Search

TPS763 Background: A meaningful subset of PDAC is characterized by a homologous recombination deficiency (HRD). The most well-defined patients within this group are those with pathogenic variants in BRCA1, BRCA2 and PALB2. In the metastatic setting, PARP inhibitor maintenance provides a progression-free survival benefit after a period of platinum based chemotherapy1,2, but the role of PARP inhibitors in the curative intent setting is undefined. The OlympiA study established one year of olaparib as the standard of care for patients with BRCA-related, early stage breast cancer who completed all other curative-intent treatment3. Therefore, we have designed a randomized, phase II double-blind study of one year of olaparib vs placebo in patients with pancreatic cancer and a germline or somatic variant in BRCA or PALB2 who have completed all curative intent therapy. Methods: We have enrolled and treated 23 of 152 planned patients on study NCT 04858334/EA2192. Eligibility criteria include: a pathogenic germline or somatic variant in BRCA1, BRCA2 or PALB2 as determined by local laboratory (central review required); completion of curative-intent resection and ≥ three months of multi-agent chemotherapy; no evidence of recurrent disease. At enrollment, patients must be within 12 weeks of their last anti-cancer intervention. Patients are randomized 2:1 to receive oral olaparib 300 mg twice daily or placebo for 12 28-day cycles. The primary endpoint is relapse-free survival. Overall survival is a secondary endpoint. Tumor tissue, fecal material (for microbiome analysis) and serial ctDNA samples are being collected. 1.Golan T, Locker GY, Kindler HL: N Engl J Med 381:1492-1493, 2019. 2. Reiss KA, Mick R, O'Hara MH, et al: J Clin Oncol 39:2497-2505, 2021. 3. Tutt ANJ, Garber JE, Geyer CE, Jr.: N Engl J Med 385:1440, 2021. Clinical trial information: NCT04858334 .

5 Background: A growing body of evidence supports the utility of circulating tumor DNA (ctDNA) as a useful biomarker for detecting molecular residual disease (MRD) in colorectal cancer (CRC). Immediately after surgery or during adjuvant therapy, high levels of cell-free DNA (cfDNA) from normal tissue may limit the detection of tumor-derived ctDNA. The optimal timing of blood collection for reliable MRD detection after surgery or adjuvant therapy remains unclear. Methods: In this retrospective, U.S.-based, multi-institutional study, data from commercial ctDNA testing in 16,347 patients with stage I-III CRC were analyzed. Complete clinical data were available for 417 patients with 2,538 plasma samples collected between 6/2019 and 4/2022. The median follow-up for relapsed and non-relapsed patients was 730 and 615 days, respectively. A personalized, tumor-informed multiplex PCR-based next-generation sequencing assay (Signatera) was used to quantify ctDNA prior to surgery and postoperatively in a longitudinal manner. We analyzed the kinetics of total cfDNA and compared it with the ctDNA MRD positivity rates at various time points after surgery. Results: Among all patients, cfDNA levels were higher immediately after surgery (0-2 weeks) and gradually declined during the subsequent 2-8 weeks (p6 months post-operatively and subsequent to any adjuvant therapy) was significantly associated with worse recurrence-free survival as compared to ctDNA negative patients (MRD: HR 14.1, 95% CI: 5.8-34; p

735 Background: We have studied the role of KRAS mutations in relation to the prognosis in patients with advanced pancreatic ductal adenocarcinoma (PDAC). KRAS is a well-described oncogenic driver in PDAC, with mutations identified in over 90% of cases, typically involving codon 12. The three predominant missense variants include G12D, G12V and G12R. PDAC has the highest rate of G12R mutations compared to other malignancies, comprising 15-20% of KRAS-mutated tumors. This study presents a new finding in the progression of advanced PDAC utilizing a large clinical and genomic database to further characterize the clinical features of pathogenic KRAS variants in PDAC with a focus on G12R. Methods: PDAC samples were tested using whole transcriptome sequencing (WTS; Illumina NovaSeq) and NextGen DNA sequencing (NextSeq, 592 Genes and NovaSEQ, WES) at Caris Life Sciences (Phoenix, AZ). Transcriptomic signatures including MPAS (MAPK activation score), T-cell inflamed score and tumor micro environment (TME) characterization were calculated on WTS data. Significance was determined by X2 and Fisher-Exact and p adjusted for multiple comparisons (q) was < 0.05 (Benjamini-Hochberg). Real-world overall survival (rwOS) was obtained from insurance claims data and calculated from tissue collection to last contact; time-on-treatment (TOT) was calculated from start to finish of specific treatments; comparison was done by Kaplan-Meier test. Results: A total of 5,555 patients with PDAC harboring either KRAS G12D (n = 2,671), G12V (n = 1,871) G12R (n = 904) or G12C (n = 109) variants were identified. The patients with KRAS G12R mutant tumors had significantly longer median real-world overall survival (mRWOS) compared to G12D (452 vs 358 days, HR 0.82, CI 0.74 – 0.9, p < 0.0001). This difference persisted regardless of treatment with FOLFIRINOX or gemcitabine plus nab-paclitaxel. There was no difference in outcomesbetween patients with KRAS G12R, G12V or G12C. PD-L1 expression was significantly lower in G12R than in G12C or G12D (13% vs 27% vs 19%,) while the prevalence of TMB-H and dMMR was comparable across isoforms. Conclusions: Patients with KRAS G12R variants has improved rwOS compared to G12D irrespective of the chemotherapy regimen administered. Immune profiling suggested that the immune contexture in G12R-driven tumors are distinct from G12D as reflected by reduced PDL1 staining, decreased levels of multiple checkpoint receptors. We aim to further explore the molecular basis for these differences with a focus on PI3K and MAPK pathways. Based on this data, survivorship studies in patients with advanced PDAC should consider reporting KRAS mutational status.

103 Background: Novel strategies to improve the efficacy of immune checkpoint inhibitors in microsatellite stable (MSS) mCRC are needed. CV301 is a vector-based vaccine that expresses carcinoembryonic antigen (CEA) and mucin 1 (MUC1), and in a phase II study in resected hepatic limited mCRC significantly improved OS compared with unvaccinated contemporary controls. Methods: In this multi-center randomized phase II study, patients with previously untreated resectable hepatic-limited mCRC were randomized to perioperative nivolumab + mFOLFOX +/- CV301 (Arm B) with a primary endpoint of 3-year OS. Treatment included mFOLFOX-nivo (+/- CV) x 4 cycles followed by resection, then 8 more cycles of mFOLFOX-nivo followed by maintenance nivo monthly for two years in both arms, and CV boosters concurrently with mFOLFOX, and then every 3 months for two years in arm B. Secondary endpoints of ORR (following induction pre-resection), PRR, and safety were determined. Correlative analyses included immune cell quantification using Immunoscore and T-cell clonality. Results: 17 patients were enrolled prior to premature closure for slow accrual (8 arm A, 9 arm B). At the time of data cutoff, 5 patients remained on treatment and no deaths had occurred. One patient was removed from study due to protocol non-compliance. The median age was 61, majority were male (59% vs 41%), and ECOG PS 0-1 (71% 0, 17% 1). All patients had complete surgical resection. Four patients (24%) experienced a SAE related to drug. The TRAE rate was 40.3%,. No AEs delayed/prevented surgical resection. The ORR in arm A was 50% (including 4 CR) and 87.5% in arm B (including 7 CR) (p=0.129, NS). There was no significant difference in pathologic response (p=0.9047). Correlative analyses demonstrated the Immunoscore CD3/CD8 predicted response to mFOLFOX + nivolumab, but did not correlate with response to CV301, though CV301 may induce a shift to predominantly cytotoxic CD8+ T cells. While there was no significant difference in T cell repertoire, clonality, fraction (TCFr) or richness, patients in arm B had significant decreases in blood TCFr and increase in tumor TCFr with treatment; those with CR had higher TCFr and clonality. Conclusions: The addition of CV301 to perioperative nivolumab and mFOLFOX was safe, did not delay or prevent surgical resection, and gave a higher response (p=ns due to sample size). Changes in T cells suggest a vaccine response. Clinical trial information: NCT03547999 . [Table: see text]

TPS764 Background: Kirsten rat sarcoma (KRAS) G12D is a point mutation observed in various cancer types including pancreatic ductal cancer, colon adenocarcinoma, and lung cancers. ASP3082 is a novel small-molecule proteolysis-targeting chimeric degrader that binds to, and selectively targets, the KRAS G12D-mutated protein for degradation via recruitment of E3 ubiquitin ligase proteins. In preclinical studies, ASP3082 selectively degraded the KRAS G12D-mutated protein and showed growth inhibitory activity in KRAS G12D-mutated cancer cells but not in KRAS -wildtype cancer cells. Notable antitumor effects of ASP3082 have been demonstrated when intravenously administered weekly in mice xenografted with KRAS G12D-mutated cancer cells. Methods: This first-in-human, open-label, multicenter, phase 1 study evaluates the safety and tolerability of ASP3082 in patients with pancreatic cancer, colorectal cancer, non-small cell lung cancer or other solid tumors. Participants with unresectable, locally advanced, or metastatic solid tumor malignancy with documented KRAS G12D mutation are eligible for enrollment. Part 1 consists of dose-escalation cohorts of 3−12 patients receiving intravenous administration of ASP3082 in a 21-day cycle. Part 2 consists of random assignment of ≤20 patients into 2 cohorts with different ASP3082 dose levels to determine the recommended phase 2 dose. Additional tumor-specific expansion cohorts may enroll ≤20 participants per tumor type. Primary endpoints will evaluate safety and tolerability as noted by dose-limiting toxicities, adverse events, serious adverse events, laboratory test results, electrocardiograms, vital signs, physical exams, and Eastern Cooperative Oncology Group performance status. Secondary endpoints will evaluate objective response rate, duration of response, disease control rate per Response Evaluation Criteria in Solid Tumors version 1.1, pharmacokinetics of single and repeated doses of ASP3082, and changes in KRAS G12D in tumor samples. Exploratory objectives will evaluate potential biomarkers that may correlate with treatment outcomes. Tumor assessment follow-up will continue for ≤45 weeks or until progressive disease. Enrollment in Cohort 1 is complete, and enrollment to Cohort 2 began in September 2022. Clinical trial information: NCT05382559 .

TPS4181 Background: The mainstay of treatment (TX) for pts with advanced or mPDAC consists of CT, with FOLFIRINOX and gemcitabine (gem)/nab-paclitaxel currently representing the front-line standards of care. TX is generally continued until either dis progression (progr) or cumulative toxicity, with pts often reaching a plateau in response after 4-6 mos. For those who have achieved dis control (stable dis or better) on front-line CT, a maint TX strategy that can effectively delay dis progr while preserving quality of life with minimal cumulative toxicity is highly desirable. However, aside from PARP inhibition in the subset of PDAC pts with gBRCA mutated dis, there is no current standard of care in this maint setting. Ival is a pan-HDAC (histone deacetylation) inhibitor that increases histone acetylation (HA), suppresses PDAC cell proliferation, and promotes apoptosis in PDAC cell lines in a dose-dependent manner. It has demonstrated synergy with 5-FU in cholangiocarcinoma cell lines and shows promising antitumor activity when combined with cap in syngeneic PDAC mouse models. On these bases, we are conducting a ph1b/randomized ph2 trial of ival plus cap vs cap alone in the maint setting for pts with mPDAC who have not progressed on front-line FOLFIRINOX. Methods: Key eligibility criteria include pts with mPDAC; no evidence of dis progr following at least 16 wks of front-line FOLFIRINOX at full or modified doses; ECOG PS 0-1; and no known gBRCA1/2 mutation. The study includes an initial dose-esc ph1b evaluating 3 dose levels of ival, (60, 125, and 250mg/m2 iv weekly on days 1 and 8) in combination (comb) with cap (1000mg/m2 po BID on days 1-14) of a 21-day cycle, using a standard 3 + 3 dose-esc design. Of note, ival 250 mg/m2 represents the RP2D identified in prior clinical studies of this agent both as monotherapy in solid tumors and in comb with gem/erlotinib in advanced PDAC pts. In the ph2 portion, pts will be randomized 1:1 to receive either ival plus cap or cap alone, in 21-day cycles, until dis progr, with tumor assessments occurring at 6-wk intervals. Blood will be collected at pre-specified serial timepoints for pharmacodynamic assessments, including HA of PBMCs. Primary endpoint for ph2 is investigator-adjudicated PFS. The primary analysis will compare PFS distributions in the ival/cap and cap alone arms using a one-sided log rank test with an alpha = 0.10. The assumed true 6-mo PFS rates are 35% (cap), based on historic data, and 60% (ival/cap), which corresponds to an HR of 0.487. Assuming an accrual duration of 18 mos and a dropout/lost to follow-up rate of 10%, the estimated total number of pts in the randomized ph2 portion is 52 (26 per arm). Enrollment is expected to being in spring 2022 across 25 U.S. sites.

3622 Background: Colorectal cancer is the third common cancer worldwide. Radical excision (RE) as total mesorectal excision for rectal cancer carries a higher risk of mortality and morbidity, while local excision (LE) could decrease these postoperative risks. However, the long-term oncologic outcomes of LE are still debatable. We aim to study the effect of LE versus RE in T1 and T2 rectal cancer. Methods: We conducted a systematic review and meta-analysis. We searched PubMed and CENTRAL databases, using an optimized search-strategy from inception until 15 June 2021, without restriction on publication date or status. We included only cohort and randomized controlled trials (RCTs). Two authors independently screened the title, abstracts, and full-text manuscripts for inclusion and data extraction. All included trials contained at least one of the primary outcomes. We used RevMan 5.4 tool for data analysis. We calculated both hazards ratio (HR) and risk ratio (RR) for the 5-years survival analyses, with their 95% confidence intervals (CI). We assessed both clinical and statistical heterogeneity of the studies; I2 >75% was considered highly heterogeneous. We used random effect model (REM). We used standardized mean difference (SMD) for hospitalization days. We conducted a subgroup analysis of patients with T1-only without adjuvant chemo/radiotherapy (CRT). Results: We retrieved from the search a total of 1243 reports. A total of 18 studies were included for final meta-analysis (4 RCTs and 14 retrospective cohorts). Nine studies were multi-central while ten were unicentral studies. We did not find any difference in risk ratio (RR) between overall survival (OVS) and disease-free survival (DFS). But there were higher HRs in OVS and DFS with LE as compared to RE. A higher recurrence rate was also seen with LE. Six studies showed absent 30-days postoperative mortality in both groups so we used peto-odds ratio. Postoperative mortality and morbidity were lower with LE rather than RE. Conclusions: LE for early stage rectal cancer has a higher risk of decreased 5-year OVS and DFS than RE, with higher local recurrence rate. However, LE is associated with lower early postoperative mortality, morbidity, and hospitalization days, as compared to RE. Patient selection is key to balance these risks for the optimal outcome. [Table: see text]

4136 Background: Mesothelin (MSLN) is overexpressed in 80-85% of pancreatic adenocarcinomas (PDAC). Anetumab ravtansine (AR) is a fully human anti-MSLN immunoglobulin G1 antibody conjugated to the anti-tubulin maytansinoid DM4. Through NCI-ETCTN, the North American Star Consortium conducted a phase I study to evaluate the safety/tolerability of AR in various combinations in patients (pts) with PDAC. Here, we report preliminary results of the escalation part. Methods: Pts with advanced PDAC after at least one line of treatment were included. AR was combined with nivolumab (ARM1), nivolumab/ipilimumab (ARM 2), or nivolumab plus gemcitabine (Gem) (ARM3), using an integrated biomarker analysis. Two dose levels (DL) of AR were evaluated, DL1=5.5mg/kg and DL2=6.5mg/kg (established RP2D). Key eligibility criterion was MSLN expression in >5% of tumor cells by immunohistochemistry. Pts with prior anti-PD1/anti-CTLA4 treatment were excluded but treatment with prior Gem was allowed. Mandatory blood and paired tumor samples were collected for investigation of the immune microenvironment, genomic/transcriptomic changes and for an in-depth description of AR pharmacokinetics. Results: Data cut-off date was 22/01/2022. A total of n=33 pts were enrolled, n=11 (ARM 1), n=13 (ARM 2) and n=9 (ARM3). Median age of pts was 66 (40-83), 33% of PS=0 and 66% of PS=1. Twenty-six pts (79%) had previously been exposed to Gem. Median number of prior lines of treatment was 3 (1-7). Twenty-eight patients were evaluable for DLT. Grade (G)3/4 TRAEs: 0% in ARM1DL1, 5.3% in ARM1DL2, 0% in ARM2DL1, 16.9% in ARM2DL2, 8.6% in ARM3DL1 and 19.5% in ARM3DL2. There were 2 dose-limiting toxicities in ARM2DL2, one G3 upper gastrointestinal haemorrhage, possibly related to AR, and one G3 thrombocytopenia and G3 anaemia, definitely related to AR. Ocular toxicity events were G1/2 blurred vision in 5/33 (15%) and G1 xerophthalmia in 1/33 (3%), related to both AR and anti-PD1; G2 keratitis in 1/33 (3%), related to AR only. Only G1 peripheral neuropathy was observed in 4/33 (12%) pts. Efficacy data is presented in the table. In ARM3, the range of tumor measurement (ΤΜ) change was ΔTM=–16.8% to +16.2% and 3/8 (36%) pts with SD had previously been exposed to Gem. Conclusions: Based on the observed disease control rate and acceptable toleratbility, ARM3 (both DL1 and DL2) will be tested in the expansion part. A further 20 patients will be recruited for dose confirmation and comprehensive biomarker evaluation. Pharmacokinetic/pharmacodynamic analysis is under way. Clinical trial information: NCT03816358. [Table: see text]

3623 Background: Minimal residual disease (MRD) assessment may effectively detect cure in patients with colorectal cancer (CRC). Emerging data have suggested that circulating tumor DNA (ctDNA) can be a reliable biomarker for MRD in patients. Recent studies have shown the ability to detect MRD using ctDNA assay to assess recurrence risk and patient selection for adjuvant chemotherapy. We performed a systematic review and metanalysis of post operative ctDNA in Stage I-IV (oligometastatic) CRC patients after curative intent resection with or without adjuvant chemotherapy. Methods: We searched PubMed/Medline, EMBASE, Web of Science, Cochrane Library, and Google from inception to February 3, 2022 using keywords related to colorectal cancer, ctDNA, and MRD. The search also includes paper and conference presentations. The search resulted in 427 studies after removing the duplicates. Data were extracted to perform a meta-analysis using RevMan 5.4. software. Subsequent subgroup analysis was performed for stages I-III and oligometastatic stage IV CRC patients. Results: After the initial abstract screening, 48 studies were identified. The final review led to the inclusion of 27 studies representing 3459 patients with evaluable ctDNA, with 623 having +ve post-surgery ctDNA levels. Seven studies had analyses of patients specifically with oligometastatic disease, the rest 20 studies subdivided within stages I-III. The pooled HR for RFS in post-surgical ctDNA +ve vs -ve patients in all stages was 7.16 (95% CI 6.12-8.36) p

6542 Background: Access to health care including clinical trials (CT) leading to paradigm-changing cancer treatments are critical for high quality cancer care and equity in society. In this report, we highlight methods in accruing to ETCTN wherein underrepresented rural, low-income, and racial minorities comprise >50% of enrollment. Methods: University of Kansas Cancer Center (KUCC) is one of eight National Cancer Institute (NCI) designated cancer centers awarded CATCH-UP.2020 (CATCH-UP), a congressionally mandated P30 supplement to enhance access for minority/underserved populations to ETCTN precision medicine CT. KUCC catchment area is 23% rural by Rural Urban Continuum Codes (RUCC); almost 90 % of counties are designated primary care HPSA’s (Health Professional Shortage Areas). KUCC Early Phase and Masonic Cancer Alliance (rural outreach network) partnered to operationalize CATCH-UP. We engaged disease-focused champion investigators in disease working groups and MCA physicians who selected scientifically sound CT that fit catchment area needs. Patient and Investigator Voices Organizing Together, a patient research advocacy group provided practical feedback. MCA navigator coordinated recruitment. Telehealth was used for rural patients that would have a significant distance to travel just to be screened. Results: CATCH-UP was initiated in September 2020. Twenty-eight CT were activated, many in community sites. Average activation time was 81 days. Delays were mainly from CT amendments. KUCC enrolled the first patient in the CATCH-UP program. In 6 months, we met accrual requirements (24/year, 50% minorities). During first year, we enrolled 47 (>50% minorities), an increase of 680% from our average accrual of 6/year (>50% minorities) in ETCTN through Early Drug Development Opportunity Program (2016-2020). To date, we have enrolled 61, 54% from rural, HPSA, race and other minorities. Although the proportion of minorities did not change but remained high, this funding allowed us to substantially increase the number of patients from a catchment area with high proportion of geographically and socioeconomically underserved minorities given access to early phase CT through ETCTN. Conclusions: Amid COVID-19 pandemic, the NCI CATCH-UP program and methods we used allowed access to novel therapies for rural, medically underserved, and other minority groups. Funded by NIH: 3P30CA168524-09S2.

e16254 Background: Neo-adjuvant therapy (NAT) and associated pathologic complete response (pCR) rates have correlated with improved survival in resected pancreatic ductal adenocarcinoma (PDAC). In this study, we explored the relationship between pathologic response, peri-operative therapy, and survival, especially the impact of change in adjuvant therapy in patients with no/poor path response to NAT. Methods: Retrospectively reviewed 66 PDAC patients who received NAT ± radiation and underwent resection at KU Cancer Center between 2011-2022. We compared DFS and OS between Path Responders vs Non-Responders based on standard Tumor Regression Scores from pathology reports. A subanalysis was performed in path non-responders based on switch in adjuvant therapy (AT) versus not. Results: Patient characteristics are summarized in the table. Among 66 PDAC patients, 50 (75.8%) achieved a path response (G0-G2), 16 (24.2%) experienced no/poor path response (G3). Of the 50 pts who achieved a path response, 4 (8.0%) had a complete path response (pCR; G0), 5 (10%) marked response (G1), 41 (82%) moderate response (G2). Median DFS (mDFS) was 17.3 months (95% CI: 12.7-22.4) in Path Responders vs 15.9m (95% CI: 9.6-35.8) in Non-Responders [p=0.59]. Median OS (mOS) was 32.9m (95% CI: 23.4-41.5) vs 27.7m (95% CI: 15.2-38.2), respectively [p=0.39). A sub-analysis in the Non-Responders (n=16) based on switch in AT (n=8) vs not (n=3), revealed mDFS 16.4m (95% CI: 9.6-41.8) when AT was switched vs mDFS 11.3m (95% CI: 5.9-16.6) when AT was not switched [p=0.24]; and mOS 30.6m (95% CI: 15.7-60.3) vs 17.2 months (95% CI: 6.7-27.7), respectively [p=0.18]. Conclusions: Our study found no statistical difference in DFS and OS between Pathologic Responders and Non-Responders to neo-adjuvant therapy. However, a sub-analysis within Pathologic Non-Responders revealed a longer DFS and OS after switching adjuvant therapy without reaching statistical significance, likely due to small sample size. Our findings warrant validation in a larger cohort as switch in adjuvant therapy could potentially change the treatment landscape for Pathologic Non-Responders.[Table: see text]

TPS4195 Background: The efficacy of chemotherapy is often compromised due to poor penetration of drugs in solid tumors. The tumor microenvironment, which is characterized by dense extracellular matrix‐rich stroma that creates a physical barrier to penetration of anti‐cancer drugs, is especially pronounced in Pancreatic Ductal Adenocarcinoma (PDAC) and in peritoneal metastases from Colorectal/Appendiceal Adenocarcinoma. CEND‐1 is a tumor‐penetrating peptide (scientifically also known as iRGD) that has preclinically demonstrated to enhance the tumor penetration of chemotherapy agents through binding and activation of alphav-integrins and neuropilin‐1 (NRP-1). The 2-step mechanism leads to a higher delivery and concentration of chemotherapeutics selectively in the tumor, while sparing normal tissue. Hence CEND-1 therapy has the potential to improve the efficacy of anti‐cancer therapies and reduce side effects through increased tumor access, specificity, and sensitivity. We hypothesize that CEND‐1 may become a powerful adjuvant that safely enhances standard anti‐neoplastic therapy in the neoadjuvant setting for the above populations. Methods: A safety lead-in 6-9 patients (Phase Ib) will be followed by an open label, single arm, parallel (3 cohorts) Phase IIa study. A total of 50 patients (20 PDAC, 15 colorectal/appendiceal with peritoneal metastases, 15 oligometastatic colorectal) will be enrolled. A starting CEND-1 dose of 3.2 mg/kg in combination with the standard doses of FOLFIRINOX (+/- Panitumumab if RAS/RAF wild type) will be used for the safety lead-in. CEND-1 dose will be lowered for Phase IIa if > 1/6 patients experienced DLTs. Participants enrolled will receive standard doses of FOLFIRINOX q2w +/- Panitumumab q2w 6mg/kg IV q2w (14-day cycles) for Cycles 1-3. After a subsequent research biopsy, the CEND-1 + chemotherapy combo will be continued at RP2D q2w for cycles 4-6, followed by CEND-1 +/- Panitumumab ̃72h prior to resection. Assessment of tumor response using RECIST v1.1 will be done every 3 cycles. Up to 10 patients may receive Panitumumab. Eligible Pts are untreated, newly diagnosed, resectable/borderline resectable PDAC or colorectal/appendiceal adenocarcinoma with peritoneal metastases or oligometastases eligible for cytoreductive surgery, as determined by multidisciplinary evaluation. Inclusion criteria also include ECOG PS 0-1, adequate organ function, measurable or evaluable disease. Primary objectives are safety and biological activity of CEND‐1. Secondary objectives include ORR, R0 resection rate, DFS, OS. Exploratory objectives include pathologic response, tissue immune response, EGFR expression, tumor tissue-to-plasma concentration of Panitumumab pre and post CEND-1 treatment. Enrollment to the CENDIFOX trial is currently ongoing. Clinical trial information: NCT05121038.

4023 Background: Metastatic pancreatic cancer (mPC) remains a deadly disease with very limited treatment options. FFX is a standard first-line therapy for mPC with a median overall survival (mOS) of 11.1 months. CPI-613 is a stable intermediate of a lipoate analog that inhibits pyruvate dehydrogenase and α-ketoglutarate dehydrogenase enzymes of the tricarboxylic cycle preferentially within the mitochondria of cancer cells. In a phase I study, CPI-613+mFFX was safe and exhibited promising signal of efficacy. Methods: A global, randomized phase 3 trial was conducted across 73 sites to investigate the efficacy and safety of CPI-613 in combination with mFFX compared to standard dose FFX in treatment-naïve patients with mPC. Treatment was administered in 2-weekly cycles until progression or intolerable toxicity. In the experimental arm, CPI-613 at 500 mg/m2 was given intravenously on days 1 and 3. The doses of irinotecan, oxaliplatin, and 5-fluorouracil in the experimental arm were 65 mg/m2, 140 mg/m2, and 2,400 mg/2, respectively. Primary endpoint was OS. Secondary endpoints were progression-free survival (PFS), overall response rate (ORR), duration of response, pharmacokinetics, patient reported outcomes and safety. Results: 528 patients were randomly assigned (266 in test and 262 in control arm). There were 362 deaths, with a mOS of 11.1 months for CPI-613+mFFX vs. 11.7 months for FFX [hazard ratio (HR), 0.95; 95% CI, 0.77 to 1.18; P = 0.655]; mPFS was 7.8 months vs. 8.0 months respectively [HR, 0.99; 95% CI, 0.76 to 1.29; P = 0.94]; ORR was 39% in the test arm vs. 34% in the control arm [ORR ratio, 1.23 (95% CI, 0.86 to 1.75)]. Grade ≥ 3 treatment-emergent adverse events with ≥ 10% frequency in CPI-613 plus mFFX vs. FFX arm were diarrhea (11.2% vs. 19.6%), hypokalemia (13.1% vs. 14.9%), anemia (13.9% vs. 13.6%), neutropenia (11.2% vs. 14.0%), thrombocytopenia (11.6% vs. 13.6%) and fatigue (10.8% vs. 11.5%). Conclusions: The addition of CPI-613 to mFFX failed to show significant improvements of ORR, PFS or OS. The mFFX in the test arm that had the lowest prospectively tested doses of FFX was without compromise on PFS or OS and may be considered as a reference for future FFX administration. Clinical trial information: NCT03504423.

4047 Background: Surgical rection is the only potentially curative intervention for locally advanced adenocarcinoma of esophagus, GEJ and stomach. Results from various studies have demonstrated the benefits of perioperative treatment including neoadjuvant and adjuvant chemotherapy or chemoradiation, however, there is lack of universally accepted standard. Recent data demonstrated the benefit of immune checkpoint inhibitor in adjuvant setting in patients who had pre-operative chemoradiation. This single arm phase 2 trial is aimed to evaluate efficacy and safety of pembrolizumab, an immune checkpoint inhibitor, in combination with mFOLFOX in patients with potentially resectable adenocarcinoma of distal esophagus, GEJ and stomach. The primary objective is pathological response rate (ypRR with tumor regression score, TRS ≤ 2). Methods: Newly diagnosed locally advanced (T1N1-3M0 or T2-3NanyM0), potentially resectable adenocarcinoma of distal esophagus, GEJ and stomach by PET, EUS, CT C/A/P and staging laparoscopy were treated with pre-operative mFOLFOX6 (oxaliplatin 85mg/m2, Leucovorin 400mg/m2, 5-FU bolus 400mg/m2, and 5-FU 2400mg/m2 infusion every 2 weeks) for 4 cycles and pembrolizumab (200 mg IV q3week) for 3 cycles. Patients with no evidence of metastatic disease by PET and CT C/A/P who are eligible for resection underwent surgery. Post-operative treatment consisted of 4 cycles of mFOLFOX and 13 cycles of pembrolizumab 4-8 weeks postoperatively. Results: Up to 2/10/2021, all 37 patients eligible for the study finished preoperative treatment. 27 had curative intended operations, and all had R0 resection. 5 of 27 (19%) achieved ypCR and 6/27 (22 %) with regression score of 0 in the primary cancer. All except 2 patients (25/27, 93%) had shown pathologic response to the treatment with TRS ≤ 2. 21 patients completed all planned treatment with an average follow-up of 27 months. 2 patients had recurrence/metastatic disease (at 9 and 10 months from the enrollment) with 1 died at 23 months, and the other is still alive at 20 month. The rest patients (19/21) are all free of disease. G3/4 toxicities were reported in 21 of all 37 treated patients. There were no unexpected toxicities. Conclusions: The combination of mFOLFOX and pembrolizumab as peri-operative (pre- and post-operative) therapy in patients with locally advanced adenocarcinoma of distal esophagus, GEJ and stomach is safe. The preliminary benefit data are very encouraging with ypRR of 93%, ypCR of 19 %, and the long-time survival. The data support the combination of chemotherapy and Immune checkpoint inhibitor at perioperative setting. In addition, the data supports the staging laparoscopy for peritoneal disease assessment as the standard in resctacbility evaluation. Clinical trial information: NCT03488667.

100 Background: CRC is a common cancer world-wide, accounting for ̃10% of cancer cases and mortality. Treatment options are limited, and survival is poor for pts with advanced disease, particularly those with mKRAS. After failure of 1L treatment for mCRC, regardless of KRAS mutation status, the ORR for FOLFIRI-bev is 5-13%, with PFS 4-6 mos, and OS 10-12 mos. Onvansertib is a highly selective, ATP-competitive, orally bioavailable PLK1 inhibitor that is synergistic with irinotecan and with 5FU in xenograft models of mKRAS CRC. We present preliminary safety, efficacy, and biomarker data from an ongoing Ph1b/2 trial of onvansertib + FOLFIRI-bev in pts with mKRAS mCRC progressing after 1L treatment with fluoropyrimidine + oxaliplatin, +/- bev. Methods: Pts with mCRC with a KRAS mutation detected by a CLIA-certified lab were eligible. In the Ph1b portion of the study, onvansertib was given on a 3+3 dose escalation at 12, 15 or 18 mg/m2 on days 1-5 and 15-19 of each 28-day cycle in combination with FOLFIRI-bev. The MTD was 15 mg/m2 and was chosen as the RP2D. The primary endpoint for the Ph2 was ORR, and radiographic response was assessed every 8 wks per RECIST v1.1. Safety was evaluated continuously, and AEs were recorded using CTCAE v5.0. Baseline and post-treatment blood samples were collected for biomarker analyses, including mutant allele frequency (MAF) of the pt’s known KRAS mutation. Results: As of 16Sep2021, a total of 50 pts had been treated: 18 on the Ph1b and 32 on the Ph2, including 35 pts at the RP2D, and median follow up was 4.7 mos (range 0.4-18). Of the 50 pts, 26 remain on treatment, as do 24 of 35 RP2D pts. The combination was well-tolerated: fatigue, neutropenia, and nausea were the most common treatment-emergent adverse events (TEAE) and were generally low-grade. Neutropenia was managed by removing the 5FU bolus from subsequent cycles of FOLFIRI and adding growth factor. Of the 50 pts, 44 were evaluable for efficacy, including 31 of 35 RP2D pts. ORR was 36% for the total group (1CR and 15 PR in 44 pts) and 35% for the RP2D group (1 CR and 10 PR in 31 pts). First responses were seen between 2 and 6 months after the start of therapy. Responses were observed across different KRAS variants. Pts achieving a CR or PR showed the greatest decreases in plasma MAF after the first cycle of therapy. Of the 50 pts, 24 pts have discontinued for the following reasons: progressive disease (13), toxicity (4), patient decision (4), proceeding to potentially curative surgery or other localized therapy (3). Conclusions: The combination of onvansertib with FOLFIRI-bev was well tolerated: observed TEAEs have been generally low-grade and manageable. The combination has demonstrated a promising ORR in 2L treatment of mCRC pts harboring various KRAS mutations, and efficacy was correlated with early changes in plasma mKRAS. Updated safety, efficacy, and biomarker analyses will be presented. Clinical trial information: NCT03829410.

504 Background: (177)Lu-Dotatate, a form of Peptide Receptor Radionuclide Therapy (PRRT), was approved by FDA for treatment of somatostatin-receptor-positive NETs in 2018. Clinical trials prior to the FDA approval of (177)Lu-Dotatate showed favorable outcomes but there is limited published real world outcomes data. Methods: After obtaining IRB approval we retrospectively evaluated the efficacy of (177)Lu-Dotatate PRRT for somatostatin receptor positive gastroenteropancreatic NET patients at University of Kansas Cancer Center between June 2018 and September 2021. Results: A total of 65 patients received PRRT of which 58 completed treatment and 7 are still undergoing therapy. The 58 patients who completed treatment had a median age of 61.5 years, included 24 females and 34 males, and were 86% Caucasian and 12% black. These patients had primarily NETs of small bowel (n=24) and pancreatic (n=14) origin. Pathology showed grades 1 (21), 2 (25), and 3 (4) with a majority of well-differentiated tumors (47). All 4 treatments of PRRT were completed by 43 patients. In the entire cohort, response assessment to PRRT revealed 14 (26.9%) partial response (PR), 31 (59.6%) with stable disease (SD) and 7 (13.5%) with progressive disease (PD). In subset analysis, we found that patients with nonfunctional disease (n=29) had a higher rate of PR 42.3% vs 11.5% (p=0.0147) and higher disease control rate 96% vs 78% (p=0.042) than functional disease (n=29). Patients with nonfunctional disease had a lower PD 3.85% vs 23% (p=0.0147) and 53% vs 65% stable disease than functional disease. The table below shows clinical characteristics of these patients. Conclusions: Our real world outcomes analysis of NET treated with PRRT shows improved PR when compared to the initial clinical trials which is promising for patients. In addition, we found that there was statistically significant improved response rate in patients with non functional tumors which has not been described in literature before. If our study findings are validated in a larger cohort then it may guide patient selection for PRRT therapy in the future.[Table: see text]

TPS209 Background: Lynch syndrome (LS) is the most common inherited colorectal cancer (CRC) syndrome and is responsible for about 3% of newly diagnosed CRC. It is caused by germline mutations in one of the DNA mismatch repair (MMR) genes, and patients with LS carry a lifetime risk of CRC ranging between 10% and 70%. The role of inflammation in driving this malignant transformation is now well established and retrospective studies have revealed a potential chemo-preventative role for omega-3 (ω-3) polyunsaturated fatty acids (PUFAs), possibly via inhibition of inflammatory pathways associated with the development of defective MMR CRC tumors. While patients with LS have the highest risk of developing CRC, the majority of chemoprevention trials are focused on sporadic CRC. Effective interventions to reduce the risk of developing CRC in this population are limited to close surveillance and surgical prophylaxis. There is an unmet need for safe, effective, and non-invasive chemo-preventive interventions in patients with LS. Methods: This pilot study is a single-arm, open-label, phase 2 clinical trial of omega-3 acid ethyl esters (generic Lovasa; 2 grams orally once daily) for adult patients (≥ 18 years of age) with confirmed LS (based on germline testing of the MMR genes panel: MLH1, MSH2, MSHS6, PMS2 or deletion in EPCAM gene). Patients who are not candidate for elective endoscopy and/or with prior history of right sided or pan-colectomy are excluded. Thirty-four patients are expected to enroll, with a primary objective to determine the feasibility (defined as 80% retention rate) of 12 months of treatment with 2 grams capsules of omega-3 acid ethyl esters daily. Secondary endpoints include safety and tolerability of the intervention. Correlative aims include pre and post treatment assessment of colon mucosal tissue proliferation (right sided colon specimens will be evaluated for markers of proliferation (Ki-67) and apoptosis (Caspase-3)), the effect of omega-3 acid ethyl esters on inflammatory markers in serum, urine and feces (PGE-2, COX-2, β-catenin levels, and EPA:AA ratios), and gene expression related to proliferation, apoptosis and cell survival in colon tissue (NF-κB/Wnt pathways). The impact of omega-3 acid ethyl esters on intestinal microbiota will also be assessed (16S rRNA-based profiling). Correlative Colon tissue, serum, urine and feces samples are collected at baseline and at 12 months. The study is actively enrolling with 20 patients enrolled at the time of submission. Clinical trial information: NCT03831698.

TPS377 Background: GEA is one of the leading causes of cancer-related deaths worldwide and in the US. In chemo refractory GEA, therapeutic options are limited with modest efficacy. About 35% of GC have DNA damage response (DDR) gene alterations and PARP inhibitors have shown to induce synthetic lethality and reduce tumor burden in in vitro and in vivo models of GEA. DDR deficient tumors have increased tumor mutational load, tumor infiltrating lymphocytes, and increased chemokines and PD-L1 expression in the tumor microenvironment. PARP inhibitors like rucaparib enhance antigen presentation, induce pro-inflammatory cytokines, influence PD-1/PD-L1 expression, and modulate T- and B- cell activity. Furthermore, results from multiple GEA early phase trials have supported the immune modulatory impact of VEGF targeted agents like ramucirumab. We hypothesize that modulation of the tumor microenvironment via PARP and VEGF inhibition will enhance anti-tumor immunity and lead to clinical synergy when combined with immune checkpoint inhibitors like nivolumab in this population. Methods: A safety lead in phase of 6-9 patients will be followed by an open label, two parallel cohorts phase II study. A total of 52 patients (26 in each cohort) will be enrolled. A starting rucaparib dose of 600 mg BID in combination with the standard doses of ramucirumab and flat doses of nivolumab will be used for the safety lead in. Rucaparib dose will be lowered for phase II if > 1/6 patients experienced DLTs. Participants enrolled to cohort A will receive rucaparib at the recommended phase II dose (RP2D) PO twice daily + ramucirumab 8mg/kg IV q2w + nivolumab 480mg IV q4w (28-day cycles). Participants in cohort B will receive rucaparib at RP2D PO twice daily + ramucirumab 8mg/kg IV q2w (28-day cycles). Assessment of tumor response using modified RECIST v1.1 will be done every 2 cycles. 50% of Pts who will be enrolled in each of the 2 cohorts must have a deleterious tumor gene alteration in at least 1 gene in a screening 17-gene homologous recombination deficient (HRD) panel. Eligible Pts received ≥1 line of therapy, ECOG PS 0-1, adequate organ function, no prior PARP inhibitors, and have mismatch repair (MMR) proficient tumors with measurable disease. Primary objectives are to determine the RP2D and overall response rate (ORR) in phase I and II respectively. Secondary objectives include safety, overall benefit rate (OBR), median PFS, OS, and ORR in HRD positive and negative cohorts. Enrollment to the RiME trial is currently ongoing. The study is funded by research grants from Bristol Myers Squibb and Clovis. Clinical trial information: NCT03995017.

598 Background: PDAC of the H and B/T differ in embryonic origin, cell composition, blood supply, lymphatic and venous drainage, and innervation. H tumors tend cause symptoms earlier and to present at earlier stages compared to B/T cancers. The impact of PDAC tumor location on patient presentation and survival has been shown in large national data-based analyses, although with conflicting results. We aimed to compare the molecular and tumor immune microenvironment (TIME) profiles of PDAC of the H vs. B/T. Methods: A total of 3499 PDAC samples were analyzed via next-generation sequencing (NGS) of RNA (whole transcriptome, NovaSeq), DNA (NextSeq, 592 genes or NovaSeq, whole exome sequencing) and immunohistochemistry (IHC, Caris Life Sciences, Phoenix, AZ). RNA deconvolution was performed using QuantiSeq (Finotello 2019, Genome Medicine) to quantify the immune cell infiltration. Pathway gene enrichment analyses were done using Gene Set Enrichment Analysis (GSEA, Subramaniam 2015, PNAS). Significance was determined as p values adjusted for multiple correction (q) of < 0.05. Results: Anatomic subsites of PDAC tumors were grouped by primary tumor sites into H (N = 2058) or B/T (N = 1384). There were significantly more metastatic tumors profiled from H vs. B/T (57% vs. 44%, p < 0.001). KRAS mutations (93.8% vs. 90.2%), genomic loss of heterozygosity (12.7% vs. 9.1%), and several copy number alterations ( FGF3, FGF4, FGF19, CCND1, ZNF703, FLT4, MUTYH, TNFRS14) trended higher in B/T when compared to H (p < 0.05 but q > 0.05). GNAS mutations (2.2% vs. 0.7%) trended higher in H vs. B/T (p < 0.05). No significant difference in immuno-oncology (IO) markers (TMB, PD-L1, MSI-H) were observed, but expression analysis of IO-related genes showed significantly higher expression of CTLA4 and PDCD1 in H (q < 0.05, fold change 1.2 and 1.3) and IDO1 and PDCD1LG2 expression trended higher in B/T (p < 0.05, fold change 0.95). When comparing median cell abundance values as part of TIME analysis, H had increased immune infiltration of B cells (0.045 vs. 0.043), M2 macrophages (0.035 vs. 0.032), neutrophils (0.056 vs. 0.052), NK cells (0.027 vs. 0.026), CD8+ T cells (% > 0: 48.2% vs. 43.2%), while B/T had increased infiltration of M1 macrophages (0.035 vs. 0.032) (all q < 0.05). GSEA showed enrichment of CTLA4 (normalized enrichment score (NES) 1.6, false discovery rate (FDR) 0.19) and primary immunodeficiency pathway enrichment (NES 1.7, FDR 0.11) in H. Conclusions: To our knowledge, this is one of the largest cohort of PDAC tumors subjected to broad molecular profiling. Differences in IO-related gene expression and TIME cell distribution suggest that response to IO therapies may differ in PDAC arising from H vs B/T. Subtle differences in the genomic profliles of H vs. B/T tumors were also observed.

329 Background: Surgical rection is the only potentially curative intervention for locally advanced adenocarcinoma of esophagus, GEJ and stomach. Results from various studies have demonstrated the benefits of perioperative treatment including neoadjuvant and adjuvant chemotherapy or chemoradiation, however, there is lack of universally accepted standard. Recent data demonstrated the benefit of immune checkpoint inhibitor in adjuvant setting in patients who had pre-operative chemoradiation. This single arm phase 2 trial is aimed to evaluate efficacy and safety of pembrolizumab, an immune checkpoint inhibitor, in combination with mFOLFOX in patients with potentially resectable adenocarcinoma of distal esophagus, GEJ and stomach with the primary objectives of pathological response rate (ypRR with tumor regression score, TRS ≤ 2). We are reporting the preliminary analyses while the study nears completion. Methods: Patients with newly diagnosed locally advanced (T1N1-3M0 or T2-3NanyM0), potentially resectable adenocarcinoma of distal esophagus, GEJ and stomach by PET, EUS, CT C/A/P and staging laparoscopy were treated with pre-operative mFOLFOX6 (oxaliplatin 85mg/m2, Leucovorin 400mg/m2, 5-FU bolus 400mg/m2, and 5-FU 2400mg/m2 infusion every 2 weeks) for 4 cycles and pembrolizumab (200 mg IV q3week) for 3 cycles. Patients with no evidence of metastatic disease by PET and CT C/A/P who are eligible for resection underwent surgery. Post-operative treatment consisted of 4 cycles of mFOLFOX and 13 cycles of pembrolizumab 4-8 weeks postoperatively. Results: Of 35 patients enrolled (age range 44-86, mean of 65 years; with male:female of 28:7), 33 finished preoperative treatment, 26 had curative intended operations with R0 resection for all, 1 is pending for surgery, and 2 are still on pre-operative treatment. 5 of 26 pts achieved ypCR (19% regression score of 0). All except 2 patients (24/26, 92%) had shown pathologic response to the treatment with TRS ≤ 2. 23/26 (88%) finished post-operative treatment. 20 patients completed all planned treatment with an average follow-up of 22.7 months. Amount them, 2 patients had recurrence/ metastatic disease (at 9 and 10 months, respectively) with 1 died 23.3 months from enrollment, and the rest are all free of disease. G3/4 toxicities were reported in 19 of all 35 treated patients. There were no unexpected toxicities. Conclusions: The combination of FOLFOX and pembrolizumab as peri-operative (pre- and post-operative) therapy in patients with locally advanced adenocarcinoma of distal esophagus, GEJ and stomach is safe and preliminary benefit data are very encouraging with ypRR of 92% and ypCR of 19% and supporting the combination of chemotherapy and Immune checkpoint inhibitor at perioperative setting. Clinical trial information: NCT03488667.

135 Background: Cabo is an anti-VEGFR2/MET/AXL drug with broad multi-kinase inhibitory spectrum. Preclinical and clinical studies in various solid tumors demonstrated favorable immune modulatory activity of Cabo with clinical synergy seen when combined with PD-1/ PD-L1 inhibitors like Durva. Upon completion of phase Ib gastrointestinal (GI) basket CAMILLA trial evaluating Cabo + Durva in 30 patients (pts) demonstrating favorable safety & efficacy, the trial was expanded to phase 2 multi-cohort, multi-center trial of 117 pts. Herein, we report results of the phase 2 CRC cohort, the first evaluation of cabo + IO in this population. Methods: Pts enrolled in this cohort were administered Cabo + Durva at the RP2D of 40mg QD and 1500mg IV Q4W respectively. Enrolled pts must have progressed on 2 or more lines of therapy. Primary outcome measure was investigator assessed overall response rate (ORR) and secondary outcomes were rate of treatment related adverse events (TRAE), investigator assessed disease control rate (DCR), progression free survival (PFS), and overall survival (OS). Subgroup analysis was done in pts with RAS wild type tumors. Exploratory analysis of pathogenic molecular tumor alterations using next generation sequencing (NGS) was done in pts who achieved confirmed partial response (PRc)/ stable disease (SD) > 6 months. Results: Of the 36 pts enrolled, 29 (16F, 13M) were evaluable for efficacy. Median age 57 years (27-76). 90% (26) had ECOG of 1. All had pMMR/MSS, 41% (12) had RAS wild type tumors, and 6.9% (2) had HER2 amplification. 52% (15) had received ≥ 3 lines of therapy. All had metastases at ≥ 3 sites and 79% (23) metastases in the liver. Among 36 pts evaluable for safety, treatment related serious adverse events (SAEs) occurred in 31% (11/36). Most common TRAEs were grade 1-2 fatigue (53%), nausea (42%), diarrhea (36%), anorexia (31%), hand-foot syndrome (22%), & hypertension (16%). Grade ≥ 3 immune-related adverse events (IRAE) occurred in 16.6% (6/36). Efficacy analysis revealed an ORR 27.6% (8/29); PRc 21% (6/29); DCR 86.2% (25/29); median DOR was not reached (NR); median PFS 4.4 months; 6-month PFS 28% & median OS 9.1 months. In the RAS wild type subgroup, ORR (PRc) was 50.0%; DCR 83.3%; median PFS 6.3 months and median OS was NR. Of the 7 pts who achieved PRc/SD > 6 months, one had KRAS G12V tumor mutation along with mutations in ARID1A & IDH1. Remaining pts had RAS wild type tumors & among those, the following NGS alterations were detected: 1 HER-2 amplification, 1 MET amplification, & 2 alterations in ATM. Conclusions: Cabo + Durva demonstrated promising efficacy and was fairly tolerated without new safety signals in heavily treated pMMR/MSS CRC pts. These encouraging results warrant further evaluation of this regimen in a randomized setting as salvage therapy in pMMR/MSS CRC. Clinical trial information: NCT03539822.

TPS620 Background: GP-2250 is a metabolic enzyme inhibitor that selectively induces oxidative stress, mitochondrial dysfunction, and apoptosis in cancer cells by inhibiting glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and limiting aerobic glycolysis, an essential process for energy production in cancer cells. In vitro, increasing concentrations of GP-2250 induced proportional dose-response effects in 5 different pancreatic cancer cell lines (Buchholtz BMC Cancer 2017). Patient-derived pancreatic xenograft studies in mice demonstrated tumor growth inhibition with GP-2250 monotherapy (Braumann J Clin Oncol 2020). In combination with gemcitabine, a synergistic effect was shown with greater inhibition of pancreatic tumor growth than with either gemcitabine or GP-2250 alone. These findings support the assessment of the therapeutic potential of GP-2250 in combination with gemcitabine in patients with pancreatic cancer. Methods: This open-label phase 1/2 trial will evaluate the safety and tolerability of escalating doses of GP-2250 and the preliminary efficacy of GP-2250 in combination with gemcitabine in patients with advanced unresectable or metastatic pancreatic adenocarcinoma who have progressed on prior treatment with FOLFIRINOX chemotherapy. In phase 1, Bayesian Optimal Interval design will be used for GP-2250 dose escalation. The dose-limiting toxicity (DLT)–assessment period will be 5 weeks at each dose level (starting dose 250 mg once weekly intravenously), consisting of a 1-week run-in period with GP-2250 monotherapy, followed by a full cycle of GP-2250 plus standard dose and schedule of gemcitabine. Single patient cohorts (100% dose escalation between cohorts) will be enrolled until the occurrence of the first DLT (or cohort 4 if no DLTs occur in cohorts 1−3), after which, cohorts will be expanded to 3 patients with 35%−45% dose escalations between cohorts. Patients will receive treatment until disease progression or development of unacceptable toxicity. Primary study endpoints include safety and tolerability of GP-2250, tumor response (RECIST 1.1), disease control rate (complete response, partial response, stable disease), and duration of response. Pharmacokinetics, incidence and severity of laboratory abnormalities and treatment-emergent adverse events, progression-free survival, and overall survival will also be assessed. As of September 2021, 19 patients have been enrolled. Following completion of the phase 1/2 trial, a phase 3 first-line maintenance study will be initiated. This study is funded by Geistlich Pharma AG. Clinical trial information: NCT03854110. Clinical trial information: NCT03854110.

e15613 Background: Fluoropyrimidines, such as 5-fluorouracil (5-FU) and capecitabine, are commonly used chemotherapies for solid tumors, and essential for curative intent treatment of colorectal cancer. But sometimes, their use may be limited by cardiac toxicity limiting the possibility of cure in some patients. Cardiotoxicity could be asymptomatic (EKG changes) or manifested as chest pain, arrhythmias, acute coronary syndrome or death. Rechallenging may be daunting and it may result in interruption or even discontinuation of planned chemo. Traditionally, nitrates and/or IV/oral calcium channel blockers (CCB) have been used for management of fluoropyrimidine-induced cardiotoxicity but without much benefit. Ranolazine, an oral antianginal drug approved for chronic angina, diminishes myocardial ischemia by reducing calcium overload caused by inhibition of late sodium current and it does not affect heart rate or blood pressure. Our objective was to evaluate the efficacy of our novel approach using ranolazine with other traditional drugs. Methods: 8 patients (median age 49.5 yrs) with fluoropyrimidine induced cardiotoxicity were retrospectively analyzed. They were rechallenged with the planned fluoropyrimidine regimen with our 3 drug cardioprotective regimen (KU protocol). This included oral Ranolazine 1000mg BID and Amlodipine 2.5 mg daily to start the day before starting 5FU infusion/oral capecitabine and to continue it until the day after completion of infusion/treatment, and Nitroglycerin paste 1 inch every 6 hours starting before infusion and continue until it was completed. These meds were discontinued upon completion of chemo. Results: 8 patients were rechallenged with fluoropyrimidine utilizing KU protocol, 6 patients (75%) were able to complete previously planned fluoropyrimidine regimen. One pt (*) discontinued capecitabine due to recurrent chest pain and treatment was switched to 5FU based regimen with KU protocol, which pt was able to complete without chest pain. Another pt (**), 5FU was stopped due to severe diarrhea, not due to cardiotoxicity. All pts tolerated KU protocol well. Conclusions: In our small, single center experience, we were able to safely and effectively rechallenge pts with fluoropyrimidines and complete curative intent treatment with our KU protocol. This protocol uses FDA approved oral and transcutaneous drugs without requiring a healthcare personnel to administer an IV CCB that can cause precipitous bradycardia and/or hypotension. Our results need validation in a larger cohort. [Table: see text]

e18578 Background: Obesity is a bona fide risk factor for ICU admission, mechanical ventilation, and mortality in patients (pts) with COVID-19 in the general population. However, whether obesity is a risk factor in cancer pts remains unknown. Herein, we have conducted a systematic review/meta-analysis of obesity and all-cause mortality in cancer pts with COVID-19. Methods: Following PRISMA guidelines,a systematic search of PubMed and Embase as well as major conference proceedings (ASCO/ESMO/AACR) was conducted for publications from inception to 14 January 2020. Observational studies that reported all-cause mortality in cancer pts with lab confirmation or clinical diagnosis of COVID-19 and BMI (obese (>30 kg/m2) vs. non-obese) were included in the analysis. The pooled odds ratio (OR) and 95% confidence interval (CI) were calculated with the fixed-effects model based on low heterogeneity. Small sample publication bias was evaluated using the Begg’s Funnel Plot and Egger’s test. Results: After reviewing 3387 studies,3 retrospective cohort studies of 419 obese and 1694 non-obese cancer pts (N=2117) with COVID-19 in both inpatient/outpatient settings that reported outcomes based on obesity were found. The 3 studies were conducted multi-nationally in North America, in France, and in the Netherlands respectively. The median ages of the cohorts ranged 66-68. All studies included various cancers of various stages and were of high quality per Newcastle Ottawa scale (scores 7-9). Fixed effects meta-analysis showed no association between obesity and all-cause mortality (OR 0.95, 95% CI 0.74-1.23) in cancer pts with COVID-19. Heterogeneity was low (I2 = 33%). No significant funnel plot asymmetry was detected per Egger’s test (P=0.2273). The reported OR of each study is outlined in the table. Conclusions: In contrast to the general population, our analysis reveals that obesity is not associated with increased all-cause mortality in cancer pts with COVID-19. Limitations of this study include a limited number of included studies, reliance on retrospective studies, non-use of ethnicity-specific WHO BMI criteria, and limited granularity of the study-reported BMI. Future prospective studies are warranted to assess the complex interplay among anthropomorphic measures, cachexia/sarcopenia, comorbidities associated with the metabolic syndrome, and COVID-19 outcomes in the cancer pt population.[Table: see text]

4091 Background: The relationship between Helicobacter pylori (H.pylori) and hepatocellular carcinoma (HCC) was first proposed in 1994. Since then, several studies have been performed to explore the association. The role of Hepatitis C (HCV) viruses coexisting with H.pylori in causing HCC was also studied. With the emergence of data in this regard, a causal relationship has been postulated, but not confirmed, and hence the relationship remains controversial. Our meta-analysis aims to summarize the research on this topic and investigate if there exists a relationship between H. pylori infection and the development of HCC and if the presence of HCV along with H.pylori plays a role in liver carcinogenesis. Methods: Following PRISMA guidelines, we performed a systematic review of all relevant studies published in the literature using keywords “Helicobacter pylori” and “Hepatocellular carcinoma” on major literature databases, including PubMed, EMBASE, Web of Science, and Cochrane controlled trials register. A total of 656 studies were identified between 1994 to March 2020, out of which 26 studies qualified under our selection criteria. Patients positive for HCC are included as cases and patients that did not have HCC under control group. In both groups, H.pylori positive patients and their HCV status, was identified. Results: Out of the 26 studies included in the final analysis, the prevalence of H. pylori infection was 64.78% (561 of 866) amongst HCC cases and 47.92% (1718 of 3585) in the non-HCC control group. The summary odds ratio for the association of H. pylori infection with the risk for HCC using the random-effects model was determined to be 4.75 (95% CI, 3.06-7.37), I²=63%. Subgroup analysis to determine the odds of developing HCC in the presence of H.pylori and HCV coinfection, was 13.97 (95% CI, 3.94-49.61), I²=81%. Whereas, the odds of developing HCC in the presence of only HCV without H.pylori was found to be 2.21 (0.70-6.94), I²=79. Subgroup analysis by study design showed no significant difference between the study groups (P= 0.5705). Conclusions: Our meta-analysis showed a positive association between H. pylori infection and the development of HCC. It showed a significantly higher risk of developing HCC in the presence of HCV infection along with H.pylori. Further prospective cohort studies are needed to prove the causal relationship, especially in cases of Hepatitis B, C coinfection, and cirrhotic patients.[Table: see text]

6556 Background: CAR T-cell therapy is a type of adoptive cell transfer (ACT). In 2017 CAR T-cell therapy was approved by the Food and Drug Administration (FDA) for management of diffuse large B-cell lymphoma refractory to at least two prior lines of therapy (DLBCL) including primary mediastinal large B-cell lymphoma (PMBCL), and B cell precursor acute lymphoblastic leukemia (ALL) up to 25 years of age that is refractory or in second or later relapse. Since its inception, several patients underwent CAR T-cell therapy but data on real world outcome is limited. In this study we aim to evaluate the in-hospital outcomes of CAR T-cell therapy in the United States in 2018. Methods: This is a cross-sectional study using National Inpatient Sample 2018 database. Discharges with the ICD-10-PCS code for CAR T-cell therapy and ICD-10-CM code of ALL, DLBCL or PMBCL were included in the study. We analyzed their in-hospital outcomes (Total discharges, length of stay in days, hospitalization cost, and mortality - number of deaths). We applied the cost to charge ratio to hospitalization charges to estimate the mean hospitalization cost. The weighted sample represents national estimates. Results: We identified 785 discharges with CAR T cell therapy and diagnosis of ALL, DLBCL or PMBCL. 155 (19.75 %) were ALL, 620 (78.98%) DLBCL, and 10 (1.27%) PMBCL. Mean length of stay for the study cohort was 23.26 days. Specifically, for ALL mean LOS was 33.67 days, DLBCL 20.76 days, and PMBCL 17 days. Mean hospitalization cost for the study cohort was $285,989, specifically for ALL it was $342,228, DLBCL $274,102, PMBCL $179,431. There were a total of 60 (7.6%) deaths in the study cohort. Diagnosis specific mortality was 20 (12.9%) in ALL, 40 (6.4%) in DLBCL and none in PMBCL. Conclusions: Majority of discharges who underwent CAR T-cell therapy were DLBCL followed by ALL and PMBCL. CAR-T Cell hospitalizations have high costs and long length of stays. Mean length of stay was highest in ALL discharges, least in PMBCL. Mean hospitalization cost was higher in DLBCL discharges compared to ALL and was least in PMBCL discharges. The in-hospital mortality with the CAR-T cell therapy appears to be higher than reported in clinical trials, especially for ALL. [Table: see text]

11523 Background: Sarcoma pts often receive aggressive, highly immunosuppressive therapy and may be at high risk for severe COVID-19. Demographics, outcomes and risk factors for pts with sarcoma and COVID-19 are unknown. We aimed to describe the course of COVID-19 in sarcoma pts and to identify factors associated with adverse outcomes. Methods: The COVID-19 and Cancer Consortium (NCT04354701) is an international registry of pts with cancer and COVID-19. Adult pts (≥18 years old) with a diagnosis of sarcoma and laboratory confirmed SARS-CoV-2 were included from 50 participating institutions. Data including demographics, sarcoma diagnosis and treatment, and course of COVID-19 infection were analyzed. Primary outcome was the composite rate of hospitalization or death at 30 days from COVID-19 diagnosis. Secondary outcomes were 30 day all-cause mortality, rate of hospitalization, O2 need, and ICU admission. Descriptive statistics and univariate Fisher tests are reported. Results: From March 17, 2020 to February 6, 2021, N=204 pts were included. Median follow up was 42 days. Median age was 58 years (IQR 43-67). 97 (48%) were male. 30 (15%) had ECOG performance status ≥2. 104 (51%) received cancer treatment, including surgery or radiation, within 3 months of COVID-19 diagnosis. 153 (75%) had active cancer, of whom 34 (22%) had lung metastases. 100 (49%) pts met the composite primary endpoint; 96 (47%) were hospitalized and 18 (9%) died within 30 days from COVID-19 diagnosis. 64 (31%) required oxygen, and 16 (8%) required ICU admission. Primary endpoint rates were similar for pts who received cytotoxic chemotherapy (38/58, 66%) or targeted therapy (16/28, 57%). Pts with higher rates of the primary endpoint included patients ≥60 years old (59% vs 40%, OR 2.04, 95% CI 1.12-3.74, p=0.016), pts with ECOG PS ≥2 vs 0-1 (90% vs 41%, OR 12.2, 95% CI 3.44-66.8, p

e16194 Background: Hepatocholangiocarcinoma (HCC-CC) is a rare form of cancer with a poor prognosis. Of all primary liver cancers, the incidence of HCC-CC ranges from 0.4 to 14.2%. HCC-CC is a mixed carcinoma with findings of both hepatocellular carcinoma and cholangiocarcinoma. Immune checkpoint inhibitors are a potent first line treatment in hepatocellular carcinoma with multiple clinical trial showing effectiveness in cholangiocarcinoma. HCC-CC has limited proven treatment options as patients are generally excluded from clinical trials. In this study we reviewed outcomes of patients with HCC-CC who received immune checkpoint inhibitor in a single center. Methods: Records of patients who had a pathological confirmed HCC-CC by a subspecialized hepatic pathologist at the University of Kansas medical center were reviewed. We identified 6 patients with locally advanced unresectable or metastatic HCC-CC that received immune checkpoint inhibitor between February 2017 and January 2021. Baseline characteristics were obtained, as well as best response, line of therapy, and duration of response. Results: Of the six patients 4 (66%) received PD-1 inhibitor alone and 2 (34%) received combination therapy with CTLA-4 inhibitor for the treatment of HCC-CC. There were 3 (50%) females and 6 (100%) with prior hepatitis C infection. four (66%) patients had metastatic disease and 2 had locally unresectable advanced disease. Objective response rate was 83.3%. One patient achieved complete response and had a treatment holiday after receiving treatment for 2 years, and restarted immunotherapy upon relapse. Four patients had a partial response, of which two passed away after disease progression. One patient had stable disease on 2 different lines of immunotherapy then progressed. Of those who responded, one patient received immunotherapy, 3 (50%) received liver directed therapy and two received chemotherapy or Lenvatinib as first line treatment (Table). Conclusions: Immune checkpoint inhibitors demonstrate potential activity in patients with HCC-CC without unexpected side effect in this unmet need high-risk population. Larger studies are needed to confirm activity and efficacy in this setting.[Table: see text]

e16273 Background: There is a critical need to develop fast, reliable, and cost-effective methods for the detection of pancreatic cancer (PC) at the earliest stage to maximize the impact of treatment. To-date, early detection of PC is close to impossible due to the location of the pancreas and the absence of characteristic symptoms in early cancer stages. Methods: Our team of clinicians and scientists has established a fast and reliable nanobiosensor technology that comprises iron/iron oxide nanoparticles attached to a protease or arginase activatable FRET pair (tetrakis (4- carboxyphenyl) porphyrin (TCPP) /cyanine 5.5). Arginase and seven proteases (MMP1, 3, and 9, cathepsin B, and E, urokinase plasminogen activator, and neutrophil elastase) were identified using the Gene Expression Omnibus (GEO) web tool based on their different expression pattern in pancreatic cancer patients, pancreatitis and healthy control subjects. Protease/arginase activities were measured in serum after 1h of incubation. Based on this data, a novel engineering approach to improved early stage detection of pancreatic cancer is reported here. This study was funded by American Cancer Society Institutional Research Grant (IRG‐16‐194‐07), awarded to the University of Kansas Medical Center. Results: In our study, 159 patients were enrolled at KU Cancer Center from 2000-2019, 47 with metastatic PC, 36 with localized PC, 26 pancreatitis and 50 healthy controls using KUCC Biospecimen Repository. The problem of early stage detection of pancreatic cancer can be modeled as a multi-class classification problem. Conventional classification approaches provide at most 77% accuracy for the dataset under consideration. A new hierarchical decision structure with specific feature engineering at each step is introduced here to improve the performance of the classifier. The fundamental premise of this information fusion-based framework involves tailoring the statistically most significant features with appropriate weights to execute an efficient binary classification task at each hierarchical step. An overall accuracy of 95% was achieved for the detection of patients with early pancreatic cancer (see table). Conclusions: Because of the dire survival statistics of pancreatic cancer, detection at the earliest possible time by means of a liquid biopsy will offer the greatest benefit. Novel nanobiosensor based protease biomarkers achieved high accuracy in early detection of pancreatic cancers by applying hierarchical decision structure. Our results need validation in a larger cohort. Predicted true class considering the following combination of classification methods: Step1 – kNN*, step2 – kNN*, step3 – RFC* (Accuracy = 94.97%).[Table: see text]

132 Background: Colorectal cancer (CRC) is the second leading cause of cancer-associated deaths in the United States. Some of the poor prognostic factors for metastatic CRC (mCRC) include BRAF V600E mutation and microsatellite instability (MSI) that result from mutation or loss of mismatch-repair genes. While the prognostic value of MSI-high CRC for early-stage patients treated with resection and adjuvant chemotherapy is favorable, the prognostic value of BRAF mutation is still unclear. Furthermore, the impact of BRAF mutation with concurrent microsatellite instability on overall survival has not been well investigated. Methods: Here, we collected BRAF mutation status and MSI status of stage II/III CRC patients (n=106) treated at the University of Kansas Cancer Center between September 2009 and July 2020 and compared overall survival between 4 subtypes:MSI-H/BRAF mutant (n=16), MSS/BRAF mutant (n=4), MSI-H/BRAF WT (n=17) and MSS/BRAF WT (n=69), further stratifying patients by age at diagnosis and tumor location. Molecular data were obtained from molecular oncology laboratory as PCR or IHC-based or acquired from outside records. Subgroup analyses were done for stage II and stage III cancers. Results: Table shows the patient characteristics. From our preliminary analysis, MSI-H CRC was found to be primarily a right-sided tumor (MSI-H/BRAF mutant: 94% and MSI-H/BRAF WT 76%). On the contrary, MSS CRC had a more heterogenous localization, spanning left colon, right colon and rectum. In our patient cohort, median survival was not reached for stage II patients whereas for stage III patients, BRAF mutation was associated with poor median survival irrespective of MSI status (MSS/BRAF mutant: 27 months and MSI-H/BRAF mutant 29 months). Median overall survival was found to be 87 months, not reached, 27 months and 29 months for MSS/BRAF WT, MSI-H/BRAF WT, MSS/BRAF mutant and MSI-H/BRAF mutant, respectively. Although associated with poor survival, MSI-H/BRAF mutant displayed later age at diagnosis (mean age 73) compared to MSS/BRAF mutant (mean age 60, p-value

e16738 Background: A phase II trial has shown improved efficacy of neoadjuvant therapy when combined with losartan (by remodeling desmoplasia) in locally advanced pancreatic ductal adenocarcinoma (PDA). However, role of losartan is unknown in metastatic PDA. We examined the relationship between the use of the angiotensin II receptor antagonist, losartan, at time of diagnosis with clinical outcomes in metastatic PDA pts that received chemo. Methods: We retrospectively evaluated 114 metastatic PDA pts treated at our center between Jan 2000 and Nov 2019. We compared OS, PFS, objective response rate (ORR), and disease control rate (DCR) between pts using losartan at time of cancer diagnosis and a control group of pts not on losartan. A subanalysis was performed based on losartan dose: 100mg dose versus control pts. and based on chemo: FOLFIRINOX or gemcitabine+abraxane. Results: Table shows baseline demographics. No significant difference was found in OS [p = 0.455] or PFS [p = 0.919] in pts on losartan (median 274d, 83d) vs control (median 279d, 111d) [p = 0.466]. No significant difference was found in ORR [p = 0.621] or in DCR [p = 0.497]. No significant difference was found in OS [p = 0.771] or PFS [p = 0.064] in losartan pts (median 347d, 350d) vs control (median 333d, 101d) treated with FOLFIRINOX. No significant difference was found in OS [p = 0.916] or PFS [p = 0.341] in losartan (median 312d, 69d) vs control (median 221d, 136d) [p = 0.916] treated with gemcitabine+abraxane. No significant difference was found in OS [p = 0.727] or PFS [p = 0.790] in 100mg losartan pts (median 261d, 84d) vs control (median 279d, 111d). Conclusions: Pts on losartan at time of diagnosis had no significant difference in OS, PFS, ORR, DCR than control pts. However, a subanalysis of pts treated with FOLFIRINOX revealed a longer PFS with losartan than control but did not meet statistical significance, likely due to small sample size. To confirm if the benefit of losartan + FOLFIRINOX seen in neoadjuvant setting for locally advanced cancer also applies to metastatic cancer, our findings need to be validated in a larger cohort. [Table: see text]

e16787 Background: Identifying novel biomarkers for early detection and prognosis would allow for improved survival outcomes in pancreatic cancer (PC). Arginase has been implicated in tumorigenesis through its production of L-ornithine. Its aberrant expression provides metabolites for tumor proliferation. Methods: In our study, 47 patients with metastatic PC, 29 with localized PC, and 50 healthy controls were enrolled at KU Cancer Center from 2000 to 2019. Overall survival (OS) was retrospectively assessed and correlated to serum protease levels at diagnosis. Protease expression was analyzed using nanobiosensors, which utilized fluorescent nanoparticles read by Spectra Scan to measure arginase, matrix metalloproteinases (MMPs), urokinase plasminogen activator (uPA), neutrophil elastase (NE), cathepsin B (CTSB), and cathepsin E (CTSE). Survival analysis was completed by Kaplan-Meier method. Results: Protease expression in localized vs. healthy cases was significant for (CTSB [p = 1.33E-10], MMP1 [p = 6.66E-21], MMP3 [p = 1.39E-13], MMP9 [p = 2.77E-08], NE [p = 0.00015], uPA [p = 2.55E-11]) and borderline significant for arginase [p = 0.05082]. Protease expression in metastatic vs. healthy cases was significant for (CTSB [p = 7.06E-06], MMP1 [p = 3.08E-12], MMP3 [p = 2.51E-14], MMP9 [p = 1.73E-08], NE [p = 0.01430], uPA [p = 0.00024]). Further, metastatic vs. localized cancer was significant for (arginase [p = 0.00034], CTSB [p = 0.00584], MMP1 [p = 2.03E-13], NE [p = 2.29E-10], and uPA [p = 6.58E-07]. Characteristics of 21 localized PC patients are illustrated in Table. Median OS was 25m in localized PC with low Arginase expression (mean < 215) vs 17.5m in high Arginase expression (p = 0.0477). Conclusions: Expression pattern of CTSB, MMP, NE, arginase, uPA proteases showed statistical significant difference for detection of localized PC vs metastatic PC vs healthy cases. In localized PC, lower arginase expression showed a statistically significant improvement in survival vs high arginase expression. Arginase expression was borderline significant for detection of localized PC vs healthy controls, likely due to small sample. Arginase as a potential biomarker for early detection and prognosis of PC warrants validation in a larger cohort. [Table: see text]

e16750 Background: GP-2250, a novel oxathiazine derivative, displayed apoptotic cytotoxicity against various tumor cell lines but not normal cells. It was therefore tested whether its antineoplastic potential - alone or in combination – could be leveraged specifically against pancreatic cancer. Methods: GP-2250 is a cancer metabolism-based therapeutic. It depleted metabolic energy through inhibition of the enzyme GAPDH (glyceraldehyde-3-phosphate dehydrogenase) which is rate limiting for aerobic glycolysis. ATP was decreased in a time- and dose-dependent manner in pancreatic tumor cell lines and ROS was increased. Mitochondrial dysfunction was triggered by an increased expression of Bax and decreased expression of Bcl2, leading to apoptosis. Cytotoxicity of GP-2250 was ROS-dependent. It was blocked by N-acetylcysteine. Results: GP-2250 substantially increased the sensitivity of pancreatic tumor cells to various chemotherapeutics in particular to gemcitabine (Gem). At doses which were inactive or barely active per se, the combination of GP-2250 and Gem caused striking cytotoxicity in patient-derived primary tumor cells in vitro, pointing to a strong synergy between the two agents. This finding was substantiated in vivo by patient-derived xenograft (PDX) studies in nude mice. While GP-2250 and Gem, given as monotherapy (500 mg/kg and 50 mg/kg respectively, 2x/week), showed only a limited antineoplastic response, the combination treatment resulted in a significantly higher anti-tumor activity as shown in further PDX. Tumor regression was found in 5 out of 9 PDX based on RECIST criteria. Stable disease was reached in 3 of the remaining grafts. In 1 xenograft, which was unresponsive to Gem, the combination treatment nevertheless achieved a reduction in tumor growth which significantly exceeded that of GP-2250 monotherapy. Conclusions: GP-2250 is a novel cancer metabolism-based therapeutic. GP-2250, in combination with Gem, strongly reduces tumor growth in patient-derived xenografts exceeding by far the response to monotherapy. GP-2250 is being evaluated in a Phase I clinical trial in patients diagnosed with advanced pancreatic cancer (Clinical Trial NCT03854110).

e16503 Background: Patients with gastrointestinal (GI) malignancies are at increased risk of developing splanchnic vein thrombosis (SVT). However, there is a dearth of information about the epidemiology of splanchnic vein thrombosis associated with specific GI malignancies. We sought to compare the differences in hospitalizations and mortality related to SVT among various GI malignancies. Methods: We included the top three discharge diagnoses to identify cases of splanchnic vein thrombosis (portal/mesenteric/hepatic and splenic vein thrombosis) along with a secondary diagnosis of a GI malignancy, namely esophagus, gastric, hepatic, colorectal, pancreatic and cholangiocarcinoma within the 2010-2014 Nationwide Inpatient Sample (NIS). Outcomes including mortality, hospitalization charges and length of stay were compared for different types of GI malignancies for SVT hospitalizations using the chi-square test. Complications related to SVT were also identified. Data was analyzed using STATA 15. Results: There were 320,804 total weighted admissions for SVT, of which 33,556 or 11.6% occurred in patients with GI malignancies. Hepatic cancer was the most common GI malignancy associated with SVT, responsible for 5.1% of all cases, followed by pancreatic cancer (2.76% cases). Hepatic vein thrombosis was the commonest type of SVT occurring in patients with GI malignancy. Hospitalizations for SVT in GI malignancies increased from 5743 in 2010 to 8415 in 2014, representing an increase of 46% over 5 years. On univariate analysis, patients with esophageal cancer and SVT had higher rates of cardiac arrest (3.5% vs 0.7% average for other cancers, p = 0.03), but there were no significant differences in rates of other complications studied including intracranial hemorrhage, requirement for blood transfusion, mechanical ventilation or variceal bleeding. Inpatient mortality rate for admission with SVT was the highest for esophageal cancer (21.1%), followed by gastric cancer (17%) against an average of 9.3% for all GI malignancies (p < 0.05).The median cost of an admission for a patient with SVT were highest for esophageal cancer $102,452, followed by colorectal cancer ($97,491) and the least for hepatic cancer ($67,007). Conclusions: We found that hospitalizations for splanchnic vein thrombosis in patients with GI related malignancies has been steadily increasing and represent a significant burden. Hospitalizations for SVT in esophageal, gastric and colorectal cancer patients bear poorer outcomes as compared to hepatic cancer.

e21048 Background: Spindle cell lung cancer (SpCC) is a rare type of NSCLC which portends a poor prognosis. Due to the rarity of diagnosis, there is a dearth of information about the epidemiology and overall survival for these patients. Methods: We performed a retrospective analysis using the SEER database from 1975-2016 to study the demographics, treatment modalities and outcomes for patients with locally advanced SpCC. Data regarding age, sex, race, pathological grade, staging, treatment, overall and disease specific survival was extracted. Hazards ratios were calculated to identify any difference in mortality between patients who received surgery alone versus those who received adjuvant chemotherapy or radiation. Results: A total of 936 cases of SpCC were identified, out of which 367 (39%) patients had locally advanced disease. 84% cases were diagnosed after the age of 60, with peak incidence occurring in the 70-74 age group. 87% were Caucasians, and 56% were males. 68% of the tumors were poorly differentiated. In terms of the treatment modalities for locally advanced SpCC, surgical resection was performed only in 58.5% cases. 27% patients received systemic chemotherapy, out of which 50% was in the adjuvant setting after surgery. 32 % patients received radiation therapy, only 38.5% of which was in the adjuvant setting. No statistically significant difference in mortality was seen in patients who received surgery alone vs adjuvant RT vs adjuvant chemotherapy. However, pts who did not receive surgery had a higher odds of mortality (OR = 4.2, p value 0.0001). Similarly, pts who only received chemotherapy alone had a higher odds of mortality vs those who received chemotherapy along with surgery (OR = 3.4, p-0.045), Overall survival was 25% for patients with localized disease, 9.5% for regional and only 2.6% for distant metastatic disease. For locally advanced SpCC, the observed cumulative 1-year survival was 54.8 % and declined to 29.2% after 5 years. Conclusions: Majority of the spindle cell carcinoma cases are poorly differentiated and present at an advanced stage at the time of diagnosis. For locally advanced SpCC, surgical resection can improve survival. Randomized trials are needed to test efficacy of adjuvant therapies.

4563 Background: Cabo targets multiple tyrosine kinases, including VEGFR, MET, and AXL, and has been reported to show immunomodulatory properties that may counteract tumor-induced immunosuppression, providing a rationale for combining it with PD-L1 inhibitors like durva. We conducted a phase Ib GI basket trial to evaluate the safety & efficacy of this regimen in advanced GE adenocarcinoma (GEA), colorectal cancer (CRC), & hepatocellular carcinoma (HCC). Methods: Patients received cabo and durva in 3+3 dose escalation then expansion to determine the dose limiting toxicity (DLT), Recommended Phase 2 Dose (RP2D), ORR, PFS & OS. Cabo was dosed at 20mg QD, 40mg QD, and 60mg QD in the first, second, and third cohorts respectively. Durva was dosed at 1500mg IV Q4W in all cohorts. DLT window was 28 days. Scans were obtained every 8 wks. Treatment beyond progression was allowed. Results: 23 Pts (16 M, 7 F), median age 60 yrs (range 33-79) were currently enrolled. 12 in the dose escalation cohort with cabo 20mg (6 pts), or 40mg (3 pts), or 60mg (3 pts). 11 pts enrolled in the dose expansion cohort with cabo 60mg. 8 pts had GEA, 13 pts had CRC, and 2 pts had HCC. Median number of prior chemotherapies was 3 (range 1-3). 3 pts were not evaluable for DLT due to missing ≥30% of DLT window doses, not related to DLT. No DLTs were observed. Drug-related Grade (G) 1&2 AEs included fatigue (83%), abnormal LFTs (39%), anorexia (26%), diarrhea (26%), nausea (13%), & hand foot syndrome (13%). One pt each developed drug related G3 hypertension, hyperthyroidism, thrombocytopenia, & thromboembolic event, all occurring outside the DLT window. 19 pts were evaluable for response: 4 PR (2 GEA & 2 CRC), 12 SD, 3 PD; ORR 21%; clinical benefit rate 84%; median time to PD 16 wks (range 8-40+). Conclusions: RP2D was determined to be Cabo 60mg QD and Durva 1500mg Q4W. Enrollment to phase I dose expansion is ongoing. RP2D may be adjusted based on additional experience & long-term tolerability. Early efficacy data was encouraging. This is an investigator-initiated trial funded by Exelixis & Astrazeneca. Clinical trial information: NCT03539822 .

e16803 Background: Pancreatic cancer has a dismal 5 year survival of 5-10%. Deaths commonly occur in-hospital as they present with acute complications. The purpose of this study is characterize this population compared to all pancreatic deaths, identify causes for admission, trends in palliative care utilization and its effect on costs and patient stay. Methods: From the years of 2002 to 2014, admissions for patients with a diagnosis of pancreatic cancer were identified using the National Inpatient Sample. Annual trends in death were compared to overalls deaths using SEER data. Trends in hospital length of stay (LOS) and total charges (TC) were assessed, as well as utilization of palliative care. The effect of palliative care utilization on hospital LOS and TC were also identified. Results: 97,389 (weighted) patient deaths occurred from 2002 to 2014, with 7,634 in 2002, compared to 7,200 in 2014. Compared to total overall deaths of 38,026 and 42,047 respectively. Signifying 25% (2002) to 21% (2014) total patients expiring in an in-patient setting. The most common billed primary diagnosis was sepsis at 15.5%, followed by acute renal failure and fluid disorder (12.5%) and liver failure (5.3%). Overall length of stay trended down from 9.0 days to 7.5 days (p < 0.001). And total charges for admission increased from $36,704 to $88,063 (p < 0.001). Palliative care consults increased from 12% in 2002 to 45% in 2014. In 2014, the TC for deaths among those who received palliative care consults was $52,612 (p < 0.001 when compared to all deaths). LOS among these patients also decreased from 7.5 days to 6.2 days. When looking at patients with sepsis who did not die, a palliative care consult decreased costs from $86,738 to $74,544 (p < 0.001). LOS was not significantly different at 8.8 days compared to 8.5 days (p = 0.15). Conclusions: A quarter of patients with pancreatic cancer die in an in-hospital setting. Palliative involvement decreased health care resource utilization. In reviewing patients who developed sepsis without in-hospital mortality, a palliative care consult decreased total charges of admission.

e16802 Background: PARP inhibitor (PARPi) has recently been approved for various cancers. However, trials have mostly recruited pts with germline BRCA (gBRCA) mutations, and it is unclear whether PARPi have similar efficacy in pts with somatic BRCA (sBRCA) mutations. We aimed to determine the efficacy of PARPi in pts with sBRCA mutations. Methods: Per PRISMA guidelines, systematic review of PubMed, Embase, Cochrane RCT, and Web of Science Collection was performed from inception thru Jan 2020 to identify studies. Our inclusion criteria were clinical trials and retrospective studies that reported use of PARPi in pts with both s and g BRCA mutations. We performed a meta-analysis comparing overall response rate and PFS with PARPi in pts with s versus g BRCA mutations. Results: After screening, 18 studies met our criteria for including both s and g BRCA mutations. Only 8 studies reported response rates for both s and g BRCA mutations (Table). In those studies, 24 out of 43 pts with sBRCA mutations (55.8%), and 69 out of 157 (43.9%) pts with gBRCA had a response to PARPi (pooled OR 1.13, p value = 0.399, I2 = 0). In all five studies that reported PFS, there was no obvious difference in outcomes between sBRCA (HR in these studies ranged 0.23 to 0.27) versus gBRCA (HR ranged 0.17 to 0.27), however a precise statistical analysis could not be done. Conclusions: Our meta-analysis and systematic review of the literature indicates similar outcomes of PARPi therapy in pts with s and g BRCA mutations. Investigation of use of PARPi therapy in a broader patient population, and the inclusion of sBRCA mutations in future clinical trials is essential. [Table: see text]

e16673 Background: Liver cancer and bile duct cancer remain among of the most deadly cancers with a dismal 5 year survival of 2-15%. Patients often end up dying in an in-hospital setting as a result of acute complication of cancer progression. The purpose of this study is to identify annual trends in this sub-population, including cause of admission, palliative care utilization and economic impact on healthcare. Methods: From the years of 2002 to 2014, admissions for patients with a diagnosis of liver and bile duct cancer were identified using the National Inpatient Sample. Annual trends in incidence were compared to overall deaths via the SEER database. Trends were identified in hospital length of stay (LOS), total charges (TC), as well as utilization of palliative care. The effect of palliative care utilization on hospital length of stay and costs were also studied. Results: 73,833 (weighted) patient deaths were recorded from 2002 to 2014, 4,577 in 2002, compared to 6,595 in 2014. Signifying 33% (2002) to 32% (2014) of total deaths related to these cancers. The most common billed primary diagnosis was liver failure at 16.9%, sepsis (15.3%) and renal failure (6.7%). Overall LOS trended down from 5.9 days to 4.8 days (p < 0.001). Palliative consults increased from 8.2% in 2002 to 39.4% in 2014. In 2014, the mean TC for those who received palliative consults versus those who did not was $52,612 and $64,388 respectively (p < 0.001). LOS among these patients did not significantly change at 5.9 and 5.6 days (p = 0.13). When looking at patients with sepsis who did not die, a palliative care consult decreased costs from $87,564 to $75,223 (p < 0.001). LOS was not significantly different at 8.8 days to 8.5 days (p = 0.15). Conclusions: A third of patients with liver and bile duct cancers die in an in-hospital setting. More effort needs to be undertaken to identify these patients and establish appropriate goals of care prior to such an event. Nonetheless, palliative involvement decreased health care resource utilization. Even among patients with sepsis who did not die, a palliative care consult decreased total charges of admission.

533 Background: Sorafenib or lenvatinib are the current frontline options for advanced HCC. Multiple biologic agents including multi-Tyrosine Kinase Inhibitors (TKI) cabozantinib & regorafenib have been recently approved for the previously treated population. Immunotherapy (IO) agent Nivolumab have also been approved in the same setting. Due to lack of prospective head to head comparison, there is no standardized way for alternating those agents beyond frontline. Methods: We performed a retrospective review of KU cancer registry. Patients with advanced HCC were divided into 2 groups based on the 2nd line systemic regimen (IO vs non-IO). Kaplan–Meier and Cox proportional hazards models were utilized to evaluate progression free survival (PFS) and overall survival (OS). Results: Between 2016-2019, 98 patients were identified, 41 received IO, while 57 received biologics. All patients had sorafinib as 1st line therapy. Most patients have ECOG 0-1 and Child-Pugh class A or B. 55% had hepatitis C, 6% hepatitis B and 27 % have history of alcohol misuse. Almost 50% of patients have received prior liver directed therapy. Comparing IO vs non-IO groups, median PFS was 3.9 months vs 3 months and median OS was 10 months vs 10 months. There was no statistically significant difference in PFS & OS but there was a delay separation in the survival curve favoring IO. Similar outcome was seen in a sub-group analysis of the hepatitis C patients. Conclusions: This retrospective comparative review of current 2nd line regimens is one of the first & largest studies reported. In this study population, IO was not superior to multi-TKI as a 2nd line regimen. The late survival curve separation favoring IO might suggest a delay IO effect in a subgroup of patients. Future studies should define specific biomarkers that could predict response and allow treatment to be optimized depending on patient and tumor characteristics. Furthermore, combining IO plus multi-TKI might be a promising next step in development. A number of ongoing trials are testing this strategy including our CAMILLA trial of Cabozantinib plus Durvalumab, currently enrolling patients at the University of Kansas Cancer Center NCT03539822 .

767 Background: High mortality associated with pancreatic ductal adenocarcinoma (PDA) warrants research into prognostic factors. We examined the relationship between the daily rate of change of CA19-9 over the first 90 days of treatment (DRC90) and pretreatment levels of neutrophils, lymphocytes, and platelets with overall survival (OS) and progression free survival (PFS) in patients with stage IV PDA that received chemotherapy. Methods: We retrospectively evaluated 102 locally advanced and metastatic PDA patients treated at KU Cancer Center between Jan 2011 and Sep 2019. We compared the ratio of pretreatment absolute neutrophil count to pretreatment absolute lymphocyte count (NLR) and the ratio between pretreatment platelet count to pretreatment absolute lymphocyte count (PLR) with OS and PFS. We also compared DRC90 to OS and PFS. Log-rank trend test using the mean of NLR, PLR, and DRC90 as the threshold for two groups within each variable. Results: Baseline demographics are shown in the table. Pts with ≥ mean NLR (4.6) had significantly lower OS [p = 0.0444] and PFS [p = 0.0483] than Pts below the mean. Pts with PLR ≥ mean (3.9) did not have significantly different OS [p = 0.507] or PFS [p = 0.643] than Pts below the mean. Pts with DRC90 ≥ mean (-1%) did not have significantly different OS [p = 0.342] or PFS [p = 0.313] than Pts below the mean. Conclusions: Pts with NLR ≥ mean (4.6) had significantly lower OS and PFS than Pts with NLR below the mean. This implies the possibility of NLR as a prognostic marker in PDA that could guide treatment approach but needs validation in a larger cohort. [Table: see text]

766 Background: Pancreatic intraepithelial neoplasms (PanINs) and intraductal papillary mucinous neoplasms (IPMNs) are common pancreatic adenocarcinoma precursor lesions. However, data regarding their respective associations with prognosis is lacking. Methods: We retrospectively evaluated 72 resected pancreatic adenocarcinoma cases at the KU Cancer Center between Aug 2009 and March 2019. Patients were divided into either one of two groups, PanIN or IPMN, based on the results of the surgical path report. We compared baseline characteristics, overall and progression free survival between the two groups, as well as OS and PFS based on local or distant tumor recurrence. Results: 52 patients had PanIN and 20 patients had IPMN. Demographic and baseline characteristics are as follows (PanIN/IPMN): Median age 62.5/69; Gender (male) 63%/65%; ECOG status (0-1) 98%/85%; pancreatic head tumors 87%/70%; pancreatic body tumors 6%/15%; pancreatic tail tumors 7%/15%; Abnormal CA19-9 at diagnosis 79%/67%; Comorbidity Index 5/5 respectively. Median PFS was 26.2 months (95% CI: 21.4-31.0) for PanIN and 74.3 months (95% CI: 15.7-132.9) for IPMN [p = 0.004]. Median OS was 70.3 months (95% CI: 35.4-105.2) for PanIN and 78.8 months (95% CI: 33.2-124.4) for IPMN [p = 0.013]. Within the PanIN group, median OS after recurrence was 71.3 months (95% CI: 68.8.-73.4) for local recurrence and 46.7 months (95% CI: 39.2-54.2) for distant recurrence [p = 0.330]. Conclusions: Patients who had a IPMN associated pancreatic cancer had better PFS and OS when compared to patients with PanIN associated pancreatic cancer. In patients with PanIN associated cancer that recurred, OS was better with local recurrence compared to distant recurrence but did not meet statistical significance. The results need to be validated in a larger cohort. [Table: see text]

e15798 Background: Understanding survival outcomes of various pathogenic mutations helps guide treatment decision making for patients. Classic pancreatic cancer mutations such as KRAS and TP53 have well documented survival outcomes while other mutations leading to DNA repair deficiency do not have well understood survival outcomes. Methods: We retrospectively evaluated survival outcomes of 70 pancreatic cancer patients who had their cancers genetically profiled by NGS methods. Patients with DNA repair deficiency harbored mutations in genes such as BRCA1 (1 Pts), BRCA2 (8 Pts), ATM (5 Pts), NBN (1 Pt), and BRIP1 (1 Pt). We compared baseline characteristics, tumor stage and clinical outcomes between patients with DNA repair deficiency versus DNA repair proficient cancer patients. Comparative survival analysis between the two groups was performed using Kaplan-Meir methods. Results: Baseline characteristics for all patients are recorded in (Table). Median OS is 24 months for DNA repair proficient group and 20 months for DNA repair deficient group. A comparison of Kaplan-Meir survival curves between the two groups yielded a p-value of 0.72. This is most likely due to sample size and different chemotherapy regimens which make it hard to retrospectively compare patient groups. Conclusions: Patients with mutated DNA repair genes did not have significantly worse survival. We are designing a clinical trial utilizing a PARP inhibitor, for these patients in order to better control for all factors in order to better ascertain any survival differences between the two groups. PARP inhibitor will create multiple single strand breaks which cancer cells deficient in DNA repair genes cannot repair and thus trigger cancer cell death[Table: see text]

TPS3165 Background: Mutations in p53 contribute to tumor progression. A rational approach is to destabilize mutant (m) p53. The group at the University of Kansas Cancer Center screened compounds that suppress m p53 in a preclinical model. Luciferase-based reporter assay identified statins as suppressors of m p53 expression. In vitro validation assay demonstrated atorvastatin (A) suppressed m p53 level and cell growth selectively; and depletion of mevalonic acid lead to degradation of m p53. These effects were limited to mutations in the conformation of p53, while wild-type and DNA contact mutations were not as sensitive to statin-induced degradation of p53. M p53 xenograft model confirmed that A could suppress tumor growth at a concentration that can decrease LDL level. The primary objective of this trial is to determine if A decreases the level of conformational m p53. The secondary objective is to assess the effects of A on Ki-67 and caspase-3 in conformational m p53 tumors. Methods: This is an open-label, window of opportunity pilot trial to see if A given for 1 to 4 weeks at a dose of 80 mg/day is sufficient to reduce the levels of conformational m p53 in the tumor tissues. Subjects with new diagnosis of malignancy with a planned surgical therapy, and subjects with previously treated AML, in between treatment regimens, are eligible. Tissues from solid tumors, and bone marrow or peripheral blood samples from AML will be used to screen for m p53 by immunohistochemistry (IHC). Subjects will receive A at 80 mg/day po for 1 to 4 weeks. Pharmacokinetics at pre-dose and 1-hour post-dose on Day 1 and on the day of surgery will be done. Mutational analysis using exome sequencing technique will be done on m p53. Using IHC, the amount of p53 in pre-treatment and post-treatment samples will be measured and compared simultaneously. The levels of Ki67 and caspace-3 will be tested and compared between pre-treatment and post-treatment samples in subjects with conformational m p53, between conformational and non-conformational m p53, and in wild-type p53 tumors. The trial is actively enrolling subjects. The results of this trial will determine further investigations on the role of atorvastatin in tumors with p53 mutations in a placebo-controlled, randomized trial. Clinical trial information: NCT03560882.

e15156 Background: There are differences in the incidence, clinical presentation, molecular pathogenesis and outcome of colorectal cancer (CRC) based on tumor location. Emerging research suggests that perioperative carcinoembryonic antigen (CEA) ratio (post-op/pre-op CEA) is a prognostic factor for CRC patients. We aimed to determine the association between tumor location, CEA ratio, smoking status and overall survival (OS) among patients with CRC. Methods: We analysed 323 patients who underwent resection for CRC at KUMC. After excluding those without pre- or post-operative CEA data, 162 patients were classified as either high ( > = 0.5) or low ( < 0.5) ratio. Primary outcomes were: 1) OS stratified by tumor location; 2) OS stratified by CEA ratio; and 3) whether the association between CEA ratio and OS differed by tumor location, after adjusting for stage and smoking status. Kaplan-Meier method was used to estimate survival rates, and Cox proportional hazards models for multivariate analysis. Results: The median age was 63 years (inter-quartile range 53-72), 61% male, 43% smokers, 73% left-sided tumors, median pre-operative CEA was 3.0 (IQR 1.5-7), and 64% had CEA ratio > = 0.5. The OS rates were 85.7% and 91.9% in patients with left-sided vs right-sided tumors respectively (log-rank p-value = 0.9). The OS rates were 83.5% and 91.5% in patients with high vs low CEA ratios respectively (log-rank p-value = 0.3). The effect of CEA ratio on OS was significantly different when stratified by tumor location (p-value for interaction < 0.001). However, in the stratified analysis, the n was too small to permit further inferential analysis. In multivariate analysis, both tumor location (HR 0.6; p = 0.5) and CEA ratio (HR 1.5; p = 0.5) were not significantly associated with OS after adjusting for smoking status and tumor stage. Smoking was significantly associated with higher rates of death (HR 3.9; p = 0.04) when adjusted for tumor location, CEA ratio, and tumor stage. Conclusions: There was no difference in OS between left versus right-sided tumors. The association between CEA ratio and OS was significantly modified by tumor location. However, to attribute this modification to left vs right warrants validation in a larger cohort as our sample size was limited. Smoking increases mortality irrespective of right vs left sided CRC.

TPS4168 Background: Attempts to improve therapy for patients with pancreatic adenocarcinoma with traditional chemotherapy have largely failed to meaningfully improve survival. Therefore, there is a critical need for identification of specific molecular changes that define prognosis and potentially guide therapy decisions. Defective DNA damage response pathways in pancreatic cancer represent a targeted opportunity for treatment. PARP inhibitors exert activity in tumor cells that may not be effectively able to repair initially single-stranded and cumulatively double-stranded DNA breaks and can have a heightened susceptibility in tumor cells over normal tissue. This concept is referred to as synthetic lethality. Niraparib is an orally available, potent, highly selective PARP-1 and -2 inhibitor. We are studying the efficacy of Niraparib in pancreatic cancer patients that harbor DNA repair defects. Methods: This study is funded by a research grant from TESARO. Pre-screening of patients to find biomarker positive patients is funded by KU Cancer Center. This is a phase II open label single arm trial in metastatic pancreatic cancer patients with germline or somatic mutations, either already known, or tested after consent to pre-screening tumor tissue analysis in BRCA1/2, PALB2, ATM, NBN, ATR, BRIP1, IDH1/2, RAD51, RAD51B/C/D, RAD54L, CDK12, BARD1, FAM175A, BAP1, CHEK1/2, GEN1, MRE11A, XRCC2, SHFM1, FANCD2, FANCA, FANCC, FANCG, RPA1, ARID1A. Patients are being treated with Niraparib 300mg or 200mg by mouth daily for 28 days (1 cycle = 28 days) (200mg dose is for participants whose baseline weight is < 77 kg [169.756 lbs] or baseline platelet count is < 150,000 µL). The primary objective is to assess antitumor efficacy of niraparib using Objective Response Rate per RECIST 1.1. Secondary objectives include PFS, OS, DCR, DOR, and safety. Eligible patients received > 1 line of therapy, no prior PARP inhibitor(s), have measurable disease, and ECOG PS 0-1. Accrual target enrollment of 18 patients over a period of 24 months with a study duration of 30 months. Correlative studies include assessment of pharmacokinetics, circulating tumor cells and storing samples for future research. The trial is currently enrolling. Clinical trial information: NCT03553004.

TPS56 Background: GI malignancies including Gastric (GC), esophageal (EAC), colon (CRC), and hepatocellular carcinoma (HCC) remain a significant health problem in the US & globally. The current standard upfront therapy has < 12 month median survival. Hypoxia mimetic agents such as VEGFR targeted therapy down regulate the angiogenesis pathway and sensitize tumors to chemotherapy. Cabozantinib targets multiple tyrosine kinases, including VEGFR, MET, and AXL, and has been reported to show immunomodulatory properties that may counteract tumor-induced immunosuppression, providing a rationale for combining it with PD-1 or PD-L1 inhibitors. We believe that modulating the tumor microenvironment with small molecule inhibitors like cabozantinib will have synergistic effect when combined with checkpoint based immunotherapy like durvaluamb in patients with chemo refractory GC, EAC, CRC and HCC. Methods: The study is funded by research grants from both Exelixis and AstraZeneca. A phase 1 dose escalation is followed by an open-label single arm expansion. Dose escalation will be conducted using a 3+3 design. Starting dose for cabozantinib is 20mg daily. Single agent durvalumab MTD has been used, given that a regimen with full dose durvalumab may be better accepted as a backbone for comparison in future studies. MTD of cabozantinib in combination with standard dose durvalumab will be defined as the highest safely tolerated dose where 0/6 or 1/6 patients experience a DLT and ≥ 2 patients have experienced a DLT at the next higher dose level. Primary objective is to determine the Recommended Phase 2 Dose (RP2D). Secondary objectives include safety, ORR, PFS and OS. Eligible patients received > 1 line of therapy, no prior checkpoint inhibitors, have measurable disease, & ECOG PS 0-1. Phase 1 dose escalation will enroll 9-18 patients and the expansion phase will enroll 12-21 patients. Correlative studies include assessment of tumor microenvironment, angiogenesis & immune molecular markers. The dose escalation phase is currently enrolling. Clinical trial information: NCT03539822.

296 Background: Early detection of pancreatic cancer would allow for improved survival outcomes. Methods: We retrospectively evaluated serum protease levels and the survival of 15 pancreatic cancer patient samples (6 localized and 9 metastatic) at the KU Cancer Center. Available serum protease assays measured matrix metalloproteinases (MMPs), urokinase plasminogen activator (uPA), arginase, neutrophil elastase (NE), cathepsin B (CTSB) and cathepsin E (CTSE). The assays utilize fluorescent nanoparticle-based nanobiosensors which increase fluorescence upon posttranslational modification or enzymatic cleavage of targeted compounds and were read by a Spectral scan plate reader. Survival analysis was performed using Kaplan-Meier methods. Results: Baseline characteristics for all 15 patients are in (Table). Median OS was 18.8m in patients with high CTSB expression (mean >51968.9) vs 9.7m in low CTSB expression (p=0.04). Similarly, median OS was 20.4m in high CTSE expression (>123264.8) vs 10m in low CTSE expression (p=0.05). Whereas, median OS was 16.3m in low NE expression (mean

165 Background: Neoadjuvant chemoradiation followed by esophagectomy is the standard of care in advanced EC. While 5FU based chemoradiation has been a common regimen in the past, its utilization has declined in recent years as the CROSS trial study regimen of carboplatin/paclitaxel has become widely adopted. A prospective evaluation of the CROSS regimen compared to the 5FU based regimen was never performed. The aim of this study is to report our institutional experience with these two chemotherapy regimens. To the best of our knowledge, this is the largest retrospective study comparing the two types of chemotherapy regimens. Methods: We performed an IRB-approved retrospective review of a prospectively maintained institutional cancer registry. EC patients who completed trimodality therapy with either carboplatin/paclitaxel or 5FU/platinum were identified and divided into groups based on their chemotherapy regimens. Multivariable logistic regression was used to analyze pathologic complete response (pCR) rates, while the Kaplan–Meier and Cox proportional hazards models were utilized to evaluate DFS and OS. Analytical models were adjusted for age, stage, radiation dose, histology sub-type, and time interval from completion of neoadjuvant therapy to surgery. Results: 224 patients treated between January of 2007 and July of 2017 were identified . Of this group, 139(62%) had received Carbo/Taxol, while 85 (37%) had received 5FU/platinum. There was no increase in the odds of pCR with 5FU based chemo compared to CROSS regimen (OR = 2.68, P = 0.671). Furthermore, the OS and DFS of 159 patients(80 5FU/platinum, 77 carbo/taxol) with median follow up of ~ 5 yrs were not statistically different with HR 1.08 (0.6-1.7) and P value 0.71. Conclusions: Neoadjuvant chemoradiation with 5FU/platinum in resectable EC is not associated with higher rates of pCR, DFS and OS compared to the CROSS regimen of carbo/taxol. Those findings will need to be validated in a larger cohort.

709 Background: Standard therapy of locally advanced rectal cancer (LARC) includes concurrent chemoradiotherapy (CRT) followed by surgery and adjuvant chemotherapy (CT). An alternative strategy known as Total Neoadjuvant Therapy (TNT) involves administration of CRT and CT prior to surgery with the goal of delivering uninterrupted systemic therapy to eradicate micrometastases. A comparison of these two approaches has not been reviewed previously. Methods: Following PRISMA guidelines, a systematic review of PubMed and Web of Science was performed from inception through September 2018 to identify studies which compared TNT with CRT in LARC. The outcomes of interest were pathologic complete response rates (PCR), sphincter-preserving surgery rates, ileostomy rates, disease-free survival (DFS) and overall survival (OS). Summary hazard ratios (HR) with 95% confidence intervals (CI) were estimated using a random effects model and heterogeneity was estimated using the inconsistency index (I2). Results: After reviewing 2,163 reports, 5 studies which compared CRT and TNT were selected for inclusion. These were reported from Europe and the United States and included a total of 1,134 patients of whom 552 received TNT. The pooled prevalence of PCR was 32.4% (range 17-28%) in the TNT group and 22.3% (range 4-21%) in the CRT group. In a meta-analysis of five studies, TNT was associated with a higher chance of achieving a pathologic complete response (OR 1.92; 95% CI 1.44- 2.57, I2= 22%). Recipients of TNT also had higher odds of receiving sphincter-sparing surgery (5 studies, OR 1.92; 95% CI 1.44- 2.57, I2= 0%) and lower odds of requiring an ileostomy (2 studies, OR 0.72; 95% CI 0.54- 0.96, I2= 0%). Only one study presented data on DFS and noted improved DFS in patients who received TNT (OR 0.72; 95% CI 0.54- 0.96). The impact of TNT relative to standard therapy on overall survival was not reported. Conclusions: Total Neoadjuvant Therapy is a promising strategy in locally advanced rectal cancer, with superior PCR rates compared to standard therapy. However, the long term impact on disease recurrence and overall survival remain unclear. These are best be studied in a prospective randomized clinical trial.