Enhancing NSCLC recurrence prediction with PET/CT habitat imaging, ctDNA, and integrative radiogenomics-blood insights

While we recognize the prognostic importance of clinicopathological measures and circulating tumor DNA (ctDNA), the independent contribution of quantitative image markers to prognosis in non-small cell lung cancer (NSCLC) remains underexplored. In our multi-institutional study of 394 NSCLC patients, we utilize pre-treatment computed tomography (CT) and 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) to establish a habitat imaging framework for assessing regional heterogeneity within individual tumors. This framework identifies three PET/CT subtypes, which maintain prognostic value after adjusting for clinicopathologic risk factors including tumor volume. Additionally, these subtypes complement ctDNA in predicting disease recurrence. Radiogenomics analysis unveil the molecular underpinnings of these imaging subtypes, highlighting downregulation in interferon alpha and gamma pathways in the high-risk subtype. In summary, our study demonstrates that these habitat imaging subtypes effectively stratify NSCLC patients based on their risk levels for disease recurrence after initial curative surgery or radiotherapy, providing valuable insights for personalized treatment approaches.

Synthetic PET from CT improves diagnosis and prognosis for lung cancer: Proof of concept

[18F]Fluorodeoxyglucose positron emission tomography (FDG-PET) and computed tomography (CT) are indispensable components in modern medicine. Although PET can provide additional diagnostic value, it is costly and not universally accessible, particularly in low-income countries. To bridge this gap, we have developed a conditional generative adversarial network pipeline that can produce FDG-PET from diagnostic CT scans based on multi-center multi-modal lung cancer datasets (n = 1,478). Synthetic PET images are validated across imaging, biological, and clinical aspects. Radiologists confirm comparable imaging quality and tumor contrast between synthetic and actual PET scans. Radiogenomics analysis further proves that the dysregulated cancer hallmark pathways of synthetic PET are consistent with actual PET. We also demonstrate the clinical values of synthetic PET in improving lung cancer diagnosis, staging, risk prediction, and prognosis. Taken together, this proof-of-concept study testifies to the feasibility of applying deep learning to obtain high-fidelity PET translated from CT.

Tumor- and circulating-free DNA methylation identifies clinically relevant small cell lung cancer subtypes

Small cell lung cancer (SCLC) is an aggressive malignancy composed of distinct transcriptional subtypes, but implementing subtyping in the clinic has remained challenging, particularly due to limited tissue availability. Given the known epigenetic regulation of critical SCLC transcriptional programs, we hypothesized that subtype-specific patterns of DNA methylation could be detected in tumor or blood from SCLC patients. Using genomic-wide reduced-representation bisulfite sequencing (RRBS) in two cohorts totaling 179 SCLC patients and using machine learning approaches, we report a highly accurate DNA methylation-based classifier (SCLC-DMC) that can distinguish SCLC subtypes. We further adjust the classifier for circulating-free DNA (cfDNA) to subtype SCLC from plasma. Using the cfDNA classifier (cfDMC), we demonstrate that SCLC phenotypes can evolve during disease progression, highlighting the need for longitudinal tracking of SCLC during clinical treatment. These data establish that tumor and cfDNA methylation can be used to identify SCLC subtypes and might guide precision SCLC therapy.

The role of APOBEC3B in lung tumor evolution and targeted cancer therapy resistance

In this study, the impact of the apolipoprotein B mRNA-editing catalytic subunit-like (APOBEC) enzyme APOBEC3B (A3B) on epidermal growth factor receptor (EGFR)-driven lung cancer was assessed. A3B expression in EGFR mutant (EGFRmut) non-small-cell lung cancer (NSCLC) mouse models constrained tumorigenesis, while A3B expression in tumors treated with EGFR-targeted cancer therapy was associated with treatment resistance. Analyses of human NSCLC models treated with EGFR-targeted therapy showed upregulation of A3B and revealed therapy-induced activation of nuclear factor kappa B (NF-κB) as an inducer of A3B expression. Significantly reduced viability was observed with A3B deficiency, and A3B was required for the enrichment of APOBEC mutation signatures, in targeted therapy-treated human NSCLC preclinical models. Upregulation of A3B was confirmed in patients with NSCLC treated with EGFR-targeted therapy. This study uncovers the multifaceted roles of A3B in NSCLC and identifies A3B as a potential target for more durable responses to targeted cancer therapy.

Distinct patterns of auto-reactive antibodies associated with organ-specific immune-related adverse events

The roles of preexisting auto-reactive antibodies in immune-related adverse events (irAEs) associated with immune checkpoint inhibitor therapy are not well defined. Here, we analyzed plasma samples longitudinally collected at predefined time points and at the time of irAEs from 58 patients with immunotherapy naïve metastatic non-small cell lung cancer treated on clinical protocol with ipilimumab and nivolumab. We used a proteomic microarray system capable of assaying antibody reactivity for IgG and IgM fractions against 120 antigens for systemically evaluating the correlations between auto-reactive antibodies and certain organ-specific irAEs. We found that distinct patterns of auto-reactive antibodies at baseline were associated with the subsequent development of organ-specific irAEs. Notably, ACHRG IgM was associated with pneumonitis, anti-cytokeratin 19 IgM with dermatitis, and anti-thyroglobulin IgG with hepatitis. These antibodies merit further investigation as potential biomarkers for identifying high-risk populations for irAEs and/or monitoring irAEs during immunotherapy treatment.

Trial registration

ClinicalTrials.gov identifier: NCT03391869.

ATM Mutations Associate with Distinct Co-Mutational Patterns and Therapeutic Vulnerabilities in NSCLC

PURPOSE

Ataxia-telangiectasia mutated (ATM) is the most frequently mutated DNA damage repair gene in non-small cell lung cancer (NSCLC). However, the molecular correlates of ATM mutations and their clinical implications have not been fully elucidated.

EXPERIMENTAL DESIGN

Clinicopathologic and genomic data from 26,587 patients with NSCLC from MD Anderson, public databases, and a de-identified nationwide (US-based) NSCLC clinicogenomic database (CGDB) were used to assess the co-mutation landscape, protein expression, and mutational processes in ATM-mutant tumors. We used the CGDB to evaluate ATM-associated outcomes in patients treated with immune checkpoint inhibitors (ICI) with or without chemotherapy, and assessed the effect of ATM loss on STING signaling and chemotherapy sensitivity in preclinical models.

RESULTS

Nonsynonymous mutations in ATM were observed in 11.2% of samples (2,980/26,587) and were significantly associated with mutations in KRAS, but mutually exclusive with EGFR (q < 0.1). KRAS mutational status constrained the ATM co-mutation landscape, with strong mutual exclusivity with TP53 and KEAP1 within KRAS-mutated samples. Those ATM mutations that co-occurred with TP53 were more likely to be missense mutations and associate with high mutational burden, suggestive of non-functional passenger mutations. In the CGDB cohort, dysfunctional ATM mutations associated with improved OS only in patients treated with ICI-chemotherapy, and not ICI alone. In vitro analyses demonstrated enhanced upregulation of STING signaling in ATM knockout cells with the addition of chemotherapy.

CONCLUSIONS

ATM mutations define a distinct subset of NSCLC associated with KRAS mutations, increased TMB, decreased TP53 and EGFR co-occurrence, and potential increased sensitivity to ICIs in the context of DNA-damaging chemotherapy.

Clinical and Molecular Features of Long-term Response to Immune Checkpoint Inhibitors in Patients with Advanced Non-Small Cell Lung Cancer

PURPOSE

We sought to identify features of patients with advanced non-small cell lung cancer (NSCLC) who achieve long-term response (LTR) to immune checkpoint inhibitors (ICI), and how these might differ from features predictive of short-term response (STR).

EXPERIMENTAL DESIGN

We performed a multicenter retrospective analysis of patients with advanced NSCLC treated with ICIs between 2011 and 2022. LTR and STR were defined as response ≥ 24 months and response < 12 months, respectively. Tumor programmed death ligand 1 (PD-L1) expression, tumor mutational burden (TMB), next-generation sequencing (NGS), and whole-exome sequencing (WES) data were analyzed to identify characteristics enriched in patients achieving LTR compared with STR and non-LTR.

RESULTS

Among 3,118 patients, 8% achieved LTR and 7% achieved STR, with 5-year overall survival (OS) of 81% and 18% among LTR and STR patients, respectively. High TMB (≥50th percentile) enriched for LTR compared with STR (P = 0.001) and non-LTR (P < 0.001). Whereas PD-L1 ≥ 50% enriched for LTR compared with non-LTR (P < 0.001), PD-L1 ≥ 50% did not enrich for LTR compared with STR (P = 0.181). Nonsquamous histology (P = 0.040) and increasing depth of response [median best overall response (BOR) -65% vs. -46%, P < 0.001] also associated with LTR compared with STR; no individual genomic alterations were uniquely enriched among LTR patients.

CONCLUSIONS

Among patients with advanced NSCLC treated with ICIs, distinct features including high TMB, nonsquamous histology, and depth of radiographic improvement distinguish patients poised to achieve LTR compared with initial response followed by progression, whereas high PD-L1 does not.

SwarmDeepSurv: swarm intelligence advances deep survival network for prognostic radiomics signatures in four solid cancers

Survival models exist to study relationships between biomarkers and treatment effects. Deep learning-powered survival models supersede the classical Cox proportional hazards (CoxPH) model, but substantial performance drops were observed on high-dimensional features because of irrelevant/redundant information. To fill this gap, we proposed SwarmDeepSurv by integrating swarm intelligence algorithms with the deep survival model. Furthermore, four objective functions were designed to optimize prognostic prediction while regularizing selected feature numbers. When testing on multicenter sets (n = 1,058) of four different cancer types, SwarmDeepSurv was less prone to overfitting and achieved optimal patient risk stratification compared with popular survival modeling algorithms. Strikingly, SwarmDeepSurv selected different features compared with classical feature selection algorithms, including the least absolute shrinkage and selection operator (LASSO), with nearly no feature overlapping across these models. Taken together, SwarmDeepSurv offers an alternative approach to model relationships between radiomics features and survival endpoints, which can further extend to study other input data types including genomics.

MCT4-dependent lactate secretion suppresses antitumor immunity in LKB1-deficient lung adenocarcinoma

Inactivating STK11/LKB1 mutations are genomic drivers of primary resistance to immunotherapy in KRAS-mutated lung adenocarcinoma (LUAD), although the underlying mechanisms remain unelucidated. We find that LKB1 loss results in enhanced lactate production and secretion via the MCT4 transporter. Single-cell RNA profiling of murine models indicates that LKB1-deficient tumors have increased M2 macrophage polarization and hypofunctional T cells, effects that could be recapitulated by the addition of exogenous lactate and abrogated by MCT4 knockdown or therapeutic blockade of the lactate receptor GPR81 expressed on immune cells. Furthermore, MCT4 knockout reverses the resistance to PD-1 blockade induced by LKB1 loss in syngeneic murine models. Finally, tumors from STK11/LKB1 mutant LUAD patients demonstrate a similar phenotype of enhanced M2-macrophages polarization and hypofunctional T cells. These data provide evidence that lactate suppresses antitumor immunity and therapeutic targeting of this pathway is a promising strategy to reversing immunotherapy resistance in STK11/LKB1 mutant LUAD.

Molecular markers of metastatic disease in KRAS-mutant lung adenocarcinoma

BACKGROUND

Prior studies characterized the association of molecular alterations with treatment-specific outcomes in KRAS-mutant (KRASMUT) lung adenocarcinoma (LUAD). Less is known about the prognostic role of molecular alterations and their associations with metastatic disease.

PATIENTS AND METHODS

We analyzed clinicogenomic data from 1817 patients with KRASMUT LUAD sequenced at the Dana-Farber Cancer Institute (DFCI) and Memorial Sloan Kettering Cancer Center (MSKCC). Patients with metastatic (M1) and nonmetastatic (M0) disease were compared. Transcriptomic data from The Cancer Genome Atlas (TCGA) were investigated to characterize the biology of differential associations with clinical outcomes. Organ-specific metastasis was associated with overall survival (OS).

RESULTS

KEAP1 (DFCI: OR = 2.3, q = 0.04; MSKCC: OR = 2.2, q = 0.00027) and SMARCA4 mutations (DFCI: OR = 2.5, q = 0.06; MSKCC: OR = 2.6, q = 0.0021) were enriched in M1 versus M0 tumors. On integrative modeling, NRF2 activation was the genomic feature most associated with OS. KEAP1 mutations were enriched in M1 versus M0 tumors independent of STK11 status (KEAP1MUT/STK11WT: DFCI OR = 3.0, P = 0.0064; MSKCC OR = 2.0, P = 0.041; KEAP1MUT/STK11MUT: DFCI OR = 2.3, P = 0.0063; MSKCC OR = 2.5, P = 3.6 × 10-05); STK11 mutations without KEAP1 loss were not associated with stage (KEAP1WT/STK11MUT: DFCI OR = 0.97, P = 1.0; MSKCC OR = 1.2, P = 0.33) or outcome. KEAP1/KRAS-mutated tumors with and without STK11 mutations exhibited high functional STK11 loss. The negative effects of KEAP1 were compounded in the presence of bone (HR = 2.3, P = 4.4 × 10-14) and negated in the presence of lymph node metastasis (HR = 1.0, P = 0.91).

CONCLUSIONS

Mutations in KEAP1 and SMARCA4, but not STK11, were associated with metastatic disease and poor OS. Functional STK11 loss, however, may contribute to poor outcomes in KEAP1MUT tumors. Integrating molecular data with clinical and metastatic-site annotations can more accurately risk stratify patients.

Predicting benefit from immune checkpoint inhibitors in patients with non-small-cell lung cancer by CT-based ensemble deep learning: a retrospective study

BACKGROUND

Only around 20-30% of patients with non-small-cell lung cancer (NCSLC) have durable benefit from immune-checkpoint inhibitors. Although tissue-based biomarkers (eg, PD-L1) are limited by suboptimal performance, tissue availability, and tumour heterogeneity, radiographic images might holistically capture the underlying cancer biology. We aimed to investigate the application of deep learning on chest CT scans to derive an imaging signature of response to immune checkpoint inhibitors and evaluate its added value in the clinical context.

METHODS

In this retrospective modelling study, 976 patients with metastatic, EGFR/ALK negative NSCLC treated with immune checkpoint inhibitors at MD Anderson and Stanford were enrolled from Jan 1, 2014, to Feb 29, 2020. We built and tested an ensemble deep learning model on pretreatment CTs (Deep-CT) to predict overall survival and progression-free survival after treatment with immune checkpoint inhibitors. We also evaluated the added predictive value of the Deep-CT model in the context of existing clinicopathological and radiological metrics.

FINDINGS

Our Deep-CT model demonstrated robust stratification of patient survival of the MD Anderson testing set, which was validated in the external Stanford set. The performance of the Deep-CT model remained significant on subgroup analyses stratified by PD-L1, histology, age, sex, and race. In univariate analysis, Deep-CT outperformed the conventional risk factors, including histology, smoking status, and PD-L1 expression, and remained an independent predictor after multivariate adjustment. Integrating the Deep-CT model with conventional risk factors demonstrated significantly improved prediction performance, with overall survival C-index increases from 0·70 (clinical model) to 0·75 (composite model) during testing. On the other hand, the deep learning risk scores correlated with some radiomics features, but radiomics alone could not reach the performance level of deep learning, indicating that the deep learning model effectively captured additional imaging patterns beyond known radiomics features.

INTERPRETATION

This proof-of-concept study shows that automated profiling of radiographic scans through deep learning can provide orthogonal information independent of existing clinicopathological biomarkers, bringing the goal of precision immunotherapy for patients with NSCLC closer.

FUNDING

National Institutes of Health, Mark Foundation Damon Runyon Foundation Physician Scientist Award, MD Anderson Strategic Initiative Development Program, MD Anderson Lung Moon Shot Program, Andrea Mugnaini, and Edward L C Smith.

Genomic and transcriptomic analysis of checkpoint blockade response in advanced non-small cell lung cancer

Anti-PD-1/PD-L1 agents have transformed the treatment landscape of advanced non-small cell lung cancer (NSCLC). To expand our understanding of the molecular features underlying response to checkpoint inhibitors in NSCLC, we describe here the first joint analysis of the Stand Up To Cancer-Mark Foundation cohort, a resource of whole exome and/or RNA sequencing from 393 patients with NSCLC treated with anti-PD-(L)1 therapy, along with matched clinical response annotation. We identify a number of associations between molecular features and outcome, including (1) favorable (for example, ATM altered) and unfavorable (for example, TERT amplified) genomic subgroups, (2) a prominent association between expression of inducible components of the immunoproteasome and response and (3) a dedifferentiated tumor-intrinsic subtype with enhanced response to checkpoint blockade. Taken together, results from this cohort demonstrate the complexity of biological determinants underlying immunotherapy outcomes and reinforce the discovery potential of integrative analysis within large, well-curated, cancer-specific cohorts.

Machine Learning Models for the Identification of Prognostic and Predictive Cancer Biomarkers: A Systematic Review

The identification of biomarkers plays a crucial role in personalized medicine, both in the clinical and research settings. However, the contrast between predictive and prognostic biomarkers can be challenging due to the overlap between the two. A prognostic biomarker predicts the future outcome of cancer, regardless of treatment, and a predictive biomarker predicts the effectiveness of a therapeutic intervention. Misclassifying a prognostic biomarker as predictive (or vice versa) can have serious financial and personal consequences for patients. To address this issue, various statistical and machine learning approaches have been developed. The aim of this study is to present an in-depth analysis of recent advancements, trends, challenges, and future prospects in biomarker identification. A systematic search was conducted using PubMed to identify relevant studies published between 2017 and 2023. The selected studies were analyzed to better understand the concept of biomarker identification, evaluate machine learning methods, assess the level of research activity, and highlight the application of these methods in cancer research and treatment. Furthermore, existing obstacles and concerns are discussed to identify prospective research areas. We believe that this review will serve as a valuable resource for researchers, providing insights into the methods and approaches used in biomarker discovery and identifying future research opportunities.

Abstract 964: Genomic and clinical predictors of early disease progression and chemoimmunotherapy benefit in advanced NSCLC

Background: Immune checkpoint inhibitors (ICIs) as monotherapy (ICI-mono) or with chemotherapy (ICI-chemo) are standard first-line treatment in NSCLC patients lacking targetable driver mutations. Biomarkers to identify patients at risk for early progression on ICI-mono or those who would maximally benefit from upfront ICI-chemo have not been defined.

Methods: We queried the GEMINI database to identify metastatic NSCLC patients without targetable EGFR/ALK alterations who were treated with ICI-mono or ICI-chemo. Mutational profiling was performed on tissue or blood using targeted NGS. Outcome measures were defined as clinical progression free survival (PFS) or early progressive disease (PD) rate (defined as rate of 3-month progression), and their association with variables was assessed via Cox Proportional Hazards regression (PFS) or logistic regression (early PD). Predictive deep learning models were used to integrate clinicopathological factors and genomic profile.

Results: 735 patients were included in this study, 269 treated with ICI-chemo and 466 with ICI-mono; 446 were treated in the first-line setting. TP53 was the most frequently altered gene (60%), followed by KRAS (37%), AR (21%), and STK11 (19%). In ICI-mono patients, alterations in STK11, ERBB2, ARID1A and CDK6 were associated with a higher likelihood of early PD; only STK11 was associated with early PD (29% vs 17%, P = 0.04) on ICI-chemo. In all patients, low PD-L1 expression and high disease burden (stage IVb and liver metastases) associated with early PD, but there were borderline significant treatment effects in favor of ICI-chemo in never smokers and patients with liver metastases and stage IVb. Shorter PFS was observed in the ICI-chemo group who had CDKN2A alterations vs wild type (median PFS: 5.1 vs 9.0 months; HR: 1.72; P = 0.01). A subgroup analysis of patients with CDKN2A alterations demonstrated preferentially worse outcomes in ICI-chemo compared to ICI-mono, with the best PFS achieved in the ICI-mono treated patients with CDKN2A point mutation. Integration of clinicogenomic features into a multivariate model with feature selection to predict early PD demonstrated a predictive performance of AUC 0.73 (vs PD-L1 alone, AUC 0.60) in the ICI-mono group, driven by liver metastases, stage IVb disease, PD-L1 expression, and STK11 alterations. These features were less predictive in ICI-chemo-treated patients, indicating a protective effect against early PD in these patients from combination chemoimmunotherapy.

Conclusions: Low PD-L1, high disease burden, and STK11 alterations are markers of early PD on ICI-mono, and patients with these features may particularly benefit from upfront combination treatment with ICI-chemo to protect against early progression.

Citation Format: Lingzhi Hong, Muhammad Aminu, Xuetao Lu, Maliazurina B. Saad, Pingjun Chen, Waree Rinsurongkawong, Amy Spelman, Yasir Y. Elamin, Marcelo V. Negrao, Ferdinandos Skoulidis, Carl M. Gay, Tina Cascone, Mara B. Antonoff, Boris Sepesi, Jeff Lewis, Don L. Gibbons, Ara A. Vaporciyan, Xiuning Le, J.Jack Lee, Sinchita Roy-Chowdhuri, Mark J. Routbort, John V. Heymach, Jia Wu, Jianjun Zhang, Natalie I. Vokes. Genomic and clinical predictors of early disease progression and chemoimmunotherapy benefit in advanced NSCLC [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 964.

Prediction Model for Tumor Volume Nadir in EGFR -mutant NSCLC Patients Treated With EGFR Tyrosine Kinase Inhibitors

PURPOSE

In patients with advanced non-small cell lung cancer (NSCLC) and oncogenic driver mutations treated with effective targeted therapy, a characteristic pattern of tumor volume dynamics with an initial regression, nadir, and subsequent regrowth is observed on serial computed tomography (CT) scans. We developed and validated a linear model to predict the tumor volume nadir in EGFR -mutant advanced NSCLC patients treated with EGFR tyrosine kinase inhibitors (TKI).

MATERIALS AND METHODS

Patients with EGFR -mutant advanced NSCLC treated with EGFR-TKI as their first EGFR-directed therapy were studied for CT tumor volume kinetics during therapy, using a previously validated CT tumor measurement technique. A linear regression model was built to predict tumor volume nadir in a training cohort of 34 patients, and then was validated in an independent cohort of 84 patients.

RESULTS

The linear model for tumor nadir prediction was obtained in the training cohort of 34 patients, which utilizes the baseline tumor volume before initiating therapy (V 0 ) to predict the volume decrease (mm 3 ) when the nadir volume (V p ) was reached: V 0 -V p =0.717×V 0 -1347 ( P =2×10 -16 ; R2 =0.916). The model was tested in the validation cohort, resulting in the R2 value of 0.953, indicating that the prediction model generalizes well to another cohort of EGFR -mutant patients treated with EGFR-TKI. Clinical variables were not significant predictors of tumor volume nadir.

CONCLUSION

The linear model was built to predict the tumor volume nadir in EGFR -mutant advanced NSCLC patients treated with EGFR-TKIs, which provide an important metrics in treatment monitoring and therapeutic decisions at nadir such as additional local abrasive therapy.

Efficacy of immunotherapy in oncogene-driven non-small-cell lung cancer

For advanced metastatic non-small-lung cancer, the landscape of actionable driver alterations is rapidly growing, with nine targetable oncogenes and seven approvals within the last 5 years. This accelerated drug development has expanded the reach of targeted therapies, and it may soon be that a majority of patients with lung adenocarcinoma will be eligible for a targeted therapy during their treatment course. With these emerging therapeutic options, it is important to understand the existing data on immune checkpoint inhibitors (ICIs), along with their efficacy and safety for each oncogene-driven lung cancer, to best guide the selection and sequencing of various therapeutic options. This article reviews the clinical data on ICIs for each of the driver oncogene defined lung cancer subtypes, including efficacy, both for ICI as monotherapy or in combination with chemotherapy or radiation; toxicities from ICI/targeted therapy in combination or in sequence; and potential strategies to enhance ICI efficacy in oncogene-driven non-small-cell lung cancers.

Association of High Tumor Mutation Burden in Non-Small Cell Lung Cancers With Increased Immune Infiltration and Improved Clinical Outcomes of PD-L1 Blockade Across PD-L1 Expression Levels

Importance

Although tumor mutation burden (TMB) has been explored as a potential biomarker of immunotherapy efficacy in solid tumors, there still is a lack of consensus about the optimal TMB threshold that best discriminates improved outcomes of immune checkpoint inhibitor therapy among patients with non-small cell lung cancer (NSCLC).

Objectives

To determine the association between increasing TMB levels and immunotherapy efficacy across clinically relevant programmed death ligand-1 (PD-L1) levels in patients with NSCLC.

Design, Setting, and Participants

This multicenter cohort study included patients with advanced NSCLC treated with immunotherapy who received programmed cell death-1 (PD-1) or PD-L1 inhibition in the Dana-Farber Cancer Institute (DFCI), Memorial Sloan Kettering Cancer Center (MSKCC), and in the Stand Up To Cancer (SU2C)/Mark Foundation data sets. Clinicopathological and genomic data were collected from patients between September 2013 and September 2020. Data analysis was performed from November 2021 to February 2022.

Exposures

Treatment with PD-1/PD-L1 inhibition without chemotherapy.

Main Outcomes and Measures

Association of TMB levels with objective response rate (ORR), progression-free survival (PFS), and overall survival (OS).

Results

In the entire cohort of 1552 patients with advanced NSCLC who received PD-1/PD-L1 blockade, the median (range) age was 66 (22-92) years, 830 (53.5%) were women, and 1347 (86.8%) had cancer with nonsquamous histologic profile. A regression tree modeling ORR as a function of TMB identified 2 TMB groupings in the discovery cohort (MSKCC), defined as low TMB (≤19.0 mutations per megabase) and high TMB (>19.0 mutations per megabase), which were associated with increasing improvements in ORR, PFS, and OS in the discovery cohort and in 2 independent cohorts (DFCI and SU2C/Mark Foundation). These TMB levels also were associated with significant improvements in outcomes of immunotherapy in each PD-L1 tumor proportion score subgroup of less than 1%, 1% to 49%, and 50% or higher. The ORR to PD-1/PD-L1 inhibition was as high as 57% in patients with high TMB and PD-L1 expression 50% or higher and as low as 8.7% in patients with low TMB and PD-L1 expression less than 1%. Multiplexed immunofluorescence and transcriptomic profiling revealed that high TMB levels were associated with increased CD8-positive, PD-L1-positive T-cell infiltration, increased PD-L1 expression on tumor and immune cells, and upregulation of innate and adaptive immune response signatures.

Conclusions and Relevance

These findings suggest that increasing TMB levels are associated with immune cell infiltration and an inflammatory T-cell-mediated response, resulting in increased sensitivity to PD-1/PD-L1 blockade in NSCLC across PD-L1 expression subgroups.

Abstract 2181: Genomic correlates of Metastasis in KRAS mutant lung adenocarcinoma

Background: Lung adenocarcinoma (LUAD) is a disease classified by molecular markers. In KRAS-mutant LUAD, STK11 and KEAP1 mutations are associated with decreased overall survival (OS), but predictors of metastasis have been poorly defined. In this study, we identify clinical and genomic predictors of metastatic KRAS-mutant LUAD.

Methods: Patients with KRAS-mutant LUAD profiled by targeted next generation sequencing (OncoPanel) were included. Stage, histology, recurrence-free and overall survival were assessed. Clinical and genomic features between metastatic vs non-metastatic samples were compared. KRAS-mutant LUAD samples profiled using MSK-IMPACT in the AACR GENIE database were used to validate our findings. Continuous variables were compared using the Mann-Whitney U test and categorical variables were compared using the Fisher’s Exact test. Survival analysis was performed using the Cox Proportional Hazards model. WExT was used to identify co-occurring and mutually exclusive genetic alterations. Benjamini-Hochberg was used to correct for multiple comparisons. P-values &lt; 0.05 and q-values &lt; 0.10 were considered significant.

Results: In the OncoPanel cohort (metastatic n=290; non-metastatic n=324), tumor mutational burden (TMB) (p = .001) and KEAP1 mutations (q = 0.05) were enriched in metastatic samples, while NFKBIA amplifications (q = 0.07) were enriched in non-metastatic samples. KEAP1/STK11 mutations significantly co-occurred (q &lt; 1e-8). Compared to double wild-type samples: KEAP1/STK11 co-mutations were significantly enriched in metastatic samples (n = 72, p = 0.0002, OR 3.4); KEAP1-mutant samples trended towards enrichment in metastatic samples, (n = 21,p = 0.07, OR 2.47); STK11 mutations did not associate with stage (n = 53, p = 0.88, OR = 0.94). In multivariable survival analysis, metastasis (p &lt; 0.005), KEAP1 mutation (p=0.01), and STK11 mutation (p=0.02) were associated with worse OS.

In the MSK-IMPACT validation cohort (metastatic site n=417, primary site n = 781), KEAP1 was the only gene enriched in metastatic samples (q &lt; 0.001) at q &lt; 0.05. Compared to double wild type samples: KEAP/STK11 co-mutations (n=138, p &lt; 0.0001, OR 2.1) and KEAP1 mutations (n=59, p = 0.04, OR 1.77) were enriched in metastatic samples; STK11-mutations did not associate with metastasis (n = 190, p = 0.34, OR 0.83). Other predictors of metastasis included Fraction Genome Altered (FGA) (p &lt; 1e-5), TMB (p &lt; 1e-5), and CDKN2A/B deletions (q &lt; 0.003).

Conclusion: While both KEAP1 and STK11 mutations are associated with decreased OS in KRAS-mutant LUAD, we find in two independent cohorts that only KEAP1 mutations and KEAP1/STK11 co-mutations, but not STK11 mutations, are associated with metastasis. We also found that FGA, TMB, CDKN2A/B deletions are strongly associated with metastasis. Further research is necessary to understand the influence of KEAP1 mutations, independent of and in-conjunction with STK11 mutations, on metastasis.

Citation Format: Daniel Boiarsky, Christine A. Lydon, Emily Chambers, Pasi A. Janne, Mark M. Awad, Eliezer M. Van Allen, David Barbie, Natalie I. Vokes. Genomic correlates of Metastasis in KRAS mutant lung adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2181.

Limited benefit from the addition of immunotherapy to chemotherapy in TKI-refractory EGFR-mutant lung adenocarcinoma

e21169

Background: The benefit of adding immunotherapy to chemotherapy in EGFR-mutant lung adenocarcinoma (LUAD) with acquired resistance to EGFR tyrosine kinase inhibitors (TKIs) is not fully investigated. This study aimed to explore the outcomes of immunotherapy combinations in such population in a real-world setting. Methods: This retrospective analysis included patients with EGFR-mutant LUAD who progressed on EGFR TKIs and received subsequent therapy with chemotherapy (chemo; n = 84), chemotherapy with immunotherapy (chemoIO; n = 30), chemotherapy plus bevacizumab with or without IO (withBev; n = 42), and IO monotherapy (IO-mono; n = 22). Clinical progression-free survival (cPFS) from the start of the therapy to discontinuation was assessed. Associations of clinical characteristics with cPFS were evaluated using univariable and adjusted Cox regression models. Results: A total of 178 patients (mean age = 62.6; 57.9% females) with a median cPFS (mcPFS) of 4.9 (95% CI: 4.4-5.4) months with 162 (91%) progression events. There was no difference in cPFS between chemoIO vs chemo groups (5.3 months vs 4.8, P = 0.8). Compared to the chemo group, patients who were treated withBev trended towards better PFS (6.1 months vs 4.8; P = 0.3; HR 0.79; 95% CI: 0.52-1.20;), while patients with IO-mono had inferior PFS (2.2 months; P = 0.001; HR 2.22; 95% CI: 1.37-3.59). Patients who were administered one-line of TKI prior to subsequent therapy acquired better PFS when compared to those with multiple prior TKIs (5.2 months vs 4.5; P = 0.011; HR 0.66; 95% CI: 0.48-0.91). The significance remained the same after adjusted by other clinicopathological variables including post-TKI treatment strategies and metastatic status (P = 0.005). Furthermore, among the chemoIO group, patients with PD-L1 expression had better PFS (n = 15, 7.7 months vs 4.5, P = 0.04; HR 0.36; 95% CI: 0.13-0.96) than those without PD-L1 expression (n = 6). However, PD-L1 was not associated with PFS among the patients treated with chemoIO plus bevacizumab. Brain or liver metastases were associated with worse PFS. Conclusions: Immunotherapy adds limited benefit when combined with chemotherapy in TKI-refractory EGFR-mutant LUAD. Chemotherapy +/- bevacizumab are reasonable treatment options for these patients.

Auto-reactive antibodies as predictive markers for immune checkpoint–induced pneumonitis

2554

Background: Certain immune-related adverse events (irAEs) that emerge with immune checkpoint blockade share clinical features of autoimmune conditions. Preexisting auto-reactive antibodies and their contribution to irAEs have not been well defined, and observations are limited. Methods: We longitudinally collected patient plasma samples from a clinical trial that combines immune checkpoint inhibitors, Ipilimumab, and Nivolumab (I+N) with subsequent radiation therapy (Lonestar, NCT03391869). Plasma samples were collected at baseline, after 12 weeks of I+N (induction), and at the time of Grade ≥ 2 pneumonitis (CTCAEv5.0). Auto reactive antibody profiles were analyzed using a fluorescence-based assay system that measures more than 130 antigens and is capable of assaying antibody reactivity for IgG and IgM fractions, including nuclear-cytosolic and tissue-specific antigens. Selected antibodies had a reportable result range, reference intervals, and reproducibility with quality controls. A paired t-test was used to compare the mean of longitudinally collected baseline and toxicity samples. An unpaired t-test was used to compare differences between groups. The False Discovery Rate was used to control the Type I error rate of multiple comparisons. Results: In the study cohort, G≥2 pneumonitis was observed in 11 patients out of 194 (5.6%). Serum was collected at baseline for all 11 patients, and 9 of the 11 patients had a serum sample collected at the time of pneumonitis event. Longitudinal serum samples (baseline and post-induction) collected from 32 patients without any irAEs were used as control. At baseline AChR3 and calmodulin antibodies were elevated in patients who developed pneumonitis, compared with baseline samples from controls (p≤0.05). At the time of pneumonitis IgM antibodies against AChR3, CXCL10, NSE, BAFF, CA242, Cytokeratin 19 were noted to be elevated in serum for pneumonitis cases compared with post induction samples from control (p≤0.005). Conclusions: We identified auto reactive antibodies associated with a higher risk of immunotherapy associated pneumonitis in patients treated with ipilimumab and nivolumab. These included auto reactive antibodies against proteins associated with lung injury (AChR3), lung inflammation (BAFF, CXCL10) and against alveolar epithelium (Cytokeratin 19). Future studies are warranted to determine if auto-reactive antibodies can be used as pre-treatment risk markers or to diagnose pneumonitis and may offer insights into to mechanisms that predispose toward pneumonitis.

Deep learning signature from chest CT and association with immunotherapy outcomes in EGFR/ALK-negative NSCLC

9061

Background: Many clinicopathological and molecular features are associate with clinical benefit from immune checkpoint inhibitors (ICIs) for patients with non-small-cell lung cancer (NSCLC), yet none was exclusive underscoring the heterogeneity of lung cancers. As images may provide a holistic view of cancer, we attempted deep learning to chest CT scans to derive a predictor of response to ICIs and test its benefit relative to known clinicopathological factors. Methods: 928 stage IV, EGFR/ALK-negative NSCLC patients treated with ICIs alone or in combination (MD Anderson GEMINI Database) were divided into training (CTtr = 572), validation (CTva = 78), and testing (CTte = 278) cohorts, balancing the distribution of clinicopathological and radiological factors. Progression-free (PFS) and overall survival (OS) were defined as outcomes. We analyzed whole lung, including tumor and normal parenchyma of chest CT images ≤ 3 months prior to ICI treatment. An ensemble learning model (CT-deep-learning) to clustering patients into high vs low risk groups of PFS or OS was developed by fusing risk scores from four independent deep learning networks (supervised, unsupervised, and hybrid). This CT-deep-learning model was further evaluated in different clinicopathological subgroups. Finally, a composite model (CT-Clinic-path) was built by combining image model with clinicopathological factors. Antolini's concordance index (C-index) was used to assess model performance. Results: Median PFS and OS were shorter in the high-risk vs low-risk group as defined by CT-deep-learning: PFS (CTtr: 4.2 vs 9.6 mons; HR 1.96; 95% CI 1.62-2.38; P < 0.0001; CTva: 3.7 vs 10.2 mons; HR 2.32; 95% CI 1.32-4.07; P = 0.0025; CTte: 3.6 vs 9.1 mons; HR 1.89; 95% CI 1.39-2.56; P < 0.0001) and OS (CTtr: 16.0 vs 31.4 mons; HR 2.19; 95% CI 1.72-2.79; P < 0.0001; CTva: 12.7 vs 28.6 mons; HR 2.01; 95% CI 1.04-3.88; P = 0.035; CTte: 14.8 vs 32.0 mons; HR 1.84; 95% CI 1.31-2.60; P = 0.0004). CT-deep-learning outperformed clinicopathologic features known to associate with ICI benefit, such as histology, smoking status, PD-L1 expression, and remained to be an independent (P < 0.001) prognostic factor on multivariate analysis. Furthermore, integrating CT-deep-learning to clinicopathological variables improved prediction performance with a net reclassification up to 7% (Clinic-path model, C-indices 0.60 – 0.62 vs CT-clinic-path model, 0.64 - 0.65 for PFS; Clinic-path model 0.64 – 0.67 vs CT-clinic-path model 0.69 – 0.71 for OS). Conclusions: We have developed and validated a deep learning signature associated with PFS and OS in ICI-treated NSCLC patients, which appears to be independent of and superior to known clinicopathological biomarkers. If validated, this signature may strengthen the predictive value of clinicopathological factors and facilitate selecting appropriate patients for ICI-based therapies.

Clinical outcome and potential benefits of post-progression immunotherapy for patients with metastatic NSCLC with primary resistance to ipilumumab and nivolumab in the LONESTAR phase III study

9049

Background: Primary resistance to immune checkpoint inhibitor (ICI) therapy remains a major challenge in clinical oncology. Here, we describe the clinical outcome of patients who experienced radiologic progression within 12 weeks of therapy with nivolumab and ipilimumab (I+N) for metastatic non-small cell lung cancer (mNSCLC). Methods: The LONESTAR study is an ongoing phase III study (NCT03391869). Study enrolls patients with immunotherapy naïve mNSCLC (prior chemotherapy is allowed). All patients receive I+N for 12 weeks and are randomized to experimental therapy vs. control arm if they did not have disease progression. Patients who experience radiologic progression per RECIST v1.1 are not randomized and removed from the study. Treatment beyond progression is allowed if they clinically benefit from the systemic therapy. We prospectively collected clinicopathologic and radiologic outcome data from patients who experienced radiologic progression within 12 weeks of I+N therapy and have not randomized to investigational therapy. We described the primary progression pattern. We collected subsequent treatment, radiologic, and toxicity data and calculated clinical outcomes, including progression-free survival (PFS) and overall survival (OS). Results: Of the 194 patients who received at least one dose of I+N therapy, 72 patients had clinical and/or radiologic progression at ≤ 12 weeks. Thirty-five (35; 48%) patients did not receive subsequent treatment, 21 (29%) patients received subsequent 2nd line systemic therapy, and 16 (22%) patients were continued on I+N beyond radiologic progression due to ongoing clinical benefit. Among patients treated with 2nd line therapy, 13 patients were treated with platinum doublet +/- anti-PD-(L)1, seven (7) patients were treated with single-agent chemotherapy +/- VEGF inhibitor, and one (1) patient was treated with targeted therapy. The PFS for the 2nd line therapy was 6.5 months (95%CI: 4.8, 8.9), and OS was 10.4 months (95%Cl: 6.6, 16.1). Among the 16 patients treated with I+N beyond progression, 13 had a mixed response to induction therapy, where primary progression was most frequently observed in mediastinal lymph nodes. LCT with radiotherapy was utilized with I+N in 10 patients. The median duration of post-progression treatment with I+N plus LCT was 8.7 months (95%Cl: 5.9, 22.3) and 5.6 months (95%Cl 4.4, 11.5) with I+N alone. The OS was 19.5 months (95% CI: 6.2,18.7). Conclusions: In this study cohort, primary resistance to I+N was observed in 37% of the patients, and in a subset of these patients treated with post-progression I+N, either alone or in combination with LCT, durable clinical benefit was observed. Further studies are warranted to identify which patients are most likely to benefit from post-progression I+N. Clinical trial information: NCT03391869.

Concurrent TP53 Mutations Facilitate Resistance Evolution in EGFR-Mutant Lung Adenocarcinoma

INTRODUCTION

Patients with EGFR-mutant NSCLC experience variable duration of benefit on EGFR tyrosine kinase inhibitors. The effect of concurrent genomic alterations on outcome has been incompletely described.

METHODS

In this retrospective study, targeted next-generation sequencing data were collected from patients with EGFR-mutant lung cancer treated at the Dana-Farber Cancer Institute. Clinical data were collected and correlated with somatic mutation data. Associations between TP53 mutation status, genomic features, and mutational processes were analyzed.

RESULTS

A total of 269 patients were identified for inclusion in the cohort. Among 185 response-assessable patients with pretreatment specimens, TP53 alterations were the most common event associated with decreased first-line progression-free survival and decreased overall survival, along with DNMT3A, KEAP1, and ASXL1 alterations. Reduced progression-free survival on later-line osimertinib in 33 patients was associated with MET, APC, and ERBB4 alterations. Further investigation of the effect of TP53 alterations revealed an association with worse outcomes even in patients with good initial radiographic response, and faster acquisition of T790M and other resistance mechanisms. TP53-mutated tumors had higher mutational burdens and increased mutagenesis with exposure to therapy and tobacco. Cell cycle alterations were not independently predictive, but portended worse OS in conjunction with TP53 alterations.

CONCLUSIONS

TP53 alterations associate with faster resistance evolution independent of mechanism in EGFR-mutant NSCLC and may cooperate with other genomic events to mediate acquisition of resistance mutations to EGFR tyrosine kinase inhibitors.

Real-world effectiveness of immune checkpoint inhibitors alone or in combination with chemotherapy in metastatic non–small cell lung cancer

9055

Background: The benefit of combination immune checkpoint inhibitor (ICI) with chemotherapy over ICI monotherapy in non-small cell lung cancer (NSCLC) remains underexplored. Methods: This retrospective cohort included patients with metastatic NSCLC from a single-institution database treated with ICI monotherapy or with chemotherapy between 1/2014-2/2020. Clinical progression-free survival (PFS) and overall survival (OS) were the primary outcomes. Propensity score adjustment for clinical and sociodemographic characteristics was used for analysis of first-line treatment outcomes. Results: A total of 1,139 patients (54% male; median age, 64.9) were included. Adenocarcinoma histology, smoking history, higher PD-L1 expression, and lower metastatic stage associated with improved PFS. However, PD-L1 expression and smoking associated with PFS only in adenocarcinoma (LUAD); squamous (LUSC) patients had shorter PFS independent of PD-L1 and smoking history (PD-L1 > 50% vs 1-49%: LUAD P < 0.001; LUSC P = 0.69; Former vs never smoker: LUAD P = 0.008; LUSC P = 0.89). In first-line patients (n = 680), treatment with ICI plus chemotherapy (ICI-chemo) associated with higher progression-free rates at 3 and 6 months compared with ICI-monotherapy (ICI-chemo vs ICI-mono: 3-month PFS, 85.2% vs 68.8%, P = 0.001; 6-month PFS, 66.4% vs 52.6%, P = 0.008). However, there was no difference overall in PFS or OS in either the full or propensity-matched cohort. Treatment with ICI and chemotherapy concurrently vs sequentially was associated with similar PFS (log-rank P = 0.12). Conclusions: In this real-world cohort, the addition of chemotherapy to ICIs may protect against early progression but does not influence long-term outcomes. Treatment with sequential vs concurrent ICI and chemotherapy produced similar outcomes. These findings suggest that combination therapy may maximally benefit patients at risk of early progression.

Integrative clinical and molecular characterization of translocation renal cell carcinoma

Translocation renal cell carcinoma (tRCC) is a poorly characterized subtype of kidney cancer driven by MiT/TFE gene fusions. Here, we define the landmarks of tRCC through an integrative analysis of 152 patients with tRCC identified across genomic, clinical trial, and retrospective cohorts. Most tRCCs harbor few somatic alterations apart from MiT/TFE fusions and homozygous deletions at chromosome 9p21.3 (19.2% of cases). Transcriptionally, tRCCs display a heightened NRF2-driven antioxidant response that is associated with resistance to targeted therapies. Consistently, we find that outcomes for patients with tRCC treated with vascular endothelial growth factor receptor inhibitors (VEGFR-TKIs) are worse than those treated with immune checkpoint inhibitors (ICI). Using multiparametric immunofluorescence, we find that the tumors are infiltrated with CD8⁺ T cells, though the T cells harbor an exhaustion immunophenotype distinct from that of clear cell RCC. Our findings comprehensively define the clinical and molecular features of tRCC and may inspire new therapeutic hypotheses.

The Role of Whole Exome Sequencing in Distinguishing Primary and Secondary Lung Cancers

Non-small cell lung cancer (NSCLC) that presents with multiple lung tumors (MLTs) poses a challenge to accurate staging and prognosis. MLTs that arise as clonally related secondary metastases from a common primary are higher stage and often require adjuvant chemotherapy or may in fact be incurable stage IV lesions. Conversely, MLTs that represent distinct primaries have a better prognosis and may be overtreated if inappropriately classified as related secondaries. Historically, pathologic and radiographic criteria were used to distinguish between primary and secondary MLTs; however, the advent of genomic profiling has demonstrated limitations to these historic classification systems. In this review, we discuss the use of molecular profiling to distinguish between primary and secondary lung cancers, with a focus on the insights gleaned from whole exome sequencing (WES) analyses. While WES is not yet feasible in routine clinical practice, WES studies have helped elucidate the clonal relationship between primary and secondary lung cancers and provide important context for the application of targeted sequencing panel-based analyses.

Tumor Growth Rate After Nadir Is Associated With Survival in Patients With EGFR-Mutant Non-Small-Cell Lung Cancer Treated With Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitor

PURPOSE

To investigate the association between tumor volume growth rate after the nadir and survival in patients with EGFR-mutant advanced non-small-cell lung cancer (NSCLC) treated with erlotinib.

MATERIALS AND METHODS

Seventy-one patients with EGFR-mutant advanced NSCLC treated with erlotinib were studied for computed tomography tumor volume kinetics during therapy. The tumor growth rate after nadir was obtained using a previously published analytic module for longitudinal volume tracking to study its relationship with overall survival (OS).

RESULTS

The median tumor volume for the cohort was 19,842 mm3 at baseline and 4,083 mm3 at nadir. The median time to nadir was 6.2 months. The tumor growth rate after nadir for logeV (the natural logarithm of tumor volume measured in mm3) was 0.11/mo on average for the cohort (SE: 0.014), which was very similar to the previously validated reference value of 0.12/mo to define slow and fast tumor growth. The OS of 48 patients with slow tumor growth (≤ 0.12/mo) was significantly longer compared with 23 patients with fast tumor growth (> 0.12/mo; median OS: 37.8 v 25.0 months; P = .0012). In Cox models, tumor growth rate was also associated with survival (regression coefficient: 3.9903; P = .0024; faster rate leads to increased hazards), after adjusting for time to nadir (regression coefficient: -0.0863; P = .0008; longer time to nadir leads to decreased hazards) and smoking history.

CONCLUSION

In patients with EGFR-mutant advanced NSCLC treated with erlotinib, slower tumor growth rates after nadir were associated with longer OS, providing a rationale for using tumor growth rates to guide precision therapy for lung cancer.

Tumor Growth Rate After Nadir Is Associated With Survival in Patients With EGFR-Mutant Non-Small-Cell Lung Cancer Treated With Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitor

PURPOSE

To investigate the association between tumor volume growth rate after the nadir and survival in patients with EGFR-mutant advanced non-small-cell lung cancer (NSCLC) treated with erlotinib.

MATERIALS AND METHODS

Seventy-one patients with EGFR-mutant advanced NSCLC treated with erlotinib were studied for computed tomography tumor volume kinetics during therapy. The tumor growth rate after nadir was obtained using a previously published analytic module for longitudinal volume tracking to study its relationship with overall survival (OS).

RESULTS

The median tumor volume for the cohort was 19,842 mm³ at baseline and 4,083 mm³ at nadir. The median time to nadir was 6.2 months. The tumor growth rate after nadir for log_eV (the natural logarithm of tumor volume measured in mm³) was 0.11/mo on average for the cohort (SE: 0.014), which was very similar to the previously validated reference value of 0.12/mo to define slow and fast tumor growth. The OS of 48 patients with slow tumor growth (≤ 0.12/mo) was significantly longer compared with 23 patients with fast tumor growth (> 0.12/mo; median OS: 37.8 v 25.0 months; P = .0012). In Cox models, tumor growth rate was also associated with survival (regression coefficient: 3.9903; P = .0024; faster rate leads to increased hazards), after adjusting for time to nadir (regression coefficient: -0.0863; P = .0008; longer time to nadir leads to decreased hazards) and smoking history.

CONCLUSION

In patients with EGFR-mutant advanced NSCLC treated with erlotinib, slower tumor growth rates after nadir were associated with longer OS, providing a rationale for using tumor growth rates to guide precision therapy for lung cancer.

Integrating molecular profiles into clinical frameworks through the Molecular Oncology Almanac to prospectively guide precision oncology

Tumor molecular profiling of single gene-variant ('first-order') genomic alterations informs potential therapeutic approaches. Interactions between such first-order events and global molecular features (for example, mutational signatures) are increasingly associated with clinical outcomes, but these 'second-order' alterations are not yet accounted for in clinical interpretation algorithms and knowledge bases. We introduce the Molecular Oncology Almanac (MOAlmanac), a paired clinical interpretation algorithm and knowledge base to enable integrative interpretation of multimodal genomic data for point-of-care decision making and translational-hypothesis generation. We benchmarked MOAlmanac to a first-order interpretation method across multiple retrospective cohorts and observed an increased number of clinical hypotheses from evaluation of molecular features and profile-to-cell line matchmaking. When applied to a prospective precision oncology trial cohort, MOAlmanac nominated a median of two therapies per patient and identified therapeutic strategies administered in 47% of patients. Overall, we present an open-source computational method for integrative clinical interpretation of individualized molecular profiles.

Abstract 22: A novel, inflamed small cell lung cancer transcriptional subtype, SCLC-I, defines a subset of patients with distinct immunotherapy vulnerability

Background

Small cell lung cancer (SCLC) is an aggressive neuroendocrine malignancy with dismal survival outcomes and no established predictive biomarkers. The landmark randomized, phase III IMpower133 trial established the new frontline standard of care for extensive-stage SCLC (ES-SCLC) as etoposide/platinum (EP) plus immune checkpoint blockade (ICB) [anti-PD-L1; atezolizumab (atezo)] based on an overall survival (OS) benefit compared to EP plus placebo. However, this survival benefit is limited in unselected populations, emphasizing the need for predictive biomarkers. Preclinically, there is emerging evidence of transcriptional heterogeneity among SCLC tumors, but the impact on therapeutic benefit remains undefined. Using non-negative matrix factorization (NMF) analysis of gene expression data from 81 SCLC tumors samples, we previously identified four subtypes, including three defined largely by differential expression of the transcription factors ASCL1 (SCLC-A), NEUROD1 (SCLC-N), and POU2F3 (SCLC-P), and a novel, fourth subtype with low expression of all three transcription factor signatures.

Method and Results

Using transcriptional and proteomic data from patient tumors and tumor-derived models, we molecularly characterized each of the four identified subtypes. The previously undescribed fourth subtype, dubbed SCLC-Inflamed (SCLC-I) showed high expression of non-neuroendocrine transcription factors (e.g. REST) and markers of EMT. Most distinctly, relative to the “cold” immune microenvironment typical of SCLC tumors, SCLC-I tumors possess markedly higher expression of interferon-γ signatures and immune checkpoints, including CD274 (PD-L1). Furthermore, cell type deconvolution using CIBERSORTx identified significantly higher infiltration into SCLC-I tumors by multiple immune cell types including T-cells, NK cells, macrophages, and dendritic cells. We predicted SCLC-I might derive disproportionate benefit from ICB due to its inflamed features. To test this, we applied our NMF-derived gene signature to 276 treatment-naïve, ES-SCLC patient tumors from the IMpower133 trial to assign patient subtype. The distribution of subtypes was as follows: SCLC-A 51%, SCLC-N 23%, SCLC-I 18% and SCLC-P 7%. While there was a trend toward OS benefit with the addition of atezo in each subtype, the benefit was numerically greater in SCLC-I. Specifically, median OS (atezo vs placebo arm) in months (mo) was 18.2 mo vs 10.4 mo for SCLC-I tumors, while median OS for the other three subtypes ranged from 9.6-10.9 mo (atezo arm) and 6.0-10.6 mo (placebo arm).

Conclusion

Unbiased transcriptional analyses identify four subtypes with distinct tumor and immune features. While all subtypes experienced improved OS with addition of anti-PD-L1 to frontline EP, SCLC-I patients appear to experience the most durable benefit.

Citation Format: Carl M. Gay, C. Allison Stewart, Lixia Diao, Barzin Y. Nabet, Junya Fujimoto, Luisa M. Solis, Wei Lu, Yuanxin Xi, Robert J. Cardnell, Natalie I. Vokes, Kavya Ramkumar, Stephen G. Swisher, Jack A. Roth, Bonnie S. Glisson, David S. Shames, Ignacio I. Wistuba, Jing Wang, John Minna, John V. Heymach, Lauren A. Byers. A novel, inflamed small cell lung cancer transcriptional subtype, SCLC-I, defines a subset of patients with distinct immunotherapy vulnerability [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 22.

Association of a very high tumor mutational load with increased CD8+ and PD-1+ T-cell infiltration and improved clinical outcomes to PD-(L)1 blockade across different PD-L1 expression levels in non-small cell lung cancer

9018

Background: Although high TMB correlates with improved outcomes to immune checkpoint inhibitors (ICI) in patients (pts) with non-small cell lung cancer (NSCLC), an optimal TMB cutoff to discriminate cancers most likely to respond to ICI has not been identified. Whether TMB impacts outcomes to ICI in different PD-L1 levels subgroups is also unclear. Methods: Unbiased recursive partitioning (URP) was used to identify an optimal TMB cutoff for objective response rate (ORR) in two independent cohorts of pts with NSCLC treated with ICI at DFCI and MSKCC. TCGA was interrogated to find differences in tumor immune cell subsets according to the TMB cutoff identified. Multiplexed immunofluorescence (IF) for CD8, PD-1, PD-L1, Foxp3, and CK7 was also performed on NSCLC samples at the DFCI. Results: In the DFCI (N=686) and MSKCC (N=672) cohorts, URP found an optimal TMB cutoff for ORR at 19 mutations/megabase (mut/Mb), corresponding to the ̃90th percentile in each cohort. Median progression-free (PFS) and overall survival (OS) were significantly longer in NSCLCs with TMB ≥19 mut/Mb vs <19 mut/Mb, in both cohorts (Table). After harmonizing TMB between DFCI OncoPanel and MSK-IMPACT NGS platforms, URP confirmed an optimal TMB cutoff for ORR at the 90th percentile in the combined cohort, which also associated with longer PFS/OS to ICI (Table). A TMB ≥90th percentile correlated with longer PFS/OS to ICI among NSCLCs with PD-L1 levels ≥50% and 1-49%, and longer PFS among those with PD-L1 <1% (Table). Cell subset transcriptome analysis from the TCGA showed higher proportions of CD8+ T cells (P=0.02) and M1 macrophages (P<0.01) among NSCLCs with a TMB ≥ vs <90th percentile. IF confirmed increased CD8+, CD8+ PD1+ T-cell infiltration (P<0.01), and increased CD8+/Foxp3+ ratio in NSCLC with very high TMB Conclusions: A very high TMB is associated with better outcomes to ICI and a distinct immunophenotype in NSCLC. Rational integration of TMB and PD-L1 expression may identify NSCLCs most likely to respond to ICI.[Table: see text]

Tumor and immune reprogramming during immunotherapy in advanced renal cell carcinoma

Immune checkpoint blockade (ICB) results in durable disease control in a subset of patients with advanced renal cell carcinoma (RCC), but mechanisms driving resistance are poorly understood. We characterize the single-cell transcriptomes of cancer and immune cells from metastatic RCC patients before or after ICB exposure. In responders, subsets of cytotoxic T cells express higher levels of co-inhibitory receptors and effector molecules. Macrophages from treated biopsies shift toward pro-inflammatory states in response to an interferon-rich microenvironment but also upregulate immunosuppressive markers. In cancer cells, we identify bifurcation into two subpopulations differing in angiogenic signaling and upregulation of immunosuppressive programs after ICB. Expression signatures for cancer cell subpopulations and immune evasion are associated with PBRM1 mutation and survival in primary and ICB-treated advanced RCC. Our findings demonstrate that ICB remodels the RCC microenvironment and modifies the interplay between cancer and immune cell populations critical for understanding response and resistance to ICB.

Patterns of transcription factor programs and immune pathway activation define four major subtypes of SCLC with distinct therapeutic vulnerabilities

Despite molecular and clinical heterogeneity, small cell lung cancer (SCLC) is treated as a single entity with predictably poor results. Using tumor expression data and non-negative matrix factorization, we identify four SCLC subtypes defined largely by differential expression of transcription factors ASCL1, NEUROD1, and POU2F3 or low expression of all three transcription factor signatures accompanied by an Inflamed gene signature (SCLC-A, N, P, and I, respectively). SCLC-I experiences the greatest benefit from the addition of immunotherapy to chemotherapy, while the other subtypes each have distinct vulnerabilities, including to inhibitors of PARP, Aurora kinases, or BCL-2. Cisplatin treatment of SCLC-A patient-derived xenografts induces intratumoral shifts toward SCLC-I, supporting subtype switching as a mechanism of acquired platinum resistance. We propose that matching baseline tumor subtype to therapy, as well as manipulating subtype switching on therapy, may enhance depth and duration of response for SCLC patients.

Transcriptional mediators of treatment resistance in lethal prostate cancer

Metastatic castration-resistant prostate cancer is typically lethal, exhibiting intrinsic or acquired resistance to second-generation androgen-targeting therapies and minimal response to immune checkpoint inhibitors¹. Cellular programs driving resistance in both cancer and immune cells remain poorly understood. We present single-cell transcriptomes from 14 patients with advanced prostate cancer, spanning all common metastatic sites. Irrespective of treatment exposure, adenocarcinoma cells pervasively coexpressed multiple androgen receptor isoforms, including truncated isoforms hypothesized to mediate resistance to androgen-targeting therapies^2,3. Resistance to enzalutamide was associated with cancer cell-intrinsic epithelial-mesenchymal transition and transforming growth factor-β signaling. Small cell carcinoma cells exhibited divergent expression programs driven by transcriptional regulators promoting lineage plasticity and HOXB5, HOXB6 and NR1D2 (refs. ^4-6). Additionally, a subset of patients had high expression of dysfunction markers on cytotoxic CD8⁺ T cells undergoing clonal expansion following enzalutamide treatment. Collectively, the transcriptional characterization of cancer and immune cells from human metastatic castration-resistant prostate cancer provides a basis for the development of therapeutic approaches complementing androgen signaling inhibition.

Abstract 5859: Genomic correlates of response to immune checkpoint inhibitors in advanced head and neck squamous cell carcinoma

Background:

Objective response rates for patients with advanced head and neck squamous cell carcinoma (HNSCC) receiving PD-1/L1 immune checkpoint inhibitors (ICI) without chemotherapy remain under 25%. Some studies suggest that HNSCC patients with human papillomavirus (HPV)-associated disease and high tumor mutational burden (TMB) derive greater benefit, although these findings do not yet guide patient selection. We hypothesized that uniform genomic analysis of a larger cohort may help place the predictive value of TMB in a broader biological context and identify more robust predictors of response.

Methods:

274 patients with biopsy-proven HNSCC treated with ICI at our institution from June 2014 to September 2018 were identified. Following targeted massively parallel sequencing, single nucleotide variant calling, and quality control measures, 127 samples were available for analysis. Clinical records were reviewed to determine radiographic response and overall survival. Responders included complete or partial responders by RECIST v1.1. Nonresponders included those with progressive or stable disease. We evaluated associations between genomic features and response using standard statistical methods.

Results:

TMB for the entire cohort was significantly higher among responders (p = 0.01, Fisher's); however, it had poor predictive power (AUROC = 0.68). Among the entire cohort, somatic mutations in NOTCH1 were enriched in responders (12/26 R, 17/101 NR, p = 0.003). After correcting for TMB, NOTCH1 remained associated with complete or partial response (p &lt; 0.05, logistic regression). After adjusting for mutations in NOTCH1, no significant difference in TMB was observed between responders and nonresponders (p &gt; 0.05, logistic regression). Among HPV-negative responders, somatic mutations in CDK12, a transcription-associated kinase whose loss is associated with genomic instability, neoantigen burden, and T cell infiltration, were enriched (3/26 R, 0/101 NR, p = 0.008, Fisher's).

Conclusion:

Through uniform genomic analysis of the largest ICI-treated HNSCC cohort to-date, we validated prior findings regarding the significance of TMB and NOTCH1 as correlates of response for ICI in HNSCC. We also show that TMB alone is insufficient as a predictor of response. We identify CDK12 as a variant with a mechanistic relationship to tumor immunology as a possible correlate of response in HPV-negative patients, which warrants further investigation.

Citation Format: Nisarg A. Patel, Natalie I. Vokes, Haitham Elmarakeby, Glenn J. Hanna, Eliezer M. Van Allen. Genomic correlates of response to immune checkpoint inhibitors in advanced head and neck squamous cell carcinoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5859.

Mammalian SWI/SNF Complex Genomic Alterations and Immune Checkpoint Blockade in Solid Tumors

Prior data have variably implicated the inactivation of the mammalian SWItch/Sucrose Non-Fermentable (mSWI/SNF) complex with increased tumor sensitivity to immune checkpoint inhibitors (ICI). Herein, we examined the association between mSWI/SNF variants and clinical outcomes to ICIs. We correlated somatic loss-of-function (LOF) variants in a predefined set of mSWI/SNF genes (ARID1A, ARID1B, SMARCA4, SMARCB1, PBRM1, and ARID2) with clinical outcomes in patients with cancer treated with systemic ICIs. We identified 676 patients from Dana-Farber Cancer Institute (DFCI, Boston, MA) and 848 patients from a publicly available database from Memorial Sloan Kettering Cancer Center (MSKCC, New York, NY) who met the inclusion criteria. Multivariable analyses were conducted and adjusted for available baseline factors and tumor mutational burden. Median follow-up was 19.6 (17.6-22.0) months and 28.0 (25.0-29.0) months for the DFCI and MSKCC cohorts, respectively. Seven solid tumor subtypes were examined. In the DFCI cohort, LOF variants of mSWI/SNF did not predict improved overall survival (OS), time-to-treatment failure (TTF), or disease control rate. Only patients with renal cell carcinoma with mSWI/SNF LOF showed significantly improved OS and TTF with adjusted HRs (95% confidence interval) of 0.33 (0.16-0.7) and 0.49 (0.27-0.88), respectively, and this was mostly driven by PRBM1 In the MSKCC cohort, where only OS was captured, LOF mSWI/SNF did not correlate with improved outcomes across any tumor subtype. We did not find a consistent association between mSWI/SNF LOF variants and improved clinical outcomes to ICIs, suggesting that mSWI/SNF variants should not be considered as biomarkers of response to ICIs.

Harmonization of Tumor Mutational Burden Quantification and Association With Response to Immune Checkpoint Blockade in Non-Small-Cell Lung Cancer

Purpose

Heterogeneity in tumor mutational burden (TMB) quantification across sequencing platforms limits the application and further study of this potential biomarker of response to immune checkpoint inhibitors (ICI). We hypothesized that harmonization of TMB across platforms would enable integration of distinct clinical datasets to better characterize the association between TMB and ICI response.

Methods

Cohorts of NSCLC patients sequenced by one of three targeted panels or by whole exome sequencing (WES) were compared (total n=7297). TMB was calculated uniformly and compared across cohorts. TMB distributions were harmonized by applying a normal transformation followed by standardization to z-scores. In sub-cohorts of patients treated with ICIs (DFCI n=272; MSKCC n=227), the association between TMB and outcome was assessed. Durable clinical benefit (DCB) was defined as responsive/stable disease lasting ≥6 months.

Results

TMB values were higher in the panel cohorts than the WES cohort. Average mutation rates per gene were highly concordant across cohorts (Pearson coefficient 0.842-0.866). Subsetting the WES cohort by gene panels only partially reproduced the observed differences in TMB. Standardization of TMB into z-scores harmonized TMB distributions and enabled integration of the ICI-treated sub-cohorts. Simulations indicated that cohorts >900 are necessary for this approach. TMB did not associate with response in never smokers or patients harboring targetable driver alterations, although these analyses were under-powered. Increasing TMB thresholds increased DCB rate, but DCB rates within deciles varied. Receiver operator curves yielded an area under the curve of 0.614 with no natural inflection point.

Conclusion

Z-score conversion harmonizes TMB values and enables integration of datasets derived from different sequencing panels. Clinical and biologic features may provide context to the clinical application of TMB, and warrant further study.

Tumor Evolution: A Problem of Histocompatibility

In this issue of Cell, two articles show that tumor-specific changes in HLA-mediated antigen presentation affect tumor immunogenicity and may play a role in shaping cancer cell survival.

Toward Molecularly Driven Precision Medicine in Lung Adenocarcinoma

<b/> Assessing the benefit of routine panel-based genomic sequencing of tumor tissue remains a critical need in clinical oncology. Jordan and coauthors report on 860 patients with metastatic or recurrent lung adenocarcinoma from a single institution with prospectively sequenced tumors using a targeted gene panel of >300 genes to guide therapy. Their results suggest that early prospective tumor sequencing, including non-standard-of-care predictive biomarkers combined with careful clinical annotation, can guide therapy, improve clinical outcomes, and accelerate the development of biomarkers and drugs. Cancer Discov; 7(6); 555-7. ©2017 AACRSee related article by Jordan et al., p. 596.

Tracing compartmentalized NADPH metabolism in the cytosol and mitochondria of mammalian cells

Eukaryotic cells compartmentalize biochemical processes in different organelles, often relying on metabolic cycles to shuttle reducing equivalents across intracellular membranes. NADPH serves as the electron carrier for the maintenance of redox homeostasis and reductive biosynthesis, with separate cytosolic and mitochondrial pools providing reducing power in each respective location. This cellular organization is critical for numerous functions but complicates analysis of metabolic pathways using available methods. Here we develop an approach to resolve NADP(H)-dependent pathways present within both the cytosol and the mitochondria. By tracing hydrogen in compartmentalized reactions that use NADPH as a cofactor, including the production of 2-hydroxyglutarate by mutant isocitrate dehydrogenase enzymes, we can observe metabolic pathway activity in these distinct cellular compartments. Using this system we determine the direction of serine/glycine interconversion within the mitochondria and cytosol, highlighting the ability of this approach to resolve compartmentalized reactions in intact cells.

Reductive glutamine metabolism is a function of the α-ketoglutarate to citrate ratio in cells

Reductively metabolized glutamine is a major cellular carbon source for fatty acid synthesis during hypoxia or when mitochondrial respiration is impaired. Yet, a mechanistic understanding of what determines reductive metabolism is missing. Here we identify several cellular conditions where the α-ketoglutarate/citrate ratio is changed due to an altered acetyl-CoA to citrate conversion, and demonstrate that reductive glutamine metabolism is initiated in response to perturbations that result in an increase in the α-ketoglutarate/citrate ratio. Thus, targeting reductive glutamine conversion for a therapeutic benefit might require distinct modulations of metabolite concentrations rather than targeting the upstream signalling, which only indirectly affects the process.

“Should I Give You My Smoking Lecture Now or Later?” Characterizing Emergency Physician Smoking Discussions and Cessation Counseling

STUDY OBJECTIVE

We determine frequency and manner in which emergency physicians address smoking with their patients.

METHODS

This was a descriptive secondary analysis of 871 audiotapes of physician-patient interactions collected during a trial assessing the effect of computer-based health risk assessment on emergency physician-patient communication. Consenting nonemergency female patients, ages 18 to 65 years, were enrolled from 2 socioeconomically diverse academic emergency department (EDs) for audiotaping of the ED visit. All audio files with any mention of smoking were independently coded with an in-depth structured coding form to characterize the nature of smoking-related discussions. Logistic regression was used to determine factors associated with emergency physician screening and discussion of tobacco exposure with women patients.

RESULTS

Overall, 484 of 871 (56%) participants were verbally screened for smoking, with 156 of 484 (32%) disclosing current smoking, with similar incidence at both sites. Tobacco screening was higher (odds ratio 2.2; 95% confidence interval 1.3 to 3.5), whereas rates of smoking-related discussions were lower (odds ratio 0.41; 95% confidence interval 0.17 to 0.98) at the urban site. At both sites, physicians tended to screen and discuss smoking when patients presented with a health condition that could be aggravated by smoking. Only 56% of discussions with current smokers contained advice to quit, 16% included assessment of readiness to quit, and a minority (13%) included a referral. Physician empathy/encouragement was associated with patients' detailing quit attempts.

CONCLUSION

Emergency physicians were likely to gather information about smoking but not to counsel or advise patients to quit. These results raise the question of whether emergency medicine resident training should include additional emphasis on smoking cessation counseling and motivational interviewing techniques.