinical characteristics and prognostic factors to evaluate an individual’s survival probability at different time points. Here, we identified risk factors for OS associated with YBCa and developed a prognostic nomogram.

First, an analysis of 12 clinical parameters in YBCa was conducted. Univariate analysis revealed that age at diagnosis, marital status, grade stage, histological type, TNM staging, surgical resection of the primary tumour, lymph node removal and chemotherapy were factors related to OS. Next, we included factors with p values lower than 0.05 in the multivariate analysis, revealing that age at diagnosis, marital status, grade stage, histological type and TNM staging were independent prognostic factors for OS in YBCa. Thus, we used these variables to construct a prognostic nomogram for YBCa.

The nomogram model indicated that T and M stages had the most significant impact on prognosis, highlighting the TNM stage as a crucial determinant of disease prognosis. Contrary to traditional treatment modalities influencing outcomes in patients with BCa, such as surgery and chemotherapy, our study found that surgery, lymph node removal and chemotherapy were not independent prognostic factors for OS in YBCa and were thus not included in our nomograms.

The model demonstrated favourable predictiveness based on calibration curve, C-index and ROC curve assessments. Nomograms with a C-index and AUC over 0.70 are considered to have good discrimination.^{14 15} Across both the training and validation sets, our model consistently exceeded this threshold (3-year survival AUC was 0.944 and 0.919, 5-year survival AUC was 0.923 and 0.890, 10-year survival AUC was 0.856 and 0.904, and the C-index values were 0.876 and 0.876, respectively). Therefore, the ROC curve analysis and C-index results collectively indicated the nomogram’s value as a prognostic indicator, demonstrating sufficient ability to predict the survival of YBCa. The prediction model, characterised by its simplicity and comprehensibility, involves drawing a vertical line from each clinicopathological parameter to the ‘points’ line and summing the scores to calculate the ‘total points’. Subsequently, a vertical line drawn from ‘total points’ to ‘3-year, 5-year and 10-year survival’ provides the survival probabilities. As an illustrative example, for a 35-year-old single male patient diagnosed with non-muscle invasive (T1) bladder TCC, poorly differentiated (grade III), without lymph node metastases and distant metastasis, the ‘total points’ amounted to 167. The corresponding 3-year, 5-year and 10-year survival rates ranged from 80% to 90%, 70% to 80% and 60% to 70%, respectively.

In our study, clinical variables such as age at diagnosis, marital status, grade stage, histological type and AJCC staging were independent risk factors for OS of YBCa.

Previous studies^16–21 have associated BCa risk with factors such as smoking, gender, ethnicity, age and arsenic content in drinking water. Our findings align with these studies, as we observed a male/female ratio of 2.73, consistent with several prior investigations.²² Several previous studies^23–28 have reported this cause. This gender difference may be attributed to interactions between oestrogen, androgen and liver function. Smoking, a known risk factor for BCa, contributes to this gender disparity as smoking prevalence is higher among males. Age is another risk factor, with our study revealing an increased risk of mortality with advancing age.²⁹ In our study, we found that the risk of mortality increased with increasing age (32~37 years vs <32 years, HR=1.752, 95% CI=1.020 to 3.010, p=0.042; >37 years vs <32 years, HR=2.791, 95% CI=1.586 to 4.912, p<0.001). This finding is consistent with previous studies,^30–32 which attribute the age-related increase in risk to declining physical function and resistance. Furthermore, the histological type is also affected by age. Compérat et al ³³ noted a decrease in papillary urothelial neoplasia of low malignant potential cases from 40.3% to 27.2% with an age group transition from under 30 years to over 30 years, emphasising the impact of age on histological characteristics and subsequent prognosis.

Marital status holds significant prognostic implications in cancer. Li et al ³⁴ observed an independent association between marital status, OS (p=0.001) and CSS (p=0.002) in patients with upper urinary tract urothelial carcinoma. Unmarried and separated/divorced/widowed patients have a worse OS (HR=1.187, 95% CI=1.016 to 1.386; and HR=1.205, 95% CI=1.094 to 1.327) and CSS (HR=1.102, 95% CI=0.877 to 1.385; and HR=1.309, 95% CI=1.131 to 1.514) than married patients. In a study³⁵ on non-metastatic urothelial bladder cancer post-radical cystectomy, findings suggested a sex-dependent prognostic effect of marital status. Unmarried men exhibited more advanced tumour stages and worse cancer-specific mortality compared with married men, whereas women’s oncological outcomes remained unaffected by marital status. Other studies³⁶ highlighted that unmarried or widowed patients with bladder urothelial carcinoma were more prone to worse cancer-specific mortality. Our study yielded similar conclusions; compared with single patients, married YBCa had a reduced risk of overall death (HR=0.533, 95% CI=0.348 to 0.816, p=0.004). Several factors might contribute to this phenomenon. Marital status and survival are influenced by psychosocial factors and social support. An epidemiological study³⁷ indicated that depression, psychological stress and lack of social support could contribute to cancer progression. The generally better health of married individuals could be attributed to lower stress levels, greater material resources and support, and engagement in fewer risky health behaviours.³⁸ Conversely, unmarried or widowed patients may lack sufficient social and emotional support.³⁹

However, our study has several limitations. First, the SEER Database lacks detailed information on crucial factors such as smoking habits, comorbidities (American Society of Anesthesiologists score, Charlson Comorbidity Index, etc), neoadjuvant chemotherapy and targeted drugs, which play a pivotal role in predicting outcomes for YBCa. Second, our study was limited to internal verification, and external validation remains to be done.

Conclusions

Our study identified several prognostic factors for YBCa through comprehensive univariate and multivariate analyses. Independent prognostic factors for the OS of YBCa included age at diagnosis, marital status, grade stage, histological type and TNM staging. Furthermore, we have developed a precise nomogram, using the aforementioned factors, to predict the 3-year, 5-year and 10-year OS of YBCa. This novel tool may offer valuable insights for future diagnostic approaches and the formulation of individualised treatment plans for YBCa. However, external validation of the predictive model remains to be done.

Data availability statement

Data are available in a public, open access repository. The data can be accessed from the SEER Database (www.seer.cancer.gov).

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

The requirement of ethical approval for this study was waived by the Institutional Review Board of Hebei Medical University and Hebei General Hospital because the data were accessed from SEER (a publicly available database). All methods were carried out in accordance with relevant guidelines and regulations (Declaration of Helsinki). Due to the retrospective nature of the study, informed consent was not required.