Diagnosis and Management of Primary Central Nervous System Lymphoma

Transcription

1 Diagnosis and Management of Primary Central Nervous System Lymphoma Catherine H. Han, MD 1,2 ; and Tracy T. Batchelor, MD, MPH 1,3 Primary central nervous system lymphoma (PCNSL) is a rare and aggressive extranodal non-hodgkin lymphoma (NHL) that is confined to the brain, eyes, spinal cord, or leptomeninges without systemic involvement. The overall prognosis, diagnosis, and management of PCNSL differ from those for other types of NHL. Prompt diagnosis and initiation of treatment are vital for improving clinical outcomes. PCNSL is responsive to radiation therapy; however, whole-brain radiotherapy (WBRT) inadequately controls the disease when it is used alone, and its delayed neurotoxicity causes neurocognitive impairment, especially in elderly patients. High-dose methotrexate (HD- MTX) based induction chemotherapy with or without autologous stem cell transplantation (ASCT) or reduced-dose WBRT leads to durable disease control and less neurotoxicity. The optimal treatment has yet to be defined; however, HD-MTX based induction chemotherapy is considered standard for newly diagnosed PCNSL. Ongoing randomized trials are addressing the roles of rituximab and consolidative treatment with ASCT or reduced-dose WBRT. Despite high tumor response rates with the initial treatment, many patients relapse with a very poor prognosis. The optimal treatment for refractory or relapsed PCNSL is poorly defined. The choice of salvage treatment depends on a patient s age, previous treatment and response, performance status, and comorbidities at the time of relapse. This review provides an overview of the clinical features, diagnosis, pathology, and management of PCNSL in immunocompetent patients, and it focuses on recent advances in treatment. Cancer 2017;123: VC 2017 American Cancer Society. KEYWORDS: autologous stem cell transplantation, high-dose methotrexate, neurotoxicity, primary central nervous system (CNS) lymphoma, whole-brain radiotherapy. INTRODUCTION Primary central nervous system lymphoma (PCNSL) is a rare and aggressive extranodal non-hodgkin lymphoma (NHL). It is confined to the brain, eyes, spinal cord, or leptomeninges without systemic involvement at the time of diagnosis. It accounts for up to 1% of NHL cases and approximately 2% of all primary central nervous system (CNS) tumors. The overall incidence rate is 0.43 per 100,000 people with a slight predilection for males. 1 Among immunocompetent patients, there is an increase in the incidence with advancing age, with the highest rates occurring in the group older than 75 years. The median age at diagnosis is 66 years. The PCNSL rates in men and women older than 65 years continue to rise. 2 Treatment has advanced significantly for PCNSL in the last 2 decades, and long-term survival is observed for approximately 15% to 20% of patients after high-dose methotrexate (HD-MTX) based chemotherapy with or without radiotherapy; this is particularly true for younger patients and those with a good performance status. However, the 5-year survival proportion remains low at 33%. 1 Untreated PCNSL carries a very poor prognosis with an overall survival (OS) of approximately 1.5 months. This review provides an overview of the clinical features, diagnosis, pathology, and management of PCNSL in immunocompetent patients, and it focuses on recent advances in treatment. HISTOPATHOLOGY AND MOLECULAR PROFILE The majority of PCNSL cases (>90%) are diffuse large B-cell lymphoma (DLBCL), with the remainder consisting of T- cell (2%), Burkitt, lymphoblastic, and low-grade lymphomas. 3 Primary central nervous system (PCNS) DLBCL is recognized as a distinct subtype of DLBCL in the World Health Organization classifications 4 and expresses pan B-cell antigens (CD19, CD20, and CD79a). 3,5 Melanoma-associated antigen (mutated) 1 (MUM1)/interferon regulatory factor 4 (IRF4) is nearly always positive, B-cell CLL/lymphoma 6 (BCL-6) is expressed in approximately 50% of cases, B-cell CLL/lymphoma 2 (BCL-2) is variably expressed, and CD10 is expressed in only a small minority of cases (approximately Corresponding author: Tracy T. Batchelor, MD, MPH, Stephen E. and Catherine Pappas Center for Neuro-Oncology, Massachusetts General Hospital, 55 Fruit Street, Yawkey 9E, Boston, MA 02114; tbatchelor@mgh.harvard.edu 1 Department of Neurology, Division of Hematology/Oncology, Massachusetts General Hospital Cancer Center and Harvard Medical School, Boston, Massachusetts; 2 Auckland Cancer Society Research Centre, School of Medical Sciences, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; 3 Department of Radiation Oncology, Division of Hematology/Oncology, Massachusetts General Hospital Cancer Center and Harvard Medical School, Boston, Massachusetts DOI: /cncr.30965, Received: May 23, 2017; Revised: July 6, 2017; Accepted: July 11, 2017, Published online September 26, 2017 in Wiley Online Library (wileyonlinelibrary.com) 4314 Cancer November 15, 2017

2 Primary CNS Lymphoma/Han and Batchelor 10%). 3,5 This means that the majority of PCNS DLBCLs most closely resemble a postgerminal center or an activated B-cell (ABC) immunophenotype (CD10, BCL- 61, and MUM1/IRF41). BCL-6, CD10, BCL-2, MUM1/IRF4, and Ki-67 may have prognostic importance. For example, BCL-6 expression was associated with inferior survival outcomes in a recent study of B-cell differentiation markers and their prognostic impact in a cohort of 119 patients from a prospective trial (G- PCNSL-SG-1). 6 PCNS DLBCL more often has a loss of human leukocyte antigen class I and/or human leukocyte antigen class II expression in comparison with primary nodal DLBCL. This could result in the evasion of neoplastic B cells from immune surveillance by T cells; this, in turn, may explain the poorer prognosis for patients with PCNS DLBCL. 5 The extent of somatic hypermutation of protooncogenes has also been shown to be greater in PCNS DLBCL versus DLBCL arising outside the CNS. 7 A recent genomic analysis of 19 PCNSL samples from immunocompetent patients identified recurrent abnormalities in PRKCD and TOX genes. 8 PRKCD, a proapoptotic protein kinase, has been implicated in cellular processes such as growth, differentiation, secretion, apoptosis, and tumor development. TOX plays a key role in T-cell development and has some effect on B-cell development as well. Mutations leading to the activation of the nuclear factor jb signaling pathway, such as activating mutations of MYD88, CARD11, and CD79 and deletions of TNFAIP3 and TBL1XR1, are observed in almost all PCNSL cases, and this suggests that nuclear factor jb activation may play a role in the pathogenesis of PCNS DLBCL and may represent a potential therapeutic target. 8,9 Gene expression profiles also demonstrate that PCNSL is characterized by the differential expression of genes related to adhesion and extracellular matrix pathways, including MUM1, CXCL13, and CHI3L1. Alterations in TP53, CDKN2A/p16, BCL-6, MYC, and PAX5 genes are also commonly observed in PCNSL. 9 CLINICAL FEATURES Presenting symptoms vary with the CNS site involved. Common symptoms include cognitive and/or behavioral changes lasting from weeks to months, focal neurologic deficits, and symptoms of increased intracranial pressure. Because PCNSL tends to involve deep brain structures, seizures are relatively uncommon. Concurrent ocular involvement occurs in approximately 10% to 20% of PCNSL patients, who may present with ocular symptoms such as floaters and/or blurred vision. However, up to half of patients with ocular involvement have no ocular symptoms. 10 Primary vitreoretinal lymphoma is rare, with approximately 300 new cases diagnosed in the United States each year. Approximately 15% to 20% of patients have concurrent leptomeningeal involvement, which is often asymptomatic and is detected only on cerebrospinal fluid (CSF) evaluation. 11 The rarest manifestation of PCNSL is spinal cord lymphoma, which presents with symptoms that correspond to the level of spinal cord involvement. 12 NEUROIMAGING Magnetic resonance imaging (MRI) with gadolinium contrast is the most sensitive imaging modality in the diagnosis of PCNSL. PCNSL is characterized by homogeneous contrast enhancement with well-defined borders 13 (Fig. 1). Nonenhancing lesions are rare. Perilesional vasogenic edema is common. A low signal on T2-weighted MRI and restricted diffusion on diffusion-weighted imaging (DWI) are other characteristics of PCNSL, and they may be explained by its high cellularity with tightly compacted cells and the high nuclear-to-cytoplasmic ratio. 14 These radiographic features may help to differentiate PCNSL from other conditions such as infections, tumefactive demyelinating lesions, and gliomas. 15 A radiographic response to corticosteroids is common, but this does not secure the diagnosis of PCNSL because inflammatory or demyelinating conditions show a response as well. Rarely, PCNSL may present with subtle focal abnormalities of cranial or radicular nerves or focal meningeal enhancement. Recent studies of advanced imaging techniques have demonstrated the usefulness of these metrics in the diagnosis and prognostication of PCNSL. 14 For example, fractional anisotropy of diffusion tensor imaging correlates with the microstructural integrity of myelinated fiber tracts, and it is significantly lower in PCNSL versus glioblastoma multiforme. 16 Similarly, on magnetic resonance spectroscopy, PCNSL differs from other glial tumors, including glioblastoma multiforme, by massively increased lipid resonances and markedly higher choline/ creatine and choline/n-acetyl aspartate ratios. 17 These imaging techniques, therefore, could facilitate differentiating PCNSL from other brain pathologies. DWI derived apparent diffusion coefficient (ADC) measurements inversely correlate with PCNSL tumor cellular density, and it has been shown that in comparison with higher ADC values, low pretherapeutic ADC values may be predictive of earlier disease progression and shorter OS for patients with PCNSL despite the same chemotherapy treatment. 18 Metabolic imaging by 18 F- Cancer November 15,

3 Figure 1. Brain magnetic resonance imaging from a patient with primary central nervous system lymphoma. (A) Axial, post gadolinium contrast, T1-weighted imaging demonstrates intense, homogeneous enhancement with well-defined borders. (B) Axial, T2- weighted fluid-attenuated inversion recovery imaging demonstrates an increased signal surrounding the tumor, and this reflects vasogenic cerebral edema. (C) Axial diffusion-weighted imaging shows an increased signal within the tumor, and this suggests high cellularity. fluorodeoxyglucose (FDG) or 11 C-methionine positron emission tomography (PET) has been suggested to be able to predict a therapy response at a very early time point and also to detect early disease recurrence during posttreatment surveillance before changes are seen on MRI. 19,20 These newer imaging techniques may ultimately provide noninvasive imaging biomarkers that might identify high-risk PCNSL patients for poor outcomes and may facilitate individualized treatment plans in the future. DIAGNOSIS PCNSL presents as a solitary brain lesion in up to 70% of immunocompetent patients. It typically occurs in the supratentorial region with a predilection for the periventricular white matter. The common sites include the cerebral hemisphere (38%), basal ganglia and thalamus (16%), and corpus callosum (14%). 13 Disease confined to the CSF, eye, or spinal cord at presentation is rare. 21 A histopathologic diagnosis, typically by stereotactic brain biopsy, is essential. Occasionally, CSF cytology and flow cytometry or vitrectomy/chorioretinal biopsy may be sufficient for a definitive diagnosis. However, because awaiting CSF or vitreous fluid diagnostic results can delay the ultimate diagnosis and treatment, it is advisable to proceed with early biopsy for brain lesions with a radiographic appearance consistent with PCNSL. Corticosteroids can interfere with a histopathological diagnosis and result in significant diagnostic delays; therefore, it is important to avoid these drugs before biopsy, if possible, when PCNSL is suspected. In addition to a diagnostic tissue biopsy, comprehensive baseline evaluations, as outlined in Table 1, are recommended to establish the extent of disease and confirm the diagnosis of PCNSL versus systemic NHL with secondary CNS involvement. 22 The extent of disease should be evaluated with a CSF analysis, a slit-lamp examination of both eyes, and a contrast-enhanced MRI scan of the spine. Evidence of systemic disease has been reported in up to 8% of patients initially thought to have PCNSL. 22 Staging FDG body PET/computed tomography (CT) may be more sensitive than conventional CT of the chest, abdomen, and pelvis for detecting systemic disease, as illustrated by a retrospective study in which 7% of the patients were found to have systemic NHL by FDG- PET imaging when the staging CT scan and bone marrow biopsies were negative. 23 Bone marrow biopsy and aspiration can also reveal subclinical, systemic (sometimes lower grade) lymphoma not detectable by other staging procedures. 24 Approximately 3% of patients with testicular lymphoma have brain metastases at diagnosis mimicking PCNSL, and body CT or PET scans may rarely miss an abnormality in the testis; therefore, testicular ultrasound for men should be considered. 25 PROGNOSTIC MODELS Age and performance status are the most important prognostic factors for PCNSL. There are 2 prognostic scoring systems developed specifically for PCNSL. The International Extranodal Lymphoma Study Group identified the following as independent survival predictors in its retrospective analysis of 378 immunocompetent PCNSL 4316 Cancer November 15, 2017

4 Primary CNS Lymphoma/Han and Batchelor TABLE 1. Recommended Baseline Evaluation Clinical Evaluation Comprehensive physical and neurologic examination Age and performance status (ECOG PS or KPS) Cognitive function evaluation (at a minimum MMSE) Corticosteroid dosing Laboratory Evaluation Serum LDH Hepatic and renal function (including creatinine clearance) HIV serologic testing Extent-of-Disease Evaluation Gadolinium-enhanced brain MRI (or contrast-enhanced CT if MRI is contraindicated) CSF analysis: cytology, flow cytometry, cell counts, protein and glucose levels, b2-microglobulin, and immunoglobulin heavy chain gene rearrangement a Ophthalmologic examination, including slit-lamp examination of both eyes Gadolinium-enhanced whole-spine MRI PET/CT (chest/abdomen/pelvis) Bone marrow biopsy with aspirate Testicular ultrasound for men Abbreviations: CSF, cerebrospinal fluid; CT, computed tomography; ECOG, Eastern Cooperative Oncology Group; HIV, human immunodeficiency virus; KPS, Karnofsky performance status; LDH, lactate dehydrogenase; MMSE, Mini-Mental State Examination; MRI, magnetic resonance imaging; PET, positron emission tomography; PS, performance status. Data were derived from Abrey et al. 22 a CSF should be sampled before or 1 week after surgical biopsy to avoid false-positive results. TABLE 2. Prognostic Scoring System From the International Extranodal Lymphoma Study Group Prognostic Factor a Age > 60 y ECOG PS 2 Elevated LDH Elevated CSF protein concentration Involvement of the deep structures of the brain Scores 2-y Overall Survival, % 0or1 80 2or3 48 4or5 15 Abbreviations: CSF, cerebrospinal fluid; ECOG, Eastern Cooperative Oncology Group; LDH, lactate dehydrogenase; PS, performance status. Data were derived from Ferreri et al. 26 a Each variable is assigned a value of 1 if it is present. The final score is the sum of these values. patients: age, Eastern Cooperative Oncology Group performance status, serum lactate dehydrogenase level, CSF protein concentration, and involvement of deep structures of the brain. 26 The International Extranodal Lymphoma Study Group scoring system derived from these factors and the corresponding 2-year OS rates are presented in Table 2. The Memorial Sloan-Kettering Cancer Center prognostic model is the other scoring system (Table 3). TABLE 3. Prognostic Model From the Memorial Sloan-Kettering Cancer Center Class Median Overall Survival, y Median Failure-Free Survival, y 1 (age 50 y) (age > 50 y 1 KPS 70) (age > 50 y 1 KPS < 70) Abbreviation: KPS, Karnofsky performance status. Data were derived from Abrey et al. 27 This was derived from a retrospective analysis of 338 patients in which age and the Karnofsky performance status (KPS) were the only independent prognostic factors. 27 MANAGEMENT OF NEWLY DIAGNOSED PCNSL Newly diagnosed PCNSL is first treated with induction chemotherapy; the goal is to achieve a complete radiographic response (CR). Consolidation therapy then follows with the objective of eliminating any residual disease and prolonging OS. The general approach to the treatment of newly diagnosed PCNSL is outlined in the algorithm (Fig. 2) and is discussed in the sections that follow. Response criteria developed by the International PCNSL Collaborative Group have been widely adopted into routine practice as well as prospective clinical trials of PCNSL (Table 4). 22 Surgery The role of surgery in PCNSL is limited to establishing a histopathologic diagnosis via stereotactic biopsy. Although an unplanned subset analysis of a phase 3 trial suggested a survival benefit of surgical resection of PCNSL versus biopsy, the study was retrospective in nature, and its results were possibly confounded by a selection bias. 28 Other studies have shown no clear benefit of surgical resection of PCNSL. 29 Moreover, it can potentially cause neurologic deficits or lead to a treatment delay. However, it may be considered under special circumstances, for example, if there is evidence of increased intracranial pressure from a large lesion with acute symptoms of brain herniation. In such rare cases, tumor debulking might be beneficial for improving symptoms and tolerance to upcoming intensive chemotherapy. Whole-Brain Radiotherapy (WBRT) and Delayed Neurotoxicity Historically, PCNSL was treated with WBRT because of its high sensitivity to radiation until improved survival was seen with HD-MTX in the 1970s. WBRT resulted in Cancer November 15,

5 Figure 2. General approach to the treatment of newly diagnosed primary central nervous system lymphoma in immunocompetent adults. ASCT indicates autologous stem cell transplantation; CNS, central nervous system; CR, complete radiographic response; ECOG, Eastern Cooperative Oncology Group; HD, high-dose; HD-MTX, high-dose methotrexate; MTX, methotrexate; PD, progressive disease; PR, partial response; PS, performance status; SD, stable disease; WBRT, whole-brain radiation therapy. TABLE 4. International PCNSL Collaborative Group Response Criteria for PCNSL Response Brain Imaging Corticosteroid Dose Eye Examination CSF Cytology CR No contrast-enhancing lesion None Normal Negative CRu No contrast-enhancing lesion Any Normal Negative Minimal abnormality Any Minor RPE abnormality Negative PR 50% decrease in enhancing lesion Irrelevant Normal or minor RPE abnormality Negative No contrast-enhancing lesion Irrelevant Decrease in citreous cells or retinal infiltrate Persistent or suspicious PD 25% increase in enhancing lesion Irrelevant Recurrent or new ocular disease Recurrent or positive Any new site of disease: CNS or systemic SD All cases not covered by responses above Abbreviations: CNS, central nervous system; CR, complete response; CRu, unconfirmed complete response; CSF, cerebrospinal fluid; PCNSL, primary central nervous system lymphoma; PD, progressive disease; PR, partial response; RPE, retinal pigment epithelium; SD, stable disease. This table has been adapted with permission from the American Society of Clinical Oncology. 22 high initial responses, but early relapse was seen in the majority of patients with a median OS of 12 to 18 months. 30 WBRT-related delayed neurotoxicity is a problem in survivors, especially patients older than 60 years. It causes cognitive impairment, incontinence, and gait disturbance. A significant decline in KPS (20-50 points) with a marked reduction in quality of life has been observed in several studies, especially in the elderly population. 31,32 Treatment-related neurotoxicity manifests as white matter abnormalities, ventricular enlargement, and cortical atrophy on MRI. There appears to be a correlation between the extent of the radiographic abnormality and clinical cognitive dysfunction. 31 The mechanism of WBRT-related neurotoxicity remains unclear but may involve toxicity to blood vessels, demyelination, and depletion of neural progenitor cells from the subventricular zone There are currently no preventive measures or treatments to reverse these neurotoxic effects. Because of its inadequate control of the disease and the risk of neurotoxicity, WBRT alone is no longer routinely recommended for patients with newly diagnosed PCNSL. Chemotherapy and Multimodality Therapy In the 1970s, HD-MTX was recognized as the most effective treatment for PCNSL. Studies of single-agent HD- MTX (3.5-8 g/m 2 ) demonstrated overall response rates (ORRs) of 35% to 74%, median progression-free survival 4318 Cancer November 15, 2017

6 Primary CNS Lymphoma/Han and Batchelor (PFS) of 10 to 12.8 months, and median OS of 25 to 55 months Improved tumor responses and survival outcomes were observed when other chemotherapeutic agents were added to HD-MTX. 39,40 A large variation exists in the survival outcomes across studies of combination chemotherapy because of the different regimens and dosing schedules used in these studies. Agents studied in addition to methotrexate include temozolomide, rituximab, procarbazine, cytarabine, thiotepa, vincristine, carmustine, etoposide, ifosfamide, and cyclophosphamide. For example, in a multicenter randomized phase 2 trial, adding cytarabine to HD-MTX as induction treatment improved the CR rate (46% vs 18%), 3-year PFS (38% vs 21%), and 3-year OS (46% vs 32%) in comparison with HD- MTX alone. 40 In this trial, all patients received consolidative WBRT. In a more recent randomized phase 2 trial, the addition of thiotepa and rituximab to HD-MTX/ cytarabine improved the CR rate (49% vs 23%), 2-year PFS (61% vs 36%), and 2-year OS (69% vs 42%). 39 It is generally agreed that the initial induction therapy should include HD-MTX based combination chemotherapy; however, there is no consensus on the optimal dose and schedule of HD-MTX or the optimal chemotherapy combination. The role of additional consolidative WBRT in patients with newly diagnosed PCNSL also remains debatable. The higher incidence of neurotoxicity observed with a combined modality treatment including standard-dose WBRT prompted studies focusing on either reducing the WBRT dose or deferring WBRT until recurrence. A multicenter phase 2 trial evaluated a regimen consisting of HD-MTX, procarbazine, vincristine, and rituximab (R- MPV) followed by consolidative reduced-dose WBRT (23.4 Gy) and cytarabine in patients who had achieved a CR after the induction chemotherapy. 41 The median PFS was 7.7 years and the 3-year OS rate was 87% for patients who achieved a CR and received all planned treatment. The median OS was not reached in this group after a median follow-up of 6 years. Comprehensive neuropsychological testing in this study showed an improvement in baseline executive function and verbal memory after induction chemotherapy that was attributable to a tumor response and relatively stable scores at 48 months of follow-up. Similar results were observed in another multicenter phase 2 trial investigating initial chemoimmunotherapy with HD-MTX, temozolomide, and rituximab (MTR) followed by hyperfractionated WBRT (36 Gy) and then 10 months of postradiotherapy maintenance temozolomide. 42 Although the lower incidence of neurotoxicity observed in these studies is promising, longer neuropsychological follow-up is necessary to definitively address the safety of these treatment regimens using lower dose consolidative WBRT. Other studies have omitted WBRT from the treatment of newly diagnosed PCNSL patients. In a multicenter, randomized phase 3 trial, patients were randomized to receive HD-MTX based chemotherapy with or without consolidative WBRT (45 Gy). 43 This was a noninferiority study, and there was a high dropout rate in the WBRT arm. Although 2-year PFS was prolonged with WBRT versus chemotherapy alone (43.5% vs 30.7%), there was no OS difference (32.4 vs 37.1 months). 44,45 The neurotoxicity rate was nearly doubled in the radiotherapy group (49% vs 26%). These findings suggest that omitting WBRT from the initial treatment of PCNSL may not compromise OS and might better preserve neurological function. However, the noninferiority endpoint was not statistically significant in this study. In a multicenter phase 2 trial of an induction MTR regimen followed by a consolidative etoposide/cytarabine combination without WBRT, the CR rate after induction therapy was 66%, and the median PFS was 29 months. 46 The median OS was not reached with a median follow-up of 5 years. These results are comparable to those seen with consolidative WBRT. An ongoing randomized phase 2 trial (Radiation Therapy Oncology Group 1114) is comparing R-MPV with or without reduced-dose WBRT (23.4 Gy) followed by cytarabine (NCT ). Another consolidative approach omitting WBRT is high-dose chemotherapy followed by autologous stem cell transplantation (ASCT). This involves leukapheresis and peripheral blood stem cell harvesting followed by conditioning chemotherapy and then reinfusion of the stem cells to restore blood cell production. Currently, there is no consensus on the optimal regimen. However, conditioning regimens that contain CNS-penetrant agents such as carmustine, thiotepa, and busulfan have demonstrated the most encouraging results. Promising 2-year OS rates were reported by 2 phase 2 trials using this approach. In a multicenter phase 2 study, patients were treated with induction chemotherapy consisting of HD-MTX, cytarabine, thiotepa, and rituximab, which was followed by high-dose carmustine and thiotepa conditioning with ASCT. 47 The ORR was 91%, and the 2-year OS rate was 87%. The treatment-related mortality rate was <10%. In another phase 2 study, patients who responded to an induction R-MPV regimen were treated with consolidative high-dose chemotherapy consisting of thiotepa, busulfan, and cyclophosphamide followed by ASCT. 48 The 2-year PFS and OS rates were almost identical at Cancer November 15,

7 79% and 81%, respectively, and this indicated that nearly all patients alive 2 years after treatment were free of disease progression. With a median follow-up of 45 months, the median PFS and OS were not reached. In this study, however, the treatment-related mortality rate among transplant patients was high at 11.5%. Long-term survival data from another phase 2 trial showed a durable response with an OS rate of 35% after 10 years of follow-up in patients treated with HD-MTX and then high-dose busulfan and thiotepa followed by ASCT. 49 Moreover, 7 of the 8 patients who were treated with chemotherapy alone had no neurotoxicity and had an excellent quality of life and functional status with Karnofsky performance scores of 90% to 100%. Overall, high-dose chemotherapy followed by ASCT appears to be highly effective with less delayed neurotoxicity; therefore, it is a promising consolidative approach, especially for younger patients with a good performance status. Several ongoing randomized trials are comparing the efficacy of different consolidation approaches: high-dose chemotherapy followed by ASCT versus WBRT (NCT and NCT ) or ASCT versus chemotherapy (NCT and NCT ). Elderly Patients Patients older than 60 years account for more than half of all PCNSL cases. Age has been identified as an independent poor prognostic factor for this type of lymphoma. Moreover, the risk of neurotoxicity is highest in elderly patients, and the majority of them develop significant neurotoxicity after a WBRT-containing regimen. Recently, a meta-analysis of 783 elderly patients with newly diagnosed PCNSL showed that 73% received HD-MTX with a median dose of 3 g/m Regimens containing HD-MTX were associated with improved survival, especially when they were combined with an oral alkylating agent such as procarbazine or temozolomide. In this meta-analysis, more aggressive intravenous chemotherapy regimens were not superior to HD-MTX pus oral chemotherapy. 50 WBRT was also associated with improved survival; however, it was associated with an increased risk of neurotoxicity (odds ratio, 5.23). The standard of care for elderly patients with PCNSL is not established; however, it is generally agreed that WBRT should be avoided or deferred to disease relapse and that multi-agent HD-MTX based chemotherapy should be considered as the initial treatment. In a multicenter, randomized phase 2 trial, 98 elderly patients were randomized to receive HD-MTX with temozolomide versus a regimen consisting of HD-MTX, vincristine, and cytarabine (MPV-A). 51 No WBRT was included in either arm. There were trends favoring the MPV-A arm with respect to the median PFS (9.5 vs 6.1 months), OS (31 vs 14 months), and ORR (82% vs 71%), but these differences were not statistically significant. The quality of life improved in both groups, and no significant difference in toxicities was observed between the 2 arms. Also, there was no evidence of neurotoxicity. In a nonrandomized phase 2 trial of HD-MTX, lomustine, and procarbazine (MCP) in 30 elderly patients, the ORR was 70%, and the median PFS and OS were 5.9 and 15.4 months, respectively. 52 In another nonrandomized phase 2 trial, rituximab was added to the MCP regimen for 28 elderly patients. 53 The efficacy endpoints appeared better with this chemo-immunotherapy regimen with an ORR of 82%, a median PFS of 16 months, and a median OS of 17.5 months. Patients younger than 80 years had a better median OS of 29 months in comparison with the median OS of 4.3 months for those who were 80 years old or older. Toxicities were manageable with a treatmentrelated mortality rate of 7%. A post hoc analysis of elderly patients treated with HD-MTX based chemotherapy in the aforementioned phase 3 trial (G-PCNSL-SG-1) 43 reported the median PFS and OS times after the achievement of a CR to be 16.1 and 26.7 months, respectively. 54 REFRACTORY OR RELAPSED PCNSL Despite high response rates with initial HD-MTX based treatment, more than half of responders relapse. Moreover, approximately a quarter of patients fail to respond to the initial treatment. The prognosis for patients with refractory or relapsed PCNSL remains poor. In a retrospective study, the median overall survival from the time of progression (OS2) was 2 months for primary refractory patients and 3.7 months for patients who relapsed within the first year of the initial therapy. 55 A better prognosis was observed for patients who received high-dose chemotherapy with ASCT at relapse; however, these patients tended to be younger with a better performance status and more chemosensitive disease. Moreover, approximately a quarter of them were asymptomatic at relapse, and this may explain the better KPS and better survival in this group. This observation supports the need for regular neuroimaging surveillance after the initial treatment. The optimal treatment for refractory or relapsed PCNSL is poorly defined because there are only a limited number of prospective trials in this setting. The choice of salvage treatment depends on the age, previous treatment and response, performance status, and comorbidities at the time of relapse. HD-MTX rechallenge is an effective 4320 Cancer November 15, 2017

8 Primary CNS Lymphoma/Han and Batchelor approach in patients who have previously achieved a prolonged response to this agent. This approach is supported by retrospective studies, which have reported response rates of 85% to 91% and median OS2 times of 41 to 62 months. 56,57 For younger patients with a good performance status who have not previously been treated with high-dose chemotherapy and ASCT, this is also an option at the time of relapse. In a phase 2 trial of high-dose etoposide and cytarabine followed by high-dose chemotherapy with a thiotepa, busulfan, and cyclophosphamide regimen and ASCT, 96% of those refractory/relapsed PCNSL patients who proceeded to transplantation achieved a CR. 58 The median PFS and OS in this group were 41.4 and 58.6 months, respectively. For patients who have not been treated with WBRT as a part of their initial treatment, this modality can be an effective option and is supported by retrospective studies in which a response was observed in approximately 75% of patients. 24,49 The median PFS and OS were 10 and 11 to 19 months, respectively. These results suggest that the efficacy of WBRT deferred until relapse is comparable to that seen when WBRT is used as the initial therapy. Delayed neurotoxicity of WBRT remains a problem with this approach because it was observed in 15% to 29% of patients who survived more than 4 months after WBRT. 32,59 An age greater than 60 years and a shorter time (<6 months) between initial HD-MTX based therapy and salvage WBRT were associated with an increased risk of neurotoxicity. There are reports of other agents, including temozolomide, topotecan, rituximab, rituximab/temozolomide, bendamustine, ifosfamide/etoposide, and cisplatin/cytarabine, with varying results. Novel Therapeutics A number of novel therapeutic agents are currently being investigated, and they include temsirolimus, pembrolizumab, nivolumab, ibrutinib, buparlisib, pemetrexed, lenalidomide, and pomalidomide. Because of the activity of programmed cell death protein 1 (PD-1) blockade in other lymphomas, 4 patients with relapsed/refractory PCNSL were treated with an anti PD-1 antibody, nivolumab. 60 All patients responded, and PFS ranged from 14 to 171 months, with 2 patients remaining progressionfree at 131 and 171 months. Phase 2 trials of PD-1 blockade (nivolumab and pembrolizumab) are ongoing (NCT and NCT ). Another promising agent is lenalidomide, an immunomodulatory agent with antiproliferative activities. An interim analysis of 13 relapsed CNS NHL patients (8 with PCNSL) treated with this agent plus rituximab in a phase 1 trial (NCT ) reported radiographic responses in 8 patients, 61 Four patients maintained a CR for more than 9 months, and 2 patients did so for more than 1.8 years. Five of 12 patients in a separate cohort of the same study, who were treated with maintenance lenalidomide after salvage, maintained remission for more than 2 years. Ibrutinib, a Bruton s tyrosine kinase (BTK) inhibitor, was recently evaluated for relapsed/refractory PCNSL on the basis of its activity in systemic ABC DLBCL. In a singlecenter, dose-escalation phase 1 trial, 20 patients with relapsed or refractory CNS lymphoma (13 with PCNSL and 7 with secondary CNS lymphoma) were treated with single-agent ibrutinib. 62 A clinical response was observed in 77% of PCNSL patients, including 5 patients with a CR. The median PFS and OS were 4.6 and 15 months, respectively, for this group of patients. In another phase 1 trial, ibrutinib was incorporated into a novel regimen consisting of dose-adjusted temozolomide, etoposide, liposomal doxorubicin, dexamethasone, ibrutinib, and rituximab (DA-TEDDI-R). 63 During the ibrutinib monotherapy window (day 14 to day 1), 94% of 18 patients (13 with relapsed/refractory PCNSL and 5 with untreated PCNSL) had a radiographic reduction of the tumor. Among 14 evaluable patients, 86% achieved a CR with DA-TEDDI-R. For 13 patients with relapsed/refractory disease, the median PFS was 15.3 months, and the median OS was not reached with 51.3% of the patients alive at 1 year. In this study, 2 patients died of invasive aspergillosis during the ibrutinib monotherapy window; this was potentially attributable to ibrutinib. A subsequent study using Btk knockout and wild-type mice demonstrated that the abrogation of innate immune control of TABLE 5. Recommended Follow-Up Schedule and Assessments Recommended Follow-Up Schedule Years 1 and 2 Years 3-5 Years 6-10 Minimum Assessments at Each Follow-Up History Physical examination Cognitive evaluation (eg, IPCG battery or MMSE) Gadolinium-enhanced MRI of the brain (CT with contrast if MRI contraindicated) Optional as Clinically Indicated Ophthalmologic examination CSF analysis At completion of therapy Every 3 mo Every 6 mo Annually Abbreviations: CSF, cerebrospinal fluid; CT, computed tomography; IPCG, International PCNSL Collaborative Group; MMSE, Mini-Mental State Examination; MRI, magnetic resonance imaging. Data were derived from Abrey et al. 22 Cancer November 15,

9 an Aspergillus infection may be an on-target effect of a BTK inhibitor. 63 Temsirolimus, a mammalian target of rapamycin inhibitor, was tested in a phase 2 trial in which an ORR of 54% was observed. 64 The relatively high response rate indicates that this agent is active in PCNSL; however, the short PFS of 2.1 months observed in this study suggests that the activity is transient, probably because of the development of drug resistance. A further evaluation in combination with other agents could be considered. MONITORING AND FOLLOW-UP The majority of recurrences occur within the CNS and within the first 5 years of treatment completion. However, continuing follow-up to 10 years is recommended because there are late recurrences. The International PCNSL Collaborative Group s recommendations for follow-up assessments and monitoring are outlined in Table CONCLUSIONS Treatment for PCNSL has advanced significantly with improved survival for this rare and aggressive lymphoma of the CNS; however, relapse is common, and long-term survival remains poor. Although the optimal treatment approach has yet to be established, HD-MTX based chemotherapy is currently considered the standard induction treatment for newly diagnosed PCNSL. Ongoing randomized trials are investigating different consolidative approaches using high-dose chemotherapy with ASCT, reduced-dose WBRT, or chemotherapy. With a better understanding of the molecular characteristics of PCNSL, individualized treatments that incorporate targeted agents may be the future direction for improving not only survival but also the quality of life for patients with PCNSL. FUNDING SUPPORT Tracy T. Batchelor reports grant funding from the National Cancer Institute that supports primary central nervous system lymphoma research (K24CA and R13CA124293). CONFLICT OF INTEREST DISCLOSURES Tracy T. Batchelor reports personal fees from Merck, Roche, Amgen, NXCD, Oxigene, Foundation Medicine, Cavion, Kirin, Proximagen, Accerta, UpToDate, Research To Practice, Imedex, Champions Biotechnology, and Oakstone Publishing and grants from Pfizer outside the submitted work. REFERENCES 1. Ostrom QT, Gittleman H, de Blank PM, et al. American Brain Tumor Association adolescent and young adult primary brain and central nervous system tumors diagnosed in the United States in Neuro Oncol. 2016;18(suppl 1):i1-i Shiels MS, Pfeiffer RM, Besson C, et al. Trends in primary central nervous system lymphoma incidence and survival in the U.S. Br J Haematol. 2016;174: Giannini C, Dogan A, Salomao DR. CNS lymphoma: a practical diagnostic approach. J Neuropathol Exp Neurol. 2014;73: Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Geneva, Switzerland: World Health Organization; Ferry JA, Pfannl R, Harris NL. Pathology of primary central nervous system lymphoma and related conditions. In: Batchelor T, DeAngelis LM, eds. Lymphoma and Leukemia of the Nervous System. 2nd ed. New York, NY: Springer; 2012: Kreher S, Johrens K, Strehlow F, et al. Prognostic impact of B-cell lymphoma 6 in primary CNS lymphoma. Neuro Oncol. 2015;17: Montesinos-Rongen M, Van Roost D, Schaller C, Wiestler OD, Deckert M. Primary diffuse large B-cell lymphomas of the central nervous system are targeted by aberrant somatic hypermutation. Blood. 2004;103: Braggio E, Van Wier S, Ojha J, et al. Genome-wide analysis uncovers novel recurrent alterations in primary central nervous system lymphomas. Clin Cancer Res. 2015;21: Ferry JA. The diversity of diffuse large B-cell lymphoma in extranodal sites: overview and update. J Hematopathol. 2014;7: Citterio G, Reni M, Ferreri AJ. Present and future treatment options for primary CNS lymphoma. Expert Opin Pharmacother. 2015;16: Fischer L, Martus P, Weller M, et al. Meningeal dissemination in primary CNS lymphoma: prospective evaluation of 282 patients. Neurology. 2008;71: Flanagan EP, O Neill BP, Porter AB, Lanzino G, Haberman TM, Keegan BM. Primary intramedullary spinal cord lymphoma. Neurology. 2011;77: Kuker W, Nagele T, Korfel A, et al. Primary central nervous system lymphomas (PCNSL): MRI features at presentation in 100 patients. J Neurooncol. 2005;72: Nabavizadeh SA, Vossough A, Hajmomenian M, Assadsangabi R, Mohan S. Neuroimaging in central nervous system lymphoma. Hematol Oncol Clin North Am. 2016;30: Ahn SJ, Shin HJ, Chang JH, Lee SK. Differentiation between primary cerebral lymphoma and glioblastoma using the apparent diffusion coefficient: comparison of three different ROI methods. PLoS One. 2014;9:e Toh CH, Castillo M, Wong AM, et al. Primary cerebral lymphoma and glioblastoma multiforme: differences in diffusion characteristics evaluated with diffusion tensor imaging. AJNR Am J Neuroradiol. 2008;29: Harting I, Hartmann M, Jost G, et al. Differentiating primary central nervous system lymphoma from glioma in humans using localised proton magnetic resonance spectroscopy. Neurosci Lett. 2003; 342: Barajas RF Jr, Rubenstein JL, Chang JS, Hwang J, Cha S. Diffusion-weighted MR imaging derived apparent diffusion coefficient is predictive of clinical outcome in primary central nervous system lymphoma. AJNR Am J Neuroradiol. 2010;31: Ogawa T, Kanno I, Hatazawa J, et al. Methionine PET for followup of radiation therapy of primary lymphoma of the brain. Radiographics. 1994;14: Palmedo H, Urbach H, Bender H, et al. FDG-PET in immunocompetent patients with primary central nervous system lymphoma: correlation with MRI and clinical follow-up. Eur J Nucl Med Mol Imaging. 2006;33: Chan CC, Rubenstein JL, Coupland SE, et al. Primary vitreoretinal lymphoma: a report from an International Primary Central Nervous System Lymphoma Collaborative Group symposium. Oncologist. 2011;16: Abrey LE, Batchelor TT, Ferreri AJ, et al. Report of an international workshop to standardize baseline evaluation and response criteria for primary CNS lymphoma. J Clin Oncol. 2005;23: Cancer November 15, 2017

10 Primary CNS Lymphoma/Han and Batchelor 23. Mohile NA, Deangelis LM, Abrey LE. The utility of body FDG PET in staging primary central nervous system lymphoma. Neuro Oncol. 2008;10: Jahnke K, Hummel M, Korfel A, et al. Detection of subclinical systemic disease in primary CNS lymphoma by polymerase chain reaction of the rearranged immunoglobulin heavy-chain genes. J Clin Oncol. 2006;24: Scott BJ, Douglas VC, Tihan T, Rubenstein JL, Josephson SA. A systematic approach to the diagnosis of suspected central nervous system lymphoma. JAMA Neurol. 2013;70: Ferreri AJ, Blay JY, Reni M, et al. Prognostic scoring system for primary CNS lymphomas: the International Extranodal Lymphoma Study Group experience. J Clin Oncol. 2003;21: Abrey LE, Ben-Porat L, Panageas KS, et al. Primary central nervous system lymphoma: the Memorial Sloan-Kettering Cancer Center prognostic model. J Clin Oncol. 2006;24: Weller M, Martus P, Roth P, Thiel E, Korfel A; German PCNSL Study Group. Surgery for primary CNS lymphoma? Challenging a paradigm. Neuro Oncol. 2012;14: Reni M, Ferreri AJ, Garancini MP, Villa E. Therapeutic management of primary central nervous system lymphoma in immunocompetent patients: results of a critical review of the literature. Ann Oncol. 1997;8: Milgrom SA, Yahalom J. The role of radiation therapy in the management of primary central nervous system lymphoma. Leuk Lymphoma. 2015;56: Correa DD, Shi W, Abrey LE, et al. Cognitive functions in primary CNS lymphoma after single or combined modality regimens. Neuro Oncol. 2012;14: Hottinger AF, DeAngelis LM, Yahalom J, Abrey LE. Salvage whole brain radiotherapy for recurrent or refractory primary CNS lymphoma. Neurology. 2007;69: Chari DM, Huang WL, Blakemore WF. Dysfunctional oligodendrocyte progenitor cell (OPC) populations may inhibit repopulation of OPC depleted tissue. J Neurosci Res. 2003;73: Lai R, Abrey LE, Rosenblum MK, DeAngelis LM. Treatmentinduced leukoencephalopathy in primary CNS lymphoma: a clinical and autopsy study. Neurology. 2004;62: Monje ML, Vogel H, Masek M, Ligon KL, Fisher PG, Palmer TD. Impaired human hippocampal neurogenesis after treatment for central nervous system malignancies. Ann Neurol. 2007;62: Batchelor T, Carson K, O Neill A, et al. Treatment of primary CNS lymphoma with methotrexate and deferred radiotherapy: a report of NABTT J Clin Oncol. 2003;21: Gerstner ER, Carson KA, Grossman SA, Batchelor TT. Long-term outcome in PCNSL patients treated with high-dose methotrexate and deferred radiation. Neurology. 2008;70: Herrlinger U, Kuker W, Uhl M, et al. NOA-03 trial of high-dose methotrexate in primary central nervous system lymphoma: final report. Ann Neurol. 2005;57: Ferreri AJ, 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- e Ferreri AJ, 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: Morris PG, Correa DD, Yahalom J, et al. Rituximab, methotrexate, procarbazine, and vincristine followed by consolidation reduced-dose whole-brain radiotherapy and cytarabine in newly diagnosed primary CNS lymphoma: final results and long-term outcome. J Clin Oncol. 2013;31: Glass J, Won M, Schultz CJ, et al. Phase I and II study of induction chemotherapy with methotrexate, rituximab, and temozolomide, followed by whole-brain radiotherapy and postirradiation temozolomide for primary CNS lymphoma: NRG Oncology RTOG J Clin Oncol. 2016;34: Thiel E, Korfel A, Martus P, et al. High-dose methotrexate with or without whole brain radiotherapy for primary CNS lymphoma (G- PCNSL-SG-1): a phase 3, randomised, non-inferiority trial. Lancet Oncol. 2010;11: Korfel A, Thiel E, Martus P, et al. Randomized phase III study of whole-brain radiotherapy for primary CNS lymphoma. Neurology. 2015;84: Swerdlow SH. Lymphoma classification and the tools of our trade: an introduction to the 2012 USCAP long course. Mod Pathol. 2013; 26(suppl 1):S1-S Rubenstein JL, Hsi ED, Johnson JL, et al. Intensive chemotherapy and immunotherapy in patients with newly diagnosed primary CNS lymphoma: CALGB (Alliance 50202). J Clin Oncol. 2013;31: Illerhaus G, Fritsch K, Egerer G, et al. Sequential high dose immuno-chemotherapy followed by autologous peripheral blood stem cell transplantation for patients with untreated primary central nervous system lymphoma a multicentre study by the Collaborative PCNSL Study Group Freiburg. Blood. 2012;120: Omuro A, Correa DD, DeAngelis LM, et al. R-MPV followed by high-dose chemotherapy with TBC and autologous stem-cell transplant for newly diagnosed primary CNS lymphoma. Blood. 2015; 125: Kiefer T, Hirt C, Spath C, et al. Long-term follow-up of high-dose chemotherapy with autologous stem-cell transplantation and response-adapted whole-brain radiotherapy for newly diagnosed primary CNS lymphoma: results of the multicenter Ostdeutsche Studiengruppe Hamatologie und Onkologie OSHO-53 phase II study. Ann Oncol. 2012;23: Kasenda B, Ferreri AJ, Marturano E, et al. First-line treatment and outcome of elderly patients with primary central nervous system lymphoma (PCNSL) a systematic review and individual patient data meta-analysis. Ann Oncol. 2015;26: Omuro A, Chinot O, Taillandier L, et al. Methotrexate and temozolomide versus methotrexate, procarbazine, vincristine, and cytarabine for primary CNS lymphoma in an elderly population: an intergroup ANOCEF-GOELAMS randomised phase 2 trial. Lancet Haematol. 2015;2:e251-e Illerhaus G, Marks R, Muller F, et al. High-dose methotrexate combined with procarbazine and CCNU for primary CNS lymphoma in the elderly: results of a prospective pilot and phase II study. Ann Oncol. 2009;20: Fritsch K, Kasenda B, Hader C, et al. Immunochemotherapy with rituximab, methotrexate, procarbazine, and lomustine for primary CNS lymphoma (PCNSL) in the elderly. Ann Oncol. 2011;22: Roth P, Martus P, Kiewe P, et al. Outcome of elderly patients with primary CNS lymphoma in the G-PCNSL-SG-1 trial. Neurology. 2012;79: Langner-Lemercier S, Houillier C, Soussain C, et al. Primary CNS lymphoma at first relapse/progression: characteristics, management, and outcome of 256 patients from the French LOC network. Neuro Oncol. 2016;18: Pentsova E, Deangelis LM, Omuro A. Methotrexate re-challenge for recurrent primary central nervous system lymphoma. J Neurooncol. 2014;117: Plotkin SR, Betensky RA, Hochberg FH, et al. Treatment of relapsed central nervous system lymphoma with high-dose methotrexate. Clin Cancer Res. 2004;10: Soussain C, Hoang-Xuan K, Taillandier L, et al. Intensive chemotherapy followed by hematopoietic stem-cell rescue for refractory and recurrent primary CNS and intraocular lymphoma: Societe Francaise de Greffe de Moelle Osseuse-Therapie Cellulaire. J Clin Oncol. 2008;26: Nguyen PL, Chakravarti A, Finkelstein DM, Hochberg FH, Batchelor TT, Loeffler JS. Results of whole-brain radiation as salvage of methotrexate failure for immunocompetent patients with primary CNS lymphoma. J Clin Oncol. 2005;23: Nayak L, Iwamoto FM, LaCasce A, et al. PD-1 blockade with nivolumab in relapsed/refractory primary central nervous system and testicular lymphoma. Blood. 2017;129: Cancer November 15,