Bayne, Harry E. Fuentes

6589 Background: Real-world utilization and comparative effectiveness of direct oral anticoagulants (DOACs), low molecular weight heparin (LMWH) and warfarin for cancer-associated venous thromboembolism (VTE) treatment remains largely unexplored. Methods: De-identified administrative claims data (OptumLabs - Data Warehouse) for adult active cancer patients with acute VTE (1/1/2012-9/30/2019) were assessed for utilization patterns, recurrent VTE, and major bleeding differences between DOACs, LMWH, and warfarin. Patients were followed up till the end of treatment. Patients who crossed over to other anti-coagulants within the follow up period were excluded. Multinomial logistic regression was used to assess predictors of anticoagulant administration. Kaplan-Meier curves were used to evaluate the differences in time to medication discontinuation among the three groups. Propensity score (PS) and inverse probability of treatment weighting were used to balance baseline differences between groups. Standardized difference (SD) was used to assess the balance of covariates after weighting and SD < 10% was considered acceptable. Weighted Cox proportional hazards regression with a robust variance estimator was then used to assess outcomes in PS weighted groups. Results: A total of 5100 patients met the inclusion criteria; 49.3% filled DOACs (n = 2512), 29.2% LMWH (n = 1488), and 28.6% warfarin (n = 1460). Median treatment duration was around 3.2 months for DOACs and warfarin, and 1.8 months for LWMH ( P-value < 0.01). Multinomial regression analysis showed that younger patients were more likely to be prescribed LMWH (OR 0.97 95% CI: 0.97-0.98) as compared to DOACs. Patients with lung (OR 2.07 95% CI: 1.12-3.65; OR 1.87 95% CI: 1.04-3.37), urological (OR: 1.94 95% CI:1.08-3.49; OR: 2.04 95% CI: 1.12-3.73), gynecological (OR 4.25, 95% CI: 2.31-7.82; OR 2.31 95% CI: 1.22-4.39) and colorectal cancer (OR 2.26 95% CI: 1.20-4.32; OR 2.51; 95% CI: 1.32-4.79) were more likely to be prescribed LMWH or warfarin respectively, compared to DOACs. VTE recurrences were more frequent for patients receiving LMWH (HR: 1.47; 95% CI: 1.14-1.90) or warfarin (HR: 1.46; 95% CI: 1.13-1.87) compared to DOACs. LMWH, but not warfarin, was associated with greater major bleeding rates compared to DOACs (HR: 2.27; 95% CI: 1.62-3.20). All-cause mortality rates were also significantly higher for patients receiving LMWH, but not warfarin compared to DOACs (HR: 1.61; 95% CI: 1.15-2.25). Conclusions: Patients with cancer associated VTE remain on anti-coagulation for a remarkably short duration in real-world clinical practice. DOACs and warfarin may offer better compliance than LMWH. Patients receiving DOACs have a lower risk of VTE recurrence, less major bleeding, and improved mortality. Warfarin may still be considered for patients with contraindications to DOACs and non-compliant to LMWH.

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) <10%. Current therapeutic agents used to treat R/M SGC have an unfavorable toxicity profile including cytotoxic CTs such as cisplatin and doxorubicin and the targeted therapy lenvatinib. Treatments with general applicability and acceptable side effect profile are urgently needed. Pemetrexed (PTX) is a multi-targeted anti-folate CT with an excellent safety and toxicity profile with promising responses reported in a case series of patients (pts) with R/M SGC (Viscuse P and Price K. Marked responses to pemetrexed chemotherapy for metastatic adenocarcinoma of the parotid gland: case series. Head Neck 2019;41(6):E99-103.). As the combination of PTX and pembrolizumab (PMB) showed enhanced responses and survival in pts with lung cancer, we hypothesize that PTX and PMB will have efficacy in pts with R/M SGC and that the combination will have improved anti-cancer activity compared with historical controls. Correlative studies on archived tissues include methylthioadenosine phosphorylase loss, PDL1 expression, and thymidylate synthase expression as potential biomarkers of response. Methods: MC200708 is a single institution (3-site), open-label, single arm phase II study of PTX and PMB in patients with R/M SGC. Pts will be treated in two cohorts: one for pts with adenoid cystic carcinoma (ACC) and the other for pts with all other types of SGC (non-ACC cohort). Eligible pts must be ≥18 years old, have a pathologically confirmed SGC that is not amenable to curative-intent therapy, ECOG 0-2, adequate organ function, and at least 1 measurable lesion by RECIST 1.1. Any number of lines of prior therapy is allowed; prior treatment with checkpoint inhibitors and/or PTX is allowed. Key exclusion criteria: serious medical co-morbidity or autoimmune disease, prior grade 3-4 adverse event (AE) or grade 2-4 lung AE with ICI, and uncontrolled brain metastases. Simon’s two-stage design will be used for each cohort separately given the expected difference in responses in pts with ACC and non-ACC. Primary endpoint is overall response rate. Secondary endpoints are progression free and overall survival and toxicity. All patients will receive the same treatment of PTX 500 mg/m2 IV + PMB 200 mg IV every 3 weeks until disease progression or treatment intolerance. Response assessments will occur every 3 cycles. Enrollment began July 2021 and 16 of the planned 45 patients (ACC cohort 9 of 20 pts, non-ACC 7 of 25) have been accrued. Clinical trial information: NCT04895735.

e15118 Background: Melanoma, non-small cell lung cancer (NSCLC), colorectal adenocarcinoma and thyroid tumors frequently harbor BRAF mutations and are tumor types eligible for treatment with FDA-approved BRAF inhibitors. Although there are many ongoing clinical trials with BRAF inhibitors in various advanced tumor types, currently eligibility requires both tumor type identification and BRAF mutation confirmation. The 92-gene assay (CancerTYPE ID) is a gene expression-based classifier of 50 tumor types and subtypes for metastatic patients with diagnostic ambiguity. Multimodal biomarker testing, including NGS, IHC and FISH, enables identification of actionable biomarkers to guide targeted therapy selection. Here, a database of metastatic cases integrating tumor type with biomarker analysis was used to identify patients eligible for BRAF inhibitor treatment. Methods: MOSAIC (Molecular Synergy to Advance Individualized Cancer Care) is an IRB-approved, de-identified database of diagnostically ambiguous cases submitted for CancerTYPE ID testing with tumor type-guided multimodal biomarker testing (Neogenomics). In this study, metastatic cancers of various tumor types with BRAF mutations were identified. Results: CancerTYPE ID determined 27 different tumor types in 2929 (92.5%) of 3168 cases. BRAF mutations were identified in 87 cases across 18 tumor types and subtypes. Out of the 87 cases, 22 (25.3%) had BRAF mutations in a tumor type eligible for FDA-approved BRAF inhibitor treatment, including melanoma, NSCLC, colorectal adenocarcinoma and thyroid cancer. Another 26 (29.9%) BRAF-mutated cases were detected in a tumor type with ongoing BRAF inhibitor clinical trials, including lymphoma, pancreaticobiliary and ovarian cancer (Table). Conclusions: This analysis identified patients with metastatic cancer with actionable BRAF mutations eligible for FDA-approved therapy, and patients who may be eligible for investigational BRAF inhibitors based on tumor type and mutation status. Notably, cancer type identification is critical given that FDA-approved BRAF inhibitor treatment regimens vary based on the tumor-specific indication. These findings support the clinical utility of molecular tumor type classification by CancerTYPE ID combined with biomarker profiling to identify additional treatment options for patients with metastatic cancer.[Table: see text]