Carlson, Robert H.
NEW YORK—Disease control rather than neurotoxicity is the primary challenge in treating primary central nervous system lymphoma (PCNSL), according to a speaker here at the International Congress on Hematologic Malignancies.
Neurotoxicity is clearly a significant problem, said Lisa DeAngelis, MD, Chair of the Department of Neurology and Co-Executive Director of the Brain Tumor Center at Memorial Sloan Kettering Cancer Center.
But while neurotoxicity “has become a huge focus in trying to determine the treatment for this disease, I would focus instead on the fact that we achieve very poor disease control. If you compare the risk of neurotoxicity versus the risk of death without neurotoxicity, which is almost exclusively death due to tumor progression, tumor progression is clearly twice the problem that neurotoxicity is.”
PCNSL is an uncommon and difficult-to-treat variant of extranodal non-Hodgkin lymphoma, she explained. Incomplete initial response to chemotherapy is often the case, and frequent relapse is seen even in patients who have responded.
Regardless of the regimen used, the response rate is typically 60 to 95 percent, with the majority complete responses. But median progression-free survival has been 10 to 24 months, median overall survival 12 to 55 months, and median overall survival for the majority of patients is approximately three years.
“If you consider PCNSL a stage 1E DLBCL [diffuse large B-cell lymphoma], this is clearly inferior to any other stage 1E tumor,” she said.
Chemoradiotherapy is clearly superior to cranial irradiation alone, DeAngelis said, but the optimal strategy is still not clear in terms of the choice of chemotherapy and dose.
Increasing chemotherapy duration and intensifying chemotherapy through the use of autologous stem cell transplant appear to increase the complete response rate and improve disease control. But there are age and performance-status limitations to aggressive treatment in older patients.
And there is the neurotoxicity issue in high-dose chemotherapy regimens, although DeAngelis noted that cognitive impairment is much more related to the presence of disease and disease progression than to whole-brain radiotherapy.
To date, she said, there has been only one Phase III randomized trial conducted in PCNSL: the German PCNSL-SG1 trial, with 551 patients, designed to assess the value of whole-brain radiotherapy, comparing regimens with and without radiotherapy (Thiel et al: Lancet Oncology 2010;11:1036). All patients received high-dose methotrexate, the majority as a single agent, although some patients also received ifosphamide.
Patients who achieved a complete response were randomly assigned to whole-brain radiotherapy at 45 Gy or to observation. Those who did not reach a complete response were randomly assigned to full-dose whole-brain radiotherapy or to high-dose cytarabine.
The trial failed to meet its endpoint, though, and there was a large dropout in the whole-brain radiotherapy group, she noted. And while there was a difference in progression-free survival—18 months with cranial irradiation versus 12 months without—overall survival was about 36 months for both study arms.
The researchers concluded that whole-brain irradiation does not increase overall survival and should not be used.
“Unfortunately, there was almost no assessment of neurotoxicity in this study, and by year six to seven there was 100 percent progressive disease,” DeAngelis said.
She noted that more than half of the patients in the trial—56 percent—had an inadequate response to therapy, and overall survival was directly related to achievement of a complete response. Overall survival of patients who achieved a complete response was 39 months versus 22 months for those who did not achieve a complete response.
“Tumor eradication is really inadequate with current therapies, so how do we do better?” DeAngelis asked. “It's clear that combination therapy is superior to single agent, but the German trial did not tell us what the optimal chemotherapy regimen is, nor what the radiotherapy dose should be.”
A trial at Memorial Sloan Kettering in which DeAngelis participated did suggest a regimen: In that trial (Morris et al: JCO 2013;31:3971-3979), 52 patients were treated initially with R-MPV (rituximab, methotrexate, procarbazine, and vincristine) chemotherapy for five to seven cycles.
LISA DEANGELIS, MD
Those achieving complete response received reduced-intensity whole-brain radiotherapy of 23.4 Gy, and those who did not have a complete response received standard 45 Gy whole-brain radiotherapy. Consolidation cytarabine was given after the radiotherapy.
“More chemotherapy improved responses” was the finding, DeAngelis said. The complete response rate was 44 percent after five cycles, versus 78 percent after seven cycles.
For the 31 patients (60%) who achieved a complete response after R-MPV and received the reduced whole-brain radiotherapy regimen, the two-year progression-free survival rate was 77 percent and median progression-free survival was 7.7 years. Median overall survival was not reached, although median follow-up for survivors was 5.9 years and three-year overall survival was 87 percent.
That compares with overall median progression-free survival for all 52 patients of 3.3 years, and median overall survival of 6.6 years.
“With up to five years of follow-up, there is no clinical or radiographic evidence of neurotoxicity, suggesting that lowering of radiotherapy may give us the benefit without the toxicity,” she said.
Dose Intensification with Transplant
DeAngelis briefly touched on the issue of chemotherapy dose intensification with transplant. She cited a retrospective German study published last year of 105 immunocompetent patients with PCNSL who underwent high-dose chemotherapy followed by autologous stem cell transplantation with or without whole-brain radiotherapy as first-line consolidation (Schorb et al: Haematologica 2013;98:765-770).
The authors reported that at a median follow-up of 47 months, median progression-free survival was reached at 85 months and overall survival at 121 months. The overall response rate was 95 percent.
“This looks extremely promising, but it is very important to keep in mind that this is a highly selected patient population and really cannot be compared with any other study,” DeAngelis said.
In conclusion, she said:
- Disease control in PCNSL is the challenge;
- Increased chemotherapy cycles and intensification increase complete response rates and disease control; and
- Reduced-intensity whole-brain radiotherapy appears safe and effective when it is combined with a methotrexate-based chemotherapy regimen.