This protocol has been superseded as it is not considered best practice for this patient population. PCNSL MATRix (methotrexate cytarabine thiotepa and rituximab) (ID 3374) is the preferred regimen.
Most modern protocols for primary CNS lymphoma (PCNSL) have included high dose intravenous methotrexate and/or intrathecal therapy with whole brain radiation therapy. O’Brien et al (2000)r reported on 46 patients who were treated with two doses of high dose methotrexate 1 g/m2 IVI on days 1 and 8 followed by radiation therapy on day 15 (whole brain 45 Gy with a boost of 5.4 Gy). Intrathecal therapy was only given if the CSF was involved. Median survival was 33 months with a 2 and 5 year probability of survival of 62% and 37%. This combined modality regimen was associated with significant neurotoxicity being 30% for all patients at 5 years and increasing to 58% for patients aged >60 years at 7 years.r
Abrey et al (2000)r used a protocol of 3.5 g/m2 methotrexate IVI every two weeks for five doses, in combination with procarbazine, vincristine, intra-Ommaya methotrexate and subsequent whole brain radiation therapy (45 Gy), followed by two doses of high dose cytarabine IVI. An overall median survival of 60 months was reported. There was however significant neurotoxicity in the older patients. 22 out of 52 patients did not receive radiation therapy. Withholding radiation therapy did not appear to impact overall survival although was associated with a higher rate of CNS recurrence. The authors recommended that radiation therapy is withheld in patients over the age of 60.
Similar findings were upheld on longer term follow-up of these and additional patients (57 patient total) treated with this regimen.r 30% of patients had treatment related neurotoxicity, with the highest risk being in the patients aged > 60 who received whole brain radiation therapy. The overall median survival was 51 months, with 47% surviving greater than or equal to 60 months. 37% of these long term survivors had evidence of neurotoxicity, and this was the commonest cause of death in this subgroup.
The high risk of neurotoxicity in patients aged >60 who receive high dose intravenous methotrexate and whole brain radiation therapy has been confirmed in other studies reviewed by Batchelor & Loeffler (2006).r Small series by Batchelor et al (2003)r have shown efficacy with reduced toxicity using high dose methotrexate alone with deferred radiation therapy. Pels et al (2003)r reported a phase I/II series of 65 patients treated with a high dose methotrexate (5 g/m2 over 24h), plus cyclophosphamide, ifosfamide, vincristine, vindesine and dexamethasone regimen, intrathecal chemotherapy and deferred radiation therapy (i.e. given on progression or relapse only). 61% had a complete response with a higher rate (76 vs 47%) in patients under the age of 60. The estimated 5 year survival was 43%. Only two patients had severe cognitive dysfunction after therapy.
Ferreri et al (2009)r reported on a multicenter, phase II study of 79 patients (age 18 to 75) who were randomized to receive four courses of methotrexate 3.5 g/m2 on day 1 or four courses of methotrexate 3.5 g/m2 on day 1 combined with cytarabine 2 g/m2 BD on days 2 and 3. Cycles were repeated every three weeks with radiation therapy starting within four weeks of the last cycle. Patients aged 60 years or younger in complete remission were treated with 36 Gy whole-brain radiation therapy; those older than 60 years in complete remission were irradiated at discretion of participating centres. Intrathecal chemotherapy was not used. The complete remission rates were 18% in the methotrexate only arm and 46% in the combined methotrexate and cytarabine arm. With relatively short periods of follow-up, three year failure free survival was 21% for the methotrexate group and 38% for the methotrexate/cytarabine group (p=0.01). A treatment related mortality of 8% was noted in the combined chemotherapy arm as opposed to 3% in the methotrexate only arm (p=0.35).
Radiation therapy is associated with an improved response rate and CNS control but with a higher risk of neurotoxicity particularly in patients over the age of 60.r However, it is still not clear in which patients' radiation therapy can be reasonably withheld. In a large trial of 551 patients from 75 European centres with a median age of 63 years,r patients were randomized following six cycles of high dose methotrexate (4 g/m2), with patients from 2000-2006 receiving additional ifosfamide based chemotherapy, to receive or not receive radiation therapy. In the group who were not randomized to radiation therapy, four cycles of high dose cytarabine (3 g/m2 daily on D1 and D2) were given every 21 days. Intrathecal chemotherapy was not administered as part of the protocol. Of the 551 patients, only 318 were treated according to the protocol. In the per protocol group, the overall survival was 32.4 months in the group receiving radiation therapy and 37.1 months in the group not receiving radiation therapy (p=0.71). Median progression free survival was 18.3 months in the group receiving radiation therapy and 11.9 months in the group not receiving radiation therapy (p=0.14). The authors, however, concluded that the non-inferiority hypothesis of omitting radiation therapy was not proven. Treatment related neurotoxicity in those with sustained complete remission was 49% in the patients receiving radiation therapy and 26% in those who did not, as assessed clinically.
What can we conclude from these studies? The median age at diagnosis is 60 to 65 years, median survival is 10 to 20 months, with survival of <20 to 30% at 5 years.rThe addition of cytarabine in one randomized study appears to confer a benefit compared to high dose methotrexate alone, but at the cost of increased treatment related mortality. Omission of intrathecal or intra-Ommaya therapy appears reasonable based on the results of Ferreri et al (2009)r and Theil et al (2010).r Radiation therapy appears to offer better CNS control but at the expense of greater neurotoxicity and should be used with particular care in patients aged >60 who have a higher risk of cognitive damage. For the purposes of this eviQ protocol we have therefore adopted the regimen of Ferreri et al (2009)r whilst acknowledging that it is highly likely that further refinements will occur in the development of an optimal protocol. There appears to be a dose-response relationship for neurotoxicity due to combined therapy, therefore the lowest radiation therapy dose associated with optimal survival outcomes has been chosen.r The addition of high dose cytarabine is however associated with an increase in treatment related mortality from 3 to 8%, and it may be appropriate to omit this in patients who are elderly, in poor physical condition or unlikely to tolerate an intensive regimen.
Radiation therapy
Details of the radiation therapy portion from Ferreri et al 2009r on which this protocol is based:
Consolidation whole-brain radiation therapy (WBRT) was started within 4 weeks from the last chemotherapy course.
Photons of 4 to 10 MeV, 1.8 Gy per day, five fractions per week were used. Whole brain radiation therapy was performed using two opposed lateral fields including the first two segments of cervical spinal cord and the posterior two-thirds of the orbits, which had to be shielded after 30 Gy (or after 36 Gy in the case of intraocular disease). Tumour bed was irradiated by two to four isocentric fields on the basis of tumour location; in the case of multifocal lesions, the boost volume included each single lesion.
Radiation dose was chosen according to age and response after chemotherapy: patients aged 60 years or younger in complete remission were treated with 36 Gy whole-brain radiation therapy; those older than 60 years in complete remission were irradiated at discretion of participating centres. Patients of any age in partial response were treated with 36 Gy whole-brain radiation therapy plus a tumour-bed boost of 9 Gy. Patients of any age in stable or progressive disease were treated with whole-brain radiation therapy with 40 Gy plus a 9 Gy boost.
Toxicity
Grade 3-4 toxicities associated with study by Ferreri et al 2009r
© The Lancet 2009