2588 Background: Optimal duration of treatment (DoT) with immune checkpoint inhibitors (ICI) in patients with metastatic melanoma (Mel) and Merkel cell carcinoma (MCC) is unclear. ICI discontinuation in Mel patients, especially those without CR, may be associated with a higher rate of progression over time, as compared to ICI continuation. Thus, extending DoT could improve outcomes. However, indefinite continuation at standard frequency doses (SFD) is not logistically or financially viable. Based on data from phase I studies suggesting sustained PD-1 receptor occupancy beyond 3 months after a single dose of nivolumab, we employed reduced frequency dosing (RFD) of anti-PD-1 antibodies every 2-3 months, to extend ICI duration beyond 2 years. Methods: This retrospective study analyzes patients in our skin cancer clinic with metastatic Mel and MCC who experienced initial clinical benefit with anti-PD-1 administered at SFD and then transitioned to RFD. We analyzed safety and efficacy endpoints including progression free survival (PFS) and rates of immune-related adverse events (irAE) with RFD. We also compared the pharmaceutical costs and patient-centered costs between 2 years of treatment at SFD versus extended DoT at RFD. Results: From 2014 – 2021, 23 patients with either metastatic Mel (N = 18) or MCC (N = 5) received anti-PD-1 therapy at RFD. Median DoT at SFD in this cohort was 1.1 years (range 0.2 – 2.2) with best objective tumor responses of CR (N =6), PR (N = 11), SD (N = 6). Median DoT at RFD was 1.2 years (range 0.2 – 3.5). The median follow-up for the entire cohort is 3.7 years (range 0.7 – 6.3) after ICI initiation. The 3-year PFS in Mel patients was 100% in those with CR (3/3), 89% with PR (8/9), and 50% with SD (3/6). The 3-year PFS in MCC was 100% in all 5 patients, including patients with CR (3/3) and with PR (2/2). Any-grade irAEs occurred in 43% of patients by 3 years on RFD, and grade 3/4 irAEs presented in 15%. Among the subset of 15 patients with DoT >2 years (median 3.4 yr, range 2.0 – 5.0), total savings amounted to $1.1 million in drugs costs and 384 hours of clinic and travel time despite the increased DoT, as compared to the calculated values for 2-year DoT at SFD. Conclusions: RFD may provide an alternative approach to extending DoT in patients receiving ICI without additional logistical and financial burden, while preserving outcomes. Efficacy and safety data suggest sustained biologic activity of ICI with RFD administration. PK/PD analyses on patient samples are ongoing to further characterize the RFD approach. The RFD approach could be utilized to expand ICI access to communities with limited healthcare resources, thereby impacting cancer outcomes at a global scale.

650 Background: Regular moderate to vigorous physical activity (MVPA) in patients with gastrointestinal (GI) cancers is associated with improved health outcomes, including greater quality of life and reduced fatigue and comorbidity. PA guidelines for cancer survivors recommend ≥150 minutes of MVPA per week for optimal health. Identifying GI oncology patients with low MVPA is important to tailor interventions aimed at increasing MVPA levels in this population. The “Exercise as a Vital Sign” (EVS) tool is a validated, two question assessment tool to quantify minutes per week (min/wk) of MVPA that was incorporated in the clinical rooming process at an academic GI oncology clinic. Here, we report characteristics of patients by amount of MVPA by EVS. Methods: Patient information was entered in the electronic medical record at a large, academic oncology clinic. Data from the first clinic visit where min/wk of MVPA EVS was collected were extracted and included demographics (e.g. age, sex, marital status, race), health status (body mass index (BMI), Eastern Cooperative Oncology Group (ECOG) performance status, comorbidities), cancer characteristics (primary tumor site, localized/metastatic, current treatment), and reason for the clinic visit. MVPA was divided into two groups based on the recommended guidelines: 0-149 min/wk (“low”) and ≥150 min/wk (“high”). Chi-square and t-tests were used to examine statistically significant differences. Results: Among 265 patients abstracted, EVS data were reported from 156 (59%) patients. Mean (standard deviation (SD)) age was 61 (13) years (range: 25-82). The primary cancer site was 35% colorectal, 23% pancreatic, 17% gastroesophageal, 10% neuroendocrine, and 15% other GI cancers. While gastroesophageal cancer patients reported high amounts of MVPA (mean (SD) = 268 (309) min/wk), comparisons with other cancer sites were not statistically different. Of the 156 with EVS data, 84 (54%) had low MVPA (mean (SD): 55 (49) min/wk) and 72 (46%) had high MVPA (mean (SD): 357 (244) min/wk). Demographic and cancer characteristics were similar across the two groups. Patients with high MVPA were more likely to have BMI 18.5-24.9 kg/m2 (56% vs. 31%) and less likely to have BMI ≥25 kg/m2 (40% vs.61%) compared to patients with low MVPA (p = 0.02). Patients with high vs. low MVPA were also more likely to have excellent performance status (ECOG 0; 47% vs. 24%, p < 0.001). Conclusions: In this study, GI oncology patients not meeting PA guidelines for cancer survivors were more likely to be overweight/obese and have worse performance status. The EVS tool was an inexpensive and accessible measure of MVPA that can be used to identify patients with low MVPA for interventions targeted to improve health outcomes. A lack of statistical significance may be due to low power. At the conference, more data will be presented (̃1000 patient visits), including longitudinal data and MVPA by active chemotherapy regimen.

Myeloid/lymphoid neoplasms associated with eosinophilia (MLN-Eo) and tyrosine kinase fusion genes include rearrangements of PDGFRB with over 30 different partner genes, most commonly ETV6-PDGFRB. Rearrangements are evident by cytogenetic analysis and fluorescence in situ hybridization (FISH) break-apart probes. Herein, we describe a novel rearrangement partner for PDGFRB. A 25-year-old man with a medical history significant for gout and hypertension presented to medical care after several days of progressive dyspnea, chest pain, fatigue, epistaxis, and rash. Physical exam revealed palpable cervical, axillary, and inguinal lymphadenopathy and massive splenomegaly. By CT scan the spleen measured 25 cm and diffuse lymphadenopathy was identified in the chest, abdomen and pelvis with the largest lymph node mass measuring 8.9 x 2.5 cm. Transthoracic echocardiogram showed normal heart function and no evidence of myocardial injury. The patient subsequently developed respiratory failure requiring intubation and mechanical ventilation, shock requiring vasopressors, and renal failure requiring renal replacement therapy. Initial laboratory evaluation demonstrated a leukocytosis to 83.3 x 10 9/L with neutrophilic predominance with left shift and eosinophilia (10% of WBCs, absolute eosinophil count of 8.2 x 10 9/L); hemoglobin of 8.3 g/dL and hematocrit of 25%; and thrombocytopenia to 50 x 10 9/L. Ferritin was 660 ng/mL, lactate dehydrogenase 384 U/L, and uric acid 10.4 mg/dL. Secondary causes of eosinophilia including infection, immunodeficiency, atopy, medications, and rheumatologic disease were ruled out. The patient was positive for strongyloides IgG antibodies and received ivermectin. Core biopsy of a lymph node was non-diagnostic and could not repeated safely due to progressive severe thrombocytopenia. Skin biopsy demonstrated a perivascular leukocytoclasis, but no abnormal B or T cell population. Bone marrow aspirate and biopsy revealed a hypercellular marrow (90%) with mild trilineage dysplasia, granulocytic hyperplasia, moderate reticulin fibrosis, and slightly increased mast cells. Blasts were not increased and no hemophagocytic histiocytes were identified. Eosinophils accounted for 6.9% of myeloid cells. Flow cytometry did not reveal an abnormal B or T lymphoid, blast or myeloid population. Peripheral blood flow cytometry did not reveal an abnormal B or T-cell population and T-cell rearrangement studies showed an oligoclonal population. No clonal B cell population was identified. FISH, which identified a rearrangement of PDGFRB in 28% of cells, confirmed the diagnosis. Karyotype analysis revealed an abnormal male karyotype with a paracentric inversion of 5q with breakpoints at 5q32 and 5q35. FusionPlex® Solid Tumor Panel (Clinical Genomics Laboratory, University of Washington SOM) identified a novel fusion between RUFY1 and PDGFRB. RUFY1 is expressed in most tissues and encodes a protein that contains a RUN domain and a FYVE-type zinc finger domain and plays a role in early endosomal trafficking (Figure). We are currently demonstrating the fusion is oncogenic and sensitive to tyrosine kinase inhibition in Ba/F3 cells. The patient's remarkable clinical response, however, supports imatinib sensitivity and is in keeping with other published reports of MLN-Eo with PDGFRB rearrangements. Initial treatment included high dose steroids for 7 days and hydroxyurea briefly. Imatinib 400 mg daily was started, then decreased to 200 mg daily due to gastrointestinal toxicity. Follow-up is limited (8 weeks). However, the patient has recovered clinically. Six weeks after starting imatinib he had renal recovery and has achieved a complete hematologic response except for mild anemia due to resolving acute kidney injury. Splenomegaly and peripheral lymphadenopathy have resolved. PDGFRB FISH has declined to 9%. Inversions of chromosome 5 have very rarely been described in association with eosinophilia, but only one report has specifically identified a rearrangement of PDGFRB. This report described a novel rearrangement involving exons 12 to 23 of PDGFRB at 5q32 and the 5′ (UTR) of G3BP1 at 5q33.1. Notably, this specific rearrangement was undetectable by PDGFRB break-apart FISH. These reports and ours highlight the clinical relevance of NGS in identifying novel rearrangement partners. Furthermore, patient-specific ddPCR assays can be designed for sensitive monitoring. Figure 1 Figure 1. Disclosures Oehler: BMS: Consultancy; OncLive: Honoraria; Pfizer: Research Funding; Takeda: Consultancy; Blueprint Medicines: Consultancy.

No abstract