Unsupervised machine learning models reveal predictive clinical markers of glioblastoma patient survival using white blood cell counts prior to initiating chemoradiation.

Background: Glioblastoma is a malignant brain tumor requiring careful clinical monitoring even after primary management. Personalized medicine has suggested the use of various molecular biomarkers as predictors of patient prognosis or factors utilized for clinical decision-making. However, the accessibility of such molecular testing poses a constraint for various institutes requiring identification of low-cost predictive biomarkers to ensure equitable care. Methods: We collected retrospective data from patients seen at Ohio State University, University of Mississippi, Barretos Cancer Hospital (Brazil), and FLENI (Argentina) who were managed for glioblastoma-amounting to 581 patient records documented using REDCap. Patients were evaluated using an unsupervised machine learning approach comprised of dimensionality reduction and eigenvector analysis to visualize the inter-relationship of collected clinical features. Results: We discovered that the serum white blood cell (WBC) count of a patient during baseline planning for treatment was predictive of overall survival with an over 6-month median survival difference between the upper and lower quartiles of WBC count. By utilizing an objective PD-L1 immunohistochemistry quantification algorithm, we were further able to identify an increase in PD-L1 expression in glioblastoma patients with high serum WBC counts. Conclusions: These findings suggest that in a subset of glioblastoma patients the incorporation of WBC count and PD-L1 expression in the brain tumor biopsy as simple biomarkers predicting glioblastoma patient survival. Moreover, machine learning models allow the distillation of complex clinical data sets to uncover novel and meaningful clinical relationships.

Clinical and descriptive characteristics associated with high-grade meningioma in a large clinical series.

PURPOSE: We studied 571 patients with intracranial meningioma for clinical characteristics and tumor location associated with high grade meningioma (WHO II/III). MATERIALS AND METHODS: Patients were participants in a multicentre epidemiologic study of risk factors for primary brain tumors including meningioma recruited from September 2005 to November 2019. We included patients 18 or older with a recent diagnosis of a primary intracranial meningioma of any subtype (ICD9/10: 9530-0, 9531-0, 9532-0, 9537-0, 9533-0, 9534-0, 9530-0, 9538-1, 9538-3) who were enrolled at neuro-oncology and neuro-surgery clinics in the southeastern U.S. RESULTS: The median patient age was 58 years (IQR: 48-68) and the majority of patients were female (n = 415; 72.7%) and Caucasian (n = 516; 90.4%). Most patients were symptomatic (n = 460; 80.6%) and their tumours more commonly occurred in a non-skull base location (n = 298; 52.2%). A total of 86 patients (15.0%) had a WHO grade II/III meningioma. Compared to patients with WHO grade I tumours, patients with WHO II/III meningiomas were over 3-times more likely to be male (odds ratio (OR): 3.25; 95% confidence interval (CI): 1.98, 5.35) adjusting for age, race, symptomatic presentation, and skull-based location. Moreover, a WHO grade II/III meningioma was substantially less likely to be observed in asymptomatic patients (OR: 0.15, 95% CI: 0.04, 0.42), and in patients with a skull-based tumour (OR: 0.40, 95% CI: 0.24, 0.66), adjusting for other factors. Male gender, symptomatic tumour, and a non-skull base location were independently associated with WHO grade II/III meningioma. CONCLUSION: These findings may shed additional light on the underlying pathogenesis of meningioma.

Unsupervised machine learning models reveal predictive markers of glioblastoma patient survival using white blood cell counts prior to initiating chemoradiation.

Purpose: Glioblastoma is a malignant brain tumor requiring careful clinical monitoring even after primary management. Personalized medicine has suggested use of various molecular biomarkers as predictors of patient prognosis or factors utilized for clinical decision making. However, the accessibility of such molecular testing poses a constraint for various institutes requiring identification of low-cost predictive biomarkers to ensure equitable care. Methods: We collected retrospective data from patients seen at Ohio State University, University of Mississippi, Barretos Cancer Hospital (Brazil), and FLENI (Argentina) who were managed for glioblastoma-amounting to nearly 600 patient records documented using REDCap. Patients were evaluated using an unsupervised machine learning approach comprised of dimensionality reduction and eigenvector analysis to visualize the inter-relationship of collected clinical features. Results: We discovered that white blood cell count of a patient during baseline planning for treatment was predictive of overall survival with an over 6-month median survival difference between the upper and lower quartiles of white blood cell count. By utilizing an objective PDL-1 immunohistochemistry quantification algorithm, we were further able to identify an increase in PDL-1 expression in glioblastoma patients with high white blood cell counts. Conclusion: These findings suggest that in a subset of glioblastoma patients the incorporation of white blood cell count and PDL-1 expression in the brain tumor biopsy as simple biomarkers predicting glioblastoma patient survival. Moreover, use of machine learning models allows us to visualize complex clinical datasets to uncover novel clinical relationships.

Alliance A071401: Phase II Trial of Focal Adhesion Kinase Inhibition in Meningiomas With Somatic NF2 Mutations.

PURPOSE: Patients with progressive or recurrent meningiomas have limited systemic therapy options. Focal adhesion kinase (FAK) inhibition has a synthetic lethal relationship with NF2 loss. Given the predominance of NF2 mutations in meningiomas, we evaluated the efficacy of GSK2256098, a FAK inhibitor, as part of the first genomically driven phase II study in recurrent or progressive grade 1-3 meningiomas. PATIENTS AND METHODS: Eligible patients whose tumors screened positively for NF2 mutations were treated with GSK2256098, 750 mg orally twice daily, until progressive disease. Efficacy was evaluated using two coprimary end points: progression-free survival at 6 months (PFS6) and response rate by Macdonald criteria, where PFS6 was evaluated separately within grade-based subgroups: grade 1 versus 2/3 meningiomas. Per study design, the FAK inhibitor would be considered promising in this patient population if either end point met the corresponding decision criteria for efficacy. RESULTS: Of 322 patients screened for all mutation cohorts of the study, 36 eligible and evaluable patients with NF2 mutations were enrolled and treated: 12 grade 1 and 24 grade 2/3 patients. Across all grades, one patient had a partial response and 24 had stable disease as their best response to treatment. In grade 1 patients, the observed PFS6 rate was 83% (10/12 patients; 95% CI, 52 to 98). In grade 2/3 patients, the observed PFS6 rate was 33% (8/24 patients; 95% CI, 16 to 55). The study met the PFS6 efficacy end point both for the grade 1 and the grade 2/3 cohorts. Treatment was well tolerated; seven patients had a maximum grade 3 adverse event that was at least possibly related to treatment with no grade 4 or 5 events. CONCLUSION: GSK2256098 was well tolerated and resulted in an improved PFS6 rate in patients with recurrent or progressive NF2-mutated meningiomas, compared with historical controls. The criteria for promising activity were met, and FAK inhibition warrants further evaluation for this patient population.

Parkinson's Syndrome After Cranial Radiotherapy: A Case Report.

Parkinson's syndrome is a group of signs and symptoms where the core issue is bradykinesia and could be a manifestation of idiopathic parkinsonism (Parkinson's disease, PD), secondary parkinsonism, or parkinsonism due to neurodegenerative disease. PD is the most common cause of Parkinson's syndrome, accounting for approximately 80% of cases. The secondary causes of Parkinson's syndrome include tumors, trauma, hydrocephalus, chemotherapy, medications including amphotericin B, metoclopramide, and radiation treatment. Parkinsonian symptoms secondary to radiation treatment are rarely reported in the literature and are usually not alleviated by carbidopa-levodopa. This report describes a 64-year-old man diagnosed with low-grade astrocytoma of the midbrain who developed unilateral parkinsonian symptoms one year after chemoradiation treatment. This case report also sheds further light on the details of reported cases and the treatment options through an extensive literature review. Clinicians need to be aware of patients developing this rare complication following radiation treatment.

Neurochondrin neurological autoimmunity.

OBJECTIVES: To describe the neurologic spectrum and treatment outcomes for neurochondrin-IgG positive cases identified serologically in the Mayo Clinic Neuroimmunology Laboratory. METHODS: Archived serum and CSF specimens previously scored positive for IgGs that stained mouse hippocampal tissue in a nonuniform synaptic pattern by immunofluorescence assay (89 among 616,025 screened, 1993-2019) were reevaluated. Antibody characterization experiments revealed specificity for neurochondrin, confirmed by recombinant protein assays. RESULTS: IgG in serum (9) or CSF (4) from 8 patients yielded identical neuron-restricted CNS patterns, most pronounced in hippocampus (stratum lucidum in particular), cerebellum (Purkinje cells and molecular layer), and amygdala. All were neurochondrin-IgG positive. Five were women; median symptom onset age was 43 years (range, 30-69). Of 7 with clinical data, 6 presented with rapidly progressive cerebellar ataxia, brainstem signs, or both; 1 had isolated unexplained psychosis 1 year prior. Five of 6 had cerebellar signs, 4 with additional brainstem symptoms or signs (eye movement abnormalities, 3; dysphagia, 2; nausea and vomiting, 1). One patient with brainstem signs (vocal cord paralysis and VII nerve palsy) had accompanying myelopathy (longitudinally extensive abnormality on MRI; aquaporin-4-IgG and myelin oligodendrocyte glycoprotein-IgG negative). The 7th patient had small fiber neuropathy only. Just 1 of 7 had contemporaneous cancer (uterine). Six patients with ataxia or brainstem signs received immunotherapy, but just 1 remained ambulatory. At last follow-up, 5 had MRI evidence of severe cerebellar atrophy. CONCLUSION: In our series, neurochondrin autoimmunity was usually accompanied by a nonparaneoplastic rapidly progressive rhombencephalitis with poor neurologic outcomes. Other phenotypes and occasional paraneoplastic causes may occur.

Analysis of Factors Influencing Accuracy of Volume Flow Measurement in Dialysis Access Fistulas Based on Duplex Ultrasound Simulation.

OBJECTIVE: We developed a duplex ultrasound simulator and used it to assess accuracy of volume flow measurements in dialysis access fistula (DAF) models. METHODS: The simulator consists of a mannequin, computer, and mock transducer. Each case is built from a patient's B-mode images that are used to create a 3-dimensional surface model of the DAF. Computational fluid dynamics is used to determine blood flow velocities based on model vessel geometry. The simulator displays real-time B-mode and color-flow images, and Doppler spectral waveforms are generated according to user-defined settings. Accuracy was assessed by scanning each case and measuring volume flow in the inflow artery and outflow vein for comparison with true volume flow values. RESULTS: Four examiners made 96 volume flow measurements on four DAF models. Measured volume flow deviated from the true value by 35 ± 36%. Mean absolute deviation from true volume flow was lower for arteries than veins (22 ± 19%, N = 48 vs. 58 ± 33%, N = 48, p < 0.0001). This finding is attributed to eccentricity of outflow veins which resulted in underestimating true cross-sectional area. Regression analysis indicated that error in measuring cross-sectional area was a predictor of error in volume flow measurement (ß = 0.948, p < 0.001). Volume flow error was reduced from 35 ± 36% to 9 ± 8% (p < 0.000001) by calculating vessel area as an ellipse. CONCLUSIONS: Duplex volume flow measurements are based on a circular vessel shape. DAF inflow arteries are circular, but outflow veins can be elliptical. Simulation-based analysis showed that error in measuring volume flow is mainly due to assumption of a circular vessel.

Clinical trial participation of patients with glioblastoma at The University of Texas MD Anderson Cancer Center.

BACKGROUND: It is estimated only 8-11% of patients with glioblastoma (GBM) enrol in clinical trials, limiting treatment development. We analysed the clinical and demographic features of patients with GBM enroled in clinical trials at the University of Texas MD Anderson Cancer Center (MDACC). METHODS: We reviewed the records of adult patients treated for primary GBM between 2007 and 2012 at the MDACC. A total of 755 patients were identified: 133 were deemed non-eligible, 111 were deemed trial eligible but received standard care and 511 participated in a clinical trial (311 for newly diagnosed glioblastoma [nGBM] and 200 for recurrent glioblastoma [rGBM]). Population characteristics were analysed using descriptive statistics, and survival end-points were evaluated with the Kaplan-Meier method. RESULTS: The median age of clinical trial participants and trial eligible patients was 53.2 years (standard deviation 12.1). Most patients (49.4%) were enroled in a clinical trial protocol for nGBM. The majority of nGBM trial participants were male patients (65.1%), white (86.3%), married (84.4%) and in state (59.9%). Employment status, education, symptoms, tumour location, performance status, extent of resection and treatment facility differed between nGBM trial participants and non-participants. Patients who were eligible but did not enrol tended to be older, have worse performance status and live farther away from the MDACC. CONCLUSION: Numerous disease and demographic barriers exist in trial enrolment in patients with GBM. This study highlights some of these obstacles, which require attention to improve patient enrolment to clinical trials. Patient and physician engagement in novel therapeutic strategies is essential to improving outcomes in this disease.

RBM10 truncation in astroblastoma in a patient with history of mandibular ameloblastoma: A case report.

Astroblastoma is a rare glial neoplasm composed of cells that have broad processes oriented perpendicular to central vessels and often demonstrate vascular sclerosis. The WHO 2016 classification does not specify a grading system for astroblastoma, and categorizes them as well-differentiated or malignant. These broad classification rubrics, however, do not accurately predict clinical outcome. Genetic profiling of astroblastoma has therefore been of particular interest in the recent years. These efforts, although in small number, have revealed heterogeneous molecular findings that may explain astroblastoma's unpredictable clinical outcome. Here, we report a case of recurrent astroblastoma in a 23-year-old female with a unique molecular characteristic. Our patient's tumor harbored an RNA-binding motif 10 (RBM10) truncation. RBM10 codes for a widely expressed RNA binding protein, and its mutation has been described in a variety of solid cancers. RBM10 is thought to be involved in stabilization of pro-apoptotic proteins in breast cancer, and its reduced protein expression is associated with advanced stages of lung adenocarcinoma. To our knowledge, this is the first report of astroblastoma harboring RBM10 truncation. Interestingly, our patient also has a history of mandibular ameloblastoma, but the link between these two rare tumors is unclear.

Evaluation of Examiner Performance Using a Duplex Ultrasound Simulator. Flow Velocity Measurements in Dialysis Access Fistula Models.

We developed a duplex ultrasound simulator for training and assessment of scanning skills. We used the simulator to test examiner performance in the measurement of flow velocities in dialysis access fistulas. Test cases were created from 3-D ultrasound scans of two dialysis access fistulas by reconstructing 3-D blood vessel models and simulating blood flow velocity fields within the lumens. The simulator displays a 2-D B-mode or color Doppler image corresponding to transducer position on a mannequin; a spectral waveform is generated according to Doppler sample volume location and system settings. Examiner performance was assessed by comparing the measured peak systolic velocity (PSV) with the true PSV provided by the computational flow model. The PSV measured by four expert examiners deviated from the true value by 7.8 ± 6.1%. The results indicate the ability of the simulator to objectively assess an examiner's measurement accuracy in complex vascular targets.

Development of a Duplex Ultrasound Simulator and Preliminary Validation of Velocity Measurements in Carotid Artery Models.

OBJECTIVE: Duplex ultrasound scanning with B-mode imaging and both color Doppler and Doppler spectral waveforms is relied upon for diagnosis of vascular pathology and selection of patients for further evaluation and treatment. In most duplex ultrasound applications, classification of disease severity is based primarily on alterations in blood flow velocities, particularly the peak systolic velocity (PSV) obtained from Doppler spectral waveforms. We developed a duplex ultrasound simulator for training and assessment of scanning skills. METHODS: Duplex ultrasound cases were prepared from 2-dimensional (2D) images of normal and stenotic carotid arteries by reconstructing the common carotid, internal carotid, and external carotid arteries in 3 dimensions and computationally simulating blood flow velocity fields within the lumen. The simulator displays a 2D B-mode image corresponding to transducer position on a mannequin, overlaid by color coding of velocity data. A spectral waveform is generated according to examiner-defined settings (depth and size of the Doppler sample volume, beam steering, Doppler beam angle, and pulse repetition frequency or scale). The accuracy of the simulator was assessed by comparing the PSV measured from the spectral waveforms with the true PSV which was derived from the computational flow model based on the size and location of the sample volume within the artery. RESULTS: Three expert examiners made a total of 36 carotid artery PSV measurements based on the simulated cases. The PSV measured by the examiners deviated from true PSV by 8% ± 5% (N = 36). The deviation in PSV did not differ significantly between artery segments, normal and stenotic arteries, or examiners. CONCLUSION: To our knowledge, this is the first simulation of duplex ultrasound that can create and display real-time color Doppler images and Doppler spectral waveforms. The results demonstrate that an examiner can measure PSV from the spectral waveforms using the settings on the simulator with a mean absolute error in the velocity measurement of less than 10%. With the addition of cases with a range of pathologies, this duplex ultrasound simulator will be a useful tool for training health-care providers in vascular ultrasound applications and for assessing their skills in an objective and quantitative manner.

Clinical discussion of the management of anaplastic oligodendroglioma/oligoastrocytoma (both codeleted and nondeleted).

Anaplastic oligodendroglioma (AO) and anaplastic oligoastrocytoma (AOA) are uncommon malignant tumors occurring in adults, but have garnered attention because of a high rate of response to chemotherapy in early studies. However, no clinical trial had demonstrated benefit with the addition of chemotherapy to radiotherapy alone until the long-term results of RTOG 9402 and EORTC 26951. These studies revealed prolonged survival in patients with anaplastic gliomas harboring the 1p/19q codeletion when treated with PCV (procarbazine, lomustine, and vincristine) and radiation therapy compared with radiation alone. These studies validated the use of 1p/19q codeletion status as a predictive biomarker in these tumors. Additional molecular characterization of these tumors may provide additional insight into treatment decisions, although these characterizations have yet to be fully elucidated. Even with the strength of the data advocating the use of combination therapy (PCV and radiotherapy), the incorporation of newer, less-toxic drugs such as temozolomide into many practices in the past decade raises important questions regarding the optimal chemotherapy regimen. Unfortunately, additional definitive phase III trials will take several years to answer remaining questions. Regardless, it is clear that patients with 1p/19q codeleted AO or AOA who can tolerate chemotherapy should not receive radiotherapy alone.

Pregnancy and glial brain tumors.

BACKGROUND: Improvements in brain tumor treatments have led to an increase in the number of young women with brain tumors who are now considering pregnancy. The aim of this study is to evaluate the influence of pregnancy on brain tumor biology. METHODS: In this institutional review board-approved retrospective study, we searched the institution's database for patients with glial brain tumors who were pregnant at the time of diagnosis or became pregnant during the course of their illness. We identified 34 such patients and reviewed their charts to determine each patient's clinical course and pregnancy outcome. RESULTS: Fifteen patients were diagnosed with a primary brain tumor during pregnancy: 3 with glioblastomas, 6 with grade III gliomas, and 6 with grade II gliomas. Pregnancy was terminated in only 2 of these patients, and the remainder delivered healthy babies. Twenty-three patients became pregnant after diagnosis (4 patients were pregnant at diagnosis and again after diagnosis). Of the patients who became pregnant after diagnosis, the 5 with grade I tumors had stable disease during and after pregnancy. However, of the 18 patients with grade II or III gliomas, 8 (44%) had confirmed tumor progression during pregnancy or within 8 weeks of delivery. CONCLUSIONS: In contrast to grade I gliomas, the tumor biology of grades II and III gliomas may be altered during pregnancy, leading to an increased risk of tumor progression. These findings support the need for increased tumor surveillance and patient counseling and for additional data collection to further refine these results.

Implications of bevacizumab discontinuation in adults with recurrent glioblastoma.

BACKGROUND: Patients with recurrent glioblastoma benefiting from bevacizumab are often treated indefinitely due to concerns regarding rebound tumor recurrence upon discontinuation. However, treatment is discontinued for reasons other than disease progression in a subset of these patients, the characteristics and outcomes of which are poorly defined. METHODS: Of 342 adults with recurrent glioblastoma in our database treated with bevacizumab, 82 received treatment for ≥ 6 months; of these, bevacizumab was discontinued for reasons other than tumor progression in 18 patients (Bev-D) and for disease progression in the remainder (Bev-S). The impact of discontinuation on outcome was assessed with discontinuation as a time-dependent covariate in a Cox hazards model for progression-free survival. RESULTS: There was no difference in hazard rates for progression between Bev-D and Bev-S groups; the adjusted hazard ratio for progression using discontinuation as a time-dependent covariate was 0.91 (95% CI:0.47, 1.78). The median PFS after bevacizumab-discontinuation was 27 weeks (95% CI:15-NR). At progression, a higher proportion of Bev-D patients had local progression compared with the Bev-S patients. Salvage therapy in Bev-D patients yielded a PFS-26 weeks of 47% (95% CI:23%-94%) with a median PFS of 23 weeks (95% CI:12-NR), vs. 5% (95% CI: 1%-21%) and 9 weeks (95% CI: 6-11) in Bev-S patients (HR:0.3;CI, 0.1-0.6) (P = .0007). CONCLUSIONS: Bevacizumab discontinuation unrelated to disease progression does not appear to cause rebound recurrence or worsen PFS in patients who benefit from bevacizumab. Additionally, Bev-D patients had an improved response to salvage therapy, findings which provide a strong basis for a prospective study.

Climate, social factors and research disturbance influence population dynamics in a declining sociable weaver metapopulation.

Population trends are determined by gains through reproduction and immigration, and losses through mortality and emigration. These demographic quantities and resulting population dynamics are affected by different external and internal drivers. We examined how these demographic quantities were affected by weather, research-induced disturbance, local density, colony site and year in a metapopulation of 17 sociable weaver (Philetairus socius) colonies over 17 years of study (4 years for reproduction). Most colonies declined, but at different rates. The four demographic quantities were related to different drivers. Survival strongly varied among years and colonies and was positively related to rainfall and negatively related to extreme temperature (together explaining 30% of variation) and disturbance (measured as number of captures conducted at a colony; 7%). There was a trend for a positive relationship between reproduction and rainfall (50%). Movement was mainly related to local density: individuals were more likely to emigrate from small to large colonies and from colonies that were either well below or above their long-term mean. They were more likely to immigrate into colonies that were nearby, and below their mean size. We then quantified the effects of these relationships on metapopulation dynamics using a multi-site matrix projection model. Rainfall was potentially a strong driver of metapopulation dynamics. In addition, field-work disturbance might have contributed to the decline of this metapopulation but could not explain its full magnitude. Hence, through a combination of analytical methods we were able to obtain information on the main drivers affecting dynamics in a declining metapopulation.

Treatment recommendations for anaplastic oligodendrogliomas that are codeleted.

Abstract: Anaplastic oligodendroglioma (AO) is a rare malignant tumor occurring in adults. Despite early indications of chemosensitivity, no clinical trial had demonstrated a benefit of chemotherapy beyond that of radiotherapy alone. Now, however, the Radiation Therapy Oncology Group (RTOG) 9402 and the European Organisation for Research and Treatment of Cancer (EORTC) 26951 studies investigating PCV (procarbazine [Matulane], lomustine [CeeNU], and vincristine) and radiation therapy vs radiation alone both show improved outcomes in patients with the 1 p/19q codeletion who received PCV and radiation therapy. These differences were detected with additional follow-up after publication of the initial results in 2006, when no differences in survival were detected. The two studies have also validated the use of the 1p/19q codeletion as a predictive biomarker in AO. Many will debate the wisdom of adopting PCV therapy as standard of care because of the greater toxicity of PCV compared with temozolomide (Temodar). Nonetheless, although important questions still remain regarding chemotherapy choice, sequence, and dosing, the answers to which will require additional large phase III trials, radiotherapy alone is no longer appropriate therapy for 1p/19q codeleted AOs.

Avian poxvirus epizootic in a breeding population of Lesser Flamingos (Phoenicopterus minor) at Kamfers Dam, Kimberley, South Africa.

Avian pox has a worldwide distribution, but prior to this investigation has not been reported in free-ranging flamingo populations. During observations of the first successful breeding of Lesser Flamingos on a purpose-built island, at Kamfers Dam near Kimberley, South Africa, multiple small, raised, crusted plaques on the legs and facial area were noticed on 30% of the fledgling flamingos. A diagnosis of avipoxvirus infection was made on the basis of the macroscopic, histologic, and electron microscopic features, and was further confirmed by DNA sequence analysis. The avipoxvirus detected was very similar to that previously detected in albatross and falcons.

The risk of flawed inference in evolutionary studies when detectability is less than one.

Addressing evolutionary questions in the wild remains a challenge. It is best done by monitoring organisms from birth to death, which is very difficult in part because individuals may or may not be resighted or recaptured. Although the issue of uncertain detection has long been acknowledged in ecology and conservation biology, in evolutionary studies of wild populations it is often assumed that detectability is perfect. We argue that this assumption may lead to flawed inference. We demonstrate that the form of natural selection acting on body mass of sociable weavers is altered and that the rate of senescence of roe deer is underestimated when not accounting for a value of detectability that is less than one. Because mark-recapture models provide an explicit way to integrate and reliably model the detection process, we strongly recommend their use to address questions in evolutionary biology.

Nonparametric estimation of natural selection on a quantitative trait using mark-recapture data.

Assessing natural selection on a phenotypic trait in wild populations is of primary importance for evolutionary ecologists. To cope with the imperfect detection of individuals inherent to monitoring in the wild, we develop a nonparametric method for evaluating the form of natural selection on a quantitative trait using mark-recapture data. Our approach uses penalized splines to achieve flexibility in exploring the form of natural selection by avoiding the need to specify an a priori parametric function. If needed, it can help in suggesting a new parametric model. We employ Markov chain Monte Carlo sampling in a Bayesian framework to estimate model parameters. We illustrate our approach using data for a wild population of sociable weavers (Philetairus socius) to investigate survival in relation to body mass. In agreement with previous parametric analyses, we found that lighter individuals showed a reduction in survival. However, the survival function was not symmetric, indicating that body mass might not be under stabilizing selection as suggested previously.