Graham, Rondell P.

4048 Background: PolyPEPI1018 is an off-the-shelf, multi-peptide vaccine containing 12 immunogenic epitopes derived from 7 cancer testis antigens (CTAs) frequently expressed in patients with CRC. Here we report the final results of the phase I study of PolyPEPI1018 vaccine as an add-on to maintenance therapy in mCRC patients. Methods: 11 patients with MSS mCRC were vaccinated with PolyPEPI1018 just after the transition to maintenance therapy with fluoropyrimidine/bevacizumab after first-line combo chemotherapy and bevacizumab. Part A: n = 5, single dose; Part B: n = 6, 3 doses, Q12W. Primary endpoint was safety. Immunomonitoring was performed at both blood and tumor levels, as well as prospectively predicted. Results: The vaccine was well tolerated; most common side effects were transient skin reactions. No vaccine-related SAE occurred. Pre-existing immune responses were boosted by the vaccine for 7/10 patients. De novo responses were also induced, overall, 80% of patients had CD8+ T cell responses against at least 3 CTAs. The magnitude of immune responses as well as the density and the ratio of CD8+/CD3+ tumor infiltrating T cells increased with multiple vaccine doses. Three patients had objective tumor response according to RECIST v1.1: one of them in the single dose group and two of them in the 3 doses group. Both patients in the 3 doses group qualified for curative surgery. One of them had no viable tumor cells in his primary tumor at the time of surgery. Post-trial follow-up revealed PFS of at least 12 months for 3 patients. mPFS was longer for patients receiving multiple doses (9.9 months) compared to single dose (6.1 months). Both measured and predicted multiantigenic immune responses tend to correlate with PFS and tumor volume reduction. Conclusions: PolyPEPI1018 was effective in restoring immunological responses to CTAs in patients’ with spontaneous immunity against. Treatment with PolyPEPI1018 vaccine and maintenance therapy was safe and demonstrated evidence of early clinical activity in MSS mCRC tumors. Data support further randomized trials with PolyPEPI1018. Clinical trial information: NCT03391232 .

e13681 Background: Cell-free tumor DNA or circulating tumor DNA tests are increasingly used in clinical care to detect somatic mutations from solid tumors. However, data on laboratory performance characteristics using standardized samples is limited. Methods: Well-characterized reference materials were used for the College of American Pathologists (CAP) cell-free tumor DNA proficiency testing surveys, which consisted of stabilized DNA fragmented to simulate cell-free DNA in a synthetic plasma matrix. For the 2018A, 2018B, 2019A and 2019B surveys, laboratories tested for hotspot mutations (single and dinucleotide sequence changes) in EGFR, BRAF, KRAS, NRAS, and IDH1 at variant allele fractions ranging from 0.1% - 1.0%. As per CAP proficiency testing standards, results were scored according to the known mutation(s) engineered at designated variant allele fractions in each PT sample. Nine laboratories were excluded from analysis because they provided incomplete results. Statistical significance was calculated using a multivariate logistic regression model. Results: In 2018 and 2019, 101 laboratories submitted survey results for at least one proficiency testing mailing. There were 5088 total proficiency testing responses for EGFR, BRAF, KRAS, NRAS, and IDH1 mutations across 12 different samples. For the 3585 responses submitted for BRAF, KRAS, NRAS, and IDH1, sensitivity ranged from 94.6 – 100%, while specificity exceeded 99%. There were no significant differences in performance between analytical methodologies for BRAF, KRAS, NRAS, and IDH1 mutations. Performance characteristics for EGFR mutations among the 1503 responses showed a combined sensitivity of 87.1% and specificity of 98.7%. For laboratories detecting mutations in EGFR, next-generation sequencing methods exhibited a sensitivity (true positivity) of 95.7% while the sensitivity of non-NGS methods was lower at 81.7% ( P= 0.02). Conclusions: These findings demonstrate high sensitivity and specificity for clinical laboratories performing cell-free tumor DNA tests. For EGFR mutations, NGS outperformed non-NGS methods. These data suggest excellent overall agreement among laboratories performing clinical cell-free tumor DNA tests. Further investigation across variant allele fractions and additional variant types is warranted.

Angiosarcoma of the colon is rare, as is colonic amyloidosis. To our knowledge, there have been no reported cases of angiosarcoma arising in association with amyloid deposition. Herein, we described a case of 77-year-old man who presented with hematochezia, and a sigmoid mass was found on colonoscopy. Histologic examination of the resected specimen showed extensive nodular deposition of AL-lambda amyloid material in the colonic wall, as well as high-grade angiosarcoma which was closely intermingled with the amyloid deposits. While the occurrence of both colonic amyloidosis and angiosarcoma in this patient may represent pure coincidence, given the intimate association of the angiosarcoma and the amyloid deposition and the rarity of both of these lesions, we hypothesize that angiosarcoma could be secondary to amyloid deposition.

751 Background: Pancreatic adenosquamous carcinomas (PASC) are rare malignancies, with limited evidence regarding best treatment options. The Mayo Clinic Pancreatic Cancer SPORE Registry was utilized to compare/contrast outcomes for pancreatic cancers with a squamous component to pancreatic ductal adenocarcinoma (PDAC). Methods: Patients were identified from the SPORE Registry (2000-2019), were reviewed and confirmed by expert pathologists. Demographic and clinical information was ascertained from medical records and risk factor questionnaires. A case control study of patients with PASC vs PDAC was constructed. PASC patients were also followed for outcome and treatment records were extracted. Results: Of 2,584 total patients with pancreatic cancers, 45 cases of PASC and 2,438 with PDAC were identified. There were no differences in age (median 69 vs 67 years, p=0.42), sex (male 64.4vs 56.6%, p= 0.29), BMI (27.41 vs 27.78 kg/m2, P=0.50), or ever-smoking (61.0 vs 55.3%, P=0.47), with a borderline association with reported diabetes (17.8 vs 29.9%, p =0.08). Compared to PDAC, PASCs were more likely to involve the body/tail (48.9 vs 33.2%, p= 0.02) and had poorer overall survival, adjusted for age, gender, and stage (median 7.1 m vs 12.8m, HR 1.89, p=0.0004). Of 9 PASC pts treated with neoadjuvant intent, 4 were surgically resected, median survival was 7 months. Eleven pts underwent upfront surgery, with variable adjuvant treatments. Median OS post- surgery was 18m (range 7-51). Of 14 patients presenting with metastatic disease, median survival was 4.5 m (range 1-22). With regard to systemic chemotherapy, for neoadjuvant or metastatic disease median duration of treatment was 3 months (range 0.5-7m) for Gemcitabine + nab-paclitaxel (N=9) and also 3 months (range 2-6m) for FOLFIRINOX (N-13). Conclusions: The diagnosis of PASC carries an even poorer outcome than pancreatic adenocarcinoma. Tumors are more likely to arise in the distal pancreas, and patients may be less likely to report associated diabetes. Limited antitumor activity was noted with multi-agent chemotherapeutic regimens. Prospective trials will be needed to clarify choice of regimen in the future.

617 Background: Currently, there is no published data on the efficacy of SSAs for well-differentiated G3 NETs. Randomized trials have demonstrated a progression-free survival (PFS) benefit and limited tumor response for lower grade 1-2 NETs, but the optimal systemic therapy for metastatic/unresectable G3 NETs is unknown. We sought to evaluate the efficacy of SSAs in G3 NETs. Methods: We performed a retrospective analysis of Mayo Clinic patients treated with SSAs for metastatic/unresectable G3 NETs, querying data from 1992 - present. Inclusion criteria were: centrally reviewed pathology confirming well-differentiated morphology, G3 based on WHO classification, SSA monotherapy, and radiological data available to assess response. Patients who had prior lines of treatment were included as long they subsequently were treated with single-agent SSA. Poorly-differentiated tumors were excluded. The primary endpoint was PFS. Best overall response was determined by radiographic regression, stabilization, or progression of tumor size. Results: Ninety patient records were reviewed, with 14 meeting inclusion criteria (diagnosed 2014 – 2018). Median Ki-67 proliferative index was 25%. Two patients (14%) had a partial response to SSA therapy, five (36%) had stable disease, and seven (50%) had progressive disease. The estimated median PFS was 4.4 months (95% CI 2.9 – 24). Of the 7 patients with stable disease or partial response, the median time to progression was 8.7 months. Three patients had stable disease for greater than 9 months (24, 29 and 42 months, respectively). Overall survival was not estimable. There was no association of Ki-67 index with PFS based on a proportional hazards model. Conclusions: This is the first report on the efficacy of SSAs for G3 NETs. Although half of the patients in our series had at least stable disease, the PFS was modest at only 4.4 months. Given their favorable side-effect profile compared to cytotoxic chemotherapy, SSAs may present an attractive option to be further explored in a prospective fashion. We are presently updating this data by reassessing response using RECIST criteria.

362 Background: Most pts with mGEA do not respond to ICI or ramucirumab/paclitaxel (RAM/TAX). Emerging data suggest that ICI may potentiate subsequent therapy (Tx). In KN059 we unexpectedly observed durable responses in 2 pts on RAM/TAX after ICI (PMID 29674442). We examined if ICI impacts efficacy of subsequent RAM/TAX in a larger cohort and explored alterations in the tumor microenvironment. Methods: All pts with mGEA at Mayo Clinic (MN, AZ, FL) who received RAM/TAX (2014-19) were included. We compared best objective response rate (ORR: complete [CR] or partial response [PR]) per RECIST1.1 and overall survival (OS) in pts who received RAM/TAX with (Group A) vs without (Group B) immediately preceding PD-1 blockade. Chi square and multivariate logistic and Cox regression were used. We adjusted for total lines of Tx received and the line of Tx in which RAM/TAX was given, to control for potential differences in the biology of heavily treated pts or those receiving RAM/TAX as 2nd vs 3rd line Tx. We also adjusted for age and ECOG PS. Results: In total 87 pts met inclusion criteria. In the 19 pts (Group A) who received RAM/TAX (usually as 3rd line Tx) after progression on ICI, there was a 77% relative risk reduction of death after initiation of RAM/TAX compared to the 68 pts (Group B) who received RAM/TAX (usually as 2nd line) without preceding ICI (median OS 15.0 vs 7.6 mo; HR 0.23; P < .001). The ORR was also significantly higher (58% vs 18%; P < .001) including the CR rate (16% vs 1%; P = .006). Two CRs in Group A are ongoing (10 - 34 mo). Of note, 95% of Group A pts did not respond to ICI, and all had irRECIST progression on ICI. Immunofluorescent analysis from a responder showed 65-fold reduction (post vs pre-Tx) in intratumoral density of immunosuppressive FOXP3+ Tregs, with preserved density of antitumor CD8+ T cells. Conclusions: Higher than expected efficacy was observed on RAM/TAX when immediately preceded by ICI, suggesting ICI may enhance efficacy of subsequent anti-VEGFR/taxane therapy. This serial immunotherapy combination may be a novel option for pts with primary resistance to ICI and will be tested prospectively in a new randomized phase 2 trial (NCT04069273).

Objective. We explore the clinicopathologic features of Ewing sarcoma (ES) presenting in older adulthood. Methods. Cases of molecularly confirmed ES arising in patients aged ≥40 years were evaluated. Results. Fifty patients were identified (33 males/17 females; 41-86 years). The majority of tumors (41) arose at extraskeletal sites, while 9 were bone primaries. Twenty-eight cases showed nested architecture, while the remaining cases showed sheet-like growth. Tumor cytology was categorized as conventional (n = 39), crushed (n = 5), clear cell (n = 4), rhabdoid (n = 3), and epithelioid (n = 2). Fifty percent had necrosis, while rosettes were noted in 1 case. Immunostains performed ranged from 1 to 28 (median = 10). Follow-up (n = 43, 1-147 months) revealed 15 patients with metastasis. Conclusion. Although rare, ES should be considered in the differential diagnosis for round cell malignancies in older adult patients. In this cohort, ES is most often extraskeletal, and may show unusual morphologic features, closely simulating more common neoplasms in this age group.

Objective This study was designed to evaluate the roles of microRNAs (miRNAs) in slow transit constipation (STC). Design All human tissue samples were from the muscularis externa of the colon. Expression of 372 miRNAs was examined in a discovery cohort of four patients with STC versus three age/sex-matched controls by a quantitative PCR array. Upregulated miRNAs were examined by quantitative reverse transcription qPCR (RT-qPCR) in a validation cohort of seven patients with STC and age/sex-matched controls. The effect of a highly differentially expressed miRNA on a custom human smooth muscle cell line was examined in vitro by RT-qPCR, electrophysiology, traction force microscopy, and ex vivo by lentiviral transduction in rat muscularis externa organotypic cultures. Results The expression of 13 miRNAs was increased in STC samples. Of those miRNAs, four were predicted to target SCN5A , the gene that encodes the Na + channel Na V 1.5. The expression of SCN5A mRNA was decreased in STC samples. Let-7f significantly decreased Na + current density in vitro in human smooth muscle cells. In rat muscularis externa organotypic cultures, overexpression of let-7f resulted in reduced frequency and amplitude of contraction. Conclusions A small group of miRNAs is upregulated in STC, and many of these miRNAs target the SCN5A-encoded Na + channel Na V 1.5. Within this set, a novel Na V 1.5 regulator, let-7f, resulted in decreased Na V 1.5 expression, current density and reduced motility of GI smooth muscle. These results suggest Na V 1.5 and miRNAs as novel diagnostic and potential therapeutic targets in STC.

INTRODUCTION: Colorectal cancer (CRC) screening is broadly endorsed for individuals 50–75 years of age. Given recent increases in CRC incidence and mortality for patients <50 years old, the American Cancer Society now recommends initiation of average-risk CRC screening at age 45 (Wolf AMD CA Cancer J Clin 2018). The performance of recommended tests, including the multi-target stool DNA (mt-sDNA; marketed as Cologuard™) test, has been minimally described in populations under 50 years old. To address this knowledge gap, we conducted a retrospective study of DNA markers (KRAS, BMP3, NDRG4) included in the mt-sDNA assay to quantify and compare tissue marker levels in sporadic CRC cases and normal colon controls from patients in 45–49 and 50–64 year-old age groups. Our primary hypothesis was that mt-sDNA marker levels would be statistically similar between cases in both age groups. METHODS: Patient samples were identified from a pathology tissue archive at Mayo Clinic, excluding those with inflammatory bowel disease or familial cancer syndromes. Formalin-fixed, paraffin-embedded tissues were reviewed by a GI pathologist for 45–49 year-old (n = 90) and 50–64 year-old (n = 96) CRC cases; normal colon samples (45–49, n = 76; 50–64, n = 92) were also included for comparison. DNA extracted from core punches was blinded and processed in a randomized order. Quantitative Allele specific Realtime Target and Signal amplification (QuARTS) assays were used to quantify KRAS mutations, and BMP3 and NDRG4 methylation. Marker level in each sample was reported as the QuARTS product of mutant KRAS , methylated BMP3 or NDRG4 relative to the ACTB product in each sample, reported as a percentage. At least 60 patients per group provided sufficient power to demonstrate differences in marker levels of ±25% between case groups and between control groups using the Wilcoxon Rank Sum test. RESULTS: CRC cases in the 45–49 vs 50–64 year-old groups did not differ significantly with respect to sex ( P = 0.77) or race/ethnicity ( P = 0.06); left-sided CRC was more common in the 45–49 vs 50–64 year-old group (87% vs. 72%; P = 0.02). None of the marker levels were statistically different between cases of the younger vs older groups or between controls of these groups (Figure 1). CONCLUSION: These data support the expectation of no age-related differences in CRC detection by mt-sDNA, as the relevant molecular markers, when they occur, are present at similar levels across age groups 45–49 and 50–64 years.

Background Pancreatoblastoma (PBL) is a rare malignant pancreatic tumor seen predominantly in childhood, and its cytologic diagnosis remains challenging. Methods Twelve fine‐needle‐aspirations from 11 adults were analyzed. Results In total, 6 men and 5 women (median age, 45 years; age range, 32‐60 years) had tumors measuring a median 5.6 cm (range, 2.5‐12 cm) located in the pancreatic head (n = 7) or tail (n = 4), including 3 with familial adenomatous polyposis (FAP)/FAP‐related syndromes and 4 with metastasis at diagnosis. The median follow‐up was 39.8 months (range, 0.8‐348 months), and 5 patients died of disease. The original cytology diagnoses were: PBL (n = 2), neuroendocrine neoplasm (n = 2), poorly differentiated neuroendocrine carcinoma (n = 2), well differentiated neuroendocrine tumor (n = 1), poorly differentiated carcinoma (n = 2), “positive for malignancy” (n = 1), acinar cell carcinoma (n = 1), and epithelioid neoplasm with endocrine and acinar differentiation versus PBL (n = 1). Universal cytopathologic findings included hypercellularity; 3‐dimensional clusters; and single, monotonous, blast‐like cells that were from 1.5 to 2.0 times the size of red blood cells with high nuclear‐to‐cytoplasmic ratio, fine chromatin, small, distinct nucleoli, and a resemblance to well differentiated neuroendocrine tumor and poorly differentiated neuroendocrine carcinoma. Branching pseudopapillae (n = 7) and grooved nuclei (n = 3) raised the differential diagnosis of solid‐pseudopapillary neoplasm, but with more atypia. Uncommon features included pleomorphism (n = 4) and numerous mitoses (n = 1). Squamoid morules were seen on smears (n = 5) or cell blocks (n = 6) in 70% of patients and were characterized by epithelioid cells with elongated, streaming nuclei, fine chromatin, absent nucleoli, and positive nuclear β‐catenin (n = 6 of 8). The median Ki‐67 index was 21% (range, 2%‐70%), and neuroendocrine marker expression was common (100%), but acinar markers were variable (63%). Conclusions A combination of cytologic findings in PBL, including a predominant population of primitive blast‐like cells, subtle squamoid morules, frequent neuroendocrine and variable acinar phenotype, should facilitate accurate cytologic diagnosis and distinction from common mimics.