Impact of Remote Symptom Monitoring With Electronic Patient-Reported Outcomes on Hospitalization, Survival, and Cost in Community Oncology Practice: The Texas Two-Step Study

PURPOSE

There is raising interest to implement electronic patient-reported outcomes (ePROs) for symptom monitoring to enhance the quality of cancer care. Step 1 of the Texas Two-Step Study demonstrated successful implementation of an ePRO system in >200 sites of service of a large community oncology practice. We now report step 2 of this study which evaluates the impact of ePROs on outcomes among patients enrolled in the Centers for Medicare & Medicaid Services' Oncology Care Model (OCM) program.

METHODS

This observational study focused on patients with metastatic cancer enrolled in OCM at large community oncology practice located in Texas between July 2020 and December 2020. Patients who completed ≥1 survey via the ePRO tool were included in the study group and were propensity score matched with patients in a control group. Adverse events (AEs; hospitalizations, emergency department visits, deaths) and total cost of care were a priori study outcomes. Mann-Whitney U and chi-square tests compared continuous and categorical variables, respectively, with multivariable logistic regression for adjustment of covariates.

RESULTS

Of 831 patients with metastatic cancer, 458 matched patients (229/group) were identified, with 52% male and a mean age of 74 years. Mean total AEs were lower in the study group compared with control (0.98 v 1.41; P = .007), with decreased hospitalizations (20% v 32.5%; P = .002), emergency visits (38.4% v 42.3%; P > .05), and deaths (11.8% v 16.6%; P > .05). Average number of hospitalizations was lower (0.28 v 0.52; P = .003) with reduced mean duration of hospitalizations (1.9 vs 3.2 d; P = .03). The total cost of care was reduced by an average of $1,146 per member per month.

CONCLUSION

Symptom monitoring with ePROs improved quality and value of cancer care delivery by reducing hospitalizations, emergency visits, and deaths while lowering cost of care in a large oncology practice.

Remote triage and the oncology patient experience

297

Background: Investments in infrastructure to enhance care delivery are often partnered with practice participation in alternative payment model contracts and innovative care delivery models. Oncology nurses routinely are physically located in a cancer practice. We sought to understand the impact of centralizing resources to work remotely building upon triage pathways already in place on staffing and symptom call resolution with the goal of optimizing outpatient symptom management and patient satisfaction. Due to a national RN shortage and a historical high number of RN vacancies in oncology, we thought this this innovative staffing solution may attract, retain, and elevate the role of RNs. Methods: A needs assessment was conducted to evaluate call volumes, staff levels and duties. A gap analysis was performed to determine which duties would be assigned to remote triage versus in person staffing. Some sites had triage; all sites upgraded to new optimization of assessment and management. Standardized evidence based care and communication pathways were implemented. RNs were trained to follow structured workflows for call ticket resolution. A regional pool supported primary triage RNs who were assigned to each site. The model allowed for flexible staffing while building and maintaining relationships with local providers. Additionally, remote triage RNs were given the opportunity to augment their assessment with audio/visual telehealth as needed. Results: In the first 4 months, 9 pilot sites decreased symptom management time by 50%; 1 site decreased by 70% (over 120 minutes down to 27 minutes). Since remote triage began, average resolution times reduced from 2.52 hours to 1.31 hours. One pilot site had an in-person triage position posted and attracted 4 applicants over 4 weeks. When the position was converted to a remote role, 38 applications were received in 1 week. Offering remote triage positions allowed RN recruitment from other states and gave tenured RNs within the organization an attractive new work model. Conclusions: Remote RN work is an attractive opportunity for RNs allowing for broader recruitment of candidates. Removing triage RNs from the clinic site allows them to focus on triage resolution and as a result call resolution times decreased substantially. Standardized communication pathways were developed, rolled out, and optimized in both virtual and in-person Triage RN roles statewide. Adherence to these communication pathways is critical to ensuring timely symptom management resolution and a sustainable workforce.

APPs provide high-quality specialty care in virtual clinic

390

Background: In our team-based approach to care delivery, Advanced Practice Providers (APPs) are a critical component of the care delivery team. In our statewide oncology practice APPs are not available at each site, limiting our ability to provide program visits that enhance patient care and limiting our flexible staffing capacity. APP program visits - Advance Care Planning (ACP), Treatment Review and Coordination (TRC) and Genetics – had limited availability due to the capacity of existing APP staff. We sought to provide both enhancements in patient care with program visits and offer flexible staffing capacity across our statewide practice by using a centralized virtual care clinic model providing high quality specialty care via Telemedicine. Methods: Four APPs were identified and trained to provide full-time telemedicine services statewide under a single collaboration agreement with the Virtual APP Medical Director. Local sites went through a brief onboarding process with the Virtual APP clinic and then submitted requests for appointment coverage by the Virtual APP (VAPP) team. The VAPPs had the same core oncology training, with a few differentiated skills which were matched with coverage requests aligning with their skill sets. To ensure continuity of care, virtual APP clinic notes were visible statewide in the practice EHR, and local providers were alerted by chart message of significant patient concerns. Results: In the first 3 months of service, the VAPPs completed 1,040 appointments for 12 clinics. The VAPPs conducted 50% Program visits and 50% established patient visits (follow-up, on treatment and urgent care). Conducting the Program visits virtually allowed patients to invite family members from any location to join the appointments virtually. The Urgent Care capability was not used frequently in the first 90 days of service, but may still grow in the future. Conclusions: The Virtual APP program ensured oncology patients received high quality, timely care through 1,040 completed visits. This prevented delays in care, resolved staffing challenges, expanded care, and supports virtual care in oncology. This program ensures APPs are available when and where needed and allows them to efficiently serve multiple clinics. The VAPPs provided education, assessment, prevention, and management of toxicities in a flexible manner.

Impact of statewide telemedicine support model in a community oncology practice

391

Background: While there was broad adoption of telemedicine during the COVID-19 pandemic, optimizing the interaction for patients and the clinical team remained a challenge. We sought to optimize delivery of telemedicine services to provide more efficient and effective patient care. Target areas of concern for improvement were scheduling, staffing, communication, technical challenges with operating the platform, a high cancellation rate, and limited copay collections. Methods: A team of 8 virtual Patient Service Coordinators (VPSCs), 8 virtual Medical Assistants (VMAs), and an RN clinical manager was created to work remotely from home to serve Providers at 8 clinics. VPSCs performed check-in duties, demographics, copay collection and technology trouble-shooting with patients. VMAs performed medical intake (medication reconciliation, depression screenings, and vital signs) with real-time EMR input. VMAs stayed in-touch with patients to communicate Provider delays. Standardized communication pathways connected virtual teams with in-clinic teams. The clinics selected to participate in the TMS program were conducting 29% - 50% of E&M visits by telemedicine. The goal of the TMS program was to reduce stress and burnout, as well as relieve in-clinic staff of telemedicine duties giving them capacity to address in-clinic COVID related staff shortages. Results: The TMS Program supported 15,500 visits (11/15/21 – 5/31/22) and increased upfront expected copay collection from 9% pre-program to 100% post program. The program reduced the time for first contact on video from 18 minutes to 1 minute and reduced the telemedicine cancellation rate by 3%. The supported TM cancellation rate was 7% lower than in-person visit cancellation rate. A geographically distributed work from home team was able to support a 66% increase in visits during inclement weather days which allowed visits to be completed that would have otherwise been canceled due to clinic closures. Additionally, the TMS program relieved workload for in-clinic staff and the VPSC and VMA positions proved highly desirable to the eligible workforce. Conclusions: The TMS Program improved patient connectivity and experience, increased upfront co-pay collection, decreased burden on in-clinic staff, allowed continuity of care during inclement weather, and was an attractive work option for staff. Due to its success, the program moved past pilot phase into an operational program.

Evaluating mass implementation of digital health solutions to improve quality and reduce disparities in a large multisite community oncology practice

1507

Background: There is a priority to accelerate the delivery of digital health solutions (DHS) to provide patients with enhanced means for accessing care, but lack of understanding of their utility in certain populations. There are concerns that equitable adoption translate into disparities. We sought to implement a portfolio of DHS across a large practice and characterize engagement across populations to enhance clinical informatics solutions that support care delivery. Methods: This is a retrospective evaluation of cancer patients who engaged with a portfolio of DHS between March 1, 2019 and January 15, 2022. We included four tools with opt-in and opt-out functionality: (1) a care management (CM) platform utilized by clinical staff to manage patient activities, (2) an electronic patient-reported outcomes (ePRO) remote monitoring program for tracking symptoms and oral adherence, (3) a patient portal (PP) for securely accessing patient health records, and (4) digital education (DE) for patients regarding disease and treatments. The engaged population was defined as the number of enrolled patients with at least one (1) record of triage activity, (2) completed ePRO assessment, (3) PP login, and (4) DE read activity, for each tool, respectively. The start of the index period was adjusted based on the first go-live date of each tool. We evaluated factors (age, gender, race/ethnicity, preferred-language, marital status, and distance from clinic) associated with patient engagement using Chi-Square test and multivariate logistic regression. Results: This analysis included a total of 267,375 unique patients. Of the enrolled population per tool, 172,840 (73.6%), 9,938 (67.7 %), 49,771 (79.2%), and 12,044 (56.9%) patients were engaged in CM, ePRO, PP and DE, respectively. The majority (>50%) of engaged patients were female, White and non-Hispanic/Latino, English-language, and aged 61-80 yrs. After adjusting for covariates, we observed that White and non-Hispanic/Latino [(CM: OR 1.15, ePRO OR 1.46, PP: OR 1.48, and DE: OR 1.36) and English-language (CM: OR 1.2, ePRO OR 1.67, PP: OR 1.8 and DE: OR 1.89) patients were significantly (p-value <0.001) more engaged compared to their counterparts. Male patients were less likely to be engaged in CM (OR: 0.79) and ePRO (OR: 0.65) but more engaged in PP (OR: 1.1) compared to females. No significant difference was observed in engagement between non-rural (<20 mile) vs. rural (≥ 20 miles) and in all age groups 21-40, 41-60, 61-80 and >80 years as compared to reference age of 0-20 years for any digital tools except CM. Conclusions: DHS can be used to support the cancer patient journey and we demonstrated high utilization in an array of sociodemographic variables in our population. However, tools designed and implemented with different populations in mind to reduce staff burden and lessen the digital divide should be further explored.

Analyzing patient engagement with digital health tools to facilitate equity across a large statewide community oncology practice

1575

Background: Digital health solutions (DHS) allow for enhanced remote communication between patients and clinical staff and the COVID-19 pandemic has brought these tools to the forefront of care delivery. Once adopted, barriers to adequate utilization still exist. Given the important need to decrease digital divides, and the diversity of patients and care settings across our clinic’s 220 sites of service, we sought to understand how utilization of oncology DHS may be limited among certain populations. Methods: We investigated utilization among cancer patients who enrolled and engaged with a portfolio of DHS between March 1, 2019 and January 15, 2022. This portfolio includes three tools: (1) an electronic patient-reported outcomes (ePRO) remote monitoring program for tracking symptoms and oral adherence, (2) a patient portal (PP) for securely accessing patient health records, and (3) digital education (DE) for patients regarding disease and treatments. ePRO completion rate, average number of PP logins, and DE read rate were used as measures of utilization for each tool, respectively, and compared among patients with different age (< 65 and ≥65 years), language preference [English (EL) or Spanish (SL)], and distance from clinic (non-rural: < 20 miles OR rural: ≥20 miles). Mann-Whitney U and Chi-Square tests were used to compare continuous and categorical variables, respectively. Results: This study included a total of 77,347 unique patients representing 651,004 digital encounters. 9,938 patients engaged in ePRO, 49,771 patients in PP, and 12,044 patients in DE. Engagement across all DHS was high in patients of age group < 65 (ePRO: 72.7%, PP: 79.67% and PE 54.7%) as compared to ≥65 years, but the ePRO completion rate is high in ≥65 age group (59.0% vs 55.6%), whereas no significant difference was observed in the PP login activity and DE read rate. EL patients were significantly (p-value < 0.01) more engaged (ePRO 68% vs. 54%, PP: 80% vs. 62%, DE: 57% vs. 37%) and had higher digital utilization (ePRO completion rate: 57.31% vs 53.23%, average PP logins: 7.48 vs 7.14 and DE read rate: 96.2% vs 90.8%) than SL patients across the DHS. Patients living in rural areas comprised roughly 25% of the population and participated across tools similarly as patients living in non-rural areas (ePRO 67% vs. 69%, PP: 79% vs. 79%, DE: 56.9% vs. 56.8%). Utilization of the portfolio was variable based on rural vs non-rural status (ePRO completion rate: 56.3% vs. 57.4%, average PP logins: 7.9 vs. 7.3, DE read rate: 96.02.7% vs 96.3%). Conclusions: Despite variable engagement based on age, language, and rural status across the portfolio, patients within these populations continue to utilize the DHS. How we understand and explore enhancements to DHS remain under investigation for tool optimization for patient-specific barriers to care.

Systemic Therapy for Advanced Human Epidermal Growth Factor Receptor 2-Positive Breast Cancer: ASCO Guideline Update

PURPOSE

To update evidence-based guideline recommendations to practicing oncologists and others on systemic therapy for patients with human epidermal growth factor receptor 2 (HER2)-positive advanced breast cancer.

METHODS

An Expert Panel conducted a targeted systematic literature review (for both systemic treatment and CNS metastases) and identified 545 articles. Outcomes of interest included efficacy and safety.

RESULTS

Of the 545 publications identified and reviewed, 14 were identified to form the evidentiary basis for the guideline recommendations.

RECOMMENDATIONS

HER2-targeted therapy is recommended for patients with HER2-positive advanced breast cancer, except for those with clinical congestive heart failure or significantly compromised left ventricular ejection fraction, who should be evaluated on a case-by-case basis. Trastuzumab, pertuzumab, and taxane for first-line treatment and trastuzumab deruxtecan for second-line treatment are recommended. In the third-line setting, clinicians should offer other HER2-targeted therapy combinations. There is a lack of head-to-head trials; therefore, there is insufficient evidence to recommend one regimen over another. The patient and the clinician should discuss differences in treatment schedule, route, toxicities, etc during the decision-making process. Options include regimens with tucatinib, trastuzumab emtansine, trastuzumab deruxtecan (if either not previously administered), neratinib, lapatinib, chemotherapy, margetuximab, hormonal therapy, and abemaciclib plus trastuzumab plus fulvestrant, and may offer pertuzumab if the patient has not previously received it. Optimal duration of chemotherapy is at least 4-6 months or until maximum response, depending on toxicity and in the absence of progression. HER2-targeted therapy can continue until time of progression or unacceptable toxicities. For patients with HER2-positive and estrogen receptor-positive or progesterone receptor-positive breast cancer, clinicians may recommend either standard first-line therapy or, for selected patients, endocrine therapy plus HER2-targeted therapy or endocrine therapy alone.Additional information is available at www.asco.org/breast-cancer-guidelines.

Management of Advanced Human Epidermal Growth Factor Receptor 2-Positive Breast Cancer and Brain Metastases: ASCO Guideline Update

PURPOSE

To provide updated evidence- and consensus-based guideline recommendations to practicing oncologists and others on the management of brain metastases for patients with human epidermal growth factor receptor 2 (HER2)-positive advanced breast cancer up to 2021.

METHODS

An Expert Panel conducted a targeted systematic literature review (for both systemic therapy for non-CNS metastases and for CNS metastases of HER2+ guideline updates) that identified 545 articles. Outcomes of interest included overall survival, progression-free survival, and adverse events.

RESULTS

Of the 545 publications identified and reviewed, six on systemic therapy were identified to form the evidentiary basis for the systemic therapy for CNS metastases guideline recommendations.

RECOMMENDATIONS

Patients with brain metastases should receive appropriate local therapy and systemic therapy, if indicated. Local therapies include surgery, whole-brain radiotherapy, and stereotactic radiosurgery. Memantine and hippocampal avoidance should be added to whole-brain radiotherapy when possible. Treatments depend on factors such as patient prognosis, presence of symptoms, resectability, number and size of metastases, prior therapy, and whether metastases are diffuse. Other options include systemic therapy, best supportive care, enrollment onto a clinical trial, and/or palliative care. There are insufficient data to recommend for or against performing routine magnetic resonance imaging to screen for brain metastases; clinicians should have a low threshold for magnetic resonance imaging of the brain because of the high incidence of brain metastases among patients with HER2-positive advanced breast cancer.Additional information is available at www.asco.org/breast-cancer-guidelines.

Patient and clinician perceptions of a digital patient monitoring program in the community oncology setting: Findings from the Texas Two-Step Study

189

Background: Digital monitoring strategies that include electronic patient reported outcomes (ePRO) measures to monitor symptoms among cancer patients have been shown to be effective in improving patient outcomes in a large academic setting and across several smaller multi-center trials. However, demonstration of clinical utility in the real-world setting must incorporate patient and provider perspectives of ePRO programs to ensure successful implementation. We sought to understand perceptions among patients and clinicians in ePRO digital symptom monitoring program. Methods: Texas Two-Step is an ongoing hybrid implementation-effectiveness study of Navigating Cancer’s ePRO digital monitoring program at Texas Oncology. Patients initiating new systemic therapy for their cancer diagnosis were introduced to the program by their oncologist and enrolled in the program by nursing staff for weekly reporting of symptoms based on a modified version of NCI’s PRO-CTCAE instrument. Feedback surveys were administered to both patients and clinic staff after 6 months of implementation of the program to evaluate the overall experience with the program. Results: 1040 (23.5%) patients and 215 (12.4%) clinicians completed the feedback survey. Of the patient responders, 90% found the program very or somewhat easy for reporting symptoms, 85% moderately-extremely beneficial for having symptoms addressed, and 84% moderately-extremely interested in utilizing the program for future treatments. Of the clinician responders, 73% indicated that that they had a good understanding of the benefit of the program; 70.6% felt confident in their ability to interpret patients’ ePRO responses; 80.3% felt confident in their ability to discuss the program with patients; 71.2% confident in their ability to counsel patients based on ePRO responses; and 55.3% felt the program enhanced communication with patients. Additionally, 59% of clinicians felt the program was beneficial for patients. Conclusions: Patients have a more favorable perception of the benefit of the ePRO program than clinicians. Methods to reduce staff burden and reinforcement of program benefits during training and implementation are imperative to improve clinical utility and will be studied further as the program is optimized. As implementation occurred during the COVID-19 pandemic, this may impact perceptions regarding the tool.

Improvement in incident resolution time with implementation of an electronic patient management solution at a community oncology practice

1578

Background: Value-based care models such as the Oncology Care Model incentivize practices to reduce hospitalizations and emergency department (ED) visits. Texas Oncology found that most ED visits occurred during regular business hours. Prolonged patient call back times were consistently rated poorly on satisfaction surveys and often led to ED visits for symptoms that could be managed in our offices. We partnered with Navigating Cancer (NC) to implement an electronic patient management technology solution. Methods: For each of our 200 locations, call volume was estimated based on clinic volume. We then reallocated or hired dedicated triage nurses and operators. Incoming calls were entered into the NC dashboard by operators as incidents which were routed based on symptom priority following system generated prompts. Incident volumes and resolution times were tracked. We instituted PDSA cycles at all locations with a goal of less than 90-minute resolution of symptom-related incidents Utilizing the electronic dashboard allowed us to continue this initiative during the COVID-19 public health emergency as our staff could work remotely. Nurses were able to document if a potential ED visit was avoided. These data points allowed our practice to establish comprehensive and strategic actions plans for quality improvement. Results: We finalized implementation of the system in February of 2020. Total incidents for 2020 were over 1 million, averaging over 5000 per location. Resolution time for all incidents started at 3.2 hours pre-implementation and improved to 2.2 hours in December of 2020. Resolution times for symptom-related incidents started at 2.3 hours pre-implementation and ended at 1.5 hours in December of 2020 with over 60% resolved under one hour. 8% of symptom-related incidents resulted in definite or probable ED avoidances by nursing assessment. Shortness of breath, vomiting, chills, and weakness were the top symptom types addressed for ED avoidances. Conclusions: An electronic patient management solution with PDSA cycles of quality improvement can markedly reduce call back times, especially for symptom related calls. We believe managing symptoms in a timely fashion will lower ED visits and hospitalizations as well as improve patient satisfaction. We will report on these outcomes once available.[Table: see text]

Real-world starting dose and outcomes of palbociclib plus an aromatase inhibitor for metastatic breast cancer

e13021

Background: Palbociclib (PA) has been approved for HR+/HER2–advanced/metastatic breast cancer (mBC) in combination with an aromatase inhibitor (AI) or fulvestrant for more than 6 years. Regardless of the labeled recommended starting dose of 125mg/day, some patients initiate palbociclib at lower doses in routine practice. This study described real-world starting dose, patient characteristics, and effectiveness outcomes of first line PA+ AI for mBC in the US clinical setting. Methods: We conducted a retrospective analysis of Flatiron Health’s nationwide longitudinal electronic health records, which came from over 280 cancer clinics representing more than 2.2 million actively treated cancer patients in the US. Between February 2015 and September 2018, 813 HR+/HER2– mBC women initiated PA+AI as first-line therapy and had ≥ 3 months of potential follow-up. Patients were followed from start of PA+AI to December 2018, death, or last visit, whichever came first. Real-world progression-free survival (rwPFS) was defined as the time from the start of PA+AI to death or disease progression. Real-world tumor response (rwTR) was assessed based on the treating clinician’s assessment of radiologic evidence for change in burden of disease over the course of treatment. Multivariate analyses were performed to adjust for demographic and clinical characteristics. Results: Of 813 eligible patients, 68.3% were white, median age was 65.0 years, and 42.9% had visceral disease (lung and/or liver). Median duration of follow-up was 21.0 months. 805 patients had records of PA starting dose, with 125mg and 75/100mg/day being 86.5% and 13.5%, respectively. Patients who started at 75/100mg/day were more likely to be ≥75 years than those who started at 125mg/day (38.5% vs 17.1%). Other baseline and disease characteristics were generally evenly distributed. Patients who started at 125mg/day had longer median rwPFS (27.8 vs 18.6 months, adjusted HR=0.74, 95%CI=0.52-1.05) and higher rwTR (54.0% vs. 40.4%) than those patients who started 100/75mg/day (adjusted OR=1.76, 95%CI=1.13-2.74). Table presents results in detail. Conclusions: Most patients in this study initiated palbociclib at 125mg/day and dose adjustment was similar regardless of starting dose. These real-world findings may support initiation of palbociclib at a dose of 125mg/day in combination with AI for the first-line treatment of HR+/HER2- mBC. [Table: see text]

Implementation of electronic patient-reported outcomes for symptom monitoring in a large multi-site community oncology practice

12103

Background: Among patients receiving chemotherapy, symptomatic adverse event monitoring with electronic patient-reported outcomes (ePRO) is associated with improved clinical outcomes, satisfaction, and compliance with therapy. Standard approaches for ePRO implementation are not established warranting evaluation in community cancer practices. Objective: Evaluate implementation of ePRO symptom monitoring across a large multi-site community oncology practice network. Methods: Patients initiating a new systemic therapy at one of 210 practice sites in the Texas Oncology Practice were invited to use in the Navigating Cancer ePRO platform, with rolling implementation from July-December 2020. Participating patients received a weekly prompt by text message or email (patient choice) to self-report common symptoms and well-being via web or smartphone. Severe self-reported symptoms triggered a real-time notification alert to nursing triage to address the symptom. Enrollment and compliance were systematically tracked weekly with evaluation of barriers and facilitators to adoption and sustainability. Results: 4375 patients planning systemic treatment enrolled and participated, with baseline characteristics are shown in Table 1. 73% (1841/2522) of enrolled patients with follow up completed at least one ePRO assessment, and among these individuals, 65% (8762/25061) of all available weekly ePRO assessments were completed. Over a 10-week period, compliance with weekly symptom reporting declined from 72% to 52%. Patients on oral therapy had higher compliance rates overall. Barriers currently being addressed include lack of a second reminder text/email prompt, inconsistent discussion of reported ePROs by clinicians at visits, and COVID-related changes in workflow. Facilitators included patient and staff engagement on the importance of PROs for symptom management. Conclusions: ePROs can be effectively implemented in community oncology practice. Utilization of ePROs is high, but diminishes over time without attention to barriers. Ongoing work to address barriers and optimize compliance are underway.[Table: see text]

Telemedicine in Community Cancer Care: How Technology Helps Patients With Cancer Navigate a Pandemic

COVID-19 places unprecedented demands on the oncology ecosystem. The extensive pressure of managing health care during the pandemic establishes the need for rapid implementation of telemedicine. Across our large statewide practice of 640 practitioners at 221 sites of service, an aggressive multidisciplinary telemedicine strategy was implemented in March by coordinating and training many different parts of our healthcare delivery system. From March to September, telemedicine grew to serve 15%-20% of new patients and 20%-25% of established patients, permitting the practice to implement safety protocols and reduce volumes in clinic while continuing to manage the acute and chronic care needs of our patient population. We surveyed practice leaders, queried for qualitative feedback, and established 76% were satisfied with the platform. The common challenges for patients were the first-time use and technology function, and patients were, in general, grateful and happy to have the option to visit their clinicians on a telemedicine platform. In addition to conducting new and established visits remotely, telemedicine allows risk assessments, avoidance of hospitalization, family education, psychosocial care, and improved pharmacy support. The implementation has limitations including technical complexity; increased burden on patients and staff; and broadband access, particularly in rural communities. For telemedicine to improve as a solution to enhance the longitudinal care of patients with cancer, payment coverage policies need to continue after the pandemic, technologic adoption needs to be easy for patients, and broadband access in rural areas needs to be a policy priority. Further research to optimize the patient and clinician experience is required to continue to make progress.

American Society of Clinical Oncology Road to Recovery Report: Learning From the COVID-19 Experience to Improve Clinical Research and Cancer Care

This report presents the American Society of Clinical Oncology's (ASCO's) evaluation of the adaptations in care delivery, research operations, and regulatory oversight made in response to the coronavirus pandemic and presents recommendations for moving forward as the pandemic recedes. ASCO organized its recommendations for clinical research around five goals to ensure lessons learned from the COVID-19 experience are used to craft a more equitable, accessible, and efficient clinical research system that protects patient safety, ensures scientific integrity, and maintains data quality. The specific goals are: (1) ensure that clinical research is accessible, affordable, and equitable; (2) design more pragmatic and efficient clinical trials; (3) minimize administrative and regulatory burdens on research sites; (4) recruit, retain, and support a well-trained clinical research workforce; and (5) promote appropriate oversight and review of clinical trial conduct and results. Similarly, ASCO also organized its recommendations regarding cancer care delivery around five goals: (1) promote and protect equitable access to high-quality cancer care; (2) support safe delivery of high-quality cancer care; (3) advance policies to ensure oncology providers have sufficient resources to provide high-quality patient care; (4) recognize and address threats to clinician, provider, and patient well-being; and (5) improve patient access to high-quality cancer care via telemedicine. ASCO will work at all levels to advance the recommendations made in this report.

FAIRLANE, a double-blind placebo-controlled randomized phase II trial of neoadjuvant ipatasertib plus paclitaxel for early triple-negative breast cancer

BACKGROUND

This hypothesis-generating trial evaluated neoadjuvant ipatasertib-paclitaxel for early triple-negative breast cancer (TNBC).

PATIENTS AND METHODS

In this randomized phase II trial, patients with early TNBC (T ≥ 1.5 cm, N0-2) were randomized 1 : 1 to receive weekly paclitaxel 80 mg/m2 with ipatasertib 400 mg or placebo (days 1-21 every 28 days) for 12 weeks before surgery. Co-primary end points were pathologic complete response (pCR) rate (ypT0/TisN0) in the intention-to-treat (ITT) and immunohistochemistry phosphatase and tensin homolog (PTEN)-low populations. Secondary end points included pCR rate in patients with PIK3CA/AKT1/PTEN-altered tumors and pre-surgery response rates by magnetic resonance imaging (MRI).

RESULTS

pCR rates with ipatasertib versus placebo were 17% versus 13%, respectively, in the ITT population (N = 151), 16% versus 13% in the immunohistochemistry PTEN-low population (N = 35), and 18% versus 12% in the PIK3CA/AKT1/PTEN-altered subgroup (N = 62). Rates of overall and complete response (CR) by MRI favored ipatasertib in all three populations (CR rate 39% versus 9% in the PIK3CA/AKT1/PTEN-altered subgroup). Ipatasertib was associated with more grade ≥3 adverse events (32% versus 16% with placebo), especially diarrhea (17% versus 1%). Higher cycle 1 day 8 (C1D8) immune score was significantly associated with better response only in placebo-treated patients. All ipatasertib-treated patients with low immune scores and a CR had PIK3CA/AKT1/PTEN-altered tumors.

CONCLUSIONS

Adding ipatasertib to 12 weeks of paclitaxel for early TNBC did not clinically or statistically significantly increase pCR rate, although overall response rate by MRI was numerically higher with ipatasertib. The antitumor effect of ipatasertib was most pronounced in biomarker-selected patients. Safety was consistent with prior experience of ipatasertib-paclitaxel. A T-cell-rich environment at C1D8 had a stronger association with improved outcomes in paclitaxel-treated patients than seen for baseline tumor-infiltrating lymphocytes. This dependency may be overcome with the addition of AKT inhibition, especially in patients with PIK3CA/AKT1/PTEN-altered tumors.

CLINICALTRIALS.GOV

NCT02301988.

Driving quality improvement: How clinical decision support can facilitate compliance with evidence-based pathways

2045

Background: Cancer care is changing rapidly with more detailed understanding of disease and more numerous therapeutic choices. As treatment choice is more complex, mechanisms to improve compliance with evidence based treatment can improve the quality of cancer care. Methods: A retrospective cohort study was conducted from January 2014-May 2016 evaluating the impact of a clinical decision support system (CDSS) on compliance with evidence based pathways (EBP) across 9 statewide community based oncology practices. These EBP are developed with physician input on efficacy toxicity and value and incorporated in to a CDSS that is used within the Electronic Health Record (EHR) at point of care to alter the choice architecture a clinician sees when prescribing therapy. A multi-level logistic regression model was used to adjust for group effects on physician or practice behavior. SAS 9.4 software was used and GLIMMIX was applied. Individual physician benchmark compliance was evaluated using McNemar's test. Results: Regimen compliance with EBP was measured pre- and post- implementation of the CDSS tool across a large network encompassing 9 statewide practices and 633 physicians who prescribed over 30,000 individual patient treatment regimens over a 6 month period. The CDSS that is incorporated within the EHR significantly improved compliance with EBP across the entire cohort of practices, and in individual practices (see Table). Individual oncologists reached a target of 75% compliance more often (58% vs 72%) after implementation of the tool (p < 0.001). Conclusions: CDSS is a tool that improves compliance with EBP that is effective at improving targets of compliance broadly, at the practice, and at the individual clinician level. Clinical informatics solutions that influence physician behavior can be inclusive of physicians in design, iterative in process, and nudge as opposed to force clinician behavior to drive quality improvement. These clinical informatics solutions grow in importance as the complexity of cancer care continues to increase and we seek to improve upon the quality and value of care delivery. [Table: see text]

A phase Ib/II study of eribulin (ERI) plus pembrolizumab (PEMBRO) in metastatic triple-negative breast cancer (mTNBC) (ENHANCE 1)

1015

Background: As monotherapies, both ERI (a chemotherapeutic microtubule inhibitor) and PEMBRO (a programmed death [PD]-1 blocking immunotherapy) have shown promising antitumor activity in mTNBC. Emerging data suggest that the addition of immunotherapy to traditional chemotherapy holds promise for mTNBC. This open-label, single-arm, phase 1b/2 study evaluated the safety and efficacy of ERI + PEMBRO in mTNBC. Methods: Patients (pts) with mTNBC and ≤2 prior systemic anticancer therapies for metastatic disease were enrolled and stratified by prior number of therapy (Stratum 1, 0; Stratum 2, 1–2). Pts received IV ERI 1.4 mg/m2 on day (d)1 and d8 and IV PEMBRO 200 mg on d1 of a 21-d cycle. The primary objectives were safety and objective response rate (ORR per RECIST 1.1 by independent imaging review). Assessments also included efficacy outcomes by PD ligand-1 (PD-L1) expression status; PD-L1+ was defined as a combined positive score ≥1 using the PD-L1 IHC 22C3 pharmDx. Results: As of data cutoff (July 31, 2019), 167 pts (Stratum 1, n=66; Stratum 2, n=101) were enrolled and treated. No dose-limiting toxicities were observed. The most common treatment-emergent adverse events were fatigue (66%), nausea (57%), peripheral sensory neuropathy (41%), alopecia (40%), and constipation (37%). No deaths were considered treatment related. The overall ORR was 23.4% (95% CI: 17.2–30.5). Efficacy outcomes by PD-L1 status (PD-L1+, n=74; PD-L1-, n=75) and stratum are presented (table). Conclusions: ERI + PEMBRO has activity in pts with mTNBC. There was a trend toward more robust activity for the combination among patients with PD-L1+ tumors compared to PD-L1- tumors in the first-line setting (Stratum 1); whereas, in the later-line setting (Stratum 2) similar survival outcomes were observed among the PD-L1+ and PD-L1- pts. ERI + PEMBRO shows promise for mTNBC with efficacy that appears greater than historical reports of either agent alone. Clinical trial information: NCT02513472 . [Table: see text]

Clinical decision support tools (CDST) provide education at the point-of-care to assist provider treatment choices

e19165

Background: Cancer care is complex and requires synthesis of increasing amounts of clinical and financial data to optimize treatment decisions. The heightened differentiation of individual cancers, therapy sequencing, and increasing number of treatment options make it more challenging for oncologists to stay current. A survey showed that 82% of Network providers validated the utility of clinical evidence and decision-making assistance at the point-of-care. Providing education on efficacy, toxicity, and cost in the form of evidence tables (ET) as CDST may assist in value-based decision-making. Methods: We retrospectively reviewed utilization of ET developed to provide clinical and financial data about regimens included within the Value Pathways powered by NCCN. Thirty ET were embedded in the electronic health record (EHR) and posted on our intranet between June-December 2019, covering over 90% of cancers. Utilization was queried from roll out through January 2020. ET include a summary of primary literature (primary/secondary endpoints, adverse events) and monthly Medicare allowable reimbursement rates. ET are updated with each change to pathways and quarterly for cost updates. We also conducted a survey to understand provider ET utilization patterns. Results: Utilization was evaluated for 1,200 physicians across 470 sites that have access to ET. ET have been accessed 1178 times by 586 providers within the EHR and accessed 1363 times by 260 providers via intranet. Our rate of repeat users of the ET is 35% in the EHR and 97% on the intranet. A survey of 200 physicians after ET release showed that 19% of physicians use ET with every new chemotherapy start and an additional 50% refer to ET only if they are uncertain about the best option. Conclusions: Utilization patterns underscore the importance of ET as a CDST within the EHR and on the intranet. While early ET use is high, continued tracking of utilization and addition of content to assist in complex clinical decisions is a priority. Providers surveyed found that clinical informatics tools like ET are useful to enhance decision-making in complex cancer care. [Table: see text]

Abstract P2-08-19: Exploratory biomarker analyses of FAIRLANE, a double-blind placebo (PBO)-controlled randomized phase II trial of neoadjuvant ipatasertib (IPAT) + paclitaxel (PAC) for early triple-negative breast cancer (TNBC)

Background: The oral AKT inhibitor IPAT is being evaluated in cancers with a high prevalence of PI3K/AKT pathway activation. In the PBO-controlled randomized phase II FAIRLANE trial (NCT02301988), adding IPAT to PAC as neoadjuvant therapy for TNBC led to a numerical increase in pathologic complete response (pCR) in unselected patients (17.1% vs 13.3%), with a greater treatment effect in patients with PIK3CA/AKT1/PTEN-altered tumors (17.9% vs 11.8%). The addition of IPAT also led to an increase in complete response (CR) by MRI (27.6% vs 13.3%) that was enhanced in patients with PIK3CA/AKT1/PTEN-altered tumors (39.3% vs 8.8%) [Oliveira, AACR 2018]. We report an exploratory analysis performed to provide better understanding of potential biomarkers for response.

Methods: Pretreatment tumor samples were evaluated for genomic alterations using the FoundationOne® (Foundation Medicine) assay (n=144) and gene expression by RNA-Seq (n=92). Samples were classified into TNBC subtypes based on the method developed by Lehmann and Pietenpol [Lehmann, J Clin Invest 2011]. Tumor-infiltrating lymphocytes (TILs) were quantified using the Salgado method [Salgado, Ann Oncol 2015] (n=135).

Results: Of 62 patients (43%) with PIK3CA/AKT1/PTEN-altered tumors, 21 had an activating mutation in PIK3CA or AKT1 and 47 had an alteration in PTEN (6 [3 in each arm] had both PIK3CA mutation and PTEN alteration). Although only 3 patients with PIK3CA/AKT1-mutant tumors achieved a pCR, there was an increased rate of MRI CR with the addition of IPAT to PAC [Table]. In patients with PTEN alterations, both pCR rate and MRI CR rate were increased with IPAT. In patients treated with PBO + PAC, all 4 pCR patients evaluable by RNA-Seq were of the immunomodulatory (IM) subtype. However, in the IPAT + PAC arm, pCRs were also seen in patients with basal-like 1 (BL-1), mesenchymal (M), and mesenchymal stem-like (MSL) subtypes. Consistent with this observation, in the PBO + PAC arm, samples from patients achieving a pCR had significantly higher levels of stromal TILs than those from patients who did not have a pCR, while no difference was observed in the IPAT + PAC arm.

Response, n (%)PIK3CA/AKT mutation (n=21)PTEN alteration (n=47) IPAT + PAC (n=11)PBO + PAC (n=10)IPAT + PAC (n=21)PBO + PAC (n=26)pCR1 (9%)2 (20%)4 (19%)3 (12%)CR by MRI5 (45%)1 (10%)8 (38%)2 (8%)

Conclusions: This retrospective exploratory biomarker analysis of the phase II FAIRLANE trial of neoadjuvant IPAT for TNBC provides insight into the potential heterogeneity of response and resistance to taxane therapy. The results also hint that response to PAC alone is dependent on baseline immune infiltration and that this dependency might be relieved with the addition of AKT inhibition.

Citation Format: Wongchenko MJ, Oliveira M, Saura C, Nuciforo P, Calvo I, Andersen J, Passos Coelho JI, Gil Gil M, Bermejo B, Patt DA, Ciruelos E, Singel SM, Maslyar DJ, Xu N, de la Peña L, Baselga J, Gendreau S, Isakoff SJ. Exploratory biomarker analyses of FAIRLANE, a double-blind placebo (PBO)-controlled randomized phase II trial of neoadjuvant ipatasertib (IPAT) + paclitaxel (PAC) for early triple-negative breast cancer (TNBC) [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P2-08-19.

Using Big Data and Predictive Analytics to Determine Patient Risk in Oncology

Big data and predictive analytics have immense potential to improve risk stratification, particularly in data-rich fields like oncology. This article reviews the literature published on use cases and challenges in applying predictive analytics to improve risk stratification in oncology. We characterized evidence-based use cases of predictive analytics in oncology into three distinct fields: (1) population health management, (2) radiomics, and (3) pathology. We then highlight promising future use cases of predictive analytics in clinical decision support and genomic risk stratification. We conclude by describing challenges in the future applications of big data in oncology, namely (1) difficulties in acquisition of comprehensive data and endpoints, (2) the lack of prospective validation of predictive tools, and (3) the risk of automating bias in observational datasets. If such challenges can be overcome, computational techniques for clinical risk stratification will in short order improve clinical risk stratification for patients with cancer.

Understanding Utilization Management Policy: How to Manage This Increasingly Complex Environment in Collaboration and With Better Data

As innovation in cancer care continues and newer costly therapies receive approval, utilization management will continue to grow as an important way that payers can attempt to control costs and value while providing service to their patients. Although utilization management may be necessary, it takes many forms and is optimized when it ensures appropriate patient access to services and minimizes administrative burdens of physicians and staff. These opportunities are best explored in collaboration with payers. Information systems today provide an excellent platform for data sharing to facilitate collaborative efforts between care delivery organizations and payers to optimize these efforts. As state and national policies differ regarding utilization management, it is important for clinicians to be both aware and involved.

Enhanced compliance with appropriate antiemetic prescribing guidelines by utilizing a medically integrated pharmacy antiemetic therapeutic interchange protocol

55

Background: Appropriate antiemetic prescribing remains an unmet need in patients receiving emetogenic chemotherapy. Non-compliance with standard antiemetic guidelines is well reported, and a focus of the American Society of Clinical Oncology Choosing Wisely campaign. Using our medically integrated pharmacies, a community oncology practice sought to maximize compliance with national guidelines by minimizing the use of NK1 inhibitors for low and minimally emetogenic regimens and optimizing medications for highly emetogenic regimens per NCCN guidelines. Methods: A retrospective review of a medically integrated pharmacy antiemetic therapeutic interchange policy. Trained oncology pharmacists imbedded within 2 cancer centers evaluated all new chemotherapy orders to determine the level of emetogenicity per NCCN guidelines. Pharmacists then used a guideline compliant prespecified therapeutic interchange protocol to adjust the antiemetic medications. Patient education to enhance medication compliance was performed by the pharmacists, nurses, and in one on one treatment review sessions with an advance practice provider. Results: Of 271 low to minimally emetogenic regimens, 99% did not receive an NK1 inhibitor on the first cycle. Of 367 highly emetogenic regimens, 97% appropriately received guideline based therapy. Additionally, the use of olanzapine was non-existent in the practice prior to this protocol and increased to 55% of patients with highly emetogenic regimens. Conclusions: We demonstrated that a therapeutic interchange policy with a medically integrated pharmacist shows enhanced compliance with guideline based antiemetic medication usage. This system also allows rapid uptake of changes in supportive care guidelines such as rapid adoption of olanzapine. Efficacy results of this protocol will be reported separately.

Efficacy of an antiemetic therapeutic interchange protocol driven by a medically integrated pharmacy

133

Background: Control of chemotherapy induced nausea and vomiting (CINV) remains an unmet need. An antiemetic therapeutic interchange protocol was developed to enhance control of CINV. This abstract reports the efficacy results of patients treated with highly emetogenic regimens as defined by NCCN guidelines. Methods: Medically integrated pharmacists imbedded within 2 centers of a community oncology practice evaluated each new chemotherapy order and adjusted antiemetics including a 5HT3 antagonist, steroids, olanzapine and an NK1 inhibitor using an NCCN based protocol. Exact medications depended on insurance coverage and copays. Medications were adjusted for patient comorbidities, drug interactions and patient refusal. All patients were given prochlorperazine and ondansetron for breakthrough nausea and vomiting. Medication education to increase adherence was provided by the pharmacist, nurse and advance practice provider. Patients were evaluated by nursing or pharmacist calls on day 2 and 3 and as needed to characterize CINV using the Multinational Association of Supportive Care in Cancer (MASCC) antiemesis tool that scores CINV on a Likert scale. Based on review of the literature, a score of 3 or higher reflected symptom management failure. Additionally, due to drowsiness as the most commonly reported side effect of olanzapine, each center sought to characterize the risk of drowsiness. Results: 367 consecutive patients treated with highly emetogenic chemotherapy regimens were identified at two centers by pharmacists and followed by nursing or pharmacy calls over a 9-month timeframe: November 2016 to August 2017. 89% of patients had complete control of nausea with only 11% reporting nausea greater than a 3. Additionally, 11% of patients reported drowsiness greater than a 3. Conclusions: Administering a pharmacist driven antiemetic protocol is an effective way to control CINV. While CINV did occur, evaluation shows CINV was mild. We have shown that utilizing an antiemetic therapeutic interchange protocol with a collaborative care team including medically integrated pharmacists facilitates excellent control of CINV.