After careful peer review of Round 3 applications submitted in the Spring of 2019, we announced the five Northern New England Clinical & Translational Research Network pilot project award winners:
Kate Ahrens (USM), Marjorie Meyer (UVM), Carole McBride (UVM)
The Health Outcomes of Pregnant Women with Opioid Use Disorder (Maine and Vermont)
During 1999-2014, Maine and Vermont had the highest annual rate increases in opioid use disorder at delivery among 28 states with data (Haight, MMWR, 2018). In addition, Maine was among the top 10 states for drug overdose death rates in 2017 and Vermont was ranked above the national average (CDC, 2018). Maine and Vermont each have an All-Payer Claims Database (APCD), which is a repository of health care claims data for the majority of residents with health insurance in each state. Both APCDs include eligibility, medical, dental, and pharmacy data from commercial payers, Medicare, Medicaid, and select self-funded/third party administrators. Claims data have not yet been used to assess the type of care received by women with opioid use disorder during pregnancy on subsequent treatment retention and health outcomes, particularly maternal health outcomes postpartum.
Sanjeev Francis (MMC), Scot Remick (MMC), Rosion Morgan (MMC), Julia Moukharskaya (MMC)
Screening and Cardio-surveillance of Cancer Patients Undergoing Immune Checkpoint Blockade
Cardio-Oncology is an emerging discipline that engages cardiologists and medical oncologists in a highly collaborative and interdisciplinary approach to manage cancer patients. Anticancer therapy with immune checkpoint blockade is an illustrative class of novel agents with a wide spectrum of cardiotoxicity, including fatal myocarditis. There is an enormous gap in our knowledge on how to optimally assess and monitor cancer patients prior to receiving these agents to further mitigate cardiotoxicity. Building on the nascent interdisciplinary clinical collaboration between the MMC Cardiovascular Institute and Cancer Institute, and the formative MaineHealth Cancer Care Network (MHCCN) we propose a NNE-CTR pilot study for screening and cardio-surveillance of cancer patients undergoing therapy with immune checkpoint inhibitors. The rationale for this pilot project is predicated on the lack of any sound clinical appreciation of perturbations in cardiac function with immune checkpoint blockade over time.
With this backdrop our team will test our central hypothesis that there are identifiable cues for cardiotoxicity in high-risk patients with immune checkpoint blockade by the pursuit of two aims: Aim 1 -- we will prospectively recruit 15 patients with an elevated cardiac risk defined as greater than 10% risk of a cardiovascular lifetime event [e.g., coronary heart disease (CHD) and/or stroke using the ACC/AHA Risk Tool] or pre-existing heart failure (either preserved or reduced ejection fraction) starting single-agent immune checkpoint blockade; and Aim 2 -- we will prospectively recruit 10 patients regardless of cardiovascular risk, starting therapy with dual immune checkpoint blockade, which is becoming more common. We will establish and monitor cardiac physiology/function [e.g., serial EKGs, baseline echocardiography (ECHO) with assessment of myocardial strain, and ambulatory telemetry monitoring, which we also hypothesize will be more discriminating than other cardiac monitoring modalities] and select biomarkers (e.g., cardiac troponins and BNP) at baseline and over a 12-month period of therapy with immune checkpoint blockade in these two cohorts.
Steven Lidofsky (UVM)
Harnessing the Electronic Health Record in Primary Care for Hepatocellular Carcinoma Surveillance in Cirrhosis
Hepatocellular carcinoma (HCC), a major cause of cancer worldwide, often escapes diagnosis until it reaches a stage that is beyond curative treatment. By the time symptoms develop, the short-term prognosis is grim. Effective screening and surveillance is currently the best way to surmount this problem. In Western nations, over 80% of HCC cases occur in individuals with cirrhosis, the end result of progressive fibrosis in chronic liver disease and a condition in which HCC incidence is approximately 2-4% per year. Driven by a rise in cirrhosis prevalence, HCC is projected to become the third leading cause of cancer-related death in the United States by 2030. Consequently, professional society practice guidelines endorse HCC surveillance in cirrhosis, with the recommended surveillance tool of abdominal ultrasound, which has been shown to increase the probability of HCC detection at a curable stage and to decrease cancer-related mortality. Yet, the minority of patients with known cirrhosis undergo ultrasound surveillance, and there are gaps in surveillance rates between specialized referral centers and the community. Thus, there is a critical need to reduce barriers to surveillance, regardless of place of residence. Without this, HCC diagnosis at an incurable stage is likely to prevail, particularly in regions at a distance from specialty centers, and geographical disparities in optimal treatment will persist.
Our long-term goal is to improve surveillance strategies for HCC in cirrhosis, which will lead to better disease-related outcomes. The objectives in this proposal are: (a) to examine geographic factors that influence HCC surveillance, and (b) to test the impact of an automated electronic health record (EHR)-based tool, for use in the primary care setting, on surveillance ultrasound adherence in individuals with cirrhosis. The central hypothesis is that an automated EHR-based reminder will increase rates of HCC surveillance, overcome geographical barriers to early diagnosis, and translate into an increased proportion of detected cases at a curable stage of disease. This hypothesis is based upon: (a) our preliminary observations that HCC surveillance occurs less frequently among rural (compared with non-rural) individuals with cirrhosis, (b) the literature that highlights geographical disparities in screening for other prevalent cancers, (c) studies that support the power of the EHR to facilitate screening for and management of a variety of health disorders, and (d) our own published findings, which have demonstrated the utility of EHR-based reminders in primary care for hepatitis C detection and linkage to treatment in at risk-patients. The rationale for this proposal is that its successful completion should establish a model for automated HCC surveillance in a high-risk population, which will streamline disease management and reduce geographical disparities in care.
At the completion of the proposed research, we expect to have found that: (a) in cirrhosis, HCC surveillance rates in primary care are low, particularly among patients from rural, compared with non-rural regions, and (b) both patient-related factors and health care-related factors contribute to lower surveillance rates in rural regions. This information is essential for the development of EHR-based strategies that will increase HCC surveillance rates in this high-risk and increasingly prevalent population. Given the widespread use of the EHR in primary care, we further expect that our findings will inform the design of systems that will have an important positive impact on access to effective HCC detection and which will improve cancer-related outcomes overall.
Deirdre O'Reilly (UVM), Alexa Craig (MMC), Tyler Hartman (Dartmouth)
Sleep Disturbance on Bedside EEG: A Biomarker for Neonatal Abstinence Syndrome
Dysregulated sleep is a hallmark of neonatal abstinence syndrome (NAS). Standardized assessment tools guide clinical decision making and rely upon sleep duration alone and do not assess sleep state. Assessment of sleep metrics by electroencephalogram (EEG) could provide new information about the degree of sleep dysregulation in infants with NAS and provide insight into best practices in care of infants with NAS. Our objective is to study sleep EEG parameters in newborns with NAS and determine if sleep is a feasible objective biomarker of NAS severity. We will perform the study at three centers, the University of Vermont Medical Center, Dartmouth-Hitchcock Medical Center, and Maine Medical Center and enroll approximately 81 patients. For aim 1, an 8 hour EEG will be performed on day of life 2 or 3 to characterize sleep states in early NAS. These sleep EEG parameters will be compared to the standardized clinical assessment tool, the Eating, Sleeping, Consoling (ESC) Care Tool. For aim 2, we will repeat an 8 hour EEG on day 4 or 5 to assess sleep states in late NAS. We will compare patients treated with postnatal opioids to patients not treated with postnatal opioids. We will also compare the late EEG to the early EEG. This goal of this pilot study is to provide evidence to support future research of sleep EEG parameters in NAS. Identification of a biomarker for NAS severity will facilitate further studies to improve care in the NAS population.
We will study sleep EEG parameters in newborns with NAS who are inborn at the University of Vermont, Dartmouth Hitchcock Medical Center, and at Maine Medical Center. We will use micro EEG portable devices with headsets that do not require placement by EEG technicians. These devices transmit digital data wirelessly to a laptop computer that stores the data on an on-board memory card.
Jesse Sammon (MMC), Paul Han (MMC), M Hansen (MMC), M Hayn (MMC), Scott Parrapato (UVM), Mark Plante (UVM)
The Development of an Enhanced Clinical Encounter as a Method of Improving Shared Decision Making at the Time of PSA Elevation
Shared decision making (SDM) has been described as most useful for decisions in which there is more than one medically reasonable option. In these instances the choice of which option is best for a given patient depends on his or her preferences and values. The decision to undergo trans-rectal ultrasound guided (TRUS) biopsy for the diagnosis of prostate cancer represents an ideal point along the prostate cancer (PCa) diagnosis-treatment arc for the implementation of SDM. Given the high rate of negative biopsies, the risks associated with biopsy, underlying uncertainty associated with definitive treatment, and the low discriminative value of PSA alone, the decision to undergo a TRUS biopsy is largely driven by patient preferences and risk perception. Nevertheless, there are no widely available decision aids for men considering TRUS biopsy.
The over-arching goals of this project are to 1) develop a process that incorporates an informational decision support tool into the clinical encounter at the time of patient decision to undertake TRUS biopsy, 2) assess the feasibility, acceptability, and effects of the enhanced clinical encounter. This project will involve a user-centered design process, involving stakeholders in all steps of process development and assessment. Process development will incorporate focus groups as well as structured interviews. Process evaluation will incorporate mixed methods to assess feasibility, acceptability, patient knowledge and the value of information given.
In the face of uncertainty around the diagnosis and treatment of prostate cancer (PCa), shared decision making (SDM) has been recommended as an approach to help patients make value-concordant treatment decisions. While SDM as a principle has been widely supported, the question of how best to operationalize SDM for prostate cancer diagnosis remains. This is a critical knowledge gap, given that prostate cancer is often an indolent disease that can be treated conservatively. It is thus medically reasonable for many men with suspected prostate cancer to forego biopsy for definitive diagnosis, particularly if their pre-biopsy risk of prostate cancer is low.
A validated clinical prediction model, the Prostate Cancer Prevention Trial (PCPT) Risk Calculator 2.0, has been developed to produce personalized estimates of a man’s pre-biopsy risk of prostate cancer given their clinical characteristics.3,4 However, there is currently no standardized decision support process to help patients and clinicians use this risk information to make informed and shared decisions about trans-rectal US guided biopsy. The proposed study addresses this gap. Its goal is to develop a decision support process, based on input from the two key stakeholders—patients and Urology physicians. To accomplish this goal, the study will employ an iterative user-centered design process, with feedback from both clinicians and patients, to create a new enhanced SDM clinical encounter and workflow incorporating use of the PCPT risk calculator to provide personalized estimates of patients’ pre-biopsy prostate cancer risk. The study will then implement the new process in high-volume Urology clinical practices at two sites, Maine and Vermont, and assess its feasibility and accessibility.
This pilot study will produce a novel prototype decision support process and preliminary data on its feasibility, acceptability, and effects, which will provide the basis for further larger-scale definitive research studies to assess the effectiveness of providing personalized risk-based decision support to patients with suspected prostate cancer who are contemplating prostate biopsy.