Impact of time to local recurrence on the occurrence of metastasis in breast cancer patients treated with neoadjuvant chemotherapy: A random forest survival approach

Machine Learning to Predict Mortality in Older Patients With Cancer: Development and External Validation of the Geriatric Cancer Scoring System Using Two Large French Cohorts

PURPOSE

Establishing an accurate prognosis remains challenging in older patients with cancer because of the population's heterogeneity and the current predictive models' reduced ability to capture the complex interactions between oncologic and geriatric predictors. We aim to develop and externally validate a new predictive score (the Geriatric Cancer Scoring System [GCSS]) to refine individualized prognosis for older patients with cancer during the first year after a geriatric assessment (GA).

MATERIALS AND METHODS

Data were collected from two French prospective multicenter cohorts of patients with cancer 70 years and older, referred for GA: ELCAPA (training set January 2007-March 2016) and ONCODAGE (validation set August 2008-March 2010). Candidate predictors included baseline oncologic and geriatric factors and routine biomarkers. We built predictive models using Cox regression, single decision tree (DT), and random survival forest (RSF) methods, comparing their predictive performance for 3-, 6-, and 12-month mortalities by computing time-dependent area under the receiver operator curve (tAUC).

RESULTS

A total of 2,012 and 1,397 patients were included in the training and validation set, respectively (mean age: 81 ± 6 years/78 ± 5 years; women: 47%/70%; metastatic cancer: 50%/34%; 12-month mortality: 43%/16%). Tumor site/metastatic status, cancer treatment, weight loss, ≥five prescription drugs, impaired functional status and mobility, abnormal G-8 score, low creatinine clearance, and elevated C-reactive protein (CRP)/albumin were identified as relevant predictors in the Cox model. DT and RSF identified more complex combinations of features, with G-8 score, tumor site/metastatic status, and CRP/albumin ratio contributing most to the predictions. The RSF approach gave the highest tAUC (12 months: 0.87 [RSF], 0.82 [Cox], 0.82 [DT]) and was retained as the final model.

CONCLUSION

The GCSS on the basis of a machine learning approach applied to two large French cohorts gave an accurate externally validated mortality prediction. The GCSS might improve decision making and counseling in older patients with cancer referred for pretherapeutic GA. GCSS's generalizability must now be confirmed in an international setting.

Comparative study of machine learning and statistical survival models for enhancing cervical cancer prognosis and risk factor assessment using SEER data

Cervical cancer is a common malignant tumor of the female reproductive system and the leading cause of death among women worldwide. The survival prediction method can be used to effectively analyze the time to event, which is essential in any clinical study. This study aims to bridge the gap between traditional statistical methods and machine learning in survival analysis by revealing which techniques are most effective in predicting survival, with a particular emphasis on improving prediction accuracy and identifying key risk factors for cervical cancer. Women with cervical cancer diagnosed between 2013 and 2015 were included in our study using data from the Surveillance, Epidemiology, and End Results (SEER) database. Using this dataset, the study assesses the performance of Weibull, Cox proportional hazards models, and Random Survival Forests in terms of predictive accuracy and risk factor identification. The findings reveal that machine learning models, particularly Random Survival Forests (RSF), outperform traditional statistical methods in both predictive accuracy and the discernment of crucial prognostic factors, underscoring the advantages of machine learning in handling complex survival data. However, for a survival dataset with a small number of predictors, statistical models should be used first. The study finds that RSF models enhance survival analysis with more accurate predictions and insights into survival risk factors but highlights the need for larger datasets and further research on model interpretability and clinical applicability.

Automatic image segmentation and online survival prediction model of medulloblastoma based on machine learning

OBJECTIVES

To develop a dynamic nomogram containing radiomics signature and clinical features for estimating the overall survival (OS) of patients with medulloblastoma (MB) and design an automatic image segmentation model to reduce labor and time costs.

METHODS

Data from 217 medulloblastoma (MB) patients over the past 4 years were collected and separated into a training set and a test set. Intraclass correlation coefficient (ICC), random survival forest (RSF), and least absolute shrinkage and selection operator (LASSO) regression methods were employed to select variables in the training set. Univariate and multivariate Cox proportional hazard models, as well as Kaplan-Meier analysis, were utilized to determine the relationship among the radiomics signature, clinical features, and overall survival. A dynamic nomogram was developed. Additionally, a 3D-Unet deep learning model was used to train the automatic tumor delineation model.

RESULTS

Higher Rad-scores were significantly associated with worse OS in both the training and validation sets (p < 0.001 and p = 0.047, respectively). The Cox model combined clinical and radiomics signatures ([IBS = 0.079], [C-index = 0.747, SE = 0.045]) outperformed either radiomics signatures alone ([IBS = 0.081], [C-index = 0.738, SE = 0.041]) or clinical features alone ([IBS = 0.085], [C-index = 0.565, SE = 0.041]). The segmentation model had mean Dice coefficients of 0.80, 0.82, and 0.78 in the training, validation, and test sets respectively. A deep learning-based tumor segmentation model was built with Dice coefficients of 0.8372, 0.8017, and 0.7673 on the training set, validation set, and test set, respectively.

CONCLUSIONS

A combination of radiomics features and clinical characteristics enhances the accuracy of OS prediction in medulloblastoma patients. Additionally, building an MRI image automatic segmentation model reduces labor and time costs.

CLINICAL RELEVANCE STATEMENT

A survival prognosis model based on radiomics and clinical characteristics could improve the accuracy of prognosis estimation for medulloblastoma patients, and an MRI-based automatic tumor segmentation model could reduce the cost of time.

KEY POINTS

• A model that combines radiomics and clinical features can predict the survival prognosis of patients with medulloblastoma. • Online nomogram and image automatic segmentation model can help doctors better judge the prognosis of medulloblastoma and save working time. • The developed AI system can help doctors judge the prognosis of diseases and promote the development of precision medicine.

Which model is better in predicting the survival of laryngeal squamous cell carcinoma?: Comparison of the random survival forest based on machine learning algorithms to Cox regression: analyses based on SEER database

Prediction of postoperative survival for laryngeal carcinoma patients is very important. This study attempts to demonstrate the utilization of the random survival forest (RSF) and Cox regression model to predict overall survival of laryngeal squamous cell carcinoma (LSCC) and compare their performance. A total of 8677 patients diagnosed with LSCC from 2004 to 2015 were obtained from surveillance, epidemiology, and end results database. Multivariate imputation by chained equations was applied to filling the missing data. Lasso regression algorithm was conducted to find potential predictors. RSF and Cox regression were used to develop the survival prediction models. Harrell's concordance index (C-index), area under the curve (AUC), Brier score, and calibration plot were used to evaluate the predictive performance of the 2 models. For 3-year survival prediction, the C-index in training set were 0.74 (0.011) and 0.84 (0.013) for Cox and RSF respectively. For 5-year survival prediction, the C-index in training set were 0.75 (0.022) and 0.80 (0.011) for Cox and RSF respectively. Similar results were found in validation set. The AUC were 0.795 for RSF and 0.715 for Cox in the training set while the AUC were 0.765 for RSF and 0.705 for Cox in the validation set. The prediction error curves for each model based on Brier score showed the RSF model had lower prediction errors both in training group and validation group. What's more, the calibration curve displayed similar results of 2 models both in training set and validation set. The performance of RSF model were better than Cox regression model. The RSF algorithms provide a relatively better alternatives to be of clinical use for estimating the survival probability of LSCC patients.

The clinical relevance of various methods of classifying ipsilateral breast tumour recurrence as either true local recurrence or new primary

PURPOSE

Describes the relevance of -various classification methods for ipsilateral breast tumour recurrence (IBTR) as either true recurrence (TR) or new primary (NP) on both disease-specific survival (DSS) and distant metastasis-free survival (DMFS).

METHOD

Two hundred and thirty-four of 4359 women undergoing breast-conserving therapy experienced IBTR. We compared the impact of four known classification methods and two newly created classification methods.

RESULTS

For three of the methods, a better DSS was observed for NP compared to TR with the hazard ratio (HR) ranging from 0.5 to 0.6. The new Twente method classification, comprising all classification criteria of three known methods, and the new Morphology method, using only morphological criteria, had the best HR and confidence interval with a HR 0.5 (95% CI 0.2-1.0) and a HR 0.5 (95% CI 0.3-1.1), respectively. For DMFS, the HR for NP compared to TR ranged from 0.6 to 0.9 for all six methods. The new Morphology method and the Twente method noted the best HR and confidence intervals with a HR 0.6 (95% CI 0.3-1.1) and a HR 0.6 (95% CI 0.4-1.2), respectively.

CONCLUSION

IBTR classified as TR or NP has a prognostic value for both DSS and DMFS, but depends on the classification method used. Developing and validating a generally accepted form of classification are imperative for using TR and NP in clinical practice.

Impact of time to first relapse on long-term outcome in adult retroperitoneal sarcoma patients after radical resection

BACKGROUND

Local recurrence of primary retroperitoneal sarcoma (RPS) is one of the major causes of treatment failure and death. We attempted to assess the effects of time to local recurrence (TLR) on the survival after recurrence (SAR) and overall survival (OS) of RPS.

METHODS

Included in this study were 224 patients who underwent R0 resection for primary RPS at our institution between January 2000 and December 2020, 118 of whom had local recurrence. Based on the median TLR (19.8 months), patients were divided into two groups: early local recurrence (ELR < 20 months) and late local recurrence (LLR > 20 months). The Kaplan-Meier method was employed to calculate the local recurrence-free survival (LRFS), SAR and OS. Univariate and multivariate analyses were conducted to explore the prognostic value of TLR.

RESULTS

The median follow-up time was 60.5 months for the entire cohort and 58.5 months for the recurrence cohort. There were 60 (50.8%) patients in the ELR group and 58 (49.2%) in the LLR group. The ELR group exhibited a worse SAR (29.2 months vs. 73.4 months, P < 0.001), OS (41.8 months vs. 120.9 months, P < 0.001), and a lower 5-year OS rate (35.9% vs. 73.2%, P = 0.004) than the LLR group. Furthermore, multivariate analysis indicated that TLR was an independent prognostic indicator for SAR (P = 0.014) and OS (P < 0.001).

CONCLUSIONS

In patients with RPS, ELR after R0 resection presents adverse effects on OS and SAR than those with LLR, and TLR could serve as a promising predictor for OS and SAR.

A pan-cancer analysis of GINS complex subunit 4 to identify its potential role as a biomarker in multiple human cancers

This study was initiated to explore the expression variation, clinical significance, and biological importance of the GINS complex subunit 4 (GINS4) in different human cancers as a shared biomarker via pan-cancer analysis through different platforms including UALCAN, Kaplan Meier (KM) plotter, TNMplot, GENT2, GEPIA, DriverDBv3, Human Protein Atlas (HPA), MEXPRESS, cBioportal, STRING, DAVID, MuTarge, Enrichr, TIMER, and CTD. Our findings have verified the up-regulation of GINS4 in 24 major subtypes of human cancers, and its overexpression was found to be substantially associated with poor overall survival (OS), relapse-free survival (RFs), and metastasis in ESCA, KIRC, LIHC, LUAD, and UCEC. This suggested that GINS4 plays a significant role in the development and progression of these five cancers. Furthermore, we noticed that GINS4 is also overexpressed in ESCA, KIRC, LIHC, LUAD, and UCEC patients with different clinicopathological characteristics. Enrichment analysis revealed the involvement of GINS4 associated genes in a variety of diverse GO and KEGG terms. We also explored few significant correlations between GINS4 expression and promoter methylation, genetic alterations, CNVs, other mutant genes, tumor purity, and immune cells infiltration. In conclusion, our results elucidated that GINS4 can serve as a shared diagnostic, prognostic biomarker, and a potential therapeutic target in ESCA, KIRC, LIHC, LUAD, and UCEC patients with different clinicopathological characteristics.

Fenchel duality of Cox partial likelihood with an application in survival kernel learning

The Cox proportional hazard model is one of the most widely used methods in modeling time-to-event data in the health sciences. Due to the simplicity of the Cox partial likelihood function, many machine learning algorithms use it for survival data. However, due to the nature of censored data, the optimization problem becomes intractable when more complicated regularization is employed, which is necessary when dealing with high dimensional omic data. In this paper, we show that a convex conjugate function of the Cox loss function based on Fenchel duality exists, and provide an alternative framework to optimization based on the primal form. Furthermore, the dual form suggests an efficient algorithm for solving the kernel learning problem with censored survival outcomes. We illustrate performance and properties of the derived duality form of Cox partial likelihood loss in multiple kernel learning problems with simulated and the Skin Cutaneous Melanoma TCGA datasets.

A Decision-Making Supporting Prediction Method for Breast Cancer Neoadjuvant Chemotherapy

Neoadjuvant chemotherapy (NAC) may increase the resection rate of breast cancer and shows promising effects on patient prognosis. It has become a necessary treatment choice and is widely used in the clinical setting. Benefitting from the clinical information obtained during NAC treatment, computational methods can improve decision-making by evaluating and predicting treatment responses using a multidisciplinary approach, as there are no uniformly accepted protocols for all institutions for adopting different treatment regiments. In this study, 166 Chinese breast cancer cases were collected from patients who received NAC treatment at the First Bethune Hospital of Jilin University. The Miller–Payne grading system was used to evaluate the treatment response. Four machine learning multiple classifiers were constructed to predict the treatment response against the 26 features extracted from the patients’ clinical data, including Random Forest (RF) model, Convolution Neural Network (CNN) model, Support Vector Machine (SVM) model, and Logistic Regression (LR) model, where the RF model achieved the best performance using our data. To allow a more general application, the models were reconstructed using only six selected features, and the RF model achieved the highest performance with 54.26% accuracy. This work can efficiently guide optimal treatment planning for breast cancer patients.

Ipsilateral Breast Cancer Recurrence: Characteristics, Treatment, and Long-Term Oncologic Results at a High-Volume Center

INTRODUCTION

Salvage mastectomy is considered the treatment of choice for ipsilateral breast cancer recurrence (IBCR), even if a second breast-conserving surgery (BCS) is feasible. The purpose of this study was to describe the characteristics of IBCR patients, to compare the 2 therapeutic options in terms of long-term outcomes, and to identify independent factors that may predict the type of treatment.

PATIENTS AND METHODS

A total of 309 IBCR patients who underwent either repeat BCS or mastectomy were identified. All the analyzed patients with IBCR had true recurrence.

RESULTS

Repeat BCS and salvage mastectomy were performed in 143 and 166 patients, respectively. Age < 65 years (59.6% vs 37.1% if age ≥ 65 years; odds ratio, 2.374; 95% confidence interval, 0.92-5.24; P = .018) and disease-free interval < 24 months (15.7% vs 10.5% if disease-free interval ≥ 24 months; odds ratio, 2.705; 95% confidence interval, 1.42-5.97; P = .007) were found to significantly increase the probability of receipt of mastectomy. Disease-free survival rates at 3, 5, and 10 years were 79.2%, 68.2%, and 36.9%; and 77.2%, 65.9%, and 55.3% in patients receiving repeat BCS or mastectomy, respectively (P = .842). Overall survival rates at 3, 5, and 10 years were 95.4%, 91.4%, and 68.5%; and 87.3%, 69.3%, and 57.9%, respectively, in patients receiving repeat BCS or mastectomy (P = .018).

CONCLUSION

Salvage mastectomy should not be considered the only treatment option for IBCR. A second BCS can still be evaluated and proposed to IBCR patients, with acceptable locoregional control and survival. The risk of poor long-term prognosis after mastectomy should be shared with the patient.

Immune-related gene data-based molecular subtyping related to the prognosis of breast cancer patients

Background

Breast cancer (BC), which is the most common malignant tumor in females, is associated with increasing morbidity and mortality. Effective treatments include surgery, chemotherapy, radiotherapy, endocrinotherapy and molecular-targeted therapy. With the development of molecular biology, immunology and pharmacogenomics, an increasing amount of evidence has shown that the infiltration of immune cells into the tumor microenvironment, coupled with the immune phenotype of tumor cells, will significantly affect tumor development and malignancy. Consequently, immunotherapy has become a promising treatment for BC prevention and as a modality that can influence patient prognosis.

Methods

In this study, samples collected from The Cancer Genome Atlas (TCGA) and ImmPort databases were analyzed to investigate specific immune-related genes that affect the prognosis of BC patients. In all, 64 immune-related genes related to prognosis were screened, and the 17 most representative genes were finally selected to establish the prognostic prediction model of BC (the RiskScore model) using the Lasso and StepAIC methods. By establishing a training set and a test set, the efficiency, accuracy and stability of the model in predicting and classifying the prognosis of patients were evaluated. Finally, the 17 immune-related genes were functionally annotated, and GO and KEGG signal pathway enrichment analyses were performed.

Results

We found that these 17 genes were enriched in numerous BC- and immune microenvironment-related pathways. The relationship between the RiskScore and the clinical characteristics of the sample and signaling pathways was also analyzed.

Conclusions

Our findings indicate that the prognostic prediction model based on the expression profiles of 17 immune-related genes has demonstrated high predictive accuracy and stability in identifying immune features, which can guide clinicians in the diagnosis and prognostic prediction of BC patients with different immunophenotypes.

Electronic supplementary material

The online version of this article (10.1007/s12282-020-01191-z) contains supplementary material, which is available to authorized users.