A phase Ib/II study (IMMCO-1) of atezolizumab plus tivozanib in castrate-resistant prostate cancer and certain other immunologically cold tumors

TPS293

Background: Castrate resistant prostate cancer (CRPC) is an immunologically cold tumor, with a 5% response rate reported to the PD-L1 inhibitor pembrolizumab. VEGF secreted by tumors may play a key role in hindering the anti-tumor immune response, leading to the development of an abnormal vasculature that may limit immune surveillance. VEGF also inhibits dendritic cell differentiation, limiting the presentation of tumor antigens. The inhibition of VEGF may therefore potentiate the effect of PD-1/L1 directed therapy by enabling immune surveillance and antigen presentation. VEGF-TKI and checkpoint inhibitor combinations are currently approved in the treatment of advanced kidney, cervical, and endometrial cancers. This signal-seeking study aims to determine whether the combination of the VEGF-TKI tivozanib and the PD-L1 inhibitor atezolizumab may be effective in CRPC and certain other immunologically cold tumors. Methods: The trial is a single center phase Ib/II basket study in multiple immunologically cold tumors. Co-primary endpoints are safety and the overall response rate as measured via RECIST v1.1. The phase Ib 3+3 dose de-escalation portion of the study has been completed. The study is designed to test for a 25%+ response rate as compared to a null hypothesis of <7% (one-sided alpha = 0.05; 80% power). A Simon’s two-stage design will be utilized and if ≥2 responses among the first 16 evaluable patients, a further 10 evaluable patients will be accrued for a total of 26. Up to 33 subjects will be enrolled to account for a 20% dropout rate. The null hypothesis will be rejected if at least 5 responses are observed. The University of Florida's Data Integrity and Safety Committee will review significant adverse events. Atezolizumab is given at a dose of 1,680 mg every 28 days. A subsequent protocol amendment decreased the starting dose of tivozanib from 1.34 mg/day for 21 days of each 28-day cycle to 0.89 mg/day for 21 days of each 28-day cycle. Treatment is until disease progression or intolerance. Key inclusion criteria include a diagnosis of certain advanced and unresectable or metastatic immunologically cold tumors (CRPC previously treated with an androgen inhibitor or cytotoxic chemotherapy in the advanced or metastatic setting, bile duct or gallbladder cancer, certain HR-negative HER2-positive breast cancers, ovarian cancer, pancreatic adenocarcinoma, soft tissue sarcoma, or vulvar cancer), at least one prior systemic therapy in this setting, ECOG 0-1 (phase Ib) or ECOG 0-2 (phase II), age ≥ 18, adequate hematologic and end-organ function, life expectancy of at least 12 weeks, and measurable disease by RECIST v1.1. Key exclusion criteria include known mismatch repair deficiency, microsatellite instability, or high tumor mutational burden. Active enrollment continues. Clinical trial information: NCT05000294 .

A phase Ib/II study of atezolizumab plus tivozanib in pancreatic, gallbladder, and bile duct malignancies

TPS625

Background: Immune checkpoint inhibitor (ICI) therapy represents a significant advance in cancer care however it is not an effective intervention in the treatment of several immunologically cold tumors including pancreatic, gallbladder, and bile duct malignancies where ICIs have produced objective response rates of 0-6%. VEGF is thought to play a key role in modulating the anti-tumor immune response as it is secreted by tumors and leads to endothelial cell proliferation, vascular permeability, and vasodilation. This in turn leads to the development of an abnormal vasculature with excessive permeability and poor blood flow, thus limiting immune surveillance. In addition, VEGF inhibits dendritic cell differentiation, limiting the presentation of tumor antigens to CD4 and CD8 T cells. Through the inhibition of VEGF, it may be possible to potentiate the effect of ICIs. Combined use of a VEGF receptor tyrosine kinase inhibitor (TKI) and ICI is already approved for advanced kidney and endometrial cancer. Methods: This is a single institution, single arm, open-label phase Ib/II study with the co-primary endpoints of safety and efficacy of the combination of the VEGFR-TKI tivozanib and the ICI atezolizumab. Eligible patients are those with metastatic cancer progressing on at least one line of therapy of the following organs: pancreatic, gallbladder, bile duct, well-differentiated grade 2 and 3 neuroendocrine tumors, ovarian, vulvar, soft tissue sarcoma, castrate resistant prostate, and HER2 positive hormone receptor negative breast. Key exclusion criteria will include patients with known mismatch repair deficiency, microsatellite instability, or high tumor mutational burden. The phase Ib portion will assess the safety profile of the combination of tivozanib and atezolizumab with a potential dose de-escalation of tivozanib using a 3+3 study design. Three patients will be treated with tivozanib 1.34 mg per day (dose level 0) for 21 days of each 28-day cycle and atezolizumab 1680 mg on day 1 of every 28-day cycle. If one dose limiting toxicity (DLT) is found then 3 more patients will be enrolled at dose level 0. If >1 DLT occurs in the first 3 patients or >1 DLT in the first 6, another 3+3 study will begin at dose level -1 (0.89 mg of tivozanib). If 1 DLT occurs in these 3 patients at dose level -1 then another 3 will be enrolled at this dose. The study will be stopped for toxicity if there is >1 DLT in the first 3 patients at dose level -1 or if there are >1 DLT in the 6 patients at dose level -1. A subsequent protocol amendment was made to reduce tivozanib dose level 0 to 0.89 mg per day and dose level -1 to 0.89 mg every other day. The phase II portion will enroll up to 26 patients using the dose of tivozanib found to be safe in the Ib portion. Disease response assessment every 12 weeks with CT Chest, Abdomen, and Pelvis via RECIST 1.1. Treatment will continue until progression or intolerance. Enrollment is ongoing. Clinical trial information: NCT05000294 .

A phase II, open-label pilot study evaluating the safety and activity of liposomal irinotecan (Nal-IRI) in combination with 5-FU and oxaliplatin (NALIRIFOX) in preoperative treatment of pancreatic adenocarcinoma (NEO-Nal-IRI Study)

TPS778

Background: Neoadjuvant treatment for potentially curable pancreatic cancer (PDAC) is increasing in acceptability, but a standard regimen has yet to be established. Multiple studies have demonstrated feasibility and effectiveness of the FOLFIRINOX (5-fluorouracil, leucovorin, oxaliplatin and irinotecan) regimen in the perioperative setting. However, FOLFIRINOX often requires dose modifications, delays and growth factor support due to excessive toxicity which can complicate care delivery when given neoadjuvantly. Liposomal irinotecan injection (Nal-IRI) is FDA approved with a well-tolerated safety profile in relapsed, refractory metastatic PDAC. The current study aims to substitute Nal-IRI for traditional irinotecan in the standard FOLFIRINOX regimen (NALIRIFOX) and to demonstrate safe and effective neoadjuvant delivery. Methods: This phase 2, open-label, multicenter single-arm study focuses on patients (pts) with operable PDAC without metastatic disease. Other key eligibility criteria include age ≥18 years, resectability confirmed by multidisciplinary GI tumor board (resectable vs. borderline), adequate cardiac, renal, hepatic function and ECOG performance status of 0 to 1. Pts receive NEO-N-IRI regimen as per Table every 2 weeks for four months followed by disease reassessment. Pts who remain surgical candidates will undergo surgical resection within 4 to 8 weeks following last dose of therapy. The primary endpoint is to assess safety and feasibility of regimen in perioperative setting. Secondary endpoints include R0 resection rate, clinical, biochemical and radiological response rate and patient-reported quality of life during treatment as measured by the NCI validated FACT-G scale. Enrollment continues to a maximum of 28 evaluable pts to demonstrate a reduction in historical 30-day postoperative complication rate. Microbiota specimens will be collected for exploratory analysis. Clinical trial information: NCT03483038 . [Table: see text]

A phase II randomized therapeutic optimization trial for patients with refractory metastatic colorectal cancer using circulating tumor DNA (ctDNA): Rapid 1 trial

TPS280

Background: Patients with advanced colorectal cancer after progressing through first line therapy, have several FDA-approved systemic therapies that are associated with clinical benefit for a substantive minority of patients. Current clinical practice is to trial these various treatments in a step-wise fashion using CT scans every 3 months to evaluate effectiveness. This process requires 3-4 months between therapeutic interventions from which the patient may ultimately derive no clinical benefit, may have a performance status decline; limiting the number of possible interventions and increases risk for physical and financial toxicity. An alternative to the traditional CT-scan guided approach for disease assessment is a circulating tumor DNA (ctDNA) intervention. The Signatera ctDNA assay which utilizes 16 truncal mutations derived from a patient’s tumor, can be assessed every 2 weeks for a rapid determination of the effectiveness of a systemic therapy. This may allow patients to be exposed to many treatments during a short time, limiting toxicity, allowing for a quicker determination of clinical benefit and personalization of treatment. The aim of this study will be to compare the traditional scan-driven approach vs an intervention guided by ctDNA assessments, both arms using a pre-specified order of chemotherapy treatments. Methods: This is a phase 2 randomized study of patients with refractory metastatic adenocarcinoma of the colon or rectum. Participants are eligible after progression or intolerance to first line chemotherapy or recurrence within 6 months of adjuvant oxaliplatin based chemotherapy. They must have RECIST measurable metastatic disease that is not eligible for definitive management. Tissue from the primary and/or metastatic deposit is required for Signatera NGS analysis and subjects must have measurable ctDNA at sampling. Participants must be ≥18 years old without major organ dysfunction and have an ECOG performance status of 0 to 2. Subjects with Microsatellite High, deficient in DNA mismatch repair genes, or BRAF V600E mutations are excluded. Subjects will be randomized 1:1 to Arm A (ctDNA guided intervention) or Arm B (scan-guided control group). Patients in both arms will undergo systemic treatments in a standardized pre-specified order. Arm A will have ctDNA assessments every 2 weeks until an intervention shows a significant decrease, then every 4 weeks until Progressive Disease (PD) by scan or significant ctDNA increase. CT imaging will be performed every 12 weeks. Those in Arm B, will have CT imaging every 12 weeks and blood collected for post-hoc analysis every 4 weeks until PD by scan. The primary endpoint is overall survival. Secondary endpoints include progression free survival and overall response. Exploratory analysis with be performed of the microbiome. Enrollment continues to a maximum of 78 patients. Clinical trial information: NCT04786600 .

Influence of clinical and tumor characteristics on survival in patients with hepatocellular carcinoma with bone metastasis

525

Background: Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer, and the incidence of liver cancer is rapidly rising in the United States. Advanced HCC generally has a dismal prognosis with an expected survival of less than 1 year. Bone metastases from HCC are infrequent with a poorer prognosis. However, the factors influencing their survival are not well-understood. We aimed to study the effects of clinical and tumor characteristics on survival in HCC patients with bone metastasis. Methods: We selected adult patients with HCC who had bone metastasis diagnosed between 2004 and 2019 using the National Cancer Database (NCDB). We described baseline characteristics using percentages. We performed the Kaplan-Meier method to calculate the median overall survival (OS). We included demographics (age at diagnosis, gender, race, insurance status), comorbidity score, and treatment characteristics. Results: Of a total of 3301 HCC patients with bone metastasis, 87.1% were male, and 12.9% were female. Among all, 72.2% were white, 19.3% were black, and 8.5% were others. A total of 59.9% of patients were <65 years. Eighty-six-point four percent of the patients had a total Charlton-Deyo comorbidity score of <3 and 13.6% had it ≥3. Among patients with known tumor grade, 25.4% had well differentiated, 37.8% moderately differentiated, 35.4% poorly differentiated, and 1.2% undifferentiated tumors. In univariate analysis, patients with well-differentiated tumors had better OS compared to poorly differentiated tumors (5.4 mo vs. 3.3 mo, p<0.05). In treatment groups, both single and multi-agent chemotherapy as first-course therapy significantly improved OS compared to patients’ chemotherapy was not administered (6.8 mo vs. 2.1 mo; 8.9 mo vs. 2.1 mo, respectively). We found no mortality difference between age, gender and race groups. Conclusions: In this cohort analysis of NCDB data, we report better OS in treatment receipt, lower tumor grade, and lower comorbidity score in HCC patients with bone metastasis. Previous studies of HCC patients with bone metastasis have been limited by small sample sizes, and lack of nationwide oncology outcomes data. Further large-scale prospective studies are needed.

Results of a phase II trial of the PARP inhibitor, niraparib, in BAP1 and other DNA damage response pathway deficient neoplasms

3122

Background: BRCA1-Associated Protein 1 (BAP1) acts as a tumor suppressor and critical regulator of the cell cycle and DNA damage response (DDR). PARP inhibitors (PARPi) demonstrate synthetic lethality in BAP1 mutant (mBAP1) preclinical models, independent of underlying germline BRCA status. mBAP1 leads to a loss of functional protein in several solid tumors. This study aimed to explore the clinical activity of niraparib in patients (pts) with advanced tumors likely to harbor mBAP1. Methods: Eligible adult pts with measurable metastatic solid tumors having exhausted approved therapies, adequate organ function, and ECOG PS 0-1 were assigned to Cohort A (histology-specific): tumors likely to harbor mBAP1 (i.e., cholangiocarcinoma, uveal melanoma, mesothelioma, or clear cell renal cell carcinoma) with tissue available for mBAP1 confirmation; or Cohort B (histology-agnostic): tumors with other known non-BRCA confirmed DDR mutations. Known BRCA1 or 2 mutations or prior PARPi exposure were excluded. All pts received niraparib 200-300mg daily, depending on weight and/or platelet count. Radiographic response was assessed by RECIST v1.1 measured every 8 weeks while on treatment. The primary endpoint was ORR with secondary endpoints of PFS, OS, clinical benefit (CR+PR+SD), toxicity, and exploratory biomarker determinations. Cohort A employed Simon's optimal two-stage design to assess a 30% ORR increase (a = 0.05; power = 90%). Cohort B aimed to assess a 40% ORR increase for this molecularly selected/enriched patient population. Results: From 08/13/2018 to 12/21/2021, 37 pts enrolled from two different centers, with 32 evaluable for response (Cohort A n = 18; Cohort B n = 14). In Cohort A, best ORR was 1 PR (6%), 8 SD (44%; median 5.7 mo; range 2 - 9.4 mo), and 9 PD (50%). Cohort A was stopped at the first stage following the pre-specified Simon’s design. mBAP1 was confirmed in 7/9 pts (78%) with PR or SD but in only 2/9 (22%) in those with PD. In Cohort B, best ORR was 6 SD (43%; median 7.5 mo; range 3.3 - 8.6 mo) and 8 PD (57%). Mutations in those with SD included ATM, CHEK2, PTEN, RAD50, and ARID1A. Common grade 3/4 AEs observed were anemia (16%), thrombocytopenia (16%), nausea (11%), and vomiting (8%). There were no unexpected nor grade 5 toxicities. Conclusions: The use of niraparib was well tolerated in pts with advanced treatment refractory solid tumors but failed to meet pre-specified efficacy threshold of ORR. However, clinical benefit was identified in 78% of patients in cohort A who had a confirmed mBAP1 tumor. Further correlative analyses are ongoing and additional clinical development restricted to mBAP1 tumors may be justified. Clinical trial information: NCT03207347.

A phase II, open-label, pilot study evaluating the safety and activity of liposomal irinotecan (Nal-IRI) in combination with 5-FU and oxaliplatin (NALIRIFOX) in preoperative treatment of pancreatic adenocarcinoma: NEO-Nal-IRI study

TPS4196

Background: Neoadjuvant treatment for potentially curable pancreatic cancer (PDAC) is increasing in acceptability, but a standard regimen has yet to be established. Multiple studies have demonstrated feasibility and effectiveness of the FOLFIRINOX (5-fluorouracil, leucovorin, oxaliplatin and irinotecan) regimen in the perioperative setting. However, FOLFIRINOX often requires dose modifications, delays and growth factor support due to excessive toxicity which can complicate care delivery when given pre-op. Liposomal irinotecan injection (Nal-IRI) is FDA approved in combination with 5-FU/LV with a well-tolerated safety profile in relapsed, refractory metastatic PDAC. The current study aims to substitute Nal-IRI for traditional irinotecan in the standard FOLFIRINOX regimen (NALIRIFOX) and to demonstrate safe and effective neoadjuvant delivery. Methods: This phase 2, open-label, multicenter single-arm study focuses on patients (pts) with operable PDAC without metastatic disease. Other key eligibility criteria include age ≥18 years, resectability confirmed by multiD GI tumor board (resectable vs. borderline), adequate cardiac, renal, hepatic function and ECOG performance status of 0 to 1. Pts receive NALIRIFOX regimen as per the table every 2 weeks for four months followed by disease reassessment. Pts who remain surgical candidates will undergo surgical resection within 4 to 8 weeks following last dose of therapy. The primary endpoint is to assess safety and feasibility of regimen in pre-op setting. Secondary endpoints include R0 resection rate, clinical, biochemical and radiological response rate and patient-reported quality of life during treatment as measured by the NCI validated FACT-G scale. Enrollment continues to a maximum of 28 evaluable pts to demonstrate a reduction in historical 30-day post-op complication rate. Clinical trial information: NCT03483038. [Table: see text]

Impact of metastasectomy for extrahepatic disease in stage IV colon cancer: A retrospective cohort NCDB analysis

e15548

Background: Stage IV colon cancer has a dismal prognosis with a 5-yr survival rate of only 14%. The role of liver metastasectomy is well established. However, it is not clear if there is a survival benefit for metastasectomy in cases of extrahepatic disease. Methods: We conducted a retrospective cohort study, US hospital-based. The NCDB version 2017 was accessed. Exposure variable: metastasectomy for extrahepatic distant disease, analyzed as binary. Outcome: overall survival. Kaplan Meier curves and log rank test were used for survival analysis. The primary endpoint was to estimate the association between metastasectomy and survival. Covariates demographic factors, age, sex, race, insurance status, site of metastatic disease – lung, liver, bone, brain. A Cox proportional hazards model was used to evaluate the impact of metastasectomy of extrahepatic disease in patients with stage IV colon cancer. 95% confidence intervals and p values of less than 0.05 were used for statistical significance. Missing data as handled with complete and available case analysis. Statistical analysis was done using Stata 16. Results: A total of 68,334 adult patients with stage IV colon cancer diagnosed in from 2010 -2016 were included in the final analysis. Median age was 64 years (SD 14.1). Prevalence of metastases in this stage IV population was as follows: Liver 72.12%, Lung 20.12%, Bone 5.06%, Brain: 1.6%. A total of 10,595 patients (15.5%) had metastasectomy. Median overall survival were as follows: Lung, 21.59 months with surgery vs 8.41 months no surgery. Bone, 7.03 mo vs 4.53 mo respectively; Brain 9.3 mo vs 3.15 mo; Liver 33.28 mo vs 10.97 mo. In univariate analysis, patients with lung metastases had an improved survival if they had metastasectomy (vs no surgery, HR 0.58, p < 0.0001, 95% CI 0.52 – 0.60). The same benefit was observed in patients with bone (HR 0.80, p 0.001, 95% CI 0.69 – 0.91) and brain metastases (HR 0.56, p < 0.001, 95%CI 0.49 – 0.65). In multivariable analysis the survival advantage was still significant. In a sensitivity analysis, excluding patients with liver metastases, the improvement in survival persisted. Conclusions: Metastasectomy for extrahepatic disease improved overall survival in patients with colon cancer in this retrospective analysis. Further research is needed before routinely incorporating it in clinical practice.[Table: see text]

Prognostic value of CA 19-9 and carcinoembrionic antigen (CEA) in duodenal adenocarcinoma (DA): An institutional retrospective cohort study

e16306

Background: Duodenal adenocarcinoma (DA) is a rare malignancy with poor outcomes. Tumor markers are used to assess disease response and to monitor for recurrence. Specifically, CA-19-9 and CEA have been validated for use in pancreatic cancer and colorectal cancer, respectively. However, these tumor markers have never been validated in patients with DA. We aim to assess the association of these biomarkers with clinical outcomes in patients with DA. Methods: This is a retrospective cohort study. After obtaining IRB approval (IRB202102705), we accessed the University of Florida medical records of patient treated for DA from January 1, 2006, until December 31, 2021. CA 19-9 and CEA were collected as continuous variables and were analyzed as binary variables: normal vs. high, using the maximum normal value as a cut-off (normal CA 19-9 < = 35 U/ml; CEA < = 3 ng/ml). Analysis was conducted using Kaplan Meyer curves, log-rank test and Cox proportional hazards model. Results: A total of 68 patients were included in the final analysis. Median age was 67 years and median follow-up was 22.2 months. CA 19-9 and CEA were elevated in 36.8% and 48.5% of patients, respectively. Patients with an elevated CA 19-9 had a median overall survival (OS) of 8.5 months vs. 27.4 months in patients with normal levels (HR 1.67; 95%CI 0.94–2.99; p = 0.081). Patients with an elevated CEA had a median OS of 13.4 months vs. 16.8 months in patients with normal level normal levels (HR 1.43; 95%CI 0.81–2.52; p value = 0.221). In a sensitivity analysis, a concomitant elevation of both tumoral markers was significantly associated with worsened OS (HR 1.9; 95%CI 1.05–3.06; p = 0.035). Conclusions: In patients with duodenal adenocarcinoma, elevation of both CA 19-9 and CEA was associated with a statistically significant worse overall survival. CA 19-9 level had a higher prognostic impact on OS than CEA levels. To our knowledge, this is the first study to evaluate the role of CA 19-9 and CEA in patients with DA. Further research is required for validation.[Table: see text]

A phase Ib/II study (IMMCO-1) of atezolizumab plus tivozanib in castrate-resistant prostate cancer and certain other immunologically cold tumors

TPS208

Background: Castrate resistant prostate cancer (CRPC) is an immunologically cold tumor, with a 5% response rate reported to the PD-L1 inhibitor pembrolizumab. VEGF secreted by tumors may play a key role in hindering the anti-tumor immune response, leading to the development of an abnormal vasculature that may limit immune surveillance. VEGF also inhibits dendritic cell differentiation, limiting the presentation of tumor antigens. The inhibition of VEGF may therefore potentiate the effect of PD-1/L1 directed therapy by enabling immune surveillance and antigen presentation. VEGF-TKI and checkpoint inhibitor combinations are currently approved in the treatment of advanced kidney, cervical, and endometrial cancers. This signal-seeking study aims to determine whether the combination of the VEGF-TKI tivozanib and the PD-L1 inhibitor atezolizumab may be effective in CRPC and certain other immunologically cold tumors. Methods: The trial is a single center phase Ib/II basket study in multiple immunologically cold tumors. Co-primary endpoints are safety and the overall response rate as measured via RECIST v1.1. The tivozanib dose will be determined via a 3+3 dose de-escalation phase Ib portion. Patients treated in the phase Ib portion will be included as study subjects for phase the II portion. The study is designed to test for a 25%+ response rate as compared to a null hypothesis of <7% (one-sided alpha = 0.05; 80% power). A Simon’s two-stage design will be utilized and if ≥2 responses among the first 16 evaluable patients, a further 10 evaluable patients will be accrued for a total of 26. Up to 33 subjects will be enrolled to account for a 20% dropout rate. The null hypothesis will be rejected if at least 5 responses are observed. The University of Florida's Data Integrity and Safety Committee will review significant adverse events. Tivozanib will be given at a dose of 1.34 mg/day for 21 days of each 28-day cycle (potential dose reduction is to 0.89 mg/day). Atezolizumab is given at a dose of 1,680 mg every 28 days. Treatment is until disease progression or intolerance. Key inclusion criteria include a diagnosis of certain advanced and unresectable or metastatic immunologically cold tumors (CRPC previously treated with an androgen inhibitor or cytotoxic chemotherapy in the advanced or metastatic setting, bile duct or gallbladder cancer, certain HR-negative HER2-positive breast cancers, ovarian cancer, pancreatic adenocarcinoma, soft tissue sarcoma, or vulvar cancer), at least one prior systemic therapy in this setting, ECOG 0-1 (phase Ib) or ECOG 0-2 (phase II), age ≥ 18, adequate hematologic and end-organ function, life expectancy of at least 12 weeks, and measurable disease by RECIST v1.1. Key exclusion criteria include known mismatch repair deficiency, microsatellite instability, or high tumor mutational burden. Active enrollment continues. Clinical trial information: NCT05000294.

A phase II randomized therapeutic optimization trial for subjects with refractory metastatic colorectal cancer using circulating tumor DNA (ctDNA): Rapid 1 trial

TPS223

Background: Patients with advanced colorectal cancer after progressing through first line therapy, have several FDA-approved systemic therapies that are associated with clinical benefit for a substantive minority of patients. Current clinical practice is to trial these various treatments in a step-wise fashion using CT scans every 3 months to evaluate effectiveness. This process requires 3-4 months between therapeutic interventions from which the patient may ultimately derive no clinical benefit, may have a performance status decline; limiting the number of possible interventions and increases risk for physical and financial toxicity. An alternative to the traditional CT-scan guided approach for disease assessment is a circulating tumor DNA (ctDNA) intervention. The Signatera ctDNA assay which utilizes 16 truncal mutations derived from a patient’s tumor, can be assessed every 2 weeks for a rapid determination of the effectiveness of a systemic therapy. This may allow patients to be exposed to many treatments during a short time, limiting toxicity, allowing for a quicker determination of clinical benefit and personalization of treatment. The aim of this study will be to compare the traditional scan-driven approach vs an intervention guided by ctDNA assessments, both arms using a pre-specified order of chemotherapy treatments. Methods: This is a phase 2 randomized study of patients with refractory metastatic adenocarcinoma of the colon or rectum. Participants are eligible after progression or intolerance to first line chemotherapy or recurrence within 6 months of adjuvant oxaliplatin based chemotherapy. They must have RECIST measurable metastatic disease that is not eligible for definitive management. Tissue from the primary and/or metastatic deposit is required for Signatera NGS analysis and subjects must have measurable ctDNA at sampling. Participants must be ≥18 years old without major organ dysfunction and have an ECOG performance status of 0 to 2. Subjects with Microsatellite High, deficient in DNA mismatch repair genes, or BRAF V600E mutations are excluded. Subjects will be randomized 1:1 to Arm A (ctDNA guided intervention) or Arm B (scan-guided control group). Patients in both arms will undergo systemic treatments in a standardized pre-specified order. Arm A will have ctDNA assessments every 2 weeks until an intervention shows a significant decrease, then every 4 weeks until Progressive Disease (PD) by scan or significant ctDNA increase. CT imaging will be performed every 12 weeks. Those in Arm B, will have CT imaging every 12 weeks and blood collected for post-hoc analysis every 4 weeks until PD by scan. The primary endpoint is overall survival. Secondary endpoints include progression free survival and overall response. Exploratory analysis with be performed of the microbiome. Enrollment continues to a maximum of 78 patients. Clinical trial information: NCT04786600.

Randomized phase II trial of postoperative adjuvant capecitabine and temozolomide versus observation in high-risk pancreatic neuroendocrine tumors: SWOG S2104

TPS515

Background: Pancreatic neuroendocrine tumors (pNETS) account for about 1-2% of all the pancreas tumors for which resection is the only curative intent modality, however, despite surgery, many patients will experience recurrence. Tumor size, positive lymph nodes and higher grade are prognostic factors for recurrence. The U.S. Neuroendocrine Tumor Study Group published a predictive score (Zaidi score, see table) demonstrating that tumors with a score≥6 had a 33% likelihood of recurrence by 24 months. Furthermore, it is known that that patients with metastatic disease who undergo curative intent resection also have a great risk of recurrence. Despite these known risk factors for recurrence, to date, no prospective study exploring the role of adjuvant cytotoxic chemotherapy has ever been performed in this population. Supported by the E2211 results using capecitabine and temozolomide (CAPTEM) in the metastatic setting, S2104 investigates the role of CAPTEM in the adjuvant setting. Methods: S2104 is a randomized phase II trial designed to evaluate recurrence-free survival (RFS) in participants with resected pNETs randomized on a 2:1 fashion to CAPTEM or observation. Patient are eligible if they had resected well-differentiated grade 2 or 3 (ki-67 up to 55%) pNETS with a Zaidi score of ≥ 3. Patients may have received resection/ablation of liver oligo-metastatic disease (up to 5 liver metastases) at the time of well-differentiated pNET resection. Key eligibility criteria: age ≥18 years, Zubrod performance status of 0-2, adequate organ and marrow function. Stratification factors include 1) Disease status prior to resection (metastatic vs. non-metastatic disease) and 2) Zaidi score (≥ 6 vs. < 6). Participants will be followed with imaging for 5 years. Funding: NIH/NCI/NCTN grants U10CA180888, U10CA180819, U10CA180821, U10CA180868. Clinical trial information: NCT05040360. [Table: see text]

A phase II, open-label pilot study evaluating the safety and activity of liposomal irinotecan (Nal-IRI) in combination with 5-FU and oxaliplatin (NALIRIFOX) in preoperative treatment of pancreatic adenocarcinoma: NEO-Nal-IRI study

TPS619

Background: Neoadjuvant treatment for potentially curable pancreatic cancer (PDAC) is increasing in acceptability, but a standard regimen has yet to be established. Multiple studies have demonstrated feasibility and effectiveness of the FOLFIRINOX (5-fluorouracil, leucovorin, oxaliplatin and irinotecan) regimen in the perioperative setting. However, FOLFIRINOX often requires dose modifications, delays and growth factor support due to excessive toxicity which can complicate care delivery when given pre-op. Liposomal irinotecan injection (Nal-IRI) in combination with 5FU/LV is FDA approved with a well-tolerated safety profile in relapsed, refractory metastatic PDAC. The current study aims to substitute Nal-IRI for traditional irinotecan in the standard FOLFIRINOX regimen (NALIRIFOX) and to demonstrate safe and effective neoadjuvant delivery. Methods: This phase 2, open-label, multicenter single-arm study focuses on patients (pts) with operable PDAC without metastatic disease. Other key eligibility criteria include age ≥18 years, resectability confirmed by multiD GI tumor board (resectable vs. borderline), adequate cardiac, renal, hepatic function and ECOG performance status of 0 to 1. Pts receive NALIRIFOX regimen as per the table every 2 weeks for four months followed by disease reassessment. Pts who remain surgical candidates will undergo surgical resection within 4 to 8 weeks following last dose of therapy. The primary endpoint is to assess safety and feasibility of regimen in pre-op setting. Secondary endpoints include R0 resection rate, clinical, biochemical and radiological response rate and patient-reported quality of life during treatment as measured by the NCI validated FACT-G scale. Exploratory ctDNA and stool microbiome analyses are also planned. Enrollment continues to a maximum of 28 evaluable pts to demonstrate a reduction in historical 30-day post-op complication rate. Clinical trial information: NCT03483038. [Table: see text]

A phase Ib/II study (IMMCO-1) of atezolizumab plus tivozanib in immunologically cold pancreatic, gallbladder, and biliary cancers

TPS491

Background: Checkpoint inhibition (CPI) represents a significant advance in cancer care however it is not effective in the treatment of several immunologically cold tumors including pancreatic, gallbladder, and biliary cancers where checkpoint inhibitors have produced objective response rates of <5%. VEGF is thought to play a key role in modulating the anti-tumor immune response. Secreted by tumors, it leads to endothelial cell proliferation, vascular permeability, and vasodilation that together leads to the development of an abnormal vasculature with excessive permeability and poor blood flow, thus limiting immune surveillance. In addition, VEGF inhibits dendritic cell differentiation, limiting the presentation of tumor antigens to CD4 and CD8 T cells. Through the inhibition of VEGF, it may be possible to potentiate the effect of immune checkpoint blockade. Combined use of a VEGF tyrosine kinase inhibitor (TKI) and checkpoint inhibitor is already standard of care in advanced kidney, cervical and endometrial cancers. There has been suggestion that such a combination may have clinical activity in some microsatellite stable (MSS) GI malignancies. This signal seeking study aims to build upon those observations by incorporating a pan-VEGF axis inhibitor (tivozanib) with CPI. Methods: This is an open-label non-randomized phase Ib/II signal seeking basket study in multiple immunologically cold tumors. The co-primary endpoints are safety and efficacy of the combination of the VEGF-TKI tivozanib and CPI atezolizumab. Key eligibility criteria includes patients with MSS pancreatic, biliary (cholangiocarcinoma and gallbladder), well-differentiated grade 2 and 3 neuroendocrine tumors, ovarian and vulvar cancer, soft tissue sarcoma, castrate resistant prostate cancer, and HER2 positive hormone receptor negative breast cancer, that is metastatic and progressed on at least one line of therapy. Key exclusion criteria will include patients with known mismatch repair deficiency, microsatellite instability, or high tumor mutational burden. The phase Ib portion will assess the safety profile of the combination of tivozanib and atezolizumab with a potential dose de-escalation of tivozanib using a 3+3 study design. Starting doses include tivozanib 1.34 mg per day for 21 days of each 28-day cycle and atezolizumab 1680 mg on day 1 of every 28-day cycle. The phase II portion will enroll up to 26 additional patients using the dose of tivozanib found to be safe in the Ib portion. Disease response assessments are every 12 weeks with CT Chest, Abdomen, and Pelvis via RECIST 1.1. Treatment will continue until progression or intolerance. This signal seeking study is looking to confirm the best objective response rate for evaluable patients increasing from <7% (null hypothesis) to 25% (one-sided alpha = 0.05; 80% power). Active enrollment continues. Clinical trial information: NCT05000294.

Trends in Use of Granulocyte Colony-Stimulating Factor Following Introduction of Biosimilars Among Adults With Cancer and Commercial or Medicare Insurance From 2014 to 2019

IMPORTANCE

The introduction of biosimilars and novel delivery devices between 2014 and 2019 may have changed the utilization of granulocyte colony-stimulating factors (G-CSF).

OBJECTIVE

To assess utilization trends of G-CSFs for primary prophylaxis of febrile neutropenia (FN) among patients with cancer receiving myelosuppressive chemotherapy with commercial or Medicare insurance.

DESIGN, SETTING, AND PARTICIPANTS

This cross-sectional study assessed G-CSF utilization trends overall and stratified by regimen febrile neutropenia risk level. Associations between patient characteristics and G-CSF use were evaluated. Patients with cancer, including breast, lung, colorectal, esophageal and gastric, pancreatic, prostate, ovarian, and non-Hodgkin lymphomas, initiating myelosuppressive chemotherapy courses were included from the 2014 to 2019 commercial insurance and 2014 to 2018 Medicare fee-for-service claims databases. Data were analyzed from March to June 2021.

EXPOSURES

Year of chemotherapy initiation.

MAIN OUTCOMES AND MEASURES

The main outcomes were use and trends of G-CSFs for primary prophylaxis, from completion to 3 days after in the first chemotherapy cycle.

RESULTS

In total, 86 731 chemotherapy courses (mean [SD] age, 57.7 [11.5] years; 57 838 [66.7%] women and 28 893 [33.3%] men) were identified from 82 410 patients in the commercial insurance database and 32 398 chemotherapy courses (mean [SD] age, 71.8 [8.3] years; 18 468 [57.0%] women and 13 930 [43.0%] men) were identified from 30 279 patients in the Medicare database. Among the commercially insured population, 39 639 patients (45.7%) received G-CSFs, and 12 562 patients (38.8%) received G-CSFs among Medicare insured patients. Overall G-CSF use increased significantly throughout the study period in both populations, from 45.1% (95% CI, 44.4%-45.7%) of patients in 2014 to 47.5% (95% CI, 46.5%-48.5%) of patients in 2019 (P = .001) in the commercially insured population and from 36.0% (95% CI, 34.2%-38.0%) of patients in 2014 to 39.1% (95% CI, 38.1%-40.1%) of patients in 2018 (P < .001) in the Medicare population. The greatest increases in G-CSF use were observed among patients with high FN risk, from 75.0% (95% CI, 74.1%-76.0%) of patients to 83.2% (95% CI, 82.0%-84.2%) of patients (P < .001) among the commercially insured population and 75.3% (95% CI, 71.8%-78.6%) of patients to 86.2% (95% CI, 84.7%-87.6%) of patients (P < .001) among the Medicare population. Use of G-CSFs decreased in the commercially insured population among patients with intermediate FN risk (from 27.5% [95% CI, 26.4%-28.5%] of patients to 20.4% [95% CI, 19.1%-21.7%] of patients; P < .001) or low FN risk (from 19.3% [95% CI, 18.3%-20.4%] of patients to 16.3% [95% CI, 14.7%-18.0%] of patients; P < .001) and remained stable in the Medicare population (intermediate risk: from 26.4% [95% CI, 23.8%-29.2%] of patients to 28.4% [95% CI, 27.0%-29.8%] of patients; P = .35; low risk: from 19.6% [95% CI, 17.0%-22.4%] of patients to 20.9% [95% CI, 19.6%-22.3%] of patients; P = .58). Factors associated with increased odds of G-CSF use included older age (commercial insurance: adjusted odds ratio [aOR], 1.50 [95% CI, 1.41-1.59]; Medicare: aOR, 1.36 [95% CI, 1.08-1.71]), receiving a regimen with high FN risk (commercial insurance: aOR, 16.01 [95% CI, 15.17-16.90]; Medicare: aOR, 17.17 [95% CI, 15.76-18.71]), and history of neutropenia (commercial insurance: 3.90 (3.67-4.15); Medicare: 3.82 (3.50-4.18).

CONCLUSIONS AND RELEVANCE

This cross-sectional study found that utilization of G-CSFs increased among patients with cancer with high FN risk in both a commercially and Medicare-insured population, but 14% to 17% of patients still did not receive preventive treatment.

A phase II, open-label pilot study evaluating the safety and activity of liposomal irinotecan (Nal-IRI) in combination with 5-FU and oxaliplatin (NALIRIFOX) in preoperative treatment of pancreatic adenocarcinoma (NEO-Nal-IRI study) (NCT03483038)

TPS4170

Background: Neoadjuvant treatment for potentially curable pancreatic cancer (PDAC) is increasing in acceptability, but a standard regimen has yet to be established. Multiple studies have demonstrated feasibility and effectiveness of the FOLFIRINOX (5-fluorouracil, leucovorin, oxaliplatin and irinotecan) regimen in the perioperative setting. However, FOLFIRINOX often requires dose modifications, delays and growth factor support due to excessive toxicity which can complicate care delivery when given neoadjuvantly. Liposomal irinotecan injection (Nal-IRI) is FDA approved with a well-tolerated safety profile in relapsed, refractory metastatic PDAC. The current study aims to substitute Nal-IRI for traditional irinotecan in the standard FOLFIRINOX regimen (NALIRIFOX) and to demonstrate safe and effective neoadjuvant delivery. Methods: This phase 2, open-label, multicenter single-arm study focuses on patients (pts) with operable PDAC without metastatic disease. Other key eligibility criteria include age ≥18 years, resectability confirmed by multiD GI tumor board (resectable vs. borderline), adequate cardiac, renal, hepatic function and ECOG performance status of 0 to 1. Pts receive NALIRIFOX regimen as per the table every 2 weeks for four months followed by disease reassessment. Pts who remain surgical candidates will undergo surgical resection within 4 to 8 weeks following last dose of therapy. The primary endpoint is to assess safety and feasibility of regimen in perioperative setting. Secondary endpoints include R0 resection rate, clinical, biochemical and radiological response rate and patient-reported quality of life during treatment as measured by the NCI validated FACT-G scale. Enrollment continues to a maximum of 28 evaluable pts to demonstrate a reduction in historical 30 day postoperative complication rate. NCT03483038. NALIRIFOX regimen components given intravenously (IV) every 14 days. Clinical trial information: NCT03483038. [Table: see text]

Phase II trial of the PARP inhibitor, niraparib, in BRCA1-Associated Protein 1 (BAP1) and other DNA damage response (DDR) pathway deficient neoplasms including cholangiocarcinoma

TPS354

Background: BRCA1-Associated Protein 1 (BAP1) is a critical regulator of the cell cycle, cellular differentiation, cell death, and DNA damage response. It also acts as a tumor suppressor. Preclinical models demonstrate significant synthetic lethality in BAP1 mutant cell lines and patient xenografts when treated with PARP inhibitors, independent of underlying BRCA status, suggesting this mutation confers a BRCA-like phenotype. BAP1 is mutated, leading to a loss of functional protein, in up to 30% of cholangiocarcinomas as well as several other solid tumors. Methods: This phase 2, open-label, single arm multicenter study aims to exploit the concept of synthetic lethality with the use of the PARP inhibitor niraparib in pts with metastatic relapsed or refractory solid tumors. Eligible pts with measurable metastatic and incurable solid tumors are assigned to one of two cohorts: Cohort A (histology-specific): tumors harboring suspected BAP1 mutations including cholangiocarcinoma, uveal melanoma, mesothelioma or clear cell renal cell carcinoma with tissue available for BAP1 mutational assessment via NGS or Cohort B (histology-agnostic): tumors with known DNA damage response (DDR) mutations (Table) confirmed by CLIA-approved NGS. Other key eligibility criteria include age ≥18 years, adequate cardiac, renal, hepatic function and Eastern Cooperative Oncology Group performance status of 0 to 1. Pts with known BRCA1 or BRCA2 mutations or prior PARPi exposure are excluded. Pts receive niraparib 200-300mg daily (depending on weight and/or platelet count) continuously. Primary endpoint is objective response rate with secondary endpoints of PFS, OS, toxicity and exploratory biomarker determinations. Radiographic response by RECIST criteria is measured every 8 weeks while on treatment. Cohort A has fully enrolled. Cohort B enrollment continues to a maximum of 47 total evaluable subjects with expansion cohorts allowable for histologic or molecular subtypes meeting pre-specified responses. NCT03207347 Clinical trial information: NCT03207347. [Table: see text]

A phase II, open-label pilot study evaluating the safety and activity of liposomal irinotecan (Nal-IRI) in combination with 5-FU and oxaliplatin (NALIRIFOX) in preoperative treatment of pancreatic adenocarcinoma (NEO-Nal-IRI Study)

TPS446

Background: Neoadjuvant treatment for potentially curable pancreatic cancer (PDAC) is increasing in acceptability, but a standard regimen has yet to be established. Multiple studies have demonstrated feasibility and effectiveness of the FOLFIRINOX (5-fluorouracil, leucovorin, oxaliplatin and irinotecan) regimen in the perioperative setting. However, FOLFIRINOX often requires dose modifications, delays and growth factor support due to excessive toxicity which can complicate care delivery when given neoadjuvantly. Liposomal irinotecan injection (Nal-IRI) is FDA approved with a well-tolerated safety profile in relapsed, refractory metastatic PDAC. The current study aims to substitute Nal-IRI for traditional irinotecan in the standard FOLFIRINOX regimen (NALIRIFOX) and to demonstrate safe and effective neoadjuvant delivery. Methods: This phase II, open-label, multicenter single-arm study focuses on patients (pts) with operable PDAC without metastatic disease. Other key eligibility criteria include age ≥18 years, resectability confirmed by multiD GI tumor board (resectable vs. borderline), adequate cardiac, renal, hepatic function and ECOG performance status of 0 to 1. Pts receive NALIRIFOX regimen as per the table below every 2 weeks for four months followed by disease reassessment. Pts who remain surgical candidates will undergo surgical resection within 4 to 8 weeks following last dose of therapy. The primary endpoint is to assess safety and feasibility of regimen in perioperative setting. Secondary endpoints include R0 resection rate, clinical, biochemical and radiological response rate and patient-reported quality of life during treatment as measured by the NCI validated FACT-G scale. Enrollment continues to a maximum of 28 evaluable pts to demonstrate a reduction in historical 30 day postoperative complication rate. Clinical trial information: NCT03483038. [Table: see text]

Outcomes of autologous blood collection prior to bone marrow harvest for matched unrelated donors

e18535

Background: Autologous (auto) blood collection prior to bone marrow (BM) harvest is a common procedure however there is little data to say whether this is beneficial. Methods: This is a retrospective study evaluating the efficacy of pre-op auto blood collection in healthy unrelated donors who underwent BM harvesting at our institution between 9/2009 and 8/2017. Unrelated donors aged 18 or older who underwent their first BM harvest were included. Comparisons were made using two sample t-tests. Results: Among the 73 BM donors, 54 (74%) underwent auto blood collection resulting in 78 units collected. The cohorts with and without auto blood collected were similar in age and gender (mean age 31.0 vs 30.2; female 59% vs 41%). Those with auto blood collected donated larger volumes of marrow (mean 1395 mL vs 799 mL, p = 0.0002). Baseline hemoglobin (Hb) was similar between the cohorts (mean 14.1 g/dL vs 14.0 g/dL, p = 0.87). However, those with auto blood collected had lower pre-harvest (mean 13.1 g/dL vs 13.8 g/dL, p = 0.0430) and post-harvest Hb (mean 10.0 g/dL vs 11.3 g/dL, p = 0.0120). Of the 78 auto blood units collected, 45 units (58%) were used with 92% of women and 56% of men receiving their auto blood post-op. 33 (42%) auto blood units were discarded. Donors who were given back their auto blood were more likely to be female and have lower pre- and post-harvest Hb. Reasons for auto blood transfusion were blood availability (31%), donor post-op symptoms (i.e. hypotension, dizziness, syncope, and bleeding; 23%), intra-operative transient hypotension (26%), post-op anemia (11%) and patient request (9%). None of the patients who were transfused auto blood, or donors without auto blood collection, required allogeneic blood. Conclusions: Based on our results, collection of auto blood prior to bone marrow harvest leads to lower pre-op Hb and increases the likelihood of post-op blood transfusion. In addition, availability of auto blood can lead to over transfusion which may be detrimental to donor health. Although there may be a subset of donors who might benefit, routine auto blood collection prior to bone marrow harvesting is unnecessary and can be potentially hazardous to donors.

Improving the assessment and documentation of constipation at UF Health Cancer Center through mandatory prompts

263

Background: Opioids are commonly prescribed to patients for management of pain and the most common and persistent side effect from opioid use is constipation. Opioid-induced constipation (OIC) affects 52% of patients with advanced cancer and 87% of terminally ill patients. In the Spring 2017 Quality Oncology Practice Initiative (QOPI) chart abstraction round, UF Health Cancer Center’s Medical Oncology clinic notes documented the assessment of constipation in only 45.45% of charts. This is below the academic hospital aggregate documentation rate of 55.61%. Methods: Our objective was to improve the rate of documentation of assessment of constipation in the Thoracic Oncology Clinic at UF Health Cancer Center by 33% to a goal rate of at least 60% by 3 months. We used a Plan-Do-Study-Act model in order to design our project. We worked with the EPIC developers to include a mandatory prompt at the end of the assessment/plan section of the clinic template notes for the Thoracic Oncology practice. Our prompt was “Constipation was addressed and @HE@ {DOES/DOES NOT} have symptoms” where the author chooses from a drop down list to select whether the patient “does” or "does not” have symptoms. We planned to assess a total of 60 random charts in the 3 month post intervention period. Results: At the end of our study, a total of 48 out of 60 charts (80%) documented constipation thus surpassing our goal of 60%. Of the 12 charts assessed that did not address constipation, 11 did not use our revised templates and were notes that were “copied forwarded” from previous encounters. Conclusions: Given that our intervention was a success, we plan to expand these revised templates to the other medical oncology subspecialties in order to better document assessment of constipation for all cancer patients. While our mandatory prompt was shown to lead to increased documentation of constipation, further studies to show whether or not this leads to decreased complaints of abdominal pain, decreased incidences of bowel obstructions, and/or decreased hospital admissions, would be interesting to pursue.