Age-Adjusted Combined Immunochemotherapy without Radiotherapy in Newly Diagnosed PCNSL – A Phase II Trial of the Nordic Lymphoma Group

From May 2007 to October 2010, 66 newly diagnosed primary central nervous system lymphoma (PCNSL) patients (M/F ratio 1: 1) were enrolled. Younger patients (≤65 yrs; N=39) received 6 three-weekly cycles of chemotherapy consisting of: high-dose (HD)-methotrexate (MTX) (cycles 1, 2, 4 and 5), HD-cytosine arabinoside (AraC) (cycles 3 and 6) in addition to Rituximab (cycle 1 only), ifosfamide (cycles 1 and 4), cyclophosphamide (cycles 2 and 5), vincristine (cycles 2 and 5), vindesine (cycles 3 and 6), and dexamethasone (all 6 cycles). Depocyte® was delivered intratechally during the HD-MTX cycles. Elderly patients (66–75 yrs; N=27) received an identical Rituximab-containing 1^st cycle. Cyclophosphamide and ifosfamide were replaced by temozolamide (cycles 2 to 6), which was also given as maintenance in patients with chemosensitive disease, and vincristine was omitted. No radiotherapy was given. Response was determined after the 2^nd, 4^th and 6^th chemotherapy cycle by cerebral MRI and assessed according to International Primary CNS Lymphoma Coordinating Group criteria. The primary endpoint was overall survival (OS), secondary endpoints were progression-free survival (PFS), overall response rate (ORR), systemic toxicity and neurotoxicity assessed as Mini Mental State Examination (MMSE) and Functional Independence Measure (FIM).

The median age was 64 yrs overall, 55 yrs (range 40–65) for younger and 70 yrs (range 66–75 years) for elderly patients. In 56 patients, the International Extranodal Lymphoma Study Group prognostic score was: 0–1 (N=5), 2–3 (N=36) and 4–5 (N=15). In the remaining 10 patients, lumbar puncture was not performed in five and spinal fluid protein concentration not reported in additional five cases. Response assessment after completion of induction treatment was performed in 43 out of 66 patients and showed complete remission (CR/CRu) in 30 patients, partial remission (PR) in 5 and progressive disease (PD) in 8. The ORR was 53 %. In 23 patients, response could not be evaluated due to early progression (n=8), toxic death (n=4), poor performance (n=3), neurotoxicity (n=5), or other causes (n=3). Of the 27 elderly patients, 15 continued to maintenance therapy. Of these, 14 have completed the maintenance schedule. Remission status at month 3 was CR in 13 and PD in 1 patient.

With a median follow-up of 11.1 months (range 0.6–40.2) the 3-yr OS was 54.6% with no significant difference between younger and elderly patients (56.4% vs 51.9% respectively, p=0.32). The 3-yr PFS was 35.1% (32.9% in younger and 38.2 % in elderly patients; p=0.96). There were four septic deaths. Grade 3–4 hematological toxicity was seen in 79 % of the patients. Arachnoditis-like symptoms occurred in 13 patients. In all but two patients, the symptoms resolved within less than a week. MMSE and FIM were recorded both before and after therapy in 32 patients. Scores improved in 18 and 20 patients, respectively.

In conclusion, the schedule applied in the present study led to a 3 yr PFS of 35%. Surprisingly, no significant outcome difference was found between the younger and the elderly patients. The majority of treatment failures were due to early progressive disease under induction therapy. Although the follow-up of our study is short, de-escalation of induction treatment intensity by introduction of a less toxic agent as temozolomide, and its subsequent use in a maintenance schedule may explain a possible survival benefit of this strategy in elderly patients.

No relevant conflicts of interest to declare.