Linfoma primário do sistema nervoso central

1812

Primary Central Nervous System Lymphoma Monocenter, Long-term, Intent-to-treat Analysis

Philipp Kiewe, MD1 Lars Fischer, MD1 Peter Martus, PhD2 Eckhard Thiel, MD1 Agnieszka Korfel, MD1

BACKGROUND. This retrospective, single-center study assessed the feasibility, outcome, and late side effects of the treatment of immunocompetent patients with primary central nervous system lymphoma (PCNSL) at the authors’ institution.

METHODS. All 72 consecutive patients diagnosed with PCNSL between January 1994 and February 2005 were scheduled to receive high-dose methotrexate (HDMTX)-based chemotherapy.

1 Department of Hematology, Oncology, and Transfusion Medicine, Charite Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany.

RESULTS. The median age of the patients was 62 years and the median Karnofsky

2

60 patients treated with HDMTX-based chemotherapy, the treatment of 9 was

Institute for Biostatistics and Clinical Epidemiology, Charite Universitätsmedizin Berlin, Campus Benjamin Franklin Berlin, Germany.

performance score (KPS) was 70. Twelve patients did not receive HDMTX-based chemotherapy because of poor physical condition or renal insufficiency. Of the followed with whole-brain irradiation. Of 54 patients who were evaluable for response, 35 (65%) responded (52% with a complete response and 13% with a partial response), and 19 patients (35%) did not. At a median follow-up of 58.7 months, the median progression-free survival was 9 months and the median overall survival (OAS) was 41.4 months. According to the Memorial Sloan-Kettering Cancer Center (MSKCC) prognosis score, patients could be divided into 3 groups with significantly different OAS: 52.9 months for patients aged 70, and 5.2 months for patients aged
50 years and with a KPS 50 years, age 60 years and >60 years, and age as a continuous variable; KPS
70 and 1 lesion; and superficial and deep lesion location. In addition, the MSKCC prognosis score, based on age and KPS, was validated. Response to first treatment was evaluated according to International Primary CNS Lymphoma Study Group (IPCG) criteria13 and grouped into ‘responders’ (complete response [CR] and partial response [PR]) and ‘nonresponders’ (stable or progressive disease) for statistical analysis. An elevated CSF cell count was defined as >5 cells/lL.14 Event-free survival (EFS) was defined as the time from histologic diagnosis to first documentation of disease progression on imaging or death from any cause in patients responding (CR and PR) to first

1814

CANCER

April 15, 2008 / Volume 112 / Number 8

therapy, progression-free survival (PFS) was defined as the time from histologic diagnosis to first documentation of disease progression on imaging or death from lymphoma in all patients, and overall survival (OAS) was defined as the time from beginning of treatment to death from any cause, according to the standardized response criteria for non-Hodgkin lymphoma.15 Depending on response to first treatment, OAS was evaluated using the landmark method, which defines the treatment response for a specific timepoint (
3 months) and only those patients alive at this timepoint were included in the analysis. This avoids potential bias because patients with complete remission usually survive at least until assessment of remission.16 Time periods for OAS and EFS were calculated in months with 1-decimalpoint accuracy; symptom duration before diagnosis was measured in days. EFS, PFS, and OAS were estimated by the Kaplan-Meier method.17 Group comparisons were made using the log-rank test.18 The independent prognostic value of variables was analyzed by Cox stepwise regression model.19 The level of significance was .05 (2-sided). In the current study, hazards ratios of at least 2.5 could be detected with a power of at least 80%. Commercially available statistical software was used (SPSS for Windows, release 14.0; SPSS Inc, Chicago, Ill).

RESULTS Patient Characteristics Seventy-two patients with a median age of 62 years (range, 16–87 years) and a median KPS of 80% (range, 10–100; mean 73%) were identified. Thirty of these patients were pooled from 2 prospective studies, 13 of whom had been published previously.6 All patients had parenchymal brain involvement except for 2 patients with isolated intraocular lymphoma and 2 patients with isolated meningeal lymphoma. Concomitant ocular involvement was detected by slit lamp examination in 2 of 12 patients examined. The median symptom duration before diagnosis was 30 days (range, 1–400 days) (n 5 51). Six of 34 patients (18%) in whom CSF analysis was performed had cytomorphologic evidence of lymphoma cells. The median cell count in the CSF was 6 (range, 0–237) and the median protein level (n 5 23) was 760 mg/L (range, 190–1990 mg/L). Patient characteristics are given in Table 1. Initial Treatment Sixty patients received the intended HDMTX-based chemotherapy; 41 patients received it as monother-

TABLE 1 Patient Characteristics No. of patients Median age (range), y KPS 100 80–90 60–70 40–50 10–30 Unknown No. of lesions 0 or 1
2 Unknown Diagnosis confirmation None Total resection Partial resection Stereotactic biopsy Vitrectomy* Cerebrospinal fluid*

% (of n 5 72)

62 (16–87) 10 30 4 11 5 12

13.9 41.7 5.6 15.3 7.0 16.7

43 20 9

59.8 27.8 12.5

3 6 17 42 2 2

4.2 8.3 23.6 58.3 2.8 2.8

KPS indicates Karnofsky performance score. * Lymphoma cells on morphologic/immunocytologic examination.

apy and 19 in combination with other cytostatics (1 each in combination with ifosfamide and high-dose cytarabine, and 17 patients in combination with carmustine and procarbazine). Twelve patients did not receive HDMTX-based chemotherapy because of poor physical condition or renal insufficiency; 4 received chemotherapy protocols without methotrexate (eg, topotecan or ifosfamide monotherapy), 3 received steroids only, and 5 were primarily treated with WBI. In patients receiving HDMTX-based chemotherapy, 9 in addition received WBI as part of their initial treatment. Fifty-four of 69 patients receiving treatment other than steroids only were evaluable for response to first treatment. Thirty-five patients (65%) responded (52% with a CR and 13% with a PR), and 19 patients (35%) were nonresponders (5% with stable disease [SD] and 30% with progressive disease [PD]). Fifteen patients could not be evaluated for response because of complete lymphoma resection (n 5 3 patients), early deterioration and death without response evaluation (n 5 2 patients), or loss to follow-up (n 5 10 patients). In the subgroup treated with HDMTX alone, 20 of 34 patients (59%) responded (7 patients were not evaluable) compared with 11 of 16 patients (69%) in the subgroup treated with HDMTX-based combination chemotherapy (3 patients were not evaluable). In the group treated without HDMTX, 4 of 12

Retrospective Monocenter Analysis in PCNSL/Kiewe et al.

1815

FIGURE 1. Distribution of the study population. HDMTX indicates high-dose methotrexate; CTx, chemotherapy; MTX, methotrexate; HD cytarabine, high-dose cytarabine; BCNU, carmustine, PCB, procarbazine; WBI, whole-brain irradiation; CR, complete remission; Med OAS, median overall survival; PR, partial remission; SD, stable disease; PD, progressive disease; NE, not evaluable.

patients responded, and in 8 patients no response evaluation was performed. The details are presented in Figure 1.

Overall Survival The median follow-up was 58.7 months (95% confidence interval [95% CI], 48.8–68.6 months). The median OAS for the entire whole study population was 41.4 months (95% CI, 32.3–50.6 months) (Fig. 2) and was 42.5 months (95% CI, 38.2–46.8 months) for the 60 patients treated with HDMTX-based chemotherapy, 42.5 months (95% CI, 38–47 months) for patients treated with HDMTX alone, and 41.4 months (95% CI, 14.2–68.6 months) for those treated with an HDMTX combination regimen. The median OAS for patients not treated with HDMTX was 7.3 months (95% CI, 0–15.4 months) and was 1.2 months for those treated with steroids alone, 10.9 months for those treated with alternative chemotherapy regimens, and 8.8 months for those treated with WBI alone. At last follow-up, 22 patients (31%) were alive and 4 patients (6%) were lost to follow-up. Forty-six patients (64%) died; the cause of death was evaluable in 35 patients, was the result of PCNSL in 28 patients (80%), from acute toxicity in 1 patient, a late treatment-related neurotoxicity in 4 patients (11%), and other causes (heart failure and pulmonary embolism)

FIGURE 2. Kaplan-Meier estimate of overall survival of the entire study population (n 5 72). The median overall survival was 41.4 months (95% confidence interval, 32.3–50.6 months) (26 events, 46 patients censored).

in 2 patients. Eleven of the surviving patients (50%) were aged 60 years (n 5 41), which was 24.3 months (95% CI, 0–57.5 months) (P 5 .014, log-rank).

1816

CANCER

April 15, 2008 / Volume 112 / Number 8

As nearly significant difference (P 5 .055) was found between the survival of patients aged 50% of All Patients

Reference

Patient no.

Median age, years

Median KPS (%)

HDMTX in primary therapy (% of all patients)

Gavrilovic 2006 (MSKCC1)20 Abrey 2006 (MSKCC2)9 Hodson 2005 (Cambridge)4

57 338 55

65 61 64

70 70 NG

100 79 55

Herrlinger 2001 (Tuebingen)5

28

59

NG

Current study

72

62

70

WBI in primary therapy (% of all patients)

Median PFS, months/ median EFS, months

Median OAS, months

129/NG NG/17 (FFS) NG/NG

51 37 8 (31 mo for 30 patients receiving HDMTX)

75

54 54 47% of 30 patients receiving HDMTX 48% of 25 patients without HDMTX 89

NG/NG

85

13

9/30.7

11 (42 mo for 5 patients receiving full regimen) 41

HDMTX indicates high-dose methotrexate; KPS, Karnofsky performance score; WBI, whole-brain irradiation; PFS, progression-free survival; EFS, event-free survival; OAS, overall survival; MSKCC, Memorial Sloan-Kettering Cancer Center; NG, not given; FFS, failure-free survival.

entire study population was not negatively selected with regard to patient age or performance status.6 However, the HDMTX infusion time of 24 hours might have been suboptimal, according to more recent findings concerning the use of methotrexate in PCNSL.21,22 Similar results have been reported in other multicenter analyses with a median OAS of 9 months and 17 months, respectively.2,3 This is most likely because of a relatively small proportion of patients treated with HDMTX (at a dose of >1 g/m2) in these trials and supports our previous conclusion that PCNSL patients should preferably be managed at highly experienced centers. We found younger age and response to first treatment to be significantly associated with survival, a finding that is consistent with previously published prognostic factors. However, the 3 prognosis classes created by MSKCC9 distinguished outcome with higher statistical significance than any single prognostic factor both on univariate and multivariate analysis. Thus, the current study data provide an external validation of the MSKCC score. It is interesting to note that, with 52.9 months, the median OAS of class 1 in our cohort was inferior to that of class 1 patients from MSKCC at a median of 102 months, most likely supporting the front-line use of more intensive therapy regimens, including intraventricular chemotherapy and WBI in younger patients, as is usually practiced at the MSKCC. With a long median follow-up of nearly 5 years, we identified 8 long-term survivors (>60 months). Two of them received chemotherapy followed by WBI, and 6 patients had chemotherapy only as their initial treatment. It is interesting to note that 5 of them developed disease recurrence after a median of 5 months and were alive at a median of 58 months

after salvage therapy; 2 patients had been treated with WBI, 2 patients had been treated with chemotherapy including topotecan, and 1 patient received high-dose chemotherapy with autologous stem cell support (n 5 1). These findings emphasize that, with adequate salvage therapy, long-term survival can be achieved in patients with PCNSL, even in those patients who develop early disease recurrence after first-line treatment. At the time of disease recurrence, not only WBI, but also an increasing number of cytostatic agents can be used successfully with a chance of long-term survival, emphasizing the role of drugs other than MTX in the management of PCNSL. The relatively short PFS noted in the current study may reflect the use of mild therapy regimens as first-line treatment and the deferral of radiotherapy in the majority of patients. However, because of effective salvage therapy, the high rate of initial treatment failure did not translate into inferior survival. This is in line with previously published studies deliberately deferring WBI at the expense of overall response and failure-free survival but without compromising OAS.8,23 Late neurotoxicity is a potentially devastating complication in PCNSL, particularly in older patients and after intensive therapy.8 Unfortunately, to our knowledge, well-conducted studies with a long-term follow-up assessing cognitive status in patients after different therapy modalities are lacking. Moreover, psychometric tests established for other conditions were not feasible in this study. Thus, we decided to evaluate radiologic changes and clinical symptoms most likely attributable to neurotoxicity using a specially designed questionnaire.11 With this method, the risk of late neurotoxicity is likely underestimated compared with prospective neuropsychologic testing.

1820

CANCER

April 15, 2008 / Volume 112 / Number 8

The high frequency of late neurotoxicity and the correlation of clinical and radiologic findings in our study are comparable to findings from studies with shorter follow-up.5 Remarkably, in the group evaluated for late neurotoxicity, 5 of 11 patients with clinical signs of neurotoxicity received WBI compared with none of 6 patients without such deficits. The median age was similar in both groups (57 years and 55 years, respectively). All 4 patients whose death was related to late neurotoxicity had previously been treated with radiotherapy, 3 of them as part of their initial therapy. Conversely, of the 12 patients treated with chemotherapy only (median age of 57 years), 4 patients had memory deficits and 2 considerable deficits of concentration, memory, and orientation. These findings highlight the problem of late neurotoxicity in PCNSL and support the deferral of radiotherapy as long as possible, particularly in elderly patients. In conclusion, the mature data of the current study demonstrate that a relatively moderate HDMTX-based chemotherapy can result in prolonged survival and most likely cure, even in older patients and those with early disease recurrence. However, the probability of late neurotoxicity with prolonged survival is considerable. Our results underscore the importance of treating PCNSL patients preferably at experienced institutions.

8.

9.

10.

11.

12.

13.

14.

15.

REFERENCES 1. 2.

3.

4.

5.

6.

7.

Batchelor T, Loeffler JS. Primary CNS lymphoma. J Clin Oncol. 2006;24:1281–1288. Panageas KS, Elkin EB, DeAngelis LM, Ben-Porat L, Abrey LE. Trends in survival from primary central nervous system lymphoma, 1975–1999: a population-based analysis. Cancer. 2005;104:2466–2472. Shenkier TN, Voss N, Chhanabhai M, et al. The treatment of primary central nervous system lymphoma in 122 immunocompetent patients: a population-based study of successively treated cohorts from the British Colombia Cancer Agency. Cancer. 2005;103:1008–1017. Hodson DJ, Bowles KM, Cooke LJ, et al. Primary central nervous system lymphoma: a single-centre experience of 55 unselected cases. Clin Oncol (R Coll Radiol). 2005;17: 185–191. Herrlinger U, Schabet M, Brugger W, et al. Primary central nervous system lymphoma 1991–1997: outcome and late adverse effects after combined modality treatment. Cancer. 2001;91:130–135. Korfel A, Martus P, Nowrousian MR, et al. German primary central nervous system lymphoma Study Group (G-PCNSLSG). Response to chemotherapy and treating institution predict survival in primary central nervous system lymphoma. Br J Haematol. 2005;128:177–183. Corry J, Smith JG, Wirth A, Quong G, Liew KH. Primary central nervous system lymphoma: age and performance

16. 17. 18.

19. 20.

21.

22.

23.

status are more important than treatment modality. Int J Radiat Oncol Biol Phys. 1998;41:615–620. Abrey LE, Yahalom J, DeAngelis LM. Treatment for primary CNS lymphoma: the next step. J Clin Oncol. 2000;18: 3144–3150. 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: 5711–5715. 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: 266–272. Jahnke K, Korfel A, Martus P, et al. on the behalf of the German Primary Central Nervous System Lymphoma Study Group (G-PCNSL-SG). High-dose methotrexate toxicity in elderly patients with primary central nervous system lymphoma. Ann Oncol. 2005;16:445–449. Fischer L, Thiel E, Klasen HA, et al. Prospective trial on topotecan salvage therapy in primary CNS lymphoma. Ann Oncol. 2006;17:1141–1145. Abrey LE, Batchelor TT, Ferreri AJ, et al. International Primary CNS Lymphoma Collaborative Group. Report of an international workshop to standardize baseline evaluation and response criteria for primary CNS lymphoma. J Clin Oncol. 2005;23:5034–5043. Fischer L, Jahnke K, Martus P, Weller M, Thiel E, Korfel A. The diagnostic value of cerebrospinal fluid pleocytosis and protein in the detection of lymphomatous meningitis in primary central nervous system lymphomas. Haematologica. 2006;91:429–430. Cheson BD, Horning SJ, Coiffier B, et al. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas.NCI Sponsored International Working Group. J Clin Oncol. 1999;17:1244 [Erratum: J Clin Oncol. 2000;18:2351]. Anderson JR, Cain KC, Gelber RD. Analysis of survival by tumor response. J Clin Oncol. 1983;1:710–719. Kaplan EL, Meier P. Non parametric estimations from incomplete observations. J Am Stat Assoc. 1958;53:457–481. Mantel N. Evaluation of survival data and 2 rank order statistics arising in its consideration. Cancer Chemother Rep. 1966;50:163–179. Cox DR. Regression models and life tables (with discussion). J R Stat Soc (B). 1972;74:187–200. Gavrilovic IT, Hormigo A, Yahalom J, DeAngelis LM, Abrey LE. Long-term follow-up of high-dose methotrexate-based therapy with and without whole brain irradiation for newly diagnosed primary CNS lymphoma. J Clin Oncol. 2006;24: 4570–4574. Hiraga S, Arita N, Ohnishi T, et al. Rapid infusion of highdose methotrexate resulting in enhanced penetration into cerebrospinal fluid and intensified tumor response in primary central nervous system lymphomas. J Neurosurg. 1999; 91:221–230. Dukic SF, Heurtaux T, Kaltenbach ML, et al. Influence of schedule of administration on methotrexate penetration in brain tumours. Eur J Cancer. 2000;36:1578–1584. Batchelor T, Carson K, O’Neill A, et al. Treatment of primary CNS lymphoma with methotrexate and deferred radiotherapy: a report of NABTT 96-07. J Clin Oncol. 2003; 21:1044–1049.