Cognitive functions in primary CNS lymphoma after single or combined modality regimens

Transcription

1 Neuro-Oncology 14(1): , doi: /neuonc/nor186 Advance Access publication October 19, 2011 NEURO-ONCOLOGY Cognitive functions in primary CNS lymphoma after single or combined modality regimens Denise D. Correa, Weiji Shi, Lauren E. Abrey, Lisa M. DeAngelis, Antonio M. Omuro, Mariel B. Deutsch, and Howard T. Thaler Departments of Neurology (D.D.C., L.E.A., L.M.D., A.M.O., M.B.D.); and Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, NY (W.S., H.T.T.) The standard treatment for primary CNS lymphoma (PCNSL) involves high-dose methotrexate-based chemotherapy (HD-MTX) alone or in combination with whole brain radiotherapy (WBRT). The combined modality regimen carries a substantial risk for cognitive impairment, and HD-MTX alone has been used more often recently in part to reduce neurotoxicity. In this study, we assessed cognitive functioning and quality of life in PCNSL survivors treated with WBRT 1 HD- MTX or HD-MTX alone. Fifty PCNSL patients in disease remission underwent a posttreatment baseline neuropsychological evaluation, and a subset of patients completed a follow-up evaluation. Quality of life and extent of white matter disease and atrophy on MRI were assessed. Comparisons according to treatment type after controlling for age and time since treatment completion showed that patients treated with HD- MTX alone had significantly higher scores on tests of selective attention and memory than patients treated with the combined modality regimen. Patients treated with WBRT 1 HD-MTX had impairments across most cognitive domains, and these were of sufficient severity to interfere with quality of life, as over 50% were not working due to their illness. Patients treated with HD-MTX alone did not meet criteria for cognitive impairment but scored within 1 SD below the normative sample on most tests. Patients with more extensive white matter disease had lower scores on tests of set-shifting and memory. Cognitive dysfunction was more prevalent in PCNSL survivors treated with WBRT 1 HD-MTX compared with patients treated with HD-MTX alone. Received May 24, 2011; accepted September 2, Corresponding Author: Denise D. Correa, PhD, ABPP-CN, Department of Neurology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY (corread@mskcc.org). Keywords: cognitive, methotrexate, neuropsychology, primary CNS lymphoma, radiation. The standard treatment for primary CNS lymphoma (PCNSL) often involves high-dose methotrexate-based chemotherapy (HD-MTX) and whole brain radiotherapy (WBRT). Although this treatment prolongs survival, 1,2 there is a risk of neurotoxicity that increases with age and in patients with prolonged disease-free survival. 3 Radiotherapy often produces damage to the CNS through vascular injury, neuronal loss, inflammation, demyelination, and necrosis 4 and can disrupt hippocampal neurogenesis. 5 Methotrexate and WBRT can each cause CNS damage, but there is synergistic toxicity when these modalities are combined. 4 To avoid the WBRT neurotoxic effects, studies have explored the use of HD-MTX alone, particularly in patients who are 60 years of age and older. 1,6,7 Recent studies suggest that HD-MTX alone can be efficacious and reduce delayed neurotoxicity 8 11 ; however, disease relapse is relatively common, and many patients require salvage therapy with WBRT or high-dose chemotherapy. 8,12 Delayed treatment related cognitive dysfunction has been recognized as a frequent and significant complication, as effective treatment for PCNSL has increased survival rates 1,13,14 and often interferes with the patient s ability to function at prediagnosis levels. In PCNSL studies that assessed cognitive functions after treatment with WBRT + HD-MTX, the majority of patients had cognitive impairment. 15 Prospective studies involving patients treated with HD-MTX alone reported either stable or improved cognitive performance in most patients. 9,15 As chemotherapy-alone regimens have been used more frequently, it is relevant to assess the possible cognitive adverse effects of this therapy, particularly in the elderly. In an earlier cross-sectional study, 16 we described cognitive outcome in a subset of PCNSL survivors # The Author(s) Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For permissions, please journals.permissions@oup.com.

2 treated with WBRT + HD-MTX (n ¼ 18) or HD-MTX alone (n ¼ 10). Patients treated with the combined modality therapy had impairments across several cognitive domains, while patients treated with chemotherapy alone had more preserved cognitive functions despite being significantly older. However, the relatively small sample size, particularly in the chemotherapy-alone group, was a limitation. We expanded our sample of PCNSL survivors and report the findings on cognitive outcome and quality of life (QoL) in a larger cohort of survivors, with approximately equal numbers of patients treated with HD-MTX in combination with the standard dose WBRT or with HD-MTX alone. Subjects Materials and Methods Ninety-five patients who completed treatment for PCNSL were identified from a cohort of survivors followed in the Department of Neurology at Memorial Sloan-Kettering Cancer Center between 1985 and Patients were contacted either by letter or by their treating physician. Among these patients, 15 were deceased, 17 were lost to follow-up, 6 were not eligible due to disease progression, and 7 refused to participate. Fifty (50) PCNSL patients treated with chemotherapy alone or in combination with WBRT, who were in remission as defined by negative MRI cerebrospinal fluid cytology, had no psychiatric or other neurological disorders, and were fluent in English participated in the study. Information on a subset of these patients was described previously. 16 Conventional doses were used for WBRT; patients treated with reduced-dose WBRT or HD-MTX with stem-cell transplant as part of ongoing clinical trials were not included in this study. The research protocol was approved by the institutional review board, and informed consent was obtained from all participants. All 50 patients completed a posttreatment baseline neuropsychological evaluation, and 33 patients were available for a follow-up cognitive assessment. Among the 17 patients who did not return to complete the follow-up visit, 7 had disease progression, 3 were in nursing homes, 3 were lost to follow-up due to relocation, and 4 refused to return for the cognitive assessment. Among the nursing home patients, 2 had received combined modality therapy and 1 had had chemotherapy alone. Measures Neuropsychological tests with documented sensitivity to the adverse effects of cancer therapy in PCNSL 15,16 were selected to evaluate auditory attention (Digit Span of the WMS-III - DF/DB), selective attention (Brief Test of Attention [BTA]), executive function (Trail Making Test Parts A and B [TMTA, TMTB]; Phonemic Verbal Fluency Test [VF]), motor speed (Grooved Pegboard Test [GP; Dominant and Nondominant Hand]), and memory (Hopkins Verbal Learning Test Revised, Learning [HVLT-R Learning, Delayed Recall]; HVLT Delay, Discrimination Index [HVLT Discrimination]). Mood and QoL were assessed using the Beck Depression Inventory (BDI) and the Functional Assessment of Cancer Therapy Brain (FACT-BR, version 4). The test battery was administered either by a neuropsychologist (D.D.C.) or by a trained research assistant. Alternate test forms were used at follow-up when available. Raw cognitive test scores were compared with published normative values according to age (and, when available, to age and education) and converted into z-scores. Cognitive impairment was defined as 2 or more z-scores 1.5 SD below the normative mean, consistent with other studies. 17 Neuroimaging White matter (WM) disease and cortical global atrophy (GA) were rated on brain MRI scans performed within a maximum of 3 months of each cognitive evaluation. The ratings were performed by neurologists who were blind to the cognitive test results. WM abnormalities were rated on a fluid-attenuated inversion recovery sequence for most patients, and if not available, T2-weighted sequences were used. Radiographic endpoints were measured according to the modified Fazekas scale 18 and included: no WM change (grade 0), minimal patchy WM foci (grade 1), start of confluence of WM disease (grade 2), large confluent areas (grade 3), confluence with cortical and subcortical involvement (grade 4), leukoencephalopathy (grade 5), and possible radiation necrosis (grade 6). GA was rated on T1-weighted MRI scans based on a 4-point scale 19 and included: no atrophy (grade 0), mild atrophy (grade 1), moderate atrophy (grade 2), and severe atrophy (grade 3). Statistical Analyses Analysis of covariance (ANCOVA) adjusting for age at baseline and time since treatment completion was used to assess whether cognitive test z-scores and QoL and mood scores differed between the 2 treatment groups. The relationship between specific demographic or clinical variables and cognitive test scores at posttreatment baseline was assessed using Spearman s correlation, t-tests, or a chi-square test. Paired t-tests, a marginal homogeneity exact test, or McNemar s test 20 were performed to compare cognitive test performance and MRI WM and GA ratings between the posttreatment baseline and follow-up evaluations. A Cochran Armitage trend test with Monte Carlo estimates for the exact test was used for group comparisons that involved ordinal clinical variables with small sample sizes. Analyses of cognitive test scores were performed with and without a Hochberg Benjamini multiplicity adjustment NEURO-ONCOLOGY JANUARY 2012

3 Results The patients were predominantly men (64%) and righthanded (82%) (Table 1). There were no significant differences between the 2 groups in education or estimated IQ, but patients treated with WBRT + HD-MTX were significantly younger than patients who received HD-MTX alone, t(48) ¼ 27.08, P, This is consistent with prior treatment practice in which most patients who were 60 years of age or older were treated with chemotherapy alone. An assessment of employment status at the time of enrollment showed that 16 patients (67%) treated with WBRT + HD-MTX were not working (13 were not employed due to their illness, 3 had retired prior to diagnosis). Twenty-one patients (81%) treated with HD-MTX alone were not working at study enrollment (17 had retired prior to diagnosis, 3 were not employed due to their illness, 1 was not working for other reasons). The 13 patients of both groups who were working at enrollment were employed in the same capacity as prior to diagnosis. Table 2 presents disease and treatment history of patients who completed the posttreatment baseline cognitive assessment according to therapy modality. Twenty-four patients had WBRT + HD-MTX and 26 patients had HD-MTX alone. The majority of patients received the HD-MTX regimen consisting of high-dose methotrexate, procarbazine and vincristine, and high-dose cytarabine (82% in the combined modality group, 65% in the chemotherapy-only group), whereas some patients received high-dose methotrexate alone or in combination with other agents. WBRT dose ranged from 3600 to 5940 cgy, and most patients received 4500 cgy (67%). Among the 10 patients who relapsed prior to enrollment, all had been treated initially with HD-MTX alone and at relapse, 9 Table 1. Patient demographic characteristics (means and standard deviations) Demographics Treatment Group 1 WBRT (n 5 24) Alone (n5 26) Sex (M/F) 17/7 15/11 Handedness (R/L) 17/7 24/2 Age at study baseline (yrs) Mean (SD) 53.5 (8.9) 70.6 (8.2)* Median (range) 52.0 (36 72) 72.5 (56 85) Mean education (yrs) 15.1 (3.3) 15.2 (3.7) Mean est.viq 110 (10.0) 113 (10.1) Occupation (prior to diagnosis) Professional 14 (58%) 16 (62%) Technical/clerical 10 (42%) 10 (38%) Occupation (at study baseline) Not working 16 (67%) 21 (81%) Working 8 (33%) 5 (19%) Abbreviations: Est.VIQ, estimated Verbal IQ (NAART or Barona Index). *P received salvage chemotherapy alone, and one had WBRT and chemotherapy. Compared with patients treated with chemotherapy alone, patients who received the combined modality regimen had significantly longer time intervals between diagnosis and the initial cognitive evaluation, t(48) ¼ 5.52, P,.0001, and between treatment completion and the initial cognitive evaluation, t(48) ¼ 4.29, P,.0001; this is consistent with the recent trend to eliminate WBRT in older patients. The calculation of time since completion of treatment was based on all therapy received, including treatment at relapse if applicable. Most patients (74%) were on antiepileptic medication at the time of enrollment, and a small number were on antidepressants or anxiolytics (22%). Posttreatment Baseline Analyses Patients treated with WBRT + HD-MTX had mean z-scores 1.5 SD below the normative mean on 7 of the 10 neuropsychological tests, including selective Table 2. Disease and treatment history Disease/Treatment History a Treatment Group 1 WBRT (n 5 24) Tumor location Frontal 4 4 Frontal-temporal 4 3 Temporal/ parietal/ occipital 5 6 Alone (n 5 26) cortical/ subcortical Predominant tumor side Left 6 10 Right 10 8 Bilateral 8 8 Time since diagnosis, mos Mean (SD) 73.2 (48.0) 23.2 (16.8)* Median (range) 71.1 (10 198) 16.5 (9 79) Time since treatment completion, mos Mean (SD) 50.5 (39.5) 14.9 (12.5)* Median (range) 42.5 (3 149) 10 (4 54) Relapse prior to study baseline (yes/no) 1/23 9/17 Biopsy/resection 11/13 7/19 Radiotherapy dose N/A (range, cgy) Antiepileptics 16/8 21/5 (yes/no) b Psychotropics (yes/no) b 5/19 6/20 a Treatment history reflects all therapy received, including treatment at relapse, if applicable. b Medication at the time of posttreatment baseline cognitive evaluation. *P,.0001 based on log base 2 transformed data. NEURO-ONCOLOGY JANUARY

4 attention (BTA), motor speed, sequencing and setshifting (GP, TMTA, TMTB), and memory (HVLT-R), and met the criteria for cognitive impairment. Patients treated with HD-MTX alone had mean z-scores ranging from 0.1 to 21.3 SD on the neuropsychological tests and did not meet criteria for cognitive impairment (Table 3; Figs 1 3). A comparison of cognitive test z-scores between treatment groups using ANCOVA with adjustments for age and time since treatment completion showed that patients who received HD-MTX alone had significantly higher mean test z-scores in selective attention (BTA), F(1, 43) ¼ 4.66, P,.04, and in memory, F(1, 46) ¼ 5.85, P,.02 for HVLT Learning, F(1, 45) ¼ 13.99, P ¼.0005 for HVLT Discrimination, compared with patients treated with WBRT + HD-MTX. A similar trend was seen on another attention test, but it did not reach significance (DB, P ¼.09) (Table 3; Figs 1 and 3). After Hochberg Benjamini adjustment for multiple testing, significant group differences remained in memory (HVLT Discrimination, P ¼.005). There were no significant differences in cognitive test scores between patients with and without a prior history of relapse. There were no significant correlations between cognitive test performance and anticonvulsant use or vascular risk factors. A comparison of mean scores on a self-report measure of QoL (FACT-BR) between treatment groups using ANCOVA with adjustments for age and time since treatment completion showed that patients who received HD-MTX alone had significantly lower mean scores, F(1, 45) ¼ 9.93, P,.003, compared with Table 3. Cognitive test z-scores, mood and quality of life (means and SD) Cognitive Tests Treatment 1 WBRT (n 5 24) Alone (n 5 26) DF 20.5 (1.5) 0.2 (1.2) DB 20.3 (1.3) 0.1 (0.9) TMTA 21.6 (1.1) 20.8 (1.4) TMTB 21.6 (1.1) 21.0 (1.3) GP 22.3 (0.8) 21.3 (1.0) BTA 21.7 (1.5) 20.9 (1.2)* VF 21.0 (1.4) 20.7 (1.1) HVLT Learning 22.0 (1.2) 21.2 (1.3)* HVLT Delay 21.8 (1.6) 21.1 (1.2) HVLT Discrimination 21.5 (1.4) 20.4 (1.2)* BDI (raw score) 7.0 (4.3) 9.6 (6.4) FACT-BR (raw score) (21.2) (29.1)* Abbreviations: DF, Digit Span Forward; DB, Digit Span Backward; BTA, Brief Test of Attention; TMTA, Trail Making Test, Part A; TMTB, Trail Making Test, Part B; GP, Grooved Pegboard, a composite mean score of GP Dominant and GP Nondominant; VF, Verbal Fluency Test; HVLT, Hopkins Verbal Learning Test Learning; BDI, Beck Depression Inventory; FACT-BR, Functional Assessment of Cancer Therapy Brain. *P,.05. patients treated with WBRT + HD-MTX (low scores indicate fewer complaints). The 2 groups did not differ significantly on a measure of depression (BDI), and mean scores did not indicate depressed mood (Table 3). Neuroimaging WM and GA ratings on MRIs performed within 3 months of the baseline cognitive evaluation are reported in Table 4 and 5. More extensive WM abnormalities (grades 2 6) were evident in 67% of patients who received WBRT + HD-MTX and in 54% of patients treated with chemotherapy alone. Moderate to severe GA (grades 2 3) was evident in 44% of patients who received WBRT + HD-MTX and in 54% of patients treated with chemotherapy alone. There were no significant differences between the 2 treatment groups on extent of WM disease or GA. Student s t-test with Hochberg Benjamini adjustment for multiple testing showed that patients with minimal or no WM disease (grades 0 1) had higher mean z-scores on the TMTB, t(42) ¼ 2.82, P ¼.03; HVLT Learning, t(48) ¼ 2.72, P ¼.03; and HVLT Delay, t(48) ¼ 2.83, P ¼.03 compared with patients with more extensive WM disease (grades 2 6). A similar trend was seen on the HVLT Discrimination Index, but it did not reach significance (P ¼.07). Student s t-tests with Hochberg Benjamini adjustment for multiple testing showed no significant differences in any of the cognitive test z-scores between patients with minimal or no GA (grades 0 1) and patients with moderate to severe GA (grades 2 3). Follow-up Analyses Thirty-three patients completed an additional follow-up cognitive evaluation. The mean time interval between the 2 evaluations was 16.2 months (SD ¼ 18.3) for patients treated with the combined modality regimen (n ¼ 15), and 14.1 months (SD ¼ 12.0) for patients treated with chemotherapy alone (n ¼ 18). Student s t-test with Hochberg Benjamini adjustment for multiple testing showed that patients who received HD-MTX alone obtained a significantly higher score on a test of attention (DSF), t(17) ¼ 3.36, P ¼.04, at follow-up. There were no significant changes between baseline and follow-up on any of the cognitive tests among patients treated with WBRT + HD-MTX or in the FACT-BR or BDI scores for either treatment group. There were no significant differences on any of the baseline cognitive test z-scores between the patients who did and did not return for follow-up in either treatment group (n ¼ 9, combined modality regimen; n ¼ 8, chemotherapy alone). A comparison of WM ratings between the posttreatment baseline and follow-up using a marginal homogeneity exact test showed that there was an increase in WM disease among patients who received WBRT + HD-MTX (P,.02). There were no significant changes in WM ratings between baseline and follow-up among 104 NEURO-ONCOLOGY JANUARY 2012

5 Fig. 1. Mean cognitive z-scores at baseline by treatment groups: Attention. Fig. 2. Mean cognitive z-scores at baseline by treatment groups: Motor speed/executive. patients treated with HD-MTX alone or in GA ratings for either treatment modality group. Discussion In this cross-sectional study, PCNSL patients treated with HD-MTX alone had higher scores than patients who received WBRT + HD-MTX across all neuropsychological tests, including attention, executive function, motor speed, and memory. After controlling for the effects of age and time since treatment completion, significant group differences were seen in selective attention and memory functions. These results support studies reporting that treatment with HD-MTX alone 9 is not associated with severe cognitive deficits in PCNSL patients, even among the elderly. Although patients treated with chemotherapy alone did not meet criteria for cognitive impairment, mean test scores were within 1 SD below the normative sample on most measures, including selective attention, motor speed, executive function, and verbal learning and delayed recall. These mild cognitive decrements are consistent with the described neurotoxicity associated with chemotherapy, particularly methotrexate and cytarabine 22 ; however, we cannot determine to what extent some of these difficulties could be disease related given the cross-sectional design. Patients treated with WBRT + HD-MTX had impairment across most cognitive tests, including selective attention, motor speed, set-shifting, verbal learning, delayed recall, and recognition memory. These findings are in agreement with studies that report a diffuse NEURO-ONCOLOGY JANUARY

6 Fig. 3. Mean cognitive z-scores at baseline by treatment groups: Verbal memory. Table 4. MRI white matter (WM) abnormalities Table 5. MRI Global Atrophy MRI Grade Treatment Atrophy Rating Treatment Minimal or no WM change (grades 0, 1) Start of confluence (grade 2) Large confluent areas (grade 3) Confluence of cortical/ subcortical; Leukoencephalopathy (grades 4, 5) 1 WBRT (n 5 24) Alone (n5 26) 8 (33%) 12 (46%) 7 (29%) 6 (23%) 5 (21%) 7 (27%) 4 (17%) 1 (4%) pattern of moderate to severe neuropsychological deficits in brain tumor patients treated with WBRT alone or in combination with chemotherapy, 15,23 and with the delayed neurotoxicity associated with combined modality regimens. 4,11,12 There were no significant changes between the initial and follow-up evaluations on any of the cognitive tests, except for an improvement in attention among patients treated with HD-MTX alone, suggesting no evidence of further cognitive decline within the relatively short follow-up period. Consistent with the cognitive findings, patients treated with HD-MTX alone had significantly lower scores than patients treated with WBRT + HD-MTX on the FACT-BR after controlling for the effects of age and time since treatment completion, indicating more intact QoL. There was no evidence of depression in either treatment group, suggesting that mood 1 WBRT (n 5 23) Alone (n 5 26) Minimal or none 13 (57%) 12 (46%) (grades 0, 1) Moderate (grade 2) 8 (35%) 13 (50%) Severe (grade 3) 2 (9%) 1 (4%) disturbance was not a contributing factor to the cognitive test performance. Additional evaluation of QoL revealed that more than 50% of patients treated with WBRT + HD-MTX were not employed due to their disease and treatment. This finding supports observations that treatment-related cognitive deficits can be of sufficient severity to interfere with QoL in brain tumor patients who are in remission from their disease. 24 The majority of patients treated with HD-MTX alone had retired prior to diagnosis; therefore, an assessment of work status was not applicable. Extensive WM abnormalities were documented in more than 60% of patients treated with WBRT + HD-MTX and in more than 50% of patients treated with HD-MTX alone, and there were no significant differences between the groups. Patients with more extensive WM disease had greater impairments in setshifting, learning, and delayed recall of information than patients with either no or minimal WM disease. In addition, there was an increase in WM disease between the initial and the follow-up evaluation among patients who received WBRT + HD-MTX, consistent with the reported progressive delayed 106 NEURO-ONCOLOGY JANUARY 2012

7 neurotoxicity associated with WBRT alone or in combination with MTX. 4 These findings support studies documenting WM disease in brain tumor patients who received radiotherapy and chemotherapy 11 and the moderate association between WM changes and cognitive impairment found in some but not all studies. 25,26 Atrophy was evident in approximately 40% of patients who received WBRT + HD-MTX and in 50% of patients treated with chemotherapy alone. There were no significant differences between the 2 groups in atrophy ratings or in cognitive test performance between patients with either minimal or no atrophy and patients with moderate to severe atrophy. The higher number of patients with atrophy in the HD-MTX alone group may be in part related to their older age. Limitations of this study include the cross-sectional design and absence of a pretreatment baseline to evaluate the specific contribution of disease and treatment adverse effects to cognitive outcome. Possible selection bias needs to be considered, as patients were not randomly assigned to treatment type. Group differences in age and time since treatment completion are also confounders, although there were significant group differences in cognitive performance after controlling for these variables. Attrition bias may have reduced the power to detect changes at follow-up and is a significant limitation in longitudinal studies of brain tumor patients. The significantly longer time since treatment completion in patients who received WBRT + HD-MTX makes it difficult to determine whether the combined regimen, as well as its more delayed effects, may have contributed to the impaired performance in these patients. Preliminary findings from a prospective study of PCNSL patients treated with an HD-MTX based regimen followed by dose-reduced WBRT and cytarabine 27 showed pretreatment impairments in motor speed, executive function, and memory, with patients improving after induction chemotherapy and remaining relatively stable for up to 2 years posttreatment. 28 These data suggest that WBRT dose may play an important role in contributing to the cognitive impairment associated with the combined regimen. The results of this study confirmed our previous findings 16 and provided further evidence that PCNSL patients treated with WBRT + HD-MTX develop diffuse cognitive impairments of sufficient severity to interfere with QoL, while patients treated with HD-MTX alone had relatively more intact cognitive functions, at least at short posttherapy intervals. The importance of including standardized and sensitive cognitive tests in prospective clinical trials of PCNSL patients has been recognized. 15,29 Randomized collaborative studies including neuropsychological assessment are needed to evaluate the relationship between disease control, treatment modality, survival, cognitive functions, and QoL in order to guide treatment choices. Moreover, as cognitive dysfunction often interferes with QoL, including the ability to return to work, the development and early implementation of appropriate intervention strategies to minimize or prevent these adverse effects are needed. Recent studies have begun to explore potential pharmacological and rehabilitation strategies for treatment-related neurotoxicity in brain tumor patients. 30 Conflict of interest statement. Genentech Scientific Advisory Board (Drs. Correa, DeAngelis, Omuro). Hoffmann-La Roche Inc, Merck, Curatio CME, Educational Concepts, Inc (Dr. Abrey). Funding Funding for this work was provided by grant R03 # CA from the National Cancer Institute to Dr. Denise D. Correa. Presented in part at the American Academy of Neurology Annual Meeting, 2002, the International Neuropsychological Society Annual Meeting, 2003, and the Society for Neuro-Oncology Annual Meeting, References 1. Abrey LE, DeAngelis LM, Yahalom J. Long-term survival in primary CNS lymphoma. J Clin Oncol. 1998;16: DeAngelis LM, Seiferheld W, Schold SC, et al. Combination chemotherapy and radiotherapy for primary central nervous system lymphoma: Radiation Therapy Oncology Group Study J Clin Oncol. 2002;20: Omuro AM, Ben-Porat LS, Panageas KS, et al. Delayed neurotoxicity in primary central nervous system lymphoma. Arch Neurol. 2005;62: DeAngelis L, Posner JB. Side effects of radiation therapy. In: DeAngelis, L, Posner, JB, eds. Neurologic Complications of Cancer. 2nd ed. New York: Oxford University Press, 2009: Monje ML, Vogel H, Masek M, et al. Impaired human hippocampal neurogenesis after treatment for central nervous system malignancies. Ann Neurol. 2007;62: 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: Omuro A, Taillandier L, Chinot O, et al. Primary CNS lymphoma in patients younger than 60: can whole-brain radiotherapy be deferred? J Neurooncol. 2010;104: 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: Juergens A, Pels H, Rogowski S, et al. Long-term survival with favorable cognitive outcome after chemotherapy in primary central nervous system lymphoma. Ann Neurol. 2010;67: 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: NEURO-ONCOLOGY JANUARY

8 11. 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: Ney DE, Reiner AS, Panageas KS, et al. Characteristics and outcomes of elderly patients with primary central nervous system lymphoma: the Memorial Sloan-Kettering Cancer Center experience. Cancer. 2010;116: Poortmans PM, Kluin-Nelemans HC, Haaxma-Reiche H, et al. High-dose methotrexate-based chemotherapy followed by consolidating radiotherapy in non-aids-related primary central nervous system lymphoma: European Organization for Research and Treatment of Cancer Lymphoma Group Phase II Trial J Clin Oncol. 2003;21: Behin A, Delattre J-Y. Neurologic sequelae of radiotheray on the nervous system. Totowa, New Jersey: Humana Press Inc., Correa DD, Maron L, Harder H, et al. Cognitive functions in primary central nervous system lymphoma: literature review and assessment guidelines. Ann Oncol. 2007;18: Correa DD, DeAngelis LM, Shi W, et al. Cognitive functions in survivors of primary central nervous system lymphoma. Neurology. 2004;62: Wefel JS, Lenzi R, Theriault RL, et al. The cognitive sequelae of standard-dose adjuvant chemotherapy in women with breast carcinoma: results of a prospective, randomized, longitudinal trial. Cancer. 2004;100: Fazekas F, Chawluk JB, Alavi A, et al. MR signal abnormalities at 1.5 T in Alzheimer s dementia and normal aging. AJR Am J Roentgenol. 1987;149: Douw L, Klein M, Fagel SS, et al. Cognitive and radiological effects of radiotherapy in patients with low-grade glioma: long-term follow-up. Lancet Neurol. 2009;8: Kuritz SJ, Landis JR, Koch GG. A general overview of Mantel Haenszel methods: applications and recent developments. Annu Rev Public Health. 1988;9: Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B. 1995;57: Dietrich J, Monje M, Wefel J, Meyers C. Clinical patterns and biological correlates of cognitive dysfunction associated with cancer therapy. Oncologist. 2008;13: Taphoorn MJ, Klein M. Cognitive deficits in adult patients with brain tumours. Lancet Neurol. 2004;3: Spelten ER, Sprangers MA, Verbeek JH. Factors reported to influence the return to work of cancer survivors: a literature review. Psychooncology. 2002;11: Postma TJ, Klein M, Verstappen CC, et al. Radiotherapy-induced cerebral abnormalities in patients with low-grade glioma. Neurology. 2002;59: Surma-aho O, Niemela M, Vilkki J, et al. Adverse long-term effects of brain radiotherapy in adult low-grade glioma patients. Neurology. 2001;56: Shah GD, Yahalom J, Correa DD, et al. Combined immunochemotherapy with reduced whole-brain radiotherapy for newly diagnosed primary CNS lymphoma. J Clin Oncol. 2007;25: Correa DD, Rocco-Donovan M, DeAngelis LM, et al. Prospective cognitive follow-up in primary CNS lymphoma patients treated with chemotherapy and reduced-dose radiotherapy. J Neurooncol. 2009;91: 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: Gehring K, Sitskoorn MM, Aaronson NK, Taphoorn MJ. Interventions for cognitive deficits in adults with brain tumours. Lancet Neurol. 2008;7: NEURO-ONCOLOGY JANUARY 2012