Search

PURPOSE To review the complex concerns of oncofertility created through increased cancer survivorship and the long-term effects of cancer treatment in young adults. DESIGN Review chemotherapy-induced ovarian dysfunction, outline how fertility may be addressed before treatment initiation, and discuss barriers to oncofertility treatment and guidelines for oncologists to provide this care to their patients. CONCLUSION In women of childbearing potential, ovarian dysfunction resulting from cancer therapy has profound short- and long-term implications. Ovarian dysfunction can manifest as menstrual abnormalities, hot flashes, night sweats, impaired fertility, and in the long term, increased cardiovascular risk, bone mineral density loss, and cognitive deficits. The risk of ovarian dysfunction varies between drug classes, number of received lines of therapy, chemotherapy dosage, patient age, and baseline fertility status. Currently, there is no standard clinical practice to evaluate patients for their risk of developing ovarian dysfunction with systemic therapy or means to address hormonal fluctuations during treatment. This review provides a clinical guide to obtain a baseline fertility assessment and facilitate fertility preservation discussions.

PURPOSE Ongoing supportive care using electronic health (eHealth) interventions has the potential to provide remote support and improve health outcomes for patients with breast cancer. This study aimed to evaluate the effectiveness of eHealth interventions on patient-reported outcomes (quality of life [QOL], self-efficacy, and mental or physical health) for patients during and after breast cancer treatment and patient-reported experience measures (acceptability and engagement). METHODS Systematic review with meta-analyses (random-effects model) of randomized controlled trials was conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. Nine databases were searched using a prespecified search strategy. Patient-directed eHealth interventions for adult patients during or after active breast cancer treatment measuring QOL, self-efficacy, and mental (depressive, anxiety, and distress symptoms) or physical (physical activity, nutrition, and fatigue) health outcomes were included. Data from eligible full-text articles were independently extracted by six observers. RESULTS Thirty-two unique studies (4,790 patients) were included. All were health self-management interventions, and most were multicomponent (videos, forums, and electronic reminder systems) websites. Meta-analyses revealed a significant effect of eHealth interventions on QOL (standardized mean difference [SMD], 0.20 [95% CI, 0.03 to 0.36]), self-efficacy (SMD, 0.45 [95% CI, 0.24 to 0.65]), distress (SMD, –0.41 [95% CI,–0.63 to –0.20]), and fatigue (SMD, –0.37 [95% CI, –0.61 to –0.13]). Twenty-five studies (78.1%) measured patient-reported experience measures. Acceptability (n = 9) was high, with high ratings for satisfaction (range, 71%-100%), usefulness (range, 71%-95%), and ease-of-use (range, 73%-92%). Engagement (n = 25) decreased over time, but disease-focused information and interactive support were most engaging. CONCLUSION eHealth interventions may provide an acceptable and effective strategy for improving QOL, distress, self-efficacy, and fatigue among patients with breast cancer.

PURPOSEEffective treatment options are limited for patients with advanced (metastatic or unresectable) melanoma who progress after immune checkpoint inhibitors and targeted therapies. Adoptive cell therapy using tumor-infiltrating lymphocytes has demonstrated efficacy in advanced melanoma. Lifileucel is an autologous, centrally manufactured tumor-infiltrating lymphocyte product.METHODSWe conducted a phase II open-label, single-arm, multicenter study in patients with advanced melanoma who had been previously treated with checkpoint inhibitor(s) and BRAF ± MEK targeted agents. Lifileucel was produced from harvested tumor specimens in central Good Manufacturing Practice facilities using a streamlined 22-day process. Patients received a nonmyeloablative lymphodepletion regimen, a single infusion of lifileucel, and up to six doses of high-dose interleukin-2. The primary end point was investigator-assessed objective response rate (ORR) per RECIST, version 1.1.RESULTSSixty-six patients received a mean of 3.3 prior therapies (anti–programmed death 1 [PD-1] or programmed death ligand 1 [PD-L1]: 100%; anticytotoxic T-lymphocyte-associated protein-4: 80%; BRAF ± MEK inhibitor: 23%). The ORR was 36% (95% CI, 25 to 49), with two complete responses and 22 partial responses. Disease control rate was 80% (95% CI, 69 to 89). Median duration of response was not reached after 18.7-month median study follow-up (range, 0.2-34.1 months). In the primary refractory to anti–PD-1 or PD-L1 therapy subset, the ORR and disease control rate were 41% (95% CI, 26 to 57) and 81% (95% CI, 66 to 91), respectively. Safety profile was consistent with known adverse events associated with nonmyeloablative lymphodepletion and interleukin-2.CONCLUSIONLifileucel demonstrated durable responses and addresses a major unmet need in patients with metastatic melanoma with limited treatment options after approved therapy, including the primary refractory to anti–PD-1 or PD-L1 therapy subset.

PURPOSE Patients with cancer experience high rates of morbidity and unplanned health care utilization and may benefit from new models of care. We evaluated an adult oncology hospital at home program's rate of unplanned hospitalizations and health care costs and secondarily, emergency department (ED) use, length of hospital stays, and intensive care unit (ICU) admissions during the 30 days after enrollment. METHODS We conducted a prospective, nonrandomized, real-world cohort comparison of 367 hospitalized patients with cancer—169 patients consecutively admitted after hospital discharge to Huntsman at Home (HH), a hospital-at-home program, compared with 198 usual care patients concurrently identified at hospital discharge. All patients met clinical criteria for HH admission, but those in usual care lived outside the HH service area. Primary outcomes were the number of unplanned hospitalizations and costs during the 30 days after enrollment. Secondary outcomes included length of hospital stays, ICU admissions, and ED visits during the 30 days after enrollment. RESULTS Groups were comparable except that more women received HH care. In propensity-weighted analyses, the odds of unplanned hospitalizations was reduced in the HH group by 55% (odds ratio, 0.45, 95% CI, 0.29 to 0.70; P < .001) and health care costs were 47% lower (mean cost ratio, 0.53; 95% CI, 0.39 to 0.72; P < .001) over the 30-day period. Secondary outcomes also favored HH. Total hospital stay days were reduced by 1.1 days ( P = .004) and ED visits were reduced by 45% (odds ratio, 0.55; 95% CI, 0.33 to 0.92; P = .022). There was no evidence of a difference in ICU admissions ( P = .972). CONCLUSION This oncology hospital at home program shows initial promise as a model for oncology care that may lower unplanned health care utilization and health care costs.

PURPOSE Presence of teratoma in patients with metastatic testicular germ cell tumor (GCT) is of unknown prognostic significance. We report survival outcomes of patients with or without teratoma in primary tumor and postchemotherapy retroperitoneal lymph node dissection (PC-RPLND) specimen and assess impact on prognosis. PATIENTS AND METHODS Patients with metastatic nonseminomatous GCT (NSGCT) who were evaluated at Indiana University between 1990 and 2016 and had primary testicular tumor specimen from orchiectomy (ORCH) were included. All patients were treated with cisplatin-based combination chemotherapy. The cohort was divided into 2 groups according to presence or absence of teratoma in ORCH specimen. Survival data were correlated with histopathologic findings. Differences in progression-free (PFS) and overall survival (OS) were evaluated using log-rank tests and Cox proportional hazards models to adjust for known adverse prognostic factors. RESULTS We identified 1,224 consecutive patients evaluated at Indiana University between 1990 and 2016 who met inclusion criteria. Median age was 27 years (range, 13-71 years); 689 patients had teratoma in ORCH specimen, and 535 did not. With median follow-up of 2.3 years, 5-year PFS was 61.9% (95% CI, 57.1% to 66.2%) for those with teratoma versus 63.1% (95% CI, 58.0% to 67.8%) for those without ( P = .66); 5-year OS was 82.2% (95% CI, 77.9% to 85.8%) versus 81.4% (95% CI, 76.5% to 85.3%; P = .91), respectively. A total of 473 patients underwent PC-RPLND; 5-year PFS for patients with pure teratoma in PC-RPLND specimen versus necrosis only was 65.9% versus 79.1% ( P = .06), and 5-year OS was 90.3% versus 93.4% ( P = .21), respectively. CONCLUSION Presence of teratoma in ORCH and PC-RPLND specimens was not a prognostic factor in this large retrospective study of patients with NSGCT.

93 Background: Clinical trials (CT) are important treatment options for patients with cancer, yet enrollment rates among underrepresented minority (URM) patients remain suboptimal. Oncology care teams need to assess barriers and facilitators of CT participation and identify practices and resources to better support patients. As part of a larger mixed-methods project, we elicited perspectives on improving URM CT enrollment from oncology research care teams and patients. Methods: We conducted four 60-minute focus groups with 12 oncology physicians, 12 research nurses, and nine clinical research coordinators, and semi-structured interviews with nine URM CT patients at a large academic medical center between January and December 2021. Results: Thematic analysis of the focus groups and interviews identified multiple barriers and potential resources and supports at the patient, healthcare team, institutional, and trial design levels. Barriers included difficulty ensuring patient understanding and informed consent, especially among patients with low health literacy and limited-English proficiency, complex logistical and financial demands of CT participation for patients, and the lack of multidisciplinary oncology care team collaboration. Collectively, these barriers undermined communication, trust, and the quality of patients’ relationships with the care team, all affecting CT participation. Suggested resources and practices included proactive needs assessments for all patients with early engagement of social workers, providing a liaison or navigator for each patient, services and support to reduce patient out of pocket costs, expansion of non-English materials availability and increased used of interpreters, increased training and diversity for all care team roles, and simplifying CT requirements by streamlining informed consent documents, eliminating unnecessary CT-related appointments, and broadening eligibility criteria. Conclusions: Findings suggest that changes in clinical trial design, care team coordination, and early assessment and monitoring of patients’ needs and experiences may help reduce access barriers and increase enrollment of URM patients into cancer CTs.

TPS6111 Background: Treatment options for recurrent and/or metastatic (R/M) salivary gland cancer (SGC) are limited with response rates to standard cytotoxic chemotherapy (CT) low at 10-30%, and responses to single-agent immune checkpoint inhibition (ICI)

e18027 Background: Response rates of immune checkpoint inhibitor (ICI) therapy for recurrent and/or metastatic head and neck squamous cell carcinoma (R/M HNSCC) are relatively low. This study evaluates the predictive and prognostic value of weight loss and changes in body composition prior to and during ICI therapy in R/M HNSCC. Methods: This retrospective multicenter cohort study enrolled ninety-eight patients. Patient, tumor and treatment characteristics were retrieved from health records, including neutrophil and platelet-lymphocyte-ratio (NLR and PLR). Cachexia was defined according to the international consensus (Fearon et al. 2011). Skeletal muscle (SM), subcutaneous adipose tissue (SAT), and visceral adipose tissue (VAT) were evaluated on baseline and follow-up computed tomography scans at the third lumbar vertebrae level. Univariable and multivariable regression analysis were performed for six months progression free survival (PFS6m) and overall survival (OS). Results: Thirty-four patients (35%) experienced significant early weight loss (> 2%) during the first six weeks of therapy. This subgroup presented a significantly higher NLR and PLR at baseline. NLR and PLR were significantly correlated with VAT and SAT index. Independent predictors of PFS6m were lower world health organization performance status (WHO PS) (HR 0.16 [95% CI 0.04-0.54] p = 0.003), higher baseline SAT index (HR 1.045 [95% CI 1.02-1.08] p = 0.003), and early weight loss < 2% (HR 0.85 [95% CI 0.74-0.98] p = 0.03). WHO PS and baseline cachexia in combination with > 2% early weight loss were independent predictors of OS (HR 2.09 [95% CI 1.11-3.92] p = 0.02, HR 2.18 [95% CI 1.13-4.21] p = 0.02). Conclusions: The combination of cachexia at baseline and weight loss during the first six weeks of ICI therapy is associated with worse OS in HNSCC patients treated with ICI therapy. Patients with early weight loss during ICI treatment show an increase in systemic inflammation. The mechanism of how adipose tissue interferes with ICI response needs further investigation.

9553 Background: RP1 is an enhanced potency oncolytic version of HSV1 that expresses human GM-CSF and the fusogenic protein GALV-GP R-. IGNYTE is a multicohort phase 1/2 study that evaluates the safety and efficacy of RP1 in combination with nivo (NCT03767348) in a range of tumor types. Preliminary data demonstrated a durable anti-tumor activity and tolerability for RP1+nivo. Here, we present updated results from the initial and melanoma (mel) and anti-PD1 naïve non-melanoma skin cancer (NMSC) cohorts with RP1+nivo. Methods: RP1 is administered via intratumoral injection Q2W, up to 10 mL/visit, first alone at a dose of 106 PFU/mL and then starting with the 2nd dose at 107 PFU/mL in combination with nivo (240 mg IV Q2W for 4 months (mos) then 480 mg IV Q4W up to 2 yrs) for up to 8 doses, with the option to re-initiate RP-1. Eligible patients (pts) must have at least one measurable & injectable tumor of ≥ 1 cm, ECOG 0-1, and no prior oncolytic therapy. For mel, both anti-PD1 naïve and failed pts were eligible, for NMSC pts who were anti-PD1 naïve. Results: As of data extraction on January 31, 2022, 13/36 pts with mel (36.1%) and 19/31 pts with NMSC (61.3%) had a best response of PR or CR. For mel this was 5/8 (62.5%), 6/16 (37.5%), 0/6 and 2/6 (33.3%) for pts with anti-PD1 naïve cutaneous, anti-PD1/anti-PD1+anti-CTLA-4 failed cutaneous, uveal and mucosal mel respectively. For the anti-PD1 naïve NMSC this included 11/17 (64.7%), 1/4 (25%), 3/4 (75%) and 4/6 (66.6%) patients with CSCC, BCC, MCC and angiosarcoma respectively, including 8/17 (47.1%) being CR for CSCC. Current immature median DOR was 13.27 mos (current range 3.67-16.93 mos) for mel, and 7.32 mos (current range 1.88-23.11mos) for anti-PD1 naïve NMSC. Any grade TEAE ( > 25%) in all cohorts combined were fatigue, nausea, pyrexia, chills, diarrhoea, pruritus, and influenza-like illness. TEAE ≥grade 3 ( > 5%) were disease progression and fatigue. No deaths related to RP1 was observed, with one death related to nivo (myocarditis). Biomarker data from paired biopsies indicated robust T cell infiltration and an increase in tumor inflammation gene signature post-treatment. Clinical responses observed were independent of baseline tumor PD-L1 expression status. Conclusions: RP1 in combination with nivo provides a durable anti-tumor activity in pts with skin cancers, including anti-PD1 failed and anti-PD1/anti-CTLA-4 failed mel. The combination continued to be generally well tolerated with no new safety signals identified. Based on this data, enrollment into both a registration-directed cohort of pts who have anti-PD1 failed cutaneous mel (n = 125) and a cohort of pts with anti-PD1 failed NMSC (n = 30) is ongoing. Up-to-date data from this ongoing trial will be reported at the conference. A randomized Ph2 trial of RP1+cemiplimab vs. cemiplimab alone in anti-PD1 naïve NMSC is also underway (NCT04050436). Clinical trial information: NCT03767348.

e21534 Background: Nivolumab is approved in the United States and other countries as an adjuvant treatment for patients with completely resected stage III–IV melanoma based on results of the phase 3 CheckMate 238 trial, though the trial enrolled a limited number of patients with stage IIIA disease (AJCC-8). The objective of this real-world study is to describe characteristics, treatment patterns, and outcomes of patients with resected stage IIIA melanoma (AJCC-8) treated with adjuvant nivolumab or monitored with observation. Methods: In this retrospective, chart-review study, physicians from Cardinal Health’s proprietary Oncology Provider Extended Network (OPEN) extracted data from electronic health records of patients who had undergone complete surgical resection of stage IIIA melanoma between Jan. 1, 2018, and Dec. 31, 2019. Recurrence-free survival (RFS) and overall survival (OS) were evaluated from the date of resection and compared between the adjuvant nivolumab and observation cohorts using log-rank tests and adjusted Cox proportional hazards models (covariates included age, sex, race, region, payer type, ECOG PS, and Charlson Comorbidity Index). Discontinuations and deaths due to adjuvant nivolumab toxicity were assessed. Results: This study included 171 patients treated with adjuvant nivolumab and 38 patients monitored with observation. In the adjuvant nivolumab and observation cohorts, respectively, mean age was 57.4 and 68.1 years; most patients were male (59% and 68%) and white (90% and 87%); and median follow-up from resection was 20.7 and 25.0 months. Sentinel lymph node tumor burden of < 1 mm was reported in 12% (n = 20) and 16% (n = 6) of patients in the adjuvant nivolumab and observation cohorts, respectively. The scheduled treatment course with adjuvant nivolumab was completed by 91% of patients (n = 155). RFS and OS rates were numerically higher with adjuvant nivolumab than with observation (Table). There was a trend toward RFS and OS benefit with adjuvant nivolumab versus observation (unadjusted RFS HR 0.53, 95% CI 0.26–1.09; adjusted RFS HR 0.62, 95% CI 0.28–1.40; unadjusted OS HR 0.55, 95% CI 0.19–1.57; adjusted OS HR 0.81, 95% CI 0.25–2.61). Discontinuation of adjuvant nivolumab due to toxicity occurred in 2% of patients (n = 4); no treatment-related deaths were reported. Conclusions: These real-world results confirm that patients with resected stage IIIA melanoma (AJCC-8) have a good prognosis. Treatment with adjuvant nivolumab may provide modest survival benefit over observation in this population, though increased sample size and additional follow-up are warranted.[Table: see text]

e18554 Background: Low-income and minority women are less likely to receive breast cancer treatments and have higher mortality rates compared with other women.We examined economic hardship, health literacy, and numeracy by race/ethnicity and whether these factors were associated with differences in receipt of recommended treatment. Methods: We conducted a telephone survey in 2018-2020 of adult women diagnosed with stage I-III breast cancer between 2013-2016 at three centers in Boston and New York. We asked women about treatment receipt and factors contributing to decision-making. We used X2 and Fischer exact tests to examine associations between economic distress, health literacy/numeracy, and treatment receipt by race/ethnicity. Results: Among 326 respondents (AAPOR cooperation rate 63-80% across sites), 55% were Non-Hispanic (NH) White, 23% were NH Black, and 14% Hispanic; 15% were Medicaid-insured. Due to item non-response, sample sizes ranged from 294-315 per question. A substantial proportion of women, and particularly Black and Hispanic women, reported economic distress, worse finances over time, and low literacy/numeracy (Table). Overall, 7% (n = 22) did not initiate at least one recommended treatment. Although we observed no differences in treatment by race/ethnicity (p = 0.70), those not initiating recommended treatment(s) reported more worry about paying large medical bills (52% vs. 27%) and covering visit costs (27 vs. 10%); p < .05 for both. Conclusions: In a diverse sample of breast cancer survivors, financial distress was common, particularly for non-White participants; non-White women also had lower literacy/numeracy. Although we observed some associations of these factors (but not race/ethnicity) with less receipt of recommended treatments, because few women declined treatments, understanding the scope of impact is limited. However, our results highlight the importance of up-front and longer-term assessments of resource needs and allocation of support for breast cancer survivors. Novelty of this work includes the granular measures on financial distress and the focus on health literacy/numeracy among a diverse population.[Table: see text]

TPS5114 Background: Patients with high-risk scores by Decipher molecular testing after prostatectomy have a 5-year metastasis rate of 28% (Decipher 0.6-0.7) and 38% (Decipher > 0.7), likely due to micrometastatic disease. Clinical trials with intensified systemic treatment are warranted to increase cure rates and address this unmet need. Previous studies of adjuvant androgen deprivation therapy (ADT) in clinically identified high-risk disease have not demonstrated substantial benefit other than in men with lymph node positive disease. Darolutamide is a novel androgen receptor antagonist with demonstrated efficacy in improving metastasis-free survival (MFS) and overall survival (OS) in patients with non-metastatic castration-resistant prostate cancer, and OS in patients with metastatic hormone-sensitive prostate cancer (mHSPC). Whether treatment with ADT and darolutamide can increase MFS versus ADT plus placebo in the adjuvant setting for men with molecularly identified high-risk prostate cancer is unknown. Methods: Patients with CAPRA-S scores ≥3 and a PSA < 0.2 after radical prostatectomy undergo Decipher testing provided by the trial. Eligible patients with high-risk Decipher scores (> 0.6) will be randomized to treatment with ADT with darolutamide or placebo for 12 months. Patients are stratified by intent to deliver adjuvant radiation and by baseline PSA (undetectable vs detectable but < 0.2 ng/mL). The primary endpoint is MFS defined by novel PET or conventional imaging. With a sample size of 810 patients, the trial has 80% power with one-sided alpha = 0.025 to detect a HR of 0.60 for the experimental arm vs control arm for the primary endpoint. Secondary endpoints include recurrence-free survival, event-free survival, and quality of life (FACT-P, FACT-Cog, and FACIT-Fatigue), overall survival, and other disease-related outcomes. Trial was activated on December 9, 2020, and is currently enrolling patients. Clinical trial information: NCT04484818.

5 Background: Hospital at Home provides acute hospital level care to patients in their homes, diminishing the need for inpatient care or emergency department (ED) use. It has been evaluated for a variety of acute and chronic conditions but not in adult oncology. Results from program evaluations have demonstrated equivalent or improved outcomes with lower health care utilization and costs and high patient satisfaction when compared to usual care. The first evaluation of an oncology Hospital at Home program in the United States, Huntsman at Home (HH) was launched in 2018 to provide acute level care after hospitalization and address emergent symptoms that arise in the home between clinic visits. We describe the structure and patient profile of HH over the first 15 months of service. Methods: We compiled data for the 169 cancer patients admitted to HH between 8/2018 and 10/2019. We examined patient characteristics, the reason for referral, team structure, services provided and provider visit pattern. Results: The 169 patients referred to HH were predominately female (62%), white (83%), with metastatic disease (79%) and an average age 62 years. The most common diagnoses were GI, lung, GU and GYN cancer. All required acute level medical care and were referred by their oncology or inpatient provider after hospitalization. The HH multidisciplinary team was led by 9 nurse practitioners (NP) partnering with palliative care hospitalists, and the patient’s oncology team. Registered nurses (RN) and other home health personnel were provided by a community-based home health agency. The most common reasons for referral were acute pain, dehydration and electrolyte imbalance, infection, unstable symptoms, failure to thrive and post-surgical needs. The most common interventions provided by HH were: assessment and management of symptoms, medication monitoring and titration, monitoring lab chemistries, intravenous management and administration of medications and fluids, and wound, drain and line assessment and care. Home visits varied based on need; generally, there were 3 NP visits with 5 RN visits in the first week. Health care utilization and cost outcomes were recently reported at the ASCO Virtual Scientific meeting and demonstrated HH patients had fewer hospitalizations, shorter length of stay, decreased ED use and lower costs than the comparison group. Conclusions: HH has proven to be an effective and efficient way to successfully manage a variety of oncology acute care needs in patients’ homes averting further hospitalization or ED use. Hospital at home shows promise as an oncology care delivery model.

10045 Background: ICIs have transformed MM mortality. Pts receiving ICIs may experience high-grade irAEs that limit continuation of treatment per current guidelines. We aimed to evaluate the safety and response rate of ICI rechallenge. Methods: 551 MM pts treated with ICI were retrospectively reviewed from Jan 2014 to Jan 2020 after IRB approval. The incidence of a recurrent irAE in pts with ICI rechallenge within the same drug class after an initial high-grade (Grade III/IV) irAE was evaluated. Age, gender, irAEs, and outcomes were descriptively analyzed within the rechallenged cohort. Results: 32.7% of pts (180/551) experienced a high-grade irAE. 60.0% of these (108/180) pts were on combination therapy with at least one ICI. 50.6% (91/180) of pts were rechallenged with ICI within the same drug class. The rechallenged cohort had a median age of 63.8 [range: 28-86] years and 48.4% was female. The cohort’s initial irAE occurred at a median of 7.6 weeks from treatment onset with Grade 3/4 severity of 60.0% /40.0% (91). Toxicities included colitis 27.5% (25/91), hepatitis 23.1% (21/91), skin toxicity 22.0% (20/91), adrenal insufficiency 5.5% (5/91) hypophysitis 5.5% (5/91), neurological abnormality 4.4% (4/91), pancreatitis 3.3% (3/91), hematological abnormality 3.3% (3/91), arthralgia 3.3% (3/91), myalgia 3.3% (3/91), pneumonitis 2.2% (2/91), insulin dependent diabetes 1.1% (1/91), fatigue 1.1% (1/91), vasculitis 1.1% (1/91), and hyponatremia 1.1% (1/91). ICI rechallenge occurred at a median of 9.7 weeks from the first Grade 3/4 irAE. 51.8% (29/56) pts initially treated with combo were rechallenged with combo, while 48.2% (27/56) were rechallenged with single agent ICI. Of pts initially treated with single ICI, 60% (21/35) were rechallenged with single agent ICI and 40% (14/35) with combo. With a median follow-up of 21.1 months after rechallenge, irAEs occurred in 75.8% (69/91), with 44.9% of irAEs (31/69) presenting as a different type from the initial event and 31.9% (22/69) as high-grade events. There were no rechallenge irAE-related deaths. Within the rechallenge cohort, 39.6% (36/91) of pts had disease progression. Clinical benefit was achieved in 60.4% (55/91) of pts: 40.7% (37/91) complete response, 11.0% (10/91) partial response and 8.8% (8/91) stable disease. Conclusions: ICI rechallenge can be safely administered in pts with MM after recovery from an initial high-grade irAE. Rechallenge irAE’s did not always reflect initial irAE’s. Close monitoring for any type or grade of IRAE is recommended.

e22019 Background: Immune-checkpoint inhibitors (ICI) have dramatically altered the prognosis of metastatic melanoma (MM); however, fifty percent of patients will not respond to ICI. For these patients, the next choice of treatment includes targeted therapy or a clinical trial if eligible. If these treatment choices have already been utilized or are not available to the patient, there may be value in attempting a course of salvage chemotherapy (CTX). Limited clinical trial evidence has demonstrated unexpected efficacy of CTX after prior progression on ipilimumab, offering higher disease control rates than expected from what is seen in first-line chemotherapy. The phenomenon of a “priming” effect of ICI on CTX efficacy has been shown in patients of various solid tumors after progression on anti-PD1/PD-L1 therapy. The purpose of this retrospective analysis is to evaluate the efficacy of salvage CTX after prior ICI therapy for patients with MM. Patients with ocular melanoma were excluded, as this tumor subtype is known to have reduced response to immunotherapy. Methods: By retrospective analysis, patients were included under an IRB approved waiver of consent. We identified patients with MM treated with ICI therapy between Jan, 2011 and July, 2019 who were subsequently treated with salvage CTX as a result of progression of disease (POD). Salvage CTX included dacarbazine, carboplatin, temozolomide, paclitaxel, or a combination. We assessed response rate, duration of response, and time to progression (TTP) from the onset of salvage CTX. Results: A total of 22 patients who satisfied the above criteria were identified. The majority of this population had a course of single agent ICI as well as a course of combination ICI prior to salvage CTX (72.7%, n = 16/22). 13 (59.1%) patients had POD on salvage CTX with a median TTP of 10.9 weeks. 9 (40.9%) patients responded to salvage CTX. 3 (13.6%) patients achieved a complete response, 4 (18.2%) patients achieved a partial response and 2 (9.1%) patients achieved stable disease. Mean durability of response was 53.6 weeks, ranging from 7-194 weeks. Conclusions: ICI “priming” prior to salvage CTX efficacy may improve disease responsiveness to CTX. This sequence of therapy may offer patients another reasonable treatment option. Despite the small sample size of this study, a prospective clinical trial in MM exploring CTX following ICI progression should be considered.

TPS12126 Background: The undertreatment of pain in patients with advanced or metastatic cancer is well described in cancer research. Overcoming barriers that prevent successful use of opioid analgesics for cancer pain requires a clear understanding of how individuals use oral medications at home. The Proteus Discover is a digital medicine program (DMP) consisting of an FDA-approved ingestible sensor made of dietary minerals co-encapsulated with patients’ medications, a wearable sensor patch, and a mobile device app that enables patients to electronically transmit their medication adherence patterns. Use of the DMP has demonstrated improved clinical outcomes vs. usual care in patients with diabetes and hypertension, shown superiority over directly-observed therapy in tuberculosis and has been studied in the treatment of patients with hepatitis C, HIV, cancer and severe mental illness, but it has not been previously studied with opioids or in monitoring cancer-related pain. Methods: We are conducting a multicenter pilot study at SWOG NCORP sites to test the feasibility of using the DMP to monitor opioid use in the treatment of metastatic cancer pain. Eligible patients must have a diagnosis of metastatic cancer, have a baseline Brief Pain Inventory worst pain score of ≥3, be deemed by their physician to need initiation or up-titration of oxycodone-acetaminophen for pain control, and be able to read English. Primary outcomes include: (1) study accrual of 60 patients within six months of study activation at all participating sites; (2) patient retention defined as ≥50 patients completing the study, and; (3) adherence to the DMP defined as ≥66% of patients wearing the sensor patch for ≥28 days of the 42-day observation period. Secondary outcomes include change in Brief Pain Inventory pain scores, opioid medication consumption, number of safety alert triggers for high consumption, hospital or emergency room visits for pain, activity levels, and frequency of changes to the pain control regimen. The study will enroll patients at six sites; the first patient was enrolled on 1/20/2020. If successful, this study will inform design of a randomized controlled trial of the DMP vs. usual care in optimizing medication utilization and controlling cancer-related pain. Clinical trial information: NCT04194528 .

e22046 Background: Immune checkpoint inhibition (ICI) improves progression-free survival (PFS) and overall survival (OS) for patients with metastatic melanoma (MM), but treatment may cause serious immune-related adverse events (irAEs) that require the use of immunosuppressive steroids. Skin toxicity (ST), the most common side effect, has been reported to be associated with better response to ICI in some studies but not others. Herein, we tested the hypothesis that the portended survival benefit associated with ICI-related ST reaches a threshold in MM patients, dependent on the severity of the ST. Methods: We analyzed MM patients enrolled prospectively in a clinicopathological database with protocol-driven follow up and treated with ICI at NYULH. The patient and/or physician reported the ST, and an immunologist-dermatologist independently confirmed the event was ICI-related. Severe ST was defined as a skin event that required treatment with systemic steroids. Other site-specific toxicities were recorded using the CTCAE v5.0. We tested the associations between ST and PFS and OS, stratified by no, mild, and severe ST, adjusting for age, gender, number of metastatic sites, LDH, and ECOG score at treatment initiation. Results: The cohort included 256 MM patients (387 lines of ICI treatments). ST was the most common irAE (34%), and was significantly associated with development of endocrine toxicity (P = 0.007). Of the ST events (n = 130), 66% were mild and 34% were severe. Compared to none, mild ST was significantly associated with improved PFS (adjusted hazard ratio [aHR] = 0.58 [0.38, 0.89], P = 0.01), and a similar trend was observed with OS (aHR = 0.68 [0.45, 1.03], P = 0.06). However, the positive effect was reduced to insignificant (PFS P = 0.53; OS P = 0.25) in patients who developed severe ST irAEs that required systemic steroids. Conclusions: Our data demonstrate that ICI-induced ST is a complex phenomenon and might explain the discordance of reported data. While the worse outcome in patients with severe ST compared to mild may stem from steroidal immunosuppression, a threshold may also exist wherein severe ST itself might contribute to a relative reduction in ICI efficacy. More research is needed to elucidate the mechanisms that link ST and response to ICI in order to better characterize and treat skin irAEs without systemically suppressing anti-tumor immunity.

3077 Background: Immune checkpoint inhibition (ICI) has improved clinical outcomes of metastatic melanoma (MM). However, 65-80% of treated patients experience immune-related adverse events (irAEs), urging for the availability of personalized and easy-access clinical biomarkers. We have previously shown that germline genetics related to host immunity affects ICI response and MM survival. In this study, we investigated if germline immune-related expression quantitative trait loci (ieQTLs) may predict ICI-induced irAEs in MM. Methods: Through a comprehensive interrogation of a healthy twin-cohort expression dataset (MuTHER), we identified 40 ieQTLs most significantly associated with the expression of 382 immune-related genes. These germline variants were genotyped using the MassARRAY system in anti-CTLA-4-treated MM patients, collected as part of a multi-institutional collaboration. Using multivariable logistic regression models, we tested the association of 40 ieQTLs with irAEs in a discovery cohort of 97 MM patients followed by a validation in additional cohort of 97 anti-CTLA-4 treated patients. Results: We found rs7036417 significantly associated with severe anti-CTLA-4 irAEs in the discovery (OR = 6.18; 95%CI = 1.61-23.74; p = 0.007) and validation (OR = 6.73, 95%CI = 1.42-31.86; p = 0.02) cohorts. Pooled analysis showed that carriers of two rs7036417 alternate alleles (TT) have a 6-fold increased risk of developing severe irAEs (OR = 6.11; 95% = 2.26-16.56;p = 0.0003). This association was not observed with ICI response or survival. The alternate allele of rs7036417 is associated with higher expression of SYK (spleen-associated tyrosine kinase), suggesting that elevated SYK contributes to developing severe irAEs. Conclusions: We report that rs7036417, an ieQTL in SYK, associates with an increased risk of severe irAEs, independent of ICI efficacy. SYK plays an important role in B-cell/T-cell expansion and increased pSYK has been reported in patients with rheumatoid arthritis or systemic lupus erythematosus. Based on our data, the over-expression of SYK likely explains the biological mechanisms of the association between rs7036417 and anti-CTLA4 irAEs. These findings propose a novel blood-based baseline biomarker stratifying the patients at increased risk of severe irAEs, with a clinical effect substantially surpassing those observed for currently available predictors. Our ongoing studies are currently investigating SYK eQTL as a novel target in toxicity-reducing therapies.

7000 Background: Unplanned hospitalizations and emergency department (ED) visits are common during cancer care. Providing acute hospital level care at home may add value by decreasing hospital and ED use. We conducted the first evaluation of an oncology Hospital-at-Home program, Huntsman at Home (H@H). Methods: The Huntsman Cancer Institute began H@H services in 2018 and accepts referral of cancer patients for acute-medical or post-surgical care at home. Patients are admitted who require continued acute level medical care after hospitalization or have emergent unstable symptoms related to treatment or disease progression that would otherwise require ED evaluation or hospitalization. Prospectively, patients referred to H@H from 8/2018 through 10/2019 were compared to a usual care comparison group (UC) drawn concurrently from patients living within the Salt Lake City metropolitan area who qualified for admission to H@H, but lived outside the service zip codes. Probability of H@H enrollment propensity scores were constructed via random forest from patient descriptors and health care utilization at admission. We used an intent-to-treat approach for analysis. Primary outcomes were hospitalizations, length of stay (LOS), ED visits and cumulative charges over 30 and 90 days post admission to either group. Comparisons were made by generalized linear models, stratified by tertiles of H@H vs. UC propensity score. Results: 367 patients, 169 H@H and 198 UC, were evaluated. The average age was 62 yrs, 85% were Caucasian, and 77% had stage IV cancer. Propensity score distributions were overlapping, demonstrating group comparability. A variety of cancers were represented; the most common being colon, gynecologic, prostate and lung cancers. Compared to UC, H@H patients were more likely to be female (61% vs 43%) and during the month prior to admission, showed a trend towards longer LOS if hospitalized (6.7 vs 5.5 days). During the first 30 days after admission, propensity stratified comparisons showed H@H patients with lower hospital LOS (mean reduction 1.19 days, p=0.022), 56% lower odds of unplanned hospitalizations (OR 0.44, p=0.001), 45% lower odds of ED visits (OR 0.55, p=0.037) and 50% lower cumulative charges (mean ratio 0.50, p

10010 Background: Several predictors of response to checkpoint inhibitors show potential, but use pathological assays and/or molecular characterization, which limits their clinical utility outside of the academic setting. We aimed to develop a streamlined approach by leveraging information immediately available through standard care. Here, we present a computational method that integrates deep learning on histology specimens with clinicodemographic variables to predict treatment outcomes in advanced melanoma. Methods: We used hematoxylin and eosin (H&E) sections from 72 patients ( n= 153 slides) from New York University (NYU) to build a Segmentation Classifier that distinguishes tumor, lymphocyte, and connective tissue. Using pre-treatment H&E slides from 121 NYU patients ( n =302 slides), we trained a Response Classifier to predict response by selectively analyzing tumor regions. We then developed a logistic regression classifier that combines neural network output with clinicodemographic variables. The classifiers were tested on an independent cohort of 32 patients from Vanderbilt University ( n =42 slides). Area under the curve (AUC) was calculated as a measure of prediction accuracy. Results: The Segmentation Classifier distinguished tumor, lymphocyte, and connective tissue with AUCs 0.886-0.984. For the Response Classifier, optimal learning conditions were identified through training on NYU patients and testing on Vanderbilt patients (AUCs 0.685 and 0.728, respectively). The fully trained Response Classifier performed with AUC 0.711 on Vanderbilt patients. The logistic regression model performed with enhanced prediction accuracy with AUC 0.803 on NYU patients and AUC 0.793 on Vanderbilt patients. Conclusions: Histology slides and patients’ clinicodemographic characteristics are readily available through routine standard of care and have the potential to predict immunotherapy response. Our approach is time-efficient, reproducible, and requires minimal resource allocation, thus overcoming multiple common barriers to generalizability for contemporary biomarkers.

e22050 Background: RP1 is an oncolytic HSV that encodes a fusogenic GALV-GP R- protein and GM-CSF. RP1 demonstrated tolerable safety and tumor regression alone and with nivolumab (nivo) in ph 1 in patients (pts) with a number of tumor types. To further define the efficacy of the combination, ph2 cohorts of 30 pts with 4 tumor types were then opened. Initial data from the melanoma (mel) and the non-mel skin cancer (NMSC) cohorts will be presented. Methods: Unresectable stage IIIb-IV mel pts for whom anti-PD-1 was indicated or who were refractory to 1 prior standard therapy including anti-PD-1 or ipi/nivo were enrolled. NMSC pts were anti-PD1 naïve. Pts received up to 8 doses of RP1 (6 PFU/mL then 107 PFU/mL). From the second RP1 dose pts also received nivo (240 mg IV Q2W for 4 mos then 480 mg IV Q4W up to 2 yrs in the absence toxicity or confirmed progressive disease (PD)). Imaging was done every 8 wks and response assessed by RECISTv1.1 (with confirmation required for PD). Results: As of Jan 22nd 2020, 30 mel pts and 9 NMSC pts had been enrolled with follow up between