Citations

Abstract Blood lipids have been associated with the development of a range of cancers, including breast, lung and colorectal cancer. For endometrial cancer, observational studies have reported inconsistent associations between blood lipids and cancer risk. To reduce biases from unmeasured confounding, we performed a bidirectional, two‐sample Mendelian randomization analysis to investigate the relationship between levels of three blood lipids (low‐density lipoprotein [LDL] and high‐density lipoprotein [HDL] cholesterol, and triglycerides) and endometrial cancer risk. Genetic variants associated with each of these blood lipid levels ( P < 5 × 10 −8 ) were identified as instrumental variables, and assessed using genome‐wide association study data from the Endometrial Cancer Association Consortium (12 906 cases and 108 979 controls) and the Global Lipids Genetic Consortium (n = 188 578). Mendelian randomization analyses found genetically raised LDL cholesterol levels to be associated with lower risks of endometrial cancer of all histologies combined, and of endometrioid and non‐endometrioid subtypes. Conversely, higher genetically predicted HDL cholesterol levels were associated with increased risk of non‐endometrioid endometrial cancer. After accounting for the potential confounding role of obesity (as measured by genetic variants associated with body mass index), the association between genetically predicted increased LDL cholesterol levels and lower endometrial cancer risk remained significant, especially for non‐endometrioid endometrial cancer. There was no evidence to support a role for triglycerides in endometrial cancer development. Our study supports a role for LDL and HDL cholesterol in the development of non‐endometrioid endometrial cancer. Further studies are required to understand the mechanisms underlying these findings.

Artificial intelligence (AI) and machine learning (ML) in medicine are currently areas of intense exploration, showing potential to automate human tasks and even perform tasks beyond human capabilities. Literacy and understanding of AI/ML methods are becoming increasingly important to researchers and clinicians. The first objective of this review is to provide the novice reader with literacy of AI/ML methods and provide a foundation for how one might conduct an ML study. We provide a technical overview of some of the most commonly used terms, techniques, and challenges in AI/ML studies, with reference to recent studies in cardiac electrophysiology to illustrate key points. The second objective of this review is to use examples from recent literature to discuss how AI and ML are changing clinical practice and research in cardiac electrophysiology, with emphasis on disease detection and diagnosis, prediction of patient outcomes, and novel characterization of disease. The final objective is to highlight important considerations and challenges for appropriate validation, adoption, and deployment of AI technologies into clinical practice.

BCL6 corepressor like-1 (BCORL1) mutation has rarely been described in thyroid cancer or in association with BRAF mutations in any malignancy. However, we report a 49-year-old woman who had aggressive follicular variant papillary thyroid carcinoma (FV-PTC) with both the BRAF K601E and BCORL1 mutations. The patient underwent a total thyroidectomy for a 3.6 cm right thyroid nodule and a smaller lesion in the left lobe in 2007; both were FV-PTCs with no lymphovascular invasion or metastases. In 2015, a positron emission tomography–CT scan showed a small defect in the left posterior lateral fifth rib with mild increased hypermetabolic activity with standardised uptake value of 3.9 and another lesion in the right hip at the junction of the femoral neck and trochanter. Tumour biopsy and genetic analysis revealed an uncommon BRAF K601E and a rare BCORL1 mutation. While rare, we report a case of aggressive FV-PTC with both the BRAF K601E and BCORL1 mutations.

Abstract Noninvasive wearable devices have great potential to aid the management of epilepsy, but these devices must have robust signal quality, and patients must be willing to wear them for long periods of time. Automated machine learning classification of wearable biosensor signals requires quantitative measures of signal quality to automatically reject poor‐quality or corrupt data segments. In this study, commercially available wearable sensors were placed on patients with epilepsy undergoing in‐hospital or in‐home electroencephalographic (EEG) monitoring, and healthy volunteers. Empatica E4 and Biovotion Everion were used to record accelerometry (ACC), photoplethysmography (PPG), and electrodermal activity (EDA). Byteflies Sensor Dots were used to record ACC and PPG, the Activinsights GENEActiv watch to record ACC, and Epitel Epilog to record EEG data. PPG and EDA signals were recorded for multiple days, then epochs of high‐quality, marginal‐quality, or poor‐quality data were visually identified by reviewers, and reviewer annotations were compared to automated signal quality measures. For ACC, the ratio of spectral power from 0.8 to 5 Hz to broadband power was used to separate good‐quality signals from noise. For EDA, the rate of amplitude change and prevalence of sharp peaks significantly differentiated between good‐quality data and noise. Spectral entropy was used to assess PPG and showed significant differences between good‐, marginal‐, and poor‐quality signals. EEG data were evaluated using methods to identify a spectral noise cutoff frequency. Patients were asked to rate the usability and comfort of each device in several categories. Patients showed a significant preference for the wrist‐worn devices, and the Empatica E4 device was preferred most often. Current wearable devices can provide high‐quality data and are acceptable for routine use, but continued development is needed to improve data quality, consistency, and management, as well as acceptability to patients.

TPS11570 Background: Soft tissue sarcoma (STS) is a heterogenous malignancy of mesenchymal origin and includes more than 50 biologically distinct subtypes. Leiomyosarcoma (LMS), a neoplasm of smooth muscle origin, represents up to 20% of STS. The uterus is the most common site of origin in women. Advanced uterine LMS (uLMS) is initially treated with gemcitabine + docetaxel or anthracycline-based chemotherapy but overall survival remains < 24 mos. Besides recurrent alterations in RB1, TP53 and ATRX, insight into cancer biology of uLMS remains limited. Recently, whole exome and transcriptomic sequencing studies suggest uLMS harbors characteristic defects in the homologous recombination (HR) DNA repair pathway and thus features of BRCAness. HR-deficient cancers are unable to efficiently repair double-stranded DNA breaks and appear sensitive to treatment with poly ADP-ribose polymerase (PARP) inhibitors. In preclinical studies, the combination of temozolomide (T), an alkylating agent, and olaparib (O), a PARP inhibitor, was synergistic and markedly suppressed proliferation of uLMS models. A recent phase II study in small cell lung cancer defined the RP2D for T + O where the chief toxicity was myelosuppression. Methods: NCI Protocol #10250 is a single-arm, open-label, multi-center phase II clinical trial of T + O in patients with advanced uLMS. Eligible pts have ECOG PS ≤ 2, progression on ≥ 1 prior line of therapy and disease measurable by RECIST v1.1 and amenable to image-guided biopsy. Pts receive T 75 mg/m2 PO daily + O 200 mg PO BID on days 1-7 in 21-day cycles. The 1° endpoint is objective response rate (ORR). A one-stage binomial design is used to evaluate for an ORR ≤ 10% (null hypothesis) versus ≥ 35% (alternative hypothesis). The design calls for 22 patients. If 5/22 respond, the treatment is promising. This design yields 93% power and 1-sided type I error of 6%. 2° endpoints include progression free survival and safety. All pts undergo tumor biopsies pre-treatment and during cycle 2. Tissue is used for correlative analysis interrogating uLMS for features of BRCAness through (a) whole exome sequencing/RNAseq to evaluate for alterations in HR pathway component genes, (b) RAD51 foci formation by immunohistochemistry as a functional marker of HR pathway activity and (c) protein expression of Schlafen family member number 11 (SLFN11), an emerging biomarker for PARPi. Tumors are also evaluated for MGMT protein expression, a known determinant of sensitivity to T. The study opened to accrual 10/2019. Clinical trial information: NCT03880019.

11564 Background: Epithelioid sarcoma (ES) is a rare, aggressive soft tissue sarcoma characterized by loss of inhibitor of integrase 1 (INI1), allowing enhancer of zeste homologue 2 (EZH2) to repress cell differentiation and promote tumorigenesis. Tazemetostat (TAZ) is a selective inhibitor of EZH2 approved by the FDA for treatment of patients (pts) aged ≥16 years with metastatic or locally advanced ES ineligible for complete resection. Methods: This open-label, multicenter, multi-cohort phase 2 study (NCT02601950) evaluated safety and efficacy of TAZ in pts with INI1-negative tumors. ES pts were enrolled in Cohorts 5 and 6; pts in Cohort 6 underwent mandatory pre-dose (at screening) and post-dose biopsies (at Day 1 of cycle 2). Herein, we report the interim efficacy and safety data from Cohort 6. Results: As of July 31 2019, 44 pts were enrolled into Cohort 6 and treated with TAZ 800 mg BID. The objective response rate (ORR) was 11.4%; 4 pts (9.1%) had a partial response and 1 pt (2.3%) had a complete response (Table). Another 17 pts (38.6%) had stable disease (SD). 18 pts had progressive disease; 13 of these pts remained on study beyond progression. Progression-free survival (PFS) and overall survival (OS) at 56 weeks were 19.4% and 59.4%, respectively. In a pooled posthoc analysis of 106 ES pts from Cohorts 5 (n = 62), and 6, ORR was 13.2%. Grade 3–4 adverse events (AEs) were reported in 16 pts (36.4%), most commonly anemia (6.8%; n = 3) and tumor pain (6.8%; n = 3). 3 pts (6.8%) experienced grade 3–4 treatment-related AEs. One pt discontinued study drug and there were no treatment-emergent dose reductions or treatment-related deaths. These safety data from Cohort 6 are consistent with previously reported data from Cohort 5. 19 paired biopsies were included for translational endpoint analyses. Preliminary RNA seq and pathway analyses are currently ongoing and updated data, including additional biomarker data will be presented. Conclusions: Consistent with previously reported data from the completed Cohort 5, TAZ demonstrated clinically meaningful, durable, single agent activity in ES pts. Efficacy data from Cohort 6 continue to mature with 8 patients still on treatment. TAZ was well tolerated with a low incidence of treatment related AEs. Clinical trial information: NCT02601950 . [Table: see text]

11562 Background: Soft tissue sarcoma (STS) is a heterogeneous malignancy of connective tissue. Although mTOR is implicated in STS pathogenesis, clinical activity from mTORC1 inhibitors is modest. M, a potent selective mTORC1/mTORC2 inhibitor, was more effective in STS preclinical models than inhibitors of mTORC1, IGF1R and mTORC1+IGF1R, owing to more complete suppression of PI3K/AKT/mTOR and abrogation of feedback AKT activation. P, an oral multikinase inhibitor, is approved for non-adipocytic STS and often used after progression (PD) on chemotherapy. In phase 1, the RP2D of M was 30 mg weekly. A091304 was to evaluate M as a novel targeted therapy for STS. Methods: In A091304, pts were randomized 1:1 to M 30 mg weekly or P 800 mg daily. Eligibility required Eastern Cooperative Oncology Group PS ≤ 1, progression on ≥ 1 prior chemotherapy and specific STS subtypes (cohort 1: UPS; 2: LMS; 3: MPNST, SS). Crossover to M was allowed after PD on P. 1° endpoint was progression-free survival (PFS). Assuming median PFS of P was 4.6 months (mo), 98 pts yielded 80% power to detect a hazard ratio of 0.66 favoring M [1-sided test, alpha = 0.15] and including 1 planned futility analysis. 2° endpoints were response rate, clinical benefit rate (CBR) at 4 mo and safety. After 4 of the first 12 pts randomized to P experienced ≥ grade (gr) 3 toxicity, the study was amended to begin at P 400 mg, allowing titration to 800 per investigator discretion. Results: After protocol amendment, 114 pt underwent randomization (M: 56, P: 58), and 111 initiated treatment. Median PFS was 2 mo for M and 2.1 mo for P (HR = 1.47; 1-sided 85% upper confidence boundary = 1.85), with 2 partial responses in each arm. CBR was 5.4% for M and 13.8% for P. Median OS was 10.7 mo for M and 13.9 mo for P (HR = 1.41; 95% CI 0.80-2.49). 26/43 pt with PD on P crossed over to M. Median PFS after crossover was 1.8 mo (95% CI 1.5-3.5). Gr 3 drug-related adverse events (AEs) occurred in 36% on M and 41% on P; gr 4 toxicity was rare. AEs were consistent with known effects of M and P. Conclusions: P at 400 mg daily (allowing escalation to 800 mg per investigator discretion) demonstrated a shorter PFS as compared prior randomized studies with P. Despite this, M failed to demonstrate superior clinical activity as compared to P at the interim analysis. Further work will examine activity within histology-specific cohorts and evaluate available tissue samples for evidence of pharmacodynamic activity. Support: U10CA180821, U10CA180882, U10CA180888, UG1CA233324 (SWOG); https://acknowledgments.alliancefound.org . Clinical trial information: NCT02601209 .

11561 Background: Angiosarcoma has a particularly poor prognosis with 5-year overall survival rates of approximately 30-40%. Treatment of locally advanced and metastatic angiosarcoma is inadequate. Data strongly suggest concurrent, potent inhibition of VEGFR and Tie2 represents an attractive therapeutic strategy in angiosarcoma. Regorafenib displays potent VEGFR and Tie2 receptor inhibition and also possesses activity against additional potential targets in angiosarcoma including PDGFRs, RAF, KIT and FGFR, amongst others. Methods: A multicenter phase II study of regorafenib in patients with locally advanced or metastatic angiosarcoma was conducted through the Midwest Sarcoma Trials Partnership. Adequate performance status, organ function, measurable disease (RECIST 1.1) and 1-4 prior therapies were required. Regorafenib 160 mg PO daily was given in 28-day cycles (21 days on, 7 days off) until disease progression (PD) or unacceptable toxicity. The primary endpoint was progression-free survival (PFS), assessed at 16 weeks. Secondary endpoints include overall response rate (ORR), clinical benefit rate (CBR), OS, and safety and tolerability. A Simon 2-stage design was used. Results: After final enrollment of the second stage, a total of 31 pts were enrolled at 6 sites, 23 are evaluable for response. Median age was 65 (range 30-91); 50% were female, 67.7% had metastatic disease. PFS at 4 months is 52.2% with a median PFS and OS of 3.55 and 11.4 months. 1 confirmed CR and 2 PR, 12 SD and 8 PD were observed. ORR and CBR are 14.29 and 65.2%, respectively. No uncommon grade 3-4 adverse events were observed. 6 pts were non-evaluable due to refusal of further therapy and 2 patients progressed prior to first evaluation. Conclusions: Regorafenib was well tolerated in this study of pretreated patients with angiosarcomas and met its primary endpoint with a median PFS > 45% at 4 months. Treatment was feasible and did not reveal any previously unreported toxicities. Efficacy outcomes were complicated by early withdrawals of patients. RECIST responses were encouraging and regorafenib has a clinically meaningful 4-month PFS. Clinical trial information: NCT02048722 .

11539 Background: Ripretinib is a novel switch-control TKI that broadly inhibits KIT and PDGFRA kinase signaling. In INVICTUS (NCT03353753), a randomized, double-blind, placebo (PBO)-controlled trial of ripretinib in ≥4th-line advanced GIST, ripretinib reduced the risk of disease progression or death by 85% vs PBO with a favorable overall safety profile. Common ( > 20%) adverse events (AEs) included, but were not limited to, alopecia and PPES. Exploratory analyses evaluated the impact of alopecia and PPES on quality of life (QoL). Methods: Patients (pts) with advanced GIST previously treated with at least imatinib, sunitinib, and regorafenib were randomized (2:1) to ripretinib 150 mg QD or PBO. AEs were graded using CTCAE v4 and PROs collected using EQ-5D-5L (EQ5D) and EORTC QLQ-C30 (C30). Repeated measures (RM) models assessed the impact of alopecia and PPES on 5 PROs (EQ5D visual analogue scale; and C30 physical functioning, role functioning, and the overall health and overall QoL questions) within the ripretinib arm. Fixed effects were sex, alopecia/PPES, and ECOG scores at baseline. Results: 128/129 randomized pts received treatment (85 ripretinib 150 mg QD; 43 PBO). Alopecia, regardless of causality, occurred in 44 (51.8%) on ripretinib (34 [40.0%] grade 1; 10 [11.8%] grade 2) and 2 (4.7%) on PBO (both grade 1). PPES occurred in 18 (21.2%) on ripretinib (11 [12.9%] grade 1; 7 [8.2%] grade 2); none on PBO. The median times in days to first occurrence and worst severity grade with ripretinib were 57.0 and 62.5 for alopecia; 56.5 and 57.0 for PPES. The RM models showed a slight trend towards improvement in PRO score over time for pts with alopecia; the only association reaching a P-value of < 0.05 was between alopecia and increased overall QoL. None of the associations between PPES and PRO scores reach P < 0.05. All PRO p-values are nominal, and no statistical significance is being claimed. Conclusions: Ripretinib had a favorable overall safety and tolerability profile. When stratified by alopecia and PPES, patient-reported assessments of function, overall health, and overall QoL were maintained over time. For both alopecia and PPES, onset and maximum severity occurred almost simultaneously, indicating that these events generally did not progressively worsen. These results suggest that alopecia and PPES are manageable and do not have a negative effect on function, overall health, and QoL. Clinical trial information: NCT03353753 .