Rituximab significantly improves complete response ... - Springer Link

3 downloads 124 Views 206KB Size Report
May 9, 2012 - MTX and ifosfamide (IFO) improves the complete response (CR) of patients with newly diagnosed PCNSL. Patients and methods. Study design ...
J Neurooncol (2012) 109:285–291 DOI 10.1007/s11060-012-0891-7

CLINICAL STUDY

Rituximab significantly improves complete response rate in patients with primary CNS lymphoma Tobias Birnbaum • Elisabeth Anne Stadler • Louisa von Baumgarten • Andreas Straube

Received: 2 March 2012 / Accepted: 23 April 2012 / Published online: 9 May 2012 Ó Springer Science+Business Media, LLC. 2012

Abstract Rituximab improves outcome for patients with systemic diffuse large B-cell lymphoma (DLBCL) and has therefore become a standard component of the treatment of this disease. However, it is unclear whether rituximab is efficacious in patients with primary CNS lymphoma (PCNSL), a rare DLBCL variant, also. The purpose of this study was to evaluate the effect of rituximab on the complete response (CR) rate after chemotherapy with methotrexate (MTX) and ifosfamide (IFO) of patients with PCNSL. This is a retrospective, observational, single-center trial analyzing 36 consecutive patients with newly diagnosed, CD-20-positive PCNSL who were treated primarily with chemotherapy between March 2007 and December 2010. We compared 19 patients who were treated with MTX and IFO with 17 patients who were treated with the same regimen plus rituximab. The addition of rituximab to MTX and IFO was correlated with a significant increase in the CR rate (100.0 vs. 68.4 %, p = 0.02). Furthermore, six-month progression-free survival was significantly higher for the rituximab group (94.1 vs. 63.2 %, p = 0.04). Toxicity did not differ significantly between the groups. Our results indicate that rituximab might be efficacious in the treatment of PCNSL and support addition of this drug to current treatment protocols until data from randomized controlled trials becomes available. Immunochemotherapy with MTX, IFO, and rituximab seems to have excellent activity as induction chemotherapy and should be further tested in prospective trials.

T. Birnbaum (&)  E. A. Stadler  L. von Baumgarten  A. Straube Department of Neurology, Ludwig-Maximilians-University, Marchioninistr. 15, 81377 Munich, Germany e-mail: [email protected]

Keywords PCNSL

Rituximab  Methotrexate  Ifosfamide  NHL 

Introduction Primary central nervous system lymphoma (PCNSL) is a rare extranodal non-Hodgkin lymphoma (NHL) confined to the brain, spinal cord, leptomeninges, and eyes. No optimum therapy has yet been defined. Primary treatment usually consists of high-dose methotrexate (MTX)-based chemotherapy with or without adjuvant whole-brain irradiation (WBI). With such treatment regimens median overall survival of 15–60 months can be achieved [1]. Because WBI leads to high incidence of severe leukoencephalopathy with cognitive impairment, many efforts have been made to improve the efficacy of chemotherapy and most centers try to postpone radiation therapy until tumor progression or recurrence [2–4]. The monoclonal CD-20 antibody rituximab has proved excellent efficacy against a variety of systemic B-cell NHLs and has become a standard component of the treatment of these patients [5]. In contrast, the efficacy of rituximab against PCNSL is unclear. Approximately 95 % of PCNSL are of B-cell origin and are characterized by extensive expression of CD-20. It is well known that rituximab cannot pass the intact blood–brain barrier (BBB), because of its large molecular size; it may therefore be not efficacious against PCNSL. Current evidence of the effect of rituximab in PCNSL treatment mainly arises from a few case series or small single-arm trials; no controlled, randomized trials have been conducted. Despite this lack of evidence, in recent years rituximab has already been incorporated into PCNSL treatment protocols in most centers in the US and in Europe. The purpose of this study

123

286

was to elucidate whether addition of rituximab to high-dose MTX and ifosfamide (IFO) improves the complete response (CR) of patients with newly diagnosed PCNSL.

Patients and methods Study design and patients This was a retrospective, observational, single center trial analyzing 36 immunocompetent patients with newly diagnosed, histologically proved, CD-20-positive B-cell PCNSL. From March 2007 all patients in our institution who qualified for high-dose chemotherapy were primarily treated with MTX and IFO according to the protocol of the G-PCNSL-SG1 trial [2]. In June 2009 we added rituximab to this protocol. Radiotherapy was restricted to patients without CR to chemotherapy. We included in this study all patients who were treated with this chemotherapy regimen between March 2007 and December 2010. Exclusion criteria were HIV infection, lymphoma manifestation outside the CNS, and pure intraocular lymphoma.

J Neurooncol (2012) 109:285–291

performed by an experienced neuroradiologist who was unaware of the treatment groups. Treatment response was assessed in accordance with the ‘‘International PCNSL Collaborative Group’’ criteria [6]. PFS and OS were defined as time between histologic diagnosis and progression or death, respectively, or last follow-up. TTF was defined as time between treatment initiation and tumor progression, death, treatment discontinuation for any reason, or last follow-up. Toxicity was recorded in accordance with the NCI common terminology criteria for adverse events (version 3.0). Statistical analysis Statistical analysis was performed by use of Fisher0 s exact test, the Wilcoxon signed rank test, or the unpaired t test, as appropriate. The Kaplan–Meier method was used for univariate analysis of survival, with assessment of differences by use of the log-rank test. p \ 0.05 was considered to represent statistical significance. Data are presented as medians and range, unless otherwise noted. Statistical analysis was performed with ‘‘Prism 4 for Windows’’ (GraphPad Software, CA, USA).

Treatment Before the start of treatment all patients were evaluated by use of contrast-enhanced computed tomography total-body scan, contrast-enhanced magnetic resonance tomography (MRI) of the brain and spinal axis, lumbar puncture (if no signs of intracranial hypertension were evident on brain MRI), slit-lamp examination, bone-marrow biopsy and aspirate, and routine laboratory testing including creatinine clearance. Tht treatment protocol consisted of six cycles of chemotherapy which were repeated every 14 days. All patients received MTX 4 g/m2 (given over 4 h) on day 1 and IFO 1.5 g/m2 (given over 3 h) on days 3, 4, and 5. Dexamethasone (3 9 8 g) was given for 10 days during the first cycle only. Patients in the rituximab group were also treated with rituximab (375 mg/m2 over 120 min) on day 0 during the first three cycles. Measures of outcome The main measure of outcome was CR to primary chemotherapy. Furthermore, we evaluated six-month progression-free survival (PFS-6), median PFS, median time to treatment failure (TTF), median overall survival (OS), WBI rate, relapse rate, and toxicity of chemotherapy. Treatment response was evaluated after the third and the sixth chemotherapy cycle by contrast-enhanced brain MRI. For patients with lymphomatous meningitis or ocular involvement, lumbar puncture and/or slit-lamp examination was also performed. Appraisal of MRI scans was

123

Results Patient characteristics A total of 42 patients with B-cell PCNSL were treated in our institution between March 2007 and December 2010. Thirty-six of these patients were eligible for the study. Six patients could not be included or had to be excluded (one patient was treated primarily with WBI only, one patient was treated with MTX mono, one patient received supportive care only, two patients had a systemic NHL in their history, and one patient had pure intraocular lymphoma). Nineteen of the 36 patients were treated with MTX and IFO until May 2009 (control group). Seventeen of the 36 patients were treated with MTX, IFO and rituximab from June 2009 (rituximab group). Three patients of the control group were treated within the G-PCNSL-SG1 trial. Two had been randomized to receive no adjuvant WBI after chemotherapy, the third patient had been randomized into the WBI arm but declined adjuvant radiation therapy. Both groups did not differ significantly in demographics or prognosis (Table 1). Treatment course Fourteen of the 19 (73.7 %) patients of the control group and 12/17 (70.6 %) patients of the rituximab group completed all six pre-planned chemotherapy cycles. Five of the

J Neurooncol (2012) 109:285–291

287

Table 1 Comparison of baseline characteristics and demographic factors Male, n (%) Age, y, median (range)

CR complete response, uCR complete response, unconfirmed, PR partial response, SD stable disease, PD progressive disease, n.a. not assessable

p

11/19 (57.9)

11/17 (64.7)

0.74

66 (36–78)

66 (39–79)

10/19 (52.6)

9/17 (52.9)

Age [50 y, n (%)

17/19 (89.5)

15/17 (88.2)

50 (10–100)

0.65 1.00 1.00

50 (20–90)

0.61 1.00

MSKCC score 1, n (%)

3 (15.8)

2 (11.8)

MSKCC score 2, n (%)

6 (31.6)

5 (29.4)

1.00

MSKCC score 3, n (%)

10 (52.6)

10 (58.8)

0.75

Ocular involvement, n (%)

1/18 (5.6)

1/15 (6.7)

1.00

Lymphomatous meningitis, n (%)

2/18 (11.1)

3/15 (20.0)

0.64

10/18 (55.6) 11/18 (61.1)

11/15 (73.3) 12/15 (80.0)

0.47 0.28

CSF pleocytosis, n (%) CSF elevated protein, n (%)

19 (26.3 %) patients of the control group discontinued treatment prematurely: one patient (no staging) with an initial KPS of only 10 % did not benefit from treatment and was referred after cycle two to palliative supportive care. One patient had PD after the second cycle and therapy was changed to cytarabine (AraC). For one patient (uCR) treatment was changed to AraC and thiotepa, because of hepatic toxicity after cycle two. Two patients (CR) declined further treatment after cycles two and four, respectively. In the rituximab group, treatment was discontinued prematurely for 5/17 (29.4 %) patients: one patient (CR) did not benefit from treatment and was referred to palliative supportive care after cycle three. For two patients (uCR and CR, respectively) treatment was changed to AraC because of nephrotoxicity after cycles three and four, respectively. For one patient (CR) treatment was stopped after cycle five because of accumulative hematologic toxicity. One patient (uCR) declined any further chemotherapy after cycle four and she was referred to WBI. Table 2 Comparison of response to chemotherapy, PFS6, WBI, and relapse rate of patients with CR to chemotherapy without WBI

Rituximab group (MTX ? IFO ? R)

Age [65 y, n (%) KPS, median (range)

Y years, KPS Karnofsky performance status, MSKCC Memorial Sloan Kettering Cancer Center, CSF cerebrospinal fluid

Control group (MTX ? IFO)

Response to chemotherapy and follow-up In the control group, 13/19 (68.4 %) patients achieved CR/ uCR, 4/19 (21.1 %) patients achieved PR, one patient (5.3 %) had PD, and one patient was not evaluable for response assessment (Table 2). In contrast, in the rituximab group, 17/17 (100 %) patients achieved CR/uCR. CR/uCR proved to be significantly different between the groups (p = 0.02). For all nine patients with uCR (two in control group, seven in the rituximab group) there was no evidence of tumor progression or recurrence in follow-up MRI 4–8 weeks after termination of treatment. This was also supported by significantly higher PFS-6 in the rituximab group (63.2 vs. 94.1 %, p = 0.04). Five of 19 (26.3 %) patients in the control group (39 PR and 19 PD after chemotherapy, 19 uCR but hepatic toxicity) and 1/17 (5.9 %) patients in the rituximab group (19 uCR, declined further chemotherapy) were treated with adjuvant WBI (p = 0.18). Of those patients who had

Control group (MTX ? IFO)

Rituximab group (MTX ? IFO ? R)

p

0.02

Response to chemotherapy CR/uCR, n (%)

13/19 (68.4)

17/17 (100)

CR, n (%) uCR, n (%)

11/19 (57.9) 2/19 (10.5)

10/17 (58.8) 7/17 (41.2)

PR, n (%)

4/19 (21.1)

0/17 (0)

SD, n (%)

0/19 (0)

0/17 (0)

PD, n (%)

1/19 (5.3)

0/17 (0)

n.a. (%)

1/19 (5.3)

PFS-6 (%)

12/19 (63.2)

0/17 (0) 16/17 (94.1)

0.04

WBI adjuvant, n (%)

5/19 (26.3)

1/17 (5.9)

0.18

WBI anytime, n (%)

8/19 (42.1)

5/17 (29.4)

0.5

Relapses following CR to chemotherapy only, n (%)

6/12 (50)

8/16 (50)

123

288

J Neurooncol (2012) 109:285–291

achieved CR to primary treatment without WBI, 6/12 (50 %) patients in the control group and 8/16 (50 %) patients in the rituximab group relapsed during follow-up. After a median follow-up time of 30 months (2–54) in the control group and 18 months (3–32) in the rituximab group we found no significant difference in median TTF (6 vs. 11 months, hazard ratio (HR) 1.21, 95 % confidence interval (95 % CI) 0.54–2.83, p = 0.62), median PFS (9 vs. 18 months, HR 1.14, 95 % CI 0.48–2.76, p = 0.75), and median OS (both not reached, HR 1.88, 95 % CI 0.52–6.94, p = 0.33) (see Fig. 1). Toxicity Overall, toxicity of chemotherapy was not significantly different between the groups (Table 3). Treatment with MTX and IFO was associated with high hematologic toxicity grades III and IV. Hepatic toxicity grade III or IV was evident in six patients. There was no case of nephrotoxicity grade III or IV in either group. Treatment had to be stopped prematurely because of toxicity more often in the rituximab group (3 vs. 1 patient) but this difference was not significant. There was no treatment-associated death in either group.

Discussion As the main result of this trial we found that addition of rituximab to high-dose MTX and IFO was associated with a significantly increased CR/uCR for patients with newlydiagnosed PCNSL. Of course, several limitations of the design of the trial must be considered and, therefore, our results must be interpreted with caution. Most importantly this was no blinded randomized controlled trial but a retrospective observational study comparing two consecutively treated patient groups. On the other hand both groups were equally balanced not only with regard to the strongest known prognostic factors, age and MSKCC prognostic score, but also with regard to all other baseline characteristics. Furthermore, treatment protocols did not differ other than the additional application of rituximab during cycles 1–3 in the rituximab group, and all patients were treated in the same institution and in most cases even by the same team. We therefore believe that comparison of these two groups is reasonable and informative. For systemic DLBCL there is strong evidence from randomized controlled trials that CR is improved by combining rituximab with the CHOP (cyclophosphamide, vincristine, prednisolone, doxorubicine) regimen, e.g. from 63 to 75 % [5]. To date no such randomized trials evaluating patients with PCNSL exist. There are only a few small single-arm studies reported which evaluated different

123

Fig. 1 There was no statistically significant difference in time to TTF (a), progression free survival (PFS; b), and OS (c) between the groups

J Neurooncol (2012) 109:285–291

289

Table 3 Comparison of toxicity and complications

ICU intensive care unit

Control group (MTX ? IFO)

Rituximab group (MTX ? IFO ? R)

p

Anemia grade IV, n (%)

2/19 (10.5)

4/17 (23.5)

0.39

Thrombopenia grade IV, n (%)

2/19 (10.5)

5/17 (29.4)

0.22

Leukopenia grade IV, n (%)

10/19 (52.6)

9/17 (52.9)

1.00

Infections grade III ? IV, n (%)

0.70

14/19 (73.7)

14/17 (82.4)

Sepsis, n (%)

2/19 (10.5)

2/17 (11.8)

1.00

Hepatic toxicity grade III ? IV, n (%)

6/19 (31.6)

6/17 (35.3)

1.00

Nephrotoxicity grade III ? IV, n (%)

0/19 (0)

0/17 (0)

n.a.

Pulmonary embolism, n (%)

0/19 (0)

1/17 (5.9)

0.47

ICU treatment, n (%)

2/19 (10.5)

3/17 (17.6)

0.65

Treatment stopped because of toxicity, n (%)

1/19 (5.3)

3/17 (17.6)

0.33

Treatment-associated death, n (%)

0/19 (0)

0/17 (0)

n.a.

chemotherapy protocols plus rituximab: in a pilot study (n = 30) Shah et al. [7] observed CR of 78 % after 5–7 cycles of rituximab, MTX, procarbazine, and vincristine (R-MPV). In a preceding study patients were treated with five cycles of the same regimen but without rituximab [8]. CR was only 56 % in this trial. Whether the improvement of CR was caused by addition of two more chemotherapy cycles for patients having achieved PR after five cycles or by addition of rituximab is unclear. Because overall response rate (RR = CR ? PR) already was [90 % after MPV without rituximab the authors could not demonstrate significant improvement of RR by addition of rituximab. A German pilot trial (n = 28) evaluating rituximab, MTX, procarbazine, and lomustine (R-MCP) in elderly PCNSL patients ([65 year) observed a CR of 64 % (RR 82 %) [9]. In a previous study using the same regimen without rituximab (n = 30) the same group had reported a CR of 44 % (RR 70 %) [3]. In a phase II study (n = 40) Chamberlain et al. [10] found a CR of 60 % after 4–6 cycles of rituximab and MTX. Previous studies using MTX monotherapy without rituximab have reported CR between only 30 and 58 % [11–13]. Of course, such comparisons between sequential single-arm trials have major limitations and these data must, therefore, be interpreted very carefully. Nevertheless, our results and data from the literature suggest that rituximab might significantly increase the number of patients with CR to primary chemotherapy not only in systemic DLBCL but also in PCNSL. This hypothesis must be tested in prospective randomized trials. This implies that rituximab could be efficacious despite its inability to cross the intact BBB. Rubenstein et al. [14] demonstrated that cerebrospinal fluid (CSF) levels of rituximab are only *0.1 % of matched serum levels after intravenous administration to humans. Another study found CSF levels ranging from 0.1 to 4.4 % of serum levels [7]. However, the BBB is disrupted in the context of parenchymal PCNSL lesions which is indicated by the

homogenous, pronounced contrast-enhancement in brain MRI. Therefore rituximab levels could be much higher within these lesions than in CSF. Supporting these considerations Batchelor et al. [15] confirmed a radiographic response to rituximab monotherapy in 4/11 patients with recurrent PCNSL. Because, in responding patients, contrast-enhancement usually decreases rapidly after the start of treatment (indicating restoration of the BBB), rituximab might be mostly efficacious before chemotherapy or glucocorticoids are given and/or during the first treatment cycles. The CR/uCR rate after immunochemotherapy with MTX, IFO, and rituximab in this retrospective study was surprisingly high; this result compares favorably with all previously reported CR after MTX-based polychemotherapy regimens without WBI [4, 7, 9, 10, 16, 17]. Because this was not a controlled multicenter trial and the sample size was small, this result must, of course, not be overestimated. But we would like to emphasize that response assessment was rechecked by a blinded, experienced neuroradiologist. There was a high rate of uCRs in the rituximab group because of minimal residual contrastenhancing lesions at the biopsy site. None of these patients was treated with corticosteroids. All uCR patients had follow-up MRI 4–8 weeks after termination of treatment and there was no case of progression or recurrence. Furthermore, high PFS-6 in the rituximab group is very consistent with a high rate of ‘‘real’’ CRs, because PCNSL which are not in complete remission usually progress rapidly after treatment has stopped. Therefore, immunochemotherapy with MTX, IFO, and rituximab could be a very promising induction chemotherapy regimen for patients with PCNSL, and should be evaluated further in prospective trials. An obvious important benefit of increased CR after initial chemotherapy of PCNSL is the less frequent need for adjuvant WBI with its substantial risk of severe neurotoxicity. To

123

290

date, most centers withhold radiation therapy from patients who achieve CR to chemotherapy until tumor recurrence. Although, in this trial, adjuvant WBI did not differ statistically significantly between the groups this might be primarily because of the small sample size. There is evidence that increased CR correlates with better clinical outcome for patients with systemic NHL [18]. Whether this might be true for PCNSL also is unclear. In our trial we found PFS-6 was significantly better in the rituximab group, which probably reflects the higher CR. PFS, OS, and relapse rate, in contrast, did not differ significantly between the groups. Of course, it must be mentioned that our study: 1 was not of sufficient power to evaluate long-term effects of rituximab on patient outcome; and 2 was not a prospective, randomized trial. Furthermore, the different lengths of follow-up and different salvage treatment regimens might have affected these results. In this study hematologic toxicity was high in both groups, without significant difference, whether rituximab was given or not, but this clearly needs confirmation by larger, prospective trials. In randomized controlled trials evaluating rituximab in systemic NHL, slightly greater incidence of neutropenia and herpes zoster reactivation was found in patients treated with rituximab [5]. Overall additional toxicity from rituximab seems to be modest, however. We detected neutropenia grade III/IV in 80.5 % of patients. In contrast, in the G-PCNSL-SG1 trial, which used the same chemotherapy regimen with MTX and IFO (without rituximab) for 125 patients, this complication was observed for 64.5 % of patients only [2]. On the other hand, CR also was substantially lower than in our trial (41.6 vs. 68.4 %). A putative explanation of the lower efficacy and lower toxicity in the G-PCNSL-SG1 trial could be our tendency, whenever possible, not to reduce chemotherapy doses in response to hematologic toxicity, because of our past experience of higher relapse rates for patients treated with reduced-dose chemotherapy. Further reasons for the greater toxicity in our trial might, of course, be the patients0 older age and lower KPS in comparison with the G-PCNSL-SG1 trial.

Conclusions We found complete remission was significantly greater after chemotherapy with high-dose MTX and IFO when we added rituximab to the first three cycles for patients with newly-diagnosed PCNSL. Our results therefore indicate that rituximab might be efficacious for treatment of this disease as well as systemic NHL. However, results from randomized controlled trials prospectively assessing the

123

J Neurooncol (2012) 109:285–291

effect of rituximab in PCNSL treatment are urgently needed. As far as we are aware, two randomized multicenter trials have recently been launched which will address, among other factors, the effect of rituximab in PCNSL treatment. The first trial is being conducted by the Dutch HOVON group (HOVON 105 PCNSL/ALLG NHL24 trial), the second by the international extranodal lymphoma study group (IELSG-32 trial, nct01011920). Until these results become available, our results support the implementation of rituximab in PCNSL treatment protocols. The combination of MTX and IFO with rituximab seems to be associated with high CR and should be further evaluated as an induction chemotherapy regimen in prospective trials. Acknowledgments Parts of this work are elements of one coauthor0 s dissertation (Elisabeth A. Stadler) presented to the Medical Faculty, LMU Munich, Germany. We thank Caroline Padgham for copy-editing the manuscript. This work was not supported by grants. Conflict of interest interest.

The authors declare they have no conflict of

References 1. Morris PG, Abrey LE (2009) Therapeutic challenges in primary CNS lymphoma. Lancet Neurol 8:581–592 2. Thiel E, Korfel A, Martus P et al (2010) 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 11:1036–1047 3. Illerhaus G, Marks R, Mu¨ller F et al (2009) 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 20:319–325 4. Pels H, Schmidt-Wolf IG, Glasmacher A et al (2003) Primary central nervous system lymphoma: results of a pilot and phase II study of systemic and intraventricular chemotherapy with deferred radiotherapy. J Clin Oncol 21:4489–4495 5. Coiffier B, Lepage E, Briere J et al (2002) CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 346:235–242 6. Abrey LE, Batchelor TT, Ferreri AJ et al (2005) Report of an international workshop to standardize baseline evaluation and response criteria for primary CNS lymphoma. J Clin Oncol 23:5034–5043 7. Shah GD, Yahalom J, Correa DD et al (2007) Combined immunochemotherapy with reduced whole-brain radiotherapy for newly diagnosed primary CNS lymphoma. J Clin Oncol 25:4730–4735 8. Abrey LE, Yahalom J, DeAngelis LM (2000) Treatment for primary CNS lymphoma: the next step. J Clin Oncol 18:3144–3150 9. Fritsch K, Kasenda B, Hader C et al (2011) Immunochemotherapy with rituximab, methotrexate, procarbazine, and lomustine for primary CNS lymphoma (PCNSL) in the elderly. Ann Oncol 22:2080–2085 10. Chamberlain MC, Johnston SK (2010) High-dose methotrexate and rituximab with deferred radiotherapy for newly diagnosed primary B-cell CNS lymphoma. Neuro Oncol 12:736–744 11. Herrlinger U, Ku¨ker W, Uhl M et al (2005) NOA-03 trial of highdose methotrexate in primary central nervous system lymphoma: final report. Ann Neurol 57:843–847

J Neurooncol (2012) 109:285–291 12. Batchelor T, Carson K, O’Neill A (2003) Treatment of primary CNS lymphoma with methotrexate and deferred radiotherapy: a report of NABTT 96-07. J Clin Oncol 21:1044–1049 13. Cobert J, Hochberg E, Woldenberg N, Hochberg F (2010) Monotherapy with methotrexate for primary central nervous lymphoma has single agent activity in the absence of radiotherapy: a single institution cohort. J Neurooncol 98:385–393 14. Rubenstein JL, Combs D, Rosenberg J et al (2003) Rituximab therapy for CNS lymphomas: targeting the leptomeningeal compartment. Blood 101:466–468 15. Batchelor TT, Grossman SA, Mikkelsen T et al (2011) Rituximab monotherapy for patients with recurrent primary CNS lymphoma. Neurology 76:929–930

291 16. Ferreri AJ, Reni M, Foppoli M et al (2009) High-dose cytarabine plus high-dose methotrexate versus high-dose methotrexate alone in patients with primary CNS lymphoma: a randomised phase 2 trial. Lancet 374:1512–1520 17. Wieduwilt MJ, Valles F, Issa S et al (2012) Immunochemotherapy with intensive consolidation for primary central nervous system lymphoma: a pilot study and prognostic assessment by diffusion-weighted MRI. Clin Cancer Res 18:1146–1155 18. Davidge-Pitts M, Dansey R, Bezwoda WR (1996) Prolonged survival in follicular non Hodgkins lymphoma is predicted by achievement of complete remission with initial treatment: results of a long-term study with multivariate analysis of prognostic factors. Leuk Lymphoma 24:131–140

123