Search

PURPOSE Patients with metastatic urothelial carcinoma have poor prognosis after failure of standard first-line chemotherapy. Immune check point programmed death 1-programmed death ligand 1 antibodies have low response rates and thus there exists a major unmet need. MATERIALS AND METHODS In this phase II trial, patients with metastatic urothelial carcinoma that recurred or progressed after platinum-based chemotherapy received soluble EphB4-human serum albumin (sEphB4-HSA) in combination with pembrolizumab. The primary end points were tolerability and overall survival (OS). The secondary end points were progression-free survival (PFS), objective response rate (ORR), duration of response, and toxicity. The expression of sEphB4-HSA target EphrinB2 was correlated with outcomes. RESULTS Seventy patients were enrolled. The median follow up was 22.9 months (range, 1.3-54.7). The regimen had acceptable toxicity. In the intent-to-treat analysis (N = 70), the median OS was 14.6 months (95% CI, 9.2 to 21.5). Twenty-six (37%) patients had an objective response (95% CI, 26 to 48). The median PFS was 4.1 (95% CI, 1.5 to 5.7) months. Forty-six (66%) patients expressed EphrinB2, and among them, the median OS was 21.5 months (95% CI, 12.4 to not reached), the ORR was 52% (95% CI, 37 to 67), including a complete response rate of 24% (11 of 46; 95% CI, 12 to 36). The median PFS was 5.7 (95% CI, 2.7 to 27.9) months. Response was maintained at 6, 12, and 24 months in 88%, 74%, and 69% of the patients, respectively. CONCLUSION The combination of sEphB4-HSA and pembrolizumab appears synergistic with improved OS and ORR compared with historical data for programmed death 1/programmed death ligand 1 monotherapy.

PURPOSE Randomized trials established the superiority of ibrutinib-based therapy over chemoimmunotherapy in chronic lymphocytic leukemia. Durability of progression-free survival (PFS) with ibrutinib can vary by patient subgroup. Clinical tools for prognostication and risk-stratification are needed. PATIENTS AND METHODS Patients treated with ibrutinib in phase II and III trials provided the discovery data set and were subdivided into discovery and internal validation cohorts. An external validation cohort included 84 patients enrolled in our investigator-initiated phase II trial. Univariable analysis of 18 pretreatment parameters was performed using PFS and overall survival (OS) end-points. Multivariable analysis and machine-learning algorithms identified four factors for a prognostic model that was validated in internal and external cohorts. RESULTS Factors independently associated with inferior PFS and OS were as follows: TP53 aberration, prior treatment, β-2 microglobulin ≥ 5 mg/L, and lactate dehydrogenase > 250 U/L. Each of these four factors contributed one point to a prognostic model that stratified patients into three risk groups: three to four points, high risk; two points, intermediate risk; zero to one point, low risk. The 3-year PFS rates for all 804 patients combined were 47%, 74%, and 87% for the high-, the intermediate-, and the low-risk group, respectively ( P < .0001). The 3-year OS rates were 63%, 83%, and 93%, respectively ( P < .0001). The model remained significant when applied to treatment-naïve and relapsed/refractory cohorts individually. For 84 patients in the external cohort, BTK and PLCG2 mutations were tested cross-sectionally and at progression. The cumulative incidences of mutations were strongly correlated with the model. In the external cohort, Richter’s transformation occurred in 17% of the high-risk group, and in no patient in the low-risk group. CONCLUSION Patients at increased risk of ibrutinib failure can be identified at treatment initiation and considered for clinical trials.

267 Background: Clinical information management is a burdensome process for oncology providers owing to the complexity of modern cancer data. The results that a clinician needs to review are often spread across several areas within the EHR, making it cumbersome to sense a broad overview of patient status. The purpose of this study is to describe a framework for oncologist-driven development of a data visualization tool to trend cancer biomarkers and assess the feasibility of this tool to query and display EHR data relevant for the treatment of prostate cancer. Methods: A clinical sponsor is selected to identify data elements necessary to make treatment decisions for patients receiving therapy for prostate cancer and to provide rapid feedback for the clinical tool interface. The commercial EHR database is queried to determine identifying codes for relevant laboratory tests, medications, and procedures. Data elements are assembled using the EHR platform for clinical synopsis and the clinical tool is made available to five genitourinary medical oncologists for initial pilot. Results: Oral prostate cancer medications were queried based on medication therapeutic and pharmaceutical class. In addition, androgen deprivation therapy (ADT) injections were separately identified based on route of administration. Prostate specific antigen (PSA) and testosterone result values were queried using laboratory base and/or common name codes. However, many duplicate entities are found varying by hospital/laboratory site and test assay. The assembled clinical data visualization tool can overlay temporal trends in PSA and testosterone over medication start/stop dates to convey treatment response and signs of early medication resistance. The clinician also has the option to overlay vital signs or other laboratory information to visualize treatment related adverse events (ex. weight gain related to ADT, anemia related to PARP inhibitor therapy, etc.) Lastly, the tool can also highlight the dates of the patient’s last imaging tests to allow clinicians to determine if the patient is due for any follow up imaging. Conclusions: The EHR can support novel data visualization tools for cancer biomarkers that can reasonably support clinical workflows. Development requires an intimate knowledge of EHR data but may still be limited by duplicate or erroneous codes for laboratory results. This issue may be addressed by using standard nomenclature for laboratory results such as LOINC codes but is not currently supported by the commercial EHR. Future work in this area will include formal usability testing from the perspective of oncology providers and patients.

154 Background: Pts with mCRPC have a poor prognosis with limited treatment options, including minimal response to immunotherapies. REGN5678 is a first-in-class, full-length anti-PSMAxCD28 bispecific costimulatory antibody designed to target prostate cancer cells and enhance T-cell activation. We report preliminary results from the dose escalation part of a first-in-human, open-label, Phase 1/2 study (NCT03972657) examining REGN5678 in combination with cemiplimab, a PD-1 blocking antibody. Methods: Pts with mCRPC had received ≥2 lines of systemic therapy in the metastatic and/or castration-resistant setting, including ≥1 second-generation anti-androgen. Pts received REGN5678 weekly at dose levels [DL] 0.1–300 mg, initially as monotherapy for 3 weeks, followed by combination with cemiplimab (350 mg Q3W) until progression or toxicity. Primary objectives are safety, tolerability, and pharmacokinetics. Preliminary efficacy measurements include decline in prostate-specific antigen (PSA) from the start of combination treatment and radiographic response from baseline. Results: At the data cutoff (DCO; July 27 2022), 35 pts had been treated. Treatment-emergent adverse events (TEAEs) ≥Grade (G)3 occurred in 54% (19/35) of pts. Cytokine release syndrome occurred in 6 pts (all G1) and there were 2 dose-limiting toxicities (both G3): pain (at 1 mg) and Guillain-Barré syndrome (at 300 mg). 4 pts (11%) experienced a ≥G3 immune-mediated adverse event (imAE; at DLs 30–300 mg). REGN5678 exposure was non-linear over the tested DLs (more than dose proportional). There were minimal signs of efficacy at lower DLs (REGN5678 0.1–10 mg), with only 1/16 pts showing a PSA decline (of 21%). More PSA declines occurred at higher DLs: 1/4 pts at 30 mg (a PSA decline of 100%), 3/8 at 100 mg (>99%, 44%, 22%), and 3/4 at 300 mg (>99%, 99%, 82%). Notably, all ≥G3 imAEs occurred in pts with PSA declines. Among pts with measurable disease and ≥1 on-treatment scan, radiographic response per RECIST 1.1 occurred in 1/3 pts at 30 mg (complete response), 1/4 at 100 mg (unconfirmed partial response [PR]), and 1/1 at 300 mg (PR confirmed after DCO). Conclusions: Preliminary data on REGN5678 plus cemiplimab in pts with mCRPC provide first evidence of clinical activity of a CD28 co-stimulatory bispecific antibody in solid tumors. Clinical activity was observed at DLs 30–300 mg. ≥G3 imAEs occurred in pts with PSA declines, suggesting a possible association. The study is ongoing to determine the maximum tolerated and recommended Phase 2 doses. Clinical trial information: NCT03972657 . [Table: see text]

TPS751 Background: Five doublet combinations incorporating immune checkpoint inhibitor (ICI) with or without tyrosine kinase inhibitor (TKI) are now approved for the first line treatment (tx) of advanced ccRCC. In COSMIC-313, the triplet combination with cabozantinib (CABO)-ipilimumab (IPI)-nivolumab (NIVO) demonstrated promising efficacy compared to IPI-NIVO in pts with intermediate/poor risk disease, but definitive overall survival (OS) benefit has yet to be shown, and is associated with higher incidence of adverse events. Currently, the optimal first /subsequent lines of therapies remain to be defined, and predictive biomarker to guide optimal tx approach is urgently needed. Pre-clinical studies allude to the ability of TKIs, such as CABO, to promote an immune-permissive environment and thus enhance ICI response. This open-label, single-arm, investigator-initiated pilot trial will evaluate the rapid sequencing from CABO to IPI-NIVO with the hypothesis of enhanced efficacy when compared to IPI-NIVO doublet, and potentially lower toxicity when compared to the concurrent triplet combination. Correlative studies, before and after administration of each systemic therapy, will be performed with the goal of identifying potential predictive biomarkers of response and resistance. Methods: Advanced, ccRCC pts will be treated with CABO 60mg PO once daily (QD) for 12 weeks (W) followed by IPI 1mg/kg + NIVO 3mg/kg IV every (Q)3W for 4 cycles. Subsequent therapy is dependent on response, and can be NIVO 480mg IV Q4W (CR/PR/SD), lenvatinib 18mg + everolimus 5mg PO QD (PD on both prior regimens), or re-initiation of CABO 60mg PO QD (CR/PR/SD at CABO but PD on IPI-NIVO). Key inclusion criteria are histologically confirmed, metastatic or unresectable RCC with predominantly clear cell component (sarcomatoid differentiation

135 Background: mCRPC is associated with (a/w) increased mortality among prostate cancer pts and new and safe txs are needed. In recent clinical trials, ARTs improved survival outcomes within this population, however, a significant portion will go on to discontinue tx. We aim to describe reasons for ART tx discontinuation and to identify predictors a/w increased risk of tx discontinuation due to tx toxicity or death. Methods: We performed a single-institution retrospective review of mCRPC pts on ART between 2010 and 2021. We screened for demographics, medical history, ART course, and tx side-effects. Our primary aim was to identify risk factors for tx discontinuation due to toxicity or death. Our secondary aim was to describe ART discontinuation among mCRPC. Fisher’s Exact was used to identify significant predictors of tx discontinuation due to toxicity or death. Significant outcomes were included in a multivariate logistic regression to determine odds ratios. Results: 133 pts with mCRPC started and discontinued ARTs. Among this cohort, 27 pts (20.3%) discontinued tx due to death or toxicity. Reasons for tx discontinuation are described in the table below. Significant predictors of tx discontinuation due to toxicity or death on bivariate analysis included pt-reported sepsis, hypertension (htn), weakness, falls, ECOG ≥2, and hospitalization during tx. Adjusting for confounders, pt-reported falls (OR 2.35; [0.91-2.59]; p=0.009), htn (OR 2.06; [0.94-2.21]; p=0.027), and weakness (OR 1.63; [0.59-2.72]; p=0.007) were a/w an increased risk of pt tx discontinuation from toxicity or death. Conclusions: Our results illustrate that the majority of mCRPC pts discontinued tx due to progression of disease. The data also indicates that mCRPC pts reporting new onset htn, falls, or weakness were more likely to discontinue tx due to toxicity or death. Early monitoring of this population is warranted to prolong duration of tx and prevent unnecessary txs. [Table: see text]

118 Background: mCRPC pts with significant comorbidities or poor performance status (PS) are often excluded from clinical trials for ART. These cohorts may not be representative of or comparable to real-world pts. We aimed to determine whether (1) baseline characteristics or (2) survival outcomes differed in pts with mCRPC categorized as eligible or ineligible for ART clinical trials. Methods: We screened for pts with mCRPC treated with abiraterone (abi-) or enzalutamide (enza-) at our institution between 2010 and 2021. Baseline characteristics were reviewed at treatment start and ART course recorded. We applied the inclusion/exclusion criteria from clinical trials, COU-AA-301, COU-AA-302, PREVAIL, and AFFIRM, to categorize pts as ineligible or eligible. Fischer’s exact was used to evaluate pt characteristics. Cox proportional hazards model was used to compare survival outcomes. Results: We identified 150 pts with mCRPC who initiated abi- or enza-. 54.7% of pts were deemed trial ineligible. Among pts on abi-, 48 of 86 (55.8%) pts were ineligible. Among those on enza-, 34 of 64 (53.1%) were ineligible. Common reasons for ineligibility included poor PS, radiation during treatment, and trial specific comorbidities. Among ineligible pts, 11% were post-chemotherapy compared to 20.6% of eligible pts (p = .162). Notably, trial ineligible pts were more likely to be non-Hispanic black when compared to trial eligible pts (29.3% vs. 17.6%, p = 0.036). Median OS was 30.2 for ineligible pts and 37.1 for eligible pts p = 0.46). for the ineligible pts and eligible pts respectively and there was no significant difference on univariate or multivariate analyses (HR: 0.84 [0.54, 1.32;]. Conclusions: Pts classified as trial ineligible had comparable survival outcomes when compared to trial eligible pts. Consistent with prior research on systemic racism in clinical trials, ineligible pts were more likely to identify as Black compared to eligible pts. Further research is warranted to confirm safety of ART in ineligible pts and identify strategies for reducing racial disparity gaps in cancer clinical trials.

627 Background: Current practice guidelines for the management of renal cell carcinoma (RCC) with brain metastases (BM) recommend a multimodal approach with systemic therapy, radiation, and/or surgery. The purpose of this study is to assess treatment patterns and identify prognostic factors for survival in this population. Methods: This retrospective, single-center study evaluated patients with RCC who were diagnosed with metastatic disease between 2015-2020 and were identified to have BM at any point in their care. Fisher's exact test and Mann-Whitney test were used to compare differences among groups. Kaplan-Meier curves were used to estimate time to events from the date of BM diagnosis. Univariate Cox proportional hazards model was used to estimate the effect of factors on survival. Results: 29 patients are included. Median follow-up was 25 months (range: 0.5-73). 16 (55%) patients presented with BM at the time of diagnosis of metastatic disease. 17 (58%) were treated with radiation. 8 (28%) were treated with surgery of which 7 also received radiation. Median overall survival was 10 months (95% CI: 7-not reached). Patients received a median of one (range 0-4) system therapy regimen after diagnosis of BM. 19 (66%) patients were exposed to at least one class of medication therapy for management of BM while 10 (34%) were exposed to two or more classes. 10 patients experienced long term survival (defined as survival ≥12 months). Patients with long term survival were more likely to have received radiation (90% vs 42%; P=0.02). However, use of radiation was not associated with survival (HR 0.47, 95% CI 0.1-1.3). Conclusions: The study is limited by sample size but suggests there is a population of patients with RCC with BM who can experience long term survival. Optimal selection for patients with BM that can be feasibly treated within radiation fields appears favorable. Most patients are not exposed to more than one systemic medication class during their care. Additional research is needed to compare efficacy in BM to guide first line systemic therapy selection. [Table: see text]

194 Background: Statins may reinforce and provide a compounded effect on ART by decreasing cholesterol levels thus decreasing de novo androgen synthesis. We aim to investigate the clinical effect of concurrent statin and ART use on outcomes of mCRPC patients. Methods: We conducted a retrospective analysis on mCRPC patients receiving ART from a single institution. Relevant demographic and clinical data was collected in addition to ART treatment course, statin treatment, and survival outcomes. Our primary outcome was PSA progression free survival (PFS) and our secondary outcomes were overall survival (OS) and associated prognostic factors for both PSA PFS and OS. Chi-squared test and Wilcoxon signed ranked test were used to compare baseline characteristics, and a Cox proportional hazards regression model was used to estimate hazard ratios (HR) with 95% confidence interval (CI) for overall survival (OS) and PSA PFS. Results: 153 patients were included in the analysis between 2010 and 2021. 67 patients (mean age 73.8 years) received concurrent statins and 86 patients (mean age 67.6 years) did not. Median PSA PFS was 37.4 months for patients that received concurrent statins and 17.4 months for patients that did not receive statins. On univariate and multivariate analyses, there was no statistically significance difference between groups for PSA PFS (HR 0.7; CI 0.44-1.1; p=0.122). Median OS was 35.6 months for patients that received concurrent statins and 24.0 months for patients who did not. There was no statistical significance between groups for OS on univariate or multivariate analyses (HR 0.67; CI 0.42-1.06; p=0.087). Conclusions: Although results were not statistically significant, our study illustrates that concurrent statin use exhibits improved time to PSA progression and OS in mCRPC patients. Larger multi-center and further prospective studies are warranted to elucidate the relationship between statin use and overall outcomes in this population.

684 Background: In the COSMIC-021 phase 1b study (NCT03170960) evaluating cabozantinib plus atezolizumab in advanced solid tumors, this combination therapy demonstrated encouraging clinical activity in patients with advanced non-clear cell renal cell carcinoma (nccRCC) with a median follow-up of 13 mo (Pal. JCO 2021). Results after extended follow-up in nccRCC are presented. Methods: Patients with advanced nccRCC and ECOG PS 0/1 who had ≤1 prior VEGFR-targeting tyrosine kinase inhibitor (TKI) were eligible. Prior treatment with TKIs targeting MET or immune checkpoint inhibitors was not allowed. Patients received cabozantinib 40 mg PO QD plus atezolizumab 1200 mg IV Q3W until unacceptable toxicity or progression; dose reductions of cabozantinib (40 mg QD to 20 mg QD, then to 20 mg QOD) were permitted to manage adverse events. The primary endpoint was objective response rate (ORR) per RECIST v1.1 by the investigator; other endpoints included safety, duration of response (DOR), PFS, and OS. Results: The study enrolled 32 patients with nccRCC (2 from dose escalation phase, and 30 from expansion phase of the study): median age, 62 y; male, 81%; ECOG PS 0/1, 75%/25%; histology, papillary/chromophobe/clear cell/other, 47%/28%/3%/22%; sarcomatoid feature, 13%; IMDC risk favorable/intermediate/poor, 50%/41%/9%; ≥3 tumor sites, 56%; tumor sites, lung/kidney/bone/liver, 50%/25%/16%/16%; prior nephrectomy, 63%; prior VEGFR TKI, 22%; 0/1 lines of prior therapy (locally advanced/metastatic setting), 81%/19%. As of July 21, 2022, median follow-up was 37.2 mo (range 32.1–58.5) with 5 (16%) patients remaining on study treatment. ORR by investigator was 31% (all PRs) and disease control rate was 94% (Table); median DOR was 8.1 mo. Median PFS was 9.3 mo (95% CI 5.5–12.3), and median OS was not reached (95% CI 23.0–NE). PFS and OS estimates at 12 mo were 34% and 84%, respectively; 24-mo estimates were 6% and 70%. Treatment-related AEs occurred in 97% (grade 3/4, 53%); the most common AEs included diarrhea (69%), palmar-plantar erythrodysesthesia (50%), fatigue (44%), dysgeusia (41%), hypertension (31%) and nausea (31%). One grade 5 treatment-related AE of pulmonary hemorrhage occurred. Treatment-related AEs leading to discontinuation of both study treatments occurred in 13% of patients. Conclusions: Extended 3-year follow-up reinforces the encouraging clinical activity of cabozantinib plus atezolizumab in advanced nccRCC with a manageable safety profile. Clinical trial information: NCT03170960 . [Table: see text]