The study combined four statistical signal-detection methods—reporting odds ratio, proportional reporting ratio, the multi-item gamma Poisson shrinker and the Bayesian information component. A preferred term was counted as a signal only if it met the criteria for all four algorithms. All 42 central nervous system preferred terms identified in the analysis met those criteria. Their reporting odds ratios ranged from 3.19 for lethargy to 101.19 for encephalopathy, while neurotoxicity, reported 117 times, had a reporting odds ratio of 67.42. Toxic encephalopathy, with 33 reports, had a similar value of 66.17. Commoner symptoms such as confusional state, disorientation, agitation, mental-status change, coma, delirium and status epilepticus also generated signals. These results indicate that ifosfamide was reported more often in connection with these terms than expected from the database’s broader reporting pattern, a finding that requires clinical and epidemiological validation.

The analysis also flagged six events that the study said were not listed in the ifosfamide prescribing information: brain oedema, hydrocephalus, spinal cord compression, cerebral ischemia, gaze palsy and Guillain–Barré syndrome. Their report counts ranged from three to six, and all met the study’s four-way signal threshold. These findings do not demonstrate that ifosfamide caused the conditions, which may instead reflect underlying disease, other medicines, metabolic disturbances or combinations of factors. Some events can also resemble tumor progression or complications of other chemotherapy agents, making attribution difficult without complete clinical information. The signals broaden the reported neurological spectrum associated with ifosfamide and identify areas requiring prospective pharmacoepidemiological investigation, rather than establishing new label indications or causal effects.

Timing data were available for 233 of the 786 central nervous system reports. Among these cases, the median time to onset was two days after treatment initiation, with an interquartile range of one to eight days and an observed range of one to 200 days. A total of 204 cases, or 87.55% of those with usable timing data, reported onset within 30 days. Smaller numbers were reported from days 31 to 60, days 61 to 90 and between 91 days and one year. A Weibull model estimated a shape parameter of 0.62, with a 95% confidence interval of 0.57 to 0.68, indicating an early-failure pattern in which documented events tended to occur earlier rather than later. Because the analysis covered only 29.64% of all included reports, the timing distribution should be interpreted as a feature of the available subset rather than the complete reporting population.

Central nervous system reports differed from the broader group of adverse drug events associated with ifosfamide. Sex was available for 669 of the 786 central nervous system reports: 372 concerned females and 297 concerned males. These represented 47.33% and 37.79% of all 786 reports, respectively, or approximately 55.61% and 44.39% among records with known sex. Age was available for 600 reports. Adults aged 18 to 64 accounted for 354 reports, or 45.04% of the complete central nervous system group and approximately 59.00% of records with known age; 131 reports involved people younger than 18 and 115 involved those older than 64. Of all central nervous system reports, 93 were associated with death, 62 with a life-threatening outcome and 269 with hospitalization.

Brain oedema had the highest reported fatal-outcome proportion, at 83.33%, although this estimate came from only six cases. Neurological decompensation was followed by coma, with reported fatal-outcome proportions of 57.14% from seven cases and 56.00% from 25 cases, respectively. Spinal cord compression had a reported fatal-outcome proportion of 50.00% from four cases. By contrast, fatal outcomes were reported in 5.83% of 120 confusional-state cases, 8.77% of 57 disorientation cases, 9.40% of 117 neurotoxicity cases and 11.11% of 252 encephalopathy cases. These high percentages describe outcomes among reported cases and are not mortality risks for the wider treated population. The denominators also differed sharply, and spontaneous reports cannot capture the number of patients who received ifosfamide without an event or the full clinical course of those who did.

Subgroup analyses suggested differences in reporting patterns by sex and age. The study detected 27 central nervous system preferred terms among females and 23 among males. Encephalopathy was reported 120 times in females, with a reporting odds ratio of 135.64 and a 95% confidence interval of 112.87 to 163.02, compared with 65 reports and a reporting odds ratio of 47.08 in males. Reporting odds ratios for confusional state, neurotoxicity and disorientation were also higher in the female cohort. Among adults aged 18 to 64, encephalopathy had the strongest signal, with a reporting odds ratio of 78.69. In patients older than 64, confusional state was most frequently reported, while psychomotor impairment, clumsiness and encephalopathy had strong disproportionality values. Reported events among minors included tonic-clonic movements, petit mal epilepsy and behavioral disorders. The observational and spontaneous-reporting design prevents these comparisons from establishing biological differences in susceptibility.

The researchers separately reviewed case reports and case series retrieved from PubMed and SpringerLink through September 2025. Two researchers screened the literature, resolving disagreements with a third, and assessed each included publication with an eight-item Joanna Briggs Institute checklist. Of 42 eligible publications, 38, or 90.5%, met all eight criteria, and the mean number of affirmative ratings was 7.9. The review contained 61 patients aged 3 to 77 years treated for sarcoma, lymphoma, ovarian cancer, breast cancer and other solid tumors. Forty-six patients, or 75.41%, developed central neurotoxicity after one to three cycles of ifosfamide-containing chemotherapy. Impaired consciousness, hallucinations, seizures, delirium, disorientation and psychiatric or behavioral abnormalities were among the manifestations described. The literature was synthesized descriptively because the evidence consisted predominantly of case reports and small case series rather than comparative cohorts.

Outcome and treatment details in the literature review showed that 29 of 46 cases with usable recovery information recovered within 48 hours and 38 recovered within 72 hours, with reported recovery times ranging from minutes to three weeks. Overall, 54 of 61 patients, or 88.52%, were reported to have recovered, while seven had a fatal outcome. Methylene blue was the most frequently reported treatment, appearing in 47.54% of cases, followed by benzodiazepines in 26.23%, antipsychotics in 18.03% and thiamine in 13.11%; several patients received more than one treatment. These figures synthesize heterogeneous reports, sometimes involving combination therapy and varying diagnostic workups, and therefore do not isolate the effectiveness of any individual treatment. They also cannot provide a comparative estimate of recovery because the review did not include a control group or standardized treatment assignment.

The results support a picture of ifosfamide-associated central neurotoxicity that is clinically heterogeneous, frequently reported soon after treatment and not confined to encephalopathy. A sensitivity analysis that removed adverse events related to cisplatin, carboplatin, epirubicin, etoposide and mesna reduced the number of events but left signals for encephalopathy, confusional state, neurotoxicity, disorientation, agitation, mental-status changes, toxic encephalopathy and coma stable. This sensitivity check addressed reporting involving those named concomitant medicines, but it did not exclude every report containing concomitant drugs or fully remove confounding from other treatments, disease-related neurological complications and metabolic disturbances. FAERS is also vulnerable to missing information, duplicate or incomplete reports, selective submission and uncertainty about renal function, albumin, electrolytes, disease progression, prior cranial radiotherapy and other drugs. The case literature is susceptible to publication bias and incomplete diagnostic workups. Prospective studies are therefore needed to determine incidence, causal relationships, risk factors and effective prevention strategies.

Subject of Research: Pharmacovigilance analysis of ifosfamide-associated central nervous system adverse events

Article Title: Exploring ifosfamide-related central nervous system adverse events based on the FAERS database and clinical feature analysis of cases

Article References: Shi, P., Mei, Q., zheng, R., & Wang, S. (2026). Exploring ifosfamide-related central nervous system adverse events based on the FAERS database and clinical feature analysis of cases. Frontiers in Oncology, 16, Article 1961478. https://doi.org/10.3389/fonc.2026.1961478

Image Credits: AI Generated

DOI: 10.3389/fonc.2026.1961478

Keywords: ifosfamide, neurotoxicity, encephalopathy, FAERS, pharmacovigilance, adverse drug events, disproportionality analysis, case reports, chemotherapy, time to onset, Exploring, ifosfamide-related

The research team, led by Aasha I. Hoogland of the Moffitt Cancer Center, pooled data from two longitudinal studies of patients treated with axicabtagene ciloleucel, restricting their analysis to people diagnosed with large B-cell lymphoma. In total, 155 participants completed neuropsychological assessments before conditioning therapy began and again at 30, 90, and 360 days after infusion. Sixty-nine of them, or 45 percent, were aged 65 or older. The sample was predominantly male and White, and participants ranged in age from 19 to 82, with a mean age of 61. Each assessment covered a battery of validated tests measuring attention, executive function, verbal ability, immediate verbal memory, delayed memory, and visuospatial abilities, which were averaged into domain scores and an overall total neuropsychological performance score.

The technical design of the study matters for interpreting its findings. Raw test scores were standardized as T-scores using age-adjusted population norms, so performance was judged against what would be expected for a person’s age. Impairment on any given test was defined as scoring two or more standard deviations below the mean, and overall cognitive impairment was flagged when a participant scored two standard deviations below the mean on one test or 1.5 standard deviations below on two tests. To track change over time, the researchers used random-effects mixed models with linear and quadratic terms for time, an approach that estimates within-person trajectories and tolerates the unequal assessment intervals and missing data that are inevitable in a population of seriously ill cancer patients.

The starting point was sobering but not surprising. Before infusion, 48 percent of older adults showed overall neurocognitive impairment compared with 34 percent of younger adults, a statistically significant difference. Older participants also performed worse than younger participants on every neurocognitive domain at baseline. Yet by one year after infusion, that gap had dissolved entirely. There were no statistically significant differences between the age groups in immediate verbal memory or delayed memory at the 360-day mark, and the overall rates of impairment no longer differed between the groups after treatment.

The most striking result emerged in the memory domains. Immediate verbal memory and delayed memory improved significantly over time in older adults, with p-values below 0.01, but remained stable in younger adults. The improvement in immediate verbal memory among older patients was large enough to be clinically meaningful, exceeding the five-point threshold the researchers set to reflect a medium effect size. This pattern is the reverse of what the team predicted. In neuropsychology, a lack of expected improvement between repeated test administrations, often called an absence of practice effects, is itself considered a signal of cognitive abnormality. The fact that younger patients showed no such improvement, while older patients did, hints at a genuine post-treatment cognitive vulnerability in the younger group that could not be explained by disease status, cytokine release syndrome, neurotoxicity, or the number of prior therapies.

The raw impairment counts reinforce that interpretation. Among younger adults with normal cognitive scores before infusion, 30 percent developed newly onset impairment by day 30, 41 percent by day 90, and fully half by day 360. Among older adults who started with normal cognition, the corresponding figures were 15, 32, and 35 percent. Persistent impairment at one year was evaluable in a subset of patients, and 61 percent of younger adults with new impairment who completed the final assessment still met impairment criteria, compared with 50 percent of older adults. These numbers suggest that the burden of new cognitive deficits after CAR T-cell therapy falls disproportionately, if unexpectedly, on the young.

The clinical context helps explain why this question is urgent. Subgroup analyses of the pivotal ZUMA-1 trial showed that patients aged 65 and older achieved high response rates to axicabtagene ciloleucel, with 92 percent clinical responses and 75 percent complete responses, and the ZUMA-7 trial found that older patients given the therapy lived significantly longer, 21.5 versus 2.5 months, with better objective response rates than standard care at the time. Real-world data echo those results. But the same trials revealed that older patients were more likely to experience severe neurologic events, grade 3 or higher, than younger patients. Severe acute neurotoxicity, however, is a different phenomenon from the subtle, measurable cognitive changes tracked by neuropsychological testing, and until now little was known about how age shapes those longer-term trajectories.

The exploratory analyses of clinical risk factors added further nuance. Older adults who had received four or more prior lines of therapy showed the greatest improvements in overall neuropsychological performance and verbal ability, a finding that challenges the so-called multiple hit hypothesis, which holds that each successive cancer treatment accumulates cognitive damage. By 12 months post-infusion, these older, heavily pretreated patients demonstrated clinically meaningful gains relative to younger patients with the same treatment history. Conversely, older adults with active disease at three months, meaning they had not responded to the CAR T-cell infusion, showed rapid deterioration in visuospatial ability and performed significantly worse than younger patients with active disease at every post-infusion timepoint. The researchers caution, however, that only four older adults in each of these subgroups provided neurocognitive data at 12 months, so these subgroup findings have limited statistical power and require replication.

The authors situate their results within a broader literature on cognition after cellular therapy. A meta-analysis of cognitive function after hematopoietic cell transplantation, another cellular therapy for blood cancers, similarly found a lack of expected practice effects over time in a population with a median age of 43. Prior work from the same Moffitt group had shown that older age was associated with worsening verbal memory and fluency after transplantation, making the current reversal in CAR T-cell recipients all the more intriguing. The researchers also note that among people without cancer, verbal working memory peaks in young adulthood, around age 30, which may create a window of vulnerability for memory-related cognitive domains in younger patients. Whether the younger patients’ flat memory trajectories reflect biological sensitivity, unmeasured treatment effects, or some other mechanism remains an open question.

The study has clear limitations. The sample was composed largely of male, White, non-Hispanic, married college graduates, which constrains generalizability, and more than a third of patients with baseline data did not provide data at one year, largely because of morbidity or mortality. The sample size also precluded comparisons between patients treated commercially and those treated on clinical trials, an important distinction since outcomes can differ between those settings. Still, the longitudinal design, validated measures, and statistical methods that accommodate missing data give the findings real weight. For patients and clinicians weighing CAR T-cell therapy, the message is cautiously reassuring for older adults: despite their higher risk of acute neurotoxicity, their memory trajectories after axicabtagene ciloleucel were not worse than those of younger patients, and in some respects were better. The authors call for larger studies of cognition in older CAR T-cell recipients, and for closer attention to the cognitive aftereffects experienced by younger patients, whose brains may bear more of the treatment’s hidden cost than anyone anticipated.

Subject of Research: Age-related differences in neurocognitive outcomes after axicabtagene ciloleucel CAR T-cell therapy for large B-cell lymphoma

Article Title: Age as a moderator of neurocognitive performance in patients with large B-cell lymphoma treated with axicabtagene ciloleucel

Article References: Hoogland, A. I., Li, X., Hu, B., Irizarry-Arroyo, N., Welniak, T., Rodriguez, Y., Jain, M. D., Oswald, L. B., Small, B. J., Snider, J. T., Wade, S. W., Chavez, J. C., Khimani, F., Lazaryan, A., Liu, H. D., Nishihori, T., Pinilla-Ibarz, J., Shah, B. D., Pidala, J. A., … Jim, H. S. L. (2026). Age as a moderator of neurocognitive performance in patients with large B-cell lymphoma treated with axicabtagene ciloleucel. Supportive Care in Cancer, 34(10), Article 943. https://doi.org/10.1007/s00520-026-11169-w

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11169-w

Keywords: CAR T-cell therapy, axicabtagene ciloleucel, large B-cell lymphoma, neurocognitive performance, memory, cognitive impairment, aging, neurotoxicity, cytokine release syndrome, immunotherapy, longitudinal study, supportive care

The drug itself is a molecular hybrid engineered for precision. A monoclonal antibody directed against CD30, a protein abundantly displayed on malignant cells in classical Hodgkin lymphoma, systemic anaplastic large cell lymphoma, peripheral T-cell lymphoma and cutaneous T-cell lymphoma, is chemically linked to monomethyl auristatin E, an agent that disrupts microtubules and halts cell division. Selective delivery limits collateral damage to healthy tissue, yet the microtubule poison still finds its way into peripheral nerves, where it accumulates with cumulative dose and treatment duration. Extended single-agent courses, such as those used for cutaneous T-cell lymphoma, produce more severe and longer-lasting neuropathy than shorter combination regimens.

Clinically, the damage follows a characteristic pattern. Patients typically develop tingling, numbness and burning sensations in a distal, symmetrical distribution affecting the hands and feet. Less frequently, motor involvement produces cramps and distal weakness, while autonomic features such as orthostatic hypotension or gastrointestinal dysmotility can appear. Although the neuropathy is often reversible, complete resolution may take months to years, and some patients experience persistent symptoms that erode quality of life. The economic toll is substantial too, with increased hospitalisation rates, emergency department visits and outpatient encounters documented across healthcare systems.

To build the consensus, the researchers turned to a modified Delphi methodology, a structured process designed for areas where robust empirical evidence is scarce. A three-member steering committee, comprising a neurologist, a haematology oncologist and a dermatology oncologist from Spain, Australia and Brazil, reviewed the protocol and led statement development without participating in voting. Fifteen experts agreed to join the panel, recruited to represent neurology, haematology-oncology, dermatology-oncology, nursing and rehabilitation, and 13 completed all three survey rounds conducted between July and November 2025. Statements were drafted from a targeted literature review of publications from 2014 to 2024 and refined across rounds, with consensus defined as at least 70 percent agreement or disagreement on six-point Likert scales.

The panel achieved consensus on 60 of 87 statements, or 69 percent, spanning risk factor assessment, symptom monitoring, pharmaceutical management and non-pharmaceutical strategies. Among the strongest agreements, 92 percent of panellists endorsed routine risk factor profiling before initiating brentuximab vedotin, including assessment of personal and family history of hereditary neuropathies. High cumulative drug dose, prior or concurrent neurotoxic therapy such as vincristine or vinblastine, and comorbidities like diabetes were identified as key risk factors, each reaching agreement rates of 77 to 85 percent. Early warning signs demanding immediate intervention included distal weakness, loss of balance resulting in falls, and rapid progression of sensation loss.

On diagnosis, the panel unanimously agreed that healthcare professionals should ask patients about neuropathy symptoms at every visit, while remaining alert to underreporting, particularly among patients with poorer prognoses who may fear losing access to the drug. Validated assessment tools should be applied at baseline and at least every three treatment cycles, with the Total Neuropathy Score clinical version, or TNSc, recognised as an effective diagnostic and grading instrument by 92 percent of panellists. The Common Terminology Criteria for Adverse Events, or CTCAE, failed to reach consensus as a standalone tool, though panellists supported using it alongside the TNSc and patient-reported outcomes. Nerve conduction studies are not routine for typical presentations but should be considered in severe neuropathy, diagnostic uncertainty, atypical courses, or suspected Guillain-Barré Syndrome, where intravenous immunoglobulin or plasma exchange is supported in confirmed cases.

Dose modification recommendations revealed both clarity and persistent uncertainty. For Grade 1 neuropathy, no dose changes are required. For Grade 2 neuropathy on the BV plus CHP regimen, reduction from 1.8 to 1.2 milligrams per kilogram, up to a maximum of 120 milligrams every three weeks, was supported. In a notable departure from the prescribing label, the panel agreed that when brentuximab vedotin is combined with doxorubicin, vinblastine and dacarbazine, vinblastine dose reduction may be considered before adjusting the antibody-drug conjugate itself. For Grade 3 neuropathy on BV plus AVD in Hodgkin lymphoma, withholding the drug until improvement to Grade 2 or lower, then resuming at 0.9 milligrams per kilogram, reached consensus. Grade 4 neuropathy warrants permanent discontinuation. Dose reduction was judged more effective at mitigating neuropathy than prolonging the interval between doses, and patients with pre-existing Grade 3 or higher neuropathy should be excluded from receiving the drug.

Intermediate-grade neuropathy remains the murkiest territory. Grade 3 and above neuropathy affects roughly six percent of patients with classical Hodgkin lymphoma, meaning direct clinical experience is limited and robust data on symptom resolution rates are lacking. Patients at greatest risk, such as those with high cumulative neurotoxic exposure in the relapsed or refractory setting, typically have poorer prognoses and fewer alternatives. Free-text responses indicated that management decisions are context-dependent, with physicians more likely to modify dosing when Grade 2 neuropathy appears early in treatment rather than near the end. The newly approved BrECADD combination regimen was not explored, as it received approval after the Delphi protocol was finalised.

Beyond pharmacology, the panel endorsed a multidisciplinary support framework. Physical therapy was identified as an effective non-pharmaceutical option by 77 percent of panellists, and all agreed that physiotherapists and occupational therapists can improve physical quality of life through interventions such as sensory re-education and graded tactile tasks. Psychologists and counsellors play a key role in supporting emotional wellbeing, while social workers may assist when severe neuropathy leads to dependence in daily living. Analgesics including duloxetine, pregabalin and gabapentin may be considered for neuropathic pain, though the panel recognised variability in their effectiveness and lacked consensus on whether pain medications risk masking symptoms during monitoring. Patients receiving the drug should avoid excessive pyridoxine intake, given its potential to exacerbate nerve damage.

Shared decision-making emerged as a unifying principle across all recommendations. The panel unanimously agreed that patient choice is central to dose modification decisions and that clinicians should consider lifestyle and occupation when discussing impact, noting that musicians, surgeons and artists reliant on fine motor skills may experience disproportionate consequences even at lower neuropathy grades. The panel called for clearer evidence-based guidelines on diagnosis, pharmaceutical and non-pharmaceutical management, and additional research to identify predictive factors for severe or irreversible neuropathy. Pharmacists were not represented on the panel, and participants from the United States and Canada were excluded because separate consensus activities were underway in those regions, meaning the recommendations should not be interpreted as a global consensus. Nevertheless, the framework lays the groundwork for future guideline development, with the ultimate goal of preserving both lymphoma control and neurological function.

Subject of Research: Expert consensus on managing brentuximab vedotin-induced peripheral neuropathy in lymphoma

Article Title: Expert consensus on the management of peripheral neuropathy in patients with lymphoma on brentuximab vedotin

Article References: Velasco, R., Sanches, J. A., Argyriou, A. A., Cavaletti, G., Domingo-Domenech, E., Fosså, A., Gállego Pérez-Larraya, J., Ruiz, A. M., Molassiotis, A., Neder Ramires Abdo, A., Ortiz-Romero, P. L., Park, S. B., Scarisbrick, J., Urwin, L., Wang, M., & Prince, H. M. (2026). Expert consensus on the management of peripheral neuropathy in patients with lymphoma on brentuximab vedotin. Supportive Care in Cancer, 34(11), Article 1116. https://doi.org/10.1007/s00520-026-11314-5

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11314-5

Keywords: brentuximab vedotin, peripheral neuropathy, lymphoma, Delphi consensus, antibody-drug conjugate, dose modification, neurotoxicity, supportive care, Hodgkin lymphoma, patient-reported outcomes, multidisciplinary care, chemotherapy-induced neuropathy

The core problem these compounds pose is biochemical and brutally simple. Both OPs and CBs target acetylcholinesterase (AChE), the enzyme that clears the neurotransmitter acetylcholine from the synaptic cleft after a nerve fires. When the enzyme is blocked, acetylcholine floods receptor sites, overstimulating the nervous system. But the two families differ in a crucial way. Organophosphates phosphorylate the serine residue in the enzyme’s catalytic gorge and remain tightly bound, producing long-lasting inhibition. Carbamates, by contrast, carbamylate the same residue reversibly, allowing the enzyme to recover within hours. This difference in binding chemistry translates directly into differences in environmental persistence: OPs, which are larger molecules of roughly 300 daltons, leave persistent residues on crops, while CBs of about 200 daltons break down into small inorganic molecules such as carbon dioxide.

Persistence, however, is a slippery concept. Pesticides are classified as non-persistent when they degrade by half in under 30 days, moderately persistent between 30 and 100 days, and persistent beyond 100 days. Yet even nominally short-lived compounds can linger far longer than expected. Organophosphates with short half-lives can persist for up to a year in soils with low pH and temperature, where adsorption to soil particles shields them from microbial degradation. Carbamates show equally dramatic pH dependence: a compound with a seven-day half-life may persist for 560 days at pH 6.0. In seawater, microbial degradation of insecticides like diazinon and chlorpyrifos is inhibited, extending their residence in marine environments. Because both families are lipophilic, they cross cell membranes and biological barriers, including the blood-brain barrier and the placenta, and accumulate in body fat, plant oils, and even human milk.

The human health toll is substantial. Pesticide intoxication is one of the most common causes of poisoning worldwide, with approximately one million cases and several hundred thousand deaths reported annually. Acute exposure, whether from accidents or intentional ingestion, produces the classic cholinergic crisis: nicotinic symptoms such as paralysis and headache, and muscarinic effects including respiratory distress, cramps, and lacrimation, which can be fatal. A particularly treacherous feature of thiophosphate organophosphates like chlorpyrifos, diazinon, and parathion is that the liver converts them during Phase I metabolism, via cytochrome P450 oxidases, into oxon derivatives that are even more potent cholinesterase inhibitors than the parent compounds. Symptoms can therefore worsen after the initial exposure as the most toxic metabolites accumulate.

Chronic, low-dose exposure is harder to pin down but arguably more worrying, because residues persist in air and food and exposure is nearly universal. Epidemiological studies have linked organophosphate exposure to an increased incidence of Parkinsonism in areas of intensive pesticide use, impaired memory, reduced psychomotor speed, and mood disturbances in adults, and neurobehavioral deficits in newborns of exposed mothers. Elevated levels of DMAP, a metabolite of thionophosphates, have been detected in children and adolescents, supporting a possible relationship between OP exposure and attention deficit hyperactivity disorder. Large cohort studies report increased risks of hormonally related cancers among spouses of pesticide applicators and associations between parental exposure and childhood leukemia. Occupational poisoning rates have reached 18.2 cases per 100,000 agricultural workers in developed countries, and domestic exposures are frequently underdiagnosed because the symptoms are nonspecific.

Perhaps the most vulnerable stage of life is the earliest. Embryonic development is a multiphasic process driven by rapid ion fluctuations that are modulated by the cholinergic system, and molecules related to acetylcholine signaling are involved in nearly every developmental event, from sperm-egg interaction through early cleavages to tissue differentiation and morphogenesis. Neurotoxic pesticides can cross the placenta and interfere with these processes, and clinical reports describe severe congenital anomalies following occupational or accidental OP exposure during pregnancy. Some compounds, such as the organophosphate fenitrothion, even display antiandrogenic activity, impairing male reproductive development and reducing sperm production and fertilizing capability. Because inductive signals in early embryos determine future body structures, the earliest developmental stages are the most sensitive, and small anomalies introduced at these stages can be amplified as development proceeds.

This is where the review’s central proposal comes in. Under European Regulation 1107/2009 and Council Directive 86/609/EEC, animal testing should be minimized and vertebrate experiments treated as a last resort, in line with the 3R principle of Replacement, Reduction, and Refinement. The authors systematically surveyed the literature, retrieving 277 records on organophosphates and 125 on carbamates from the Web of Science database, to identify which non-mammalian and in vitro models can reliably stand in for vertebrate testing. The answer is a tiered panel spanning nearly every level of biological organization: stabilized cell lines such as NT2 neurons and C6 glioma cells, primary cultures of nervous cells, immune cells from invertebrates, three-dimensional neural spheroids, and whole organisms including protists, fruit flies, brine shrimp, and sea urchins.

The cellular data are striking. C6 glioma cells exposed to diazinon at concentrations as low as 200 nanograms per liter, levels comparable to those measured in the blood and cord samples of mothers and children living in agricultural areas, showed dose-dependent micronuclei induction and significant AChE inhibition, suggesting that doses non-toxic to intact organisms may still weaken brain defense mechanisms. The NTera2/D1 human neuronal line, which expresses a complete cholinergic toolkit, showed time-dependent enhancement of apoptosis at diazinon concentrations of 10^-4 to 10^-5 M, with a paradoxical viability boost at 10^-6 M that the authors attribute to a dynamic balance between activated and blocked acetylcholine receptors. Three-dimensional retinal spheroids from chick embryos exposed to diazinon at 20 to 120 micromolar showed reduced AChE activity, inhibited differentiation of inner plexiform layer areas, and enhanced apoptosis, demonstrating that 3D models can capture tissue-level architecture that flat cultures miss.

Whole-organism models add ecological realism. The fruit fly Drosophila melanogaster, itself a target species, is so sensitive that its AChE has been engineered into biosensors for rapid detection of paraoxon and malaoxon. Brine shrimp cysts rehydrated in water containing pesticides at concentrations from 10^-11 to 10^-5 M showed dose-dependent decreases in hatching speed and larval survival, with chlorpyrifos oxon and carbaryl causing the strongest effects. But the standout model is the sea urchin. Species such as Paracentrotus lividus develop synchronously in transparent envelopes, allowing researchers to watch anomalies unfold in real time, and they possess the full set of cholinergic molecules from gamete interaction onward. Their AChE serves as a biomarker of environmental contamination, and researchers have proposed integrated toxicity indices that combine the number of anomalous embryos with the severity of skeletal and developmental defects to assess the health of coastal waters. Notably, deficits in skeletal spicule mineralization after exposure to the organophosphate Basudin open possible lines of inquiry into osteoporosis, while long-term effects have been associated with cardiovascular disease.

The review’s conclusion is a call for a panel approach rather than a single-model fix. Because cholinesterase enzymes exist in multiple molecular forms whose responsiveness to pesticides varies among species, tissues, and even physiological states, no one model can capture the full spectrum of risk. Electric eel AChE, for instance, is far more resistant to chlorpyrifos than rat brain or fruit fly enzyme, with inhibitory concentrations differing by an order of magnitude. By combining models of graded complexity, from molecules to spheroids to embryos, and measuring a broad range of endpoints including AChE inhibition, apoptosis, micronuclei, synaptogenesis, and developmental anomalies, toxicologists can build a more protective and predictive risk assessment. In an era when pesticide residues are nearly ubiquitous in environmental compartments, the humble sea urchin embryo may prove to be one of the most powerful sentinels we have.

Subject of Research: Hazard identification of organophosphate and carbamate pesticides using bioethical alternative models in toxicology

Article Title: Hazard identification and the harmful effects of organophosphates and carbamates in bioethical models: a review

Article References: Gambardella, C., Soro, P., Paladino, O., Falugi, C., Misurale, F., & Faimali, M. (2026). Hazard identification and the harmful effects of organophosphates and carbamates in bioethical models: a review. Frontiers in Cell Death, 5, Article 1680195. https://doi.org/10.3389/fceld.2026.1680195

Image Credits: AI Generated

DOI: 10.3389/fceld.2026.1680195

Keywords: organophosphates, carbamates, acetylcholinesterase, pesticides, neurotoxicity, 3R principle, bioassays, sea urchin embryos, brine shrimp, neural spheroids, risk assessment, emerging contaminants

The review’s central argument is that toxicity prediction is no longer a single-technique discipline but a layered pipeline in which every stage matters. Traditional safety assessment relies on in vitro assays and animal studies, methods the authors describe as resource-intensive, slow, and limited by interspecies physiological differences that can mask human-specific toxicity. Machine learning and deep learning models, by contrast, can extract complex nonlinear patterns from high-dimensional chemical and biological datasets, enabling rapid screening of large compound libraries during early discovery. This shift builds on decades of quantitative structure-activity relationship (QSAR) modeling, which links chemical structure to biological activity, but modern frameworks go further by learning molecular features directly from data rather than relying on hand-crafted rules.

At the foundation of any predictive model lies the question of how to represent a molecule for computation. The review catalogs the dominant approaches: molecular fingerprints such as MACCS structural keys and extended connectivity fingerprints (ECFPs), which encode substructures as bit strings; numerical descriptors calculated by tools like RDKit, Mordred, PaDEL, and the FDA-developed Mold2; and graph-based representations that treat atoms as nodes and bonds as edges. Graph representations have become especially important because they preserve molecular connectivity and allow graph neural networks (GNNs) to learn directly from structure. More recently, pre-trained chemical language models such as ChemBERTa, SMILES-BERT, MolBERT, and MoLFormer generate learned embeddings that capture richer contextual information than fixed fingerprints ever could.

The authors critically compare conventional machine learning algorithms, including random forests, support vector machines, decision trees, k-nearest neighbors, naive Bayes, XGBoost, and LightGBM, against deep learning architectures spanning deep neural networks, convolutional neural networks, GNNs, and transformer-based models. Their conclusion is deliberately unsentimental: no single architecture consistently wins across toxicity tasks. Conventional methods remain competitive on small to medium datasets with informative descriptors, while deep learning approaches benefit from larger datasets and richer representations. Performance differences reported across studies often reflect variations in dataset composition, endpoint definition, class distribution, and validation strategy rather than genuine algorithmic superiority.

One of the review’s most technically significant sections addresses validation rigor, an issue that quietly undermines many published benchmarks. Random train/test splitting can produce optimistic results when structurally similar compounds or close analogues appear in both sets, a problem compounded by data leakage when preprocessing, feature selection, or hyperparameter optimization occurs before the final split. Scaffold-based, cluster-based, temporal, and independent external validation provide far more honest assessments of generalization to novel chemical space. The authors also stress that accuracy alone is insufficient for imbalanced datasets, where toxic compounds are typically rare; they recommend complementary metrics including sensitivity, specificity, F1-score, balanced accuracy, Matthews correlation coefficient, ROC-AUC, and precision-recall AUC, with calibration considered when predicted probabilities inform risk assessment.

The review then organizes its evidence around five organ-specific toxicity domains, drawing on 40 primary studies published between 2021 and 2026. In respiratory toxicity, random forest models using PubChem fingerprints achieved 93.3% accuracy, while a LightGBM model for chemical lung toxicity reached 86.88% accuracy with a ROC-AUC of 0.931. In cardiotoxicity, transformer models using Morgan fingerprints hit a ROC-AUC of 0.93 for hERG channel interactions, and a multi-view GNN framework called CardiacTox achieved a ROC-AUC of 0.938 for CaV1.2 prediction. Nephrotoxicity models reached a ROC-AUC of 0.93 in consensus approaches, while hepatotoxicity studies reported ResNet18 deep neural networks achieving a test ROC-AUC of 0.958 for drug-induced liver injury. Neurotoxicity studies spanned blood-brain barrier permeability, neuronal cytotoxicity, and acetylcholinesterase inhibition, with hybrid deep learning models reaching ROC-AUC values up to 0.964.

These headline numbers, however, come with caveats the authors refuse to bury. Dataset sizes varied enormously, from 2,670 chemicals in a neurotoxicity study to just 83 breast cancer patients in a radiotherapy cardiotoxicity analysis, and KidneyTox_v1.0 relied on only 565 compounds. Class imbalance, uncertain negative labels, limited chemical-space coverage, and incomplete external validation recurred across all five domains. The authors emphasize that a higher ROC-AUC in one organ domain does not indicate a superior modeling approach, because biological endpoints, dataset characteristics, and prediction tasks differ fundamentally across systems. Hepatotoxicity studies typically target drug-induced liver injury, cardiotoxicity studies focus on specific molecular targets like hERG and CaV1.2, and neurotoxicity encompasses heterogeneous endpoints from blood-brain barrier permeability to mammalian neurotoxicity.

Interpretability receives equally critical treatment. Explainable artificial intelligence techniques such as SHAP, attention mechanisms, saliency methods, and structural-alert identification can reveal which molecular features drive a prediction, but the authors draw a sharp line between explaining a model and establishing biological causation. High SHAP values or atom-level attribution maps identify features associated with a prediction without proving those features cause the toxic effect, and attribution results can shift with molecular representation, correlated descriptors, and model architecture. True mechanistic interpretation requires independent evidence from experimental validation, target-interaction data, toxicogenomics, pathway analysis, or adverse outcome pathways. The review also highlights the applicability domain concept, which defines the region of chemical space where a model can be trusted; predictions for compounds far from training data may be unreliable even when internal performance metrics look impressive.

On the regulatory front, the authors argue that strong retrospective performance alone cannot establish suitability for real-world safety assessment. Regulatory confidence depends on independent external validation, clearly defined endpoints, transparent and reproducible workflows, applicability-domain assessment, uncertainty evaluation, and adequate documentation of input data, preprocessing, algorithms, and known limitations. The review catalogs major toxicity databases including SIDER, FAERS, ADReCS, Pneumotox, PubChem, DrugBank, ChEMBL, ICE, LINCS L1000, and LiverTox, noting that clinical adverse-event sources differ fundamentally from experimental toxicity resources and should not be treated as interchangeable. Data quality, endpoint harmonization, and duplicate removal emerge as prerequisites before any model development begins.

Looking forward, the review identifies active learning and federated learning as strategies to overcome limited labeled data, foundation models as a path to stronger molecular representation, and multimodal learning as a way to combine molecular graphs, SMILES strings, structural images, and spectroscopic information. The authors call for improved negative-data construction, assay and endpoint harmonization, causal and mechanistic modeling, and prospective validation to build tools that are reliable, interpretable, and generalizable enough for regulatory applications. Their overarching message is measured but firm: artificial intelligence has demonstrated competitive and sometimes improved predictive performance in toxicology, but it should not be considered inherently superior to conventional methods. The technology’s real value depends on disciplined data curation, rigorous validation, and honest acknowledgment of where predictions can and cannot be trusted.

Subject of Research: Artificial intelligence methods for predicting chemical and drug toxicity across organ-specific endpoints

Article Title: A review on advances in chemical compound toxicology prediction

Article References: Hagag, A., Elsayed, N. A., & Abdelbaky, I. (2026). A review on advances in chemical compound toxicology prediction. Frontiers in Bioinformatics, 6, Article 1974935. https://doi.org/10.3389/fbinf.2026.1974935

Image Credits: AI Generated

DOI: 10.3389/fbinf.2026.1974935

Keywords: toxicity prediction, machine learning, deep learning, graph neural networks, QSAR, hepatotoxicity, cardiotoxicity, nephrotoxicity, neurotoxicity, explainable AI, drug safety, molecular fingerprints

Synthetic pyrethroids are synthetic derivatives of pyrethrins, the natural insecticides found in chrysanthemum flowers, and they are prized in agriculture and aquaculture for their potency against pests and relatively short environmental persistence. Yet they are also highly lipophilic, meaning they dissolve readily into fats rather than water, and they act on the nervous systems of fish by disrupting sodium channels in nerve membranes. Because of their chemical affinity for organic surfaces, pyrethroids bind tightly to plastic debris, and the review argues that this binding turns microplastics into mobile reservoirs of pesticide contamination within aquaculture systems.

The physicochemical mechanics of this partnership hinge on sorption and desorption dynamics, the processes by which chemicals attach to and detach from particle surfaces. Polymer crystallinity, or the degree to which plastic chains are ordered rather than amorphous, governs how much chemical a given particle can hold. Weathering, the combined effect of sunlight, abrasion, and microbial activity, alters surface chemistry and roughness, often increasing the available binding sites over time. Shifting water parameters such as salinity, pH, and temperature further modulate how strongly pyrethroids cling to plastic or release back into the water column, meaning that vector efficiency is not a fixed property but a moving target that responds to the environment.

Once contaminated particles are ingested, the classic assumption has been the so-called Trojan Horse effect, in which microplastics smuggle their chemical cargo deep into animal tissues and release it under digestive conditions, amplifying internal exposure beyond what the water alone would deliver. The review challenges this paradigm, pointing out that the picture is more nuanced than the label suggests. Some studies indicate that microplastics can dilute rather than concentrate chemical exposure under certain conditions, and the authors emphasize significant gaps in commercially relevant in vitro digestion models that attempt to simulate what actually happens inside a fish gut. Without better models, the true magnitude of the vector effect remains uncertain.

What is clearer is that vulnerability is not evenly distributed across farmed species. Benthic organisms, those living on or near the pond bottom, and species that are consumed whole rather than filleted face disproportionate exposure because they ingest particles directly from sediments or from the water column without the partial filtering that larger, selectively feeding animals perform. Tissue-specific bioaccumulation patterns suggest that pyrethroids carried on plastic surfaces may concentrate in particular organs rather than distributing uniformly, complicating both toxicological assessment and food safety monitoring.

The review also dissects how different aquaculture architectures shape exposure pathways. Recirculating aquaculture systems, or RAS, which reuse water through mechanical and biological filtration, concentrate particles and dissolved chemicals in ways that open-sea cages do not, since the latter rely on natural currents to dilute and disperse contaminants. Semi-intensive ponds, which blend natural productivity with supplemental feeding, sit somewhere in between, with hydrodynamics driven largely by wind, aeration, and fish movement rather than engineered flow. Each configuration produces a distinct profile of particle residence time, chemical partitioning, and trophic structure, meaning that risk cannot be generalized across the industry.

Trophic transfer, the movement of contaminants through food webs as predators consume prey, adds another layer of concern. When smaller organisms ingest contaminated particles and are then eaten by larger fish, the associated pyrethroids can move upward through the food chain, potentially reaching concentrations that exceed what direct water exposure alone would produce. The review highlights evidence that this pathway matters in aquaculture settings, particularly where natural forage organisms supplement formulated feeds, and it underscores the vulnerability of whole-consumed species whose digestive tracts, and therefore particle burdens, are eaten along with the muscle tissue.

The toxicological consequences compound across multiple biological systems. Pyrethroids are neurotoxic, and co-exposure with microplastics has been linked in experimental studies to heightened neuro-stress, disrupted behavior, and altered neurotransmitter activity in fish. Beyond the nervous system, the combination appears to compromise immune function, leaving animals more susceptible to disease, and to disturb gut microbiota, the microbial community that supports digestion and immune regulation. These sublethal effects translate directly into economic consequences for farm performance, including reduced growth rates, elevated mortality, and increased treatment costs, all of which erode the margins on which aquaculture operations depend.

Against this backdrop, the review identifies a significant regulatory blind spot. Maximum Residue Limits, or MRLs, the legal ceilings on pesticide residues in food products, are typically set for individual chemicals assessed in isolation, without accounting for the possibility that microplastic carriage may alter bioavailability, tissue distribution, or persistence of the residue in edible tissue. The authors argue that current frameworks do not capture the synergistic threat posed by the convergence of plastic pollution and routine pyrethroid application, leaving a gap between what regulations assume and what aquaculture ecosystems actually deliver to the consumer.

Looking forward, the review proposes a set of mitigation strategies aimed at closing this gap. Targeted green pharmacology, the development of plant-derived or biologically based alternatives to synthetic pesticides, could reduce the chemical burden entering farm systems in the first place, and the authors point to growing evidence that certain botanical feed additives can help fish cope with xenobiotic stress. Advanced filtration technologies, including improved mechanical separation and possibly biofilm-based capture, offer engineering routes to reduce particle loads before they accumulate. Ultimately, the authors call for a holistic paradigm shift in aquaculture risk assessment, one that treats microplastics and pesticides not as separate problems but as an interacting system whose combined behavior determines the real exposure profile of farmed fish and, by extension, the safety of the seafood that reaches the market.

Subject of Research: Microplastics as vectors for synthetic pyrethroid pesticides in aquaculture ecosystems

Article Title: The accelerating role of microplastics as vectors for synthetic pyrethroids: a critical review of bioaccumulation and trophic transfer in aquaculture ecosystems

Article References: Üstüner, E., Öz, M., İleri, E., & Dikel, S. (2026). The accelerating role of microplastics as vectors for synthetic pyrethroids: a critical review of bioaccumulation and trophic transfer in aquaculture ecosystems. Environmental Monitoring and Assessment, 198(9), Article 999. https://doi.org/10.1007/s10661-026-15850-9

Image Credits: AI Generated

DOI: 10.1007/s10661-026-15850-9

Keywords: microplastics, synthetic pyrethroids, aquaculture, bioaccumulation, trophic transfer, Trojan horse effect, food safety, neurotoxicity, pesticide vectors, recirculating aquaculture systems, maximum residue limits, green pharmacology

The durability of those responses is as consequential as their depth. Multiple myeloma is a cancer of plasma cells, the antibody-producing arm of the immune system, and it is notorious for relapsing after initially effective treatment. The anito-cel trial therefore tracked patients well beyond the initial response window. More than half of the participants remained progression-free two years after receiving the therapy, and 65 percent were still alive three years out. In a population whose disease had already resisted multiple prior lines of therapy, those figures suggest that a single treatment can reset the course of the disease for a meaningful fraction of patients rather than merely buying time between relapses. Long-term follow-up from larger trials will be needed to determine how many of these responses endure indefinitely, but the trajectory of the phase 1 cohort provides the kind of early evidence that cellular therapy researchers look for when deciding whether a new construct deserves expansion into pivotal testing.

Equally important to the efficacy numbers was the safety picture, which distinguished anito-cel from earlier generations of BCMA-directed CAR-T therapies. Serious immune-related and neurological side effects were uncommon in the trial, and no delayed neurological complications were reported at all. That last point matters because some existing CAR-T cell therapies targeting BCMA have been linked to rare but serious neurological side effects, including delayed syndromes that can emerge weeks after infusion and cause lasting impairment. Patients with relapsed or refractory myeloma already endure substantial treatment burden, and the prospect of a therapy whose most feared complications appear to be largely absent represents a genuine shift in the risk-benefit calculation. The investigators reported that the absence of neurotoxic and inflammatory side effects was particularly encouraging given the serious and potentially lasting impact such complications can have on patients’ lives.

The apparent explanation lies in the molecular design of the construct itself. CAR-T therapy begins with a patient’s own T cells, which are extracted, genetically engineered to recognize a specific target on cancer cells, multiplied in the laboratory, and infused back into the patient. The recognition element, called the binding domain, determines both how effectively the engineered cells kill cancer and how vigorously they activate the immune system. Excessive activation drives the inflammatory toxicities that have shadowed CAR-T therapy since its inception, including cytokine release syndrome and neurotoxicity. According to the researchers’ laboratory analyses, anito-cel’s engineered BCMA-binding domain, described as a novel D-domain binder, maintained the cancer-killing activity required for efficacy while limiting the excessive immune activation and other features that can impair CAR-T cell function or contribute to toxicity. In other words, the design appears to have decoupled potency from inflammation, a goal that has eluded many previous construct engineering efforts.

Senior author Michael R. Bishop, MD, Professor of Medicine and Director of the David and Etta Jonas Center for Cellular Therapy at UChicago Medicine, framed the findings in the context of what phase 1 trials usually deliver. He described the results as great for a phase 1 study, noting that even at a low dose of anito-cel the team was seeing complete responses in the majority of patients right away, and that not a single patient developed severe toxicities. That combination, he said, prompted the investigators to investigate further why the safety profile was so favorable. Lead author Matthew Frigault, MD, vice chair of Clinical Research in Hematology/Oncology and deputy program director of the Cellular Immunotherapy Program at the Mass General Brigham Cancer Institute, highlighted the absence of neurotoxic and inflammatory side effects as particularly encouraging, and suggested that the novel D-domain binder may contribute to that finding. He added that the combination of compelling efficacy and encouraging safety data lays important groundwork for continuing to investigate this new treatment option for patients.

To understand why a BCMA-targeted therapy is so significant for myeloma specifically, it helps to consider the biology of the disease. BCMA, or B-cell maturation antigen, is a surface protein expressed on myeloma cells and largely restricted to the late B-cell lineage, which makes it an attractive bullseye: attacking it spares most healthy tissue while striking the malignant plasma cells directly. Two BCMA-directed CAR-T products are already approved for myeloma, and they have demonstrated that redirecting T cells against this target can produce deep remissions. The field’s outstanding questions concern durability, accessibility, and above all toxicity. The delayed neurological syndromes associated with some BCMA constructs, along with the cytokine-driven inflammatory storms that can accompany infusion, have constrained how these therapies are used and pushed clinicians to weigh carefully which patients should receive them. A construct that preserves efficacy while reducing these risks could widen the population for whom CAR-T therapy is a reasonable choice.

The trial itself was a phase 1 study, the first step in the clinical evaluation pathway, typically designed to establish dosing and identify side effects in a small cohort rather than to prove efficacy against a control group. That all 38 participants responded is not a randomized comparison, and phase 1 results by definition come with caveats about patient selection and cohort size. The investigators and the therapy’s developer, Kite, a Gilead company, have responded to the results in the standard way: anito-cel is currently being evaluated in phase 2 and phase 3 clinical trials intended to confirm the findings in larger populations of patients with multiple myeloma. Those pivotal studies will subject the construct to the scrutiny of broader enrollment, longer follow-up, and, in some designs, direct comparison against existing standards of care, which is the evidence regulators require before approving a new therapy.

The study also illustrates the growing sophistication of CAR-T engineering as a discipline. Early CAR-T constructs were largely borrowed from antibody fragments grafted onto T-cell signaling machinery, and the field learned through experience that the fine details of the binding domain, including its affinity, the geometry of its engagement with the target, and the signaling architecture it feeds into, profoundly shape the behavior of the resulting cells. The D-domain binder characterized in this trial reflects a deliberate design strategy: rather than accepting whatever toxicity profile a construct happens to produce, researchers are now engineering the recognition module itself to tune immune activation downward while preserving cytotoxic function. The laboratory analyses accompanying the clinical results, in which the researchers characterized the binder’s properties, point toward a future in which CAR-T products are designed with toxicity management built in at the molecular level rather than managed afterward with supportive drugs.

For patients and their physicians, the practical meaning of the trial is straightforward. Multiple myeloma remains incurable with current approaches, and treatment typically proceeds through a sequence of proteasome inhibitors, immunomodulatory drugs, antibody therapies, and eventually BCMA-directed options, each of which the disease may eventually evade. Anito-cel’s phase 1 data indicate that a next-generation BCMA CAR-T may deliver deep, durable remissions even in patients whose myeloma has already exhausted many prior therapies, with a safety profile that avoids the delayed neurological complications that have made some existing products difficult to deploy. The investigators caution that confirmation in larger trials is essential, and the ongoing phase 2 and phase 3 studies will determine whether the striking phase 1 experience holds up at scale. But the results, published in The New England Journal of Medicine under the title describing anito-cel as a D-domain BCMA CAR-T with phase 1 results and binder characterization, establish a credible new candidate in the cellular therapy pipeline for one of hematologic oncology’s most persistent challenges, and they demonstrate that rational engineering of the binding domain can translate into measurable clinical benefit at the bedside.

Subject of Research: A phase 1 trial of the BCMA-targeted CAR-T cell therapy anito-cel for relapsed or refractory multiple myeloma

Article Title: Novel CAR-T therapy shows promising safety and durable responses in hard-to-treat multiple myeloma

Article References: Novel CAR-T therapy shows promising safety and durable responses in hard-to-treat multiple myeloma. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: multiple myeloma, CAR-T cell therapy, anito-cel, BCMA, phase 1 clinical trial, immunotherapy, D-domain binder, neurotoxicity, Mass General Brigham, University of Chicago Medicine, New England Journal of Medicine, relapsed refractory cancer

The economic stakes could hardly be higher. Developing an innovative drug takes on average more than ten years, roughly eight of which are consumed by clinical trials, and the out-of-pocket cost per approved drug is estimated at approximately 173 million dollars, rising to about 879 million once failures and capital costs are included. Between 2014 and 2023, the likelihood that a candidate entering Phase I would ultimately win FDA approval was a mere 6.7 percent, with Phase II acting as the deadliest filter, advancing only 28 percent of the molecules it receives. A comprehensive review of AstraZeneca’s small-molecule projects from 2005 to 2010 showed that neural toxicity in the central nervous system accounts for nearly one quarter of failures across the entire discovery process, yet only a small fraction of these problems becomes evident during preclinical Good Laboratory Practice toxicity testing.

The reason most functional neural side effects slip through is that they do not leave the structural fingerprints that standard animal models are designed to catch. Disturbances of central nervous system function such as dizziness, anxiety, seizures or depression are far harder to predict than structural damage visible in histological sections. The Northeastern team’s Neuro-Behavioral Assay, originally developed to evaluate anesthetics, attacks this gap from a different direction by exploiting a biological quirk of the brown ghost knifefish, Apteronotus macrostomus of the Apteronotus leptorhynchus species group. This teleost generates a continuous, extraordinarily regular train of electric organ discharges, with amplitudes below one volt and frequencies approaching one kilohertz, and every single discharge is driven one-to-one by command spikes from an endogenous oscillator in the medulla oblongata called the pacemaker nucleus.

That tight coupling is what makes the assay so powerful: the frequency of the electric organ discharge is identical to the frequency of the neural oscillations of the pacemaker nucleus, so the fish’s electric behavior serves as a real-time, non-invasive proxy of activity in a specific brainstem structure. Electrodes built into an opaque PVC shelter tube record the signal continuously, which is then digitized at 50 kilohertz and parsed by automated procedures. Because the discharge frequency is extremely stable at constant temperature, with coefficients of variation in this study ranging from only 0.074 to 0.303 percent, even small drug-induced signals stand out clearly against the baseline. Two further readouts are extracted from the same recording: the rate of chirps, transient amplitude-frequency modulations driven by diencephalic input that probe higher brain function, and a locomotor activity index derived from amplitude changes as the fish moves relative to the electrodes.

The fourth and newest readout emerged from this study. After a 15-minute immersion in a 0.002 percent weight-per-volume solution of dextromethorphan, fish began quivering their anal and pectoral fins within two to six minutes, followed by jerking motions and belly-up postures, and then the most conspicuous effect: complete rotations about the longitudinal axis in the transverse plane, including full 360-degree turns that the researchers named Full Transverse Plane Rotations. These rotations, essentially absent in untreated and vehicle-treated fish, erupted in bouts and reached staggering rates of up to 165 per five minutes. Scoring was performed by an investigator blinded to treatment condition, and in five of the six fish the increase over baseline was highly significant, with rates persisting for hours and, in some individuals, for at least two days.

The dose-response data cemented the finding. Across four concentrations including a water-only control, a Spearman correlation revealed a significant moderate negative relationship between dextromethorphan concentration and electric organ discharge frequency, with all six fish showing frequency decreases at the highest dose ranging from 1.3 to 7.1 hertz against a mean control change of just 0.4 hertz. Chirp rate, which rose three- to eighteen-fold above baseline in three of the six fish, showed a strong positive correlation with concentration, and the net rotation rate correlated even more strongly. Notably, the chirp response appeared in only about half the animals, echoing a bimodal pattern previously seen with the anesthetics tricaine methanesulfonate and eugenol, a variability the researchers suggest may reflect the genomic and phenotypic diversity of the wild-type fish they deliberately use, a factor gaining recognition in drug discovery more broadly.

Crucially, the pattern of readouts produced by dextromethorphan differs from the signatures of every compound tested before. The anesthetics triggered dramatic locomotor suppression and discharge frequency drops exceeding 100 and reaching nearly 300 hertz, while diphenhydramine produced a more modest 40-to-100 hertz reduction without chirp or locomotor changes, and none of these drugs ever induced rotations. Dextromethorphan, by contrast, produced its characteristic trio of mild frequency decline, elevated chirping and vigorous rotational behavior, and its effects lingered far longer than the few-hour recovery typical of the other drugs, in one fish still clearly elevated on the third day after exposure.

Because the electric readouts are controlled exclusively by the pacemaker nucleus and its upstream inputs, any compound that alters them has demonstrably crossed the blood-brain barrier, which in ray-finned bony fish is built from the same tight-junctioned brain microvascular endothelial cells as in mammals and relies on conserved molecules such as Claudin-5. The authors therefore hypothesize that readout changes predict blood-brain barrier permeability in humans with reasonable likelihood, a capability current in silico and in vitro technologies have achieved only with limited success. The well-mapped circuitry of the pacemaker system also opens the door to target deconvolution: roughly 110 cells, about 90 pacemaker cells and 20 relay cells, generate the oscillation, and transient frequency changes are mediated by NMDA receptors while chirp-like modulations depend on AMPA receptors at the relay cells. The team proposes that dextromethorphan, a known NMDA receptor blocker, lowers discharge frequency by dampening excitatory drive at the relay cells, while a resulting glutamate surge stimulates AMPA receptors and drives chirping, a mechanism paralleling the drug’s antidepressant action that led to its repurposing for depression.

Translational parallels abound. Dextromethorphan is metabolized by the polymorphic enzyme CYP2D6, with 84 percent of studied volunteers classed as fast metabolizers and 9 percent as slow metabolizers whose drug half-life stretches from 2 to as long as 19 hours, and the researchers speculate that the fish homologue’s genetic variation might underlie the inter-individual differences observed in the assay. The prolonged behavioral impairment in the fish also contrasts strikingly with the popular assumption that intoxication wears off in three to six hours; the only controlled human study of supratherapeutic doses found psychomotor impairment, including a balance task, still measurable 360 minutes after administration. If the hypothesis holds that the diversity of affected readouts reflects a broad spectrum of side effects, as dextromethorphan’s documented profile of dizziness, restlessness, hallucinations and seizures suggests, the quadrimodal assay could let developers catch functional neural toxicity long before it detonates in a Phase II trial, quantify recovery on a continuous rather than binary scale, and even optimize molecules for brain penetration, all with a fish that broadcasts its brain activity through the water.

Subject of Research: A quadrimodal neuro-behavioral in vivo assay in weakly electric fish for detecting adverse central nervous system effects of drugs, validated with dextromethorphan

Article Title: Quadrimodal in vivo assay for detection of adverse neuro-behavioral effects: validation using dextromethorphan

Article References: Zupanc, G. K. H., Zodl, A. L., & Sabha, N. M. (2026). Quadrimodal in vivo assay for detection of adverse neuro-behavioral effects: validation using dextromethorphan. Frontiers in Drug Discovery, 6, Article 1917655. https://doi.org/10.3389/fddsv.2026.1917655

Image Credits: AI Generated

DOI: 10.3389/fddsv.2026.1917655

Keywords: dextromethorphan, neuro-behavioral assay, weakly electric fish, drug discovery, neurotoxicity, blood-brain barrier, pacemaker nucleus, electric organ discharge, NMDA receptors, AMPA receptors, preclinical screening, Apteronotus

The particles at the center of the study came from Okhla Industrial Area-I in New Delhi, a location that blends industrial emissions with residential sources and therefore represents a realistic mixture of what millions of residents breathe daily. The researchers sampled continuously for twenty-four-hour periods between November 22, 2018 and June 30, 2019, capturing seasonal variation across Delhi’s notorious winter pollution episodes and the comparatively cleaner summer months. PM1 refers to particles with an aerodynamic diameter of one micrometer or less, a size class so fine that individual particles are invisible under ordinary light microscopes and can penetrate deep into lung tissue. Because these particles carry adsorbed toxic metals on their surfaces, toxicologists consider the chemical composition of PM1, not merely its mass, to be the primary driver of its harm.

To prepare the exposure material, the collected PM1 samples were extracted from their filters and processed both for chemical analysis and for injection. Eighteen male Sprague-Dawley rats were divided into three equal groups: a control group receiving saline, a low-dose group receiving 80.5 micrograms per kilogram of body weight, and a high-dose group receiving 244.6 micrograms per kilogram. The two exposed groups were dosed intravenously once a week for seventy days, a regimen designed to simulate the sustained systemic delivery of particle-associated metals that occurs when inhaled ultrafine particles cross from the lungs into the circulation. This route of administration is a deliberate departure from conventional inhalation studies, and the researchers are explicit that it does not fully replicate the physiology of breathing polluted air, a caveat they flag as a key limitation of the design.

After the seventy-day exposure window closed, the animals were euthanized and their organs harvested for analysis. The team used inductively coupled plasma optical emission spectroscopy, or ICP-OES, a technique that excites metal atoms in a plasma torch and reads the characteristic wavelengths of light they emit, allowing precise quantification of twelve metals: aluminum, iron, nickel, chromium, copper, lead, manganese, molybdenum, zinc, cobalt, cadmium and vanadium. To test whether differences between groups were statistically meaningful, the researchers applied one-way analysis of variance with a significance threshold of p less than 0.05, followed by Tukey’s post hoc test for multiple comparisons between the control, low-dose and high-dose cohorts. This analytical pipeline let them rank the metals not only by how much they accumulated overall but by where in the body each one preferentially settled.

The resulting hierarchy, ordered from highest to lowest concentration across the rats’ tissues, placed aluminum and iron at the top, followed by nickel, chromium, copper, lead, manganese, molybdenum, zinc, cobalt, cadmium and vanadium. Aluminum showed particularly high accumulation in the kidneys, the spleen and the cerebellum, the part of the brain responsible for motor coordination. Iron, an element whose overload is known to catalyze the production of damaging reactive oxygen species, concentrated significantly in the liver, spleen, heart and cerebellum. The convergence of both elements in the cerebellum is notable, because iron-driven oxidative stress and aluminum’s association with neurodegenerative processes have each been examined extensively in prior toxicological literature, and the metal-ranking framework in this study provides a systematic basis for pursuing those links in the context of real-world particulate exposures.

Perhaps the most consequential finding concerns lead. The study documented substantial accumulation of lead in the bloodstream and in vital organs including the lungs, heart, kidneys, liver and brain, indicating that once it detaches from its particle carrier, lead is highly mobile and disseminates widely through the body. Lead is a neurotoxin with no known safe exposure threshold, and its molecular targets include synaptotagmin I, a protein central to neurotransmitter release. The observation that intravenously delivered PM1-borne lead reached both the cerebrum and cerebellum underscores why lead remains the pollutant of greatest concern in this exposure scenario, even though it ranked only sixth in overall concentration across tissues.

Chromium and copper told their own organ-specific stories. Chromium accumulated significantly in the blood, kidneys, heart, cerebrum and cerebellum, consistent with its status as a carcinogenic metal whose hexavalent form can cross cellular membranes and damage DNA. Copper concentrated in the blood, lungs and heart, a distribution that resonates with clinical observations linking elevated myocardial copper to heart failure. The study also identified clear single-organ specialists: cobalt localized primarily in the spleen, nickel in the heart, and zinc in the kidneys. Vanadium, another transition metal associated with cardiopulmonary toxicity in earlier inhalation studies, was found at high levels in the lungs, kidneys and heart. Each of these patterns suggests that the health consequences of breathing metal-laden air may depend less on the total metal dose than on which organs each specific element finds hospitable.

Not every notorious toxin behaved as expected. Cadmium, a metal infamous for kidney damage, bone disease and endocrine disruption, showed no significant organ-level accumulation, and neither did certain other toxic trace metals. The authors suggest three non-exclusive explanations: constrained bioavailability of the metal in the particle matrix, limited cellular uptake once in circulation, or distinct metabolic clearance pathways that removed the element before it could lodge in tissue. This null result is scientifically valuable, because it demonstrates that the toxicity of particulate matter cannot be predicted simply by summing the hazardous reputations of its constituents. The same particle that delivers lead to the brain may carry cadmium that the body efficiently eliminates, meaning risk assessments must be element-specific rather than mass-based.

The intravenous delivery method deserves careful interpretation. By bypassing the respiratory tract entirely, the design isolates what happens after particles and their dissolved metals enter the blood, which is precisely the step most relevant to the emerging paradigm that air pollution’s deadliest effects, from heart attacks to strokes, arise from particles that escape the lungs rather than damage them directly. But the trade-off is realism: inhaled particles face filtration in the nose and airways, deposition patterns in the alveoli, interactions with lung immune cells and variable rates of translocation into circulation. The doses delivered by injection likely exceed what translocation alone would achieve, and the weekly bolus schedule differs from continuous low-level exposure. The authors acknowledge this limitation directly, positioning the study as a controlled probe of systemic metal fate rather than a model of the full inhalation pathway.

Even with those caveats, the study’s organ-specific metal map carries weighty implications for public health in Delhi and other megacities where PM1 concentrations routinely soar. It suggests that chronic exposure to submicron particulate matter could quietly load the kidneys, spleen, cerebellum, liver and heart with aluminum and iron, seed the heart with nickel, and circulate lead and chromium throughout the vasculature and brain. Research cited by the team connects these metals to oxidative stress, inflammation, hypertension, atherosclerosis, hepatic fibrosis, kidney damage and neurodegeneration, so each localization pattern hints at a plausible disease pathway waiting to be tested. As the authors and their collaborators continue this line of work, the challenge will be bridging the gap between this controlled murine model and the messier reality of human exposure, where inhalation, diet, genetics and comorbidities all shape the same toxic ledger that seventy days of injections so cleanly revealed.

Subject of Research: Organ-specific accumulation of metals associated with submicron particulate matter (PM1) following intravenous exposure in rats

Article Title: Preferential accumulation of PM₁-associated metals in Sprague-Dawley rats’ tissues/organs as a consequence of intravenous administration of Delhi’s airborne PM₁

Article References: Mehta, D., Kumar, M., Kumar, N., & Srivastava, A. (2026). Preferential accumulation of PM₁-associated metals in Sprague-Dawley rats’ tissues/organs as a consequence of intravenous administration of Delhi’s airborne PM₁. Air Quality, Atmosphere & Health, 19(10), Article 236. https://doi.org/10.1007/s11869-026-02125-y

Image Credits: AI Generated

DOI: 10.1007/s11869-026-02125-y

Keywords: PM1, particulate matter, air pollution, Delhi, heavy metals, toxicology, Sprague-Dawley rats, intravenous exposure, ICP-OES, organ accumulation, lead, neurotoxicity

The study, published in the journal iScience, is the first preclinical evaluation of minibeam radiation therapy combined with CAR T cell therapy in an orthotopic model of primary CNS lymphoma. The researchers, working at Institut Curie and collaborating institutions, implanted luciferase-labeled human lymphoma cells into the caudate nucleus of immunodeficient RRGS rats, which carry human SIRPα and can accept human tumor and immune cells. Seven days after implantation, once bioluminescence imaging confirmed tumor establishment, the animals received either conventional radiotherapy or minibeam radiation therapy, and CAR T cells were infused intravenously either three or five days later.

The technical distinction between the two radiation modalities lies at the heart of the finding. Conventional radiotherapy delivers a uniform field of radiation across the target volume. Minibeam radiation therapy instead splits the dose into an array of narrowly collimated beams, in this case 700 micrometers wide with a center-to-center spacing of about 1.4 millimeters, producing a striking peak-and-valley dose landscape. At a mean dose of 30 Gy, the peak regions received roughly 83 Gy while the valleys received only about 4.5 Gy. This spatial modulation allows tumor cells to be effectively irradiated while normal tissue, which benefits from the low-dose valleys and the ability of healthy cells to migrate in and repair, is largely spared.

That difference in physics translated into a difference in biology. Earlier work by the same group and others had shown that minibeam irradiation, unlike conventional radiotherapy, increases intratumoral CD8-positive tissue-resident memory T cells, B cells, and type 1 conventional dendritic cells, while conventional beams tend to promote immunosuppressive signaling. Because CAR T cell therapy depends on a permissive immune microenvironment, the researchers hypothesized that the choice of radiation modality could determine whether bridging radiotherapy helps or hinders the cellular therapy.

The survival data bore this out. In the definitive six-arm experiment, untreated animals survived a median of just 18 days. CAR T cell monotherapy at a dose of 5 million cells extended median survival to 43 days, with long-term survivors beyond 90 days in about a third of that group. Minibeam radiation alone at 30 Gy produced a median survival of 36 days, statistically indistinguishable from conventional radiotherapy alone at 32 days. But when minibeam was combined with CAR T cells infused five days after irradiation, median survival jumped to 83 days, the longest of any irradiated group, and a quarter of those animals survived long term. By contrast, adding CAR T cells to conventional radiotherapy yielded a median survival of only 37 days, essentially no better than conventional radiotherapy by itself.

The authors are careful about what this does and does not mean. Because the minibeam-plus-CAR T combination did not significantly outperform CAR T monotherapy, the data do not demonstrate pharmacological synergy. What they do show, the researchers argue, is that minibeam radiation provides a permissive, non-antagonistic context that preserves CAR T efficacy, whereas conventional radiotherapy appears to blunt it. In a clinical landscape where bridging radiotherapy between leukapheresis and CAR T infusion is increasingly common, that distinction could matter enormously. Clinical intervals between radiation and CAR T infusion range from 2 to 38 days, and the new results suggest that the type of beam, not merely the dose or timing, may shape the outcome.

Histopathology reinforced the survival curves. At the study endpoint, only 37.5 percent of animals in the minibeam-plus-CAR T group had detectable residual tumor, the lowest rate of any group, compared with 100 percent of untreated and conventionally irradiated animals and 64.3 percent of those receiving CAR T alone. The residual lesions in minibeam-treated animals were structurally loose, with fewer vascular abnormalities, less inflammatory infiltration, and minimal neural damage. Conventional radiotherapy, in contrast, was associated with diffuse spread, more frequent cavitation, greater neuronal damage, and fibrotic scarring. Strikingly, the combination of conventional radiotherapy with CAR T cells was linked to extensive brain necrosis, suggesting that rapid tumor clearance can outpace normal tissue repair and worsen neurotoxicity.

Tolerability data added a further point in minibeam’s favor. Animals across all groups generally gained weight during follow-up, and no overt systemic toxicity was observed. Skin injuries, a common radiation side effect, were variable in the conventionally irradiated groups and required topical treatment, whereas only a handful of minibeam animals showed minor lesions attributed mainly to self-scratching. The authors caution, however, that the study did not include dedicated neurological, behavioral, or detailed normal-brain toxicity assessments, so the safety profile remains incomplete.

The timing question yielded a suggestive but not definitive answer. Infusing CAR T cells five days after minibeam irradiation produced a median survival of 83 days versus 35.5 days for the three-day interval, and residual tumor incidence fell from 87.5 percent to 37.5 percent with the longer delay, but the difference did not reach statistical significance. The researchers describe the five-day interval as a reasonable scheduling strategy that warrants further investigation rather than an established optimum.

Several limitations frame the conclusions. The model used immunodeficient rats, so endogenous immune contributions to minibeam’s effects could not be assessed. Only CD19 immunohistochemistry was used to track tumor, meaning reduced staining could reflect antigen loss or modulation rather than true tumor clearance. The CAR T product was predominantly CD4-positive with relatively few cytotoxic CD8 cells, and only male animals were studied. The study was also designed as a therapeutic proof of concept, not a mechanistic investigation, so questions of CAR T trafficking, persistence, and tumor microenvironment remodeling remain open. Even so, the work makes a compelling case that radiotherapy modality should be a deliberate design choice when combining radiation with adoptive cell therapies. With minibeam radiation therapy already advancing through a canine trial that reported pathological complete remission in 71 percent of de novo brain tumors and its first human treatments, the prospect of a radiation technology that debulks tumors without sabotaging the immune cells sent to finish the job is moving steadily closer to the clinic.

Subject of Research: Combining minibeam radiation therapy with CD19 CAR T cell therapy in a rat model of primary central nervous system lymphoma

Article Title: Minibeam radiation therapy supports CAR T therapy in a rat model of CNS lymphoma

Article References: Iturri, L., Potiron, S., Sun, L., Juchaux, M., Gilbert, C., Espenon, J., Anegon, I., Ménoret, S., Soussain, C., Amigorena, S., Alcantara, M., & Prezado, Y. (2026). Minibeam radiation therapy supports CAR T therapy in a rat model of CNS lymphoma. iScience, 29(11), Article 117827. https://doi.org/10.1016/j.isci.2026.117827

Image Credits: AI Generated

DOI: 10.1016/j.isci.2026.117827

Keywords: minibeam radiation therapy, CAR T cells, primary CNS lymphoma, radiotherapy, immunotherapy, CD19, brain tumor, preclinical study, spatially fractionated radiation, tumor microenvironment, neurotoxicity, iScience