We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
To save content items to your account,
please confirm that you agree to abide by our usage policies.
If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account.
Find out more about saving content to .
To save content items to your Kindle, first ensure no-reply@cambridge.org
is added to your Approved Personal Document E-mail List under your Personal Document Settings
on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part
of your Kindle email address below.
Find out more about saving to your Kindle.
Note you can select to save to either the @free.kindle.com or @kindle.com variations.
‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi.
‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.
Background: TERT promoter mutation (TPM) is an established biomarker in meningiomas associated with aberrant TERT expression and reduced progression-free survival (PFS). TERT expression, however, has also been observed even in tumours with wildtype TERT promoters (TP-WT). This study aimed to examine TERT expression and clinical outcomes in meningiomas. Methods: TERT expression, TPM status, and TERT promoter methylation of a multi-institutional cohort of meningiomas (n=1241) was assessed through nulk RNA sequencing (n=604), Sanger sequencing of the promoter (n=1095), and methylation profiling (n=1218). 380 Toronto meningiomas were used for discovery, and 861 external institution samples were compiled as a validation cohort. Results: Both TPMs and TERTpromoter methylation were associated with increased TERT expression and may represent independent mechanisms of TERT reactivation. TERT expression was detected in 30.4% of meningiomas that lacked TPMs, was associated with higher WHO grades, and corresponded to shorter PFS, independent of grade and even among TP-WT tumours. TERT expression was associated with a shorter PFS equivalent to those of TERT-negative meningiomas of one higher grade. Conclusions: Our findings highlight the prognostic significance of TERT expression in meningiomas, even in the absence of TPMs. Its presence may identify patients who may progress earlier and should be considered in risk stratification models.
Background: Meningiomas are the most common intracranial tumors. Radiotherapy (RT) serves as an adjunct following surgical resection; however, response varies. RTOG-0539 is a prospective, phase 2, trial that stratified patients risk groups based on clinical and pathological criteria, providing key benchmarks for RT outcomes. This is the first study that aims to characterize the molecular landscape of an RT clinical trial in meningiomas. Methods: Tissue from 100 patients was analyzed using DNA methylation, RNA sequencing, and whole-exome sequencing. Copy number variations and mutational profiles were assessed to determine associations with meningioma aggressiveness. Tumors were molecularly classified and pathway analyses were conducted to identify biological processes associated with RT response. Results: High-risk meningiomas exhibited cell cycle dysregulation and hypermetabolic pathway upregulation. 1p loss and 1q gain were more frequent in aggressive meningiomas, and NF2 and non-NF2 mutations co-occurred in some high-risk tumors. Molecular findings led to the reclassification of several cases, highlighting the limitations of histopathologic grading alone. Conclusions: This is the first study to comprehensively characterize the molecular landscape of any RT trial in meningioma, integrating multi-omic data to refine treatment stratification. Findings align with ongoing genomically driven meningioma clinical trials and underscore the need for prospective tissue banking to enhance biomarker-driven treatment strategies.
Background: The WHO grade of meningioma was updated in 2021 to include homozygous deletions of CDKN2A/B and TERT promotor mutations. Previous work including the recent cIMPACT-NOW statement have discussed the potential value of including chromosomal copy number alterations to help refine the current grading system. Methods: Chromosomal copy number profiles were inferred from from 1964 meningiomas using DNA methylation. Regularized Cox regresssion was used to identify CNAs independenly associated with post-surgical and post-RT PFS. Outcomes were stratified by WHO grade and novel CNAs to assess their potential value in WHO critiera. Results: Patients with WHO grade 1 tumours and chromosome 1p loss had similar outcomes to those with WHO grade 2 tumours (median PFS 5.83 [95% CI 4.36-Inf] vs 4.48 [4.09-5.18] years). Those with chromosome 1p loss and 1q gain had similar outcomes to those with WHO grade 3 cases regardless of initial grade (median PFS 2.23 [1.28-Inf] years WHO grade 1, 1.90 [1.23-2.25] years WHO grade 2, compared to 2.27 [1.68-3.05] years in WHO grade 3 cases overall). Conclusions: We advocate for chromosome 1p loss being added as a criterion for a CNS WHO grade of 2 meningioma and addition of 1q gain as a criterion for a CNS WHO grade of 3.
Background: We previously developed a DNA methylation-based risk predictor for meningioma, which has been used locally in a prospective fashion. As a follow-up, we validate this model using a large prospective cohort and introduce a streamlined next-generation model compatible with newer methylation arrays. Methods: The performance of our next-generation predictor was compared with our original model and standard-of-care 2021 WHO grade using time-dependent receiver operating characteristic curves. A nomogram was generated by incorporating our methylation predictor with WHO grade and extent of resection. Results: A total of 1347 meningioma cases were utilized in the study, including 469 prospective cases from 3 institutions and a retrospective cohort of 100 WHO grade 2 cases for model validation. Both the original and next-generation models significantly outperformed 2021 WHO grade in predicting postoperative recurrence. Dichotomizing into grade-specific risk subgroups was predictive of outcome within both WHO grades 1 and 2 tumours (log-rank p<0.05). Multivariable Cox regression demonstrated benefit of adjuvant radiotherapy in high-risk cases specifically, reinforcing its informative role in clinical decision making. Conclusions: This next-generation DNA methylation-based meningioma outcome predictor significantly outperforms 2021 WHO grading in predicting time to recurrence. This will help improve prognostication and inform patient selection for RT.
Background: Meningiomas exhibit considerable heterogeneity. We previously identified four distinct molecular groups (immunogenic, NF2-wildtype, hypermetabolic, proliferative) which address much of this heterogeneity. Despite their utility, the stochasticity of clustering methods and the requirement of multi-omics data limits the potential for classifying cases in the clinical setting. Methods: Using an international cohort of 1698 meningiomas, we constructed and validated a machine learning-based molecular classifier using DNA methylation alone. Original and newly-predicted molecular groups were compared using DNA methylation, RNA sequencing, whole exome sequencing, and clinical outcomes. Results: Group-specific outcomes in the validation cohort were nearly identical to those originally described, with median PFS of 7.4 (4.9-Inf) years in hypermetabolic tumors and 2.5 (2.3-5.3) years in proliferative tumors (not reached in the other groups). Predicted NF2-wildtype cases had no NF2 mutations, and 51.4% had others mutations previously described in this group. RNA pathway analysis revealed upregulation of immune-related pathways in the immunogenic group, metabolic pathways in the hypermetabolic group and cell-cycle programs in the proliferative group. Bulk deconvolution similarly revealed enrichment of macrophages in immunogenic tumours and neoplastic cells in hypermetabolic/proliferative tumours. Conclusions: Our DNA methylation-based classifier faithfully recapitulates the biology and outcomes of the original molecular groups allowing for their widespread clinical implementation.
Background: The combination of PARP inhibitor and immune checkpoint inhibitors have been proposed as a potentially synergistic combinatorial treatment in IDH mutant glioma, targeting dysregulated homologous recombination repair pathways. This study analyzed the cell-free DNA methylome of patients in a phase 2 trial using the PARP inhibitor Olaparib and the PD-1 inhibitor Durvalumab. Methods: Patients with recurrent high-grade IDH-mutant gliomas were enrolled in a phase II open-label study (NCT03991832). Serum was collected at baseline and monthly and cell-free methylated DNA immunoprecipitation and high-throughput sequencing (cfMeDIP-seq) was performed. Binomial GLMnet models were developed and model performance was assessed using validation set data. Results: 29 patients were enrolled between 2020–2023. Patients received olaparib 300mg twice daily and durvalumab 1500mg IV every 4 weeks. The overall response rate was 10% via RANO criteria. 144 plasma samples were profiled with cfMeDIP-seq along with 30 healthy controls. The enriched circulating tumour DNA methylome during response periods exhibited a highly specific signature, accurately discriminating response versus failure (AUC 0.98 ± 0.03). Additionally, samples that were taken while on treatment were able to be discriminated from samples off therapy (AUC 0.74 ± 0.11). Conclusions: The cell-free plasma DNA methylome exhibits highly specific signatures that enable accurate prediction of response to therapy.
Background: We’ve adopted a novel approach that combines cellular barcoding with CRISPR/Cas-9 technology and single-cell RNA sequencing known as continuous lineage tracing to track the development, treatment and inevitable recurrence of glioblastoma. Methods: Patient derived glioma initiating cell lines were engineered with expressed DNA barcodes with CRISPR/Cas-9 targets and engrafted into NOD scid-mice. Clonal and relationships are surmised through identification of expressed barcodes, and cells were characterized by their transcriptional profiles. Phylogenetic lineage trees are created using lineage reconstructive algorithms to define cell fitness and expansion. Results: Our work has revealed a significant amount of intra-clonal cell state heterogeneity, suggesting that tumour cells engage in phenotype switching prior to therapeutic intervention. Phylogenetic lineage trees allowed us to define a gene signature of cell fitness. GBMs exist along a transcriptional gradient between undifferentiated but “high-fit” cells and terminally differentiated, “low-fit” cells, lending further evidence that these tumours consist of pools of cells that are capable of recapitulating the tumour microenvironment after treatment. Conclusions: We have successfully engineered a set of glioma initiating tumours with a novel lineage tracing technique, creating a powerful tool for real-time tracing of tumour growth through the analysis of highly detailed singe-cell RNA sequencing data with associated clonal and phylogenetic relationships.
Liouville-type theorems for the steady incompressible Navier–Stokes system are investigated for solutions in a three-dimensional (3-D) slab with either no-slip boundary conditions or periodic boundary conditions. When the no-slip boundary conditions are prescribed, we prove that any bounded solution is trivial if it is axisymmetric or $ru^r$ is bounded, and that general 3-D solutions must be Poiseuille flows when the velocity is not big in $L^\infty$ space. When the periodic boundary conditions are imposed on the slab boundaries, we prove that the bounded solutions must be constant vectors if either the swirl or radial velocity is independent of the angular variable, or $ru^r$ decays to zero as $r$ tends to infinity. The proofs are based on the fundamental structure of the equations and energy estimates. The key technique is to establish a Saint-Venant type estimate that characterizes the growth of the Dirichlet integral of non-trivial solutions.
Two studies were conducted in 2022 and 2023 near Rocky Mount and Clayton, NC, to determine the optimal granular ammonium sulfate (AMS) rate and application timing for pyroxasulfone-coated AMS. In the rate study, AMS rates included 161, 214, 267, 321, 374, 428, and 481 kg ha−1, equivalent to 34, 45, 56, 67, 79, 90, and 101 kg N ha−1, respectively. All rates were coated with pyroxasulfone at 118 g ai ha−1 and topdressed onto 5- to 7-leaf cotton. In the timing study, pyroxasulfone (118 g ai ha−1) was coated on AMS and topdressed at 321 kg ha−1 (67 kg N ha−1) onto 5- to 7-leaf, 9- to 11-leaf, and first bloom cotton. In both studies, weed control and cotton tolerance to pyroxasulfone-coated AMS were compared to pyroxasulfone applied POST and POST-directed. The check in both studies received non-herbicide-treated AMS (321 kg ha−1). Before treatment applications, all plots (including the check) were maintained weed-free with glyphosate and glufosinate. In both studies, pyroxasulfone applied POST was most injurious (8% to 16%), while pyroxasulfone-coated AMS resulted in ≤4% injury. Additionally, no differences in cotton lint yield were observed in either study. With the exception of the lowest rate of AMS (161 kg ha−1; 79%), all AMS rates coated with pyroxasulfone controlled Palmer amaranth ≥83%, comparably to pyroxasulfone applied POST (92%) and POST-directed (89%). In the timing study, the application method did not affect Palmer amaranth control; however, applications made at the mid- and late timings outperformed early applications. These results indicate that pyroxasulfone-coated AMS can control Palmer amaranth comparably to pyroxasulfone applied POST and POST-directed, with minimal risk of cotton injury. However, the application timing could warrant additional treatment to achieve adequate late-season weed control.
Novel management strategies for controlling smutgrass have potential to influence sward dynamics in bahiagrass forage systems. This experiment evaluated population shifts in bahiagrass forage following implementation of integrated herbicide and fertilizer management plans for controlling smutgrass. Herbicide treatments included indaziflam applied PRE, hexazinone applied POST, a combination of PRE + POST herbicides, and a nonsprayed control. Fertilizer treatments included nitrogen, nitrogen + potassium, and an unfertilized control. The POST treatment reduced smutgrass coverage regardless of PRE or fertilizer application by the end of the first season and remained low for the 3-yr duration of the experiment (P < 0.01). All treatments, including nontreated controls, reduced smutgrass coverage during year 3 (P < 0.05), indicating that routine harvesting to remove the biomass reduced smutgrass coverage. Bahiagrass cover increased at the end of year 1 with POST treatment (P < 0.01), but only the POST + fertilizer treatment maintained greater bahiagrass coverage than the nontreated control by the end of year 3 (P < 0.05). Expenses associated with the POST + fertilizer treatment totaled US$348 ha−1 across the 3-yr experiment. Other smutgrass control options could include complete removal of biomass (hay production) and pasture renovation, which can cost 3-fold or greater more than POST + fertilizer treatment. Complete removal of biomass may reduce smutgrass coverage by removing mature seedheads, but at a much greater expense of US$2,835 to US$5,825 ha−1, depending on herbicide and fertilizer inputs. Bahiagrass renovation is US$826 ha−1 in establishment costs alone. When pasture production expenses are included for two seasons postrenovation, the total increases to US$1,120 ha−1 across three seasons. The importance of hexazinone and fertilizer as components of smutgrass control in bahiagrass forage was confirmed in this study. Future research should focus on the biology of smutgrass and the role of a PRE treatment in a long-term, larger-scale forage system.
An experiment was conducted in 2022 and 2023 near Rocky Mount and Clayton, NC, to evaluate residual herbicide-coated fertilizer for cotton tolerance and Palmer amaranth control. Treatments included acetochlor, atrazine, dimethenamid-P, diuron, flumioxazin, fluometuron, fluridone, fomesafen, linuron, metribuzin, pendimethalin, pyroxasulfone, pyroxasulfone + carfentrazone, S-metolachlor, and sulfentrazone. Each herbicide was individually coated on granular ammonium sulfate (AMS) and top-dressed at 321 kg ha−1 (67 kg N ha−1) onto 5- to 7-leaf cotton. The check plots received the equivalent rate of nonherbicide-treated AMS. Before top-dress, all plots (including the check) were treated with glyphosate and glufosinate to control previously emerged weeds. All herbicides except metribuzin resulted in transient cotton injury. Cotton response to metribuzin varied by year and location. In 2022, metribuzin caused 11% to 39% and 8% to 17% injury at the Clayton and Rocky Mount locations, respectively. In 2023, metribuzin caused 13% to 32% injury at Clayton and 73% to 84% injury at Rocky Mount. Pyroxasulfone (91%), pyroxasulfone + carfentrazone (89%), fomesafen (87%), fluridone (86%), flumioxazin (86%), and atrazine (85%) controlled Palmer amaranth ≥85%. Pendimethalin and fluometuron were the least effective treatments, resulting in 58% and 62% control, respectively. As anticipated, early season metribuzin injury translated into yield loss; plots treated with metribuzin yielded 640 kg ha−1 and were comparable to yields after linuron (790 kg ha−1) was used. These findings suggest that with the exception of metribuzin, residual herbicides coated onto AMS may be suitable and effective in cotton production, providing growers with additional modes of action for late-season control of multiple herbicide–resistant Palmer amaranth.
Background: Liquid biopsy represents a major development in cancer research, with significant translational potential. Similarly, the integration of multiple molecular platforms has yielded novel insights into disease biology and heterogeneity. We hypothesise that applying contemporary multi-omic approaches to liquid biopsies will improve the power of current models. Methods: We have compiled a cohort of 51 patients with glioblastoma, brain metastasis, and primary CNS lymphoma who underwent CSF sampling as part of clinical care. Cell free methylated DNA and shotgun proteomic profiling was obtained from the CSF of each patient and used to build tumour-specific classifiers. Integrated classifiers were compared with single platform classifiers using multiple approaches. Results: In this study, we show that the DNA methylation and protein profiles of cerebrospinal fluid can be combined to fully discriminate lymphomas from their major diagnostic counterparts with perfect AUCs of 1 (95% confidence interval 1-1) and 100% specificity. Each integrated lymphoma classifier significantly outperforms single-platform classifiers, suggesting synergistic biology is obtained using multiple molecular platforms. Conclusions: We present the most specific and accurate CNS lymphoma classifier to date by integrating the methylome and proteome of CSF. This has important implications for the future of cancer diagnostics and generates immediate utility for patients with CNS lymphoma.
Background: Meningiomas have significant heterogeneity between patients, making prognostication challenging. For this study, we prospectively validate the prognostic capabilities of a DNA methylation-based predictor and multiomic molecular groups (MG) of meningiomas. Methods: DNA methylation profiles were generated using the Illumina EPICarray. MG were assigned as previously published. Performance of our methylation-based predictor and MG were compared with WHO grade using generalized boosted regression modeling by generating time-dependent receiver operating characteristic (ROC) curves and computing area under the ROC curves (AUCs) along with their 95% confidence interval using bootstrap resampling. Results: 295 meningiomas treated from 2018-2021 were included. Methylation-defined high-risk meningiomas had significantly poorer PFS and OS compared to low-risk cases (p<0.0001). Methylation risk increased with higher WHO grade and MG. Higher methylome risk (HR 4.89, 95%CI 2.02-11.82) and proliferative MG (HR 4.11, 95%CI 1.29-13.06) were associated with significantly worse PFS independent of WHO grade, extent of resection, and adjuvant RT. Both methylome-risk and MG classification predicted 3- and 5-year PFS and OS more accurately than WHO grade alone (ΔAUC=0.10-0.23). 42 cases were prescribed adjuvant RT prospectively although RT did not significantly improve PFS in high-risk cases (p=0.41). Conclusions: Molecular profiling outperforms conventional WHO grading for prognostication in an independent, prospectively collected cohort of meningiomas.
Despite increasing evidence regarding the efficacy of risk assessment and intervention in community supervision, there continues to be relatively weak adherence to such evidence in its application to decision-making. Regarding assessment, such a lack of fidelity purportedly leads to higher rates of overrides, degraded accuracy, and inefficiency in resource allocation. Regarding intervention, failure to implement evidence-based practice and recognize the role of probation officers as agents of change further diminishes supervision effectiveness. With probation populations at an all-time high and a burgeoning violation rate, this chapter summarizes the emerging probation research to situate the need for clarity of purpose and improved decision-making to maintain public confidence in probation. The chapter presents a decision framework utilizing empirically informed domains as a logic model to ensure probation officer decisions are fair, transparent, and defensible. This framework incorporates both static and dynamic information to ensure decisions are accurate and contextual.
The concept of cognitive reserve (CR) explains why individuals with higher education, intelligence, or occupational attainment exhibit less severe cognitive changes in the presence of age-related or neurodegenerative pathology. CR may be a useful construct in understanding the cognitive performance of patients with late life depression (LLD), a cohort who are twice as likely to later receive a clinical diagnosis of dementia. It follows that depressed older adults with low CR may be at greater risk of cognitive decline compared to non-depressed older adults matched for CR. However, the literature on CR and LLD is limited to cross-sectional studies with mixed findings as to whether proxies of CR moderate cognitive outcomes in LLD. For example, both higher and lower education levels in LLD are associated with greater cognitive impairment in LLD compared to similarly educated non-depressed older adults. Longitudinal studies may help disentangle the association between CR and cognitive outcomes in LLD. The current study investigated the interaction between proxies of CR (e.g., education) and depression status on cognitive functioning over three years. We hypothesized that depressed older adults with low CR would demonstrate greater cognitive impairment and decline compared to depressed elders with high CR and non-depressed older adults with comparable CR.
Participants and Methods:
Older adults with LLD and non-depressed older adults age 59+ participated. All participants were free of dementia at baseline. We divided both patients and non-depressed participants into low (<16) and high (>=16) education groups based upon the median years of education (16) of the entire sample. All participants underwent detailed neuropsychological testing. Composite measures of episodic memory (CERAD Wordlist and recall, LM I and LM II, BVRT), processing speed-executive functioning (SDMT and Trail Making Part B), working memory (forward, reverse, ascending Digit Span), and verbal fluency (Animal Naming and COWA) were calculated based on the non-depressed older adults.
Results:
The baseline sample included 210 non-depressed older adults and 465 older adults with major depression (LLD). 150 non-depressed older adults and 235 LLD patients provided three-year follow-up data. Separate ANOVAs revealed a statistically significant interaction between education and depression status at baseline on the composite score of executive functioning F (1, 668) = 8.74, p <.003. Consistent with our hypothesis, LLD with low education performed significantly worse compared to non-depressed with low education (z-score difference -1.35) and this effect was significantly greater than the difference between LLD patients and non-depressed with high education (z-score difference -0.36). No other interactions were found at baseline. Longitudinal analyses also revealed significant interactions between education and depression on memory over time, although sensitivity analyses did not suggest findings consistent with our hypothesis.
Conclusions:
Cognitive reserve contribute to group differences between LLD and non-depressed older adults in cognitive performance but may not alter cognitive trajectories over time. Future studies should further explore structural and functional brain changes associated with CR in LLD.
Background: Isocitrate dehydrogenase (IDH) mutation status is a key diagnostic and prognostic feature of gliomas. There are conflicting reports regarding the stability of IDH mutations throughout glioma evolution and treatment. Here, we provide an institutional experience of patients with conflicting IDH mutation status longitudinally in order to determine if IDH mutation status changes over time. Methods: We retrospectively identified patients from 2009-2018 with immunohistochemistry (IHC)-recorded IDH mutation status discrepancies longitudinally. Archived frozen tissue samples were analyzed using methylation profiling, Sanger sequencing, and droplet digital PCR (ddPCR). Results were compared to the IHC-reported IDH mutation status. Results: We reviewed 1491 archived glioma samples including 91 patients with multiple tumour samples collected longitudinally. In all instances of IDH mutation discrepancy, we found reasonable explanations through multi-platform profiling that resolved the discrepancies. This included the presence of non-canonical IDH2 mutations identified through Sanger sequencing and perilesional tumour samples or reactive brain tissue identified through methylation profiling. Conclusions: Our findings support the hypothesis that IDH mutations occur early in gliomagenesis and are stable throughout glioma treatment and evolution. Our study highlights the importance of accurate surgical sampling and the role of DNA methylome profiling in diagnostically uncertain cases for integrated pathological and molecular diagnosis.
As a form of land ownership, condominium enables subdivision and produces local government. Designed to facilitate the production of apartments as distinct parcels of land, ownership within condominium now dominates many urban housing markets. In some jurisdictions, including British Columbia, condominium (labelled strata property) may also be deployed to subdivide land for single-house lots within a structure of private local government. The principal effect of extending condominium to unbuilt land is not to enable subdivision, which is something that was already possible and common, but, rather, to endow groups of single-house lot owners with fiscal capacity and governing authority to assume important aspects of local government. Through an analysis of bare land strata property in British Columbia, we reveal how the condominium form, which brought an architecture of ownership and government from the homeowners association of the American suburbs to the North American city, has spread back from the city into the suburban, exurban, and rural, producing a sprawl of ownership within private local government.
Trolley dilemmas were first developed by moral philosophers engaged in reflection on the ethics of permissible harm. But they have since become central to psychological research into morality. One reason why psychologists have paid so much attention to trolley dilemmas is that they see them as a key way to investigating the contrast between deontological and utilitarian approaches to ethics. This framing, however, departs from the original philosophical purpose of trolley dilemmas, and can lead psychological research astray. In this chapter, we question the assumption that trolley dilemmas can shed general light about the psychological bases of utilitarian decision-making. Some lay responses to trolley dilemmas that psychologists routinely classify as "utilitarian" in fact have little meaningful relation to what philosophers mean by this term. Even when what underlies lay responses to trolley dilemmas partly echoes aspects of a utilitarian approach to ethics, this doesn’t generalize to other moral domains, and tells us little about the psychological roots of other aspects of utilitarianism. Properly used, trolley dilemmas have a useful role to play in psychological research. But once we get clear about what we can, and cannot, learn from them, the current centrality of the trolley paradigm in moral psychology will seem overblown.
The giant Cape zebra (Equus capensis) is one of the extinct Quaternary large mammal species of southern Africa, and the largest equid from the Quaternary of Africa. Twenty-six Pleistocene equid tracksites have been identified in aeolianites on the Cape south coast of South Africa. An age range of 161 ± 12 ka to 43 ± 4 ka has been established through Optically Stimulated Luminescence. More than half of the sites contain large-equid tracks, representing the first ichnosites attributed to E. capensis. Smaller equid tracks may have been registered by the quagga (E. quagga quagga). The abundance of E. capensis tracksites on the Cape south coast contrasts with the paucity of body fossils of the species from the region, contrasting with the impression obtained from the body fossil record that E. capensis was predominantly a west coast species in the region. The new data illustrate the capacity of the body fossil and trace fossil records to complement each other. The loss of suitable habitat provided by the Palaeo-Agulhas Plain was probably a contributing factor in the extinction of this large-bodied grazer. A long trackway at Driefontein, attributed to E. capensis, adds to a sparse global record of fossil horse trackways.