Carbone, David Paul

Importance Outcomes for patients with unresectable stage III non–small cell lung cancer (NSCLC) treated with chemoradiation therapy (CRT) have improved with adjuvant immune checkpoint inhibitors, with a reported 5-year overall survival benefit of approximately 10% for adjuvant durvalumab vs placebo after completion of CRT without progression and with preserved performance status. Starting atezolizumab prior to CRT may allow more patients to benefit from immunotherapy. Objective To evaluate clinical outcomes of patients treated with atezolizumab before and after CRT for unresectable stage III NSCLC. Design, Setting, and Participants This single-cohort, phase II, nonrandomized controlled trial was conducted at 11 US sites. Patients with pathologically confirmed, unresectable stage III NSCLC who were treatment naive and had good performance status were enrolled between January 3, 2018, and July 24, 2019. Data were locked on March 21, 2023. Interventions Patients received four 21-day cycles of atezolizumab, 1200 mg intravenously, with therapy administered on day 1 of each cycle. Patients not experiencing tumor progression continued to CRT (60 Gy to involved fields) concurrent with weekly carboplatin area under the curve of 2 and paclitaxel, 50 mg/m 2 , followed by planned consolidation carboplatin area under the curve of 6 and paclitaxel, 200 mg/m 2 , for two 21-day cycles. Patients not experiencing progression continued atezolizumab, 1200 mg, every 21 days to complete 1 year of therapy. Main Outcomes and Measures The primary end point was the disease control rate at 12 weeks. Secondary end points were progression-free survival, overall survival, overall response rate, safety, and translational science end points. Results A total of 62 patients (median [range] age, 63.9 [38.1-86.5] years; 32 female [51.6%]) were enrolled and received at least 1 dose of atezolizumab. The disease control rate at 12 weeks was 74.2% (80% CI, 65.7%-81.4%). Median progression-free survival was 30.0 months (95% CI, 15.8 to not evaluable), and the median overall survival was not reached. The overall survival rate at 24 months was 73.7% (95% CI, 63.4%-85.7%), and the overall response rate was 66.2%. Seventeen patients (27.4%) experienced grade 3 or higher immune-related adverse events, including 1 with grade 5 pneumonitis and 1 with grade 4 Guillain-Barré syndrome. Thirty patients (48.4%) experienced grade 3 or higher treatment-related adverse events. Conclusions and Relevance These findings suggest that neoadjuvant atezolizumab merits further study based on safety and encouraging outcomes. Trial Registration ClinicalTrials.gov Identifier: NCT03102242

Abstract Background: Immune checkpoint inhibitors (ICIs) have significantly improved outcomes for patients with metastatic non-small cell lung cancer (NSCLC). However, clinical management is challenging. The PD-L1 biomarker does not optimally guide the choice between ICI monotherapy and combination ICI-chemotherapy, and there are currently no biomarkers to predict immune-related adverse events (irAEs). Here, we describe a versatile platform based on plasma proteomics and machine learning for generating predictive tools to support treatment decisions. Methods: Pre-treatment plasma samples were collected from 616 NSCLC patients receiving ICI-based therapy (NCT04056247). Clinical benefit (CB) data and irAEs were recorded. Proteomic profiling of plasma samples was performed using the SomaScan® assay, covering ~7000 proteins per sample. Proteins associated with Clinical Benefit (progression-free survival at 12 months) were classified as resistance-associated proteins (RAPs). Proteins associated with grade ≥3 irAEs occurring within the first 100 days of treatment were classified as toxicity-associated proteins (TAPs). RAPs and TAPs were integrated into two separate machine learning-based models, yielding CB and irAE probabilities, respectively. The CB model used a CB probability threshold to assign patients a PROphet-positive or PROphet-negative result. Bioinformatic analysis was performed on RAPs and TAPs. Results: CB and irAE models displayed strong predictive performance with a high correlation between predicted probability and observed CB or irAE rate (CB model: R2=0.98; p-value<0.0001; irAE model: R2= 0.92; p-value <0.0001). In the CB model, patients classified as PROphet-positive achieved significantly longer overall survival (OS) than PROphet-negative patients, with a median OS of 25.9 vs. 10.8 months (hazard ratio, HR = 0.51; p-value<0.001). Furthermore, patients with PD-L1 ≥50% tumors and a PROphet-negative result had significantly longer OS with ICI-chemotherapy in comparison to ICI monotherapy (HR=0.23, p-value = 0.0003), suggesting that combination therapy is preferable for such patients despite high PD-L1 levels. Bioinformatic analysis showed that RAPs related to immune regulation, angiogenesis, chemo-resistance, and other potential resistance mechanisms displayed higher expression levels in patients who did not benefit from treatment. Neutrophil- and inflammation-related proteins were significantly enriched in patients who experienced irAEs. Conclusions: We describe two machine-learning based predictive tools that may be used separately or in combination to inform treatment decisions based on the patient’s likelihood to benefit from ICI therapy and to develop serious irAEs. Our findings also provide biological insights related to treatment resistance and immune-related toxicity. Citation Format: Michal Harel, Petros Christopoulos, Igor Puzanov, Jair Bar, Iris Kamer, Niels Reinmuth, Ina Koch, Mor Moskovitz, Adva Levy-Barda, Alona Zer, Michal Lotem, David Farrugia, Rivka Katzenelson, Gillian Price, Abed Agbarya, Helen Cheley, Adam Hassani, Mahmoud Abu-Amna, Tom Geldart, Anirban Chatterjee, Andreas Polychronis, Maya Gottfried, Yanyan Lou, Tatiana Harkovsky, Alison Brewster, Ido Wolf, Ella Tepper, Ben Yellin, Itamar Sela, Coren Lahav, Yehonatan Elon, Raya Leibowitz, Adam P. Dicker, Young Kwang Chae, Ryan J. Sullivan, David Carbone, David Gandara, Jarushka Naidoo. Plasma proteomics-based models for predicting therapeutic benefit and immune-related adverse events in non-small cell lung cancer patients treated with immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1208.

PURPOSE Current guidelines for the management of metastatic non–small cell lung cancer (NSCLC) without driver mutations recommend checkpoint immunotherapy with PD-1/PD-L1 inhibitors, either alone or in combination with chemotherapy. This approach fails to account for individual patient variability and host immune factors and often results in less-than-ideal outcomes. To address the limitations of the current guidelines, we developed and subsequently blindly validated a machine learning algorithm using pretreatment plasma proteomic profiles for personalized treatment decisions. PATIENTS AND METHODS We conducted a multicenter observational trial (ClinicalTrials.gov identifier: NCT04056247 ) of patients undergoing PD-1/PD-L1 inhibitor–based therapy (n = 540) and an additional patient cohort receiving chemotherapy (n = 85) who consented to pretreatment plasma and clinical data collection. Plasma proteome profiling was performed using SomaScan Assay v4.1. RESULTS Our test demonstrates a strong association between model output and clinical benefit (CB) from PD-1/PD-L1 inhibitor–based treatments, evidenced by high concordance between predicted and observed CB ( R 2 = 0.98, P < .001). The test categorizes patients as either PROphet-positive or PROphet-negative and further stratifies patient outcomes beyond PD-L1 expression levels. The test successfully differentiates between PROphet-negative patients exhibiting high tumor PD-L1 levels (≥50%) who have enhanced overall survival when treated with a combination of immunotherapy and chemotherapy compared with immunotherapy alone (hazard ratio [HR], 0.23 [95% CI, 0.1 to 0.51], P = .0003). By contrast, PROphet-positive patients show comparable outcomes when treated with immunotherapy alone or in combination with chemotherapy (HR, 0.78 [95% CI, 0.42 to 1.44], P = .424). CONCLUSION Plasma proteome–based testing of individual patients, in combination with standard PD-L1 testing, distinguishes patient subsets with distinct differences in outcomes from PD-1/PD-L1 inhibitor–based therapies. These data suggest that this approach can improve the precision of first-line treatment for metastatic NSCLC.

Background: Immune checkpoint inhibitors (ICIs) are a first-line and perioperative treatment for lung cancer. Pneumonitis is a potentially life-threatening complication of ICI treatment in 2% to 5% of patients; however, risk factors for developing ICI pneumonitis (ICI-p) remain undefined. Methods: We conducted a retrospective cohort study of consecutive patients with lung cancer who received at least one dose of ICI from 2015 through 2020 at The Ohio State University. Pneumonitis cases were documented by the treating oncologist and retrospectively evaluated for agreement between an oncologist and a pulmonologist. Patient demographic and clinical characteristics were recorded and summarized between those with and without pneumonitis for the overall cohort. Univariate and multivariable survival analyses using the Fine-Gray competing risk model were used to examine the associations. Results: A total of 471 patients with lung cancer were included, of which 402 had non–small cell lung cancer and 69 had small cell lung cancer; 39 (8%) patients in the overall cohort developed ICI-p. Preexisting interstitial abnormalities and prior chest radiation were both significantly associated with ICI-p on univariate analysis (hazard ratio [HR], 8.91; 95% CI, 4.69–16.92; P<.001; and HR, 2.81; 95% CI, 1.50–5.28; P=.001). On multivariable analyses, interstitial abnormalities remained a strong independent risk factor for ICI-p when controlling for chest radiation and type of immunotherapy (HR, 9.77; 95% CI, 5.17–18.46; P<.001). Among patients with ICI-p (n=39), those with severe (grade 3–5) pneumonitis had worse overall survival compared with those with mild (grade 1 or 2) pneumonitis (P=.001). Abnormal pulmonary function test results at both 12 and 18 months prior to ICI initiation were not significantly associated with ICI-p. Conclusions: Preexisting interstitial abnormalities on chest CT and prior chest radiation are independent risk factors that are strongly associated with ICI-p in patients with lung cancer. These findings highlight a potential need for closer observation for ICI-p among patients with these risk factors.

9122 Background: Initial treatment selection for advanced NSCLC patients without driver mutations is mainly based on evaluating PD-L1 expression levels in the tumor tissue. However, PD-L1 assays are only moderately predictive. In addition, the guidelines for the PD-L1≥50% subpopulation are not definitive, enabling the usage of immune checkpoint inhibitors (ICI) either as a monotherapy or combined with chemotherapy. Here we aim to provide a decision-making tool for the first-line treatment of advanced NSCLC patients based on plasma derived biomarkers obtained before treatment initiation. Methods: Pre-treatment plasma samples were collected from 545 NSCLC patients receiving ICI-based therapy. Clinical benefit was evaluated based on progression-free survival (PFS) at 12 months as a threshold. Deep plasma proteomic profiling was performed using aptamer technology. Based on the proteomic profiles, a computational model, termed PROphet, was developed and evaluated in a blinded manner on the validation subset (n = 272). The model output, clinical benefit probability, was used to stratify the patients into two groups, PROphet-positive or -negative. Results: The model displayed good performance with a high correlation between the predicted clinical benefit and the observed clinical benefit rate (R2= 0.97), outperforming a PD-L1-based prediction model (R2= 0.35). Patients classified as PROphet-positive achieved significantly longer overall survival (OS) than PROphet-negative patients, with a median OS of 25.9 versus 10.8 months (hazard ratio, HR = 0.51; 95% confidence interval, CI = 0.37-0.70; p-value < 0.001). Next, we examined the clinical utility of combining the PROphet output with PD-L1 levels by comparing different treatment modalities. Focusing on PD-L1≥50%, patients with PROphet-positive results displayed similar OS and PFS when receiving ICI monotherapy or ICI-chemotherapy (OS HR = 0.77; CI = 0.42-1.43; p-value = 0.4096), suggesting that these patients may consider monotherapy, thus avoiding chemotherapy-related toxicity. Conversely, PD-L1≥50% patients with a PROphet-negative result displayed a significantly longer OS and PFS when receiving ICI-chemotherapy in comparison to ICI monotherapy, with a median OS that was not reached versus 11.10 months in the combination therapy and monotherapy groups, respectively (HR = 0.29; CI = 0.14-0.59; p-value < 0.001). These patients should consider combination therapy, avoiding a treatment that is suboptimal for them. Results for PD-L1 1-49% and < 1% provide additional clinical information for each subgroup. Conclusions: Plasma proteomic profiling can provide biomarkers that accurately predict the benefit of ICI-based treatment. The model output complements tissue PD-L1 expression levels as a tool to assist therapeutic decisions.

Abstract Introduction: Treatment modality selection for metastatic NSCLC patients (immunotherapy alone vs. combination of immunotherapy with chemotherapy) relies mainly on determining the programmed cell death 1 (PD-L1) expression levels in the tumor. However, available assays are only moderately predictive. Here we set to develop PROphet®, a plasma proteomics-based predictive model for informing treatment decisions for NSCLC patients receiving immune checkpoint inhibitor (ICI)-based therapy. Methods: Pre-ICI plasma samples were collected in 12 centers in a clinical trial (NCT04056247) from 367 advanced-stage NSCLC patients. Clinical benefit (CB) was assessed at 12 months based on the occurrence of progression-free survival (PFS) until this time point. Deep proteomic profiling of the plasma samples was performed using aptamer-based technology. A novel machine learning model was developed to determine the CB probability for each patient, and the performance was successfully evaluated on an independent validation set, followed by training and prediction over the entire cohort using cross-validation. The resulting PROphet® score (positive or negative) was determined by setting the median CB rate probability as a threshold. The patients were divided into four subgroups based on their PD-L1 expression level combined with their PROphet® score prediction, and the overall survival (OS) was examined for each subgroup. Results: The PROphet® computational model was evaluated in a blinded manner on a subset of 85 patients and displayed strong predictive capability with area under the curve (AUC) of the receiver operating characteristics (ROC) plot of 0.78 (p-value = 5.00e-05), outperforming a PD-L1-based predictive model (AUC = 0.62; p-value 2.76e-01). When combining PROphet® score with PD-L1 expression levels, four different outcome patterns were identified: (i) Patients with PD-L1 ≥50% and PROphet® negative score, who displayed significantly longer OS when treated with ICI-chemotherapy combination therapy compared to ICI monotherapy and may consider combination therapy. (ii) Patients with PD-L1 ≥50% and PROphet® positive score, who benefit similarly from either treatment modalities, and may consider monotherapy to avoid the potential toxicity of the combination therapy. (iii) Patients with PD-L1&lt;50% and PROphet® negative score, who do not benefit from either treatment modalities and may consider chemotherapy alone or treatment beyond standard of care. (iv) Patients with PD-L1&lt;50% and PROphet® positive score who benefit from combination therapy. Conclusions: Altogether, the PROphet® model, when combined with PD-L1 test, stratifies the patients into four subgroups, providing additional resolution to the PD-L1 biomarker currently used to guide treatment selection. Furthermore, the model accurately predicts CB at 12 months based on proteomic analysis of a pre-treatment plasma sample. Citation Format: Michal Harel, Petros Christopoulos, Coren Lahav, Itamar Sela, Nili Dahan, Niels Reinmuth, Ina Koch, Alona Zer, Mor Moskovitz, Adva Levy-Barda, Michal Lotem, Hovav Nechushtan, Rivka Katzenelson, Abed Agbarya, Mahmoud Abu-Amna, Maya Gottfried, Ido Wolf, Ella Tepper, Yanyan Lou, Raya Leibowitz, Adam P. Dicker, David Gandara, David P. Carbone. A pretreatment blood-based proteomic biomarker for enhanced decision-making in non-small cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2159.

e21134 Background: Therapies with monoclonal antibodies targeting PD-1 and its ligands are associated with remarkable outcomes and have revolutionized cancer treatment (Honey 2017). However, patients treated with PD-1/PD-L1 blockade may develop “a primary or secondary resistance” to therapy (Sharma, Hu-Lieskovan et al. 2017). Contrary to monoclonal antibodies, chimeric B-cell cancer vaccines have the advantage of producing polyclonal B-cell antibodies that can potentially induce memory B- and T-cell responses, while reducing immune evasion and suppression. The hypothesis is that a polyclonal induced B-cell antibody response will be more effective or as effective with improved safety over current monoclonal antibody therapy. IMU-201 (PD1-Vaxx) is being developed using an active immunization approach to treat cancers that overexpress PD-L1 by inducing the production of anti-PD-1 antibodies with a peptide epitope designed to stimulate polyclonal antibodies against PD-1 (Kaumaya et al. 2020). Methods: The IMPRINTER study is an ongoing open-label dose escalation study of IMU-201 as monotherapy (Phase 1) or in combination with atezolizumab (Phase 1b) for patients with PD-L1 expressing non-small cell lung cancer (NSCLC). All patients enrolled in Phase 1 of the study must have previously received an immune checkpoint inhibitor and experienced disease progression. The primary objective is to evaluate the safety and tolerability of IMU-201 and identify the optimal biological dose (OBD). The secondary objective is to evaluate the efficacy of IMU-201 as monotherapy and in combination with atezolizumab. Humoral and cellular immunogenicity data will be evaluated, including IMU-201 and PD-1 specific antibodies (IgG, IgM), vaccine-specific cytokine levels, and regulatory and effector T and B cells. IMU-201 is administered by intramuscular (IM) injection on Day 1, Day 15, and Day 29. Dose-limiting toxicity (DLT) assessment is completed after 29 days on treatment. Tumor progression is evaluated according to RECIST 1.1 at Day 43 then every 42 days until progression or withdrawal. Results: In Phase 1, four patients were enrolled into each of the three cohorts at 10 μg/dose, 50 μg/dose and 100 μg/dose IMU-201 with no DLTs observed. In the 10 μg/dose cohort, one patient achieved CR and one patient SD; in the 50 μg/dose cohort, four patients achieved SD; and in the 100 μg/dose cohort, one patient achieved PR and two patients achieved SD. Within the 100 μg/dose cohort, one patient experienced an immune related pneumonitis after two IMU-201 administrations and discontinued from study treatment. Conclusions: IMU-201 had no observed DLT and demonstrated preliminary signs of efficacy. The study will therefore move into Phase 1b with IMU-201 being assessed in combination with atezolizumab. Clinical trial information: NCT04432207.

Abstract Background: Current management of lung nodules is complicated by nontherapeutic resections and missed chances for cure. We hypothesized that a serum proteomic signature may add diagnostic information beyond that provided by combined clinical and radiographic data. Methods: Cohort A included 265 and cohort B 114 patients. Using multivariable logistic regression analysis we calculated the area under the receiver operating characteristic curve (AUC) and quantified the added value of a previously described serum proteomic signature beyond clinical and radiographic risk factors for predicting lung cancer using the integration discrimination improvement (IDI) index. Results: The average computed tomography (CT) measured nodule size in cohorts A and B was 37.83 versus 23.15 mm among patients with lung cancer and 15.82 versus 17.18 mm among those without, respectively. In cohort A, the AUC increased from 0.68 to 0.86 after adding chest CT imaging variables to the clinical results, but the proteomic signature did not provide meaningful added value. In contrast, in cohort B, the AUC improved from 0.46 with clinical data alone to 0.61 when combined with chest CT imaging data and to 0.69 after adding the proteomic signature (IDI of 20% P = 0.0003). In addition, in a subgroup of 100 nodules between 5 and 20 mm in diameter, the proteomic signature added value with an IDI of 15% (P ≤ 0.0001). Conclusions: The results show that this serum proteomic biomarker signature may add value to the clinical and chest CT evaluation of indeterminate lung nodules. Impact: This study suggests a possible role of a blood biomarker in the evaluation of indeterminate lung nodules. Cancer Epidemiol Biomarkers Prev; 21(5); 786–92. ©2012 AACR.