OBJECTIVE:Interictal epileptiform discharges (IEDs) are transients observed on the electroencephalogram (EEG) of patients with epilepsy. IEDs have traditionally been recorded from scalp or intracranial EEG macrocontacts, which coarsely sample neural activity. Here, we investigated the use of flexible, high-resolution microelectrocorticographic (μECoG) arrays for measuring IEDs with greater spatiotemporal precision to test whether there exist microscale patterns of IED activity that may be missed on standard intracranial EEG. METHODS:We used liquid crystal polymer thin-film μECoG arrays with both high resolution (.76-1.72-mm spacing, 200-μm diameter) and large cortical coverage (144-1596 mm2) to record from seven patients undergoing surgical treatment of epilepsy. We identified IEDs by a combination of expert review and automated detection. We quantified the spatial extent of IEDs, mapped patterns of repeated IED activity, and quantified IED propagation direction using multilinear fit models. We also compared IED detection rates and propagation measurements between μECoG arrays and simulated macroarrays (10-mm spacing, 2.3-mm diameter). RESULTS:We demonstrated successful use of μECoG arrays to map intraoperative microscale patterns of IEDs. The majority of patients (5/7) exhibited elevated IED activity that was highly localized (subcentimeter localization). Across all patients, 40% of detected IEDs were observed within a 4-mm radius of cortex. μECoG arrays also mapped the direction of IED propagation. An average of 39% (range = 4.2%-96.5%, SD = ±36.8%) of the IED events captured by the μECoG arrays were not detectable by simulated macrocontacts. SIGNIFICANCE:These intraoperative data demonstrate that μECoG arrays can map the microscale spatiotemporal activity of IEDs. These patterns of IEDs may be poorly captured by standard, macroscale recording devices. Our findings support the use of high-resolution, large area coverage μECoG arrays for the presurgical and intraoperative mapping of epileptic cortex.

Although ctDNA-based liquid biopsies show promise for multi-cancer early detection (MCED), current performance evaluations using aggregate metrics across cancers mask biological heterogeneity and cancer-specific test performance, limiting clinical validity and utility. To better elucidate the clinical validity of MCED testing, we assessed the detection performance of a novel sequential reflex ctDNA-based test in a case-control study, focusing on cancer-specific intrinsic accuracy and incidence-adjusted PPV of each tissue of origin (TOO). Peripheral blood samples (NCT05435066) from treatment-naïve cancer patients (N=1039) and individuals with no reported cancer (N=1,000) were analyzed using a sequential reflex test by Harbinger Health. The Primary Test profiles ctDNA methylation for cancer signal detection followed by a Reflex Test analyzing a broader set of biomarkers for cancer signal confirmation and TOO localization. Test performance was established using 10-fold cross-validation with patient-level splits. Cancer incidence for age>50 was estimated at 1.32% based on SEER 2022 data. Prospective PPV was calculated using cancer-specific intrinsic accuracy estimated from the case-control study in combination with SEER incidence values. In this cohort (mean age 57.6 ± 14.3 years; 61.5% female; 71.6% White; 17.4% Black or African American), sequential reflex testing achieved 98.7% specificity. In a representative selection of common or high-mortality cancers lacking organized screening, TOO-specific prospective PPVs were 94% for upper gastrointestinal (UGI), 85% for colorectal, 80% for hepatobiliary (HB: liver, biliary duct), 39% for lung, 38% for pancreaticobiliary (PB: pancreas, gallbladder), and 34% for head & neck (H&N). Corresponding odds of correct case-type were 16:1 (UGI), 6:1 (colorectal), 4:1 (HB), 2:3 (lung), 3:5 (PB), and 1:2 (H&N). Early-stage (I-II) intrinsic accuracy (sensitivity) was 70% for H&N, 51% for HB, and 40% for PB, with lower sensitivities for lung and GI cancers. In a modeled cohort of 100,000 individuals receiving the test, 107 pancreaticobiliary cases are expected (38 early stage); the test identifies 67 of the 107 (15 early stage) with odds in favor of correct classification of 3:5. This analytical framework for MCED testing quantifies misclassification proportions corresponding to each TOO readout, informing diagnostic pathways in healthcare systems. Our approach can facilitate MCED test optimization, standardize cross-platform comparisons, and clarify risk-benefit implications of follow-up; potentially expediting diagnosis, reducing costs, and improving equity in access. The reported sequential reflex test demonstrates high performance for early cancer detection and localization, warranting prospective validation in diverse populations. Elie Massaad, A. Gregory DiRienzo, Johanna B. Withers, Dorna Kashef, Jocelyn Charlton, James X. Sun, Franziska Michor, Kieran Chacko, Hutan Ashrafian. Novel performance quantification of MCED testing to aid clinical decisions: Analysis of a sequential reflex blood-based methylated ctDNA test [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB252.

OBJECTIVE Maximal safe resection is the goal of surgical treatment for high-grade glioma (HGG). Deep-seated hemispheric gliomas present a surgical challenge due to safety concerns and previously were often considered inoperable. The authors hypothesized that use of tubular retractors would allow resection of deep-seated gliomas with an acceptable safety profile. The purpose of this study was to describe surgical outcomes and survival data after resection of deep-seated HGG with stereotactically placed tubular retractors, as well as to discuss the technical advances that enable such procedures. METHODS This is a retrospective review of 20 consecutive patients who underwent 22 resections of deep-seated hemispheric HGG with the Viewsite Brain Access System by a single surgeon. Patient demographics, survival, tumor characteristics, extent of resection (EOR), and neurological outcomes were recorded. Cannulation trajectories and planned resection volumes depended on the relative location of white matter tracts extracted from diffusion tractography. The surgical plans were designed on the Brainlab system and preoperatively visualized on the Surgical Theater virtual reality SNAP platform. Volumetric assessment of EOR was obtained on the Brainlab platform and confirmed by a board-certified neuroradiologist. RESULTS Twenty adult patients (18 with IDH-wild-type glioblastomas and 2 with IDH-mutant grade IV astrocytomas) and 22 surgeries were included in the study. The cohort included both newly diagnosed (n = 17; 77%) and recurrent (n = 5; 23%) tumors. Most tumors (64%) abutted the ventricular system. The average preoperative and postoperative tumor volumes measured 33.1 ± 5.3 cm3 and 15.2 ± 5.1 cm3, respectively. The median EOR was 93%. Surgical complications included 2 patients (10%) who developed entrapment of the temporal horn, necessitating placement of a ventriculoperitoneal shunt; 1 patient (5%) who suffered a wound infection and pulmonary embolus; and 1 patient (5%) who developed pneumonia. In 2 cases (9%) patients developed new permanent visual field deficits, and in 5 cases (23%) patients experienced worsening of preoperative deficits. Preoperative neurological or cognitive deficits remained the same in 9 cases (41%) and improved in 7 (32%). The median overall survival was 14.4 months in all patients (n = 20) and in the newly diagnosed IDH-wild-type glioblastoma group (n = 16). CONCLUSIONS Deep-seated HGGs, which are surgically challenging and frequently considered inoperable, are amenable to resection through tubular retractors, with an acceptable safety profile. Such cytoreductive surgery may allow these patients to experience an overall survival comparable to those with more superficial tumors.

OBJECTIVE:Effective surgical treatment of drug-resistant epilepsy depends on accurate localization of the epileptogenic zone (EZ). High-frequency oscillations (HFOs) are potential biomarkers of the EZ. Previous research has shown that HFOs often occur within submillimeter areas of brain tissue and that the coarse spatial sampling of clinical intracranial electrode arrays may limit the accurate capture of HFO activity. In this study, we sought to characterize microscale HFO activity captured on thin, flexible microelectrocorticographic (μECoG) arrays, which provide high spatial resolution over large cortical surface areas.METHODS:We used novel liquid crystal polymer thin-film μECoG arrays (.76-1.72-mm intercontact spacing) to capture HFOs in eight intraoperative recordings from seven patients with epilepsy. We identified ripple (80-250 Hz) and fast ripple (250-600 Hz) HFOs using a common energy thresholding detection algorithm along with two stages of artifact rejection. We visualized microscale subregions of HFO activity using spatial maps of HFO rate, signal-to-noise ratio, and mean peak frequency. We quantified the spatial extent of HFO events by measuring covariance between detected HFOs and surrounding activity. We also compared HFO detection rates on microcontacts to simulated macrocontacts by spatially averaging data.RESULTS:We found visually delineable subregions of elevated HFO activity within each μECoG recording. Forty-seven percent of HFOs occurred on single 200-μm-diameter recording contacts, with minimal high-frequency activity on surrounding contacts. Other HFO events occurred across multiple contacts simultaneously, with covarying activity most often limited to a .95-mm radius. Through spatial averaging, we estimated that macrocontacts with 2-3-mm diameter would only capture 44% of the HFOs detected in our μECoG recordings.SIGNIFICANCE:These results demonstrate that thin-film microcontact surface arrays with both highresolution and large coverage accurately capture microscale HFO activity and may improve the utility of HFOs to localize the EZ for treatment of drug-resistant epilepsy.

Abstract Background Early detection of cancer has significant potential to impact human health and society by decreasing cancer-related morbidity and mortality. While previous approaches to identify cancer-informative biomarkers are predominantly statistical, Harbinger Health has utilized foundational discoveries from developmental biology to design a targeted methylation assay for early cancer detection from cell-free DNA (cfDNA) extracted from plasma. Utilizing this biologically informed approach, we developed a fixed multi-layered logistic regression-based machine learning algorithm, trained with an in-house generated dataset of 1046 samples (621 cancer, 425 non-cancer) that predicts a binary classification (yes/no) for cfDNA samples processed through our assay. We have previously reported high sensitivity for multi-cancer detection, including for early-stage disease. Methods Here, we perform a comprehensive independent analytical validation of our assay and algorithm, encompassing 69 subjects: 19 with newly diagnosed treatment-naïve cancer (8 different cancer types) and 50 individuals with no history, diagnosis, or cancer symptoms. In total, we utilized 122 replicate samples to assess reproducibility and precision, 8 non-template controls (water) to determine limit of blank (LOB), and cohorts of matched biopsy and cfDNA to determine tumor content limit of detection (LOD). Results Precision was assessed within five different sub-studies, by comparing concordance of predicted binary cancer classification between replicate samples, giving results of 0.90 (0.95 CI: 0.764–0.959) for inter-run precision, 1.00 (0.95 CI: 0.796–1.000) for intra-run precision, 1.00 (0.95 CI: 0.871–1.000) for inter-operator precision, 0.96 (0.95 CI: 0.930–0.998) for inter-instrument precision and 0.83 (0.95 CI: 0.641–0.933) for inter-day precision. To determine LOB, we carried 8 non-template controls (water) through the entire assay and detected on average ∼0.02% unique aligned reads of a true sample on the same sequencing run. Finally, to assess tumor content LOD, we developed methodology that uses methylation signal to estimate the amount of tumor-derived DNA in each cfDNA sample and validated our estimates using whole exome sequencing, an orthogonal gold-standard approach. We then assessed the relationship between tumor content and classifier sensitivity using our training data of 625 cancer cfDNA samples and determined that our tumor content LOD whereby 95% true cancer samples were correctly predicted to be 0.037%. Conclusion Our assay shows high performance and high technical reproducibility. Our previously reported high sensitivity for stage 1 and stage 2 cancers, as well as extremely low tumor content LOD reported here supports our ability to perform early-stage multi-cancer detection, where the levels of circulating tumor DNA are low.

e15035 Background: Early cancer detection has the potential to significantly improve patient outcomes and reach the Cancer Moonshot goals of reducing the death rate from cancer by at least 50 percent over the next 25 years. Harbinger Health is pioneering early cancer detection with a blood-based test that combines recent genomic and epigenomic discoveries of early cancer and biology-informed artificial intelligence. Unlike prior approaches that are purely statistical to identify informative biomarkers, Harbinger’s approach is informed by insights into specific biological events early during tumorigenesis and is therefore optimized for detecting cancer in patients with very low levels of circulating tumor DNA. Methods: Here, we present data generated using the Harbinger Health assay from 1,046 subjects, 621 with newly diagnosed, treatment-naive cancer (15 different cancer types) and 425 individuals with no history, diagnosis, or cancer symptoms. Using this sample cohort, we developed and applied a rigorous framework for training, calibration and predicting likelihood of cancer using a multi-layered logistic regression-based machine learning algorithm, generating final outputs of binary classification (cancer yes/no). Our framework involved multiple iterations of 10-fold cross-validation of the full dataset, and we report solely on samples that appear in the held-out test set in each iteration. Additionally, we developed a similar framework to predict tissue of origin (TOO) for cancer samples. Results: The overall sensitivity of cancer detection was 82% (95% confidence interval (CI): 72.7-91.0%) at 95% specificity. Notably, the sensitivity was 74% (95% CI: 54.8-92.7%) for stage 1 and 84% (95% CI: 65.3-100%) for stage 2. Furthermore, we were able to correctly predict cancer in 95% of patient samples with at least 0.037% tumor fraction. The overall sensitivity for high incident cancers were: breast (73%), prostate (82%), lung (85%), and colorectal (96%) at 95% specificity and the overall accuracy of TOO prediction was 86% when the tumor fraction was greater than 0.1% in the top 3 most prevalent cancer types (breast, colorectal and lung). Importantly, we noted that technical variability was introduced when performing assay optimization strategies, yet we observed comparable performance when assessing subsets of stably processed samples, indicating that our performance is robust. Conclusions: Overall, these results demonstrate that the Harbinger Health platform, with its biology-informed approach, has the potential for highly sensitive multi-cancer diagnostic accuracy and specifically those with early-stage cancer. The platform is now being validated in a 10,000-subject prospective clinical trial (CORE-HH/NCT05435066).

Abstract Background Multi-cancer early detection products are poised to change cancer treatment and survival. However, current applications lack the ability to translate analytical findings to clinical outcomes. Quantitatively tracking cancer signals is the next frontier for early cancer detection and data-driven intervention. Harbinger Health has pioneered a platform that combines novel epigenomic insights with artificial intelligence to detect cancer and monitor signal over time to track disease progression and/or response to treatment. Methods We designed a comprehensive enrichment panel (17.8 Mb) which provides simultaneous methylation and mutational state for cancer and tissue-of-origin informative regions as well as a panel of tumor suppressor genes and oncogenic drivers. We also defined the most significant methylation motifs (MMs) that contribute to pan-cancer signal and developed an algorithm to estimate the amount of tumor-derived reads in a cell-free DNA (cfDNA) sample. To verify our methodology, we performed whole exome sequencing (WES) for a cohort of 46 patients with matched FFPE and cfDNA and compared variant allele frequency of somatic mutations to tumor content estimates from MMs. We used this tumor content (%TC) estimate to quantify signal over time and monitor dynamic changes in the tumor. Simultaneously, we tracked orthogonal mutations across treatment. To clinically assess our quantitative analysis, we sourced longitudinal samples of 21 cancer patients over the course of treatment with four timepoints, averaging 38 days apart. Timepoint 0 (TP0) was before any treatment, while TP1, TP2, and TP3 were taken before subsequent treatment. For most timepoints, we had endpoint RECIST data which identified n = 8 samples as progressive disease (PD), n = 1 as stable disease (SD), n = 9 as partial response (PR), and n = 3 as complete response (CR). We also sourced 21 non-cancer subjects with two timepoints, 60 days apart. All samples across these timepoints were captured with the 17.8 Mb panel and analysis was done to assess longitudinal tracking—predict prognosis, identify minimal residual disease (MRD), and/or detect disease progression. Results We confirmed that our %TC estimates derived using MMs were highly correlated to estimates derived using WES. We therefore utilized our methylation data to estimate %TC for all timepoints across treatment. The %TC ranged from 0.02% to 60%, with Stage IV disease showing the greatest tumor burden, as expected. Notably, %TC showed dynamic fold-changes during treatment, with differential levels between response groups and non-cancers. In a small subset that included three Stage IV prostate cancer patients with high initial %TC, our biomarkers were able to identify one patient that was reported PR but showed an increase of 18.9% tumor content, indicating disease progression. This patient was deceased at last follow-up. [Additional data to be reported.] Conclusion Our analytical approach of identifying %TC based on Harbinger Health’s proprietary enrichment panel was successful in quantitatively monitoring cancer patients through treatment and disease progression. Our data can provide valuable insights to help clinicians make more informed decisions (e.g. increased monitoring or interventions) about patient care. Further development of techniques enriching specifically for these MMs are in progress.

One-third of epilepsy patients suffer from medication-resistant seizures. While surgery to remove epileptogenic tissue helps some patients, 30-70% of patients continue to experience seizures following resection. Surgical outcomes may be improved with more accurate localization of epileptogenic tissue. We have previously developed novel thin-film, subdural electrode arrays with hundreds of microelectrodes over a 100-1,000 mm2 area to enable high-resolution mapping of neural activity. Here we used these high-density arrays to study microscale properties of human epileptiform activity. We performed intraoperative micro-electrocorticographic recordings within epileptic cortex (the site of seizure onset and early spread) in nine patients with epilepsy. In two of these patients, we obtained recordings from cortical areas distal to the epileptic cortex. Additionally, we recorded from two non-epileptic patients with movement disorders undergoing deep brain stimulator implantation as non-epileptic tissue controls. A board-certified epileptologist identified microseizures, which resembled electrographic seizures normally observed with clinical macroelectrodes. Epileptic cortex exhibited a significantly higher microseizure rate (2.01 events/min) than non-epileptic cortex (0.01 events/min; permutation test, P=0.0068). Using spatial averaging to simulate recordings from larger electrode contacts, we found that the number of detected microseizures decreased rapidly with increasing contact diameter and decreasing contact density. In cases in which microseizures were spatially distributed across multiple channels, the approximate onset region was identified. Our results suggest that micro-electrocorticographic electrode arrays with a high density of contacts and large coverage are essential for capturing microseizures in epilepsy patients and may be beneficial for localizing epileptogenic tissue to plan surgery or target brain stimulation.

BACKGROUND:Glioma is a family of primary brain malignancies with limited treatment options and in need of novel therapies. We previously demonstrated that the adhesion G protein-coupled receptor GPR133 (ADGRD1) is necessary for tumor growth in adult glioblastoma, the most advanced malignancy within the glioma family. However, the expression pattern of GPR133 in other types of adult glioma is unknown.METHODS:We used immunohistochemistry in tumor specimens and non-neoplastic cadaveric brain tissue to profile GPR133 expression in adult gliomas.RESULTS:We show that GPR133 expression increases as a function of WHO grade and peaks in glioblastoma, where all tumors ubiquitously express it. Importantly, GPR133 is expressed within the tumor bulk, as well as in the brain-infiltrating tumor margin. Furthermore, GPR133 is expressed in both isocitrate dehydrogenase (IDH) wild-type and mutant gliomas, albeit at higher levels in IDH wild-type tumors.CONCLUSION:The fact that GPR133 is absent from non-neoplastic brain tissue but de novo expressed in glioma suggests that it may be exploited therapeutically.

Midface advancement by distraction osteogenesis (DO) is commonly performed in patients with craniosynostosis for indications including midface hypoplasia, exorbitism, obstructive sleep apnea, class III malocclusion, and overall aesthetic facial deficiency. There is evidence to suggest that maxillary LeFort I advancement increases the risk of velopharyngeal dysfunction in the cleft palate population, yet few studies have investigated changes in speech following LeFort III or monobloc midface advancement in patients with syndromic craniosynostosis. The purpose of this study was to examine the effect of midface DO on speech as indicated by the Pittsburgh Weighted Speech Score in patients with Apert, Crouzon, and Pfeiffer Syndrome. Among 73 midface advancement cases performed during the study period, 19 cases met inclusion criteria. Overall, the highest post-advancement Pittsburgh Weighted Speech Score (PWSS) was significantly higher than the pre-advancement PWSS (0.52 versus 2.42, P = 0.01), indicating an acute worsening of VPI post-advancement. Specifically, the PWSS components nasal emission and nasality were significantly higher post-advancement than pre-advancement (nasal emission: 1.16 versus 0.21, P = 0.02) (nasality: 0.68 versus 0.05, P = 0.04). However, there was no significant difference between pre-advancement PWSS and the latest post-advancement PWSS (P = 0.31). Midface distraction is associated with an acute worsening of VPI post-operatively that is followed by improvement, and often resolution over time. Future work with additional patient accrual is needed to determine the effect of different advancement procedures and syndromes on VPI rates and profundity.

Abstract A key goal in immuno-oncology is the identification of tumor antigens recognized by T cells. Significant progress has been made in predicting MHC class I presentation of tumor-specific antigens (peptides) recognized by CD8 reactive T cells. However, predicting antigens recognized by CD4 T cells that are presented by the MHC class II has proven to be more challenging. Studies have shown binding affinity to be less predictive for MHC class II presentation than for class I. Class II antigen-directed therapeutics require accurate antigen identification from patient samples, which remains elusive today. Methods: We focused initially on antigen presentation by the HLA-DR and generated a dataset of human tumor transcriptomes and HLA-DR immunopeptidomes from resections of B cell lymphomas (N=39). Transcriptomes were obtained by NGS of exome-captured cDNA and immunopeptidomes by immunoprecipitation using the HLA-DR specific Ab L243 and MS/MS. Each sample was typed for HLA-DRB1,3,4,5 using standard methods. Additionally, we obtained published class II mass spectrometry data for two B cell lines, each of which expressed a single common HLA class II allele (HLA-DRB1*15:01 and HLA-DRB5*01:01). RNA sequencing data was not available for either cell line; therefore, we substituted RNA-sequencing data from a different B cell line, B721.221. We combined these data to train a deep learning model of Class II HLA peptide presentation. Our model addressed two key challenges: (1) learning HLA-allele-specific models from tumor and normal data where each sample expressed up to 4 unique HLA-DR alleles and (2) reflecting information about all aspects of HLA presentation, including gene expression, antigen processing and stable binding of peptides to HLA. We evaluated the performance of the model on two independent test datasets. First, we tested the model on HLA peptides from a held-out sample from the lymphoma training dataset. Second, we tested the model on a separate public cell line expressing both HLA-DRB1*15:01 and HLA-DRB5*01:01. Results: An average of 567 training and 203 testing peptides at q<0.01 and >50M unique transcriptome reads were obtained from the lymphoma samples. From cell-lines, 433 peptides for training and 223 peptides for testing were used. The model demonstrated a significant improvement in prediction accuracy, achieving >10%-point gain in area under the ROC curve (ROC AUC) vs a standard binding affinity-based predictor. On the held-out lymphoma test data, it achieved a ROC AUC of .95 vs the binding affinity model ROC AUC .80. On the cell-line test data, it achieved a ROC AUC of .90 vs the binding affinity model ROC AUC of .79. Conclusion: We used a large dataset of transcriptomes and HLA peptidomes to train a deep learning model for HLA class II antigen presentation. The new model significantly outperforms standard methods and advances in silico HLA class II antigen selection for personalized cancer immunotherapy. Citation Format: Tommy Boucher, Matthew Davis, Christine Palmer, Tyler Murphy, Andrew Clark, Fujiko Duke, Aaron Yang, Lauren Young, Karin Jooss, Mojca Skoberne, Josh Francis, Roman Yelensky, James Sun, Jennifer Busby. MHC class II antigen identification for cancer immunotherapy by deep learning on tumor HLA peptides [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4445.

Background: Due to the compelling predictive value of companion diagnostic (CDx) biomarkers tied to targeted and immune-based therapies, well-characterized robust analytic and clinical validation of genomic assays has become mandatory. An NGS-based CGP (comprehensive genomic profiling) platform was developed in compliance with FDA guidelines for CDx indications. Methods: DNA extracted from FFPE tumor tissue underwent whole-genome shotgun library construction and hybridization-based capture, followed by sequencing using Illumina HiSeq 4000. Sequence data were processed using a proprietary analysis pipeline designed to identify sub substitutions, indels, copy number alterations, genomic rearrangements, microsatellite instability (MSI), and tumor mutational burden (TMB) in 324 genes. Results: Clinical validity was demonstrated by establishing statistical non-inferiority between CGP and the respective approved CDx, e.g. cobas EGFR and BRAF mutational testing, ALK rearrangements with FISH and IHC, ERBB2 amplification with FISH, and others. For analytical validity, concordance with an orthogonal NGS platform was 94.6% for substitutions and indels, and within-assay reproducibility had positive percent agreement (PPA) of 99.4%. TMB was analytically validated via concordance with whole-exome sequencing. For the first 616 patients (25% non-small cell lung cancer) assayed in clinical care, 6.8% of cases had TMB exceeding 20 mut/Mb, with 25% of these also harboring microsatellite instability. For 143 NSCLC cases, >50% harbored 10 mut/Mb. Of 354 cases with CDx findings possible, 25.6% had such findings, which were split nearly evenly between indications to benefit from and contraindications to targeted therapies. Conclusions: We developed a CGP assay and demonstrated clinical and analytical validity for CDx biomarkers for targeted therapy, with clinical validation for TMB in progress via correlation with prospective immunotherapy trials. Initial oncologist feedback indicates impact of assay results on course of treatment decisions in patient care. Legal entity responsible for the study: Foundation Medicine, Inc. Funding: Foundation Medicine, Inc. Disclosure: Y. Li: Employee of and stockholder: Foundation Medicine Inc. J.X. Sun: Stockholder: Foundation Medicine Inc. J. Skoletsky, C. Milbury, C. Burns, W-K. Yip, N. Dewal, J. He, J. Tuesdell, J.A. Elvin, G. Otto, D. Lipson, J.S. Ross, V.A. Miller, M. Doherty, C. Vietz: Employee and stockholder: Foundation Medicine Inc. E. Peters, E. Schleifman, J. Noe: Employee and stockholder: Genentech Inc. S. Jenkins: Employee and stockholder: AstraZeneca.

Purpose: The aim of this study is to evaluate the effect of timing of surgery and spring characteristics on correction of scaphocephalic deformity in patients undergoing spring-mediated cranioplasty (SMC) for sagittal craniosynostosis. Methods: The authors conducted a review of patients with sagittal craniosynostosis who underwent SMC at a tertiary referral center between July 2011 and March 2017, with a primary outcome measure of head shape, both preoperatively and postoperatively, determined by cephalic index (CI). Patient demographics and operative details including timing of surgery and spring characteristics were collected. Differences in CI preoperation and postoperation were compared using Wilcoxon signed-rank test. Ordinary least-squares linear regression was used to assess the impact of timing, number of springs, maximum single spring force, and total spring force on postoperative change in CI. Results: Thirty-six subjects (12 males and 24 females) were included in the study. Mean age at spring placement was 3.9 months (range: 1.9–9.2) with a mean follow-up of 1.4 years (range: 0.3–5.2). The mean number of springs used was 3 (range: 2–4). The mean maximum single spring force was 9.9 Newtons (N) (range: 6.9–13.0) and the mean total spring force was 24.6 N (range: 12.7–37.0). Mean CI increased from 70 ± 0.9 preoperatively to 77 ± 1.0 postoperatively (P < 0.001). Age at spring placement was significantly associated with change in CI: for every month increase in age, the change in CI decreased by 1.3 (P = 0.03). The number of springs used, greatest single spring force, and total spring force did not correlate with changes in CI (P = 0.85, P = 0.42, and P = 0.84, respectively). Conclusion: In SMC, earlier age at time of surgery appears to correlate with greater improvement in CI, at least in the short-term. While spring characteristics did not appear to affect head shape, it is possible that the authors were underpowered to detect a difference, and spring-related variables likely deserve additional study.

Objective: Neuroendocrine carcinomas (NEC) of gynecologic and breast origin are aggressive histologic subtypes that confer a worse prognosis. We hypothesized that a single comprehensive genomic profiling (CGP) assay could better match breast and gynecologic NEC subsets to mechanistically driven treatment modalities by simultaneously assessing tumor microsatellite instability (MSI), tumor mutation burden (TMB), and targetable genomic alterations (GA).

Purpose: The purpose of this study was to evaluate the indications, safety, and short-term outcomes of posterior vault distraction osteogenesis (PVDO) in patients with no identified acrocephalosyndactyly syndrome (study) and to compare those to a syndromic cohort (controls). Methods: Demographic and perioperative data were recorded and compared across the study and control groups for those who underwent PVDO between January 2009 and December 2016. Univariate analysis was conducted using &khgr;2 and Fisher exact tests for categorical variables, and Mann–Whitney U test for continuous variables. Results: Sixty-three subjects were included: 19 in the nonsyndromic cohort, 44 in the syndromic cohort. The cohorts had similar proportion of subjects exhibiting pansynostosis (42.1% of nonsyndromic versus 36.4% of syndromic, P = 0.667). The nonsyndromic cohort was significantly older (4.04 ± 3.66 years versus 2.55 ± 3.34 years, P = 0.046) and had higher rate of signs of raised intracranial pressure (68.4% versus 25.0%, P = 0.001) than the syndromic cohort. There was no significant difference in perioperative variables or rate of complications (P > 0.05). The mean total advancement distance achieved was similar, 27 ± 6 mm in the nonsyndromic versus 28 ± 8 mm in the syndromic cohort (P = 0.964). All nonsyndromic subjects with signs of raised intracranial pressure demonstrated improvement at an average follow-up of 22 months. Conclusion: As in the syndromic patient, PVDO is a safe and, in the short-term, effective modality for cranial vault expansion in the nonsyndromic patient. The benefits and favorable perioperative profile of PVDO may therefore be extended to patient populations other than those with syndromic craniosynostosis.

Background: Microsatellite instability (MSI) is a hallmark of mismatch repair (MMR) deficiency and can be attributed to alterations in MMR-related genes including MSH2, MLH1, MSH6, and PMS2. Although alterations in PI3K pathway genes have been reported in MSI-High (MSI-H) colorectal carcinoma (CRC), a comprehensive enrichment analysis of the genomic landscape in MSI-H and MSI-stable (MSS) populations across tumor types is lacking. To better understand the molecular signatures of MSI and investigate new avenues for therapeutic opportunities, we sought to define the genomic landscape of MSI-H tumors across cancer types. Methods: Comprehensive genomic profiling of 395 cancer-related genes, including MSI status, was performed on ∼70,000 tumors. To identify potential driver alterations enriched in MSI-H tumors, variants in regions likely to be affected by polymerase slippage were excluded. Results: As expected, alterations in MSH2, MHL1, MSH6, and PMS2 as well as MMR deficiency variants were enriched in MSI-H specimens regardless of tumor type. We confirmed that variants in PI3K genes were enriched in MSI-H tumors in CRC. Importantly, this was observed across all MSI-H tumors, with 57% of pan-solid MSI-H tumors harboring a PI3K pathway variant compared to 24% of MSS tumors. WNT pathway variants were also enriched specifically in MSI-H tumors, except for CRC, in which frequent APC variants in MSS resulted in WNT enrichment in MSS tumors. Together, 84% of MSI-H tumors have at least one PI3K or WNT pathway variant (compared to 48% of MSS samples). Finally, although ERBB2 alterations occur in both MSS and MSI-H tumors, we found that ERBB2 amplifications occur nearly exclusively in MSS tumors, while ERBB2 missense mutations are enriched in MSI-H tumors. Conclusions: The genomic landscapes of MSI-H and MSS tumors suggest that they acquire alterations in distinct pathways. MSI-H tumors appear to share signaling pathway alterations across diseases, suggesting that MSI-H tumors may be more molecularly similar to one another than they are to MSS tumors of the same disease histology. These data may provide new avenues for exploration of targeted therapies in MSI-H tumors. Legal entity responsible for the study: Foundation Medicine Inc. Funding: Foundation Medicine Inc. Disclosure: S.E. Trabucco: Current employee at Foundation Medicine. S.L. Maund, P.S. Hegde, S-M.A. Huang: Current employee and has ownership interest in Genentech. R. Hartmaier, K. Gowen, KJ. Sun, G.M. Frampton, P.J. Stephens: Current employee and has ownership interest in Foundation Medicine.

Abstract High-grade epithelial ovarian carcinomas containing mutated BRCA1 or BRCA2 (BRCA1/2) homologous recombination (HR) genes are sensitive to platinum-based chemotherapy and PARP inhibitors (PARPi), while restoration of HR function due to secondary mutations in BRCA1/2 has been recognized as an important resistance mechanism. We sequenced core HR pathway genes in 12 pairs of pretreatment and postprogression tumor biopsy samples collected from patients in ARIEL2 Part 1, a phase II study of the PARPi rucaparib as treatment for platinum-sensitive, relapsed ovarian carcinoma. In 6 of 12 pretreatment biopsies, a truncation mutation in BRCA1, RAD51C, or RAD51D was identified. In five of six paired postprogression biopsies, one or more secondary mutations restored the open reading frame. Four distinct secondary mutations and spatial heterogeneity were observed for RAD51C. In vitro complementation assays and a patient-derived xenograft, as well as predictive molecular modeling, confirmed that resistance to rucaparib was associated with secondary mutations. Significance: Analyses of primary and secondary mutations in RAD51C and RAD51D provide evidence for these primary mutations in conferring PARPi sensitivity and secondary mutations as a mechanism of acquired PARPi resistance. PARPi resistance due to secondary mutations underpins the need for early delivery of PARPi therapy and for combination strategies. Cancer Discov; 7(9); 984–98. ©2017 AACR. See related commentary by Domchek, p. 937. See related article by Quigley et al., p. 999. See related article by Goodall et al., p. 1006. This article is highlighted in the In This Issue feature, p. 920

More than 25 inherited neurological disorders are caused by the unstable expansion of repetitive DNA sequences termed short tandem repeats (STRs). A fundamental unresolved question is why specific STRs are susceptible to unstable expansion leading to severe pathology, whereas tens of thousands of normal-length repeat tracts across the human genome are relatively stable. Here, we unexpectedly discover that nearly all STRs associated with repeat expansion diseases are located at boundaries demarcating 3-D chromatin domains. We find that boundaries exhibit markedly higher CpG island density compared to loci internal to domains. Importantly, disease-associated STRs are specifically localized to ultra-dense CpG island-rich boundaries, suggesting that these loci might be hotspots for epigenetic instability and topological disruption upon unstable expansion. In Fragile X Syndrome, mutation-length expansion at the Fmr1 gene results in severe disruption of the boundary between TADs. Our data uncover higher-order chromatin architecture as a new dimension in understanding the mechanistic basis of repeat expansion disorders.