Abstract
Background
The intricate task of diagnosing and managing small renal masses (SRMs) has become progressively convoluted within the realm of clinical practice. Contemporary clinical prediction instruments may succumb to a gradual decay in precision, coupled with an absence of unambiguous guidelines to navigate patient management.
Methods
This investigation was devised to formulate and authenticate nomograms for the overall survival (OS) and cancer- specific survival (CSS) among patients afflicted with SRMs. The study encompassed a cohort of 2558 pediatric patients diagnosed with SRMs over the period of 2000 to 2019. Independent prognostic indicators for OS and CSS, encompassing historical staging, chemotherapy regimens, surgical interventions, and pathological classifications, were ascertained through the employment of multivariate Cox proportional hazards regression analysis and backward stepwise selection.
Results
Through the utilization of multivariate Cox regression models, nomograms for OS and CSS were meticulously crafted, demonstrating commendable discrimination and calibration within the training set (OS C-index: 0.762, CSS C-index: 0.779). The validation set further corroborated the exemplary discrimination and calibration of the nomograms. Moreover, these nomograms adeptly differentiated between patient groups at elevated and diminished risk levels.
Conclusion
The nomograms delineated in this research provide propitious predictive accuracy for overall survival and cancer-specific survival in patients suffering from pediatric SRMs, thereby contributing to refined risk stratification and steering the optimal therapeutic course of action. The necessity for supplementary validation prevails before the translation of these findings into clinical practice.
Keywords: small renal masses, overall survival, cancer- specific survival, nomograms
Introduction
Small renal masses (SRMs) are increasingly being recognized in clinical practice, not only due to their potential for malignant transformation but also due to the complexities associated with their management. SRMs are often detected incidentally and can present a therapeutic dilemma, particularly in patients with competing risks such as advanced age or significant comorbidities.1-3 While active surveillance (AS) has become a recommended initial treatment option for certain SRMs,1,2 the biological heterogeneity of SRMs and the lack of clear guidance in managing continued growth on AS add to the clinical challenge.4,5
In the pediatric population, renal tumors remain a significant concern, with Wilms tumor being the predominant form, accounting for nearly 95% of cases.6,7 The overall survival rate has improved, but late complications related to treatment, such as kidney impairment, cardiotoxicity, and secondary cancers, necessitate careful consideration of therapeutic interventions. 8
The management of SRMs, whether in children or adults, requires nuanced decision-making, balancing oncological outcomes with long-term health considerations. The growth kinetics of SRMs, with an average growth rate of 0.28-0.31 cm/year and variable growth rates in the first 6-12 months of AS,9-11 further complicate the treatment landscape. Existing clinical prediction tools may lose precision over time, and there is minimal guidance for patient management after a period of time on AS. Moreover, retrospective analyses have shown that overall survival (OS) in patients with SRM is primarily driven by age and competing risks of medical comorbidities. 12
Despite these complexities, studies on AS for SRM have shown the feasibility of this approach,11,13 and multi-institutional prospective clinical studies, such as the Delayed Intervention and Surveillance for Small Renal Masses (DISSRM) Registry, are emerging to validate retrospective findings and determine predictors of malignancy and progression for SRM. However, robust nomograms targeting both OS and Cancer- specific Survival (CSS) for SRMs, particularly those incorporating conditional survival estimates, are still relatively underexplored.14,15
Given these considerations, the present study aimed to construct and validate nomograms for OS and CSS in pediatric SRMs, incorporating comprehensive variables such as historical stage, chemotherapy, surgery, and pathology types. Our intention was to create practical and clinically relevant tools that can facilitate more precise risk stratification, guiding optimal therapeutic decisions for patients with SRMs, and potentially impacting long-term outcomes and quality of life.
Methods and Materials
Patients
In an exhaustive investigation, a cohort of 2558 patients diagnosed with small renal masses was ascertained between the years 2000 and 2019, utilizing the SEER database. ICD-O-3 Hist/behave is a code in the SEER database that represents the histology or pathological type of the patient's tumor. The ICD-O-3 (International Classification of Diseases for Oncology, Third Edition) is used to code the morphology and behavior of tumors. Patients diagnosed with renal tumors with a diameter of less than 4 cm. These patients were meticulously enrolled in this study, adhering to specific inclusion and exclusion criteria. The inclusion criteria were as follows: (i) ICD-O-3 Hist/behav, (ii) nephroblastoma, renal cell carcinoma, soft tissue sarcomas of other genitourinary tract among others. The exclusion criteria consisted of: (i) absence of complete laterality information, (ii) patients aged ≥ 18 years, (iii) absence of complete information on Survival months. The study design was showed in the Figure 1, while the methodology of patient selection is illustrated in Figure S1. Following enrollment, the patients were systematically divided into two cohorts: 2047 patients were assigned to the training set, while 511 were delegated to the validation set. An array of clinicopathological data was extracted for each case, including but not limited to age at diagnosis, race, sex, laterality, SEER historic stage A (1973-2015), radiation treatment (yes, no/unknown), chemotherapy treatment (yes, no/unknown), surgical intervention, LN positivity, pathology. Furthermore, follow-up data were compiled, incorporating for each case aspects such as survival duration, vital status records, and SEER cause-specific death classification. Overall Survival (OS) duration was precisely defined as the time sourced from diagnosis until death or last follow-up. Cancer- specific survival (CSS) was similarly defined as the probability of surviving a specific type of cancer in the absence of other causes of death. This study utilized de-identified data from the SEER database, which does not require Institutional Review Board (IRB) approval as it involves no direct interaction with human subjects or identifiable human tissue samples.
Figure 1.
Study design for the discovery and validation of a prediction model based on available clinical information to assess pediatric patients’ outcomes in small renal masses.
Nomograms Construction and Performance Assessment
Clinicopathological candidate predictors, encompassing histological type, were rigorously evaluated using a multivariate Cox proportional hazards regression algorithm within the training set. The process of backward stepwise selection, guided by Akaike's Information Criterion (AIC), was employed to discern the significant predictors of OS and CSS. 16 Subsequently, an OS nomogram and a CSS nomogram were meticulously constructed, each based on their respective multivariate Cox proportional hazards regression models. The efficacy of the nomograms was assessed with regard to their discrimination and calibration within the training set. Harrell's C-index, a widely recognized metric for evaluating the discriminative ability of prognostic models, was applied for a quantitative analysis of their performance. 17 It should be noted that bootstrapping, with 1000 resampling procedures, was utilized to derive the C-index, thereby correcting for any potential overfitting. The calibration of the nomograms was further examined through the plotting of calibration curves, facilitating a comparison between the nomogram-predicted survival probability and the observed survival probability.
Validation of the Nomograms
The efficacy of the two nomograms was meticulously evaluated within the validation set, with each being assessed independently. The multivariate Cox proportional hazards regression formulas, originally derived from the training set, were applied to the entirety of the validation set. Risk scores were scrupulously computed for each patient, serving to encapsulate the risk of both all-cause mortality and cancer-specific mortality. Thereafter, Cox proportional hazards regression was executed, utilizing the risk score as an integral factor within the validation set. In the final stage of analysis, the C-indices were calculated, and the calibration curves were plotted, thereby providing a rigorous validation of the performance of the nomograms.
Categorization of Patients into High- or Low-Risk Groups
A risk score for each patient was meticulously computed utilizing the multivariate Cox proportional hazards regression formula. Subsequently, all patients were classified into high-risk and low-risk cohorts, guided by the optimal risk score cutoff value, which was ascertained using X-tile plots within the training set. 16 The variance in the survival curves between the high-risk and low-risk groups was rigorously evaluated employing the log-rank test. In addition, stratified analyses were conducted within diverse subgroups, encompassing both the training and validation sets.
Clinical Usefulness of the Nomograms
The decision curve analysis (DCA) was deployed to gauge the clinical applicability of the proposed nomograms, through the calculation of net benefits across varying threshold probabilities. Serving as an integrative methodology, the DCA algorithm offers a comprehensive framework for the examination and comparison of disparate diagnostic and prognostic models. 17
Statistical Analyses
The X-tile software version 3.6.1 (Yale University School of Medicine, New Haven, CT, USA) was used to create the X-tile plots. X-tile plots provide a single method to automatically select the optimum cutoff based on the highest χ² value (ie, minimum P value) defined using a Kaplan-Meier survival analysis and the log-rank test. 18 All data analyses were performed using R software (version 4.0.3). Continuous variables are presented as means ± standard deviations (SD), and categorical variables are presented as frequencies and percentages. Survival rates were estimated using the Kaplan-Meier method, and survival curves were compared using the log-rank test. Univariate and multivariate Cox proportional hazards regression models were used to evaluate prognostic factors. The multivariate model employed a stepwise selection method, with the optimal model determined based on the Akaike Information Criterion (AIC). To validate the accuracy and stability of the models, we used calibration curves and ROC curves to assess the predictive ability of the models, and calculated Harrell's C-index and the area under the curve (AUC).
Results
Patient Clinicopathological Characteristics
Based on the inclusion and exclusion criteria, our study included a total of 2558 cases (Figure S1). All SRM patients were randomly assigned at an 8:2 ratio (training group = 2047, validation group = 511). Table 1 lists the clinical baseline characteristics of SRM (Small Renal Mass) patients, with no significant differences in each clinical characteristic between the training and validation groups. As Table 2 showed, Historic stage, Chemotherapy, Surgery, and Pathology were significantly related with the prognosis of SRM patients by univariable cox analysis. Historic stage, Chemotherapy, Surgery, and Pathology were significant in multivariable cox analysis by backward stepwise selection method.
Table 1.
Baseline Characteristics of Pediatric Small Renal Masses Patients.
| Training cohort (N = 2047) | Validation cohort (N = 511) | Total cohort | P value | |
|---|---|---|---|---|
| Age (year) | 4.50 ± 3.80 | 4.65 ± 3.90 | 4.50 ± 3.80 | .455 |
| Sex | .198 | |||
| Male | 942 (46%) | 252 (49.3%) | 1194 (46.7%) | |
| Female | 1105 (54%) | 259 (50.7%) | 1364 (53.3%) | |
| Race | .343 | |||
| White | 1523 (74.4%) | 403 (78.9%) | 1926 (75.3%) | |
| Black | 367 (17.9%) | 77 (15.1%) | 444 (17.4%) | |
| Asian or Pacific Islander | 103 (5%) | 21 (4.1%) | 124 (4.8%) | |
| AI/AN a | 26 (1.3%) | 5 (1%) | 31 (1.2%) | |
| Unknown | 28 (1.4%) | 5 (1%) | 33 (1.3%) | |
| Laterality | .087 | |||
| Left | 981 (47.9%) | 250 (48.9%) | 1231 (48.1%) | |
| Right | 923 (45.1%) | 239 (46.8%) | 1162 (45.4%) | |
| Bilateral | 143 (7%) | 22 (4.3) | 165 (6.5%) | |
| Historic stage | .596 | |||
| Distant | 514 (25.1%) | 126 (24.7%) | 640 (25.0%) | |
| Localized | 784 (38.3%) | 207 (40.5%) | 991 (38.7%) | |
| Regional | 676 (33%) | 165 (32.3%) | 841 (32.9%) | |
| Unstaged | 73 (3.6%) | 13 (2.5%) | 86 (3.4%) | |
| Surgery | .407 | |||
| No | 96 (4.7%) | 19 (3.7%) | 115 (4.5%) | |
| Yes | 1951 (95.3%) | 492 (96.3%) | 2443 (95.5%) | |
| Chemotherapy | .170 | |||
| No | 269 (13.1%) | 55 (10.8%) | 324 (12.7%) | |
| Yes | 1778 (86.9%) | 456 (89.2%) | 2234 (87.3%) | |
| Radiotherapy | .811 | |||
| No | 1075 (52.5%) | 272 (53.2%) | 1347 (52.7%) | |
| Yes | 972 (47.5%) | 239 (46.8%) | 1211 (47.3%) | |
| Lymph Node | .389 | |||
| Negative | 1619 (79.1%) | 399 (78.1%) | 2018 (78.9%) | |
| Positve | 362 (17.7%) | 100 (19.6%) | 462 (18.1%) | |
| Unknown | 66 (3.2%) | 12 (2.3%) | 78 (3.0%) | |
| Pathology | .55 | |||
| Nephroblastoma | 1741 (85.1%) | 439 (85.9%) | 2180 (85.2%) | |
| Renal cell carcinoma | 144 (7%) | 39 (7.6%) | 58 (2.3%) | |
| Soft tissue sarcomas (Other GUT) a | 116 (5.7%) | 21 (4.1%) | 137 (5.4%) | |
| Others | 46 (2.2%) | 12 (2.3%) | 183 (7.2%) |
AI/AN: American Indian/Alaska Native; Soft tissue sarcomas (Other GUT): Soft tissue sarcomas of other genitourinary tract.
Categoric data are expressed as number (%) and continuous data as median (IQR). Bold values are statistically significant (P < .05).
Table 2.
Univariable and Multivariable cox Regression Analysis for OS (Overall Survival) and CSS (Cancer-Specical Survival) Patients with Pediatric Small Renal Masses in Training Cohort.
| Characteristics | OS | CSS | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Univariable analysis | Multivariable analysis | Univariable analysis | Multivariable analysis | |||||||
| HR (95% CI) | P-Value | HR (95% CI) | P-Value | Type-3 test P-Value |
HR (95% CI) | P-Value | HR (95% CI) | P-Value | Type-3 test P-Value |
|
| Age (years) | 1.079 (1.048- 1.111) | <.001 | 1.085 (1.051-1.121) | <.001 | ||||||
| Sex | ||||||||||
| Female | ref | ref | ||||||||
| Male | 1.337 (1.013-1.764) | .0402 | 1.352 (0.994-1.84) | .055 | ||||||
| Race | ||||||||||
| White | 0.432 (0.177-1.055) | .065 | 0.601 (0.191-1.893) | .384 | ||||||
| Black | 0.673 (0.268-1.691) | .588 | 0.889 (0.274-2.880) | .844 | ||||||
| Asian or Pacific Islander | 0.727 (0.262-2.018) | .541 | 0.864 (0.238-3.140) | .824 | ||||||
| AI/AN a | ref | ref | ||||||||
| Unknown | 0.224 (0.026-1.913) | .171 | 0.367 (0.038-3.525) | .385 | ||||||
| Laterality | ||||||||||
| Left | 0.790 (0.4846-1.289) | .346 | 0.777 (0.456-1.325) | .354 | ||||||
| Right | 0.619 (0.375-1.021) | .060 | 0.578 (0.333-1.002) | .051 | ||||||
| Bilateral | ref | ref | ||||||||
| Historic stage | <.001 | <.001 | ||||||||
| Distant | ref | ref | ||||||||
| Localized | 0.146 (0.098-0.217) | <.001 | 0.134 (0.087-0.204) | <.001 | 0.090 (0.054-0.151) | <.001 | 0.081 (0.047-0.140) | <.001 | ||
| Regional | 0.248 (0.175-0.351) | <.001 | 0.253 (0.177-0.363) | <.001 | 0.222 (0.152-0.326) | <.001 | 0.224 (0.150-0.333) | <.001 | ||
| Unstaged | 0.412 (0.201-0.843) | .015 | 0.190 (0.088-0.412) | <.001 | 0.351 (0.154-0.799) | <.001 | 0.167 (0.068-0.408) | <.001 | ||
| Surgery | <.001 | <.001 | ||||||||
| No | ref | ref | ref | |||||||
| Yes | 0.198 (0.137-0.286) | <.001 | 0.207 (0.136-0.316) | <.001 | 0.221 (0.145-0.339) | <.001 | 0.231 (0.142-0.374) | <.001 | ||
| Chemotherapy | .008 | .058 | ||||||||
| No | ref | ref | ref | |||||||
| Yes | 1.122 (0.733-1.719) | .596 | 2.024 (1.179-3.474) | .011 | 1.298 (0.785-2.144) | .309 | 1.804 (0.963-3.377) | .065 | ||
| Radiotherapy | ||||||||||
| No | ref | ref | ||||||||
| Yes | 1.109 (0.841-1.462) | .464 | 1.181 (0.869-1.605) | .289 | ||||||
| Lymph Node | ||||||||||
| Negative | ref | |||||||||
| Positive | 1.977 (1.441-2.711) | <.001 | 2.085 (1.476-2.945) | <.001 | ||||||
| Unknown | 2.778 (1.572-4.910) | <.001 | 2.393 (1.212-4.723) | .011 | ||||||
| Pathology | <.001 | <.001 | ||||||||
| Nephroblastoma | ref | ref | ||||||||
| Renal cell carcinoma | 2.897 (1.953-4.299) | <.001 | 6.595 (4.069-10.688) | <.001 | 2.380 (1.497-3.786) | <.001 | 5.737 (3.290-10.005) | <.001 | ||
| Soft tissue sarcomas (Other GUT) a | 2.548 (1.625-3.994) | <.001 | 3.403 (2.160-5.361) | <.001 | 2.161 (1.282-3.642) | .003 | 2.952 (1.743-5.002) | <.001 | ||
| Others | 2.724 (1.434-5.175) | .002 | 1.502 (0.773 - 2.919) | .230 | 2.205 (1.029-4.726) | .042 | 1.207 (0.551 - 2.647) | <.001 | ||
AI/AN: American Indian/Alaska Native; Soft tissue sarcomas (Other GUT): Soft tissue sarcomas of other genitourinary tract.
Bold values are statistically significant (P < 0.05). HR Hazard ratio, CI confidence interval, ref reference.
Nomograms Construction and Performance Assessment
Historic stage, Chemotherapy, Surgery, and Pathology were identified as independent predictors of OS based on the multivariate Cox proportional hazards regression algorithm (Table 2). Then, the OS nomogram was constructed by incorporating these four predictors based on the multivariate Cox proportional hazards regression model (Figure 2A). The OS nomogram showed favorable discrimination with a C-index of 0.762 (95% CI, 0.726-0.798) in the training set (Figure 2G). The calibration curves for the 1-, 3- and 5-year OS showed favorable agreement between the nomogram predicted OS probability and actual OS probability, indicating good calibration of the OS nomogram in the training set (Figure S2). The proposed model had an area under the curve (AUC) of 0.850 (95% confidence interval [CI], 0.795-0.906) for predicting the 1-year OS in the training cohort, while 0.787 for 3-year OS and 0.776 for 5-year OS (Figure 2B).
Figure 2.
Performance of the prognostic nomograms for overall and cancer-specific survival in pediatric small renal masses. (A) Nomograms for predicting 1-year, 3-year, and 5-year overall survival of pediatric small renal masses. (B-C) Receiver operator characteristic (ROC) curves of the OS prediction model. (D) Nomograms for predicting 1-year, 3-year, and 5-year cancer-free survival of pediatric small renal masses. (E-F) Receiver operator characteristic (ROC) curves of the CSS prediction model. (G) C-indexes of the OS and CSS prediction models.
The four variables, including Historic stage, Chemotherapy, Surgery, and Pathology, were also found to be independent predictors of CSS based on the multivariate Cox proportional hazards regression algorithm (Table 2). The CSS nomogram was developed by incorporating these predictors (Figure 2D). The CSS nomogram yielded a C-index of 0.779 (95% CI, 0.743-0.815) (Figure 2G). The calibration curves for the 1-, 3- and 5-year CSS also showed favorable calibration of the CSS nomogram in the training set (Figure S2). The proposed model had an AUC of 0.868 (95% CI, 0.814-0.922) for predicting the 1-year CSS in the training cohort, while 0.800 for 3-year CSS and 0.791 for 5-year CSS (Figure 2E).
Validation of the Nomograms
The favorable discrimination of the OS nomogram was confirmed using the validation set (C-index [95% CI], 0.685 [0.613-0.757]). And good calibration of the OS nomogram was also observed in the validation set (Figure S2). As for the CSS nomogram, the C-index was 0.701 (95% CI, 0.626-0.776) (Figure 2G). The calibration curves for the 1-, 3- and 5-year CSS in the validation set also confirmed the good calibration of the CSS nomogram (Figure S2). Similarly, the 1-year OS and DFS models achieved high levels of discriminability with AUCs of 0.912 and 0.909 in the validation cohorts, respectively. While the AUCs for 3-year OS and CSS modes were 0.707 and 0.723, for 5-year OS and CSS were 0.685 and 0.692 (Figure 2C and F).
Categorization of Patients into High- or Low-Risk Groups
OS risk score = Historic stage (10 for Distant, 3.178 for Regional, 1.753 for Unstaged, and 0 for Localized) + Chemotherapy (3.502 for yes) + Surgery (7.818 for yes) + Pathology (9.371 for Renal Cell Carcinoma, 6.085 for Soft tissue sarcomas of other genitourinary tract, 2.02 for others and 0 for Nephroblastoma). CSS risk score = Historic stage (10 for Distant, 4.032 for Regional, 2.869 for Unstaged, and 0 for Localized) + Chemotherapy (2.350 for yes) + Surgery (5.844 for yes) + Pathology (6.962 for Renal Cell Carcinoma, 4.315 for Soft tissue sarcomas of other genitourinary tract, 0.750 for others and 0 for Nephroblastoma). The optimal OS risk score cutoff generated by the X-tile plots was 12.55 (Figure S4). All patients were classified into high-risk and low-risk groups according to the optimal cutoff value. We assessed the distributions of the OS risk score and OS status in the combined training and validation set, and found that patients with higher risk scores were more likely to have death (Figure 3A). There was a significant discrimination between the OS of the high-risk and low-risk patients in the training set (Figure 3A), which was confirmed in the validation set (Figure 3B). Thus, the OS nomogram can successfully distinguish patients with high risk of all-cause mortality.
Figure 3.
Kaplan-Meier analyses of CSS and OS according to the risk score in pediatric patients with small renal masses. Significant discrimination between the OS of the high-risk and low-risk patients was observed in the training set (A) and the validation set (B). Significant discrimination between the CSS of the high-risk and low-risk patients was also observed in the training set (C) and the validation set (D).
As for the CSS risk score, we defined an optimal cutoff value of 11.06 based on the X-tile plots (Figure S5). Accordingly, the patients were categorized into high-risk and low-risk groups for CSS. Patients with higher risk scores were more likely to have death due to small renal masses. Significant discrimination between the CSS of the high-risk and low-risk patients was observed both in the training and validation sets (Figure 3C and D). Therefore, those patients with high risk of cancer-specific mortality can be identified by using our CSS nomogram.
Clinical Usefulness of the Nomograms
The outcomes from the Decision Curve Analysis (DCA) demonstrate that employing the OS and CSS nomogram for clinical decision-making offers superior net benefits compared to either the universal treatment strategy or the no-treatment strategy (Figure 4), underscoring the clinical utility of this nomogram. Analogous observations were noted in both the training and validation cohorts (Figure S6).
Figure 4.
DCA for the OS and CSS prediction models.
Discussion
The present investigation unveils a multifaceted and nuanced understanding of small renal masses (SRMs), illuminating the complex interplay of Historic stage, Pathology, Chemotherapy, and Surgery in determining patient outcomes. Through rigorous analysis and development of predictive nomograms for Overall Survival (OS) and Cancer-specific Survival (CSS), this work adds substantive weight to the existing body of literature on SRMs.
At the crux of our investigation is the paramount significance of the historic stage in prognostic evaluations of SRMs. The relevance of our study stems from its resonance with the well-established consensus in the oncological landscape. 19 By delineating the precise influence of Historic stage on both OS and CSS, our findings forge a deeper connection between patient characteristics, treatment interventions, and ultimate outcomes. 20 These insights may pave the way for more targeted interventions in SRM management.
Moreover, we included unstaged historic stage and unknown lymph node data in our multivariate Cox regression analysis to comprehensively evaluate their impact on prognosis and ensure that our analysis reflects real-world clinical scenarios. In clinical practice, some patients may lack complete staging information due to emergency surgeries, limited diagnostic resources, or other factors. Excluding these patients from our analysis could introduce selection bias, making our results less representative of actual clinical outcomes. Including unstaged and unknown data helps provide a more accurate reflection of clinical reality. Furthermore, incorporating these data allows us to assess the potential impact on prognosis when complete staging information is unavailable, offering a more comprehensive risk assessment. Even if certain variables are not significant in the multivariate analysis, their inclusion helps control for the effects of other covariates, enhancing the robustness of our model. It is important to emphasize that this does not imply tumor staging should be omitted in clinical practice. On the contrary, accurate tumor staging is crucial for prognosis and patient survival.
Adding layers to our understanding of SRMs, our research offers an in-depth examination of the interlinkages between diverse pathological categorizations and associated risk assessments. By delving deep into the nuances of Pathology's influence, we advocate for a rigorous pathological scrutiny, harmonizing our findings with contemporary investigations that stress the significance of molecular markers in prognostic delineations.21-23 This unearths the ever-evolving nuances of pathological assessments pivotal to SRM therapeutic strategies.
We also considered the potential bias introduced by including both Chemotherapy and Pathology in the final model. Although we attempted a stratified analysis to separately evaluate the effect of chemotherapy in different pathology types (Wilms tumor and renal cell carcinoma), the limited sample size led to unstable and statistically insignificant results. The smaller sample size in each subgroup post-stratification reduced the statistical power and made it challenging to control for all potential confounders effectively. To ensure the reliability of our findings, we opted to include both Chemotherapy and Pathology in the multivariable Cox regression model. This comprehensive approach allows us to consider all patient characteristics holistically, providing a more robust and realistic prognostic assessment, thus minimizing potential bias.
The implications of Chemotherapy and Surgery as robust predictors are expounded in our study. Our findings echo the surgical literature's emphasis on primary surgical resection followed by chemotherapy for specific SRM stages.19-21 Moreover, the discernible impact of chemotherapy, especially its adverse effects, accentuates the imperative of strategic planning and execution.24-26 This alignment with current therapeutic practices underscores the holistic approach needed for effective SRM management.
What distinguishes our research is the adept synthesis of predictive nomograms, which seamlessly incorporate Historic stage, Pathology, Chemotherapy, and Surgery. This synthesis promises a leap in clinical practices, paving the way for tailored therapeutic roadmaps, enhancing the prospects of improved patient trajectories. Such advances mirror the contemporary shift towards bespoke oncological interventions. 4 Notwithstanding our contributions, we acknowledge the limitations of our research. The omission of pivotal components such as molecular and genetic markers begs further examination.22,23 The SEER database does not provide detailed information about specific surgical methods, such as partial nephrectomy, radical nephrectomy, and local therapy, which constrained our analysis to the presence or absence of surgery. Additionally, our analysis did not include tumor size, a well-established prognostic factor, due to limitations in data availability. In future studies, we plan to increase the sample size and incorporate multicenter data to further validate and refine our findings. This will enable us to more accurately assess the effect of chemotherapy in different pathology types and improve our prognostic models. Prospective explorations can delve into these arenas, further honing the predictive precision and expanding the utility spectrum of the nomograms.
Conclusion
Our study constitutes a seminal contribution to the field of urological oncology, offering a composite framework for understanding SRM prognosis. By integrating key variables of Historic stage, Pathology, Chemotherapy, and Surgery, and crafting nomograms that may guide clinical decision-making, this work stands as a beacon for personalized treatment strategies in SRM patients. It encapsulates the dynamism and complexity of SRM management, with ripples likely to extend across the broader landscape of oncological care.
Supplemental Material
Supplemental material, sj-docx-1-tct-10.1177_15330338241284845 for Crafting and Authenticating Prognostic Nomograms for Overall and Cancer-specific Survival in Pediatric Small Renal Masses: A Comprehensive Two-Decade Cohort Study by Jianqiu Kong, Yitong Zou, Yi Huang, Yuhui Yao, Qinghua Gan, Zhijian Chen, Weibin Xie and Xinxiang Fan in Technology in Cancer Research & Treatment
Footnotes
Contributors: JK, YZ, and XF conceptualized the research and drafted the initial manuscript; YZ, JK, YH, QG and ZC curated and interpreted the clinical datasets; YZ, WX and YY provided manuscript revisions and refinements; JK and XF oversaw the research design and endorsed the final manuscript.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethics Approval and Ethics Approval: Ethical committee clearance was not mandated for this research due to the open-access nature of the dataset. The investigation did not encompass identifiable personal data or engagement with human subjects, negating the need for informed consent.
Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Guangdong Basic and Applied Basic Research Foundation (2024A1515013201) and the National Natural Science Foundation of China (82002682, 81972731, 81773026, 81972383, 82203720, 82203188).
ORCID iD: Xinxiang Fan https://orcid.org/0000-0001-8495-6488
Supplemental Material: Supplemental material for this article is available online.
References
- 1.Kutikov A, Egleston BL, Wong YN, Uzzo RG. Evaluating overall survival and competing risks of death in patients with localized renal cell carcinoma using a comprehensive nomogram. J Clin Oncol. 2010;28(2):311-317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kunkle DA, Egleston BL, Uzzo RG. Excise, ablate or observe: The small renal mass dilemma–a meta-analysis and review. J Urol. 2008;179(4):1227-1233. discussion 1233-1224. [DOI] [PubMed] [Google Scholar]
- 3.Lane BR, Abouassaly R, Gao T, et al. Active treatment of localized renal tumors may not impact overall survival in patients aged 75 years or older. Cancer. 2010;116(13):3119-3126. [DOI] [PubMed] [Google Scholar]
- 4.Kutikov A, Fossett LK, Ramchandani P, et al. Incidence of benign pathologic findings at partial nephrectomy for solitary renal mass presumed to be renal cell carcinoma on preoperative imaging. Urology. 2006;68(4):737-740. [DOI] [PubMed] [Google Scholar]
- 5.Thompson RH, Kurta JM, Kaag M, et al. Tumor size is associated with malignant potential in renal cell carcinoma cases. J Urol. 2009;181(5):2033-2036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kalapurakal JA, Dome JS, Perlman EJ, et al. Management of Wilms’ tumour: Current practice and future goals. Lancet Oncol. 2004;5(1):37-46. [DOI] [PubMed] [Google Scholar]
- 7.Lowe LH, Isuani BH, Heller RM, et al. Pediatric renal masses: Wilms tumor and beyond. Radiographics. 2000;20(6):1585-1603. [DOI] [PubMed] [Google Scholar]
- 8.Termuhlen AM, Tersak JM, Liu Q, et al. Twenty-five year follow-up of childhood Wilms tumor: A report from the childhood cancer survivor study. Pediatr Blood Cancer. 2011;57(7):1210-1216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chawla SN, Crispen PL, Hanlon AL, Greenberg RE, Chen DY, Uzzo RG. The natural history of observed enhancing renal masses: Meta-analysis and review of the world literature. J Urol. 2006;175(2):425-431. [DOI] [PubMed] [Google Scholar]
- 10.Smaldone MC, Kutikov A, Egleston BL, et al. Small renal masses progressing to metastases under active surveillance: A systematic review and pooled analysis. Cancer. 2012;118(4):997-1006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Uzosike AC, Patel HD, Alam R, et al. Growth kinetics of small renal masses on active surveillance: Variability and results from the DISSRM registry. J Urol. 2018;199(3):641-648. [DOI] [PubMed] [Google Scholar]
- 12.Pierorazio PM, Johnson MH, Patel HD, et al. Management of renal masses and localized renal cancer. Agency for Healthcare Research and Quality (US). 2016. Report No.: 16-EHC001-EF. [PubMed] [Google Scholar]
- 13.Pierorazio PM, Johnson MH, Ball MW, et al. Five-year analysis of a multi-institutional prospective clinical trial of delayed intervention and surveillance for small renal masses: The DISSRM registry. Eur Urol. 2015;68(3):408-415. [DOI] [PubMed] [Google Scholar]
- 14.Karakiewicz PI, Suardi N, Capitanio U, et al. Conditional survival predictions after nephrectomy for renal cell carcinoma. J Urol. 2009;182(6):2607-2612. [DOI] [PubMed] [Google Scholar]
- 15.Bianchi M, Becker A, Hansen J, et al. Conditional survival after nephrectomy for renal cell carcinoma (RCC): Changes in future survival probability over time. BJU Int. 2013;111(8):E283-E289. [DOI] [PubMed] [Google Scholar]
- 16.Camp RL, Dolled-Filhart M, Rimm DL. X-tile: A new bio-informatics tool for biomarker assessment and outcome-based cut-point optimization. Clin Cancer Res. 2004;10(21):7252-7259. [DOI] [PubMed] [Google Scholar]
- 17.Vickers AJ, Elkin EB. Decision curve analysis: A novel method for evaluating prediction models. Med Decis Making. 2006;26(6):565-574. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Harrell FE, Jr, Lee KL, Mark DB. Multivariable prognostic models: Issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med. 1996;15(4):361-387. [DOI] [PubMed] [Google Scholar]
- 19.Frazier AL, Shamberger RC, Henderson TO, Diller L. Decision analysis to compare treatment strategies for stage I/favorable histology Wilms tumor. Pediatr Blood Cancer. 2010;54(7):879-884. [DOI] [PubMed] [Google Scholar]
- 20.Shamberger RC, Anderson JR, Breslow NE, et al. Long-term outcomes for infants with very low risk Wilms tumor treated with surgery alone in national wilms tumor study-5. Ann Surg. 2010;251(3):555-558. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Fernandez CV, Perlman EJ, Mullen EA, et al. Clinical outcome and biological predictors of relapse after nephrectomy only for very low-risk Wilms tumor: A report from children’s oncology group AREN0532. Ann Surg. 2017;265(4):835-840. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Perlman EJ, Grundy PE, Anderson JR, et al. WT1 Mutation and 11P15 loss of heterozygosity predict relapse in very low-risk wilms tumors treated with surgery alone: A children’s oncology group study. J Clin Oncol. 2011;29(6):698-703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.D’Angio GJ. Pre- or postoperative therapy for Wilms’ tumor? J Clin Oncol. 2008;26(25):4055-4057. [DOI] [PubMed] [Google Scholar]
- 24.Fernandez CV, Mullen EA, Chi YY, et al. Outcome and prognostic factors in stage III favorable-histology wilms tumor: A report from the children’s oncology group study AREN0532. J Clin Oncol. 2018;36(3):254-261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Metzger ML, Dome JS. Current therapy for Wilms’ tumor. Oncologist. 2005;10(10):815-826. [DOI] [PubMed] [Google Scholar]
- 26.Dome JS, Liu T, Krasin M, et al. Improved survival for patients with recurrent Wilms tumor: The experience at St. Jude children’s research hospital. J Pediatr Hematol Oncol. 2002;24(3):192-198. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material, sj-docx-1-tct-10.1177_15330338241284845 for Crafting and Authenticating Prognostic Nomograms for Overall and Cancer-specific Survival in Pediatric Small Renal Masses: A Comprehensive Two-Decade Cohort Study by Jianqiu Kong, Yitong Zou, Yi Huang, Yuhui Yao, Qinghua Gan, Zhijian Chen, Weibin Xie and Xinxiang Fan in Technology in Cancer Research & Treatment