Clinical and Biological Correlates of Neurotoxicity Associated with CAR T-cell Therapy in Patients with B-cell Acute Lymphoblastic Leukemia.

CD19-specific chimeric antigen receptor (CAR) T-cell therapy is highly effective against relapsed or refractory acute lymphoblastic leukemia (ALL), but is hindered by neurotoxicity. In 53 adult patients with ALL, we found a significant association of severe neurotoxicity with high pretreatment disease burden, higher peak CAR T-cell expansion, and early and higher elevations of proinflammatory cytokines in blood. Patients with severe neurotoxicity had evidence of blood-cerebrospinal fluid (CSF) barrier disruption correlating with neurotoxicity grade without association with CSF white blood cell count or CAR T-cell quantity in CSF. Proinflammatory cytokines were enriched in CSF during severe neurotoxicity with disproportionately high levels of IL6, IL8, MCP1, and IP10, suggesting central nervous system-specific production. Seizures, seizure-like activity, myoclonus, and neuroimaging characteristics suggested excitatory neurotoxicity, and we found elevated levels of endogenous excitatory agonists in CSF during neurotoxicity.Significance: We detail the neurologic symptoms and blood, CSF, and neuroimaging correlates of neurotoxicity associated with CD19 CAR T cells and identify neurotoxicity risk factors. Our findings implicate cellular components other than T cells and suggest novel links between systemic inflammation and characteristic neurotoxicity symptoms. Cancer Discov; 8(8); 958-71. ©2018 AACR.This article is highlighted in the In This Issue feature, p. 899.

A case of acyclovir neurotoxicity presenting with atypical cerebrospinal fluid findings.

An 82-year-old man with a history of end-stage renal disease presented with progressively worsening confusion and somnolence for the past 4-5 days. The patient was diagnosed with herpes zoster by his primary care physician 5 days ago and was started on a course of valacyclovir 1 g three times a day (dose not adjusted for renal impairment).A lumbar puncture was performed and cerebrospinal fluid (CSF) studies revealed 37 white blood cells (WBCs)/hpf (100% monocytes), protein 64 mg/dL and glucose 52 mg/dL. He was started on ceftriaxone, ampicillin and acyclovir. MRI of the brain was done and was unremarkable. Acyclovir-induced encephalopathy was high on differential, but his CSF findings were concerning for viral encephalitis. Nonetheless, all antimicrobials were discontinued and he was scheduled for a 5-hour dialysis session. The very next day, he showed immense improvement and eventually recovered completely. CSF PCR tests for both herpes simplex virus and varicella zoster virus came back negative.

Novel clinical grading of delayed neurologic sequelae after carbon monoxide poisoning and factors associated with outcome.

INTRODUCTION: Delayed neurologic sequelae (DNS) after carbon monoxide (CO) poisoning manifest as a relapse of neurologic deficits. However, the long-term outcome of DNS has not been fully clarified. Myelin basic protein (MBP) levels in the cerebrospinal fluid (CSF) have been reported to be elevated in DNS. However, the precise timing and clinical value of the CSF examination have not been fully evaluated. We aimed to clarify the long-term outcome and the factors predicting the outcome of DNS and to evaluate the utility of CSF-MBP for predicting the development and severity of DNS. METHODS: This work was designed as a single-center, prospective, observational study. We graded DNS severity as Grade 1 (consistent independence), Grade 2 (temporary dependence), or Grade 3 (persistent dependence). We analyzed the percentage categorized in each grade and the parameters associated with outcome. RESULTS: Of 100 patients experiencing acute CO poisoning (median age: 46 years; 69% male), 20 (20%) developed DNS, including six Grade 1 (30%), ten Grade 2 (50%), and four Grade 3 (20%) cases. The Grade 3 patients [median: 77 years; interquartile range (IQR): 76-82] were older than the Grade 1 patients [42; 30-46] (P<0.01); the DNS onset of the Grade 1 patients [median interval after poisoning: 35 days; IQR: 32-56] occurred later than that of the Grade 3 patients [10; 9-13] P<0.001) and the Grade 2 patients [25; 23-27] (P<0.05). The CSF-MBP levels of the DNS patients were higher than those of the non-DNS patients (P<0.0001). The 1-month CSF-MBP levels of the Grade 3 patients were higher than those of the Grade 1 patients (P<0.05); the MBP index, defined as [(Age)×(1-month CSF-MBP)], was higher in the Grade 3 patients than in the Grade 1 patients (P<0.01). Severe DNS were associated with advanced age (>72.5 years), earlier onset (<18 days), higher 1-month CSF-MBP (>252 pg/ml), and higher MBP index (>20.9 year × ng/ml). CONCLUSIONS: Poor DNS outcomes were associated with advanced age and earlier onset. CSF-MBP can serve as a sensitive predictor of both the development and outcomes of DNS.

Severe neurotoxicity following intrathecal methotrexate with nitrous oxide sedation in a child with acute lymphoblastic leukemia.

Systemic and intrathecal methotrexate is widely used in treatment protocols for childhood acute lymphoblastic leukemia. Its side effects vary in characteristics, intensity and time of onset, and depend on the administration route. Interactions with several drugs are known. Side effects of nitrous oxide sedation, often used for moderately painful procedures, typically occur after long time use and include neurological symptoms. We present a child who experienced a severe and long-lasting neurotoxicity after the third intrathecal application of methotrexate with short sedation by nitrous oxide during induction therapy for acute lymphoblastic leukemia. Symptoms completely resolved after 12 months.

Study of blood and brain lithium pharmacokinetics in the rat according to three different modalities of poisoning.

Lithium-induced neurotoxicity may be life threatening. Three patterns have been described, including acute, acute-on-chronic, and chronic poisoning, with unexplained discrepancies in the relationship between clinical features and plasma lithium concentrations. Our objective was to investigate differences in plasma, erythrocyte, cerebrospinal fluid, and brain lithium pharmacokinetics using a multicompartmental approach in rat models mimicking the three human intoxication patterns. We developed acute (intraperitoneal administration of 185 mg/kg Li2CO3 in naive rats), acute-on-chronic (intraperitoneal administration of 185 mg/kg Li2CO3 in rats receiving 800 mg/l Li2CO3 in water during 28 days), and chronic poisoning models (intraperitoneal administration of 74 mg/kg Li2CO3 during 5 days in rats with 15 mg/kg K2Cr2O7-induced renal failure). Delayed absorption (4.03 vs 0.31 h), increased plasma elimination (0.65 vs 0.37 l/kg/h) and shorter half-life (1.75 vs 2.68 h) were observed in acute-on-chronically compared with acutely poisoned rats. Erythrocyte and cerebrospinal fluid kinetics paralleled plasma kinetics in both models. Brain lithium distribution was rapid (as early as 15 min), inhomogeneous and with delayed elimination (over 78 h). However, brain lithium accumulation was more marked in acute-on-chronically than acutely poisoned rats [area-under-the-curve of brain concentrations (379 ± 41 vs 295 ± 26, P < .05) and brain-to-plasma ratio (45 ± 10 vs 8 ± 2, P < .0001) at 54 h]. Moreover, brain lithium distribution was increased in chronically compared with acute-on-chronically poisoned rats (brain-to-plasma ratio: 9 ± 1 vs 3 ± 0, P < .01). In conclusion, prolonged rat exposure results in brain lithium accumulation, which is more marked in the presence of renal failure. Our data suggest that differences in plasma and brain kinetics may at least partially explain the observed variability between human intoxication patterns.

Mass spectrometric analysis of rat cerebrospinal fluid proteins following exposure to the neurotoxicant carbonyl sulfide.

RATIONALE: Using a proteomic-based approach we have investigated possible altered expression of a range of cerebral spinal fluid (CSF) proteins following exposure to the neurotoxicant carbonyl sulfide (COS). CSF is ideal for the investigation of markers of brain injury or disease since it is secreted from several central nervous system structures and changes in the CSF composition may reflect brain insult and many pathological processes. METHODS: Animals were placed in exposure chambers and were exposed to 0 ppm or 500 ppm COS for 1, 2 or 3 days, 6 h per day. After the last inhalation exposure, 50-70 µL CSF sample was obtained by lumbar puncture. CSF samples were analyzed by electrospray ionization mass spectrometry (ESI-MS) on either a Premier quadrupole time-of-flight (QTOF) or an Agilent 6340 ion trap and by matrix-assisted laser desorption/ionization (MALDI)-MS on a 4800 MALDI-TOF/TOF analyzer. RESULTS: The dynamic range of abundance of the identified proteins spanned over more than three orders of magnitude. The four most abundant proteins identified (albumin, cystatin C, serotransferrin, transthyretin) are major proteins that are present in both CSF and blood at high levels but the fifth most abundant protein identified (prostaglandin H2D isomerase) is the second most abundant protein in human CSF and is secreted and synthesized in the rat central nervous system. No significant differences were observed between COS-treated CSF samples and the control CSF samples because of blood contamination. CONCLUSIONS: Quantitative MS protein analyses of rat CSF is limited by the low sample volumes that can practicably be obtained from rats and the low protein concentrations in rat CSF. Results of this work suggest a clear need for CSF collection that would minimize blood contamination. Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.

Initial biological qualification of SBDP-145 as a biomarker of compound-induced neurodegeneration in the rat.

Detection of compound-related neurodegeneration is currently limited to brain histopathology in veterinary species and functional measurements such as electroencephalography and observation of clinical signs in patients. The objective of these studies was to investigate whether concentrations of spectrin breakdown product 145 (SBDP-145) in cerebrospinal fluid (CSF) correlate with the severity of neurodegeneration in rats administered neurotoxic agents, as part of a longer term objective of developing in vivo biomarkers of neurotoxicity for use in non-clinical and clinical safety studies. Non-erythroid alpha-II spectrin is a cytoskeletal protein cleaved by the protease calpain when this enzyme is activated by dysregulation of calcium in injured cells. Calcium dysregulation is also associated with some toxicological responses in animals, and may be sufficient to activate neuronal calpain and produce SBDPs that can be released into CSF. Neurotoxicants (kainic acid, 2-chloropropionic acid, bromethalin, and pentylenetetrazole) known to affect different portions of the brain were administered to rats in dose-response and time-course studies in which neurodegeneration was measured by histopathology and SBDP-145 concentrations in CSF were measured by ELISA. We consistently observed >3-fold increases in SBDP-145 concentration in rats with minimal to slight neurodegenerative lesions, and 20 to 150-fold increases in animals with more severe lesions. In contrast, compounds that caused non-degenerative changes in central nervous system (CNS) did not increase SBDP-145 in CSF. These data support expanded use of SBDP-145 as a biomarker for monitoring compound-induced neurodegeneration in pre-clinical studies, and support the investigation of clinical applications of this biomarker to promote safe dosing of patients with compounds that have potential to cause neurodegeneration.

Cerebrospinal fluid protein biomarker panel for assessment of neurotoxicity induced by kainic acid in rats.

Glutamate excitotoxicity plays a key role in the etiology of a variety of neurological, psychiatric, and neurodegenerative disorders. The goal of this study was to investigate spatiotemporal distribution in the brain and cerebrospinal fluid (CSF) concentrations of ubiquitin C-terminal hydrolase-1 (UCH-L1), glial fibrillary acidic protein (GFAP), αII-spectrin breakdown products (SBDP150, SBDP145, and SBDP120), and their relationship to neuropathology in an animal model of kainic acid (KA) excitotoxicity. Triple fluorescent labeling and Fluoro-Jade C staining revealed a reactive gliosis in brain and specific localization of degenerating neurons in hippocampus and entorhinal cortex of KA-treated rats. Immunohistochemistry showed upregulation of GFAP expression in hippocampus and cortex beginning 24h post KA injection and peaking at 48h. At these time points concurrent with extensive neurodegeneration all SBDPs were observed throughout the brain. At 24h post KA injection, a loss of structural integrity was observed in cellular distribution of UCH-L1 that correlated with an increase in immunopositive material in the extracellular matrix. CSF levels of UCH-L1, GFAP, and SBDPs were significantly increased in KA-treated animals compared with controls. The temporal increase in CSF biomarkers correlated with brain tissue distribution and neurodegeneration. This study provided evidence supporting the use of CSF levels of glial and neuronal protein biomarkers to assess neurotoxic damage in preclinical animal models that could prove potentially translational to the clinic. The molecular nature of these biomarkers can provide critical information on the underlying mechanisms of neurotoxicity that might facilitate the development of novel drugs and allow physicians to monitor drug safety.

39-week toxicity and toxicokinetic study of ponezumab (PF-04360365) in cynomolgus monkeys with 12-week recovery period.

Ponezumab (PF-04360365) is a novel humanized IgG2Δa monoclonal antibody that binds to amyloid-ß (Aß). It is designed to have reduced immune effector function compared to other passive immunotherapies for Alzheimer's disease (AD). Toxicity was evaluated in cynomolgus monkeys treated intravenously with vehicle or 10, 30, or 100 mg/kg of ponezumab every 10th day for up to 39 weeks, and after a 12-week recovery phase. The Aß peptide sequence of monkeys is identical to that of humans. No substantial difference in test article exposure between sexes was observed, and mean plasma Cmax and AUC0-n were approximately dose-proportional. Ponezumab was detectable approximately 9 weeks after cessation of dosing. All animals, except two males given 10 mg/kg, maintained exposure to test article. One of these males tested positive for anti-ponezumab antibodies. Ponezumab was detected in the cerebrospinal fluid (CSF) of animals given active treatment. The estimated CSF/plasma ponezumab concentration ratio was <0.008 after multiple doses. At the end of the dosing and recovery phases, plasma Aß1-40 and Aß1-x were increased in treated animals versus controls. No test article-related effects were seen after ophthalmogical, cardiovascular, physical examinations, and clinical and anatomic pathology evaluations. Plasma concentrations of ponezumab on day 261 at the no observed adverse effect level of 100 mg/kg were 22.4 and 5.3 times greater on a Cmax and AUC basis, respectively, than human exposures at the highest dose (10 mg/kg) in a single-dose Phase I trial. These data suggest an acceptable safety profile for ponezumab as an immunotherapy for AD.

[Clinical significance of 5-HT and DA levels in serum and cerebrospinal fluid of the patients with delayed encephalopathy after acute carbon monoxide poisoning].

OBJECTIVE: To explore the changes and the clinical significance of 5-hydroxytryptamine (5-HT), dopamine (DA) levels in serum and cerebrospinal fluid (CSF) of patients with delayed encephalopathy (DEACMP) after acute carbon monoxide poisoning. METHODS: The dynamic detection of 5-HT and DA levels in serum and CSF from 42 patients with DEACMP was performed with high performance liquid chromatography (HPLC). The condition changes of patients with DEACMP were analyzed with three types of scales: the activity of daily living scale (ADL), information memory concentration test (IMCT) and Hasegawa's dementia scale (HDS); these changes were compared with those from 38 other encephalopathy patients and 38 non-encephalopathy patients, respectively. RESULTS: Before treatment, the serum 5-HT and DA levels [(662.61 ± 178.50) and (155.74 ± 60.32) nmol/L, respectively] of DEACMP group were both significantly lower than those [(914.08 ± 198.04) and (225.70 ± 48.53) nmol/L] of non-encephalopathy group (P < 0.05); the serum DA level of DEACMP group was also significantly lower than that [(243.57 ± 66.94) nmol/L] of other encephalopathy group (P < 0.05); the serum 5-HT level of DEACMP group was not significantly different from that [(729.54 ± 299.87) nmol/L] of other encephalopathy group (P > 0.05). After treatment, the serum 5-HT and DA levels [(714.08 ± 170.47) and (192.18 ± 33.07 nmol/L, respectively)] of DEACMP group elevated to various extent, but only serum DA level was significantly higher than that before treatment (P < 0.05). Before treatment, the CSF 5-HT and DA levels of DEACMP group were significantly lower than those of non-encephalopathy group and those of other encephalopathy group (P < 0.05). After treatment, the CSF 5-HT level (232.44 ± 54.28 nmol/L) was similar to normal level and significantly higher than that before treatment (P < 0.05); the CSF DA level [(56.83 ± 12.85) nmol/L] of DEACMP group increased only slightly (P > 0.05). In DEACMP group, ADL score (50.64 ± 7.23), HDS score (8.55 ± 8.08) and IMCT score (4.95 ± 7.30) before treatment were significantly different from those (8.5 ± 8.08, 4.95 ± 7.30 and 15.64 ± 10.90) after treatment (P < 0.01). In DEACMP group, there wasa negative correlation between DA level changes and HDS score changes, when the DA levels and HDS scores before treatment were compared with those after treatment (P < 0.05). CONCLUSION: The dynamic changes of 5-HT and DA levels in serum and CSF of patients with DEACMP consisted basically with the patient's condition change. The dynamically detected 5-HT and DA levels can be used as the biological indicators to reflect the condition change and treatment effects of DEACMP patients.

5-FU-induced neurotoxicity in cancer patients with profound DPD deficiency syndrome: a report of two cases.

PURPOSE: 5-Fluorouracil (5-FU) is a mainstay for treating various solid tumours in adults, including digestive and head and neck cancers. 5-FU-related toxicities usually include haematological, digestive and cutaneous features. Additionally, 5-FU has been described as being potentially neurotoxic in patients, but these side effects are quite rare in clinical practice. Here, we report two cases of sudden and unpredictable drug-induced neurotoxicities that occurred in patients undergoing their first course of 5-FU-based chemotherapy. PATIENTS AND METHODS: None of these patients had any previous neurological disorder history, and both were treated following standard regimen (LV-5-FU2 and TPF for patient 1 and 2, respectively). Neurotoxicity included drowsiness, acute confusion plus dysarthria for the first patient and seizure, confusion and signs of metabolic encephalopathy for the second one. In addition, typical 5-FU-related severe toxicities (e.g. neutropenia and mucosities) were observed. Both patients slowly recovered from these neurological toxicities under supportive treatment. It was assumed that overexposure to 5-FU could explain the severe toxicities encountered. To test this hypothesis, we retrospectively evaluated the dihydropyrimidine dehydrogenase (DPD) activity of these patients on a phenotypic basis. RESULTS: Evaluation of the uracil-to-di-hydrouracil (U/UH2) ratio in plasma revealed a profound DPD deficiency syndrome in both patients. CONCLUSION: These cases suggest that 5-FU standard dosage administration may lead to strong overexposure, responsible for the severe toxicities observed, including the neurological features. It implies that DPD deficiency can cause neurotoxicity in 5-FU-treated patients and advocates for the prospective screening of DPD deficiency before starting any 5-FU-containing chemotherapy so as to prevent such side effects in the future.

Safety and pharmacokinetics of intrathecal administration of pemetrexed in rats.

PURPOSE: Leptomeningeal carcinomatosis is a devastating complication of malignant disease. In this study, we evaluated the safety and pharmacokinetics of intrathecally administered pemetrexed in rats. METHODS: Three levels of pemetrexed (0.3, 1, and 3 mg/kg) were administered to 15 rats per level (45 rats in total) twice a week for 2 weeks through specifically designed indwelling subarachnoid catheters. Presence of clinical and pathological neurotoxicity was evaluated. To evaluate the pharmacokinetics of pemetrexed, independent cohorts of 30 rats were treated with 1 mg/kg of pemetrexed and its concentration in cerebrospinal fluid (CSF) and blood was measured using UPLC/MS/MS. RESULTS: There were no cases of clinical or pathologic neurotoxicity after intrathecal administrations of pemetrexed at levels of 0.3 and 1 mg/kg; however, 5 of 15 (33%) rats died after administration of 3 mg/kg pemetrexed. The distribution/elimination of pemetrexed in CSF was best described by a two-compartment model, with initial and terminal half-lives of 0.43 and 1.43 h, respectively. The predicted maximal concentration in CSF was 588 µM, and high levels of pemetrexed appeared to be maintained for a long time. Area under the curve and volume of distribution at steady state were 560 µM h and 1.14 ml, respectively. CONCLUSIONS: The no observed adverse effect level of intrathecal administration of pemetrexed was 1 mg/kg in rats. At this level, therapeutically high and durable pemetrexed concentrations could be achieved. Based on these results, further research on intrathecal pemetrexed in humans or non-human primates should be considered.

Fatal cerebral edema associated with serine deficiency in CSF.

Two young girls without a notable medical history except for asthma presented with an acute toxic encephalopathy with very low serine concentrations both in plasma and cerebrospinal fluid (CSF) comparable to patients with 3-phosphoglycerate dehydrogenase (3-PGDH) deficiency. Clinical symptoms and enzyme measurement (in one patient) excluded 3-PGDH deficiency. Deficiencies in other serine biosynthesis enzymes were highly unlikely on clinical grounds. On basis of the fasting state, ketone bodies and lactate in plasma, urine and CSF, we speculate that reduced serine levels were due to its use as gluconeogenic substrate, conversion to pyruvate by brain serine racemase or decreased L-serine production because of a lack of glucose. These are the first strikingly similar cases of patients with a clear secondary serine deficiency associated with a toxic encephalopathy.

The early elevation of interleukin 6 concentration in cerebrospinal fluid and delayed encephalopathy of carbon monoxide poisoning.

OBJECTIVE: This study was designed to investigate whether interleukin 6 (IL-6) in cerebrospinal fluid (CSF) in the early phase of carbon monoxide (CO) poisoning can be a predictive marker of delayed encephalopathy (DE). METHODS: Nine patients with CO poisoning were included in the study. Cerebrospinal fluid was sampled within 24 hours of the last exposure to CO, on hospital day 4, and once a week for at least 1 month to determine IL-6 and myelin basic protein concentrations. All patients were followed at least 3 months. RESULTS: Three patients demonstrated significant early IL-6 elevation in CSF, normal IL-6 level in CSF on day 4, and significant delayed myelin basic protein elevation in CSF. The 2 patients with the highest early IL-6 elevation in CSF developed DE. Interleukin 6 in serum was not related to DE. CONCLUSION: Interleukin 6 in CSF at the early phase of CO poisoning may be a predictive marker of DE.

Encephalopathy secondary to capecitabine chemotherapy: a case report and discussion.

INTRODUCTION: Capecitabine is an oral prodrug of 5-fluorouracil (5-FU) chemotherapy and is licensed for the treatment of breast and gastrointestinal cancers. Multifocal inflammatory leukoencephalopathy has been associated with intravenous 5-FU, but only a few cases of capecitabine-induced encephalopathy have been reported. SETTING: We describe here a case of encephalopathy following administration of Epirubicin/ Cisplatin/Capecitabine chemotherapy, review those cases previously described and suggest recommendations for management. RESULTS: Symptoms of neurotoxicity from 5-FU and capecitabine usually include confusion, ataxia, nystagmus, dysarthria, sensory loss, and memory loss. Withdrawal of the drug generally leads to improvement of symptoms and steroids are of only questionable benefit. CONCLUSION: Patients on fluouropyrimidine drugs with new neurological symptoms should be investigated with brain MRI scan and the drug should be withdrawn until symptoms resolve. Dihydropyrimidine dehydrogenase deficiency should be excluded and ideally an alternative chemotherapy regime sought. We would encourage reporting of such incidences to gain a clearer picture of the incidence and optimal management.

Myelin basic protein in cerebrospinal fluid: a predictive marker of delayed encephalopathy from carbon monoxide poisoning.

This study was designed to investigate whether myelin basic protein (MBP) in cerebrospinal fluid (CSF) can be a predictive marker of delayed encephalopathy from carbon monoxide (CO) poisoning. Five patients with CO poisoning were included in the study. The CSF was serially sampled to determine the MBP concentration. All patients were classified into group DE or group non-DE according to whether delayed encephalopathy developed or not. In all 3 patients in group DE, the MBP levels in the CSF were markedly elevated preceding the clinical manifestations of delayed encephalopathy. In both group non-DE patients, the MBP concentrations in the CSF were never elevated. Elevated MBP concentrations in the CSF may represent a predictive marker of delayed encephalopathy from CO poisoning, leading to a more appropriate triage of patients with CO poisoning.

High signal in the cerebrospinal fluid following prior gadolinium administration in a patient with renal impairment.

Increased signal intensity in the cerebrospinal fluid (CSF) on magnetic resonance imaging due to the presence of gadolinium is rarely observed, but has been seen in patients with brain or spinal pathology or underlying renal impairment. We report this phenomenon in a 66-year-old woman with diabetic nephropathy and discuss the possible pathogenesis of the scan findings. Recognition of this unusual finding, and features distinguishing it from other causes of high CSF signal intensity, such as subarachnoid haemorrhage and protein in the CSF, are emphasised to help prevent diagnostic errors.