Bleeker, Jonathan

TPS4175 Background: Platinum/fluoropyrimidine (FP) with PD-1 inhibitor is now established as the standard first-line therapy for most patients with advanced HER2 negative gastric, esophageal, and gastroesophageal junction (GEJ) adenocarcinoma. The FP doublet chemotherapy backbone, most commonly FOLFOX, is recommended based on marginal survival benefits of taxane-containing triplet therapy at the cost of increased toxicity. Although several agents are approved as standard treatment options in later lines, only about 10% of patients in the US receive a third line of therapy, and very few have the opportunity to benefit from all approved agents in this disease. This challenge will only increase as more agents and targeted therapies are approved. FOLFIRINOX is a triplet regimen commonly used in gastrointestinal malignancies with established safety and often with superior efficacy. Two single arm phase II trials of FOLFIRINOX have reported promising activity in GEA with manageable toxicities. Reported median overall survival (OS) from these trials is around 15 months, which is superior to what is expected from FP doublet first-line chemotherapy (about 11 months). We hypothesize that using upfront triplet therapy with a more effective and less toxic triplet combination will result in improved patient outcomes by optimization of available therapies in GEA. Methods: This study is a randomized phase III, open-label, multicenter clinical trial with the primary objective to determine whether modified FOLFIRINOX (FU 2400 mg/m2, leucovorin 400 mg/m2, oxaliplatin 85 mg/m2, irinotecan 150 mg/m2 every 2 weeks) as first-line treatment improves OS compared to mFOLFOX in patients with advanced HER2 negative GEA. Eligible patients will have unresectable or metastatic adenocarcinoma of esophagus, GEJ, or stomach with no prior systemic treatment. Patients must have adequate organ function, and measurable or evaluable disease as defined by RECIST 1.1. Prior neoadjuvant or adjuvant therapy is allowed if completed at least 1 year prior to registration. Patients who will receive nivolumab (mandatory for PD-L1 combined positive score > 5) in addition to chemotherapy must not have any contraindications to immune checkpoint inhibitors. Patients with treated, asymptomatic and stable brain metastases are eligible. The total sample size is 382 evaluable patients for this study (191 patients per arm). The study accrual began January 2023. Clinical trial information: NCT05677490 .

BACKGROUND A patient-derived organoid (PDO) platform may serve as a promising tool for translational cancer research. In this study, we evaluated PDO’s ability to predict clinical response to gastrointestinal (GI) cancers. METHODS We generated PDOs from primary and metastatic lesions of patients with GI cancers, including pancreatic ductal adenocarcinoma, colorectal adenocarcinoma, and cholangiocarcinoma. We compared PDO response with the observed clinical response for donor patients to the same treatments. RESULTS We report an approximately 80% concordance rate between PDO and donor tumor response. Importantly, we found a profound influence of culture media on PDO phenotype, where we showed a significant difference in response to standard-of-care chemotherapies, distinct morphologies, and transcriptomes between media within the same PDO cultures. CONCLUSION While we demonstrate a high concordance rate between donor tumor and PDO, these studies also showed the important role of culture media when using PDOs to inform treatment selection and predict response across a spectrum of GI cancers. TRIAL REGISTRATION Not applicable. FUNDING The Joan F. & Richard A. Abdoo Family Fund in Colorectal Cancer Research, GI Cancer program of the Mayo Clinic Cancer Center, Mayo Clinic SPORE in Pancreatic Cancer, Center of Individualized Medicine (Mayo Clinic), Department of Laboratory Medicine and Pathology (Mayo Clinic), Incyte Pharmaceuticals and Mayo Clinic Hepatobiliary SPORE, University of Minnesota-Mayo Clinic Partnership, and the Early Therapeutic program (Department of Oncology, Mayo Clinic).

PURPOSE To update the ASCO guideline on the recommended prevention and treatment approaches in the management of chemotherapy-induced peripheral neuropathy (CIPN) in adult cancer survivors. METHODS An Expert Panel conducted targeted systematic literature reviews to identify new studies. RESULTS The search strategy identified 257 new references, which led to a full-text review of 87 manuscripts. A total of 3 systematic reviews, 2 with meta-analyses, and 28 primary trials for prevention of CIPN in addition to 14 primary trials related to treatment of established CIPN, are included in this update. RECOMMENDATIONS The identified data reconfirmed that no agents are recommended for the prevention of CIPN. The use of acetyl-l-carnitine for the prevention of CIPN in patients with cancer should be discouraged. Furthermore, clinicians should assess the appropriateness of dose delaying, dose reduction, substitutions, or stopping chemotherapy in patients who develop intolerable neuropathy and/or functional impairment. Duloxetine is the only agent that has appropriate evidence to support its use for patients with established painful CIPN. Nonetheless, the amount of benefit from duloxetine is limited. Additional information is available at www.asco.org/survivorship-guidelines .

Abstract Abstract 4332 Background: Hemolytic anemia caused by the passive transfer of isoagglutinins, typically anti-A or anti-B antibodies, is a rare and likely under-recognized complication of intravenous immunoglobulin (IVIG) therapy. The Immunohematology Reference Laboratory at Mayo Clinic Rochester has detected 12 cases of passively acquired isoagglutinins following IVIG administration during the first seven months of this calendar year (January-July 2011). This frequency of detection is much higher than that seen over the three prior calendar years, when only two cases were detected. Given this “cluster” of isoagglutinin detection, we analyzed these 12 cases to determine the clinical significance of this finding as well as characteristics of resulting cases of hemolytic anemia. Materials/Methods: Charts of patients identified as having passively acquired isoagglutinins following IVIG therapy were analyzed for evidence of a significant drop in hemoglobin (defined as >1 g/dL) as well as evidence of hemolysis (elevated lactate dehydrogenase, reticulocytosis, spherocytosis, decreased haptoglobin, indirect hyperbilirubinemia, hemoglobinemia and hemoglobinuria). Cases with a >1 g/dL drop in hemoglobin and two or more of the above findings consistent with hemolysis were considered “confirmed” cases. Those with passively acquired isoagglutinins but without evidence of hemolysis or with an alternative explanation for anemia were considered “asymptomatic”, and cases without a full hemolytic evaluation but still clinically suspicious were categorized as “suspected” cases. Results: Passively acquired isoagglutinins were detected in 11 patients following 12 separate treatments with IVIG (one patient had antibodies noted after two separate treatment episodes >3 weeks apart). Four of these 12 cases were categorized as “confirmed” cases, five as “suspected” and three as “asymptomatic”. The indication for IVIG use was a neurologic disorder in three of the four confirmed cases, with immune thrombocytopenic purpura (ITP) being the indication in the other case. In suspected cases, ITP was the indication in four of the five cases, with the fifth patient carrying a diagnosis of Kawasaki’s disease. During the 7 month period reviewed, 250 unique patients were treated with IVIG, meaning 1.6% of patients (4/250) had confirmed IVIG-associated hemolytic anemia and another 1.6% (4 patients; 5 events) had suspected hemolytic anemia during this period. The mean decrease in hemoglobin was 4.1 g/dL in confirmed cases (range 1.3–7.1 g/dL) and 2.7 g/dL in suspected cases (range 1.7–4.2 g/dL). Eight of nine patients reviewed had blood type A positive; the other was B positive. The median time from initiation of therapy to hemoglobin nadir was eight days in confirmed and three days in suspected cases. The mean dose of IVIG administered in confirmed cases was 300 g (range 180–400 g) and was 162 g (range 50–300 g) in suspected cases. Red cell transfusion was required in three of the four confirmed cases. Privigen™ was the IVIG product utilized in all four confirmed and the first three suspected cases. After removal of Privigen™ from the Mayo Clinic formulary given concerns regarding these cases, two suspected cases of IVIG-associated hemolytic anemia occurred with Gamunex-C™, including one in a patient who previously suffered confirmed hemolysis after receipt of Privigen™. Conclusions: Hemolytic anemia is a rare complication of therapy with IVIG. The rate of hemolytic anemia following IVIG administration over the evaluated period is consistent with previous reports, but it should be noted that these cases were detected without the aid of an active surveillance policy, indicating the actual rate of hemolytic anemia associated with IVIG use was likely greater than that detected by this review. Our series suggests that those with blood type A may be at increased risk for this complication; further work is ongoing to better define risk factors for development of this potentially significant complication of IVIG therapy. Disclosures: Off Label Use: IVIG use in various off-label indications.

4009 Background: Interim analysis of a global, phase 2 KEYNOTE-059 study (NCT02335411) reported manageable safety and promising antitumor activity for pembro alone or pembro + chemo in pts with G/GEJ cancer. Here we report long-term efficacy and safety data of all cohorts. Methods: Pts with recurrent or metastatic G/GEJ adenocarcinoma were enrolled in 3 cohorts. Cohort 1 pts (PD-L1–positive or –negative) received pembro alone after ≥2 prior lines of therapy. Cohort 2 pts (PD-L1–positive or –negative) received pembro + cisplatin (80 mg/m2 day 1) + 5-fluorouracil (800 mg/m2 days 1-5 Q3W) or capecitabine (in Japan only, 1000 mg/m2 twice daily) as first-line. Cohort 3 pts (PD-L1–positive, combined positive score ≥1% using the PD-L1 IHC 22C3 pharmDx assay) received pembro alone as first-line. All pts received pembro 200 mg Q3W for up to 2 years. End points included safety, ORR, DOR, and OS. Results: At data cutoff (Aug 8, 2018), median (range) follow-up was 6 (1-38), 14 (2-40), and 21 (2-36) months for cohorts 1 (n = 259), 2 (n = 25), and 3 (n = 31), respectively. In cohort 1, confirmed ORR (95% CI) was 11.6% (8-16) overall, 15.5% (10-22) in PD-L1–positive, and 6.4% (3-13) in PD-L1–negative tumors. In cohort 2, confirmed ORR was 60.0% (39-79) overall, 73.3% (45-92) in PD-L1–positive, and 37.5% (9-76) in PD-L1–negative tumors. In cohort 3, confirmed ORR was 25.8% (12-45). Median (range) DOR in months was 16.1 (2-35+), 4.6 (3-37+), and not reached (2.1-32.5+) in cohorts 1, 2, and 3, respectively. OS at 1 year/2 years was 24.6%/12.5%, 52%/32%, and 63.6%/40.1% in cohorts 1, 2, and 3, respectively. In cohorts 1, 2, and 3, grade 3-5 treatment-related adverse event (TRAE) incidence was 46 (18%), 20 (80%), and 8 (26%) respectively. TRAEs led to discontinuation in 6 (2%) and 3 (12%) pts in cohorts 1 and 2, respectively, and to death in 2 (1%) pts in cohort 1. No TRAEs led to discontinuation or death in cohort 3. Conclusions: These updated results demonstrate manageable safety, durable clinically meaningful activity of pembro in heavily pretreated pts, and promising efficacy of first-line pembro (alone or + chemo) in pts with advanced G/GEJ cancer. Clinical trial information: NCT02335411.

511 Background: Guidelines do not recommend routine FDG PET/CT (PET) as preoperative staging for pancreatic cancer, although many single center series have demonstrated that PET can lead to changes in management in a sizable minority of patients. We performed a retrospective analysis of patients undergoing PET for potentially resectable pancreatic adenocarcinoma at our institution to help define the utility of PET in this setting. Methods: We reviewed patients with pancreatic adenocarcinoma diagnosed at our center from June 2010 to May 2017 and included patients with pancreatic adenocarcinoma felt to be potentially resectable following standard staging studies [computed tomography (CT), magnetic resonance imaging (MRI) and endoscopic ultrasound (EUS)] who also underwent preoperative PET. Data collected and analyzed included: demographics, pre-PET staging, CA19-9 levels, PET results and surgical outcomes. Results: Forty eight patients with pancreatic adenocarcinoma felt to be surgically resectable underwent PET. PET changed management in 4/48 (8.3%) of these patients. In all 4 of these patients, hepatic metastatic disease was detected on PET and planned surgery was canceled; metastatic disease was confirmed by biopsy in 1 of these patients. 1/48 (2.1%) of patients had a false positive PET scan, where a focus of suspected metastatic disease on PET was biopsied and found to be benign, allowing the patient to proceed to surgery. 3/48 (6.3%) of patients had a false negative PET; 2 patients had hepatic metastatic disease and one had peritoneal disease discovered during surgery. Mean time from negative PET to surgery in these 3 patients was 31 days (range 21-45). Degree of CA19-9 elevation and primary tumor FDG avidity did not correlate with detection of metastatic disease on PET. Conclusions: PET changed management in a smaller number of patients in this cohort than in many previously reported series with a nearly equal number of patients with false negative PET results proceeding to unnecessary surgery. These results are consistent with the currently uncertain role of PET in preoperative staging for pancreatic cancer and further work must be undertaken to optimize presurgical staging in this population.

Abstract Purpose: Immunotherapeutic strategies to treat patients with renal cell carcinoma (RCC) offer new opportunities for disease management. Further improvements to immunotherapy will require additional understanding of the host response to RCC development. Experimental Design: Using a novel approach to understanding the immune status of cancer patients, we previously showed that patients with a certain immune profile had decreased overall survival. Here, we examine in more detail the phenotypic changes in peripheral blood and the potential consequences of these changes in RCC patients. Results: We found that CD14+HLA-DRlo/neg monocytes were the most predominant phenotypic change in peripheral blood of RCC patients, elevated nearly 5-fold above the average levels measured in healthy volunteers. Intratumoral and peritumoral presence of CD14 cells was an independent prognostic factor for decreased survival in a cohort of 375 RCC patients. The amount of peripheral blood CD14+HLA-DRlo/neg monocytes was found to correlate with the intensity of CD14 staining in tumors, suggesting that the measurement of these cells in blood may be a suitable surrogate for monitoring patient prognosis. The interaction of monocytes and tumor cells triggers changes in both cell types with a loss of HLA-DR expression in monocytes, increases of monocyte survival factors such as GM-CSF in tumors, and increased production of angiogenic factors, including FGF2. Conclusions: Our results suggest a model of mutually beneficial interactions between tumor cells and monocytes that adversely affect patient outcome. Clin Cancer Res; 21(18); 4224–33. ©2015 AACR.

AbstractBackgroundCardiac complications following hematopoietic stem cell transplantation (HSCT) are emerging as a significant concern given the increasing utilization of HSCT for a variety of hematologic malignancies.MethodsWe utilized an existing database to determine the frequency of cardiac dysfunction (CD), namely a decrease in left ventricular ejection fraction, following conditioning with high‐dose melphalan (HDM) and autologous HSCT for multiple myeloma (MM) and systemic amyloidosis (AL). We then performed a case–control study to examine variables associated with increased risk of CD in this population.ResultsIn MM patients undergoing HSCT, the rate of CD was 1.6% (17/1050, 95% CI: [0.9, 2.6]). None of the examined pre‐HSCT variables or HDM dose were significantly associated with development of CD in this population. In patients with AL, the rate of CD was 5.6% (24/426, 95% CI: [3.6, 8.3]). On univariate analysis, decision to administer an HDM dose <200 mg/m2 [odds ratio (OR): 4.59 (1.27–16.57) P = 0.02], pretransplant left ventricular ejection fraction <60% [OR: 17.78 (2.29–138.33) P = 0.006], and documented amyloid involvement of ≥3 organs [OR: 4.0 (1.03–15.6) P = 0.046] were associated with the development of CD in the AL population. No other examined peri‐transplant factors were associated with development of CD.ConclusionTo our knowledge, this is the first series to report a significant rate of CD following HDM conditioning and autologous HSCT in patients with AL.