Start Of Immune Checkpoint Inhibitors Research Articles

Performance status, survival, and end-of-life care in adults with non-small cell lung cancer (NSCLC) treated with immunotherapy.

49 Background: Immune checkpoint inhibitors (ICI) improve survival for adults with metastatic NSCLC. Despite limited data on ICI efficacy in adults with poor performance status (PS), oncologists often elect to treat this population with ICI given their relatively favorable toxicity profile and potential for durable response. We aimed to describe the timing of ICI and patients’ survival based on PS, and explore the effect of late ICI use on end-of-life (EOL) care. Methods: Retrospective study of 235 adults with metastatic NSCLC at a single academic center who initiated ICI from 2015-2017. We compared overall survival (OS) among adults with Eastern Cooperative Oncology Group (ECOG) PS ≥ 2 at ICI start to those with ECOG PS < 2, using the log-rank test and Cox regression, adjusted for age, sex, comorbidity, time from diagnosis and line of therapy. We used logistic regression to analyze the association between ICI in the last 30 days of life and EOL care. Results: The median age at ICI start was 67 (range 37-91), and 83/235 (35%) had ECOG PS ≥ 2. Patients received ICI as first- (19%), second-line (56%) or later (25%) therapy. Median OS was 4.0 months in adults with ECOG PS ≥ 2 and 14.3 months in ECOG PS < 2 (p < 0.0001; HR = 2.5 [95% CI 1.8–3.5]). Among adults who died (n = 165), 17% of those with ECOG PS ≥ 2 started ICI in last 30 days of life and 24% started or continued ICI in their last 30 days, compared to 4% and 7% of ECOG PS < 2 (p = 0.005, p = 0.001, respectively). Receipt of ICI in last 30 days of life was associated with decreased hospice referral (OR 0.29, p = 0.006), decreased odds of hospice stay > 7 days (OR 0.15, p < 0.001), and increased in-hospital death (OR 6.8, p = 0.001). Conclusions: Adults with metastatic NSCLC and ECOG PS ≥ 2 experience significantly shorter survival than those with ECOG PS < 2 and more often receive ICI near the end of life, and late ICI use is associated with decreased hospice use and increased in-hospital death. Clinicians should thus use caution in extrapolating data from clinical trials, which are limited to ECOG PS < 2, to inform the care of adults with ECOG PS ≥ 2. Further, these results highlight potential tradeoffs of ICI and underscore the need for efforts to improve communication about ICI risks and benefits.

Association of gastric acid suppression with efficacy of immune checkpoint inhibitors (ICIs) in advanced cancer patients

Abstract Background Recent studies have shown that proton pump inhibitor (PPI) negatively impacts on the clinical benefit from ICIs. Gut microbiome modulated by PPI may contribute to decreased efficacy. It has been also reported that H2 blocker (H2B) or gastrectomy affected gut microbiome. We assessed in this study the association of GAS including PPI, H2B and gastrectomy with clinical outcome from ICI. Methods We retrospectively investigated advanced cancers pts treated with ICIs as the 1st- to the 3rd-line treatment at our institution between July 2014 and September 2018. GAS was defined as using PPI or H2B within 30 days before start of ICI, or having past history of gastrectomy. Response to ICIs was assessed by RECIST v1.1. Performance status (PS), body mass index, antibiotics, steroid, LDH level, albumin level, CPR and neutrophil-lymphocyte ratio (NLR) were compared between the GAS and non-GAS arms. Associations of nominal variables with the response were evaluated. Results Total of 115 pts (lung 69.9%, skin 11.3%, head and neck 7.8%, renal cell 6.9%, stomach 4.2%, and urothelium 2.5%) were evaluated. The types of ICIs were nivolumab of 72.2%, pembrolizumab of 22.6%, and atezolizumab of 5.2%. Sixty of 115 pts recieved GAS (PPI 83.3%, H2B 11.6%, and gastorectomy 5%) before start of ICI. Patients’ characteristics between the GAS and non-GAS arms was almost well balanced. However, pts with lower PS (>2) or high NLR were more frequently observed in the GAS arm (20% vs. 0%, 40% vs. 16%, respectively). The response rate (RR) was almost significantly lower in the GAS arm than the non-GAS arm (12% vs. 25%, P = 0.055). GAS did not remain statistically significant in a multivariate analysis (P = 0.091), while LDH was significant in both uni- and multivariate analyses. Conclusion Pts with GAS and pts without GAS have different patients’ background such as PS and NLR, and the RR in pts with GAS was numerically lower compared to that without GAS in advanced cancer treated with ICI.

P1.04-70 Application of Tumor Kinetics for Evaluation of Hyperprogression in Immune Checkpoint Inhibitor Treatment for Non-Small Cell Lung Cancer

Immune checkpoint inhibitors (ICIs) have suggested substantial options for treatment in advanced non-small cell lung cancer (NSCLC) without driver oncogenes. However, existing criteria for response assessment have limitations in differentiating categories of disease progression upon ICI treatment: primary resistance, pseudoprogression or hyperprogression (HPD). This study was aimed to investigate a real-world feasibility of pre- and on-treatment tumor kinetics for ICIs-specific response assessment in NSCLC. We retrospectively recruited stage III/IV NSCLC patients treated with anti-PD-1/PD-L1 monotherapy after failure of first-line platinum-doublet chemotherapy (n=91) between June 2016 and October 2018. Based on medical records from response-evaluable patients, we analyzed tumor kinetics prior and upon ICI treatment using ratio of tumor growth rate (TGR ratio, TGRR), tumor growth kinetics (TGK ratio, TGKR) and difference in TGR (delta TGR, ΔTGR), respectively. HPD was defined as TGRR≥2, TGKR≥2 and ΔTGR >50%. After excluding 18 patients without CT scan after start of ICIs, 73 patients were enrolled. Of response-evaluable patients, 47 patients (64%) stopped ICI treatment due to disease progression or death. Overall response rate by RECIST v1.1 was 16.4% (12/73) and disease control rate was 62% (45/73). HPD patients was 12% (9/73) and 15% (11/73) when defined by TGRR and TGKR, respectively. There was no HPD patients defined by ΔTGR. HPD patients classified by each criterion were associated with shorter median progression-free survival (PFS) according to iRECIST (TGRR ≥2, 1.6 vs 2.1 months, p<0.001; TGKR ≥2, 1.6 vs 2.1 months, p<0.001) and median overall survival (OS) than non-HPD-PD patients (TGRR ≥2, 2.4 vs 5.2 months, p=0.002; TGKR ≥2, 2.4 vs 5.2 months, p=0.002). HPD was not associated with age, sex, histology, stage, brain metastasis, nor PD-L1 expression at baseline. There was no pseudoprogression in patients classified as HPD. HPD defined by pre- and on-treatment tumor kinetics was observed in 12-15% of patients with advanced NSCLC treated with second-line ICI monotherapy, and correlated with shorter PFS and OS. TGR and TGK could classify HPD patients from non-HPD PD with similar discriminability. Further studies are needed to investigate clinical and molecular determinants that distinguish HPD from non-HPD PD and pseuoprogression.

976P - Efficacy of immune checkpoint inhibitors (ICI) and genomic alterations by body mass index (BMI) in advanced renal cell carcinoma (RCC)

BackgroundAn elevated BMI is associated with improved prognosis in certain solid tumors treated with ICI; however, real-world data and genomic analysis has been lacking. We investigated the effect of BMI on RCC patients treated with PD-1/PD-L1 based ICI and explored potential genomic alterations (GA). MethodsUsing the International Metastatic RCC Database Consortium (IMDC) database, we performed a retrospective analysis on patients treated with ICI alone or in combination with other therapies. The association of BMI with overall survival (OS), time to treatment failure (TTF), and objective response rate (ORR) was evaluated using Cox and logistic regressions respectively, adjusted for IMDC risk groups, histology, line and type of therapy, age, gender, and race. In an exploratory analysis of patients with clear cell RCC and available NGS panel data (275-447 genes), GA frequencies and tumor mutational burden (TMB) were compared by BMI status using Fisher’s exact and Mann-Whitney U tests. Results were considered statistically significant if p<0.05 or q<0.10. ResultsOf 1055 eligible patients, 735 had BMI data at start of ICI. Median follow up was 13.5 (<1-78.6) months (mos). Patients were mostly male (76%), had clear cell histology (85%), and were intermediate risk (60%). Overall, 31% received first-line ICI and 31% received combination ICI (19% with VEGF, 12% with CTLA-4/other therapies). At ICI initiation, 274 (37%) patients were considered low BMI (<25kg/m2) and 461 (63%) considered high BMI (≥25kg/m2). Patients with high BMI had better OS compared to those with low BMI, with 1-yr OS 79% vs 66% (adjusted HR=0.75, 0.57-0.97, p=0.03). ORR (30% vs 21%, adjusted p=0.06) and TTF (median 7.4 vs 4.9 mos, adjusted p=0.8) did not statistically differ. In a subset of 319 patients with clear cell RCC and available NGS data, GA and TMB (6.81 vs 6.81 mut/Mb, p=0.9) were found to be similar between those with high BMI compared to low BMI. ConclusionsHigh BMI is associated with improved OS in advanced RCC patients treated with ICI. In an exploratory analysis, there were no differences in genomic alterations on NGS by BMI status. Further correlative work is ongoing. Legal entity responsible for the studyThe authors. FundingHas not received any funding. DisclosureS.K. Pal: Advisory / Consultancy: Astellas; Advisory / Consultancy: Aveo; Advisory / Consultancy: BMS; Advisory / Consultancy: Eisai; Advisory / Consultancy: Exelixis; Advisory / Consultancy: Genentech; Advisory / Consultancy: Ipsen; Advisory / Consultancy: Myriad; Advisory / Consultancy: Novartis; Advisory / Consultancy: Pfizer; Research grant / Funding (institution): Medivation. A.R. Hansen: Advisory / Consultancy: AstraZeneca; Advisory / Consultancy, Research grant / Funding (institution): Boehringer Ingelheim; Advisory / Consultancy, Research grant / Funding (institution): BMS; Advisory / Consultancy, Research grant / Funding (institution): GSK; Advisory / Consultancy, Research grant / Funding (institution): Merck; Advisory / Consultancy: Pfizer. F. Donskov: Research grant / Funding (institution): Ipsen; Research grant / Funding (institution): Novartis; Research grant / Funding (institution): Pfizer. T. Yuasa: Advisory / Consultancy: BMS; Advisory / Consultancy: Novartis; Advisory / Consultancy: Ono; Advisory / Consultancy: Pfizer. U. Vaishampayan: Advisory / Consultancy, Research grant / Funding (institution): Astellas; Advisory / Consultancy: Bayer; Advisory / Consultancy, Research grant / Funding (institution): BMS; Advisory / Consultancy, Research grant / Funding (institution): Exelixis; Advisory / Consultancy: Genentech; Advisory / Consultancy, Research grant / Funding (institution): Pfizer; Advisory / Consultancy: Sanofi. N.S. Basappa: Advisory / Consultancy: Astellas; Advisory / Consultancy: AstraZeneca; Advisory / Consultancy: Bayer; Speaker Bureau / Expert testimony: BMS; Advisory / Consultancy: Eisai; Advisory / Consultancy: Ipsen; Advisory / Consultancy: Janssen; Advisory / Consultancy: Merck; Advisory / Consultancy: Pfizer; Advisory / Consultancy: Roche. D.Y.C. Heng: Advisory / Consultancy: Astellas; Advisory / Consultancy, Research grant / Funding (institution): BMS; Research grant / Funding (institution): Exelixis; Advisory / Consultancy, Research grant / Funding (institution): Novartis; Advisory / Consultancy, Research grant / Funding (institution): Pfizer; Advisory / Consultancy: Janssen. T.K. Choueiri: Advisory / Consultancy: Alexion; Advisory / Consultancy: Alligent; Advisory / Consultancy: Analysis Group; Research grant / Funding (institution): Agensys; Advisory / Consultancy, Research grant / Funding (institution): AstraZeneca; Advisory / Consultancy, Research grant / Funding (institution): Bayer; Advisory / Consultancy, Research grant / Funding (institution): BMS; Advisory / Consultancy, Research grant / Funding (institution): Cerulean; Advisory / Consultancy, Research grant / Funding (institution): Corvus; Advisory / Consultancy, Research grant / Funding (institution): Eisai; Advisory / Consultancy, Research grant / Funding (institution): Exelixis; Advisory / Consultancy, Research grant / Funding (institution): Foundation Medicine; Advisory / Consultancy, Research grant / Funding (institution): GSK; Advisory / Consultancy, Research grant / Funding (institution): Ipsen; Advisory / Consultancy, Research grant / Funding (institution): Merck; Advisory / Consultancy, Research grant / Funding (institution): Novartis; Advisory / Consultancy, Research grant / Funding (institution): Peloton; Advisory / Consultancy, Research grant / Funding (institution): Pfizer; Advisory / Consultancy, Research grant / Funding (institution): Prometheus; Advisory / Consultancy, Research grant / Funding (institution): Roche/Genentech. All other authors have declared no conflicts of interest.

THER-14. INFLUENCE OF CLINICAL PARAMETERS ON THE EFFICACY OF IMMUNE CHECKPOINT INHIBITORS IN PATIENTS WITH NON-SMALL CELL LUNG CANCER BRAIN METASTASES (NSCLCBM)

INTRODUCTION: Non-small cell lung cancer brain metastases (NSCLCBM) patients have a dismal prognosis. Immune checkpoint inhibitors (ICI) have resulted in improved outcomes in a subset of patients, although limited information exists on the impact of ICI in patients with NSCLCBM. METHODS: We reviewed 121 NSCLCBM (2012–2018) patients treated at our tertiary care center. All patients received at least 2 cycles of ICI therapy after diagnosis of NSCLCBM. Overall survival (OS) and progression-free survival (PFS) were calculated from the start of ICI therapy to date of death, progression or last follow up. Kaplan-Meier curves were used to estimate survival and were analyzed using the Wilcoxon test. RESULTS: Median age was 62 years (39–81) and median KPS was 90. Eighty-six patient received Nivolumab, 7 Atezolizumab, 25 Pembrolizumab, and 3 patients received multiple ICI over the course of their treatment for NSCLCBM. One hundred and twelve patients underwent stereotactic radiosurgery. Nine patients were treated with ICI alone and 25 patients underwent surgical resection. Median OS for the entire cohort was 558 (303–1159) days and median PFS was 220 (114–512) days. Twenty-four patients received oral steroids within the first 28 days of ICI (median prednisone equivalent dose of 27 mg). Patients on upfront steroid therapy had a median PFS of 148 days vs 301 days in patients not on upfront steroids (p-value .0095). Complete blood count at the start of ICI was available for 87 patients and neutrophil to lymphocyte ratios (NLR) were calculated. Patients with NLR at the start of ICI above 5 (n=33) had a median overall survival of 337 days compared to 558 days when NLR was below 5 (p-value .038). CONCLUSION: Use of steroids at initiation or within first 28 days of ICI therapy and NLR of greater than 5 are associated with worse outcomes in NSCLCBM treated with ICI.

Survival of patients with non-small cell lung cancer having leptomeningeal metastases treated with immune checkpoint inhibitors

IntroductionPatients with non-small cell lung cancer (NSCLC) experience leptomeningeal metastases (LM) in 3–9% of cases. Because overall survival (OS) and performance status are very poor, they are mostly excluded from clinical trials. Here, we evaluated survival of patients with NSCLC having LM treated with immune checkpoint inhibitors (ICIs). MethodsA prospectively collected list of patients with advanced NSCLC treated with ICIs between November 2012 and July 2018 in 7 European centres was merged. All patients with LM before ICI start were selected, data were retrospectively added and patients were classified according to the National Comprehensive Cancer Network (NCCN) LM prognostic classification (good/poor). Progression-free survival (PFS) and OS on ICIs were evaluated. ResultsNineteen of 1288 (1.5%) patients had LM; 73.7% had synchronous brain metastases; 73.7% had neurological symptoms at the start of ICIs and 52.6% were in the NCCN LM good prognosis group. Programmed death ligand-1 (PD-L1) expression was known for 42.1% of patients (87.5% positive). Median follow-up was 13 months from the start of ICIs, and median (95% confidence interval [CI]) PFS on ICIs was 2.0 (1.8–2.2) months. Six-month PFS rate was 21.0% and was significantly higher in the NCCN good versus poor prognostic group: 40% vs 0% (p = 0.05). Twelve-month PFS rate was 0%. Median (95% CI) OS from the start of ICIs was 3.7 (0.9–6.6) months. Six-month OS rate was 36.8%, and 12-month OS rate was 21.1%; both were not statistically significantly different for the good versus poor NCCN prognostic group (p = 0.40 and p = 0.56, respectively). ConclusionSome patients with NSCLC having LM do benefit from ICI treatment; specifically, those in the NCCN LM good prognosis group can obtain a long survival.

Efficacy and safety of immune checkpoint inhibitors in elderly patients (≥70 years) with squamous cell carcinoma of the head and neck.

6035 Background: Recent meta-analysis showed that immune checkpoint inhibitors (ICI) have comparable activity in younger vs older patients (pts) (≥65 years (y)). However little is known about efficacy and safety of ICI in elderly pts with relapsed/metastatic squamous cell carcinoma of the head and neck (R/M SCCHN). The aim of this study is to compare efficacy and grade ≥3 immune-related adverse events (irAEs) of ICI in pts ≥70 y with R/M SCCHN to younger pts. Methods: A retrospective study was conducted at 4 French hospitals. Eligibility criteria were pts treated with ICI for R/M SCCHN between September 2014 and December 2018. Clinical and radiological data and outcome were collected from review of medical records. Results: Two hundred twenty six pts were enrolled including 67 pts ≥ 70 y. Median age of elderly pts was 75y (range 70-87). Elderly pts received ICI as first-line treatment in 21% of pts vs 17% in younger pts. In elderly pts, 9% had ECOG of 0, 72% had ECOG of 1 and 15% had ECOG of 2 at ICI initiation vs 34%, 62% and 4% respectively in younger pts (p = 0.0006). In elderly pts, 22% had only loco-regional relapse at ICI initiation, 30% only distant recurrence and 49% had both vs 42%, 32% and 26% respectively (p = 0.0014). Elderly pts received ICI as monotherapy in 73% of pts vs 52% (p = 0.0027). The ORR in elderly pts was 23% vs 13% in younger pts (p = 0.071). After a median follow-up of 16.8 months (m) (range 10.7-23.7), median OS was 9.7m in elderly pts vs 8.7m in younger pts (p = 0.87). Median PFS was 2.7 m in elderly pts vs 1.9 m (p = 0.2). After adjustment for ECOG, type of evolution, number of ICI drugs, time between initial diagnosis and ICI start and number of previous lines, age ≥70 years was significantly associated with a better PFS (HR = 0.66 (95%CI = 0.47;0.93), p = 0.02) but was not significantly associated with OS (HR = 0.91 (95%CI = 0.61;1.34), p = 0.62). Grade ≥3 irAEs occurred in 15% of elderly pts vs 8% of younger pts (p = 0.13). Patients with grade ≥3 irAEs had a significantly higher ORR than pts without Grade ≥3 irAEs (36% vs 14%, p = 0.007). Conclusions: Elderly pts treated with ICI had significantly higher PFS but not OS after adjustment. Grade ≥3 irAEs were associated with significantly higher ORR to ICI in the whole population.

Clinical outcomes of EGFR+ NSCLC pts treated with immune checkpoint inhibitors (ICI).

9069 Background: ICIs have limited efficacy in EGFR+ NSCLC with ORR ~10% if PDL1 &gt; 25% in ATLANTIC, yet ICIs are often used in later lines of therapy as pts and providers feel there may be little risk. The impacts of ICI in this setting are poorly understood. We describe our institutional experience of ICI use in EGFR+ NSCLC. Methods: MGH pts with advanced EGFR+ NSCLC treated with ICI (any line) were retrospectively reviewed for demographics, PDL1, treatment duration and patient outcomes. Disease flare was defined as hospital/hospice admission due to progression or death (Chaft CCR 2011) within 30d of ICI. Results: 40 pts with EGFR+ NSCLC (22 del19, 11 L858R, 5 ins20, 2 other) received ICI between 7/2012-12/2018. 13 were on a clinical trial. 4 had SCLC transformation. Median # of prior therapies = 3 (range, 0-8). Of 16 with PDL1 quantified, 8 had PDL1 &gt; 25%. ICI regimens were: nivolumab (nivo; n = 16), pembrolizumab (pembro; 9), atezolizumab (atezo; 3), ipilimumab/nivo (7), carboplatin/pemetrexed (pem)/pembro (3), pem/pembro (1), paclitaxel/atezo (1). 18 pts stopped TKI ≤21d prior to ICI start. Median duration of treatment (DOT) was 25 days (range, 1-1482). DOT was &gt; 1 yr for 2 pts (5%), treated with 1st-line nivo/erlotinib (erlo) and 3rd-line nivo. All 8 pts with PDL1 &gt; 25% had DOT &lt; 2 mos. Disease flare within 30d of ICI occurred in 16/40 (40%) overall, 8/18 (44%) who stopped TKI ≤21d of ICI start, and 14/26 (54%) who received ICI in 4thline or later. 8 pts had concurrent TKIs (4 erlo/nivo, 2 erlo/pembro, 1 erlo/atezo, 1 osimertinib (osi)/nivo); 3/8 discontinued ICI for toxicity (all 3 treated first-line). 5 pts received osi immediately post-ICI. There was no pneumonitis on osi post-ICI; 1 pt developed gr3 LFTs and gr4 hypoNa. Conclusions: In this real world cohort of EGFR+ NSCLC, clinical benefit from ICI (assessed by DOT) was rare, including pts with high PDL1. 5% had durable benefit (both pts received ICI in earlier lines of therapy). A previously underappreciated negative outcome of ICI is that admission to hospital, hospice or death within 30d of ICI occur in up to 54% pts. This may be related to disease flare or hyperprogression and suggests that use of ICI in heavily pretreated EGFR+ NSCLC may negatively impact outcomes at end-of-life and should be used with caution.

Organ dysfunction (dys) and clinical outcomes in patients (pts) treated with immune checkpoint inhibitors (ICIs).

2569 Background: ICIs (anti-PD-L1/PD-1/CTLA-4) are approved in multiple cancers. The impact of organ dys on the pharmacokinetics of ICIs is known, but associated clinical outcomes are not well characterized. We compared real-world (rw) clinical outcomes in ICI-treated pts by liver and renal function. Methods: This retrospective study used longitudinal, patient-level data from community practices in the Flatiron Health electronic-health record (EHR)-derived database. We included pts diagnosed with advanced cancers (NSCLC, renal cell, melanoma, gastric/esophageal, or head and neck) on or after 1/1/2011, treated with an ICI with follow-up through 12/31/2018 and with baseline liver or renal function results in the EHR ≤30 days prior to ICI start. Organ function was stratified as normal, mild, moderate, or severe dys based on NCI CTCAE. We computed unadjusted median estimates for rw time to treatment discontinuation (rwTTD) for any reason and overall survival (OS) across baseline groups using the Kaplan-Meier method. Results: Of 15,979 pts, we identified 12,978/12,840 pts with evaluable renal/liver function, respectively; median follow-up was 5.1 mos and median age was 69.0 yrs (IQR: 61.0, 76.0) for both. Most pts had NSCLC (69.4/69.0%), were men (60.1/60.0%), white (73.5/73.6%), and diagnosed at stage IV (58.7%/58.6%). Most ICI was given in 1st-line (42.3/42.1%) (outcomes in Table). Conclusions: Pts with categorically worse baseline liver function had progressively worse on-treatment outcomes, including shorter OS, which differed from trends in renal dys. Whether baseline dys is prognostic or predictive of ICI outcomes should be further investigated in addition to reasons for discontinuation. Clinical outcomes (unadjusted median times, mos [95% CI]) by organ function. [Table: see text]

Association between age, risk of severe (Grade 3-4) immune-related adverse events (sirAE), and mortality in patients receiving immune checkpoint inhibitors (ICI).

6597 Background: ICI are used in the treatment of advanced malignancies with on-target adverse events of non-tumor inflammation. Many studies have shown conflicting safety results with regard to older vs younger adults where 65 - 70 years of age has been used as the discrete cut-off variable. We sought to investigate the incidence and the association between age and sirAE using age as a continuous variable at a large tertiary cancer center. Methods: Under IRB approval, our ICI outcomes database was queried for those who were hospitalized and received an immunosuppressant. Charts were individually reviewed to identify hospital admissions due to a sirAE (a grade 3 or 4 AE per CTCAE v4.0) and all-cause mortality at 12 months post ICI start. Non-linear analyses using Cox regression models with penalized smoothed splines were performed to explore association between age and sirAE. Results: There were 6.3% of 1043 patients who had a sirAE, and a total of 83 sirAE events. Mean age was 64± 13 years.ICI included anti-PD-1 (77.8%), anti-CTLA-4 (18.1%), and anti PD-L1 (4.1%). Pts with sirAE had a thirty day, and one year mortality of 12% and 49%, respectively. These events included: colitis (30.1%), pneumonitis (16.9%), hepatitis (10.8%), hypophysitis (6%), and thyroiditis (6%). There were 5 neurologic and 4 myocarditis/myositis sirAEs. The 1-year cumulative incidences of sirAE and all-cause mortality were 8.4%, and 48%, respectively. Spline analysis showed a U-shaped association between age and hazard of sirAE (P = 0.03). Every 10 years above age 60 was associated with increased sirAE (HR 1.65 [1.14-2.40], P = 0.008), while every 10 years below 60 was associated with increased sirAE (HR 1.50 [0.99-2.27], P = 0.054). However, age and mortality showed a linear association (P = 0.003). Conclusions: We observed a curvilinear U-shaped association between age and the risk of sirAE, with minimum risk at age of 60, compared with a linear relationship between age and mortality. Further studies are needed to understand this relationship and its impact on outcomes, clinical care, and underlying host immune context.

Dynamic changes of neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and lactate dehydrogenase (LDH) during treatment with immune checkpoint inhibitors (ICIs) in non-small cell lung cancer (NSCLC).

2596 Background: ICIs have revolutionized the therapeutic landscape of NSCLC. However, with the exception of PD-L1 expression, predictive biomarkers are lacking. The aim of this study was to evaluate the dynamic changes of some markers of inflammation over time and the outcome of NSCLC patients (pts) treated with nivolumab (N) or pembrolizumab (P). Methods: All consecutive NSCLC pts treated with N or P between Aug. 2015-Dec. 2018 were analyzed. Laboratory results were collected at baseline, 6 weeks (6-wk), and 12 weeks (12-wk) and correlated with the outcome. NLR and PLR were defined as absolute neutrophil and platelet count divided by lymphocyte count, respectively. NLR ≥5, PLR ≥200, and LDH levels ≥ upper normal limit were considered high. Overall survival (OS) was defined as time from ICI start to death and Progression Free Survival (PFS) as time from treatment start to progression disease or death for any cause. OS and PFS curves were estimated using the Kaplan–Meier method and compared with the log-rank test. Results: We included 71 consecutive NSCLC pts treated with either N (75%) or P (25%). Baseline characteristics: median age 69 years (range 46-80), sex male 76%, squamous histology in 39%. PD-L1 expression (39/71): &lt; 1% in 20%, 1-49% in 45%, and ≥50% in 35%. NLR ≥5 was associated with lower PFS and OS, with an increased predictive value over time ( p =0.01 and p =0.009 at baseline; p =0.007 and p &lt;0.001 at 6-wk; p &lt;0.001 and p &lt;0.001 at 12-wk, respectively). PLR ≥200 at baseline and 12-wk was significantly associated with shorter OS ( p =0.05 and p =0.004, respectively), but no in terms of PFS at all the three time points. Finally, LDH ≥UNL at baseline was associated with shorter PFS and OS ( p =0.02 and p =0.03), as well as a reduction of LDH levels at 12-wk compared with baseline values ( p =0.006 and p =0.004). Conclusions: Baseline evaluation of NLR, PLR and LDH levels is significantly associated with outcome in NSCLC treated with single agent ICIs. Moreover, dynamic changes of LDH levels at 12 weeks significantly predicted outcome. These easy to determine parameters may have a place in the selection process of pts candidate for immunotherapy.

Effect of immunotherapy (IO) on primary renal tumor in metastatic renal cell cancer (mRCC).

649 Background: Recent data showed that cytoreductive nephrectomy (CN) should not be performed prior to systemic therapy in intermediate and poor risk groups. Immune checkpoint inhibitors (ICI) are now approved in first and second line in mRCC, but ICI effect on the primary tumor is unknown. Methods: All patients (pts) treated with ICI for mRCC without prior CN were retrospectively collected and sequential computed tomography (CT) scan evaluated for response. Primary objective was Objective Response Rate (ORR) in the primary renal tumor in the primary RECIST 1.1. Secondary objective included systemic ORR and nephrometry score changes. Results: A total of 20 pts were treated with ICI with primary tumor in situ between 12/2012 and 10/2018. Among these, CT scans were available for evaluation of response for 19 primary tumors in 18 pts. Median age was 59.5 years. Pathology was 16 clear cell RCC, 2 translocation RCC, 1 Collecting duct carcinoma. Baseline median primary renal tumor size was 7.3 cm. ICI was used in first line setting in 7 pts, in second line or beyond in 13 pts. IMDC risk group at ICI start was favourable in 2 pts, intermediate in 9 pts and poor in 9 pts. ICI used was NIVOLUMAB(N) single agent in 13 pts or in combination in 7 pts (6 N-IPILIMUMAB and 1 N-VEGFRTKI). After a median follow up of 6.5 month from ICI start and a median duration of immunotherapy of 3.0 month, ORR on the primary tumor, was 3/19 (16%) (1 CR and 2 PR), 14 tumors (74%) were stable, and 2 had PD as best response. Median change in primary renal tumor was -1%. ICI use did not change nephrometry score. At extra renal assessment, ORR (systemic) was 5/18 (28%), (2 CR, 3 PR), 5 (28%) SD, and 8 (44%) PD (including 1 with new lesion and 1 with non-target lesion PD). Interestingly, 4 pts/18 (22%) had mixed response with opposite response between primary and systemic response. Furthermore, 11/18 (61%) would not classified in same RECIST group (CR, PR, SD, PD) between primary and metastatic sites assessment. Conclusions: To our knowledge, we report the first cohort of patients treated with ICI investigating the response on the primary tumor. ICI can induce a significant reduction in primary renal tumor, especially complete response and therefore raise the question of delayed CN in responders under ICI.

Influenza vaccination and myocarditis among patients receiving immune checkpoint inhibitors

BackgroundInfluenza vaccination (FV) is recommended for patients with cancer. Recent data suggested that the administration of the FV was associated with an increase in immune-related adverse events (irAEs) among patients on immune checkpoint inhibitors (ICIs). Myocarditis is an uncommon but serious complication of ICIs and may also result from infection with influenza. There are no data testing the relationship between FV and the development of myocarditis on ICIs.MethodsPatients on ICIs who developed myocarditis (n = 101) (cases) were compared to ICI-treated patients (n = 201) without myocarditis (controls). A patient was defined as having the FV if they were administered the FV from 6 months prior to start of ICI to anytime during ICI therapy. Alternate thresholds for FV status were also tested. The primary comparison of interest was the rate of FV between cases and controls. Patients with myocarditis were followed for major adverse cardiac events (MACE), defined as the composite of cardiogenic shock, cardiac arrest, hemodynamically significant complete heart block and cardiovascular death.ResultsThe FV was administered to 25% of the myocarditis cases compared to 40% of the non-myocarditis ICI-treated controls (p = 0.01). Similar findings of lower rates of FV administration were noted among myocarditis cases when alternate thresholds were tested. Among the myocarditis cases, those who were vaccinated had 3-fold lower troponin levels when compared to unvaccinated cases (FV vs. No FV: 0.12 [0.02, 0.47] vs. 0.40 [0.11, 1.26] ng/ml, p = 0.02). Within myocarditis cases, those administered the FV also had a lower rate of other irAEs when compared to unvaccinated cases (36 vs. 55% p = 0.10) including lower rates of pneumonitis (12 vs. 36%, p = 0.03). During follow-up (175 [IQR 89, 363] days), 47% of myocarditis cases experienced a MACE. Myocarditis cases who received the FV were at a lower risk of cumulative MACE when compared to unvaccinated cases (24 vs. 59%, p = 0.002).ConclusionThe rate of FV among ICI-related myocarditis cases was lower than controls on ICIs who did not develop myocarditis. In those who developed myocarditis related to an ICI, there was less myocardial injury and a lower risk of MACE among those who were administered the FV.

Impact of Palliative Radiation on Lymphocyte Count and Neutrophil-to-Lymphocyte Ratio in Patients Receiving PD-1 Inhibitors

Immune checkpoint inhibitors (ICI) are increasingly used in patients receiving palliative radiation (RT). However, studies suggest that RT, and particularly protracted RT courses, may decrease circulating lymphocytes, which could adversely impact ICI efficacy. We therefore evaluated circulating lymphocytes in a contemporary cohort of melanoma, non-small cell lung cancer and renal cell carcinoma patients treated with ICI and radiation. We reviewed consecutive patients treated with palliative RT and ICI. We recorded absolute lymphocyte count (ALC), neutrophil count, and neutrophil to lymphocyte ratio (NTL) before and after palliative RT, and start of ICI. Patients who received other intervening systemic therapy were excluded. Survival was analyzed using the Kaplan-Meier method and log-rank tests. Chi-squared and non-parametric tests evaluated differences between groups. We identified 110 patients treated with PD-1 inhibitors; 38 also received CTLA-4 inhibitors (23 sequentially, 15 concurrently). The most common site irradiated was the brain (71 patients, 120 courses, 40% SRS). Among all patients, median change in ALC (dALC) and NTL (dNTL) after RT was -161 and 0.8, respectively. Changes were more pronounced after extracranial RT (dALC -384 vs. -48 for brain-directed RT p=0.01, dNTL 0.2 vs. 2.5, p=0.001), RT to the spine, lung, or chest wall vs. brain, extremity, or abdomen/pelvis (dALC -438, -384 or -470 vs. 252, 204, or -7, p=0.002, dNTL 11.6, 15.8 or 10.8 vs. 0.2, 1.2, or 3.4, p=0.05) or RT courses >5 fractions (dALC -328 vs. -11 p=0.003, dNTL 1.6 vs. 0.05, p=0.02). The association of dALC with >5 fractions RT remained significant after excluding brain-directed RT (dALC -443 vs. -26, p=0.01, dNTL 2.7 vs. 0.03, p=0.09), but not in brain-directed RT patients (dALC -128 vs 11, p=0.5, dNTL 0.5 vs 0.1, p=0.97). There was no association between timing of RT in relation to ICI on dALC and dNTL. Median survival (MS) was 319 days from PD-1 therapy onset. ALC<500 and NTL>4 at PD-1 therapy onset was associated with decreased MS (359 vs. 100 days, p<.0001, and 472 vs 250 days, p=0.01, respectively). In a MVA correcting for performance status and serum albumin, this association of ALC with survival persisted (HR 2.3, p=0.01). RT before PD-1 therapy onset was associated with ALC<500 or NTL>4 at onset when RT was extracranial (OR 3.9, p=0.0008 and 2.4, p=0.02) or >5 fractions (OR 2.9, p=0.01 and 2.3, p=0.03). There was no such association with RT performed after initiation of PD-1 therapy. Among RT patients treated <30d prior to PD-1 therapy, dALC after RT was associated with decreased MS (p=0.009). In a contemporary cohort of patients treated with palliative RT and PD-1 inhibitors, we confirm that RT can decrease ALC and increase NTL, particularly extracranial RT and more protracted regimens. These parameters are associated with worse outcomes following treatment with ICI, suggesting the need for further optimization of combined RT/ICI regimens.

A data driven approach to immunotherapy toxicity management.

326 Background: The immune checkpoint inhibitors (ICIs) confer a risk of unique inflammatory immune-related adverse events (irAEs), which are highly distinct from the adverse events historically observed with cytotoxic therapy. To develop a strategy for easier identification and mitigation of irAEs, we sought to understand the frequency of ED visits and hospitalizations in 90 days following ICI start by implementing a rapid learning system (RLS). Methods: We convened an Immunotherapy Toxicities Management Committee with representatives from the Center for Immuno-Oncology, Quality and Patient Safety, and Informatics to draft a series of recommendations for the development of an irAE rapid learning system. The Committee requested an audit of all irAEs between June 2015 and April 2018 by drug, event type (clinical diagnosis), outcomes (including ED visits, hospitalization with length of stay, and death), and time-frame (days since beginning ICI course). An automated pipeline was created to merge structured data from the Electronic Health Record (Epic) and billing system (EPSi). This data was used to design a tool which consisted of an automated dashboard to monitor patient and enable interventions. Results: Over the course of 3 years, a total of 2,020 unique patients receiving ICIs were seen. 918 were treated with Pembrolizumab (45.4%), 768 with Nivolumab (38.0%), 234 with Atezolizumab (11.6%), 111 with Nivolumab &amp; Ipilmumab (5.6%), 68 with Ipilimumab (3.4%), 9 with Durvalumab (0.4%), and 9 with Avelumab (0.4%). ED visits and hospitalization rates over 90 days were similar among the three most prescribed therapies, ranging from 332 unique patient events in the Pembrolizumab Cohort (36.1%) to 96 unique patient events in the Atezolizumab cohort (41.0%). Conclusions: The dashboard is effective tool to build a RLS for irAEs. The immediate output of this tool is using natural language processing (NLP) to distinguish between irAEs related ED visits and hospitalizations or regular disease progression, and measure the impact of interventions including (a) developing standardized algorithms for monitoring for irAEs, (b) designing an educational program for providers, and (c) developing an inpatient and outpatient immunotherapy toxicity management service.

P1.15-03 Clinical Characteristics of Long-Term Survivors With Nivolumab in Pretreated Advanced NSCLC from Real-World Data (RWD)

We have previously analyzed real-world data (RWD) from patients with metastatic NSCLC that progressed to chemotherapy and were subsequently treated with immune-checkpoint inhibitors (ICI). This included patients with performance status of 2, brain metastases, concomitant use of steroids, unknown PD-L1 status, systemic antibiotics during the previous month, and without restriction in the number of previous lines received. Now, with a longer follow-up, we aimed to analyze the baseline clinical characteristics in long-term survivor patients. We performed a retrospective study of previously treated advanced NSCLC patients that received nivolumab at 3 mg/kg every 2 weeks outside clinical trials from our institution in Madrid (Spain) between January 2015 and April 2018. We used RWD to analyze the clinical characteristics of patients who were alive 2 years after the start of ICI. A total of 38 pts fulfilled inclusion criteria. 7 pts (18%) where alive 2 years after the first dose of nivolumab. Median age at diagnosis was 63 years (53-72 years). 6 pts (86%) had history of tobacco smoking, and 6 pts (86%) had non-squamous histology. 3 pt had ECOG 0 (43%), 3 pts had ECOG 1 (43%) and one pt had ECOG 2 (14%). Median number of previous lines was 3; 3 pts received nivolumab in second line, 3 pts in fourth line and one pt in fifth line. All pts were diagnosed with primary advanced disease. 2 of the pts (29%) had brain metastases and 1 pt (14 %) had ≥ 4 metastatic locations. 2 pts had use antibiotics in the month before and 1 pt used steroids. Best response to nivolumab per RECIST 1.1 criteria included 3 partial responses (43%), and 2 each (29%) had stable and progressive disease. Only 2 pts had no evidence of disease progression during > 2 years at last follow-up when the database was lock (April 2018). Five pts discontinued treatment due to disease progression. After progression, 1 pt continued treatment with nivolumab, 1 pt started osimertinib (EGFR mutant previously treated with TKI and chemotherapy, and confirmed T790M+), and 3 pts started chemotherapy followed by retreatment with ICI. No patients discontinued due to adverse events. In our RWD experience, nivolumab resulted in a 2-years survival rate of 18% in previously treated advanced NSCLC patients. Clinical characteristics from real world patients do not predict for long term benefit of nivolumab. Immunological biomarkers are necessary to better select pts who will derive long-term benefit from immunotherapy.

Immune-Related Thyroiditis with Immune Checkpoint Inhibitors.

Although immune-related thyroiditis (irT) with immune checkpoint inhibitors (ICI) is a common consequence, its natural course and management recommendations are not well characterized in existing guidelines. This study sought to investigate the evolution of irT and describe its course and sequelae. This was a retrospective study of cancer patients treated with ICI between November 2014 and July 2016 at MD Anderson Cancer Center and referred for endocrinology evaluation for suspected irT. Patients included had normal baseline thyroid function tests prior to starting ICI and developed thyrotoxicosis due to irT. Of 657 patients treated with ICI during the study period, 43(6.5%) met the inclusion criteria. ICI included: ipilimumab + nivolumab (40%), nivolumab (33%), pembrolizumab (21%), and other (7%). Cancer diagnoses observed were melanoma (23%), renal-cell carcinoma (21%), lung cancer (19%), bladder cancer (12%), colon cancer (9%), and other cancers (15%). Median time from ICI start to thyrotoxicosis was 5.3 weeks (range 0.6-19.6 weeks). Clinically, patients presented with painless thyroiditis, and 67% were asymptomatic during the thyrotoxicosis phase. Thyrotoxicosis lasted a median of six weeks (range 2.6-39.7 weeks). Hypothyroidism developed in 37 (84%) patients at a median of 10.4 weeks (range 3.4-48.7 weeks) after starting ICI. These patients remained on levothyroxine and ICI at a median follow-up of 57.4 weeks (range 1-156.7 weeks) from hypothyroidism onset. Four patients recovered without initiating levothyroxine and remained euthyroid at a median follow-up of 11.35 months (range 4.43-14.43 months). Subgroup analysis of ipilimumab + nivolumab versus nivolumab alone showed a median time to thyrotoxicosis of two weeks [confidence interval (CI) 3.5-8.4] versus six weeks ([CI 1.2-2.8]; p = 0.26) and time to hypothyroidism of 10 weeks [CI 8.1-11.9] versus 17 weeks ([CI 8.8-25.2]; p = 0.029) after starting ICI. Thyroid peroxidase and thyroglobulin antibodies were present in 45% and 33% at the time of irT diagnosis. IrT manifests as an early onset of thyrotoxicosis, which is largely asymptomatic, followed by rapid transition to hypothyroidism requiring long-term levothyroxine substitution. The evolution of irT is more rapid with combination ICI. Frequent monitoring of thyroid function tests during ICI is warranted. Future guidelines need to recognize this entity and incorporate their management.

The Impact of Radiation Therapy on Lymphocyte Count and Survival in Metastatic Cancer Patients Receiving PD-1 Immune Checkpoint Inhibitors

Therapeutic radiation has conflicting immune effects: radiation therapy (RT)-induced immunogenic cell death can contribute to immune response, but lymphocytes are also sensitive to RT. It is unknown whether palliative RT leads to lymphopenia in patients treated with immune checkpoint inhibitors (ICIs) and whether this affects outcomes. As such, we sought to assess the impact of palliative RT on circulating lymphocyte count and neutrophil-to-lymphocyte ratio in patients being treated with PD-1-directed ICI and associations with survival. We identified patients from 5 radiation oncology centers, treated with palliative RT and either pembrolizumab or nivolumab with non-small cell lung cancer, metastatic melanoma, and renal cell carcinoma. Patients who received intervening cytotoxic chemotherapy were excluded. We recorded absolute lymphocyte count (ALC) and neutrophil-to-lymphocyte ratio before and after palliative RT and at the start of ICI. Survival was analyzed using the Kaplan-Meier method and Cox proportional hazard models. One hundred ten patients received 225 courses of palliative RT. Median change in ALC after RT was -161cells/mL. Decreases in ALC were greater with RT to the spine, lung/mediastinum, and chest wall compared with the brain, extremity, or abdomen/pelvis (P=.002) and after courses >5 fractions (P=.003). Extracranial and >5-fraction RT was associated with increased odds of severe lymphopenia (ALC <500) at the end of RT (odds ratio [OR], 3.7; P=.001; and OR, 3.9; P=.001, respectively). Patients who developed RT-induced severe lymphopenia were more likely to have severe lymphopenia when ICI was initiated (OR, 6.4; P=.0001), particularly when RT was administered in the previous 3months (OR, 189; P<.0001). Severe lymphopenia at onset of ICI therapy was associated with increased mortality on multivariable analysis (hazard ratio, 2.1; P=.03). Extracranial or prolonged courses of RT increase the risk of severe lymphopenia, which is associated with poorer survival in patients treated with ICI.

P.2.d.033 - Social cognition and exposure to trauma in individuals with bipolar disorder

Recent meta-analysis showed that immune checkpoint inhibitors (ICIs) have comparable activity between younger and older patients. However, little is known about efficacy and safety of ICI in elderly patients with relapsed/metastatic squamous cell carcinoma of the head and neck (R/M SCCHN). The aim of this study is to compare the efficacy of ICI for patients aged ≥70 y to that for younger patients, while taking into account potential confounding factors.A retrospective study was conducted at four hospitals in France. Patients treated with ICI for R/M SCCHN between September 2014 and December 2018 were eligible. Patients’ charts were reviewed for clinical and radiological data as well as oncologic outcomes.We included 226 patients, of whom 67 were aged ≥70 years. Objective response rate (ORR), median overall survival (OS) and median progression-free survival (PFS) were 23%, 9.7 months and 2.7 months, respectively, for elderly patients, compared to 13%, 8.7 months and 1.9 months for younger patients (respective p-values: 0.071, 0.87 and 0.21). After adjustment for performance status, site of progression, number of ICI drugs, time between initial diagnosis and ICI start and number of previous lines, age ≥70 years was significantly associated with a better PFS (hazard ratio [HR], 0.66; p = 0.021) but not OS (HR, 0.91; p = 0.59). Grade 3-5 adverse events (AEs) occurred in 15% of patients aged ≥70 years and in 8% of younger patients (p = 0.13).Patients aged ≥70 years with R/M SCCHN may respond to ICI similarly as younger patients in terms of ORR, OS and PFS, while maintaining comparable rate of AEs.

Importance of hypovolaemic shock and endoscopic signs in predicting recurrent haemorrhage from peptic ulceration: a prospective evaluation.

The incidence of rebleeding was studied prospectively in 177 patients with acute gastrointestinal haemorrhage from peptic ulceration with reference to the degree of haemodynamic insult on admission and the presence of endoscopic signs of recent haemorrhage on initial endoscopy. Rebleeding occurred in two (2%) of 114 patients without shock, in seven (18%) of 38 with tachycardia (pulse rate greater than 100 beats/min, systolic blood pressure greater than 100 mm Hg), and in 12 (48%) of 25 with shock (systolic blood pressure less than 100 mm Hg). A similar gradient was noted with the presence of endoscopic signs alone. Rebleeding occurred in four (5%) of 79 patients with no endoscopic signs, in none of 40 with black spot only, in 11 (23%) of 48 with a clot, and in five (50%) of 10 with a visible vessel on endoscopy. When the incidence of rebleeding was assessed in shocked patients, however, it was significantly higher in those with important signs (clot or visible vessel), in 11 (79%) out of 14 patients, than in those with no signs, in one (9%) out of 11 (p less than 0.001). These data suggest that the association of shock with important endoscopic signs is a stronger predictor of rebleeding than either shock or important signs alone. More aggressive treatment may be warranted in this small group of patients.