Primary central nervous system lymphomas: EHA–ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up

•This EHA–ESMO Clinical Practice Guideline provides key recommendations for managing primary DLBCL of the CNS.•The guideline covers clinical, imaging and pathological diagnosis, staging and risk assessment, treatment and follow-up.•Algorithms for first-line and salvage treatments are provided.•The author group encompasses a multidisciplinary group of experts from different institutions and countries in Europe.•Recommendations are based on available scientific data and the authors' collective expert opinion. Primary diffuse large B-cell lymphoma (DLBCL) of the central nervous system (CNS), termed primary CNS lymphoma (PCNSL), is an aggressive neoplasm presenting with disease limited to the CNS. PCNSL was recognised as a distinct entity by the 2017 World Health Organization (WHO) Classification of Tumours of Haematopoietic and Lymphoid Tissues.1Kluin P.M. Deckert M. Ferry J.A. Primary diffuse large B-cell lymphoma of the CNS.in: Swerdlow S.H. Campo E. Harris N.L. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. IARC, Lyon, France2017: 300-302Google Scholar In the 2022 edition of the WHO classification,2Alaggio R. Amador C. Anagnostopoulos I. et al.The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: lymphoid neoplasms.Leukemia. 2022; 36: 1720-1748Crossref PubMed Scopus (1099) Google Scholar this neoplasm is classified in the 'Large B-cell lymphomas of immune-privileged sites' group, whereas it is considered a specific entity in the International Consensus Classification of Mature Lymphoid Neoplasms.3Campo E. Jaffe E.S. Cook J.R. et al.The International Consensus Classification of Mature Lymphoid Neoplasms: a report from the Clinical Advisory Committee.Blood. 2022; 140: 1229-1253Crossref PubMed Scopus (528) Google Scholar This entity is also recognised by the WHO classification of CNS tumours.4Deckert M. Ferry J.A. Primary diffuse large B-cell lymphoma of the CNS.in: WHO Classification of Tumours Editorial Board, ed. WHO Classification of Tumours: Central Nervous System Tumours. 5th ed. IARC, Lyon, France2021: 351-355Google Scholar PCNSL can arise in both immunocompetent individuals and in those who are immunosuppressed (e.g. individuals living with human immunodeficiency virus and patients receiving immunosuppressive therapies following organ transplant). While no clear predisposing factors have been recognised in immunocompetent individuals, the nature, intensity and duration of immune suppression can influence the risk of PCNSL in those who are immunocompromised.5Engels E.A. Biggar R.J. Hall H.I. et al.Cancer risk in people infected with human immunodeficiency virus in the United States.Int J Cancer. 2008; 123: 187-194Crossref PubMed Scopus (638) Google Scholar This European Hematology Association (EHA)–ESMO Clinical Practice Guideline (CPG) includes recommendations for the management of immunocompetent patients with PCNSL. In this population, PCNSL accounts for 2% of all primary CNS tumours and 4%-6% of extranodal lymphomas, with an incidence of 0.47/100 000 person-years.6Villano J.L. Koshy M. Shaikh H. et al.Age, gender, and racial differences in incidence and survival in primary CNS lymphoma.Br J Cancer. 2011; 105: 1414-1418Crossref PubMed Scopus (389) Google Scholar PCNSL is typically diagnosed in the sixth or seventh decade of life, with a median age at diagnosis of 68 years and a slightly higher frequency in males.7Houillier C. Soussain C. Ghesquieres H. et al.Management and outcome of primary CNS lymphoma in the modern era: an LOC network study.Neurology. 2020; 94: e1027-e1039Crossref PubMed Scopus (121) Google Scholar Notably, the recent increase in incidence is limited to patients of >60 years. The incidence of PCNSL in African-American males of <50 years is more than twofold higher than that in Caucasian males of the same age.6Villano J.L. Koshy M. Shaikh H. et al.Age, gender, and racial differences in incidence and survival in primary CNS lymphoma.Br J Cancer. 2011; 105: 1414-1418Crossref PubMed Scopus (389) Google Scholar Among elderly patients, however, incidence in Caucasian males is twofold higher than that in African-American males. Similar patterns, but with a lesser magnitude, are evident among females.6Villano J.L. Koshy M. Shaikh H. et al.Age, gender, and racial differences in incidence and survival in primary CNS lymphoma.Br J Cancer. 2011; 105: 1414-1418Crossref PubMed Scopus (389) Google Scholar A proposed algorithm for the diagnosis of PCNSL is shown in Figure 1. Patients usually present with a range of neurological or neuropsychiatric symptoms corresponding to the location and extent of the tumour, while systemic symptoms (fever, night sweats and weight loss) are exceptionally rare.8Ferreri A.J. Reni M. Pasini F. et al.A multicenter study of treatment of primary CNS lymphoma.Neurology. 2002; 58: 1513-1520Crossref PubMed Google Scholar The brain is by far the most common location, with frequent involvement of the corpus callosum, basal ganglia and periventricular areas. Up to 40%-50% of patients have multifocal disease on standard magnetic resonance imaging (MRI),7Houillier C. Soussain C. Ghesquieres H. et al.Management and outcome of primary CNS lymphoma in the modern era: an LOC network study.Neurology. 2020; 94: e1027-e1039Crossref PubMed Scopus (121) Google Scholar resulting in a more complex pattern of symptoms. The eye is an important but less common site of disease: vitreous fluid and/or the retina are involved in 15%-20% of patients at presentation, which is asymptomatic in half of the patients while manifesting with blurred vision or floaters in the other half (Figure 1).7Houillier C. Soussain C. Ghesquieres H. et al.Management and outcome of primary CNS lymphoma in the modern era: an LOC network study.Neurology. 2020; 94: e1027-e1039Crossref PubMed Scopus (121) Google Scholar,8Ferreri A.J. Reni M. Pasini F. et al.A multicenter study of treatment of primary CNS lymphoma.Neurology. 2002; 58: 1513-1520Crossref PubMed Google Scholar Ocular involvement is commonly bilateral and can be detected in two different conditions: as the exclusive site of disease or concomitant to other CNS sites, usually brain parenchymal lesions. When disease is limited to the eyes, patients may be initially diagnosed with uveitis, resulting in significant diagnostic delays. If comprehensive staging (see below) excludes the involvement of systemic organs and CNS localisations other than the eye, a diagnosis of primary vitreoretinal lymphoma (PVRL) should be considered. In patients with PVRL, an expert ophthalmologist should exclude differential diagnoses among the 'masquerade syndromes', and definitive diagnosis requires cytology examination of vitreous humour. Often, PVRL precedes brain lesions for months or years. When ocular disease is concomitant to other CNS sites, ocular involvement is detected during staging in patients with PCNSL diagnosed on brain biopsy. In both forms, ocular evaluation should be carried out with a slit lamp, fundoscopy and, if required, retinal angiography or tomography.9Soussain C. Malaise D. Cassoux N. Primary vitreoretinal lymphoma: a diagnostic and management challenge.Blood. 2021; 138: 1519-1534Crossref PubMed Scopus (50) Google Scholar In patients with a suitable volume of vitreous sample, flow cytometry and PCR assessment of genes encoding immunoglobulin (Ig) heavy chains can be carried out. Although not diagnostic, detection of the MYD88 L265P mutation and elevated interleukin-10 (IL-10) levels in the vitreous and aqueous humours are indicators of ocular lymphoma.10Miserocchi E. Ferreri A.J.M. Giuffre C. et al.MYD88 L265P mutation detection in the aqueous humor of patients with vitreoretinal lymphoma.Retina. 2019; 39: 679-684Crossref PubMed Scopus (10) Google Scholar,11Bonzheim I. Giese S. Deuter C. et al.High frequency of MYD88 mutations in vitreoretinal B-cell lymphoma: a valuable tool to improve diagnostic yield of vitreous aspirates.Blood. 2015; 126: 76-79Crossref PubMed Google Scholar PCNSL can spread through the cerebrospinal fluid (CSF). Concurrent leptomeningeal involvement, which is often asymptomatic, is detected by conventional CSF cytology examination in 16% of patients, while isolated leptomeningeal lymphoma represents <5% of all PCNSLs.8Ferreri A.J. Reni M. Pasini F. et al.A multicenter study of treatment of primary CNS lymphoma.Neurology. 2002; 58: 1513-1520Crossref PubMed Google Scholar Spinal cord lymphoma is the rarest manifestation of PCNSL and is often associated with a delayed diagnosis and poor prognosis.6Villano J.L. Koshy M. Shaikh H. et al.Age, gender, and racial differences in incidence and survival in primary CNS lymphoma.Br J Cancer. 2011; 105: 1414-1418Crossref PubMed Scopus (389) Google Scholar Lymphomas arising primarily in the spinal nerves and ganglia, cauda equina and peripheral nerves ('neurolymphomatosis') are extremely rare and should be distinguished from neural infiltration of a systemic lymphoma. Early recognition of imaging features suggestive of PCNSL is essential to expedite definitive tissue diagnosis and avoid corticosteroids before biopsy. The imaging method of choice is contrast-enhanced MRI including diffusion- and perfusion-weighted scans with volumetric protocols in line with the published recommended minimum and desirable standards of the International PCNSL Collaborative Group (IPCG) (Supplementary Table S1, available at https://doi.org/10.1016/10.1016/j.annonc.2023.11.010).12Barajas R.F. Politi L.S. Anzalone N. et al.Consensus recommendations for MRI and PET imaging of primary central nervous system lymphoma: guideline statement from the International Primary CNS Lymphoma Collaborative Group (IPCG).Neuro Oncol. 2021; 23: 1056-1071Crossref PubMed Scopus (63) Google Scholar Patients with contraindications to MRI can be assessed by contrast-enhanced computed tomography (CT) scan. Typical MRI findings are summarised in Supplementary Table S2, available at https://doi.org/10.1016/10.1016/j.annonc.2023.11.010. The combined use of modern neuroimaging and biomarkers can be useful in distinguishing PCNSL from neoplastic and non-neoplastic lesions, including neuroinflammatory diseases, brain metastases, non-necrotic brain tumours, abscesses or tumefactive demyelination. A delay in diagnosis of weeks to months following the onset of symptoms is common in patients with PCNSL.13Abrey L.E. Batchelor T.T. Ferreri A.J. et al.Report of an international workshop to standardize baseline evaluation and response criteria for primary CNS lymphoma.J Clin Oncol. 2005; 23: 5034-5043Crossref PubMed Scopus (673) Google Scholar A critical consideration in the diagnostic pathway is to avoid corticosteroids before biopsy, given their rapid effect on tumour cell viability. Accordingly, if PCNSL is suspected based on imaging and corticosteroids have already been administered, it may be necessary to stop them before stereotactic biopsy of an enhancing lesion. A repeat MRI scan may be required after stopping corticosteroids and before a biopsy. The gold standard diagnostic method in PCNSL relies on the histopathological examination of specimens obtained by stereotactic biopsy. Surgical resection and/or cytological examination of CSF should not be considered first-choice diagnostic methods due to the higher morbidity rate conferred by the former and the low diagnostic reliability of the latter.13Abrey L.E. Batchelor T.T. Ferreri A.J. et al.Report of an international workshop to standardize baseline evaluation and response criteria for primary CNS lymphoma.J Clin Oncol. 2005; 23: 5034-5043Crossref PubMed Scopus (673) Google Scholar Histopathological and molecular findings are summarised in Supplementary Table S3, available at https://doi.org/10.1016/10.1016/j.annonc.2023.11.010. CSF samples should be collected from all patients with suspected or confirmed PCNSL for diagnosis and staging. However, lumbar puncture is not always safe, particularly in patients with concurrent brain masses and/or extensive perilesional oedema. CSF examinations include physical–chemical parameters, conventional cytology and flow cytometry. CSF from patients with PCNSL often has a normal glucose concentration, increased leukocyte count and increased protein concentration. Although CSF examination facilitates diagnosis of PCNSL in <20% of patients, this source of material should be exploited diagnostically. Flow cytometry allows the detection of monotypic B cells, increasing diagnostic sensitivity. Combined assessment of the MYD88 L265P mutation, IL-10 levels and PCR for Ig heavy chain variable (IgVH) rearrangement in the CSF allows efficient discrimination of PCNSL from glial neoplasms and non-neoplastic disorders of the CNS.14Ferreri A.J.M. Calimeri T. Lopedote P. et al.MYD88 L265P mutation and interleukin-10 detection in cerebrospinal fluid are highly specific discriminating markers in patients with primary central nervous system lymphoma: results from a prospective study.Br J Haematol. 2021; 193: 497-505Crossref PubMed Scopus (47) Google Scholar Assessment of these biomarkers in CSF represents a promising tool when biopsy is not possible (e.g. poor clinical condition, brainstem lesions). This approach may also more efficiently identify candidates for stereotactic biopsy among patients with confounding radiological features, i.e. during corticosteroid therapy. PCNSL displays perturbation of pathways related to signalling of B-cell receptors (BCRs), toll-like receptors (TLRs) and nuclear factor-κB (NF-κB), as well as terminal B-cell differentiation, deregulation of the cell cycle, immune escape and protection from apoptosis. Among these, the BCR–TLR–NF-κB axis, mammalian target of rapamycin (mTOR)–protein kinase B (Akt) pathway and programmed cell death protein 1 (PD-1) have already been targeted therapeutically in PCNSL (see Novel therapies subsection). The biological rationales for potential future therapies are summarised in Supplementary Table S4, available at https://doi.org/10.1016/10.1016/j.annonc.2023.11.010. •Contrast-enhanced cranial MRI is the recommended imaging modality for patients with PCNSL [II, A]. The IPCG protocol based on 3T or 1.5T MRI is recommended [V, A].•PCNSL diagnosis must be confirmed by histopathological examination of tumour biopsy [III, A].•Corticosteroid therapy before tissue biopsy should be avoided whenever clinically possible [IV, D]. In case of clinical deterioration, urgent biopsy should be carried out before the start of corticosteroids [IV, A].•Tissue samples should be collected by stereotactic biopsy in patients with brain lesions [IV, A].•Tumour resection is not recommended, except in carefully selected patients with rapidly increasing intracranial pressure who may benefit from surgical debulking at the time of tumour biopsy [IV, D].•Diagnosis is based on morphology and immunohistochemistry [minimum stain panel includes cluster of differentiation (CD)20, CD3, CD10, B-cell lymphoma (Bcl)-6, Bcl-2, multiple myeloma 1 (MUM1) and Ki-67 antibodies]. Molecular analysis of Ig heavy and light chain loci can be used in selected cases where diagnosis is difficult [V, A].•When brain biopsy is contraindicated, CSF examination is a valid option. Flow cytometry, MYD88 L265P mutation analysis and IL-10 levels in CSF samples may support a diagnosis of PCNSL [IV, B].•Suspicion of PVRL should be confirmed by conventional cytology examination of the vitreous humour and, when possible, by flow cytometry. Although not diagnostic, MYD88 L265P mutation and IL-10 levels may be assessed in the vitreous and aqueous humours as indicators of ocular lymphoma [III, A]. A comprehensive assessment of the extent of lymphoma involvement (see Supplementary Table S5, available at https://doi.org/10.1016/10.1016/j.annonc.2023.11.010) is mandatory to determine both the compartments involved within the CNS and the presence of concomitant systemic disease, as recommended by the IPCG guidelines.13Abrey L.E. Batchelor T.T. Ferreri A.J. et al.Report of an international workshop to standardize baseline evaluation and response criteria for primary CNS lymphoma.J Clin Oncol. 2005; 23: 5034-5043Crossref PubMed Scopus (673) Google Scholar Full neurological and oncohaematological evaluation is crucial before treatment planning. Gadolinium-enhanced MRI is the most relevant tool to define an extension of disease in the brain and spinal cord. Brain MRI should be repeated after biopsy and ideally within 14 days before starting treatment12Barajas R.F. Politi L.S. Anzalone N. et al.Consensus recommendations for MRI and PET imaging of primary central nervous system lymphoma: guideline statement from the International Primary CNS Lymphoma Collaborative Group (IPCG).Neuro Oncol. 2021; 23: 1056-1071Crossref PubMed Scopus (63) Google Scholar; this is supported by extremely high proliferative activity, often with >90% of tumour cells expressing the Ki-67 antigen, which could potentially affect therapeutic response definition.1Kluin P.M. Deckert M. Ferry J.A. Primary diffuse large B-cell lymphoma of the CNS.in: Swerdlow S.H. Campo E. Harris N.L. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. IARC, Lyon, France2017: 300-302Google Scholar The involvement of spinal cord parenchyma is rare and specific MRI should be carried out only in patients with symptoms suggestive of spinal cord injury. Meningeal dissemination is often asymptomatic; thus, CSF analysis is advised in every patient with suspected or confirmed PCNSL, unless clinically contraindicated. Physical–chemical features in the CSF (i.e. normal glucose concentration, increased leukocyte count, high protein concentration) are not specific for PCNSL but may suggest meningeal dissemination and blood–CSF barrier disruption. Conventional cytology examination underestimates CSF involvement and should be coupled with flow cytometry to improve diagnostic sensitivity.15Hegde U. Filie A. Little R.F. et al.High incidence of occult leptomeningeal disease detected by flow cytometry in newly diagnosed aggressive B-cell lymphomas at risk for central nervous system involvement: the role of flow cytometry versus cytology.Blood. 2005; 105: 496-502Crossref PubMed Scopus (314) Google Scholar Accurate ophthalmological examination should be carried out in every patient with PCNSL. Vitrectomy, however, is not mandatory in patients with histopathological diagnosis of PCNSL carried out on brain biopsies. Assessment of IL-10 level, MYD88 L265P and monoclonal IgVH rearrangement on vitreous and aqueous humours offers diagnostic potential as a conservative procedure to confirm intraocular disease during staging, but its precise role in routine practice remains to be defined. Assessment for extra-CNS disease is relevant as patients with PCNSL and secondary CNS lymphoma (SCNSL) exhibit different prognoses and require different treatment protocols.16Ferreri A.J.M. Doorduijn J.K. Re A. et al.MATRix-RICE therapy and autologous haematopoietic stem-cell transplantation in diffuse large B-cell lymphoma with secondary CNS involvement (MARIETTA): an international, single-arm, phase 2 trial.Lancet Haematol. 2021; 8: e110-e121Abstract Full Text Full Text PDF PubMed Google Scholar Conventional staging with [18F]2-fluoro-2-deoxy-d-glucose–positron emission tomography (FDG–PET), preferably combined with contrast-enhanced CT scan, can identify systemic disease in 4%-12% of patients with a presumptive diagnosis of PCNSL.16Ferreri A.J.M. Doorduijn J.K. Re A. et al.MATRix-RICE therapy and autologous haematopoietic stem-cell transplantation in diffuse large B-cell lymphoma with secondary CNS involvement (MARIETTA): an international, single-arm, phase 2 trial.Lancet Haematol. 2021; 8: e110-e121Abstract Full Text Full Text PDF PubMed Google Scholar When FDG–PET is not available, bone marrow biopsy and aspiration and testicular ultrasound (US) are recommended to accompany CT imaging. To predict outcomes and better stratify patients in clinical trials, two scoring systems have been proposed: the International Extranodal Lymphoma Study Group (IELSG) score17Ferreri A.J. Blay J.Y. Reni M. et al.Prognostic scoring system for primary CNS lymphomas: the International Extranodal Lymphoma Study Group experience.J Clin Oncol. 2003; 21: 266-272Crossref PubMed Scopus (655) Google Scholar and the Memorial Sloan Kettering Cancer Center prognostic score.18Abrey L.E. Ben-Porat L. Panageas K.S. et al.Primary central nervous system lymphoma: the Memorial Sloan-Kettering Cancer Center prognostic model.J Clin Oncol. 2006; 24: 5711-5715Crossref PubMed Scopus (469) Google Scholar Validation of other proposed scores is pending. Before starting treatment, bone marrow status and cardiac, liver and renal functions should be assessed (see Supplementary Table S5, available at https://doi.org/10.1016/10.1016/j.annonc.2023.11.010). A battery of cognitive functions and quality of life (QoL) measures has been proposed by the IPCG.19Correa D.D. Maron L. Harder H. et al.Cognitive functions in primary central nervous system lymphoma: literature review and assessment guidelines.Ann Oncol. 2007; 18: 1145-1151Abstract Full Text Full Text PDF PubMed Scopus (149) Google Scholar Its use outside clinical trials remains to be defined. •Spinal cord imaging should be carried out in symptomatic patients or in case of CSF positivity [V, B].•Unless lumbar puncture is clinically contraindicated, physical–chemical features of CSF as well as conventional cytology and flow cytometry should be assessed in all patients [IV, B].•Ophthalmological assessment by slit lamp fundoscopy should be carried out in all patients to exclude intraocular involvement [IV, A]. When available, retinal angiography or tomography is advisable [IV, C].•All patients should undergo systemic imaging to exclude extra-CNS disease using FDG–PET, preferably combined with contrast-enhanced CT scan [V, B].•If FDG–PET–CT is not feasible, contrast-enhanced total-body CT scan, bone marrow aspiration and biopsy and testicular US should be carried out [V, B]. A proposed algorithm for the treatment of newly diagnosed PCNSL is shown in Figure 2. Age is the main prognostic factor in PCNSL. A reliable age cut-off by which to distinguish young and elderly patients remains to be defined. This has become an increasingly pressing issue in the last decade due to the wider use of high-dose chemotherapy (HDC) and autologous stem cell transplantation (ASCT), and the increasing incidence of PCNSL with age.7Houillier C. Soussain C. Ghesquieres H. et al.Management and outcome of primary CNS lymphoma in the modern era: an LOC network study.Neurology. 2020; 94: e1027-e1039Crossref PubMed Scopus (121) Google Scholar,20Mendez J.S. Ostrom Q.T. Gittleman H. et al.The elderly left behind-changes in survival trends of primary central nervous system lymphoma over the past 4 decades.Neuro Oncol. 2018; 20: 687-694Crossref PubMed Scopus (150) Google Scholar The management of older patients with PCNSL remains a clinical challenge, with disappointing survival figures. Comorbidities and neurocognitive dysfunction influence individualised treatment approaches, particularly in patients aged 65-75 years. Moreover, the use of high-dose (HD) methotrexate (MTX), the most important component of chemotherapy (ChT) regimens used as first-line treatment (see Supplementary Table S6, available at https://doi.org/10.1016/10.1016/j.annonc.2023.11.010), requires suitable renal (creatinine clearance >50 ml/min), hepatic and cardiac (left ventricular ejection fraction >45%) functions. Accordingly, stratification between 'young' and 'elderly' patients should not be made considering exclusively the patient's age but also the ability to tolerate intensified treatments, informed by performance status (PS), organ function, comorbidities and frailty. Importantly, given the rarity of PCNSL and the complexity of its management, the overall evaluation and treatment of every patient should be carried out in specialist centres by an experienced multidisciplinary team, which should involve neurosurgeons, neuroradiologists, haematopathologists, haematologists, oncologists, radiation oncologists and ophthalmologists with specialist knowledge of PCNSL. Moreover, enrolment in a prospective clinical trial should always be prioritised. Most of the following recommendations concern treatment in routine practice. Response to treatment should follow IPCG criteria13Abrey L.E. Batchelor T.T. Ferreri A.J. et al.Report of an international workshop to standardize baseline evaluation and response criteria for primary CNS lymphoma.J Clin Oncol. 2005; 23: 5034-5043Crossref PubMed Scopus (673) Google Scholar: gadolinium-enhanced MRI of the brain should be carried out every two courses during induction ChT and 2 months after consolidation, and compared with baseline MRI, with the addition of ocular and CSF exams if involved at baseline. Due to superior efficacy, intensive systemic ChT protocols have replaced the historical therapeutic standard of whole-brain radiotherapy (WBRT) as initial treatment of patients with PCNSL. Empirically adopted from systemic lymphoma protocols, combination ChT regimens such as cyclophosphamide–doxorubicin–vincristine–prednisone (CHOP) have proven ineffective due to their insufficient ability to cross the blood–brain barrier.21Mead G.M. Bleehen N.M. Gregor A. et al.A medical research council randomized trial in patients with primary cerebral non-Hodgkin lymphoma: cerebral radiotherapy with and without cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy.Cancer. 2000; 89: 1359-1370Crossref PubMed Scopus (0) Google Scholar HD-MTX is widely established as a key component of current PCNSL remission induction protocols. HD-MTX doses of at least 3 g/m2 with a rapid infusion time of 2-4 h are recommended to achieve sufficient drug levels in the CNS; some experts advise preceding MTX infusion with a fast bolus (MTX 500 mg/m2 infused over 15 min). The central role of HD-MTX has been confirmed in several prospective non-randomised clinical studies either as monotherapy22Batchelor T. Carson K. O'Neill A. et al.Treatment of primary CNS lymphoma with methotrexate and deferred radiotherapy: a report of NABTT 96-07.J Clin Oncol. 2003; 21: 1044-1049Crossref PubMed Scopus (480) Google Scholar or as polychemotherapy (polyChT).23Ghesquieres H. Ferlay C. Sebban C. et al.Long-term follow-up of an age-adapted C5R protocol followed by radiotherapy in 99 newly diagnosed primary CNS lymphomas: a prospective multicentric phase II study of the Groupe d'Etude des Lymphomes de l'Adulte (GELA).Ann Oncol. 2010; 21: 842-850Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar,24Ferreri A.J. Reni M. Foppoli M. et al.High-dose cytarabine plus high-dose methotrexate versus high-dose methotrexate alone in patients with primary CNS lymphoma: a randomised phase 2 trial.Lancet. 2009; 374: 1512-1520Abstract Full Text Full Text PDF PubMed Scopus (544) Google Scholar Recently, the approach to upfront induction therapy has evolved to be increasingly intensive, translating into improved efficacy and higher chances of long-term remission or cure. HD cytarabine (AraC) 2 g/m2 every 12 h for 2 days has also been shown to be an important component of the treatment of PCNSL. This was first demonstrated in the randomised phase II IELSG20 trial.24Ferreri A.J. Reni M. Foppoli M. et al.High-dose cytarabine plus high-dose methotrexate versus high-dose methotrexate alone in patients with primary CNS lymphoma: a randomised phase 2 trial.Lancet. 2009; 374: 1512-1520Abstract Full Text Full Text PDF PubMed Scopus (544) Google Scholar Compared with HD-MTX monotherapy (3.5 g/m2), the HD-MTX–HD-AraC combination was associated with a significantly increased overall response rate (ORR) (40% versus 69%, respectively) and progression-free survival (PFS) (3-year PFS 21% versus 38%), as well as a trend in favour of the combination in overall survival (OS) (3-year OS 32% versus 46%). Based on these results, the international randomised phase II IELSG32 trial compared three different induction protocols: HD-MTX–HD-AraC (standard), HD-MTX–HD-AraC–rituximab and HD-MTX–HD-AraC–rituximab–thiotepa (MATRix regimen) followed by a second randomisation comparing consolidation with HDC–ASCT versus WBRT.25Ferreri A.J. Cwynarski K. Pulczynski E. et al.Chemoimmunotherapy with methotrexate, cytarabine, thiotepa, and rituximab (MATRix regimen) in patients with primary CNS lymphoma: results of the first randomisation of the International Extranodal Lymphoma Study Group-32 (IELSG32) phase 2 trial.Lancet Haematol. 2016; 3: e217-e227Abstract Full Text Full Text PDF PubMed Google Scholar Two hundred and twenty-seven patients were registered across 53 centres in five European countries. The MATRix regimen achieved significantly better response, PFS and OS rates.25Ferreri A.J. Cwynarski K. Pulczynski E. et al.Chemoimmunotherapy with methotrexate, cytarabine, thiotepa, and rituximab (MATRix regimen) in patients with primary CNS lymphoma: results of the first randomisation of the International Extranodal Lymphoma Study Group-32 (IELSG32) phase 2 trial.Lancet Haematol. 2016; 3: e217-e227Abstract Full Text Full Text PDF PubMed Google Scholar An updated analysis at a median follow-up of 88 months confirmed the benefit of MATRix,26Ferreri A.J.M. Cwynarski K. Pulczynski E. et al.Long-term efficacy, safety and neurotolerability of MATRix regimen followed by autologous transplant in primary CNS lymphoma: 7-year results of the IELSG32 randomized trial.Leukemia. 2022; 36: 1870-1878Crossref PubMed Scopus (30) Google Scholar with significantly improved 7-year PFS (52% versus 29% versus 20%) and 7-year OS (56% versus 37% versus 26%) rates compared with HD-MTX–HD-AraC–rituximab and HD-MTX–HD-AraC, respectively. Patients who received the MATRix regimen followed by consoli