Management of secondary central nervous system involvement in systemic aggressive B cell lymphoma using R-MIADD chemotherapy: a single-center experience

Introduction Secondary central nervous lymphoma(SCNSL) was defined as lymphoma involvement of both within and outside CNS at initially diagnosis or CNS relapse of a systemic disease. The prognosis of SCNSL was poor and the most appropriate treatment remained unestablished. Methods We conducted a retrospective study addressing the feasibility of R-MIADD regimen which comprised rituximab, high dose methotrexate, ifosfamide, cytarabine, liposomal formulation of doxorubicin, dexamethasone in 19 consecutive SCNSL patients. Results Nineteen SCNSL patients with newly diagnosed CNS lesions were included with median age of 58 years (range 20 to 72 years). Eleven out of 19(57.9%) patients achieved complete remission(CR) and 2(10.5%) patients achieved partial remission by the end of induction treatment, the overall response rate (ORR) was 68.4%. The median follow-up time after the onset of CNS was 11.1 (3.2-35.5) months, the median progression-free survival after CNS was 28.0 months (95% CI: 11.0-44.9), and the median overall survival after CNS were 34.5months, by the time of this report, 8 patients remained CR. Treatment-related deaths was found in only one patient. Conclusions This is the largest series of SCNSL patients in China, and these date underscore the feasibility and efficacy of R-MIADD as induction treatment of SCNSL, further investigation is warranted.

Background Secondary central nervous system lymphoma (SCNSL) refers to secondary involvement of brain, eye, spine, meningeal by systemic lymphoma. [1,2] It's a devastating complication of systemic lymphoma which occurred in 5-10% diffuse large B cell lymphoma (DLBCL) patients with very few long-term survivors under conventional treatment. [3] SCNSL can occur in combination of systemic disease or presented as an isolated relapse. The prognostic model to assess the risk of CNS disease included the International Prognostic Score, involvement of the kidneys and/or adrenal gland. [4,5] In addition, dual expression of MYC and BCL2, [6] breasts/testis involvement [7] were also considered substantial risk factors of SCNSL according to recent researches. The prognosis of SCNSL was poor comparing to PCNSL, in the current largest cohort of SCNSL, less than half patients reached complete remission by the end of the induction treatment, and the median post-CNS overall survival(OS) was only 3.9 months. [8] Therefore, further explorations on the treatment of SCNSL especially initial induction are needed to improve CR rate and improve outcomes.
High dose Methotrexate(HD-MTX) based regimens were the most commonly used therapy, HD-MTX in combination with high dose cytarabine(Ara-C) can improve survival significantly. Procarbazine, etoposide, ifosfamide(IFO), thiotepa, carmustine and other drugs which can cross the blood-brain barrier (BBB) were also included in combination with HD-MTX and/or Ara-C to further improve outcomes. Doxorubicin which is a vital component for the treatment of systemic DLBCL, was not included in regimens primary central nervous system lymphoma(PCNSL) for its incapability of penetrate the BBB, [9] but with an alternative liposomal formulation, doxorubicin can overcome this intervention and was proved an promising drug for CNSL in several studies. [10] In addition, high dose chemotherapy followed by autologous stem cell transplantation (ASCT) were introduced and showed benefit effect. However, ASCT was offered to only a small part of younger patients with favorable response to induction treatment. Therefore, several studies explored the effectiveness of non-transplant regimens in SCNSL and indicated that this can be an effective treatment for SCNSL. [11][12][13] Nijland et al retrospectively investigated the outcome of SCNSL patients treated with HD-MTX and R-CHOP(rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone regimen) without ASCT, the 3-year OS was 49% which was comparable to patient outcomes with more intensive treatment including ASCT. [11] Moreton et al carried out a pilot study using IDARAM regimen in 16 SCNSL among which 12 achieved CR, seven remained CR at median follow-up of 24 months. [14] We observed an promising response in SCNSL treated with the combination of rituximab, HD-MTX, ifosfamide, cytarabine, liposomal formulation of doxorubicin, thus we retrospectively analyze the effectiveness of this regimen in order to explore novel treatment combination for SCNSL patients.

Methods And Materials
Study design and patient identification

Statistical analysis
Survival analyses were performed using the Kaplan-Meier methodology with SPSS statistics 24.0 software. Univariate Cox regression was used to analyze the impact of baseline characteristics on PFS and OS. There were not enough subjects to perform reliable multivariate analysis. All tests were two sided and a p-value of < 0.05 was considered statistically significant.

Results
Clinical characteristics of systemic disease in SCNSL patients (Table 1)   The median age at onset of CNS disease was 59 (20-76) years. The most common symptoms at the initial of CNS disease were increased intracranial pressure symptoms and dizziness, which were seen in five(26.3%) SCNSL patients respectively. Other symptoms included blurred vision(n = 2), limb weakness(n = 2), seizure(n = 1), somnolence(n = 1) and focal neurological deficits(n = 3). For radiological features, brain parenchymal lesions were found in all nineteen patients, six (31.6%) were single lesion and in thirteen (68.4%) patients the lesions were multiple; Ten (47.4%) had tumors involving the deep part of the brain including cerebellum, basal ganglia, corpus callosum, and brain stem.
As for diagnostic approaches, four(21.1%) patients were diagnosed with typical radiological manifestations on enhanced MRI, ten(52.6%) were diagnosed with stereotactic biopsy and four ( (Table 3). Patients with SD and PD proceeded with whole brain radiotherapy(WBRT), one patient with PR also turned to WBRT due to financial difficulties. All 5 patients achieved CR after WBRT. The median follow-up time after the onset of CNS disease was 11.1 (3.2-35.5) months, the median post CNS PFS was 28.0 months (95% CI: 11.0-44.9) (Fig. 2), and the post CNS OS was 34.5months (Fig. 3). One patient with refractory disease died during the induction treatment due to myelosuppression and severe pneumonia after chemotherapy, which resulted in toxic shock of infection. Of 11 patients who obtain CR with R-MIADD alone, 3 had isolated relapsed in CNS (2 new disease and 1 relapse disease initially) of which 1 relapse in spine 2 in brain parenchymal, and 1 patients died of progression of CNS disease (Fig. 4).

Discussion
With the emerging of novel treatment strategies, more regimens were introduced to CNS lymphoma. However, comparing to the general favorable treatment response and survival of PCNSL, SCNSL have inferior outcome and remained a fatal complication of aggressive Bcell lymphoma with a post CNS OS between 3.9-7.2months according to previous reports.
[ [17][18][19] In this research, we use the combination of R-MAIDD to treat SCNSL patients with an ORR of 68.4%, and post CNS PFS 28.0months(95%CI: 11.0-44.9), this result indicated that non-transplant regimen of R-MIADD could be a potential effective treatment for SCNSL.
There's no current randomized study define the optimal regimen of SCNSL due to the In our research, the regimen contain HD-MTX/cytarabine/IFO, in addition to liposomal doxorubicin, but the dose of all regimen was reduced comparing to the study mentioned above which lead to better treatment response without additional treatment toxicity. By the end of the induction therapy, 57.9% patients achieved CR, post CNS PFS was 28.0months(95%CI: 11.0-44.9) which was comparable to ASCT containing regimen. Much longer survival than median PFS were seen in two patients(Patient 6 and 7) at the age of 39 and 20, there's previous reports indicated that long-term survival was seen in a small albeit clinically relevant proportion of younger SCNS patients, younger patients may have better prognosis despite the overall disappointing survival of SCNSL. [20] In this study, patients with new disease possessed higher CR rate than relapse/refractory patients(70% vs 50% and 40%). We postulated this due to patients with relapse and refractory disease were more likely to be drug resist and more likely to develop treatment toxicity, this indicate that SCNSL patients require more individualized treatment regarding their disease condition and regimen prior to CNS disease. In addition, timely treatment was crucial, the only treatment related death in our study were found in an refractory SCNSL patient with atypical MRI manifestation who were initially diagnosed as anti-NMDAreceptor encephalitis, while treated by high dose corticosteroids, his neurological symptoms continuous deteriorate, by the time the patients was diagnosed SCNSL by biopsy, he was already in coma and had hypostatic pneumonia making treatment option extremely challenging. Eventually, he received reduced dose R-MIADD and became conscious after the first cycle, therefore we proceeded with the second cycle of R-MIADD.
Unfortunately, even with carefully supportive treatment, the patient died of severe sepsis.
This case suggested the significance of early recognize and early CNS targeting treatment, for patients with inferior ECOG-PS score, even they can response to CNS targeting treatment, adverse event of treatment may greatly influence outcome of the patient.
Benefit from the use of Rituximab in PCNSL is controversial. [25] In a meta-analysis published in 2019, two randomize control studies were included, and it was concluded that Rituximab could improve PFS, but without significant improvement in OS (HR 0.76; 95%CI 0.52-1.12, low certainty); Nevertheless, the majority research on SCNSL treatments included Rituximab (some were as part of the peripheral lymphoma treatment). [11-13, 20, 22] T.C El -Galaly et al concluded that Rituximab may reduce the risk of death in patients with isolated SCNSL, they also emphasized that in this subgroup, Rituximab plays a role even comparable to ASCT. [8] Thus we also included Rituximab as part of SCNSL treatment.
In conclusion, treatment of SCNSL has always been challenging, this is current largest cohort report of SCNSL treatment in Chinese population, our single center experience indicated that R-MIADD were effective in SCNSL, the outcome of patients receiving R-MIADD were comparable to previous reports of ASCT contained regimens. This study naturally has some limitations for it's a retrospective study with relative small sample size. Larger cohort prospective studies were needed to define the most effective treatment strategies in SCNSL.

Conflict of interest:
The authors declare that they have no competing interests.

Ethics approval and consent to participate
Ethical approval was provided by Beijing Tiantan Hospital Ethics Committee, Capital medical university (Ethical approval reference number: KYSB2016-170).

Consent to participate
Informed consent was written obtained when patients were admitted to Department of

Consent for publication
Not applicable.

Availability of data and materials
The datasets supporting the conclusions of this study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This study was supported by the Natural Science Foundation of Beijing Municipality(7172071) and the National Natural Science Foundation of China (81500157).

Author's contributions
LYB and WYC designed the study; JN provided the patient samples; SSJ revised neuroimaging; WYC analyzed the data and wrote the manuscript; SXF, CQ, ZH and QJ performed the experiments; BXY, XRX, CYD, LQ and WYL collected and analyzed the data; and all the authors have read the manuscript and approved its submission.
system lymphoma-A systematic review and meta-analysis. Hematological oncology 2019. Figure 1 Course of therapy, responses and clinical outcomes. CR, complete response; PR, partial response; PD, progressive disease; SD, stable disease; WBRT, whole brain radiotherapy.

Figure 2
Post CNS progression free survival of 19 SCNSL patients.  Swim lane plot of treatment duration and response for all patients. CR, complete response; PR, partial response; PD, progressive disease; SD, stable disease.