Anti-PD-1 monotherapy has shown a significant survival benefit for patients with advanced-stage melanoma. In this study, we explored whether long-term clinical benefit is determined by abundance or changes in such abundances of CD4 and CD8 T-cell subsets in blood. Peripheral blood mononuclear cells from 58 patients with advanced-stage melanoma treated with anti-PD-1 monotherapy were collected prior to first, second, and third administration. Overall survival (OS) was categorized into 3 groups: <1 yr (short term); 1-3 yr; and >3 yr (long term). The abundance of different T-cell subsets was determined via multiplex flow cytometry and transcriptomics analysis. Long- and short-term survivors demonstrated no differences in frequencies of CD4 and CD8 T-cell subsets at baseline. However, on-treatment, long-term survivors showed an early increase in the frequency of CD4 T cells expressing 4-1BB, whereas this subset frequency was decreased in short-term survivors. Notably, this early increase in the 4-1BB+ CD4 T-cell frequency was associated with an increase in the frequency of LAG3+TIM3+ CD4 T cells and followed by a later decrease in the frequency of FOXP3+ CD4 T cells. Notably, 4-1BB+ CD4 T cells of long-term survivors demonstrated a distinctively high expression of IL-17. In short, long-term survivors have unique on-treatment changes in the frequency of different CD4 T-cell subsets, particularly those related to co-signaling and regulatory T cells. These results show that CD4 T cell-mediated stimulation and differentiation are associated with long-term benefit of anti-PD-1 treatment, which may guide development of future treatment strategies.

Extended interval dosing regimens of immune checkpoint inhibitors have been implemented widely. However, their approval was mainly based on pharmacokinetic modeling and simulations. Consequently, comparative safety data of extended interval dosing regimens in a real-world setting are limited. This study compares grade ≥3 immune-related adverse events between standard and extended interval dosing of nivolumab in patients with melanoma and explores associated risk factors. This retrospective cohort study included patients with melanoma treated with nivolumab monotherapy from April 2016-September 2021 in the MULTOMAB study. Data on baseline characteristics and immune-related adverse events were collected from patients' electronic medical records with a maximum follow-up of 6 months. A total of 236 patients (125 metastatic and 111 adjuvant) were included, with 146 in the standard cohort and 90 in the extended interval cohort. Grade ≥3 immune-related adverse events occurred in 20 patients: 13 (8.9%) patients in the standard cohort and 7 (7.8%) in the extended interval cohort. No significant differences in grade ≥3 immune-related adverse events were observed between the two cohorts (P = 0.763). This study suggests that the risk of severe immune-related adverse events is comparable between both interval dosing regimens of nivolumab in patients with melanoma. Therefore, the extended interval dosing of nivolumab appears to be safe for application in standard care.

The development of immune checkpoint inhibitors (ICIs) has a tremendous effect on the treatment options for multiple types of cancer. Nonetheless, there is a large interpatient variability in response, survival, and the development of immune-related adverse events (irAEs). Pharmacogenetics is the general term for germline genetic variations, which may cause the observed interindividual differences in response or toxicity to treatment. These genetic variations can either be single-nucleotide polymorphisms (SNPs) or structural variants, such as gene deletions, amplifications or rearrangements. For ICIs, pharmacogenetic variation in the human leukocyte antigen molecules has also been studied with regard to treatment outcome. This review presents a summary of the literature regarding the pharmacogenetics of ICI treatment, discusses the most important known genetic variations and offers recommendations on the application of pharmacogenetics for ICI treatment.

Aim: With increasing use of immune checkpoint inhibitors (ICIs) more patients will develop severe and potentially life-threatening immune-related adverse events (irAEs). So far, predictive models for the occurrence of grade >= 3 irAEs are lacking. Therefore, we analysed associations between patient and disease characteristics, and the occurrence of grade >= 3 irAEs. Methods: Patients with cancer who were treated with anti-PD-1 (+/-anti-CTLA-4) between July 2015 and February 2020, and who were prospectively included in the MULTOMAB-trial, were eligible for this cohort study. Time to and occurrence of grade >= 3 irAEs according to CTCAE v5.0 were retrospectively registered. The associations between patient and disease characteristics and irAE occurrence were analysed using the competing risk cox-regression model of Fine and Gray. Analyses were performed separately in patients treated with monotherapy (anti-PD-1) and combination therapy (anti-PD-1 + anti-CTLA-4). Subgroup analyses were performed in tumour types with the highest number of patients; melanoma and NSCLC. Results: Out of 641 patients, 106 patients (17%) experienced grade >= 3 irAEs. None of the analysed factors were associated with grade > 3 irAE occurrence in the monotherapy (n = 550) or the combination therapy (n = 91) groups, nor in the subgroup analyses. Of interest, none of the patients with NSCLC with a WHO performance status of 0 (n = 34) experienced grade >= 3 irAEs. Most common NSCLC histology types were adenocarcinoma (n = 99/55%) and squamous cell carcinoma (n = 39/22%). Concluding statement: This study shows that patient and disease characteristics are not able to predict the occurrence of serious AEs in patients treated with ICIs. (c) 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

INTRODUCTION:Immune checkpoint inhibitor (ICI) associated diabetes is a harmful adverse event (AE) in patients with cancer following anti-programmed (cell) death protein-1 (PD-1) treatment. There are no available biomarkers able to predict this AE. The primary aim of this study was to investigate C-peptide levels as potential predictor for the occurrence of ICI-related diabetes. The secondary aim was to describe the presence of islet autoantibodies and course of pancreatic enzymes in patients with and without ICI-related diabetes.METHODS:From a total of 1318 patients with cancer who started anti-PD-1 treatment 8 cases and 16 controls were studied in this nested case-control study. C-peptide levels, islet autoantibodies, and pancreatic enzymes were measured in prospectively collected blood serum.RESULTS:In cases versus controls, median C-peptide levels were comparable at baseline and before toxicity or at the corresponding time point in controls. No patient had C-peptide levels below reference range before toxicity onset. Two out of eight patients in the ICI-related diabetes group had positive islet autoantibodies, whereas one out of 16 patients in the control group had positive islet autoantibodies. Pancreatic enzymes were elevated before diabetes onset in one patient (13%) and in one control (6%) at the corresponding time point.CONCLUSIONS:In patients developing ICI-related diabetes, changes in C-peptide levels, islet autoantibody positivity, and pancreatic enzymes before ICI-related diabetes onset seem comparable to patients without ICI-related diabetes. (NTR: NL6828).

Background Dosing schemes of pembrolizumab (anti-programmed cell death protein 1 monoclonal antibody) are solely based on pharmacokinetic (PK) modelling derived from phase I–III trials. The current study aimed to determine factors affecting PK and its relationship with clinical outcome in the real-world setting.Methods Advanced-stage cancer patients, who were treated with pembrolizumab monotherapy (2 mg/kg Q3W or 200 mg flat Q3W), were prospectively included for serial sampling to obtain trough concentrations. A PK model was generated, covariate effects assessed and internally validated by a bootstrap procedure. PK parameters were related to overall survival (OS) and the occurrence of immune-related adverse events (irAEs).Results 588 serum samples derived from 122 patients with (non-)small-cell lung cancer ([N]SCLC), malignant pleural mesothelioma (MPM), melanoma and urothelial cell cancer (UCC) were analyzed. Median follow-up was 2.2 years. A one-compartment PK model was generated: body surface area (BSA) and serum albumin had a significant effect on drug clearance (CL; covariate estimate 1.46 and −1.43, respectively), and serum lactate dehydrogenase (LDH) on the distribution volume(Vd; 0.34). A significant inverse CL–OS relationship was determined for NSCLC (HR:1.69; 95%CI1.07–2.68; p=0.024) and MPM (HR: 3.29; 95% CI 1.08 to 10.09; p=0.037), after correction for prognostic factors, which could not confirmed for melanoma (p=0.22) or UCC (p=0.34). No relationship could be determined between CL and grade >3 irAEs (p=0.70).Conclusions High interpatient variability of pembrolizumab PK is determined by BSA and serum albumin (on CL) and LDH (on Vd). A strong inverse CL–OS relationship was demonstrated for NSCLC and MPM, which could not be observed for melanoma and UCC. The findings suggest that personalized dosing should be prospectively explored.

Immune checkpoint inhibitors blocking programmed cell death 1 (PD-1) have been associated with the occurrence of severe (grade ≥3) adverse events (AEs) in 7-20% of patients. A better understanding of risk factors for the development of grade ≥3 AEs may aid clinical monitoring and decision-making. We therefore aimed to explore determinants of the occurrence of grade ≥3 AEs during anti-PD-1 monotherapy. Patients starting nivolumab or pembrolizumab between July 2015 and Jan 2020 and prospectively included in the Dutch MULTOMAB-trial (NL6828), were followed for grade ≥3 AEs until May 2020. Multivariable (backward selection) cox regression analysis was used to explore the associations between occurrence of a first grade ≥3 AE with patient- and treatment-related characteristics, first for the total cohort, and separately for the melanoma and NSCLC cohorts. Covariates used in the analysis were: tumor type, age, gender, prior treatment, WHO performance status (WHO PS), type of PD-1 inhibitor, histology (NSCLC only), and baseline LDH and treatment setting (melanoma only). 554 patients treated with anti-PD1 monotherapy for various tumor types were prospectively included. Overall, 64 patients (12%) experienced at least one grade ≥3 AE during follow-up (75 grade ≥3 AEs in total), most commonly being colitis (n=22), hepatitis (n=10), skin toxicity (n=9) or pneumonitis (n=9). None of the covariates were associated with the occurrence of a first grade ≥3 AE in the total cohort. Within the melanoma cohort (n=231), age ≥65 years was significantly associated with the occurrence of grade ≥3 AEs (HR=2.4; 95%CI:1.1-5.5;p=0.04). None of the covariates were predictive of grade ≥3 AEs in the NSCLC cohort (n=181). However, grade ≥3 AEs were unevenly distributed across WHO PS, as none of the WHO PS=0 patients experienced high grade AEs. Surprisingly, despite the large number of co-variates, no patient- or treatment-related characteristics were predictive for the occurrence of grade ≥3 AEs during anti-PD-1 therapy across tumor types. The increased risk of grade ≥3 AEs in elderly patients with melanoma and impact of WHO PS on incidence of grade ≥3 AEs in NSCLC requires further investigation.

e21135 Background: There is an unmet need for reliable biomarkers predictive of treatment outcome with immunotherapy in patients with non-small cell lung cancer (NSCLC). The aim of our study is to examine, in front-line patients, the potential utility of a proteomic signature, PIR (primary immune response), developed and validated in 2 nd line patients (M. Muller Clin Cancer Res. 2020 Oct 1;26(19):5188-5197. doi: 10.1158/1078-0432.CCR-20-0538. Epub 2020 Jul 6) as a predictive biomarker of response to treatment with pembrolizumab in a first-line setting. Methods: Pretreatment serum samples were collected from stage IV NSCLC patients treated with a pembrolizumab-containing regimen in first-line at the Netherlands Cancer Institute and Erasmus Medical Center between 2016 and 2020. Data on demographics, ECOG performance status (PS), PD-L1 expression, treatment, progression free survival (PFS) and overall survival (OS) were collected. The PIR test was performed and classified in two groups: resistant vs. intermediate/sensitive (not-resistant). Results: Serum samples of 119 patients were available for analysis. Median age was 65 (36-82) years; 49% female; PS 0-1 80%. 45 samples failed QC and were excluded allowing for the analysis of 74 patients: 24 receiving pembrolizumab monotherapy, and 50 receiving a combination of chemotherapy and pembrolizumab. Overall, median PFS was 5.3 months in the resistant group vs. 9.0 months in the not-resistant group (HR 0.60 (95%CI 0.33-1.08; p = 0.09). In patients receiving pembrolizumab monotherapy, the resistant group had a trend toward shorter PFS (median PFS 3.4 months vs. 10.3 months) then the non-resistant group, (HR 0.42 (95% CI 0.15-1.21; p = 0.11))). No difference was observed in patients treated with a combination of chemotherapy and pembrolizumab (median PFS 5.8 months vs. 9.0 months, respectively; HR = 0.65 (95% CI 0.31-1.36; p = 0.25). Median OS was not reached vs 17.0 months, respectively (HR 0.90 (95% CI 0.34-2.36; p = 0.83)). We did not observe any significant association between performance score and PD-L1 expression, and the PIR test. Conclusions: These data encourage further study of PIR as a predictive biomarker for first-line treatment with pembrolizumab in NSCLC patients.

BACKGROUND:A minority of patients with advanced non-small-cell lung cancer (NSCLC) benefit from treatment with immune checkpoint inhibitors (ICIs). Ineffective effector function of activated T and NK cells may lead to reduced tumor cell death, even when these activated effector cells are released from their immune checkpoint brake. Hence, in this study we aimed to assess the association of baseline serum granzyme B, as well as germline variation of the GZMB gene, with clinical outcome to programmed cell death protein 1 (PD-1) blockade.METHODS:A total of 347 patients with stage IV NSCLC who started nivolumab treatment between June 2013 and June 2017 were prospectively included. Baseline serum and whole blood was available, allowing for protein quantification and targeted DNA sequencing. Clinical outcome was based on best overall response (BOR) according to Response Evaluation Criteria in Solid Tumors, V.1.1, progression-free survival (PFS), and overall survival (OS).RESULTS:Patients with low serum levels of granzyme B had worse PFS (HR: 1.96; 95% CI: 1.12 to 3.43; p=0.018) and worse OS (HR: 2.08; 95% CI: 1.12 to 3.87; p=0.021) than patients with high baseline serum levels. To validate the findings, germline variation of GZMB rs8192917 was assessed. Patients with homozygous and heterozygous variants of GZMB rs8192917 had worse BOR (OR: 1.60; 95% CI: 1.01 to 2.52; p=0.044) and worse PFS (HR: 1.38; 95% CI:1.02 to 1.87; p=0.036) than wild types.CONCLUSIONS:A low baseline serum level of granzyme B and germline variation of GZMB was associated with worse clinical outcome in NSCLC, emphasizing the relevance and additional value of monitoring germline genetic variations which mirror cytotoxic functions of T cells in ICI therapy.TRAIL REGISTRATION NUMBER:Dutch Trial Registry (NL6828).

Background: Thyroid dysfunction is among the most common adverse effects during anti-programmed cell death 1 (PD-1) immunotherapy, and alongside correlations with elevated anti-thyroid antibodies (ATAb), studies have found correlations with survival. However, the exact relations remain to be clarified. We, therefore, aimed at clarifying the relationship between thyroid dysfunction, ATAbs, and survival in anti-PD-1 treated cancer patients. Methods: We included 168 patients with nonsmall-cell lung carcinoma, renal cell carcinoma, and metastatic melanoma treated with nivolumab or pembrolizumab. Thyrotropin and free T4 (fT4) levels were measured before each anti-PD-1 infusion. ATAb levels (anti-thyroid peroxidase [TPO] and anti-thyroglobulin [Tg]) were measured at baseline and after two months of treatment. Although the vast majority of patients had detectable levels of ATABs, only a few patients had positive ATAbs when using conventional cut-offs. To study the consequences of detectable ATABs, the cut-off levels were a priori set at the median concentrations at baseline in the study population. Tumor progression was classified according to RECIST v1.1. Results: Patients who acquired overt thyroid dysfunction during treatment had significantly higher overall survival (OS) (hazard ratio [HR] = 0.18 confidence interval [CI: 0.04-0.76]; p = 0.020) and progression-free survival (PFS) (HR = 0.39 [0.15-0.998]; p = 0.050) than patients without thyroid dysfunction with 1-year OS rates of 94% vs. 59% and 1-year PFS rates of 64% vs. 34%. During treatment, patients with ATAb levels above the median had a higher OS (HR = 0.39 [0.21-0.72]; p = 0.003) and PFS (HR = 0.52 [0.33-0.81]; p = 0.004) than patients with ATAb levels below the median, with 1-year OS rates of 83% vs. 49% and PFS rates of 54% vs. 20%, respectively. When analyzing ATAb levels over time, patients with a persistent ATAb level above the median had a higher OS (HR = 0.41 [0.19-0.89], p = 0.025) and PFS (HR = 0.54 [0.31-0.95], p = 0.032) compared with patients with a persistent ATAb level below the median. Patients whose ATAb levels increased above the median during treatment had an improved OS (HR = 0.27 [0.06-1.22], p = 0.088) and PFS (HR = 0.24 [0.07-0.77], p = 0.017) compared with patients whose ATAb levels decreased below the median. Conclusions: Acquired overt thyroid toxicity and above median ATAb levels during anti-PD-1 treatment are associated with improved PFS and OS. In addition, our results suggest that ATAb levels at baseline are of clinical relevance for PFS and OS.

Background Many cancer patients do not obtain clinical benefit from immune checkpoint inhibition. Checkpoint blockade targets T cells, suggesting that tyrosine kinase activity profiling of baseline peripheral blood mononuclear cells may predict clinical outcome. Methods Here a total of 160 patients with advanced melanoma or non-small-cell lung cancer (NSCLC), treated with anti-cytotoxic T-lymphocyte-associated protein 4 (anti-CTLA-4) or anti-programmed cell death 1 (anti-PD-1), were divided into five discovery and cross-validation cohorts. The kinase activity profile was generated by analyzing phosphorylation of peripheral blood mononuclear cell lysates in a microarray comprising of 144 peptides derived from sites that are substrates for protein tyrosine kinases. Binary grouping into patients with or without clinical benefit was based on Response Evaluation Criteria in Solid Tumors V.1.1. Predictive models were trained using partial least square discriminant analysis (PLS-DA), performance of the models was evaluated by estimating the correct classification rate (CCR) using cross-validation. Results The kinase phosphorylation signatures segregated responders from non-responders by differences in canonical pathways governing T-cell migration, infiltration and co-stimulation. PLS-DA resulted in a CCR of 100% and 93% in the anti-CTLA-4 and anti-PD1 melanoma discovery cohorts, respectively. Cross-validation cohorts to estimate the accuracy of the predictive models showed CCRs of 83% for anti-CTLA-4 and 78% or 68% for anti-PD-1 in melanoma or NSCLC, respectively. Conclusion Blood-based kinase activity profiling for response prediction to immune checkpoint inhibitors in melanoma and NSCLC revealed increased kinase activity in pathways associated with T-cell function and led to a classification model with a highly accurate classification rate in cross-validation groups. The predictive value of kinase activity profiling is prospectively verified in an ongoing trial.

Background: Checkpoint inhibitors have become standard care of treatment for non-small cell lung cancer (NSCLC), yet only a limited fraction of patients experiences durable clinical benefit, highlighting the need for markers to stratify patient populations. Methods: To prospectively identify patients showing response to therapy, we have stained peripheral blood samples of NSCLC patients treated with 2nd line nivolumab (n = 71), as well as healthy controls, with multiplex flow cytometry. By doing so, we enumerated 18 immune cell subsets and assessed expression for 28 T cell markers, which was followed by dimensionality reduction as well as rationale-based analyses. Results: In patients with a partial response (PR), representing best overall response (BOR) according to RECIST v1.1, the number of CD8 T cells at baseline and during treatment is similar to those of healthy controls, but 2-fold higher than in patients with progressive and stable disease (PD and SD). CD8 T cell populations in PR patients show enhanced frequencies of T effector memory re-expressing CD45RA (TEMRA) cells, as well as T cells that express markers of terminal differentiation (CD95+) and egression from tumor tissue (CD69-). In PR patients, the fraction of CD8 T cells that lacks co-stimulatory receptors (CD28, ICOS, CD40L, 4-1BB, OX40) correlates significantly with the total numbers and differentiated phenotype of CD8 T cells. Conclusions: This study demonstrates that high numbers of peripheral CD8 T cells expressing differentiation markers and lacking co-stimulatory receptors at baseline are associated with response to nivolumab in NSCLC patients.

Introduction: Nivolumab treatment is subject to large interpatient variability in both efficacy and toxicity, which may partly be explained by differences in nivolumab exposure. Exposure-response relationships in regular healthcare have not been extensively investigated for nivolumab. Therefore, we aimed to identify possible exposure-response relationships in nivolumab-treated patients with non-small-cell lung cancer (NSCLC). Methods: Patients with NSCLC who started second-line nivolumab therapy (3 mg/kg Q2W) between May 5th 2016 and August 1st 2017, and from whom serial blood samples, toxicity data and outcome data were prospectively collected, were included. Follow-up was carried out until November 1st 2017. Patients were classified according to the best overall response (BOR) based on the Response Evaluation Criteria in Solid Tumours, v1.1, and toxicities according to the Common Terminology Criteria for Adverse Events. Nivolumab trough concentrations were measured after 2, 4 and 10 weeks of treatment, excluding dose delays, and calculated geometric means were tested versus BOR or toxicity using analysis of variance and an independent samples t-test, respectively. Overall survival (OS) and progression-free survival were compared between high and low trough concentration groups. Results: Seventy-six patients were evaluable for analyses. Responders (n = 15) had higher mean trough concentrations than patients with progression (n = 33): 47% higher after 2 weeks (p = 0.001), 53% higher after 4 weeks (p = 0.008) and 73% higher after 10 weeks (p = 0.002). Higher trough concentrations were associated with longer OS (p = 0.001). Conclusions: This study shows that patients with NSCLC with a response to nivolumab had a higher nivolumab exposure than patients with progression, indicating a potential exposure-response relationship. Further clinical research should focus on clarifying these exposureeresponse relationships. (C) 2018 Elsevier Ltd. All rights reserved.

Background Nivolumab is administered in a weight-based or fixed-flat dosing regimen. For patients with non-small cell lung cancer (NSCLC), a potential exposure-response relationship has recently been reported and may argue against the current dosing strategies. The primary objectives were to determine nivolumab pharmacokinetics (PK) and to assess the relationship between drug clearance and clinical outcome in NSCLC, melanoma, and renal cell cancer (RCC). Methods In this prospective observational cohort study, individual estimates of nivolumab clearance and the impact of baseline covariates were determined using a population-PK model. Clearance was related to best overall response (RECISTv1.1), and stratified by tumor type. Results Two-hundred-twenty-one patients with metastatic cancer receiving nivolumab-monotherapy were included of whom 1,715 plasma samples were analyzed. Three baseline parameters had a significant effect on drug clearance and were internally validated in the population-PK model: gender, BSA, and serum albumin. Women had 22% lower clearance compared to men, while the threshold of BSA and albumin that led to > 20% increase of clearance was > 2.2m 2 and < 37.5 g/L, respectively. For NSCLC, drug clearance was 42% higher in patients with progressive disease (mean: 0.24; 95% CI: 0.22–0.27 L/day) compared to patients with partial/complete response (mean: 0.17; 95% CI: 0.15–0.19 L/day). A similar trend was observed in RCC, however, no clearance-response relationship was observed in melanoma. Conclusions Based on the first real-world population-PK model of nivolumab, covariate analysis revealed a significant effect of gender, BSA, and albumin on nivolumab clearance. A clearance-response relationship was observed in NSCLC, with a non-significant trend in RCC, but not in melanoma. Individual pharmacology of nivolumab in NSCLC appears important and should be prospectively studied.

Background Thyroid dysfunction is one of the most common adverse effects during anti-PD-1 therapy, and alongside elevated anti-thyroid antibodies (ATAb) it is correlated with overall (OS) and progression free survival (PFS). The objective and novelty of our study was the simultaneous investigation of thyroid dysfunction and ATAbs on survival. Methods We included 168 patients with non-small cell lung carcinoma (n = 93), renal cell carcinoma (n = 12), and advanced and metastatic melanoma (n = 63) treated with nivolumab or pembrolizumab. TSH and fT4 serum levels were measured prior to each anti-PD-1 infusion and ATAb titers in sera, i.e. anti-TPO and anti-Tg (cut-off: 3.05 IU/ml and 22.35 IU/ml, respectively; based on median value at baseline), at baseline, and after 2 months of treatment. Thyroid dysfunction was based on TSH and fT4 and classified as subclinical or overt. Tumor progression was classified according to RECIST v1.1 and was monitored until progression, death or withdrawal of the study. Cox regression was used with correction for tumor type. Results Patients who acquired overt thyroid toxicity during anti-PD-1 treatment had significantly higher OS (HR = 0.17 [95% CI: 0.04-0.74]; p = 0.018) and PFS (HR = 0.38 [0.15-0.98]; p = 0.05) than patients without thyroid toxicity with one-year OS rates of 95% vs 64% and one-year PFS rates of 65% vs 33%. Moreover, patients with positive ATAb status during treatment had higher OS (HR = 0.39 [0.21-0.72]; p = 0.003) and PFS (HR = 0.52 [0.33-0.81]; p = 0.004) than patients with negative ATAb status with one-year OS rates of 83% vs 49% and PFS rates of 54% vs 20%, respectively. For 84% of patients the ATAb status at baseline was the same as during treatment. Patients with persistent positive ATAb status (48%) had higher OS (HR = 0.41 [0.19-0.89], p = 0.03) and PFS (HR = 0.54 [0.31-0.95], p = 0.03) compared to patients with persistent negative ATAb status (36%). Conclusions Acquired overt thyroid toxicity and positive ATAb status during anti-PD-1 treatment are associated with improved PFS and OS. Moreover, our results indicate that ATAb status at baseline is of clinical relevance for PFS and OS. If validated, these parameters may serve as a novel positive predictive markers. Clinical trial identification Dutch trial register: MULTOMAB, NL6828. Legal entity responsible for the study Erasmus Medical Center. Funding Erasmus Medical Center. Disclosure A.A.M. Van der Veldt: Advisory / Consultancy: Bristol-Myers Squibb; Advisory / Consultancy: Merck Sharp & Dohme. R.H. Mathijssen: Research grant / Funding (institution), Travel / Accommodation / Expenses: Astellas; Research grant / Funding (institution): Bayer; Research grant / Funding (institution): Boekringer; Research grant / Funding (institution): Cristal Therapeutics; Honoraria (self), Research grant / Funding (institution): Novartis; Research grant / Funding (institution): Pamgene; Research grant / Funding (institution), Travel / Accommodation / Expenses: Pfizer; Research grant / Funding (institution): Roche; Research grant / Funding (institution): Sanofi; Honoraria (self): Servier. All other authors have declared no conflicts of interest.

9057 Background: Treatment with anti-PD-1 therapy is subject to large inter-individual variation in clinical outcome. This may be influenced partly by differences in nivolumab exposure between patients. The objective of the current analysis was to investigate whether an exposure-response relationship exists for nivolumab-treated NSCLC patients. Methods: 84 patients started nivolumab treatment between May 5th 2016 and August 1st 2017 and were included for prospective collection of serum samples prior to each nivolumab cycle. Clinical data were collected until November 1st 2017. Patients were classified according to best objective response (BOR) according to RECIST v1.1 and the occurrence of grade ≥3 toxicities according to CTCAE v4.03. Geometric mean nivolumab trough concentrations after 2, 4, and 10 weeks of treatment were compared using ANOVA with respect to BOR and t-test with respect to toxicity in patients without dose delays until that particular time point. Results: 76 patients were evaluable for analysis; 4 patients had no follow-up tumor evaluation, and 4 patients had no follow-up blood sample. At each time point, partial responders (PR; n = 15) had higher geometric mean trough concentrations when compared to patients with progression (PD; n = 33): at 2 weeks : 27.4 µg/mL (95% CI: 22.3 – 33.6 µg/mL) vs. 18.7 µg/mL (95% CI: 16.7 – 20.9 µg/mL; p = 0.001; 47% higher), at 4 weeks: 46.2 µg/mL (95% CI: 37.4 – 57.0 µg/mL) vs. 30.2 µg/mL (95% CI: 25.0 – 36.4 µg/mL; p = 0.008; 53% higher), at 10 weeks: 79.4 µg/mL (95% CI: 60.7 – 103.8 µg/mL) vs. 45.8 µg/mL (95% CI: 35.6 – 58.9 µg/mL; p = 0.002; 73% higher). Moreover, PR patients had higher trough concentrations when compared to patients with stable disease (SD; n = 28) at week 2 (p = 0.034) and at week 4 (p = 0.047). Exposure was not significantly related to the occurrence of grade ≥3 toxicity (n = 15) at any time point. Conclusions: This analysis shows that NSCLC patients with an objective response to nivolumab have significantly higher nivolumab exposure than patients with early progressive disease, indicating an exposure-response relationship. Further clinical research is needed to explain and quantify this relation. If confirmed, more rational and individualized dosing-strategies can improve patient outcome.

9579 Background: Prediction of clinical responses to checkpoint inhibitor therapies is urgently needed. Notably, a significant proportion of patients does not benefit from the treatment, agents are costly and may cause serious toxicity. The kinase activity of peripheral blood cells (PBMCs) may reflect biological mechanisms underlying response to immunotherapy. We hypothesized that kinase activity profiles from PBMCs may constitute a predictive marker for clinical response to CTLA4 and/or PD1 blockade immunotherapy in patients with advanced melanoma. Methods: In a multi-center effort, data were prospectively collected from 6 cohorts of anti-CTLA4- or anti-PD1-treated advanced melanoma patients (n = 138). Kinase activity profiles were generated by analyzing phosphorylation signatures of PBMC lysates on a peptide micro-array. The PamChip (PamGene, Netherlands) microarray comprises 144 different peptides derived from protein phosphorylation sites that are substrates for protein tyrosine kinases. Performance of the predictive model (PLS-DA) was estimated using cross-validation and described by correct classification rate (CCR). Analyses were based on binary grouping of best overall response (RECIST v1.1; CR/PR/SD vs PD) and early/late progression using PFS data (cut-off 140 days). Results: Predictive signatures were discovered for anti-CTLA4 in cohort 1 (anti-CTLA4; n = 10; CCR = 100%; 95%CI 69-100%) and confirmed in cohort 2 (anti-CTLA4; n = 28; CCR = 82%; 95%CI 63-94%), as well as for anti-PD1 in cohort 3 (anti-PD1; n = 17; CCR = 76%; 95%CI 50-93%), which was confirmed in cohort 4 (anti-PD1; n = 29; CCR = 72%; 95%CI 53-87%), cohort 5 (anti-PD1; n = 38; CCR = 75%; 95%CI 57-87%), and cohort 6 (anti-PD1; n = 16; non-evaluable due to the low number of responders). Conclusions: In advanced melanoma patients, kinase activity profiles of baseline PBMCs samples can predict the likelihood of response to anti-PD1 or anti-CTLA4 therapy. This assay may serve as a rapid and fast predictive liquid biomarker to stratify patients prior to treatment. Involvement of receptor tyrosine kinases underlying the mechanism are being further elucidated and a larger validation study is underway.

Background Treatment with PD-1 inhibitors can be hampered by severe auto-immune-related toxicities. Our objective was to identify single-nucleotide polymorphisms (SNPs) in genes previously associated with auto-immunity, which are associated with toxicities in nivolumab-treated NSCLC patients. This was in order to identify patients prone to develop severe toxicities and to gain more insight into the underlying pathobiology. Methods We analysed 322 nivolumab-treated patients and assessed the association with toxicities for seven SNPs in four genes, which are considered contributors to PD-1-directed T-cell responses, i.e., PDCD1 , PTPN11 , ZAP70 and IFNG . Every SNP was tested for its association with toxicity endpoints. Significant associations were tested in a validation cohort. Results A multivariable analysis in the exploration cohort showed that homozygous variant patients for PDCD1 804C>T (rs2227981) had decreased odds for any grade treatment-related toxicities ( n = 96; OR 0.4; 95% CI 0.2–1.0; p = 0.039). However, this result could not be validated ( n = 85; OR 0.9; 95% CI 0.4–1.9; p = NS). Conclusions Our results show that it is unlikely that the investigated SNPs have a clinical implication in predicting toxicity. A finding, even though negative, that is considered timely and instructive towards further research in biomarker development for checkpoint inhibitor treatments.

Background: Treatment with monoclonal antibodies (mAbs) against programmed cell death protein 1 receptor is subject to high variation in treatment outcome among cancer patients. For these agents, no exposure–response (ER) relationships have been investigated in routine health care settings. However, ER relationships have been identified for several other mAbs used in oncology. Methods to conveniently measure serum concentrations of anti–programmed cell death protein 1 mAbs in routine health care may clarify possible ER relationships. Therefore, the authors aimed to develop an enzyme-linked immune sorbent assay (ELISA) for the measurement of both nivolumab and pembrolizumab serum concentrations of treated cancer patients. Methods: Optimal capture antigen and detection antibody concentrations were selected based on titrations. Nivolumab calibration standards ranging from 0.2 to 300 ng/mL were tested in duplicate. Accuracy was assessed in 2 recovery experiments. Intra- and interassay variations were assessed on 3 different days by 2 independent technicians. The developed ELISA was also set up for pembrolizumab calibration curves. Cross-reactivity of nivolumab measurements with ipilimumab was assessed. Of one nivolumab treated patient, serum concentrations in follow up samples were measured and presented. Results: Nivolumab calibration standards of 0.20–25 ng/mL were used. Nivolumab trough concentrations after 1 cycle in 8 patients ranged from 17.3 to 31.1 mcg/mL. The range of accuracy was 84%–105%, whereas intra- and interassay variations showed a coefficient of variation of 5.5% and 10.1%, respectively. No cross-reactivity with ipilimumab was detected. Pembrolizumab trough concentrations (n = 8) ranged from 9.1 to 19.7 mcg/mL after 1 infusion. Conclusions: The in-house–developed ELISA provides the opportunity to measure both nivolumab and pembrolizumab serum concentrations. This may help identify possible ER relationships in treated cancer patients and may potentially lead to dose adjustments in the future.

e21239 Background: Checkpoint inhibitors, such as the PD-1 antibody nivolumab, have been approved for the treatment of several cancer types, including NSCLC. As only a limited subset of NSCLC patients experiences durable benefit, there is a need for early prediction of clinical response. Here, we aim to identify immune markers in peripheral blood samples of NSCLC patients treated with nivolumab. Methods: In 71 NSCLC patients treated with 2nd line nivolumab (2-weekly), whole blood samples were collected at 3 time points (pre-treatment; prior to 2nd and 3rd administration of nivolumab). Multiplex flow cytometry was applied to enumerate 19 immune cell subsets and determine frequencies of T cell subsets positive for 28 markers (single + combinations), enabling classification of T cells according to activation, maturation, co-signaling, and chemotaxis. Best overall response (BOR) within 90 days was assessed using RECIST (v1.1). Standard statistical tests were applied. Results: Patients showing partial response (PR) displayed increased numbers of CD8 T cells prior to and throughout therapy with a median (range) of 560 cells/μl (170-1900), while patients with progressive disease (PD) showed a median of 220 cells/μl (90-1070) (p = 0.022). PR patients also possessed significantly higher frequencies of CD8T cells positive for the maturation marker CD95 (60% (9-87)) at baseline and throughout therapy when compared to PD patients (45% (5-82)) (p = 0.041). With regard to co-signaling, PD patients displayed a significantly lower median frequency of CD8 T cells positive for CD279 (PD-1) and CD366 (TIM3) at baseline (4.8% (1.2-17.6)) and, unlike PR patients, a drop in CD4 T cells positive for multiple co-inhibitory receptors from 24% (8-36) to 17% (8-36) following therapy. In addition, PD patients showed a higher frequency of CD4 T cells positive for co-stimulatory CD28 and CD278 (ICOS) (35% (12-48)). Conclusions: We demonstrate that total numbers of CD8 T cells as well as T cells subsets in peripheral blood, in particular those expressing co-signaling receptors, associate with response to nivolumab in NSCLC patients. When confirmed, these findings will facilitate clinical decision making.