Search

Primary central nervous system lymphoma (PCNSL) is a rare form of extranodal non-Hodgkin lymphoma that is typically confined to the brain, eyes, and cerebrospinal fluid (CSF) without evidence of systemic spread. PCNSL is an uncommon tumor, and only four randomized trials and one phase III trial have been completed so far, all in the first-line setting. The prognosis of patients with PCNSL has improved during the past few decades with the introduction of high-dose methotrexate (HD-MTX), which now serves as the backbone of all first-line treatment regimens. Despite recent progress, results after treatment are durable in half of patients, and therapy can be associated with late neurotoxicity. Novel insights into the pathophysiology of PCNSL have identified the B-cell receptor (BCR) pathway as a key mechanism in the pathogenesis of PCNSL. The use of novel agents targeting components of the BCR pathway, namely the Bruton tyrosine kinase (BTK) inhibitor ibrutinib, and immunomodulatory drugs (IMIDs) like lenalidomide and pomalidomide, has so far been limited to patients who have recurrent/refractory PCNSL with promising high response rates. Within the past 5 years, there has been a peak in clinical trials investigating small molecules and novel reagents in the recurrent/refractory setting, including immune checkpoint inhibitors, IMIDs, and BTK and PI3K/AKT/mTOR inhibitors.

2501 Background: MTX-based chemoradiotherapy is effective in PCNSL, but carries a risk of severe neurotoxicity (NT), especially in the elderly. In a phase II single arm study, R-MPV-A chemotherapy was combined with substantially reduced doses of radiotherapy (23.4 Gy), achieving prolonged progression free survival (PFS) and overall survival (OS) with acceptable NT. Because R-MPV-A had never been tested without radiotherapy, we conducted a randomized study to determine if the low doses of radiation played a role in the observed disease control, and to characterize NT as compared to chemotherapy alone. Methods: Patients were stratified by MSK RPA class and randomized to receive R-MPV-A with LD-WBRT (chemoRT arm) versus R-MPV-A alone (chemo arm). MTX dose was 3.5g/m2 infused over 2 hours. Filgrastim and pegfilgrastim support was given to all patients. LD-WBRT dose was 23.4 Gy (1.8 Gy X 13). The primary endpoint was intent-to-treat (ITT) PFS. A sample size of 89 would provide 80% power to detect a hazard ratio (HR) of 0.63, with one-sided alpha level of 0.15. Results: A total of 91 patients were randomized, of whom 4 were ineligible. Among eligible patients, 43 were enrolled in the chemoRT arm and 44 in the chemo arm. Median age was 66 (chemoRT) and 59 (chemo). Median KPS was 80 for both arms. Response rates following R-MPV were 81% (chemoRT) and 83% (chemo). In the chemoRT arm, 37 patients (86%) received LD-WBRT. After median follow-up of 55 months (m), the median ITT PFS was 25 m in the chemo arm and not reached in the chemoRT arm (HR 0.51; 95% CI [0.27, 0.95]; p = 0.015). The 2-year PFS was 54% (chemo) and 78% (chemoRT). Salvage radiotherapy has been given to 11 patients in the chemo arm. Median OS was not reached in either arm, with data still maturing. In both arms, most common grades 3 or 4 toxicities were anemia (27%), lymphopenia (41%), neutropenia (35%), thrombocytopenia (26%), ALT (23%) and AST (13%). One patient died from sepsis (chemo arm). As per investigators’ assessment, the rate of clinically defined moderate to severe NT was 11.4% (chemo) and 14% (chemoRT), p = 0.75. Conclusions: The study met the primary endpoint, demonstrating the addition of LD-WBRT to R-MPV-A improves PFS in newly diagnosed PCNSL. As per investigator’s assessment, NT rates were not statistically significantly increased, but further neuropsychological testing and neuroimaging analyses are ongoing to characterize cognitive decline and how it compares to other consolidation treatments. Clinical trial information: NCT01399372 .

7516 Background: PCNSL is a diffuse large B cell lymphoma confined to the CNS. Pomalidomide (POM) is a novel immunomodulatory agent with excellent CNS penetration (~40%) based on CNS PK analysis in rats and pre-clinical therapeutic activity against CNS lymphoma. Methods: A Phase I clinical trial was undertaken to determine the maximal tolerated dose (MTD) of POM, safety profile and overall response rate (ORR). Treatment consists of POM daily for 21 days every 28 days in combination with dexamethasone 40 mg PO weekly for two cycles followed by POM alone in subsequent cycles until progression or intolerance. 4 dose escalation levels of POM (3mg, 5mg, 7mg, and 10 mg) were planned. Thromboprophylaxis with oral anticoagulant or aspirin was required. MTD determination has been completed and expansion of the MTD cohort is ongoing. Therapeutic responses were evaluated per the international PCNSL collaborative group (IPCG) criteria after 2 cycles of treatment. The trial is registered with ClinicalTrials.gov (#NCT01722305). Results: 21 of 25 patients accrued were eligible for assessment. The MTD was determined to be 5 mg qd for 21 days every 28 days. Two DLTs were seen at dose level 3 (Grade 3 dyspnea and grade 4 thrombocytopenia). One DLT was seen in the expanded MTD cohort (Grade 4 neutropenia and lymphocytopenia). ORR for the study (9/21) was 43% (95% CI- 22%, 66%) with 4 CR, 1 CRu and 4 PR. 3 responders completed 2, 4, and 6 cycles before progression. 6 responders have completed 4, 5, 6, 10, 12, and 32 cycles and remain on treatment. ORR for the MTD dose level was (5/12) 42% (95% CI- 15%, 72%) with 3 CR and 2 PR. 2 patients had stable disease (SD). Pseudo-progression was seen in 1 patient. Overall, grade 3/4 toxicity was hematologic (neutropenia, anemia, and thrombocytopenia) in 38.1% and non-hematologic in 33.3% (fatigue, pneumonia, sepsis, syncope, dyspnea, hypoxia, respiratory failure and maculo-papular rash). Percent CSF/blood ratio of POM was determined to be 19% in 1 patient. Conclusions: Pomalidomide treatment is feasible with therapeutic activity against relapsed/refractory PCNSL and should be further developed. Clinical trial information: NCT01722305.

7515 Background: PCNSL is an aggressive primary brain tumor with median progression free survival (PFS) after upfront methotrexate-based chemotherapy of 2-3 years. Outcome and treatment options are poor for recurrent/refractory (r/r) disease. Ibrutinib has shown promising clinical response in Mantle cell lymphoma, CLL, Marginal Zone, and Waldenström. This trial investigates Ibrutinib in patients with r/r PCNSL and SCNSL. Methods: Eligible patients had r/r PCNSL or SCNSL, age≥18, ECOG≤2, normal end-organ function, and unrestricted number of CNS directed prior therapies. In patients with SCNSL disease, systemic disease needed to be absent. Results: Twenty-five patients were enrolled (3 at 560 mg; 22 at 840 mg). Median age was 68 (range 21-85); 15 were women. Median ECOG was 1 (0: 2, 1: 15, 2: 8). 64% had PCNSL and 36% SCNSL; 68% had recurrent disease. Seventeen had parenchymal disease, 3 isolated cerebrospinal fluid (CSF) involvement and 5 both. Seven grade 4 adverse events were observed in 7 patients neutropenia (in 3 patients), lymphopenia (2), sepsis (1), and ALT elevation (1). Fourteen patients developed 20 grade 3 toxicities, including lymphopenia in 5 patients, hyperglycemia in 3, ALT elevation in 2, thrombocytopenia in 2, lung infection in 2, AST elevation in 1, neutropenia in 1, urinary tract infection in 1, colitis in 1, febrile neutropenia in 1 and fungal encephalitis in 1. The most common toxicities at any grade were hyperglycemia, thrombocytopenia and anemia of which most were grade 1/2. No grade 5 events have been observed. After a median follow-up of 414 days (range 289-674), 22/25 patients were evaluated for response (3 did not complete at least 15 days of drug treatment). Over all response was 68% (17/22; 77% (17/22) in patient that completed at least 15 days of drug treatment) with 10 CR, 7 PR, 2 SD and 3 PD as best response. The median PFS is 4.6 months (5.4 months in patients that completed at least 15 days of drug treatment; longest: 15.3 months). The median overall survival has not been reached. Conclusions: Patients with CNS lymphoma tolerate Ibrutinib with manageable adverse events. Clinical response was seen in 68% of CNS lymphoma patients. Clinical trial information: NCT02315326.

2040 Background: There has been significant improvement in treatment outcomes of Primary Central Nervous System Lymphoma (PCNSL) at specialized centers over past decades, and it is unclear if these changes have translated to benefits in the general population. In this study, we utilized national databases to examine survival trends over time. Methods: Incidence rates were obtained from the Central Brain Tumor Registry of the United States (CBTRUS, 2000 – 2013) and 18 Surveillance, Epidemiology and End Results (SEER, 1973 – 2013) registries. Data for survival analysis was obtained from SEER and analyzed using SEER*STAT. To focus on non-HIV-associated PCNSL, patients with “other infectious and parasitic diseases including HIV” as cause of death and follow up were excluded. CBTRUS identified 19,027 patients over 13 years and SEER 6,765 over 40 years. Results: The annual incidence of PCNSL in 2013 was 0.4 per 100,000 population (CBTRUS/SEER). Incidence increased from 0.1 per 100,000 in the 1970s to 0.4 per 100,000 in the 1980s, correlating with an increase in the diagnosis of elderly patients, ≥70 (1973:0.2 vs 2013:2.1 – SEER). Incidence rates differed greatly between young and elderly patients (20-29: 0.08 vs 70-79: 4.32 – CBTRUS). The median overall survival (mOS) of all patients is 17 months (m) with no survival benefit based on sex. Survival doubled from 12.5 m in the 1970s to 26 m in the 2010s. There was a significant difference in survival based on age: < 50: 83 m vs 50-69: 25 m vs ≥70: 6 m (p-value < 0.0001). In patients < 50, mOS increased from 35.5 m in the 1970s to 134 m in the 2000s (mOS not achieved in 2010s). In patients 50-69, mOS increased from 8 m in the 1970s to 35 m in the 2010s. However, mOS in the elderly population, ≥70, has not changed in the last 40 years (6 m in the 1970s vs 7 m in the 2010s, p-value = 0.1). Conclusions: PCNSL is a disease that more frequently affects the elderly. Although overall survival has increased over the past 4 decades, reflecting current literature in PCNSL, survival in the elderly has not changed since the 1970s. Identification of this vulnerable patient population highlights the need for clinical trials targeting the elderly in hopes of improving treatment strategies and ultimately outcomes.

2033 Background: Treatment options for refractory high grade gliomas (HGG) are limited. Programmed cell death ligand-1 (PD-L1) expression has been reported in 0-61% of HGGs and therefore might be a suitable target in HGG. The purpose of this study was to describe safety and efficacy of PD-1 inhibition in patients with refractory HGGs. Methods: This IRB approved single center retrospective study at Memorial Sloan Kettering Cancer Center included pathologically confirmed HGG with an age ≥18 years who received a PD-1 inhibitor between 9/2014 and 10/2016 outside of a clinical trial. Results: Twenty five HGGs were identified. All patients received the PD-1 inhibitor pembrolizumab (pembro) as part of compassionate use. Median age was 49 years (range 30-72); 44% were men; 13 had glioblastoma (52%), 7 anaplastic astrocytoma (28%), 2 anaplastic oligodendroglioma (8%), 2 unspecified HGG (8%), and 1 gliosarcoma (4%). Patients received a median of 4 prior lines of therapy (range 1-9). Nineteen (76%) previously failed bevacizumab. Median baseline KPS was 80 (range 50-100). Concurrent treatment included bevacizumab in 17 (68%) or bevacizumab and temozolomide in 2 (8%) patients. Median number of doses administered was 3 (range 1-14). Treatment toxicity and response was assessed in 24 patients. PD-1 inhibitor related adverse events (AEs) included LFT elevations (33%), hypothyroidism (17%), diarrhea (17%), myalgias/arthralgias (13%), and rash (8%). Other common AEs were hyperglycemia, fatigue, thrombocytopenia, lymphopenia, headache, and nausea in the setting of concomitant therapy and additional supportive care (dexamethasone). Grade 3 AEs included seizure (4%), headache (4%), nausea (4%), and vomiting (4%). Best radiographic response was partial response (n = 2), stable disease (n = 5), and progressive disease (n = 17). Median progression free survival (PFS) was 42 days (range 7-282) and median overall survival was 121 days (range 15-415). Three patients (12%) had a PFS > 90 days; of these, 2 received single agent pembro. Conclusions: Patients with HGG had low response rates. However, a small number of patients had prolonged PFS. Pembro was tolerated with few serious AEs, even in patients receiving concomitant therapy.

e19020 Background: Ibrutinib (IBRU) is a Bruton tyrosine kinase (BTK) inhibitor that has been FDA approved for chronic lymphocytic leukemia (CLL), mantle cell lymphoma (MCL) and Waldenstrom's Macroglobulinemia (WM). BTK inhibition may contribute to immunosuppression through B- and T-cell inhibition, resulting in an increase in infections. We therefore characterized IBRU-related infections retrospectively. Methods: The study was an IRB approved retrospective review. Patients (pts) treated with IBRU between 4/2014-11/2016 who developed any infection were identified using ICD10 codes in a clinical database. Study population was defined using descriptive statistics. Results: 200 pts were identified: 78 pts had CLL (39%), 30 diffuse large B-cell lymphoma (15%), 28 MCL (14%), 19 WM (9.5%), 15 (7.5%) marginal zone lymphoma, 14 (7%) follicular lymphoma, 7 (3.5%) multiple myeloma, 7 (3.5%) T-cell lymphoma and 2 (1%) primary central nervous system lymphoma (PCNSL). Median age was 68 (range 28-96), 34% were men and median ECOG was 1. The majority of pts received IBRU as second line treatment (172, 86%) with a median of 2 prior lines of treatment; 23 pts (11.5%) were post transplant. Median IBRU dose was 420mg daily (range 140-840mg), administered a median of 316 days (range 3-1780). Single agent IBRU was used in 162 pts (81%). 105 pts (52%) developed an infection, with pneumonia (30%), upper airway infection (26%); skin infection (18%); and sinusitis (13%) being the most common. 10 (9.5%) had neutropenic fever and 1 (1%) cryptococcal pneumonia. Seven (7%) developed fungal infections, with invasive aspergillosis in 5 (5%). 34 (32%) developed ≥3 infections. 46 pts (44%) were hospitalized, 10 (9.5%) interrupted IBRU and 3 pts (2.9%) died due to infections (2 pneumonia; 1 neutropenic fever). The median time to infection after starting IBRU was 70 days (range 2-1261). The highest infection rate was seen in pts with PCNSL (100%, 2/2), MZL (67%, 10/14), followed by CLL (64%, 50/78). Conclusions: We identified infections in half of the pts treated with IBRU. Pts treated with IBRU should be monitored closely for the development of infections, particularly airway infections.

2046Background: PCNSL is an aggressive primary brain tumor. Outcome and treatment options are poor for patients with recurrent/refractory disease. Response rates range between 30-60% with a progression free survival (PFS) of 2-6 months. Ibrutinib has shown promising clinical response in some B-cell malignancies. This phase I trial investigates ibrutinib in patients with recurrent/refractory (r/r) PCNSL and secondary CNS lymphoma (SCNSL). Methods: Eligible patients had r/r PCNSL or SCNSL, age ≥ 18, KPS ≥ 50, normal end-organ function, and unrestricted number and type of prior therapies. In patients with SCNSL disease, systemic disease needed to be absent. Results: Ten patients were enrolled (3 at 560 mg; 7 at 840 mg). Median age was 70 (range 21-80); 7 were women. Median ECOG was 1 (0: 1, 1: 5, 2: 4). 70% had PCNSL and 30% SCNSL. Six had parenchymal disease, 3 isolated cerebrospinal fluid (CSF) involvement and 1 both. Patients were highly pretreated: 2 (both SCNSL) had prior stem cell transplant; 3 (2 PCNS...

2008 Background: In our previous study in newly diagnosed PCNSL, induction chemotherapy with MTX and cytarabine followed by consolidation HDC (carmustine, etoposide, cytarabine, melphalan [BEAM]) with ASCT without radiotherapy resulted in only 50% of pts transplanted, reflecting low efficacy of induction chemotherapy, and short intent-to-treat (ITT) median PFS (=6m). In this phase II trial, we sought to optimize this strategy by utilizing a more effective induction regimen (R-MPV) and a more aggressive HDC regimen (Soussain et al). Methods: Pts received 5-7 cycles of R-MPV (MTX: 3.5g/m2) and if a partial or complete response was achieved, HDC with thiothepa, cyclophosphamide and busulfan was given, followed by ASCT and no radiotherapy. The primary endpoint was ITT 1 year event-free survival (promising: 75%, non-promising: 50%; 90% power, significance=0.05). Follow-up included comprehensive neuropsychological evaluation. Results: Accrual has been completed (N=32 pts, median age 57 [range 23-67], median KPS=80). Following R-MPV, 17 pts achieved a CR, 13 pts a PR and two pts progressed. A total of 25 (78%) pts were transplanted; the reasons for not receiving transplant were progressive disease (N=2), poor performance status/ physician’s decision (N= 2), mobilization failure (N=1) and consent withdrawn (N= 2). One pt who withdrew consent relapsed and received HDCASCT for salvage. Two (8%) pts died from early complications of ASCT (Stevens-Johnson: one, sepsis: one) and one pt experienced a fatal late colitis of unknown etiology. In the ITT population, the median EFS and OS have not been reached after a median follow-up of 22 months. The 1 year EFS was 78% (95%CI 58-90) and the 2y OS was 76% (95% CI 54-89). No pt has developed delayed neurotoxicity. Conclusions: R-MPV induction regimen resulted in improved response rates, allowing 78% of pts to receive HDC-ASCT. Although more toxic, this regimen resulted in excellent disease control and survival in the ITT population, far exceeding the efficacy of our previous transplant study. The primary endpoint was met, warranting further investigation.

2048 Background: In GBM, VEGF inhibition decreases vascular permeability and MRI contrast enhancement (CE), which may or may not reflect meaningful anti-tumor activity. We sought to evaluate the relationships between different imaging modalities and anti-tumor activity in a prospective phase II trial of BEV, TMZ and HFSRT in GBM. Methods: Newly diagnosed GBM pts (N=40) received HFSRT with concomitant/ adjuvant BEV and TMZ. All pts underwent dynamic susceptibility contrast MR perfusion (DSC) at baseline, after HFSRT and every 2m. An FDG-PET scan was done at cycle 6. McDonald and RANO response criteria were applied. Results: DSC showed early/progressive decreases in relative cerebral blood volume (rCBV): baseline mean rCBV: 2.77; 6 weeks: 1.65 (p=0.014); 2m: 1.07 (p

2028 Background: Pre-clinical data suggest BEV may sensitize tumor endothelia to RT and disrupt vascular niches harboring cancer stem cells. Due to radioprotective effects, BEV also allows for more...

2031 Background: Testicular GCT rarely causes ESCC. Most GCT metastases respond to CT but ESCC is usually treated with radiation therapy (RT). We investigated the use of CT alone to treat ESCC in t...