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.

OBJECTIVE: Research into the effectiveness and applicability of deep learning, radiomics, and their integrated models based on Magnetic Resonance Imaging (MRI) for preoperative differentiation between Primary Central Nervous System Lymphoma (PCNSL) and Glioblastoma (GBM), along with an exploration of the interpretability of these models.
METHODS: A retrospective analysis was performed on MRI images and clinical data from 261 patients across two medical centers. The data were split into a training set (n = 153, medical center 1) and an external test set (n = 108, medical center 2). Radiomic features were extracted using Pyradiomics to build the Radiomics Model. Deep learning networks, including the transformer-based MobileVIT Model and Convolutional Neural Networks (CNN) based ConvNeXt Model, were trained separately. By applying the \"late fusion\" theory, the radiomics model and deep learning model were fused to produce the optimal Max-Fusion Model. Additionally, Shapley Additive exPlanations (SHAP) and Grad-CAM were employed for interpretability analysis.
RESULTS: In the external test set, the Radiomics Model achieved an Area under the receiver operating characteristic curve (AUC) of 0.86, the MobileVIT Model had an AUC of 0.91, the ConvNeXt Model demonstrated an AUC of 0.89, and the Max-Fusion Model showed an AUC of 0.92. The Delong test revealed a significant difference in AUC between the Max-Fusion Model and the Radiomics Model (P = 0.02).
CONCLUSIONS: The Max-Fusion Model, combining different models, presents superior performance in distinguishing PCNSL and GBM, highlighting the effectiveness of model fusion for enhanced decision-making in medical applications.
CONCLUSIONS: The preoperative non-invasive differentiation between PCNSL and GBM assists clinicians in selecting appropriate treatment regimens and clinical management strategies.

BACKGROUND: This study aimed to investigate what treatment are selected for malignant brain tumors, particularly glioblastoma (GBM) and primary central nervous system lymphoma (PCNSL), in real-world Japan and the costs involved.
METHODS: We conducted a questionnaire survey regarding treatment selections for newly diagnosed GBM and PCNSL treated between July 2021 and June 2022 among 47 institutions in the Japan Clinical Oncology Group-Brain Tumor Study Group. We calculated the total cost and cost per month of the initial therapy for newly diagnosed GBM or PCNSL.
RESULTS: The most used regimen (46.8%) for GBM in patients aged ≤74 years was \'Surgery + radiotherapy concomitant with temozolomide\'. This regimen\'s total cost was 7.50 million JPY (Japanese yen). Adding carmustine wafer implantation (used in 15.0%), TTFields (used in 14.1%), and bevacizumab (BEV) (used in 14.5%) to the standard treatment of GBM increased the cost by 1.24 million JPY for initial treatment, and 1.44 and 0.22 million JPY per month, respectively. Regarding PCNSL, \'Surgery (biopsy) + rituximab, methotrexate, procarbazine, and vincristine (R-MPV) therapy\' was the most used regimen (42.5%) for patients of all ages. This regimen incurred 1.07 million JPY per month. The three PCNSL regimens based on R-MPV therapy were in ultra-high-cost medical care (exceeding 1 million JPY per month).
CONCLUSIONS: Treatment of malignant brain tumors is generally expensive, and cost-ineffective treatments such as BEV are frequently used. We believe that the results of this study can be used to design future economic health studies examining the cost-effectiveness of malignant brain tumors.

Double-hit lymphoma (DHL) formerly referred to high-grade B-cell lymphoma with concurrent MYC and BCL2 or BCL6 rearrangements, however, the updated 2022 World Health Organization Classification (5th edition online) excludes those with MYC and BCL 6 rearrangements from the high-grade category. DHL confined to the central nervous system (CNS), known as double-hit primary CNS lymphoma (DH-PCNSL), is rare with poorly understood clinical features. Here, we report a case of a 64-year-old man with multiple brain tumors diagnosed with DH-PCNSL who showed bone marrow (BM) infiltration early in the clinical course. The histological diagnosis was high-grade B-cell lymphoma with MYC and BCL6 rearrangements. Fluorodeoxyglucose positron emission tomography (FDG-PET) revealed no abnormal accumulation except in the CNS. The patient received whole-brain radiotherapy following the failure of high-dose methotrexate. After completion of radiotherapy, the patient developed thrombocytopenia, and BM biopsy showed infiltration of DHL cells, which were not detected by repeated FDG-PET. This is the first report of DH-PCNSL where identical gene rearrangements were confirmed in both the resected CNS tumor and BM tissue. Patients with DH-PCNSL require careful follow-up because they may be at a potential risk of BM infiltration, which may be undetectable by FDG-PET, particularly early in the disease course.

Background: The incidence of lymphomatous involvement of the central nervous system (CNS) has been increasing in recent years. However, the rarity of the disease has resulted in a scarcity of available data regarding its clinical presentation, natural history, and prognosis. We aimed to investigate the neurological characteristics of uncommon lymphomatous involvements confined to the CNS and to identify key variables that could serve as predictive biomarkers for treatment outcomes. Methods: We identified patients presenting with neurological symptoms and diagnosed with CNS-restricted lymphomatous involvement between 2005 and 2023. Results: We identified 44 cases, 93% of which were diagnosed with primary central nervous system lymphoma (PCNSL) and 7% with intravascular lymphoma. The median time from symptom onset to diagnosis was 47 days (range: 6-573 days), with no statistically significant difference between patients older and younger than 60 years (p = 0.22). The median follow-up time was 1144 days (range: 27-3501 days). Cognitive deterioration was the most common presenting symptom, occurring in 19 out of 44 patients (43%). Brain MRI revealed that lobar lesions were the most frequent location of lesions, found in 24 out of 44 patients (55%). By the end of the study period, 30 patients (68%) had died, with a median survival of 666 days (range: 17-3291 days). Death was significantly more common in patients who experienced relapses (p = 0.04; 95% CI: 0.99-0.03), with these patients having a four times higher chance of death (HR = 4.1; 95% CI: 1.01-16.09). The time to diagnosis significantly correlated with survival (p = 0.02; 95% CI: 0.005-0.54), as did the Eastern Cooperative Oncology Group (ECOG) performance status at the last follow-up (p = 0.006; 95% CI: 0.0012-0.62). Patients aged over 60 years did not exhibit a higher likelihood of death (p = 0.19; HR = 2.3; 95% CI: 0.63-8.61); however, the threshold age at diagnosis for the maximally predicted mortality was 64 years (ROC = 0.73; p = 0.03). Conclusions: Patients had significant delays in diagnosis, affecting patient outcomes. Cognitive deterioration and lobar lesions were prominent clinical and radiological features. Mortality was notably higher in patients with relapses and those who had a longer time to diagnosis.

Histopathologic examinations of primary central nervous system lymphoma (PCNSL) reveal concentric accumulation of lymphocytes in the perivascular area with fibrosis. However, the nature of this fibrosis in \"stiff\" PCNSL remains unclear. We have encountered some PCNSLs with hard masses as surgical findings. This study investigated the dense fibrous status and tumor microenvironment of PCNSLs with or without stiffness. We evaluated by silver-impregnation nine PCNSLs with stiffness and 26 PCNSLs without stiffness. Six of the nine stiff PCNSLs showed pathological features of prominent fibrosis characterized by aggregation of reticulin fibers, and collagen accumulations. Alpha-smooth muscle actin (αSMA)-positive spindle cells as a cancer-associated fibroblast, the populations of T lymphocytes, and macrophages were compared between fibrous and control PCNSLs. Fibrous PCNSLs included abundant αSMA-positive cells in both intra- and extra-tumor environments (5/6, 87% and 3/6, 50%, respectively). Conversely, only one out of the seven control PCNSL contained αSMA-positive cells in the intra-tumoral area. Furthermore, the presence of extra-tumoral αSMA-positive cells was associated with infiltration of T lymphocytes and macrophages. In conclusion, recognizing the presence of dense fibrosis in PCNSL can provide insights into the tumor microenvironment. These results may help stratify patients with PCNSL and improve immunotherapies for these patients.

Primary central nervous system lymphoma (PCNSL) is a rare neoplasm that can affect the brain, eyes, and, rarely, the spinal cord. Clinical presentation and MRI findings can mimic a variety of diseases, including high-grade gliomas, infectious and granulomatous diseases, and demyelinating diseases. We describe three cases where the diagnosis of PCNSL was difficult due to an ambiguous clinical, radiological and laboratory results. The role of stereotactic biopsy remains leading in differential diagnosis; however, the invasiveness and frequent limitations of this method determine the search for additional biological markers of the disease. New evidence suggests a potential role for cerebrospinal fluid (CSF) cytokine profiles and proteomic analysis in differential diagnosis, disease progression, and treatment response.
Первичная лимфома ЦНС (ПЛЦНС) — редкое новообразование, которое может поражать головной мозг, глаза и, реже, спинной мозг. Клиническая картина и данные МРТ могут имитировать различные заболевания, включая глиомы высокой степени злокачественности, инфекционные и гранулематозные, а также демилинизирующие заболевания. В статье представлены три клинических случая, где постановка диагноза ПЛЦНС была затруднительной в связи с неоднозначной клинической, радиологической и лабораторной картиной. Роль стереотаксической биопсии остается ведущей в проведении дифференциальной диагностики, однако инвазивность и нередко ограниченная доступность этого метода обусловливают поиск дополнительных биологических маркеров заболевания. Новые данные свидетельствуют о потенциальной роли маркеров изменения цитокинового профиля цереброспинальной жидкости, а также протеомного анализа в проведении дифференциальной диагностики, прогнозировании течения заболевания и ответа на лечение.

UNASSIGNED: To investigate the prognostic significance of pan-immune-inflammation value (PIV) and PILE score (based on PIV, lactate dehydrogenase (LDH), and Eastern Cooperative Oncology Group Performance Status (ECOG PS)) in patients with primary central nervous system lymphoma (PCNSL).
UNASSIGNED: A total of 109 patients were enrolled. PIV was calculated as follows: (neutrophil count × platelet count × monocyte count)/lymphocyte count. The PILE score was incorporated based on PIV, LDH levels, and ECOG PS. The Kaplan-Meier curves and Cox hazards regression models were applied for survival analyses. The relationship between PIV, PILE, and therapeutic response was examined.
UNASSIGNED: Baseline high PIV was significantly associated with worse overall survival (OS) in univariate (HR 3.990, 95% CI 1.778-8.954, p < 0.001) and multivariate (HR 3.047, 95% CI 1.175-7.897, p = 0.022) analyses. High PIV was also associated with worse progression-free survival (PFS) in univariate (HR 2.121, 95% CI 1.075-4.186, p = 0.030) but not significant in multivariate analyses. PIV outperformed other systemic inflammation parameters. The patients in the high PILE group (PILE score 2-3) had worse OS (p = 0.008) and PFS (p < 0.001) compared to the low PILE group (PILE score 0-1). PILE was independently associated with therapeutic response to initial treatment (OR 0.17, 95% CI 0.05-0.46; p < 0.001).
UNASSIGNED: High PIV and PILE were correlated with worse clinical outcomes in PCNSL patients, indicating that PIV and PILE might be a powerful predictor of prognosis and a potential predictive indicator for therapeutic response in PCNSL.

Primary Central Nervous System Lymphoma (PCNSL) is an aggressive brain tumour with a median survival rarely exceeding 3 months without treatment when seen in association with advanced HIV. High dose methotrexate (HD-MTX) in association with combination antiretroviral therapy (cART) is the recommended chemotherapy. However, HD-MTX may be not feasible due to poor performance status and concerns about toxicity. The 2023 guidelines from the European Association of Neuro-Oncology recommend that for people living with HIV (PLWH) presenting with PCNSL who have morbidities and/or poor functional status precluding the safe use of HD-MTX, other agents with a more favorable toxicity profile such as temozolomide might be considered. However, reports of temozolomide use for PLWH presenting PCNSL are exceedingly rare and this recommendation is extrapolated from its use in immunocompetent patients. We report here an elderly man living with HIV, with PCNSL and poor performance status who achieved long lasting remission with temozolomide plus cART. Our case illustrates the potential effectiveness of temozolomide in association with cART as first line treatment for PCNSL in a patient with poor functional status.

Cerebrospinal fluid (CSF) cytology of primary central nervous system lymphoma arising in the immune deficiency/dysregulation setting (IDD-PCNSL) has not been described. This study presented a case of IDD-PCNSL-DLBCL, a GCB phenotype who was successfully diagnosed by CSF cytology in conjunction with ICC, ISH, FCM and clinical information.