Blinatumomab in Relapsed/Refractory Burkitt Lymphoma

Simple Summary 20–40% of patients with Burkitt lymphoma (BL) have relapsing or refractory (r/r) disease, and standard treatment for such patients is poorly established. An unexplored treatment option is the bispecific T-cell engager blinatumomab, as used for the treatment of r/r and minimal residual disease (MRD) positive B-cell precursor acute lymphoblastic leukemia (BCP-ALL). So far, data on the use of blinatumomab in r/r BL are limited. In this retrospective multi-center case series, we investigated blinatumomab treatment in nine patients with r/r BL after 1–3 previous therapy lines. Data on safety and efficacy were collected. No high-grade (≥grade 3) adverse effects (AEs) occurred, and use of blinatumomab was found to be safe. The best response to blinatumomab and survival data varied considerably among patients, but with five from nine patients responding, blinatumomab seems to have activity in patients with r/r BL. Our data suggest that blinatumomab could be further explored in r/r BL. Abstract In patients with relapsed/refractory Burkitt lymphoma (r/r BL), overall survival (OS) is poor, and effective therapies and evidence for the best therapy are lacking. The monoclonal antibody blinatumomab may represent a novel option. However, only limited data on the use of blinatumomab in r/r BL are so far available. This multi-center, retrospective case series investigated nine patients with r/r BL treated with blinatumomab. The safety of blinatumomab was assessed with respect to frequency and severity of adverse effects (AEs) infections, cytokine release syndrome (CRS) and neurotoxicity. Progression-free survival (PFS), OS and overall response rate (ORR) were analyzed to assess efficacy. No AEs > grade 2 occurred, and AEs were generally treatable and fully reversible. The best response to blinatumomab was complete remission in 3/9 patients and partial remission in 2/9, whilst 4/9 presented with progressive disease. Median PFS and OS were 2 and 6 months, respectively, ranging from 5 days to 32 months and 11 days to 32 months, respectively. Blinatumomab treatment was a successful bridging treatment to stem cell transplantation in 3/9 patients. The response to blinatumomab varied widely, and only one patient survived longer term, but activity in patients with r/r BL was evident in some patients, with its use being safe, warranting its prospective investigation.


Introduction
Due to its high proliferation rate, Burkitt lymphoma (BL) is a very chemosensitive disease and the majority of patients can be cured with an intensive chemoimmunotherapy refractory disease. Patients had to be at least 18 years old and had to have relapsed or refractory BL after at least one first-line therapy. Relapse and refractoriness were evaluated according to bone marrow biopsy and PET-CT. No upper limit of prior therapy lines was set for inclusion, and no specific requirements for types of prior treatments were determined. All patients were described as one group without a control or comparison group. Follow-up was conducted until death or February 2022, whichever was first. All data were collected by the authors from December 2021 until February 2022 from chart review using EMR. Detailed characteristics of included patients at diagnosis are summarized in Table 1.

Treatment
All patients were treated off-label, according to the manufacturer's instructions. Dosing schedule was identical to the phase 2 study, investigating the dosing of blinatumomab in patients with diffuse large B-cell lymphoma. A weekly stepwise dose escalation, until the target dose was reached (9, 28 and 112 mcg/day) and was preferably used instead of flat dosing (112 mcg/day from day one) [18]. One cycle of blinatumomab lasted 28 days (d) and started with 9 mcg/d in the 1st week, then the dose was increased to 28 mcg/d in the 2nd week, and in the 3rd week the target dose of 112 mcg/d was reached. This therapy scheme was the planned treatment in all patients of this cohort. Due to its short halflife, blinatumomab was administered as a continuous intravenous infusion via portable pumps [12]. The blinatumomab treatments of this cohort occurred between January 2018 and August 2021.

Assessment of Safety and Efficacy
Safety was evaluated according to the frequency of infection, CRS, and neurotoxicity, and the severity of CRS and neurotoxicity was graded from 1-5. Efficacy was evaluated for the endpoints overall response rate (ORR), OS, and PFS since the start of blinatumomab therapy. The response to blinatumomab was evaluated according to bone marrow biopsy, as well as radiologic assessment, preferably PET-CT, using Lugano criteria. Data on safety and efficacy were collected from medical reports.

Measurements and Definitions
Initial staging of patients was according to the Ann Arbor classification, and the International Prognostic Index (IPI) was determined for risk assessment [19,20]. Bulky disease was defined as lymphoma size >10 cm or mediastinal width >1/3 of chest diameter. Extranodal involvement was evaluated according to PET-CT, and infiltration of bone marrow and central nervous system (CNS) were assessed by biopsy or puncture of cerebrospinal fluid, respectively. The grading of CRS (grade 1-4) and neurotoxicity (grade 1-4) was obtained according to the ASTCT consensus grading for CRS and ICANS [21].
OS was defined as the duration from the start of blinatumomab therapy until death of any cause or last follow-up, whichever occurred first. PFS was defined as the duration from start of blinatumomab therapy until progression/relapse of disease or last follow up, whichever occurred first. The overall response rate (ORR) comprised patients achieving a complete remission (CR) or partial remission (PR) of disease following blinatumomab treatment.

Statistical Analysis
OS and PFS were evaluated according to Kaplan-Meier and their graphical representation was generated with GraphPad Prism ® (San Diego, CA, USA). Statistical calculations were made with Microsoft Excel and data of median was rounded to whole numbers. Data cut-off was on 4 March, 2022.

Patient Characteristics at First Diagnosis
At the University Hospital Bern, Switzerland, five patients met the criteria for inclusion. The study cohort was enlarged with four patients from three other Swiss hospitals (Lausanne, Lucerne, Basel), resulting in a total number of nine patients (9/9) in the study cohort. Patients' characteristics at first diagnosis are summarized in Table 1. Five of nine patients (5/9) were male, four were female. The median age at first diagnosis was 33 years, ranging from 25-62 years. All nine patients had sporadic BL, and MYC translocation was identified in all patients [22,23].

Previous Lines of Treatment before Blinatumomab
The treatment given before blinatumomab is summarized in Table 2. Five patients (5/9) had one, two patients (2/9) had two, and two patients (2/9) had three different lines of treatment before blinatumomab. The median number of previous treatment lines was one (range 1-3); all patients had previous rituximab exposure.
The best result of first-line therapy was CR in three patients (3/9) and PR in six (6/9) patients. The disease subsequently relapsed in seven patients (7/9) and was refractory to therapy in two patients (2/9). The median duration from start of first-line therapy to relapse/refractoriness was 5 months, with a range from 3 to 11 months.

Blinatumomab Therapy
None of the previous treatment lines achieved a longer remission duration; thus, blinatumomab was given in rapidly progressive patients, reflecting the aggressive nature of the disease. Eight patients (8/9) started blinatumomab therapy in the setting of relapsed/refractory disease with progressive Burkitt lymphoma. In one patient, 2nd line chemotherapy led to a clearance of BL in the bone marrow with persistent disease documented by PET-CT in lymph nodes and spleen.
As planned, blinatumomab therapy was started at a dose of 9 mcg/d in all nine patients. In seven patients (7/9), the dose was increased to 28 mcg/d in week 2 of therapy, and in six (6/9) patients the target dose of 112 mcg/d was reached in the third week. The reason for not proceeding to the highest dose level in these two patients was the progression of the disease and subsequent initiation of palliative treatment in five patients, whereas dose was maintained in one patient at 28 mcg/d, based on decision of the treating physician, although blinatumomab was well tolerated. In one patient, two episodes of neurotoxicity led to a dose reduction from 112 mcg/d to 56 mcg/d, firstly, and then to a reduced dose of 28 mcg/d.
The duration of blinatumomab therapy was adapted to the individual situation (response to blinatumomab, disease progression, planned consolidation therapy), which led to administered cycles ranging from one to five cycles. Due to the rapid progression of the disease, the first cycle of therapy needed to be stopped in two patients. The median of given blinatumomab cycles was three, and except for one cycle in one patient (42 days), all cycles lasted 28 days. In patients with more than one cycle (6/9), there was a preplanned twoweek therapy-free interval between cycles. During blinatumomab, two patients received additional radiotherapy.
Blinatumomab was stopped in three patients (3/9) when allogeneic HSCT was started and because of the progressive disease in the other six patients. Data on blinatumomab therapy are summarized in Table 3.

Adverse Effects Infection, CRS and Neurotoxicity
Infections during blinatumomab treatment occurred in four (44%) patients. Identified pathogens and manifestation are listed in Table 4. All infections were manageable with adequate antibacterial or antiviral therapy. CRS was observed in five (5/9) patients. The severity of CRS was classified as grade 1 in one patient (1/9) and grade 2 in the remaining four patients (4/9). One patient had two episodes of CRS, both with a severity of grade 2. According to the grading, all patients had fever (body temperature ≥38 • C) with/without constitutional symptoms, and those with grade 2 additionally had hypoxia (requiring low-flow oxygen delivery (≤6 L/min)), and/or hypotension (not requiring vasopressors) related to CRS [21]. No patient had severe CRS (≥ grade 3). In all five affected patients, CRS led to hospitalization, but no ICU (intensive care unit) admission was required. All patients with CRS were treated with steroids, and blinatumomab infusion was temporarily interrupted until symptoms resolved completely. Three patients (3/9) with CRS grade 2 received tocilizumab.
The symptoms of neurotoxicity caused by blinatumomab occurred in three (3/9) patients, whereas one patient had two episodes of neurotoxicity. Considering all four episodes in the three patients, three were grade 1 and one episode was grade 2. The symptoms of grade 1 neurotoxicity were aphasia, headache and tremor, and in grade 2 they were neurotoxicity dysarthria, emesis, paresthesia, headache, apraxia/ataxia and weakness. The grade 2 neurotoxicity required hospitalization but no ICU admission. Neurotoxicity was treated with a temporary interruption of blinatumomab infusion until symptoms resolved, and two out of the three patients received steroids. In all three patients, the symptoms of neurotoxicity were completely reversible. Data on adverse effects are depicted in Table 4.

Outcome of Blinatumomab Therapy
Three patients (3/9) achieved a CR in the bone marrow and PET-CT as best response, and two (2/9) had a PR with either PET Deauville score >3 or persisting disease in the bone marrow. CR in bone marrow was determined by morphologic bone marrow assessment (biopsy and aspirate). Unfortunately, no flow-cytometric MRD assessments were available in these three patients. In four patients (4/9), BL was progressive during blinatumomab therapy. Adding up the patients presenting with CR or PR, the ORR to blinatumomab was 5/9 patients. At the end of blinatumomab therapy, two patients (2/9) had achieved second CR, one patient (1/9) was in PR and six (6/9) had progressive disease.
Eight patients (8/9) in this cohort relapsed during or after the completion of blinatumomab therapy, with the median duration from the start of blinatumomab until relapse being 2 months (range 5 days to 13 months). The median PFS for the entire cohort was 2 months, ranging from 5 days to 32 months. All eight relapsed patients (8/9) died: seven (7/9) due to progression of disease and one (1/9) of acute liver failure, related to graftversus-host disease (GvHD) after allogeneic HSCT. Outcomes of the individual patients are summarized in Table 5. The median OS of the cohort following initiation of blinatumomab was 6 months, with a wide range from 11 days to 32 months. PFS and OS data are provided in Figure 1 and Table 6. Median duration until death since start of blinatumomab, m (range) 6 (11 d-16 m) Median OS, m (range) 6 (11 d-32 m) Median OS, censored at allogeneic HSCT, m (range) 4 (11 d-7 m) Median follow up since start of blinatumomab, m (range) 6 (11 d-32 m) CR: complete remission, PR: partial remission, PD: progressive disease, d: days, m: months, PFS: progression free survival, HSCT: hematopoietic stem cell transplantation, OS: overall survival, GvHD: graft-versus-host disease.    At the end of blinatumomab therapy, four patients (4/9) underwent allogeneic HSCT. Accordingly, Figure 2 provides survival outcomes censoring patients at the time of allogeneic HSCT. Hereby, OS was defined as the time from the start of blinatumomab until death or allogeneic HSCT, whichever occurred first, and PFS was defined as the time from the start of blinatumomab until relapse/progression of disease or allogeneic HSCT, whichever occurred first.     In eight patients (8/9), the follow-up was terminated by death, and in the 9th patient it ended in December 2021 at last follow-up, 32 months after start of blinatumomab. At this time, this patient was still in CR. The median duration of follow-up since the start of blinatumomab was 6 months (range 11 days to 32 months).

Treatment after Blinatumomab Therapy
Five patients (5/9) received one line of further therapy after the end of blinatumomab. One patient had radiotherapy and four patients (4/9) had allogeneic HSCT. At the time of allogeneic transplantation, two patients were in CR, one in PR and one had PD. All four patients received BEAM-Flu without TBI as conditioning regimen and immunosuppression was based on cyclosporine/steroids. Two more patients had been planned to undergo allogeneic HSCT, but relapse prevented these patients from transplant. Four relapsing patients (4/9) had no further therapy after blinatumomab except palliative symptomatic treatment. The final outcome of the treatments after blinatumomab was unfavorable. Three of the four patients with allogeneic HSCT relapsed and either died from progression of disease (two patients) or from acute liver failure related to GvHD after allogeneic HSCT (one patient). The one patient with radiotherapy after blinatumomab died from the progression of the disease. One patient with allogeneic HSCT is still alive and had been relapse-free for 32 months at the last follow-up. These data are summarized in Table 7.

Discussion
So far, there has been no effective and standardized treatment for r/r BL. Discussed options are salvage chemotherapy; autologous or allogeneic HSCT, PI3K-, CDK6-and MYC-inhibitors; and CAR T-cells [8][9][10]. The monoclonal antibody blinatumomab may provide another option for this situation. Blinatumomab has been extensively investigated in the B-ALL settings, where its efficacy for the r/r and MRD-positive situation was confirmed by many studies [25][26][27][28][29][30]. In contrast, data on its application in the r/r BL setting are scarce, with one study describing the in vitro activity of blinatumomab against BL cell lines and one report of three r/r BL patients treated with blinatumomab [16,17]. Aiming to improve insights in the potential of blinatumomab for this indication, we here retrospectively analyzed data in nine patients with a focus on data of safety and efficacy.
Overall, the tolerability of blinatumomab, administered in a weekly dose escalation (1st week 9 mcg/day, 2nd week 28 mcg/day) until the target dose of 112 mcg/day (3rd week) was reached, was favorable, and no toxicological mortalities occurred. The administration of blinatumomab via portable pumps worked without complications and could be performed mostly in an outpatient setting. Most patients of the cohort (6/9) experienced at least one of the AEs infection, CRS, and neurotoxicity, as well as 2/9 showing all three. CRS was the most frequently observed AE (5/9). A third of patients (3/9) suffered from none of these adverse effects; however, it must be considered that 2/9 received blinatumomab for only 7 days, whereas one cycle usually lasts 28 days. Interestingly, the only patient receiving blinatumomab for a longer period (12 weeks) and with none of the adverse effects assessed was the one who never received the target dose of 112 mcg/d, instead only receiving 28 mcg/d as the highest dose. In one patient, there was a need for dose reduction from 112 mcg/day to 56 mcg/day, firstly, and then 28 mcg/day due to two episodes of neurotoxicity with aphasia in the first episode and then emesis, paresthesia, headache, apraxia/ataxia and weakness in the second episode. No patient experienced CRS or neurotoxicity higher than grade 2, so there were no high-grade (≥ grade 3) AEs.
The ORR to blinatumomab in r/r Burkitt lymphoma was encouraging with 5/9 patients responding in this study. In more detail, there were 3/9 with CR and 2/9 with PR, whereby the best response was observed in patients with infiltration of the bone marrow (4/9 with CR or PR). Thus, blinatumomab showed the highest activity in patients with Burkitt leukemia in our cohort. However, it must be mentioned that in 4/9 patients, no response to blinatumomab was seen and that responses were mostly transient, with 4/5 responders relapsing within 2-13 months after the start of blinatumomab. Thus, when the whole cohort was considered, the median PFS and OS were rather short at two and six months. Still, responders showed a longer PFS and OS, as compared to the non-responders. Strikingly, 3/9 of the cohort were successfully bridged to allogeneic HSCT, which indicates that blinatumomab may be useful to prepare patients for an early allogeneic HSCT as potential curative therapy in this adverse situation. Finally, the outcome of the patients in our cohort was poor, with only one patient surviving longer term (last follow up at 32 months). This is consistent with the adverse prognosis of patients with r/r BL observed in previous studies with different salvage chemotherapy regimens and HSCT [7, 8,31]. According to our results, blinatumomab seems to have activity in patients with r/r BL, but we do not think it is suitable as a curative monotherapy approach in patients with r/r BL, especially given its possible toxicity and cumbersome mode of administration (continuous intravenous infusion), and, in light of emerging CAR-T-cells, its role in r/r BL should be questioned critically.
The observation that the individual activity of blinatumomab was very variable in this study cohort might lead to the hypothesis that response to blinatumomab depends on individual patient or disease characteristics. Since the best effect of blinatumomab was seen in patients with bone marrow infiltration >25%, blinatumomab was more appropriate for patients with Burkitt leukemia in our study, as compared to non-leukemic patients.
In the context of current research, this case series can be compared with the study by Duell et al. who also investigated the safety and activity of blinatumomab in three patients with r/r BL [17]. With nine patients, the cohort of our case series was three times larger, making it the largest study of blinatumomab in patients with r/r BL to date. Concerning safety and efficacy, the two studies agree that blinatumomab could be applied safely, that there was a high interindividual variability of response and that the activity of blinatumomab in patients with r/r BL could be observed.
The use of blinatumomab should also be compared to the only officially approved therapies for r/r BL, which is so far salvage chemotherapy (e.g., Hyper-CVAD (cyclosphosphamide, vincristine, doxorubicin, dexamethasone), (R-)ICE (rituximab), ifosfamide, carboplatin, etoposide) and EPOCH (etoposide, prednisone, vincristin, cyclophosphamide, doxorubicin). The outcomes of such approaches were analyzed by Short 31]. The comparison of this quantitative data with our study is not appropriate due to the small number of patients in our study, but it can be compared qualitatively. Blinatumomab has some advantages compared to chemotherapy: it penetrates the CNS, so no additional intrathecal prophylaxis is needed; its short half-life allows AEs to be simply treated with a temporary interruption of intravenous infusion; and it can be applied via portable pumps in an outpatient setting, thus guaranteeing a higher quality of life [11,12,32].
Since the number of patients in this cohort was small, no definitive conclusions should be drawn. However, considering the limited experience with blinatumomab in r/r BL in the previous literature, this study makes an important contribution to the amount of available data. Given the rarity of BL, and even more so in the refractory or relapsed situation, multi-center and international collaboration efforts would be needed to investigate blinatumomab in r/r BL in a representative number of patients.

Conclusions
The AEs infection, CRS and neurotoxicity occurred in a relevant proportion of patients (infections 4/9, CRS 5/9 and neurotoxicity 3/9) but no AE was severe (highest was grade 2), and all were treatable and reversible. Therefore, in our study, we rate the use of blinatumomab in r/r BL to be safe. The response to blinatumomab and duration until relapse and death varied widely among patients in the cohort, but with an ORR of 5/9 patients and 2/9 achieving a CR at the end of blinatumomab, the activity of blinatumomab as monotherapy in r/r Burkitt lymphoma/leukemia was clearly documented in this study. Blinatumomab proved valuable for bridging patients (3/9) to allogeneic HSCT as consolidation therapy. Unfortunately, the outcome of our cohort was unfavorable with 8/9 patients relapsing and dying within 16 months of the initiation of blinatumomab and only one patient being alive at the last follow up (32 months).
Thus, we doubt the use of blinatumomab as a curative monotherapy with long-lasting effects. Rather we see blinatumomab as a possible part of combined therapy, particularly as a bridging therapy until HSCT in curative intent. Further investigation should be made in combination with other treatments in the future by collaborative approaches, with the aim of improving the outcomes of patients with r/r Burkitt lymphoma.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee Bern, Switzerland (#2018-00162).

Informed Consent Statement:
Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.