Application of random survival forest to establish a nomogram combining clinlabomics-score and clinical data for predicting brain metastasis in primary lung cancer
Zhongxiang ShiYixin ChenAoyu LiuJingya ZengWanlin XieLin XinYangyang ChengHuimin XuJialing ZhouShan GaoChunyuan FengHongxia ZhangYihua Sun
0
Citation
37
Reference
10
Related Paper
Abstract:
Abstract Purpose To establish a nomogram for predicting brain metastasis (BM) in primary lung cancer at 12, 18, and 24 months after initial diagnosis. Methods In this study, we included 428 patients who were diagnosed with primary lung cancer at Harbin Medical University Cancer Hospital between January 2020 and January 2022. The endpoint event was BM. The patients were randomly categorized into two groups in a 7:3 ratio: training ( n = 299) and validation ( n = 129) sets. Least absolute shrinkage and selection operator was utilized to analyze the laboratory test results in the training set. Furthermore, clinlabomics-score was determined using regression coefficients. Then, clinlabomics-score was combined with clinical data to construct a nomogram using random survival forest (RSF) and Cox multivariate regression. Then, various methods were used to evaluate the performance of the nomogram. Results Five independent predictive factors (pathological type, diameter, lymph node metastasis, non-lymph node metastasis and clinlabomics-score) were used to construct the nomogram. In the validation set, the bootstrap C-index was 0.7672 (95% CI 0.7092–0.8037), 12-month AUC was 0.787 (95% CI 0.708–0.865), 18-month AUC was 0.809 (95% CI 0.735–0.884), and 24-month AUC was 0.858 (95% CI 0.792–0.924). In addition, the calibration curve, decision curve analysis and Kaplan–Meier curves revealed a good performance of the nomogram. Conclusions Finally, we constructed and validated a nomogram to predict BM risk in primary lung cancer. Our nomogram can identify patients at high risk of BM and provide a reference for clinical decision-making at different disease time points.Keywords:
Nomogram
Clinical endpoint
Primary tumor
Up to 15% of all patients with brain metastases have no clearly detected primary site despite intensive evaluation, and this incidence has decreased with the use of improved imaging technology. Radiosurgery has been evaluated as one of the treatment modality for patients with limited brain metastases. In this study, we evaluated the effectiveness of radiosurgery for brain metastases from unknown primary tumors.We retrospectively evaluated 540 patients who underwent gamma knife radiosurgery (GKRS) for brain metastases radiologically diagnosed between August 1992 and September 2007 in our institution. First, the brain metastases were grouped into metachronous, synchronous, and precocious presentations according to the timing of diagnosis of the brain metastases. Then, synchronous and precocious brain metastases were further grouped into 1) unknown primary; 2) delayed known primary; and 3) synchronous metastases according to the timing of diagnosis of the primary origin. We analyzed the survival time and time to new brain metastasis in each group.Of the 540 patients, 29 (5.4%) presented precocious or synchronous metastases (34 GKRS procedures for 174 lesions). The primary tumor was not found even after intensive and repeated systemic evaluation in 10 patients (unknown primary, 34.5%); found after 8 months in 3 patients (delayed known primary, 1.2%); and diagnosed at the same time as the brain metastases in 16 patients (synchronous metastasis, 55.2%). No statistically significant differences in survival time and time to new brain metastasis were found among the three groups.Identification of a primary tumor before GKRS did not affect the patient outcomes. If other possible differential diagnoses were completely excluded, early GKRS can be an effective treatment option for brain metastases from unknown primary tumor.
Primary tumor
Brain tumor
Cite
Citations (15)
Nomogram
Cite
Citations (235)
Nomogram
Cite
Citations (102)
Metastasis to the brain is frequent in adult cancer patients but rare among children. Advances in primary tumor treatment and the associated prolonged survival are said to have increased the frequency of brain metastasis in children. The authors present a series of cases of brain metastases in children diagnosed with a solid primary cancer, evaluate brain metastasis trends, and describe tumor type, patterns of occurrence, and prognosis.Patients with brain metastases whose primary cancer was diagnosed during childhood were identified in the 1990-2012 Tumor Registry at The University of Texas M.D. Anderson Cancer Center. A review of their hospital records provided demographic data, history, and clinical data, including primary cancer sites, number and location of brain metastases, sites of extracranial metastases, treatments, and outcomes.Fifty-four pediatric patients (1.4%) had a brain metastasis from a solid primary tumor. Sarcomas were the most common (54%), followed by melanoma (15%). The patients' median ages at diagnosis of the primary cancer and the brain metastasis were 11.37 years and 15.03 years, respectively. The primary cancer was localized at diagnosis in 48% of patients and disseminated regionally in only 14%. The primary tumor and brain metastasis presented synchronously in 15% of patients, and other extracranial metastases were present when the primary cancer was diagnosed. The remaining patients were diagnosed with brain metastasis after initiation of primary cancer treatment, with a median presentation interval of 17 months after primary cancer diagnosis (range 2-77 months). At the time of diagnosis, the brain metastasis was the first site of systemic metastasis in only 4 (8%) of the 51 patients for whom data were available. Up to 70% of patients had lung metastases when brain metastases were found. Symptoms led to the brain metastasis diagnosis in 65% of cases. Brain metastases were single in 60% of cases and multiple in 35%; 6% had only leptomeningeal disease. The median Kaplan-Meier estimates of survival after diagnoses of primary cancer and brain metastasis were 29 months (95% CI 24-34 months) and 9 months (95% CI 6-11 months), respectively. Untreated patients survived for a median of 0.9 months after brain metastasis diagnosis (95% CI 0.3-1.5 months). Those receiving treatment survived for a median of 8 months after initiation of therapy (95% CI 6-11 months).The results of this study challenge the current notion of an increased incidence of brain metastases among children with a solid primary cancer. The earlier diagnosis of the primary cancer, prior to its dissemination to distant sites (especially the brain), and initiation of presumably more effective treatments may support such an observation. However, although the actual number of cases may not be increasing, the prognosis after the diagnosis of a brain metastasis remains poor regardless of the management strategy.
Primary tumor
Brain tumor
Cite
Citations (29)
Nomogram
Biochemical recurrence
Cite
Citations (80)
Abstract This 1:5 case‐control study aimed to identify the risk factors of hospital‐acquired pressure injuries (HAPIs) and to develop a mathematical model of nomogram for the risk prediction of HAPIs. Data for 370 patients with HAPIs and 1971 patients without HAPIs were extracted from the adverse events and the electronic medical systems. They were randomly divided into two sets: training (n = 1951) and validation (n = 390). Significant risk factors were identified by univariate and multivariate analyses in the training set, followed by a nomogram constructed. Age, independent movement, sensory perception and response, moisture, perfusion, use of medical devices, compulsive position, hypoalbuminaemia, an existing pressure injury or scarring from a previous pressure injury, and surgery sufferings were considered significant risk factors and were included to construct a nomogram. In both of the training and validation sets, the areas of 0.90 under the receiver operating characteristic curves showed excellent discrimination of the nomogram; calibration plots demonstrated a good consistency between the observed probability and the nomogram's prediction; decision curve analyses exhibited preferable net benefit along with the threshold probability in the nomogram. The excellent performance of the nomogram makes it a convenient and reliable tool for the risk prediction of HAPIs.
Nomogram
Univariate
Pressure injury
Cite
Citations (13)
Nomogram
Cite
Citations (9)
Objective: To compare the diagnostic accuracy of various transcutaneous bilirubin (TcB) nomograms for predischarge screening. Methods: The paired total serum bilirubin (TSB) and TcB measurements collected in neonates ≥35 weeks and ≥2000 g birth weight were analyzed. BiliCare™ bilirubinometer was used for TcB measurement. We chose the following nomograms for the study: Bhutani nomogram, Maisel's nomogram, Agarwal nomogram, Thakkar nomogram, American Academy of Pediatrics (AAP) nomogram within 3 mg/dl of phototherapy cutoff, AAP nomogram >70% of phototherapy cutoff and if TcB value is above 13 mg/dl. The diagnostic accuracy of these nomograms for TcB was compared with TSB plotted in the Bhutani nomogram. Results: TcB showed a positive correlation with TSB (Pearson correlation coefficient = 0.783). Bhutani nomogram, Maisel's nomogram and AAP (using within 3 mg/dL cutoff) nomogram showed good sensitivity and low false-negative rate while avoiding blood draws in most neonates. Conclusion: Bhutani nomogram, Maisel's nomogram, and AAP (using within 3 mg/dL of phototherapy cutoff) nomograms have comparable diagnostic accuracy for predischarge bilirubin screening in neonates.
Nomogram
Cut-off
Cite
Citations (0)
Nomogram
Predictive modelling
Cite
Citations (96)
BACKGROUND The prognosis of patients with brain metastasis as the only manifestation of an undetected primary tumor generally is considered to be poor. Therefore, most treatment is palliative. The authors reviewed the clinical outcomes and treatment results of patients presenting with brain metastasis from an undetected primary tumor at The University of Texas M. D. Anderson Cancer Center. METHODS Between 1977-1996, 220 patients were referred to the study department for the treatment of brain metastasis from an undetected primary tumor. The patients' records were reviewed to identify those for whom brain metastasis was the only manifestation of the primary tumor. The majority of patients were excluded from the current analysis because extracranial metastasis also were present. Thirty-nine patients qualified for this retrospective review. The level of neurosurgical excision varied, but all patients received radiotherapy. Tumor control in the brain and survival were analyzed by various tumor-related and treatment-related factors. RESULTS In 31 patients, the brain metastasis were adenocarcinomas, whereas the remaining patients had tumors of various other histologies. In 12 patients, the primary tumor eventually was found, most commonly in the lung. The median survival time for all patients was 13.4 months. Overall survival rates (OS) at 1, 3, and 5 years were 56%, 19%, and 15%, respectively. Intracranial disease control was 72% at 5 years. Patients who received gross total resection (GTR) and radiotherapy had significantly better OS than patients who received radiotherapy alone. The OS of patients whose primary tumor was identified was similar to that of patients in whom the primary tumor remained occult. CONCLUSIONS Brain metastasis as the only manifestation of an unknown primary tumor is a distinct clinical entity. The prognosis for patients with this presentation is better than that of patients with brain metastasis in general. Although the majority of patients die of extracranial disease, a few will achieve long term survival. Treatment to the brain is effective in controlling local disease; aggressive treatment with GTR and radiotherapy is recommended. Cancer 1998;83:2181-2184. © 1998 American Cancer Society.
Primary tumor
Brain tumor
Cite
Citations (86)