Background Plasma cell-free Human Papillomavirus DNA (cfHPVDNA) is a biomarker for oropharyngeal carcinoma. Existing diagnostics may be limited by inadequate sensitivity or high cost/complexity for longitudinal monitoring. Objectives We hypothesized that sensitive and specific plasma cfHPVDNA detection may be achieved via a highly-multiplex qPCR method. Study Design We designed and validated a single-tube one-step genotype-specific qPCR assay for detection of cfHPV16DNA in human plasma using >8,000 genomes spanning 18 genotypes. Amplicons were optimized for cfHPVDNA fragment size. Results The cfHPV16DNA qPCR amplicons spanned 16% of the HPV16 genome. Amplicons were conserved in a median of 99.0% of 3,944 genomes in silico. The 95% lower limit of detection was 0.35 genome copies/reaction and the limit of blank was 0. Multiplexing achieved a tenfold improvement in sensitivity compared with single amplicons using in silico simulations of cfHPVDNA fragmentation, which was in close agreement with experimental observations. An assay was replicated for HPV18 with similar observations.Among 36 patients with head/neck mucosal carcinomas (26 HPV-positive, 12 HPV-negative), there was 100% concordance with tissue HPV status and with NavDx digital PCR. Pre-treatment specimens with sub-genomic cfHPVDNA concentration were detected. False negatives were observed with single amplicons but not with this multiplexed method. Among 17 patients with post-treatment landmark specimens, there was 100% PPV and 100% NPV for recurrence. Conclusions This assay is specific for plasma cfHPVDNA detection and prognostic for recurrence. Sub-genomic sensitivity was in close agreement with in silico simulations. The format might be more accessible than dPCR or NGS for longitudinal testing.

Older patients have similar immune checkpoint inhibitor efficacy and rates of adverse events as younger patients, but appear to have decreased tolerability, particularly in the oldest patient cohort (>80 years), often leading to early cessation of therapy. We aimed to determine whether early discontinuation impacts efficacy of anti-PD-1 therapy in patients ≥80 years old. In this retrospective, multicenter, international cohort study, we examined 773 patients with 4 tumor types who were at least 80 years old and treated with anti-PD-1 therapy. We determined response rate, overall survival (OS), and progression-free survival (PFS) in patients who discontinued therapy early (<12 months) for reasons other than progression or death. We used descriptive statistics for demographics, response, and toxicity rates. Survival statistics were described using Kaplan Meier curves. Median (range) age at anti-PD-1 initiation was 83.0 (75.8-97.0) years. The cancer types included were melanoma (n=286), non-small cell lung cancer (NSCLC) (n=345), urothelial cell carcinoma (UCC) (n=108), and renal cell carcinoma (RCC) (n=34). Of these, 102 met the primary endpoint of <12 months to discontinuation for reasons other than death or progression. Median PFS and OS, respectively, for these patients were 34.4 months and 46.6 months for melanoma, 15.8 months and 23.4 months for NSCLC, and 10.4 months and 15.8 months for UCC. This study suggests geriatric patients who have demonstrated therapeutic benefit and discontinued anti-PD-1 therapy at less than 12 months of duration for reasons other than progression may have durable clinical benefit without additional therapy.

9585 Background: Neoadjuvant immunotherapy is a potential curative approach to patients with locally advanced cutaneous squamous cell carcinoma (cSCC) that offers an attractive alternative to traditional, often morbid surgery and/or radiation. In this study, we report our initial institutional experience treating patients with primary immunotherapy monotherapy (PRIMO) and reserving surgery or radiation for progression only. Methods: Patients with primary or recurrent locally advanced cSCC (AJCC 8 T3-4 or node positive or in-transit metastases) in whom surgical resection and/or definitive radiation were deemed excessively morbid or futile and were treated with PRIMO were included in an IRB-approved database. Patients were treated with IV cemiplimab (350mg q 3week) or pembrolizumab (200mg q3week or 400 q6week). Complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD) was scored according to iRECIST criteria. Kaplan Meier analysis was used to estimate overall survival (OS) and progression free survival (PFS). Univariate analysis (UVA) for PD was performed using Chi square for categorical variables and Kruskal-Wallis for continuous variables. Results: This study included 36 patients treated between 2017-2023, with a median age of 80 (61-96), and a median follow up of 13.5 months (8-20.5). Most patients had lesions on the head and neck (32; 88.9%), recurrent disease (26; 72.2%), and T3/4N0 disease (20; 55.6%), while 18 patients (50.0%) had nodal disease or in-transit metastases. Cemiplimab was used in 31 patients (86.1%) while pembrolizumab was used for 5 patients (13.9%). Twelve patients (33.3%) stopped PRIMO due to an immune-related adverse event (irAE). 1 and 2yr OS and 1 and 2yr PFS were 76%, 64%, 72% and 51%, respectively. Best initial response to PRIMO was a CR in 15 (41.7%), PR in 14 (38.9%), SD in 3 (8.3%) and PD in 4 (11.1%) patients. All 3 lesions (100%) with SD and 3/14 lesions (21.4%) with PR ultimately progressed with a median duration of response of only 3 months (2.0-6.8), while all 15 lesions (100%) with a CR and 11/14 lesions with a PR (78.5%) remain controlled at last follow up with a median duration of response of 15.5 months (8.8 – 23.3) (p<0.001). The median duration of months of ongoing response after the completion of PRIMO was 11 (0-57). The median number of treatment cycles was 14 (2-36) in all patients, and 16 (2-36) in the 26 patients that did not progress. The only variable significantly associated with PD on UVA was the lack of irAE; 10/26 (38.5%) patients who did not have an irAE experienced PD, while 0/12 patients (0%) who had an irAE experienced PD (p=0.035). Conclusions: The use of PRIMO for locally advanced cSCC produces impressive response rates that appear durable without any additional therapy. This attractive alternative to the emerging neoadjuvant paradigm deserves prospective validation with longer term follow up.

e15034 Background: Circulating tumor DNA (ctDNA) has emerged as a potential noninvasive biomarker to assess tumor burden throughout cancer treatment and to aid in clinical decision-making. However, data on the use of ctDNA in cutaneous squamous cell carcinoma (cSCC) is currently lacking. In this study, we report our first experience with ctDNA in patients with cSCC. Methods: In this IRB-approved study at a tertiary care center, an initial cohort of patients with intermediate to high-risk cSCC had tissue and plasma sent for ctDNA analysis using Natera®’s proprietary bioinformatics pipeline. This cohort included 21 patients treated from 2023 to the present. Initial tissue samples were obtained through biopsy (i.e., excisional, punch, or shave), Mohs surgery, or surgical resection. Some patients had additional plasma sample(s) drawn later during or after the treatment course. Patients either received a “positive,” “positive below analytical range,” or “negative” result or had ‘insufficient sample’ to receive a result based on Natera®’s analysis. All patients were treated curatively with surgical resection with or without adjuvant radiation as per the standard of care. The assessment of gross disease was determined based on whether the patient had disease present on imaging or physical exam at the time of the plasma draw. A comparison of ctDNA results and the presence of gross disease was done using Fisher’s exact test. Results: Of the samples submitted for Natera® ctDNA analysis, 18/21 had sufficient tissue for analysis of tumor DNA. All 3 tumors with insufficient tissue were from biopsies (6/9 sufficient), while 12/12 of those from Mohs or surgical resection had sufficient tissue. The average tumor volume in these patients with insufficient samples for tumor analysis was relatively low (0.35 cm3) compared to those that resulted (17-30 cm3). 7/9 (78%) patients with gross disease had a positive ctDNA result, which was significantly greater than those with no gross disease (1/9 patients, 11%) (p = 0.004). Three patients in our cohort with an initially positive ctDNA result had multiple plasma draws, with all having a later draw demonstrating a negative ctDNA result after having surgery and/or radiation. Conclusions: Circulating tumor DNA is a reliable marker for patients with cSCC with a high tumor burden and is feasible in those with adequate tissue. However, in patients where it is only possible to obtain a small volume of tumor tissue and/or low tumor burden, ctDNA may be a less reliable biomarker. [Table: see text]

Background The combination of ipilimumab and nivolumab is a highly effective treatment for metastatic cutaneous melanoma. However, immune-related adverse events (irAEs) are common, often necessitating treatment interruption and the use of immunosuppressive agents. There is no data on the impact of resuming nivolumab on survival following recovery from the irAE and completion of immunosuppressive treatment.Patients and methods In this retrospective analysis, we examined a cohort of patients treated with ipilimumab/nivolumab who developed irAEs requiring treatment interruption and immunosuppressive therapy. The differences in physician practice patterns at our institution allowed us to examine the survival effect of restarting single-agent nivolumab. A multivariate analysis of clinical factors associated with improved survival was performed.Results We identified 165 patients who were treated with ipilimumab/nivolumab and developed irAEs requiring treatment interruption and immunosuppressive therapy. Patients with the best overall response of progressive disease were excluded. Of the remaining 122 patients, 46 resumed single-agent nivolumab. When stratified by age and adjusted for sex, M-stage, lactate dehydrogenase (LDH), therapy duration, and irAE type, the effect of resumption of nivolumab on survival was highly significant (p=0.02). Patients who resumed nivolumab had a 68% reduction in the hazard of death compared with patients who had not yet or never resumed nivolumab (HR: 0.32, 95% CI: 0.12 to 0.84). Of the patients who resumed nivolumab, 12 (26%) patients had subsequent irAEs, with five patients having grade 3 irAEs. No grade 4 or 5 irAEs were noted.Conclusions Resuming single-agent nivolumab following a treatment interruption for ipilimumab/nivolumab-associated irAE and completion of immunosuppressive therapy increased overall survival compared with discontinuing nivolumab permanently in patients with metastatic melanoma. Toxicity observed post-resumption of single-agent nivolumab was manageable with no severe irAEs observed.

Patients with treatment-naive mRCC are treated with IO-based regimens, usually a combination of either IO/IO or IO/TKI. Both IO and TKI therapies are associated with an increased risk of thromboembolism, however, there are no head-to-head trials comparing IO/IO to IO/TKI regimens. Our study did not show a significant difference in the rate of TE between patients treated with IO/IO or IO/TKI. Background: Most patients with treatment-naive metastatic renal cell carcinoma (mRCC) receive combination-based immunotherapy with either 2 immune-oncology checkpoint inhibitors (IO/IO) or an IO agent in combination with a vascular endothelial growth factor receptor (VEGF-R) tyrosine kinase inhibitor (IO/TKI). The rates of thromboembolism (TE) in these cohorts are not clear ly descr ibed and can potentially impact decision-making between IO/IO and IO/TKI. Methods: We conducted a retrospective investigation of patients with treatment-naive mRCC treated with IO-based combinations between January 2015 and April 2021 at the Cleveland Clinic. TE events, including venous and arterial, were identified in each group. Competing risk regression was done to identify factors associated with the development of TE following therapy, with all -cause mortality treated as a competing event. Results: Of 180 patients identified, 77 (43%) received IO/TKI and 103 (57%) received IO/IO. Median age was 65 years, 75% were male, and 80% had clear cell histology. Baseline characteristics were similar between the 2 groups. At a median follow-up of 22.0 months, 10.0% of all patients had a TE. The one-year incidence of TE was 8.1% (95% CI: 3.3%-15.8%) with IO/TKI and 9.8% (95% CI: 5.0%-16.5%) with IO/IO and was not significantly different between the 2 groups (HR 0.89, 95% CI: 0.35%-2.28%). Occurrence of TE was associated with decreased overall survival regardless of IO/IO or IO/TKI therapy (HR 2.80, 95% CI: 1.57-5.02). There was no difference in incidence of TE based on patient age, gender, prior history of TE, International Metastatic Renal Cell Carcinoma (IMDC) risk group, or Khorana score. Conclusions: Incidence of TE is similar between IO/IO and IO/TKI regimens in treatment-naive mRCC and is also associated with decreased overall survival. While risk of TE may not guide decision-making in choice of front-line mRCC therapy, careful attention should be given to the high risk of TE in this population.

BackgroundIn the era of immune checkpoint blockade, the role of cancer vaccines in immune priming has provided additional potential for therapeutic improvements. Prior studies have demonstrated delayed type hypersensitivity and anti-tumor immunity with vaccines engineered to secrete granulocyte-macrophage colony-stimulating factor (GM-CSF). The safety, efficacy and anti-tumor immunity of GM-CSF secreting vaccine in patients with previously treated stage III or IV melanoma needs further investigation.MethodsIn this phase II trial, excised lymph node metastases were processed to single cells, transduced with an adenoviral vector encoding GM-CSF, irradiated, and cryopreserved. Individual vaccines were composed of 1x106, 4x106, or 1x107 tumor cells, and were injected intradermally and subcutaneously at weekly and biweekly intervals. The primary endpoints were feasibility of producing vaccine in stage III patients and determining the proportion of patients alive at two years in stage IV patients.ResultsGM-CSF vaccine was successfully developed and administered in all 61 patients. Toxicities were restricted to grade 1-2 local skin reactions. The median OS for stage III patients (n = 20) was 71.1 (95% CI, 43.7 to NR) months and 14.9 (95%CI, 12.1 to 39.7) months for stage IV patients. The median PFS in stage III patients was 50.7 (95%CI, 36.3 to NR) months and 4.1 (95% CI, 3.0-6.3) months in stage IV patients. In the overall population, the disease control rate was 39.3% (95%CI, 27.1 to 52.7%). In stage III patients, higher pre-treatment plasma cytokine levels of MMP-1, TRAIL, CXCL-11, CXCL-13 were associated with improved PFS (p<0.05 for all). An increase in post-vaccination levels of IL-15 and TRAIL for stage III patients was associated with improved PFS (p=0.03 for both). Similarly, an increase in post-vaccination IL-16 level for stage IV patients was associated with improved PFS (p=0.02) and clinical benefit.ConclusionsVaccination with autologous melanoma cells secreting GM-CSF augments antitumor immunity in stage III and IV patients with melanoma, is safe, and demonstrates disease control. Luminex data suggests that changes in inflammatory cytokines and immune cell infiltration promote tumor antigen presentation and subsequent tumor cell destruction. Additional investigation to administer this vaccine in combination with immune checkpoint inhibitors is needed.

Background: Immune checkpoint inhibitor therapy (ICI) has been variably associated with increased risk of VTE, with some reports suggesting increased risk and others suggesting risk is not different from the baseline venous thromboembolism (VTE) risk in unselected ambulatory patients with cancer. Methods: Patients treated with ICI at our institution between 2009 and 2022 were retrospectively identified. Standard baseline characteristics (age, sex, race/ethnicity, BMI), cancer type and stage, type of ICI, KS risk score variables not already captured including white cell count, hemoglobin, and platelet count at time of initiating ICI therapy, as well as other patient and treatment characteristics were collected. Those with history of VTE prior to start of ICI were removed from the analysis population; anticoagulation after starting ICI was a time dependent factor. Cumulative incidence of VTE with death as a competing risk was determined using Fine and Gray technique. Univariate analysis for baseline characteristics was completed. Characteristics that were statistically significant at the 0.1-level or less were carried forward as candidate predictors in multivariable, competing risks models of VTE. Results are presented as hazard ratios with 95% confidence intervals adjusted for multiple comparisons against a reference group. Results: Of 12,007 patients starting ICI therapy, 10,638 were analyzed with median follow-up of 25.1 months. The majority of tumor types included lung (30.3%), skin (15.8%; melanoma, squamous cell), gastrointestinal (11.7%), and genitourinary (11.6%). The most frequently used ICI agents were the anti-PD-1 monoclonal antibodies pembrolizumab (49.7%) and nivolumab (31.8%). The overall cumulative incidence estimates of VTE are summarized in Figure 1. The estimate at six months was 7.6% (95% CI: 7.1 to 8.1%); at one year: 11.1% (95% CI: 10.5 to 11.8%); and at two years 13.8% (95% CI: 13.0 to 14.5%). Among the 1,328 patients who developed VTE, the median time after starting ICI therapy was 5.4 months [IQR: 1.8 to 12.0 months]. Univariate comparison findings carried forward as candidate predictors in the multivariable analyses included sex, age, BMI, race, history of prior anticoagulant therapy (before ICI start), cancer type, disease stage, ICI therapy (by drug), KS, and platelets. The first multivariable model using all predictors identified gender, Khorana risk score, and race as no longer significant. The second model examined the inter-relationships between remaining predictors and the incidence of VTE; as the KS is partially comprised of cancer type, BMI, and platelets, having these as separate predictors removed the influence of the Khorana score since they would provide similar information. Predictors of VTE identified in this model included type of ICI with the use of ipilumumab associated with the highest risk with HR of 1.90 (95% CI: 1.33-2.71) and durvalumab the lowest HR 0.59 (95% CI:0.36-0.96), when compared with pembrolizumab. Type of cancer, stage of cancer, BMI, and high platelet count were also predictive. Starting anticoagulation at the time of or after beginning ICI but not for treatment of VTE was associated with approximately a 40% reduced hazard of VTE compared to patients who had not yet or never started anticoagulation. (HR: 0.59, 95% adj. CI: 0.48 to 0.73). (Table 1) Conclusion In this large sample of patients, the use of ICI in VTE naïve patients was associated with an increased risk of VTE, with a cumulative incidence of 7.6% at 6 months compared with a baseline 3-4% risk in general ambulatory patients with cancer. Type of ICI treatment was associated with VTE risk, as was stage of cancer. The individual KS risk predictors of cancer type and BMI were predictive but only elevated platelet count was associated with increased VTE risk-hemoglobin and white cell count were not. Importantly, those that started anticoagulation for reasons other than development of VTE had a significant reduction in VTE risk, suggesting that anticoagulant prophylaxis may be indicated in those patients with multiple risk factors in the setting of ICI treatment. Further analysis of these data will help identify at risk patients starting treatment with ICI who will benefit from VTE prophylaxis.

9520 Background: The combination of ipilimumab (ipi), and nivolumab (nivo) has been shown to be a highly effective treatment for metastatic melanoma. However, immune-related adverse events (irAEs) are common with this treatment, leading to treatment interruption and the use of immunosuppressive agents. There are no data on the impact of resuming single agent nivo on survival outcomes following recovery from the irAE and completion of immunosuppressive treatment. Some physicians hold therapy permanently as done on clinical trials and other physicians resume nivo after steroid taper. Methods: In this retrospective analysis, we examined a cohort of patients treated with ipi/nivo who developed irAEs that required treatment interruption and immunosuppressive therapy. The differences in practice patterns among physicians at our institution allowed us to examine the effect of single agent nivo resumption on survival after treatment of irAEs. Multivariate analysis of clinical factors associated with improved survival was performed. Results: We identified 165 patients who were treated with ipi/nivo and developed irAEs requiring treatment interruption and immunosuppressive therapy. To remove any impact of immediate disease progression on the analysis patients who progressed on ipi/nivo prior to or during treatment hold were removed from analysis. Of the remaining 131 patients, 45 resumed single agent maintenance nivo following the resolution of toxicity and completion of steroid taper. When stratified by age and adjusted for sex, M-stage, LDH, duration of therapy, and type of irAE causing the treatment pause, the effect of resumption of nivo upon survival remained highly significant (p = 0.001). Patients who resumed nivo had an 82% reduction in the hazard of death compared with patients who had not yet or never resumed nivo (HR: 0.18, 95% CI: 0.06 to 0.52). Of the patients who resumed nivo, 12 (26%) patients had subsequent irAEs with 5 patients having grade 3 irAEs (only one of whom had a recurrence of the prior ipi/nivo associated toxicity) and no grade 4 or 5 irAEs were noted. Conclusions: The resumption of single agent maintenance nivolumab following a treatment hold for ipi/nivo associated irAE and completion of immunosuppressive therapy increased overall survival compared to discontinuing nivo permanently in patients with metastatic melanoma after stratification for many factors. Toxicity observed post resumption of single agent nivolumab was manageable with no severe irAEs observed. These results suggest that restarting maintenance nivo post irAEs should be studied prospectively.

12082 Background: Emerging reports suggest high rates of venous thromboembolism (VTE) and arterial thromboembolism (ATE) with immune checkpoint inhibitors (ICI) in pts with melanoma, but it is unclear whether these are truly increased compared to older systemic therapy approaches. We assessed the incidence of TE in melanoma pts on ICI, cytokine therapy (CY), and chemotherapy (chemo), and evaluated its impact on survival. Methods: We conducted a cohort study using the SEER-Medicare database to evaluate rates of TE in pts with melanoma treated from 2008-2019 with ICI (ipilimumab, nivolumab, pembrolizumab), CY (IL2, IFN), and/or chemo (antineoplastic agents, BRAF/MEK inhibitors) within two years of treatment initiation. TE including VTE events of deep venous thrombosis, pulmonary embolism, and ATE of MI, ischemic stroke, and transient ischemic attack were identified by at least two outpatient claims or one inpatient claim. Overall survival (OS) from treatment start was analyzed by time-varying Cox analysis. Results: The cohort comprised 13,124 pts with median age 75 (24-101) years and 68% male. Of these, 14.8% received ICI, 48.9% chemo, 1.7% CY, 31.8% chemo+ICI, and 2.8% chemo+CY. At treatment start, comorbidities included 72% of pts with hypertension, 33% cerebrovascular disease, 17% atrial fibrillation, 7.3% history of VTE, and 13.0% history of ATE. Overall, 11.4% were on anticoagulation and 5.2% on antiplatelet agents. Incidence rates of ATE and VTE after treatment start are shown (Table). VTE was highest at 3 months after starting therapy with 19.6 events per 100 person-years in those receiving chemo+ICI, 17.1 events with chemo, 15.4 events with ICI, 10.1 events with chemo+CY, and 7.7 events with CY. In multivariable analysis, VTE and ATE were associated with worse OS compared to patients without (HR 2.77 [95%CI, 2.50-3.08], and HR 2.53 [95%CI, 2.23-2.86], respectively). Conclusions: Systemic therapy with ICI and chemo, especially when exposed to both sequentially or concurrently, demonstrates a high incidence of TE in pts with melanoma. TE is associated with substantial worsening of survival. TE rates are not substantially increased with ICI in comparison to chemo. Further studies are needed to identify benefit of thromboprophylaxis.[Table: see text]

Background To evaluate factors affecting the utilization of immunotherapy and to stratify results based on the approval of ipilimumab in 2011 and programmed death-1 inhibitors in 2014, an analysis of available data from the National Cancer Database (NCDB) was performed.Methods The NCDB was analyzed to identify patients with stage IV melanoma from 2004 to 2016. Patients were categorized during the time periods 2004–2010, 2011–2014, and 2015–2016. Overall survival (OS) was analyzed by Kaplan-Meier, log-rank, and Cox proportional hazard models; IO status was analyzed using logistic regression.Results 24,544 patients were analyzed. Overall, 5238 patients (21.3%) who received IO had improved median OS compared with those who did not (20.2 months vs 7.4 months; p<0.0001). Between 2004 and 2010, 9.7% received immunotherapy; from 2011 to 2014, 21.9% received immunotherapy; and from 2015 to 2016, 43.5% received immunotherapy. Three-year OS significantly improved in patients treated with IO across treatment years: 31% (95% CI 29% to 34%) from 2004 to 2010, 35% (95% CI 33% to 37%) from 2011 to 2014, and 46% (95% CI 44% to 48%) from 2015 to 2016 (p<0.0001). Survival was worse in patients who did not receive IO during these treatment years: 16% (15%–17%), 21% (20%–22%), and 27% (25%–28%), respectively. In the overall cohort, age <65 years, female gender, private insurance, no comorbidities, residence in metropolitan area, and treatment at academic centers were associated with better OS (p<0.0001 for all). In the multivariate analysis, receipt of IO from 2015 to 2016 was associated with age <65 years (OR 1.27, 95% CI 1.08 to 1.50), African American race (OR 5.88, 95% CI 1.60 to 28.58), lack of comorbidities (OR 1.43, 95% CI 1.23 to 1.66), and treatment at academic centers (OR 1.44, 95% CI 1.26 to 1.65) (p<0.05 for all).Conclusions OS improved in patients with stage IV melanoma receiving IO, with the highest OS rate in 2015–2016. Our findings, which represent a real-world population, are slightly lower than recent trials, such as KEYNOTE-006 and CheckMate 067. Significant socioeconomic factors may impact receipt of IO and survival.

BACKGROUND:Immunotherapy has become one of the mainstays for metastatic urothelial carcinoma treatment. Whether immune checkpoint inhibitor therapy increases thromboembolism (TE) risk is unknown. OBJECTIVE:We investigated the incidence of arterial thromboembolism (ATE) and venous thromboembolism (VTE) events and its associated outcomes in patients with metastatic urothelial cancer treated with immune checkpoint inhibitors. METHODS:Patients with urothelial cancer treated with immune checkpoint inhibitors at the Cleveland Clinic from 1/1/2015 to 12/31/2019 were identified. The Kaplan-Meier method estimated overall survival and Cox proportional hazards regression evaluated the impact of TE on overall survival. RESULTS:Of 279 patients, 72% were men with pure urothelial cancer (62%) who started atezolizumab (40%), nivolumab (3%), or pembrolizumab (57%). At a median follow-up of 5.6 months (range 0.3-51.6), 42 patients developed a TE (VTE n = 37, 13%, ATE n = 5, 2%). The cumulative incidence of TE after immune checkpoint inhibitor therapy was 9.1% (95% confidence interval 6.0-13.0) at 6 months and 13.6% (95% confidence interval 9.6-18.4) at 12 months. Most TE (VTE 62%, ATE 100%) occurred within 6 months of immune checkpoint inhibitor initiation (median doses 5, range 1-59), and the majority (VTE 81%, ATE 100%) resulted in hospitalization (median: 5 days, 4 days, respectively). Thromboembolism (hazard ratio 2.296, p = 0.0004), Bajorin score 1 or 2 (hazard ratio 1.490, p = 0.0315), and Bajorin score 2 (hazard ratio 3.50, p < 0.0001) were associated with worse overall survival. CONCLUSIONS:Immune checkpoint inhibitors are associated with a high TE risk. Thromboembolism is associated with worsened survival, among other poor outcomes. Further investigation into the mechanism behind immune checkpoint inhibitor-associated TE is needed.

IMPORTANCE Geriatric (aged >= 80 years) patients are historically underrepresented in cancer clinical trials. Little is known about the efficacy of immune checkpoint inhibitors (ICIs) in geriatric patients. These agents are associated with immune-related adverse events (irAEs), which may be particularly associated with morbidity in this population. OBJECTIVE To provide insight into the clinical outcomes and safety of ICIs among geriatric patients (aged >= 80 years) with cancer. DESIGN, SETTING, AND PARTICIPANTS A Multicenter, international retrospective study of 928 geriatric patients with different tumors treated with single-agent ICIs between 2010 to 2019 from 18 academic centers in the US and Europe. Analyses were conducted from January 2021 to April 2021. MAIN OUTCOMES AND MEASURES Clinical outcomes and irAE patterns in geriatric patients treated with single-agent ICIs. RESULTS Median (range) age of the 928 patients at ICI initiation was 83.0 (75.8-97.0) years. Most patients (806 [86.9%]) were treated with anti-programmed cell death 1 therapy. Among the full cohort, the 3 most common tumors were non-small cell lung cancer (NSCLC, 345 [37.2%]), melanoma (329 [35.5%]), and genitourinary (GU) tumors (153 [16.5%]). Objective response rates for patients with NSCLC, melanoma, and GU tumors were 32.2%, 39.3%, and 26.2%, respectively. Median PFS and OS, respectively, were 6.7 and 10.9 months (NSCLC), 11.1 and 30.0 months (melanoma), and 6.0 and 15.0 months (GU). Within histologically specific subgroups (NSCLC, melanoma, and GU), clinical outcomes were similar across age subgroups (aged <85 vs >= 85 years). Among all 928 patients, 383 (41.3%) experienced >= 1 irAE(s), including 113 (12.2%) that were reported to be grade (G) 3 to 4 based on Common Terminology Criteria for Adverse Events (version 5.0). The median time to irAE onset was 9.8 weeks; 219 (57%) occurred within the first 3 months after ICI initiation. Discontinuation of treatment with ICIs owing to irAEs occurred in 137 (16.1%) patients. There was no significant difference in the rate of irAEs among patients aged younger than 85, 85 to 89, and 90 years or older. Despite the similar rate of G3 or higher irAEs, ICIs were discontinued owing to irAEs more than twice as often among patients aged 90 years or older compared with patients younger than 90 years (30.9% vs 15.1%, P =.008). CONCLUSIONS AND RELEVANCE The findings of this international cohort study suggest that treatment with ICIs may be effective and generally well tolerated among older patients with cancer, though ICI discontinuation owing to irAEs was more frequent with increasing age.

e21520 Background: The majority of data for IO in melanoma stems from trials that have specific inclusion criteria and often exclude important populations. In this analysis, we present the first real-world evidence of outcomes for stage IV patients (pts) with cutaneous melanoma receiving IO from 2015–2017 (addition of anti-PD-1 therapies) and factors associated with receipt of IO, and compare these outcomes with pts receiving IO (likely interferon and interleukin-2) from 2004–2010, and IO (addition of ipilimumab) from 2011–2014. Methods: The NCDB was analyzed to identify pts with stage IV melanoma from 2004–2017. Pts were categorized into receipt of IO or not during time periods 2004–2010, 2011–2014, and 2015–2017. Overall survival (OS) was analyzed by Kaplan-Meier, log-rank, and Cox proportional hazard models; IO status was analyzed using logistic regression. Results: 24,544 pts were analyzed from 2004–2017. Overall, 5,238 pts (21.3%) that received IO had improved median OS compared to those that did not (20.2 vs. 7.4 mos; p<0.0001). 4.1% received IO from 2004-2010, 7.8% from 2011-2014, and 9.4% from 2015-2017. Three-year OS significantly improved in pts treated with IO across treatment years; p<0.0001 (Table). Similarly, median time to IO initiation improved over time: 62 days in 2011-2014 vs. 49 days in 2015-2017 (p<0.0001). In the overall cohort, age<65 years, female gender, private insurance, no comorbidities, residence in metropolitan area, and treatment at academic centers were associated with better OS (p<0.0001 for all). Race was not associated with differences in OS (p=0.07). Multivariate analysis for 2015-2017 demonstrated that lack of treatment with IO [HR 1.98, 95%CI 1.81-2.17], radiation therapy [HR 1.20, 95%CI 1.09-1.32], presence of liver metastases [HR 1.87, 95%CI 1.63-2.13], and lack of surgery [HR 1.68, 95%CI 1.45-1.93] were independently associated with worse OS (p<0.0001 for all). In multivariate analysis, receipt of IO from 2015-2017, was associated with age<65 years [OR 1.27, 95%CI 1.08-1.50], African American race [OR 5.88, 95%CI 1.60-28.58], lack of comorbidities [OR 1.43, 95%CI 1.23-1.66], and treatment at academic centers [OR 1.44, 95%CI 1.26-1.65] (p<0.05 for all). Conclusions: OS was improved in stage IV melanoma pts receiving IO, with the highest OS rate in 2015-2017. Our findings that represent a real-world population are consistent with recent trials, like KEYNOTE 006 and CheckMate 067 where 3-year OS for anti-PD-1 therapy was 50% and 52%, respectively. In our study, African Americans demonstrate a 5-fold increase in likelihood of receiving IO, a population underrepresented in clinical trials. Significant socioeconomic factors may impact receipt of IO and survival.[Table: see text]

Over one third of American adults are at high risk for developing diabetes, which can be delayed or prevented using interventions such as medical nutrition therapy (MNT) or metformin. Physicians’ self-reported rates of prediabetes treatment are improving, but patterns of actual referral, prescription, and MNT visits are unknown. To characterize treatment of prediabetes in primary care. We conducted a retrospective cohort study using electronic health record data. We described patterns of treatment and used multivariable logistic regression to evaluate the association of patient factors and PCP-specific treatment rate with patient treatment. We included overweight or obese outpatients who had a first prediabetes-range hemoglobin A1c (HbA1c) during 2011–2018 and had primary care provider (PCP) follow-up within a year. We collected patient characteristics and the following treatments: metformin prescription; referral to MNT, diabetes education, endocrinology, or bariatric medicine; and MNT visit. We did not capture within-visit physician counseling. Of 16,713 outpatients with prediabetes, 20.4% received treatment, including metformin prescriptions (7.8%) and MNT referrals (11.3%), but only 7.4% of referred patients completed a MNT visit. The strongest predictor of treatment was the patient’s PCP’s treatment rate. Some PCPs never treated prediabetes, but two treated more than half of their patients; 62% had no patients complete a MNT visit. Being younger or female and having higher body mass index or HbA1c were also positively associated with treatment. Compared to white patients, black patients were more likely to receive MNT referral and less likely to receive metformin. Almost 80% of patients with new prediabetes never received treatment, and those who did receive referrals had very poor visit completion. Treatment rates appear to reflect provider rather than patient preferences.

Background: Patients with renal cell carcinoma (RCC) are at a higher risk of thromboembolism (TE) compared to other cancers. Immunotherapy (IO) is a mainstay in the treatment of advanced or metastatic RCC (aRCC). It is still unknown whether IO further increases the hypercoagulable status in this already high risk population. We present the first analysis of the incidence and outcomes of venous (VTE) and arterial (ATE) events in patients with aRCC treated with immunotherapy. Methods: All patients with aRCC treated with IO at the Cleveland Clinic from 1/2015 to 12/2019 were identified. Cumulative incidence analysis was done to calculate rates of TE over time, and Gray's test was used to determine difference in rates of TE among groups. Factors associated with overall survival (OS) were identified using Cox proportional hazards regression, and the impact of TE on OS was evaluated with TE analyzed as a time dependent covariate. Results: Of 351 pts, 75% were men (median age 65 years, 92% white). 81% had clear cell histology and 77% had International Metastatic RCC Database Consortium intermediate to poor risk disease. Median follow up 12.8 months (range 0.1-53 months). From RCC diagnosis to IO initiation, the cumulative incidence of TE was 6.1% (95%CI 3.3-9.1) at 6 months and 6.1% (95%CI 3.8-9.1 months) at 12 months. At the time of IO start, 12% of patients were on anticoagulation, not specifically for TE related events. IO was used as first line therapy in 43% of patients (median doses 8, range 1-81) and included single agent (54% nivolumab, 13% pembrolizumab, 0.6% durvalumab, 1% atezolizumab, 0.2% avelumab) or doublet (30% ipilimuamb and nivolumab). VTE occurred in 11% (n=37), of which 62% were DVT, 27% PE, or 11% both. DVT locations were 74% lower limb, 7% upper limb, 19% visceral vein. ATE occurred in 2% (n=6), 4 had cerebral vascular events and 2 had myocardial infarctions. There was no difference in TE rates in patients receiving single versus doublet IO (p= 0.7475) For all patients, the incidence of TE after initiation of IO was 4.4% (95%CI 2.6%-6.9%) at 6 months and 9.8% (95%CI 6.8%-13.4%) at 12 months. Of all TE events, 72% resulted in hospitalization (mean duration of hospitalization 4 days), 9% of which resulted in clot related mortality. TE related dose delay occurred in 21% of patients (mean delay of 13 days). There were no TE-related IO discontinuations. The impact of IMDC score and Khorana score with respect to OS were evaluated in separate multivariable models given overlap in variables for score calculation. TE (HR 3.23, 95% CI 2.07-5.03, p <0.0001) and poor IMDC score were associated with worse OS. In the second model, TE (HR 3.16, 95% CI 2.0-4.92, p <0.0001) and higher Khorana scores were associated with worse OS. (Figure 1) Conclusion: Patients treated with IO therapy had a high incidence of TE at 12 months from treatment initiation. TE is associated with risk of treatment delay, hospitalization and mortality. TE, IMDC poor risk, and Khorana score ≥ 2 are associated with poorer survival. Further investigations into IO-associated TE are needed to identify benefit from primary thromboprophylaxis. Disclosures Gupta: Astra Zeneca: Consultancy; BMS: Consultancy; Merck: Consultancy; Pfizer: Consultancy; Seattle Genetics: Consultancy; Exelixis: Other; Janssen: Other. Khorana:Leap: Research Funding; BMS: Honoraria, Research Funding; Merck: Research Funding; Array: Other: Research funding (to institution); Janssen: Honoraria; Bayer: Honoraria; Pfizer: Honoraria; Sanofi: Honoraria; Medscape: Honoraria; Leo Pharma: Honoraria; Seattle Genetics: Honoraria; Pharmacyte: Honoraria; Pharmacyclics: Honoraria.

BackgroundThromboembolism (TE) in cancer significantly contributes to morbidity and mortality. Little is known about the incidence of arterial TE (ATE) and venous TE (VTE) in patients with melanoma on immune checkpoint inhibitor (ICI) therapy.MethodsWe conducted a retrospective cohort study of patients with melanoma receiving ICI from July 2015 through December 2017 at the Cleveland Clinic. TE, including VTE events of deep venous thrombosis, pulmonary embolism, visceral vein thrombosis, and ATE events of myocardial infarction, stroke, peripheral arterial embolism, or transient ischemic attack after ICI initiation were identified. Overall survival (OS) from ICI initiation was estimated by Kaplan-Meier and Cox hazard models; associations between TE, ICI regimen, and clinical risk factors were evaluated using log-rank test.ResultsThe study population comprised 228 patients with median age of 65 years (23–91 years), 67% male, and median follow-up of 27.3 months. Pembrolizumab was most commonly used (38.7%), followed by combination of ipilimumab plus nivolumab (29.4%), ipilimumab (20%), and nivolumab (12.3%). Most had stage IV disease (81.1%) and 11% had brain metastases (BM) at treatment initiation. Fifty-one TE events occurred in 47 patients (20.6%), including 37 (16.2%) VTE and 14 (6.1%) ATE. Cumulative incidence of TE after ICI initiation was 9.3% (95% CI: 6.0% to 13.6%) at 6 months, and 16.0% (95% CI: 11.6% to 21.2%) at 12 months. The 6-month and 12-month VTE cumulative incidence rates were higher with combination ICI than single agent (16.7% vs 5.0% and 21.3% vs 9.5%, respectively; p=0.02). Risk factors significantly associated with VTE in multivariate analysis included combination ICI (HR 2.70; 95% CI: 1.28 to 5.70; p=0.009), Khorana Score ≥1 (HR 2.24; 95% CI: 1.06 to 4.74; p=0.03), history of coronary artery disease (HR 2.71; 95% CI: 1.16 to 6.29); p=0.02), and anticoagulation at treatment start (HR 4.14; 95% CI: 1.60 to 10.7; p=0.003). Of patients without BM, OS was worse in patients with TE compared with those without (2-year OS 50.8% vs 71.3%; HR 2.27; 95% CI: 1.36 to 3.79; p=0.002), when adjusted for age and stage.ConclusionsICI is associated with a high incidence of TE in patients with melanoma, with higher rates with combination therapy; TE is associated with substantial worsening of survival. Further studies are needed to identify pathophysiology, biomarkers, and preventive approaches.

5042 Background: Most pts with bladder cancer will be treated with immunotherapy. There is concern for increased TE risk with CPIs in this already high risk population. We present the first analysis of the incidence and outcomes of venous (VTE) and arterial (ATE) thromboembolism in pts with bladder cancer treated with CPIs. Methods: Consecutive pts with bladder cancer treated with CPIs at the Cleveland Clinic from 1/2015 to 12/2019 were identified and TE events noted. Overall survival (OS) was estimated using Kaplan-Meier method and the impact of VTE on OS was evaluated using Cox proportional hazards regression. Results: Of 274 pts, 72% were men (median age 73.3 years, 89% white), 82% had pure UC, 92% had lower tract disease, and 67% had a Bajorin score ≥1 (median KPS 90, 61% visceral metastases), 59% had prior systemic therapy (median 1, range 0-4) and 36% had prior TE (14% ATE, 19% VTE, 0.4% both). At CPI initiation, 24% were on antiplatelet therapy, and 15% on therapeutic anticoagulation. CPI (median doses 5, range 8.5-59) included: 40% atezolizumab, 3% nivolumab, 57% pembrolizumab. VTE occurred in 14% (n = 37), including 8% DVT, 4% PE, 2% both. DVT locations were 56% lower limb, 26% upper limb, 15% visceral vein, 4% visceral+upper limb. 2% (n = 5) had ATE (1% CVA, 0.4% visceral, 0.4% left subclavian). 92% of VTE and all ATE occurred within 6 months of CPI initiation. The incidence of TE was 10.9% (95%CI 6.6%—15.1%) at 6 months and 19.8% (95%CI 13.3%-26.4%) at 12 months. 82% of VTE (mean 6 days) and all ATE (mean 5 days) resulted in hospitalization. Multivariate analysis showed TE (HR 2.296, 95%CI 1.451-3.632, p = 0.0004), Bajorin score 1 (HR 1.490, 95%CI 1.036-2.142, p = 0.0315), and Bajorin score 2 (HR 3.50, 95%CI 2.14-5.74, p < 0.0001) were independently associated with worse OS. Conclusions: CPIs in bladder cancer pts are associated with a high TE risk, especially within six months of initiation. TE is associated with worsened survival. Further investigation into the risk factors for CPI-associated TE is needed to identify if benefits exist from thromboprophylaxis.

To evaluate factors affecting the utilization of immunotherapy and to stratify results based on the approval of ipilimumab in 2011 and PD-1 inhibitors in 2014, an analysis of available data from the National Cancer Database (NCDB) was performed. Stage IV melanoma patients were identified. Effects of immunotherapy on overall survival (OS) were assessed using Kaplan–Meier curves and Cox proportional hazards model. A total of 19 233 patients were analyzed and 1998 received immunotherapy. Between 2011 and 2013, and in 2014, 18.6 and 28.9% of patients received immunotherapy, respectively. Patients who received immunotherapy from 2011 to 2013 had a 33% (95% CI, 30–35%) 3-year OS compared to 23% (95% CI, 21–24%). In 2014, 3-year OS was 37% (95% CI, 32–43%) for those who received immunotherapy compared to 22% (95% CI, 18–26%) for those who did not (P < 0.0001). This is the first analysis of a large cancer database for melanoma patients with stratification based on utilization and availability of immunotherapy. Immunotherapy increased yearly and improved OS. With combination immunotherapy now more widely employed, it is expected these results will continue to improve. This is the first analysis of a large cancer database for melanoma patients with stratification based on utilization and availability of immunotherapy demonstrating that immunotherapy increased yearly and improved OS.