Radiomics-Based Machine Learning Model for Predicting Overall and Progression-Free Survival in Rare Cancer: A Case Study for Primary CNS Lymphoma Patients

Automated Lesion and Feature Extraction Pipeline for Brain MRIs with Interpretability

This paper introduces the Automated Lesion and Feature Extraction (ALFE) pipeline, an open-source, Python-based pipeline that consumes MR images of the brain and produces anatomical segmentations, lesion segmentations, and human-interpretable imaging features describing the lesions in the brain. ALFE pipeline is modeled after the neuroradiology workflow and generates features that can be used by physicians for quantitative analysis of clinical brain MRIs and for machine learning applications. The pipeline uses a decoupled design which allows the user to customize the image processing, image registrations, and AI segmentation tools without the need to change the business logic of the pipeline. In this manuscript, we give an overview of ALFE, present the main aspects of ALFE pipeline design philosophy, and present case studies.

MRI-based radiomics virtual biopsy for BCL6 in primary central nervous system lymphoma

AIM

To establish a machine learning model based on a radiomic signature for predicting B-cell lymphoma 6 (BCL-6) rearrangement in primary central nervous system lymphoma (PCNSL).

MATERIALS AND METHODS

Retrospective study on 102 PCNSL patients (31 with BCL-6 rearrangement positive, 71 with BCL-6 rearrangement negative) were randomly divided into the training and validation sets at a ratio of 7:3. Radiomics models based on contrast-enhanced T1-weighted imaging (CE-T1WI) and fluid-attenuated inversion recovery (FLAIR) in different regions, including VOItumour core and VOIperitumoural oedema. Radiomics features were extracted and selected using LASSO regression, and radiomics score (rad-score) were calculated using the weighted coefficients. Four machine learning models (logistic regression, random forest, support vector machine, K-nearest neighbours) were developed and evaluated based on rad-score. The optimal radiomics model was integrated into the clinical or radiological factors to construct a predictive model through logistic regression analysis. A nomogram was constructed based on independent significant features for individualised prediction.

RESULTS

All rad-scores based on CE-T1WI and FLAIR sequences were significantly associated with BCL6 rearrangement (p < 0.05) in univariate regression analysis. The logistic regression machine learning model performed best with AUCs of 0.935 (training) and 0.923 (validation). Rad-scores from CE-T1WI tumour core and peritumoural oedema were independent significant predictors.

CONCLUSION

Radiomics signatures based on CE-T1WI and FLAIR sequences have significant value in distinguishing BCL6 rearrangement. The CE-T1WI radiomics model based on VOItumour core and VOIperitumoural oedema are robust markers for identifying BCL6 rearrangement.

Optimizing radiomics for prostate cancer diagnosis: feature selection strategies, machine learning classifiers, and MRI sequences

OBJECTIVES

Radiomics-based analyses encompass multiple steps, leading to ambiguity regarding the optimal approaches for enhancing model performance. This study compares the effect of several feature selection methods, machine learning (ML) classifiers, and sources of radiomic features, on models' performance for the diagnosis of clinically significant prostate cancer (csPCa) from bi-parametric MRI.

METHODS

Two multi-centric datasets, with 465 and 204 patients each, were used to extract 1246 radiomic features per patient and MRI sequence. Ten feature selection methods, such as Boruta, mRMRe, ReliefF, recursive feature elimination (RFE), random forest (RF) variable importance, L1-lasso, etc., four ML classifiers, namely SVM, RF, LASSO, and boosted generalized linear model (GLM), and three sets of radiomics features, derived from T2w images, ADC maps, and their combination, were used to develop predictive models of csPCa. Their performance was evaluated in a nested cross-validation and externally, using seven performance metrics.

RESULTS

In total, 480 models were developed. In nested cross-validation, the best model combined Boruta with Boosted GLM (AUC = 0.71, F1 = 0.76). In external validation, the best model combined L1-lasso with boosted GLM (AUC = 0.71, F1 = 0.47). Overall, Boruta, RFE, L1-lasso, and RF variable importance were the top-performing feature selection methods, while the choice of ML classifier didn't significantly affect the results. The ADC-derived features showed the highest discriminatory power with T2w-derived features being less informative, while their combination did not lead to improved performance.

CONCLUSION

The choice of feature selection method and the source of radiomic features have a profound effect on the models' performance for csPCa diagnosis.

CRITICAL RELEVANCE STATEMENT

This work may guide future radiomic research, paving the way for the development of more effective and reliable radiomic models; not only for advancing prostate cancer diagnostic strategies, but also for informing broader applications of radiomics in different medical contexts.

KEY POINTS

Radiomics is a growing field that can still be optimized. Feature selection method impacts radiomics models' performance more than ML algorithms. Best feature selection methods: RFE, LASSO, RF, and Boruta. ADC-derived radiomic features yield more robust models compared to T2w-derived radiomic features.

MRI-Based Machine Learning for Prediction of Clinical Outcomes in Primary Central Nervous System Lymphoma

A portion of individuals diagnosed with primary central nervous system lymphomas (PCNSL) may experience early relapse or refractory (R/R) disease following treatment. This research explored the potential of MRI-based radiomics in forecasting R/R cases in PCNSL. Forty-six patients with pathologically confirmed PCNSL diagnosed between January 2008 and December 2020 were included in this study. Only patients who underwent pretreatment brain MRIs and complete postoperative follow-up MRIs were included. Pretreatment contrast-enhanced T1WI, T2WI, and T2 FLAIR imaging were analyzed. A total of 107 radiomic features, including 14 shape-based, 18 first-order statistical, and 75 texture features, were extracted from each sequence. Predictive models were then built using five different machine learning algorithms to predict R/R in PCNSL. Of the included 46 PCNSL patients, 20 (20/46, 43.5%) patients were found to have R/R. In the R/R group, the median scores in predictive models such as support vector machine, k-nearest neighbors, linear discriminant analysis, naïve Bayes, and decision trees were significantly higher, while the apparent diffusion coefficient values were notably lower compared to those without R/R (p < 0.05). The support vector machine model exhibited the highest performance, achieving an overall prediction accuracy of 83%, a precision rate of 80%, and an AUC of 0.78. Additionally, when analyzing tumor progression, patients with elevated support vector machine and naïve Bayes scores demonstrated a significantly reduced progression-free survival (p < 0.05). These findings suggest that preoperative MRI-based radiomics may provide critical insights for treatment strategies in PCNSL.

Radiomic prediction for durable response to high-dose methotrexate-based chemotherapy in primary central nervous system lymphoma

BACKGROUND

The rarity of primary central nervous system lymphoma (PCNSL) and treatment heterogeneity contributes to a lack of prognostic models for evaluating posttreatment remission. This study aimed to develop and validate radiomic-based models to predict the durable response (DR) to high-dose methotrexate (HD-MTX)-based chemotherapy in PCNSL patients.

METHODS

A total of 159 patients pathologically diagnosed with PCNSL between 2011 and 2021 across two institutions were enrolled. According to the NCCN guidelines, the DR was defined as the remission lasting ≥1 year after receiving HD-MTX-based chemotherapy. For each patient, a total of 1218 radiomic features were extracted from prebiopsy T1 contrast-enhanced MR images. Multiple machine-learning algorithms were utilized for feature selection and classification to build a radiomic signature. The radiomic-clinical integrated models were developed using the random forest method. Model performance was externally validated to verify its clinical utility.

RESULTS

A total of 105 PCNSL patients were enrolled after excluding 54 cases with ineligibility. The training and validation cohorts comprised 76 and 29 individuals, respectively. Among them, 65 patients achieved DR. The radiomic signature, consisting of 8 selected features, demonstrated strong predictive performance, with area under the curves of 0.994 in training cohort and 0.913 in validation cohort. This signature was independently associated with the DR in both cohorts. Both the radiomic signature and integrated models significantly outperformed the clinical models in two cohorts. Decision curve analysis underscored the clinical utility of the established models.

CONCLUSIONS

This radiomic signature and integrated models have the potential to accurately predict the DR to HD-MTX-based chemotherapy in PCNSL patients, providing valuable therapeutic insights.

MRI-based radiomics signatures for preoperative prediction of Ki-67 index in primary central nervous system lymphoma

PURPOSE

The aim of this study was to develop and validate radiomics signatures based on MRI for preoperative prediction of Ki-67 proliferative index (PI) expression in primary central nervous system lymphoma (PCNSL).

METHODS

A total of 341 patients with PCNSL were retrospectively analyzed, including 286 patients in one center as the training set and 55 patients in another two centers as the external validation set. Radiomics features were extracted and selected from preoperative contrast-enhanced T1-weighted images, fluid attenuation inversion recovery to build radiomics signatures according to the Ki-67 PI. The predictive performances of the radiomics model were evaluated using four classifiers including random forest, K-Nearest Neighbors, Neural Network and Decision Tree. A combined model was built by incorporating radiomics signature, clinical variables and MRI radiological characteristics using multivariate logistic regression analysis, and a nomogram was established to predict the expression of Ki-67 individually. The predictive performances of the models were evaluated using area under receiver operating characteristic curve (AUC) and decision curve analysis (DCA).

RESULTS

Radiomics signatures were independent predictors of the expression level of Ki-67 (OR: 2.523, P < 0.001). RF radiomics models had the highest accuracy (0.934 in the training set and 0.811 in the external validation set) and F1 Score (0.920 in the training set and 0.836 in the external validation set). The clinic-radiologic-radiomics nomogram showed better predictive performance with AUCs of 0.877(95 % CI: 0.837-0.918) in the training set and 0.866(95 % CI: 0.774-0.957) in the external validation set. The calibration curve and DCA demonstrated goodness-of-fit and improved benefits in clinical practice of the nomogram.

CONCLUSIONS

Nomograms integrating MRI-based radiomics and clinical-radiological characteristics could effectively predict Ki-67 PI in primary PCNSL.

Applications of Multimodal Artificial Intelligence in Non-Hodgkin Lymphoma B Cells

Given advancements in large-scale data and AI, integrating multimodal artificial intelligence into cancer research can enhance our understanding of tumor behavior by simultaneously processing diverse biomedical data types. In this review, we explore the potential of multimodal AI in comprehending B-cell non-Hodgkin lymphomas (B-NHLs). B-cell non-Hodgkin lymphomas (B-NHLs) represent a particular challenge in oncology due to tumor heterogeneity and the intricate ecosystem in which tumors develop. These complexities complicate diagnosis, prognosis, and therapy response, emphasizing the need to use sophisticated approaches to enhance personalized treatment strategies for better patient outcomes. Therefore, multimodal AI can be leveraged to synthesize critical information from available biomedical data such as clinical record, imaging, pathology and omics data, to picture the whole tumor. In this review, we first define various types of modalities, multimodal AI frameworks, and several applications in precision medicine. Then, we provide several examples of its usage in B-NHLs, for analyzing the complexity of the ecosystem, identifying immune biomarkers, optimizing therapy strategy, and its clinical applications. Lastly, we address the limitations and future directions of multimodal AI, highlighting the need to overcome these challenges for better clinical practice and application in healthcare.

Prognostic Assessment in Patients With Primary Diffuse Large B-Cell Lymphoma of the Central Nervous System Using MRI-Based Radiomics

BACKGROUND

Primary central nervous system lymphoma (PCNSL) carries a poor prognosis. Radiomics may hold potential value in prognostic assessment.

PURPOSE

To develop and validate an MRI-based radiomics model and combine it with clinical factors to assess progression-free survival (PFS) and overall survival (OS) of patients with PCNSL.

STUDY TYPE

Retrospective and prospective.

POPULATION

Three hundred seventy-nine patients (179 female, 53 ± 7 years) from 2014 to 2022.

FIELD STRENGTH/SEQUENCE

T2/fluid-attenuated inversion recovery, contrast-enhanced T1WI and diffusion-weighted echo-planar imaging sequences on 3.0 T.

ASSESSMENT

Radiomics features were extracted from enhanced tumor regions on preoperative multi-sequence MRI. Using a least absolute shrinkage and selection operator (LASSO) Cox regression model to select radiomic signatures in training cohort (N = 169). Cox proportional hazards models were constructed for clinical, radiomics, and combined models, with internal (N = 72) and external (N = 32) cohorts validating model performance.

STATISTICAL TESTS

Chi-squared, Mann-Whitney, Kaplan-Meier, log-rank, LASSO, Cox, decision curve analysis, time-dependent Receiver Operating Characteristic, area under the curve (AUC), and likelihood ratio test. P-value <0.05 was considered significant.

RESULTS

Follow-up duration was 28.79 ± 22.59 months (median: 25). High-risk patients, determined by the median radiomics score, showed significantly lower survival rates than low-risk patients. Compared with NCCN-IPI, conventional imaging and clinical models, the combined model achieved the highest C-index for both PFS (0.660 internal, 0.802 external) and OS (0.733 internal, 0.781 external) in validation. Net benefit was greater with radiomics than with clinical alone. The combined model exhibited performance with AUCs of 0.680, 0.752, and 0.830 for predicting 1-year, 3-year, and 5-year PFS, and 0.770, 0.789, and 0.863 for OS in internal validation, with PFS AUCs of 0.860 and 0.826 and OS AUCs of 0.859 and 0.748 for 1-year and 3-year survival in external validation.

DATA CONCLUSION

Incorporating a multi-sequence MR-based radiomics model into clinical models enhances the assess accuracy for the prognosis of PCNSL.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 2.

Deep representation learning of tissue metabolome and computed tomography annotates NSCLC classification and prognosis

The rich chemical information from tissue metabolomics provides a powerful means to elaborate tissue physiology or tumor characteristics at cellular and tumor microenvironment levels. However, the process of obtaining such information requires invasive biopsies, is costly, and can delay clinical patient management. Conversely, computed tomography (CT) is a clinical standard of care but does not intuitively harbor histological or prognostic information. Furthermore, the ability to embed metabolome information into CT to subsequently use the learned representation for classification or prognosis has yet to be described. This study develops a deep learning-based framework -- tissue-metabolomic-radiomic-CT (TMR-CT) by combining 48 paired CT images and tumor/normal tissue metabolite intensities to generate ten image embeddings to infer metabolite-derived representation from CT alone. In clinical NSCLC settings, we ascertain whether TMR-CT results in an enhanced feature generation model solving histology classification/prognosis tasks in an unseen international CT dataset of 742 patients. TMR-CT non-invasively determines histological classes - adenocarcinoma/squamous cell carcinoma with an F1-score = 0.78 and further asserts patients' prognosis with a c-index = 0.72, surpassing the performance of radiomics models and deep learning on single modality CT feature extraction. Additionally, our work shows the potential to generate informative biology-inspired CT-led features to explore connections between hard-to-obtain tissue metabolic profiles and routine lesion-derived image data.

New hopes in relapsed refractory primary central nervous system lymphoma

PURPOSE OF REVIEW

Patients with relapsed/refractory primary central nervous system lymphoma (rrPCNSL) have poor prognosis, with a median survival after relapse of 6.8 months. In this review, we discuss the evolving landscape and the possible future directions related to this important unmet clinical need.

RECENT FINDINGS

The modern two-phase approach for newly diagnosed PCNSL based on an induction using high-dose methotrexate (HD-MTX) combinations and a subsequent consolidation, has significantly improved the outcome in this setting. However, this strategy is able to cure more or less 50% of patients. rrPCNSL patients have a very poor prognosis with a reported 5-year overall survival of 18%. Late relapses (after third year) and use of high-dose chemotherapy and autologous stem cell transplantation (HDT-ASCT) represent important factors associated with a better outcome in this setting. On the basis of the growing acquisition of knowledge on the molecular characteristics of PCNSL, the use of non-chemotherapeutic drugs such as bruton tyrosine kinase inhibitors (BTK-is), immunomodulatory drugs (IMiDs) and immune checkpoint blockers (ICBs) is increasing in the last years along with the introduction of novel approaches (CAR-T cells and blood--brain barrier disruption). However, despite high responses in some cases, durations are often short, translating in outcome results still unsatisfactory.

SUMMARY

Treatment of rrPCNSL patients is challenging. As no standard of care exist in this setting, it is of paramount importance to acquire new knowledge related to this condition and start multidisciplinary collaboration in order to improve pts outcome.

Early MRI Predictors of Relapse in Primary Central Nervous System Lymphoma Treated with MATRix Immunochemotherapy

Primary Central Nervous System Lymphoma (PCNSL) is a highly malignant brain tumour. We investigated dynamic changes in tumour volume and apparent diffusion coefficient (ADC) measurements for predicting outcome following treatment with MATRix chemotherapy in PCNSL. Patients treated with MATRix (n = 38) underwent T1 contrast-enhanced (T1CE) and diffusion-weighted imaging (DWI) before treatment, after two cycles and after four cycles of chemotherapy. Response was assessed using the International PCNSL Collaborative Group (IPCG) imaging criteria. ADC histogram parameters and T1CE tumour volumes were compared among response groups, using one-way ANOVA testing. Logistic regression was performed to examine those imaging parameters predictive of response. Response after two cycles of chemotherapy differed from response after four cycles; of the six patients with progressive disease (PD) after four cycles of treatment, two (33%) had demonstrated a partial response (PR) or complete response (CR) after two cycles. ADC_mean at baseline, T1CE at baseline and T1CE percentage volume change differed between response groups (0.005 < p < 0.038) and were predictive of MATRix treatment response (area under the curve: 0.672-0.854). Baseline ADC and T1CE metrics are potential biomarkers for risk stratification of PCNSL patients early during remission induction therapy with MATRix. Standard interim response assessment (after two cycles) according to IPCG imaging criteria does not reliably predict early disease progression in the context of a conventional treatment approach.

Radiomic features define risk and are linked to DNA methylation attributes in primary CNS lymphoma

Background

The prognostic roles of clinical and laboratory markers have been exploited to model risk in patients with primary CNS lymphoma, but these approaches do not fully explain the observed variation in outcome. To date, neuroimaging or molecular information is not used. The aim of this study was to determine the utility of radiomic features to capture clinically relevant phenotypes, and to link those to molecular profiles for enhanced risk stratification.

Methods

In this retrospective study, we investigated 133 patients across 9 sites in Austria (2005-2018) and an external validation site in South Korea (44 patients, 2013-2016). We used T1-weighted contrast-enhanced MRI and an L1-norm regularized Cox proportional hazard model to derive a radiomic risk score. We integrated radiomic features with DNA methylation profiles using machine learning-based prediction, and validated the most relevant biological associations in tissues and cell lines.

Results

The radiomic risk score, consisting of 20 mostly textural features, was a strong and independent predictor of survival (multivariate hazard ratio = 6.56 [3.64-11.81]) that remained valid in the external validation cohort. Radiomic features captured gene regulatory differences such as in BCL6 binding activity, which was put forth as testable treatment target for a subset of patients.

Conclusions

The radiomic risk score was a robust and complementary predictor of survival and reflected characteristics in underlying DNA methylation patterns. Leveraging imaging phenotypes to assess risk and inform epigenetic treatment targets provides a concept on which to advance prognostic modeling and precision therapy for this aggressive cancer.