Clinicopathologic and Genomic Factors Impacting Efficacy of First-Line Chemoimmunotherapy in Advanced NSCLC (original) (raw)

• Journal List
• HHS Author Manuscripts
• PMC10500613

J Thorac Oncol. Author manuscript; available in PMC 2023 Sep 14.

Published in final edited form as:

PMCID: PMC10500613

NIHMSID: NIHMS1923169

Joao V. Alessi, MD,a Arielle Elkrief, MD,b,c,d Biagio Ricciuti, MD,a Xinan Wang, PhD, MS,e Alessio Cortellini, MD,f,g Victor R. Vaz, MD,a Giuseppe Lamberti, MD,a Rosa L. Frias, MD,a Deepti Venkatraman, MPH,a Claudia A. M. Fulgenzi, MD,f,g Federica Pecci, MD,a Gonzalo Recondo, MD, PhD,a Alessandro Di Federico, MD,a Adriana Barrichello, MD,a Hyesun Park, MD,h,i Mizuki Nishino, MD,i Grace M. Hambelton, BA,j Jacklynn V. Egger, MD,c Marc Ladanyi, MD,c Subba Digumarthy, MD,k Bruce E. Johnson, MD,a David C. Christiani, MD,e Xihong Lin, PhD,l Justin F. Gainor, MD,j Jessica J. Lin, MD,j David J. Pinato, MD,f Adam J. Schoenfeld, MD,b and Mark M. Awad, MD, PhDa,*

Joao V. Alessi

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Arielle Elkrief

bThoracic Oncology Service, Memorial Sloan Kettering Cancer Center, New York, New York

cDepartment of Pathology, Memorial Sloan Kettering Cancer Center, New York, New York

dHuman Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York

Biagio Ricciuti

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Xinan Wang

eDepartment of Environmental Health, Harvard T.H. Chan School of Public Health, Harvard University, Boston, Massachusetts

Alessio Cortellini

fDivision of Cancer, Department of Surgery and Cancer, Hammersmith Hospital Campus, Imperial College London, London, United Kingdom

gDepartment of Medical Oncology, University Campus Bio-Medico of Rome, Italy

Victor R. Vaz

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Giuseppe Lamberti

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Rosa L. Frias

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Deepti Venkatraman

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Claudia A. M. Fulgenzi

fDivision of Cancer, Department of Surgery and Cancer, Hammersmith Hospital Campus, Imperial College London, London, United Kingdom

gDepartment of Medical Oncology, University Campus Bio-Medico of Rome, Italy

Federica Pecci

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Gonzalo Recondo

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Alessandro Di Federico

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Adriana Barrichello

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

Hyesun Park

hDepartment of Radiology, Lahey Hospital and Medical Center, Burlington, Massachusetts

iDepartment of Radiology, Brigham and Women’s Hospital and Department of Imaging, Dana-Farber Cancer Institute, Boston, Massachusetts

Mizuki Nishino

iDepartment of Radiology, Brigham and Women’s Hospital and Department of Imaging, Dana-Farber Cancer Institute, Boston, Massachusetts

Grace M. Hambelton

jCenter for Thoracic Cancers, Department of Medicine, Massachusetts General Hospital Cancer Center, Boston, Massachusetts

Jacklynn V. Egger

cDepartment of Pathology, Memorial Sloan Kettering Cancer Center, New York, New York

Marc Ladanyi

cDepartment of Pathology, Memorial Sloan Kettering Cancer Center, New York, New York

Subba Digumarthy

kDepartment of Radiology, Massachusetts General Hospital Cancer Center, Boston, Massachusetts

Bruce E. Johnson

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

David C. Christiani

eDepartment of Environmental Health, Harvard T.H. Chan School of Public Health, Harvard University, Boston, Massachusetts

Xihong Lin

lDepartment of Biostatistics, Harvard T.H. Chan School of Public Health, Harvard University, Boston, Massachusetts

Justin F. Gainor

jCenter for Thoracic Cancers, Department of Medicine, Massachusetts General Hospital Cancer Center, Boston, Massachusetts

Jessica J. Lin

jCenter for Thoracic Cancers, Department of Medicine, Massachusetts General Hospital Cancer Center, Boston, Massachusetts

David J. Pinato

fDivision of Cancer, Department of Surgery and Cancer, Hammersmith Hospital Campus, Imperial College London, London, United Kingdom

Adam J. Schoenfeld

bThoracic Oncology Service, Memorial Sloan Kettering Cancer Center, New York, New York

Mark M. Awad

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

aLowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts

bThoracic Oncology Service, Memorial Sloan Kettering Cancer Center, New York, New York

cDepartment of Pathology, Memorial Sloan Kettering Cancer Center, New York, New York

dHuman Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York

eDepartment of Environmental Health, Harvard T.H. Chan School of Public Health, Harvard University, Boston, Massachusetts

fDivision of Cancer, Department of Surgery and Cancer, Hammersmith Hospital Campus, Imperial College London, London, United Kingdom

gDepartment of Medical Oncology, University Campus Bio-Medico of Rome, Italy

hDepartment of Radiology, Lahey Hospital and Medical Center, Burlington, Massachusetts

iDepartment of Radiology, Brigham and Women’s Hospital and Department of Imaging, Dana-Farber Cancer Institute, Boston, Massachusetts

jCenter for Thoracic Cancers, Department of Medicine, Massachusetts General Hospital Cancer Center, Boston, Massachusetts

kDepartment of Radiology, Massachusetts General Hospital Cancer Center, Boston, Massachusetts

lDepartment of Biostatistics, Harvard T.H. Chan School of Public Health, Harvard University, Boston, Massachusetts

Dr. Alessi, Dr. Elkrief, and Dr. Ricciuti are listed as co-first authors.

Dr. Schoenfeld and Dr. Awad are listed as co-senior authors.

*Address for correspondence: Mark M. Awad, MD, PhD, Lowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, 450 Brookline Avenue, Dana 1240, Boston, MA 02215. ude.dravrah.icfd@dawa_kram

Supplementary Materials

Supplementary_methods.

GUID: 04081B7C-F3E1-47FB-9610-80D5A6E660D0

Supplementary_Figures.

GUID: 87AE0943-2258-4DD1-B594-EDDCCDE1E9EE

Table_S1.

GUID: AB054D92-7172-4747-B843-C41375738B7A

Table_S3A-B.

GUID: 432128F3-A0BC-4A32-8CA6-4D16BBA13BE8

Table_S2.

GUID: 4B678F76-DF3B-4585-98FD-F197C9E3A23B

Data Availability Statement

Data are available on reasonable request. The data that support the finding of our study are available on request from the corresponding author.

Abstract

Introduction:

Although programmed cell death protein 1 and programmed death-ligand 1 (PD-L1) blockade in combination with platinum-doublet chemotherapy has become a mainstay of first-line treatment for advanced NSCLC, factors associated with efficacy of chemoimmunotherapy (CIT) are not well characterized.

Methods:

In this multicenter retrospective analysis, clinicopathologic and genomic data were collected from patients with advanced NSCLC (lacking sensitizing genomic alterations in EGFR and ALK) and evaluated with clinical outcomes to first-line CIT.

Results:

Among 1285 patients treated with CIT, a worsening performance status and increasing derived neutrophil-to-lymphocyte ratio in the blood were associated with a significantly reduced objective response rate (ORR), median progression-free survival (mPFS), and median overall survival (mOS). With increasing PD-L1 tumor proportion scores of less than 1%, 1% to 49%, 50% to 89%, and greater than or equal to 90%, there was a progressive improvement in ORR (32.7% versus 37.5% versus 51.6% versus 61.7%, p < 0.001), mPFS (5.0 versus 6.1 versus 6.8 versus 13.0 mo, p < 0.001), and generally mOS (12.9 versus 14.6 versus 34.7 versus 23.1 mo, p = 0.009), respectively. Of 789 NSCLCs with comprehensive genomic data, NSCLCs with a tumor mutational burden (TMB) greater than or equal to the 90th percentile had an improved ORR (53.5% versus 36.4%, p = 0.004), mPFS (10.8 versus 5.5 mo, p < 0.001), and mOS (29.2 versus 13.1 mo, p < 0.001), compared with those with a TMB less than the 90th percentile. In all-comers with nonsquamous NSCLC, the presence of an STK11, KEAP1, or SMARCA4 mutation was associated with significantly worse ORR, mPFS, and mOS to CIT (all p < 0.05); this was also observed in the _KRAS_-mutant subgroup of NSCLCs with co-occurring mutations in STK11, KEAP1, or SMARCA4 (all p < 0.05). In KRAS wild-type NSCLC, KEAP1 and SMARCA4 mutations were associated with a significantly shorter mPFS and mOS to CIT (all p < 0.05), but STK11 mutation status had no significant impact on mPFS (p = 0.16) or mOS (p = 0.38).

Conclusions:

In advanced NSCLC, better patient performance status, low derived neutrophil-to-lymphocyte ratio, increasing PD-L1 expression, a very high TMB, and STK11/KEAP1/SMARCA4 wild-type status are associated with improved clinical outcomes to first-line CIT.

Keywords: NSCLC, Chemoimmunotherapy, First-line, PD-L1 expression, Tumor mutational burden, KRAS

Introduction

For patients with advanced NSCLC lacking targetable genomic alterations, first-line treatment options include immune checkpoint inhibitors (ICIs) alone1-5 or in combination with platinum-doublet chemotherapy.6-9 Having a deeper understanding of clinical, pathologic, and genomic features that influence outcomes to first-line therapy will improve regimen selection for each patient. For example, patients who are less likely to respond to ICIs alone may need treatment intensification with ICI plus chemotherapy, whereas those who are likely to benefit from ICIs alone may not need to be exposed to potential side effects of chemotherapy.

Recent studies have begun to elucidate the factors that affect the efficacy of programmed cell death protein 1 (PD-1) and programmed death-ligand 1 monotherapy in NSCLC.10-13 Patients with better performance status, a history of tobacco use, low peripheral blood neutrophil-to-lymphocyte ratios, and tumors14 with greater PD-L1 tumor proportion score (TPS) levels and very high tumor mutational burden (TMB) were found to have improved clinical outcomes to PD-(L)1 monotherapy.15-21 By contrast, patients with NSCLCs harboring mutations in STK11 or KEAP1 have worse outcomes to PD-(L)1 blockade, particularly among those with _KRAS_-mutant NSCLCs.12,14,22

Currently, little is known about predictors of outcomes to PD-(L)1 blockade in combination with platinum-doublet chemotherapy, hereafter referred to as chemoimmunotherapy (CIT). From the KEYNOTE-189 study of pembrolizumab in combination with platinum-doublet chemotherapy in nonsquamous NSCLC, overall survival improved with increasing PD-L1 TPS categories from less than 1% to 1% to 49% to greater than or equal to 50%, whereas TMB does not seem to affect outcomes to CIT.23,24 Nevertheless, because other factors associated with CIT efficacy have not been characterized in detail, we explored clinical outcomes to CIT from four large academic medical centers.

Materials and Methods

Participating academic medical centers included the Dana-Farber Cancer Institute (DFCI), the Memorial Sloan Kettering Cancer Center (MSKCC), the Imperial College London, and the Massachusetts General Hospital. Patients in this multicenter retrospective analysis were included if they had advanced NSCLC (lacking sensitizing genomic alterations in EGFR or ALK) and received at least one dose of first-line PD-(L)1 checkpoint blockade in combination with platinum-doublet chemotherapy. Detailed methods, including statistical analysis, methods used for PD-L1 and TMB assessment, clinicopathologic and genomic analysis, are reported in the Supplementary Methods.

Results

Clinicopathologic Factors Associated With Outcomes to CIT

A total of 1285 patients with advanced NSCLC lacking sensitizing EGFR or ALK alterations treated with first-line PD-(L)1 inhibition in combination with platinum-doublet chemotherapy at DFCI (n = 412), MSKCC (n = 748), Imperial College London (n = 78), and Massachusetts General Hospital (n = 47) were identified (Supplementary Fig. 1). At a median follow-up of 24.1 months (95% confidence interval [CI]: 22.6–25.1), the objective response rate (ORR) in the whole cohort of 1285 cases was 37.8% (95% CI: 35.2%–40.5%), the median progression-free survival (mPFS) was 5.8 months (95% CI: 5.5–6.2), and the median overall survival (mOS) was 14.1 months (95% CI: 13.0–15.8) calculated from the start date of CIT. The median age of the cohort was 67 (range: 24–93) years; 51.8% of the patients were male, 86.8% had a history of tobacco use, and 80.6% had lung adenocarcinoma. The baseline clinicopathologic and genomic characteristics of the patients are detailed in Table 1; Supplementary Table 1 provides information from each academic center. Among 1005 patients who experienced disease progression to CIT during the study period, subsequent systemic therapy was administered in 44.8% (n = 450) of the cases, as summarized in Supplementary Table 2.

Table 1.

Clinical and Pathological Characteristics of the 1285 Patients

To determine clinicopathologic factors that affect CIT efficacy, we evaluated the association of age, sex, Eastern Cooperative Oncology Group performance status (ECOG PS), smoking status, tumor histology, baseline derived neutrophil-to-lymphocyte ratio (dNLR), PD-L1 expression level, and oncogenic drivers with patient outcomes in univariable analyses (Fig. 1A-​C). Compared with patients above or equal to 70 years old, patients less than age 70 years had a significantly higher ORR to CIT (40.1% versus 34.5%; p = 0.046), significantly longer mPFS (6.0 versus 5.7 mo, hazard ratio [HR] = 0.88 [95% CI: 0.77–0.99]; p = 0.04), and significantly longer mOS (15.0 versus 13.5 mo, HR = 0.85 [95% CI: 0.73–0.98]; p = 0.03) (Supplementary Fig. 2A). There was no significant difference in ORR to CIT by sex (36.1% for men versus 39.7% for women, p = 0.18), but compared with men, women had a significantly longer mPFS (6.7 versus 5.0 mo, HR = 0.86 [95% CI: 0.76–0.98]; p = 0.02) and mOS (16.1 versus 12.6 mo, HR = 0.83 [95% CI: 0.72–0.96]; p = 0.01) (Supplementary Fig. 2B). Worsening ECOG PS from 0 (n = 359) to 1 (n = 691) to greater than or equal to 2 (n = 172) was associated with progressively lower ORR (44.6% versus 38.5% versus 24.4%, p < 0.001), shorter mPFS (7.4 versus 5.7 versus 3.2 mo, p < 0.001), and shorter mOS (19.4 versus 13.5 versus 7.4 mo, p < 0.001), respectively (Supplementary Fig. 2C). Comparing current or former smokers to never smokers, there was no difference in terms of ORR (37.8% versus 43.8%, p = 0.11), mPFS (5.6 versus 6.4 mo, HR = 1.02 [95% CI: 0.82–1.18]; p = 0.87), or mOS (13.9 versus 15.4 mo, HR = 1.09 [95% CI: 0.88–1.35]; p = 0.44) (Supplementary Fig. 2D) to CIT, respectively. According to tumor histologic subtype, although there was a higher response rate in squamous versus nonsquamous NSCLC (46.6% versus 36.7%, p = 0.02), this did not translate to a difference in mPFS (4.9 versus 5.9 mo, HR = 1.02 [95% CI: 0.84–1.25]; p = 0.82) or mOS (11.7 versus 14.5 mo, HR = 1.10 [95% CI: 0.88–1.39]; p = 0.41) (Supplementary Fig. 2E).

(A) Objective response rate to chemoimmunotherapy by age, sex, ECOG PS, smoking status, tumor histology (nonsquamous versus squamous), dNLR tertiles (lower [0.10-2.10], middle [2.11-3.33], and upper [3.34-36.8]), and PD-L1 TPS groups: less than 1% versus 1% to 49% versus 50% to 89% versus greater than or equal to 90%. Forest plot for (B) PFS and (C) OS with chemoimmunotherapy according to sex, ECOG PS, smoking status, tumor histology, dNLR tertiles, and PD-L1 expression level groups. CI, confidence interval; dNLR, derived neutrophil-to-lymphocyte ratio; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; NS, not significant; OS, overall survival; PD-L1, programmed death-ligand 1; PFS, progression-free survival; TPS, tumor proportion score.

Among 1211 cases with a complete blood count and differential before CIT initiation, the median dNLR was 2.6. Clinical outcomes improved with decreasing dNLR values divided into tertiles; moving from the upper (dNLR 3.34–36.8), to the middle (dNLR 2.11–3.33), to the lower (dNLR 0.10–2.10) tertile, we observed improvements in ORR (28.6% versus 38.2% versus 48.6%, p < 0.001), mPFS (3.5 versus 6.4 versus 8.0 mo, p < 0.001), and mOS (7.9 versus 15.0 versus 23.9 mo, p < 0.001), respectively (Supplementary Fig. 3A). Highlighting the continuous nature of dNLR, we also observed improvements in outcomes to CIT with decreasing dNLR values when dNLR was divided into quartiles or quintiles (Supplementary Fig. 3B and C).

Higher PD-L1 TPSs were associated with improvements in ORR (TPS <1%: 32.7%, TPS 1%–49%: 37.5%, TPS 50%–89%: 51.6%, TPS ≥90%: 61.7%, p < 0.001), mPFS (TPS <1%: 5.0 mo, TPS 1%–49%: 6.1 mo, TPS 50%–89%: 6.8 mo, TPS ≥90%: 13.0 mo, p < 0.001), and generally mOS (TPS <1%: 12.9 mo, TPS 1%–49%: 14.6 mo, TPS 50%–89%: 34.7 mo, TPS ≥90%: 23.1 mo, p = 0.009) (Supplementary Fig. 4). Because we observed a numerically longer mOS in patients with PD-L1 50% to 89% compared with PD-L1 greater than or equal to 90%, we evaluated whether the two groups were balanced in terms of clinicopathologic characteristics and the subsequent line of therapy on progression to CIT in this population. Patients with an ECOG PS of 0 were more common in the PD-L1 50% to 89% group compared with NSCCLs with a PD-L1 greater than or equal to 90%; other baseline clinical and pathologic characteristics were balanced in terms of age, sex, tumor histology, dNLR level, and oncogenic drivers (Supplementary Table 3A). Among 151 patients with NSCLCs and a high PD-L1 expression, a sizable number of patients are still on treatment with CIT; 41.7% (25 of 60) in PD-L1 greater than or equal to 90% and 31.9% (29 of 91) in the PD-L1 50% to 89% subset. Among patients who experienced disease progression, no difference in terms of receiving or not second-line therapy or type of treatment was observed between PD-L1 greater than or equal to 90% and PD-L1 50% to 89% subgroups (Supplementary Table 3B).

Efficacy of CIT in Genomic Subsets of NSCLC With KRAS, TP53, STK11, KEAP1, and SMARCA4 Alterations

We evaluated the impact of mutations in KRAS, TP53, STK11, KEAP1, and SMARCA4 on outcomes to CIT in nonsquamous NSCLC, given the prevalence of these mutations and their prior associations with treatment outcomes in NSCLC.12,14,25 Among the nonsquamous cohort of 877 cases with available KRAS mutation status (Supplementary Fig. 5), 351 (40.0%) were KRAS mutant (_KRAS_MUT) and 526 (60.0%) were KRAS wild type (_KRAS_WT). KRAS mutation status had no impact on CIT efficacy in terms of ORR, mPFS, or mOS (Fig. 2A-​C and Supplementary Fig. 6). Among 782 nonsquamous NSCLCs with available TP53 mutation status (Supplementary Fig. 7), _TP53_MUT cases (n = 412) had a significantly higher ORR (42% versus 31%; p = 0.001), a significantly longer mPFS (6.1 versus 5.5 mo, HR = 0.83 [95% CI: 0.71–0.97]; p = 0.02), but no significant increase in mOS (15.6 versus 13.1 mo, HR = 0.87 [95% CI: 0.73–1.04]; p = 0.13), compared with _TP53_WT cases (n = 370) (Fig. 2A-​C and Supplementary Fig. 8).

(A) Objective response rate, (B) PFS, and (C) OS to chemoimmunotherapy by KRAS, TP53, STK11, KEAP1, and SMARCA4 mutation status in nonsquamous NSCLC. Outcomes by TP53, STK11, KEAP1, and SMARCA4 mutation status in _KRAS_WT and _KRAS_MUT nonsquamous NSCLC are illustrated. CI, confidence interval; HR, hazard ratio; NS, not significant; OS, overall survival; PFS, progression-free survival.

A total of 707 nonsquamous tumors with comprehensive genomic profiling from DFCI (n = 291) and MSKCC (n = 416) were fully assessed for STK11, KEAP1, and SMARCA4 mutation status (Supplementary Fig. 9). Patients with _STK11_MUT tumors, compared with _STK11_WT tumors, had a significantly reduced ORR (25.1% versus 40.5%; p < 0.001), mPFS (3.9 versus 6.3 mo, HR = 1.46 [95% CI: 1.22–1.76]; p < 0.001), and mOS (10.4 versus 15.2 mo, HR = 1.36 [95% CI: 1.10–1.67]; p = 0.004) to CIT (Fig. 2A-​C and Supplementary Fig. 10A). This observation was also noted when looking at the DFCI and MSKCC cohorts individually (Supplementary Fig. 10B and C). Similarly, in all-comers with nonsquamous NSCLC, KEAP1 mutation was associated with significantly impaired outcomes to CIT when compared with _KEAP1_WT in terms of ORR (23.3% versus 41.1%; p < 0.001), mPFS (3.3 versus 6.7 mo, HR = 1.53 [95% CI: 1.28–1.84]; p < 0.001), and mOS (8.1 versus 16.9 mo, HR = 1.71 [95% CI: 1.40–2.10]; p < 0.001) (Fig. 2A-​C and Supplementary Fig. 11). Likewise, among patients with _SMARCA4_MUT nonsquamous NSCLC, CIT use was associated with significantly decreased ORR (21.9% versus 39.1%; p < 0.001), mPFS (2.7 versus 6.1 mo, HR = 1.62 [95% CI: 1.30–2.01]; p < 0.001), and mOS (8.1 versus 15.0 mo, HR = 1.70 [95% CI: 1.33–2.17]; p < 0.001) from the start of CIT (Fig. 2A-​C and Supplementary Fig. 12).

Because KRAS mutations define a subset of NSCLCs with unique clinical and genomic features, with heterogeneous outcomes to PD-(L)1 monotherapy on the basis of co-mutation status,14 we next explored the impact of co-mutations in TP53, STK11, KEAP1, and SMARCA4 in KRASMUT and KRASWT cases. Among _KRAS_WT nonsquamous NSCLCs, tumors harboring _TP53_MUT compared with _TP53_WT had a significantly higher ORR (44% versus 30%; p = 0.003) to CIT; however, this did not translate to a difference in PFS or OS (Fig. 2A-​C). Tumors harboring _KRAS_MUT_TP53_MUT compared with _KRAS-MUTTP53_WT had no difference in ORR (38% versus 31%; p = 0.27), a significantly longer mPFS (6.1 versus 5.5 mo, HR = 0.75 [95% CI: 0.58–0.97]; p = 0.027), and no significant difference in mOS (15.1 versus 13.2 mo, HR = 0.85 [95% CI: 0.63–1.13]; p = 0.26) (Fig. 2A-​C and Supplementary Fig. 13). We next evaluated the impact of STK11 mutation on outcomes to CIT by KRAS mutation status. The deleterious effect of STK11 mutations on CIT efficacy that we observed in all-comers with nonsquamous NSCLC seemed to be largely driven by the _KRAS_MUT subgroup. In the combined (DFCI + MSKCC) cohort, STK11 mutation was associated with significantly lower ORR and shorter mPFS and mOS among _KRAS_MUT, but not _KRAS_WT, nonsquamous NSCLCs (Fig. 2A-​C and Supplementary Fig. 14). Similar results were observed in the individual DFCI and MSKCC cohorts (Supplementary Figs. 15 and 16, respectively). In contrast to STK11, KEAP1 mutation negatively affected CIT efficacy both in _KRAS_MUT and in _KRAS_WT nonsquamous NSCLCs in the combined cohort (Fig. 2A-​C and Supplementary Fig. 17) and in the DFCI and MSKCC individual cohorts (Supplementary Figs. 18 and 19, respectively). Similar to KEAP1 mutation, _SMARCA4_MUT tumors have worse outcomes to CIT both in _KRAS_WT and _KRAS_MUT nonsquamous NSCLCs compared with _SMARCA4_WT cancers (Fig. 2A-​C and Supplementary Fig. 20). The individual DFCI and MSKCC cohorts are found in Supplementary Figures 21 and 22, respectively.

Given that mutations in STK11, KEAP1, and SMARCA4 can co-occur with each other,25,26 we also analyzed outcomes to CIT with concurrent alterations in these genes in nonsquamous NSCLC. The frequency of co-occurrence of these three genes in our cohort is found in Supplementary Figure 23. Compared with NSCLCs which were wild type in all three genes (n = 404), the ORR, mPFS, and mOS to CIT were progressively worse with an increasing number of alterations in these genes. For example, the response rate to CIT dropped from 43.3% to 33.1% to 25.7% to 7.7% in NSCLCs with mutations in zero, one, two, or all three of these genes, respectively, as found in Supplementary Figure 24A to C. The impact of mutations in one, two, or three of these genes in _KRAS_MUT and _KRAS_WT nonsquamous NSCLCs is found in Supplementary Figure 24D-I, respectively.

Impact of TMB on Clinical Outcomes to CIT in NSCLC

A total of 789 patients with NSCLCs, including tumors with squamous and nonsquamous histology, treated with first-line CIT had TMB values available at DFCI (n = 314) and MSKCC (n = 475). Because TMB was determined using two different platforms (OncoPanel at DFCI and MSK-IMPACT at MSKCC), we harmonized the TMB distributions between the two institutions by applying a normal transformation followed by standardization to Z-scores, as previously described,27 which brought the TMB distributions of the two cohorts into alignment (Supplementary Fig. 25). We initially explored the impact of increasing TMB cutoffs on ORR, PFS, and OS to CIT and observed improved outcomes with higher levels in the combined TMB cohort (DFCI + MSKCC) (Fig. 3A and ​B) and in each individual cohort (Supplementary Figs. 26 and 27). As this gradual improvement in outcomes could be influenced by TMB outliers, we also evaluated the HRs for PFS and OS of each TMB decile independently relative to the lowest decile as reference. Only patients with a TMB at the uppermost decile (≥90th percentile, termed “very high TMB,” corresponding to ≥19.0 mutations per megabase for the MSKCC and ≥19.3 mutations per megabase for the DFCI cohort) had improved PFS and OS with CIT (Fig. 3C). Forest plots in the individual DFCI and MSKCC cohorts are found in Supplementary Figure 28.

(A) ORR in each TMB decile in the combined TMB cohort. (B) Forest plot for PFS and OS to chemoimmunotherapy according to increasing TMB thresholds in the combined cohort. (C) Forest plot for PFS and OS to chemoimmunotherapy in each TMB decile versus the lowest decile as reference in the combined cohort. (D) ORR, PFS, and OS in NSCLCs with TMB greater than or equal to the 90th percentile versus TMB less than the 90th percentile. CI, confidence interval; HR, hazard ratio; NR, not reached; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; TMB, tumor mutational burden.

Tobacco use and KRAS wild-type status were more common in the very high TMB group compared with NSCLCs with a TMB less than the 90th percentile; other baseline clinical and pathologic characteristics were balanced in terms of age, sex, ECOG PS, tumor histology, dNLR level, and PD-L1 expression levels (Table 2). Compared with NSCLCs with a TMB Z-score less than the 90th percentile (n = 712), those with a very high TMB (TMB Z-score ≥90th percentile, n = 77) had a significantly higher ORR (53.5% versus 36.4%; p = 0.004), a significantly longer mPFS (10.8 versus 5.5 mo, HR = 0.62 [95% CI: 0.47–0.82]; p < 0.001), and a significantly longer mOS (29.2 versus 13.1 mo, HR = 0.54 [95% CI: 0.37–0.77]; p < 0.001) to CIT (Fig. 3D). Analyses for the individual DFCI and MSKCC cohorts are found in Supplementary Figure 29.

Table 2.

Clinical and Pathologic Characteristics of the 789 Patients by TMB Level (<90th Versus ≥90th Percentile)

Multivariable Analysis

Having found that different clinical and genomic factors are associated with clinical outcomes to CIT in NSCLC, we next performed multivariable Cox regression analysis to identify factors that retained association with CIT efficacy, after adjusting for potential confounders. In all-comers with NSCLC, we confirmed that a very high TMB (≥90th percentile), increasing PD-L1 TPS levels, low dNLR, and good ECOG PS were independently associated with improved PFS and OS (Supplementary Fig. 30). These results were confirmed in a sensitivity analysis conducted using inverse probability weighting (IPW) to account for potential selection bias resulting from PD-L1 missingness (Supplementary Fig. 31).

Because several genomic factors that are unique to nonsquamous NSCLCs also affected clinical efficacy of CIT in this NSCLC subtype, we next performed multivariable analysis in the subgroup of patients with nonsquamous histology. Importantly, we confirmed that a good ECOG PS, low dNLR, very high TMB (≥90th percentile), and increasing PD-L1 TPS levels were independently associated with improved PFS and OS (Fig. 4A and ​B). In addition, we also confirmed a significant association between KEAP1 and SMARCA4 mutations and shorter PFS (_KEAP1_MUT HR = 1.32, p = 0.01; _SMARCA4_MUT HR = 1.61, p < 0.001) and OS (_KEAP1_MUT HR = 1.69, p < 0.001; _SMARCA4_MUT HR = 1.66, p < 0.001) to CIT (Fig. 4A and ​B), whereas the presence of an STK11 mutation did not affect PFS or OS (Fig. 4A and ​B). Multivariable analysis using IPW was conducted also in this cohort and confirmed that KEAP1 and SMARCA4 mutations were independent factors associated with decreased PFS and OS (Supplementary Fig. 32).

Forest plot for (A) PFS and (B) OS in multivariable Cox regression analysis in the cohort of patients with advanced nonsquamous NSCLC treated with chemoimmunotherapy. CI, confidence interval; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; OS, overall survival; PD-L1, programmed death-ligand 1; PFS, progression-free survival; TMB, tumor mutational burden; TPS, tumor proportion score.

To evaluate whether _STK11_MUT had distinct impact in _KRAS_WT versus _KRAS_MUT NSCLC, multivariable analysis was performed in both subgroups. As observed in univariable analysis, STK11 mutations were also associated with impaired PFS and OS to CIT among KRAS mutant, but not KRAS wild-type, nonsquamous NSCLCs in multivariable models (Supplementary Figs. 33 and 34).

Discussion

Thus far, limited evidence has been available regarding baseline factors that influence outcomes to first-line PD-(L)1 blockade in combination with platinum-doublet chemotherapy in advanced NSCLC. Here, in a large cohort of 1285 patients, we found that a better performance status, a low dNLR level, and increasing levels of PD-L1 expression are associated with greater benefit from first-line CIT. In addition, cancers with STK11, KEAP1, or SMARCA4 mutations have less favorable outcomes to CIT in _KRAS_-mutant NSCLC; KEAP1 and SMARCA4 mutations are also associated with impaired outcomes to treatment in KRAS wild-type nonsquamous NSCLC. We also reveal that a very high TMB level (≥90th percentile) is independently associated with better clinical outcomes than for patients whose cancers had lower TMB values. Our findings have important implications for treatment decisions and clinical trial interpretation and design.

PD-L1 expression levels often affect treatment decisions in the first-line setting for patients with NSCLC lacking targetable genomic alterations. NSCLCs with a PD-L1 TPS greater than or equal to 50% are typically treated with PD-(L)1 monotherapy, whereas those with PD-L1 TPS less than 50% often receive a combination of a PD-(L)1 inhibitor plus platinum-doublet chemotherapy. Consistent with previous studies of PD-1 monotherapy,3,13 we also observed that clinical outcomes are improved with increasing PD-L1 expression levels greater than or equal to 90% compared with 50% to 89%, suggesting that PD-L1 expression acts as a continuous variable in predicting outcomes also for CIT. Our findings suggest that more granular PD-L1 levels (e.g., 50%–89% versus ≥90%) should routinely be reported in immunotherapy clinical trial results and potentially introduced as a stratification factor for new studies to ensure that outcomes are associated with treatment interventions, rather than imbalances in PD-L1 distributions. Although no difference in mOS was observed between the PD-L1 TPS greater than or equal to 90% and TPS 50% to 89% groups, a longer follow-up is needed to fully address the survival outcomes because a sizable number of patients are still on treatment with first-line CIT. Whether combining platinum-doublet chemotherapy with ICI improves clinical outcomes in NSCLCs with PD-L1 TPS 50% to 89% or greater than or equal to 90% compared with ICI alone needs to be explored.

STK11, KEAP1, and SMARCA4 mutations have been reported to correlate with impaired clinical outcomes in patients treated with immunotherapy among _KRAS_MUT NSCLC.12,25 Although a recent post hoc analysis of the KEYNOTE-189 and KEYNOTE-407 studies revealed that clinical outcomes to CIT were generally better than to chemotherapy alone,28 that study did not explore factors that affect the efficacy of CIT in NSCLC, which our current analysis seeks to address. Another recent study revealed that among patients with _KRAS_-mutant NSCLC treated with the IMpower 150 CIT regimen, the presence of STK11 or KEAP1 mutations was generally associated with worse outcomes to carboplatin plus paclitaxel with atezolizumab with or without bevacizumab.29 Although the co-mutation sample sizes in that study were relatively small, those findings corroborate our results which reveal, in a larger cohort of patients, that STK11, KEAP1, and/or SMARCA4 mutations have less favorable outcomes to CIT in _KRAS_-mutant NSCLC. Although genomic alterations in STK11, KEAP1, and SMARCA4 affect CIT efficacy among _KRAS_-mutant NSCLCs, we found that KEAP1 and SMARCA4 mutations are also associated with impaired outcomes among KRAS wild-type nonsquamous NSCLCs. In contrast, STK11 mutation did not affect outcomes to CIT among _KRAS_WT tumors, similar to a recent report among NSCLCs treated with immunotherapy alone.14 Mechanistically, both KEAP1 and SMARCA4 mutations can activate the Nrf2 signaling pathway,30,31 resulting in potential platinum-based treatment resistance through up-regulation of cytoprotective genes that enable cells to resist oxidative stress,32 and this may explain our observations that these genomic alterations impaired clinical outcomes to CIT. As more effective treatment options become available for _KRAS_MUT NSCLC, STK11/KEAP1/SMARCA4 mutation status may be a useful biomarker to determine the optimal treatment sequence, including covalent KRASG12C inhibitors that might be better used before ICI-based therapy in genomic subsets, which are less likely to respond to PD-(L)1 alone or CIT.

Along with genomic alterations in STK11, KEAP1, and SMARCA4, the predictive value of TMB as a biomarker for outcomes with anti-PD-(L)1 therapy may vary when administered as monotherapy or in combination with chemotherapy. A review of multiple studies of anti-PD-(L)1 blockade given as single agents across different tumor types, including NSCLC, revealed a significant association between increasing TMB levels and response to ICIs.33,34 Nonetheless, this parallel has not been found in NSCLCs treated with CIT. In contrast to an exploratory preliminary analysis of KEYNOTE-021, −189, and −407 studies where no association between TMB and outcomes to pembrolizumab plus chemotherapy was detected,35 we observed improved ORR, PFS, and OS to CIT in NSCLCs with very high TMB levels. Although the prior studies were conducted prospectively, the pre-specified cutpoint of 175 mutations per exome (approximately 10 mutations per megabase [mut/Mb] used by FoundationOne36) was derived using gene expression profiling and whole exome sequence from a training set of patients with multiple tumor types treated with pembrolizumab monotherapy36-39; whether higher TMB levels in these KEYNOTE studies might be associated with improved outcomes is thus far unknown. Consistent with our findings, a recent analysis of 1552 patients with advanced NSCLC treated with ICIs alone revealed that patients with high TMB levels (approximately the 90th percentile) derived the greatest improvement in terms of response to treatment and survival and that lower cutpoint closer to 10 mut/Mb may not significantly or adequately identify long-term survivors to immunotherapy.21 Therefore, exploring a cutoff of 10 mut/Mb might be too low to clearly distinguish patients most likely to benefit from ICIs plus platinum-doublet chemotherapy.

Our study is limited by its retrospective nature, lack of validation from prospective clinical trials of CIT, and some subgroup analyses with relatively small sample sizes. In addition, although we excluded lung cancers with EGFR and ALK alterations from our analyses, other genomic subsets with generally low response rates to immunotherapy were included in our cohort (e.g., those with ROS1 or RET rearrangements11), which may influence the clinical outcomes to CIT in this study; however, the frequency of these molecular subtypes of NSCLC in our data set was quite low. Nevertheless, this is the largest report of clinicopathologic and genomic correlates of CIT efficacy in patients with advanced NSCLCs to date. Last, PD-L1 expression was not available in 12% of the samples. Nonetheless, to account for the potential selection bias resulting from PD-L1 TPS missingness, we used IPW in Cox regression analysis.

In conclusion, our report of 1285 cases represents the largest retrospective cohort to date of cases treated with first-line CIT in NSCLC and provides important in-sights on the clinicopathologic and genomic features that influence CIT efficacy, with implications for clinical decision-making and trial design. Our work revealed an association between very high TMB level and benefit to CIT; very high TMB values may represent a novel biomarker for CIT efficacy among NSCLCs. In addition, with an expanding number of targeted KRAS therapies on the horizon, how best to sequence small molecule inhibitors, immunotherapy alone, or CIT in _KRAS_-mutant NSCLC according to STK11, KEAP1, and SMARCA4 co-mutation status warrants prospective study.

Supplementary Material

Supplementary_methods

Supplementary_Figures

Table_S1

Table_S3A-B

Table_S2

Acknowledgments

The authors acknowledge the Elva J. and Clayton L. McLaughlin Fund for Lung Cancer Research.

Disclosure:

Awad serves as a consultant to Merck, Bristol-Myers Squibb, Genentech, AstraZeneca, Nektar, Maverick, Blueprint Medicines, Syndax, AbbVie, Gritstone, ArcherDX, Mirati, NextCure, and EMD Serono; and receives research funding from Bristol-Myers Squibb, Lilly, Genentech, and, AstraZeneca. Gainor has served as a compensated consultant or received honoraria from Bristol-Myers Squibb, Genentech, Ariad/Takeda, Loxo, Pfizer, Incyte, Novartis, Merck, Agios, Amgen, Jounce, Karyopharm, GlydeBio, Regeneron, Oncorus, Helsinn, Array, and Clovis Oncology; has an immediate family member who is an employee with equity in Ironwood Pharmaceuticals; has received research funding from Novartis, Genentech/Roche, and Ariad/Takeda; and has received institutional research support from Tesaro, Moderna, Blueprint, Bristol-Myers Squibb, Jounce, Array, Adaptimmune, Novartis, Genentech/Roche, Alexo, and Merck. J. Lin has served as a compensated consultant for Genentech, C4 Therapeutics, Blueprint Medicines, Nuvalent, Turning Point Therapeutics, and Elevation Oncology; received honorarium and travel support from Pfizer; received institutional research funds from Hengrui Therapeutics, Turning Point Therapeutics, Neon Therapeutics, Relay Therapeutics, Bayer, Elevation Oncology, Roche, and Novartis; and received CME funding from OncLive, MedStar Health, and Northwell Health. Nishino serves as a consultant to Daiichi Sankyo and AstraZeneca; receives research grant from Merck, Canon Medical Systems, AstraZeneca, and Daiichi Sankyo; receives honorarium from Roche; and is supported by R01CA203636 and U01CA209414 (National Cancer Institute). Wang is supported by 5U01CA209414. Christiani is supported by 5U01CA209414. X. Lin is supported by R35-CA197449, U19-CA203654, and U01-CA209414 from the National Cancer Institute and U01-HG009088 and U01HG012064 from the National Human Genome Research Institute. Pinato received lecture fees from ViiV Healthcare, Bayer HealthCare, Bristol-Myers Squibb, Roche, Eisai, and Falk Foundation; received travel expenses from Bristol-Myers Squibb and Bayer Healthcare; received consulting fees for Mina Therapeutics, H3B, EISAI, Roche, DaVolterra, Mursla, Exact Sciences, Avamune, and AstraZeneca; received research funding (to institution) from Merck Sharp & Dohme, Bristol-Myers Squibb, and GlaxoSmithKline; is supported by grant funding from The Wellcome Trust Strategic Fund (PS3416), the Foundation for Liver Research, and the Associazione Italiana per la Ricerca sul Cancro (AIRC MFaG Grant ID 25697); and acknowledges support by the National Institute for Health and Care Research Imperial Biomedical Research Centre, the Imperial Experimental Cancer Medicine Centre, and the Imperial College Tissue Bank. Recondo reports receiving institutional research grants from Amgen and Janssen; receiving personal fees and serving on the advisory boards from Amgen, AstraZeneca, Bayer, Biocartis, Bristol-Myers Squibb, Janssen, Merck Serono, Merck Sharp & Dohme, Pfizer, Roche, and Takeda; serving as partner at PxMedica; and having clinical trials from Amgen, AstraZeneca, Bayer, Janssen, and Roche. Elkrief is supported by the Canadian Institutes of Health Research, Royal College of Physicians and Surgeons of Canada Detweiler Travelling Fellowship, and the Cedar’s Cancer Center Henry R. Shibata Fellowship. Schoenfeld has had a consulting or advisory role for Johnson & Johnson, Bristol-Myers Squibb, Merck, Enara Bio, KSQ Therapeutics, Lyell Immunopharma, Iovance Biotherapeutics, Perceptive Advisors, and Heat Biologics; and receives research funding from GlaxoSmithKline, PACT Pharma, Inc., Iovance Biotherapeutics, Achilles Therapeutics, Merck, Amgen, Bristol-Myers Squibb, and Harpoon Therapeutics. Johnson has received research support from Cannon Medical Systems; has served as a consultant/advisor for Hengrui Therapeutics, Novartis, GlaxoSmithKline, and Bluedot Bio; and participated in advisory boards for Boston Pharmaceuticals, Checkpoint Therapeutics, AstraZeneca, Daiichi Sankyo, G1 Therapeutics, Genentech, Janssen Scientific, Jazz Pharmaceuticals, and Hummingbird Diagnostics. The remaining authors declare no conflict of interest.

Footnotes

Supplementary Data

Note: To access the supplementary material accompanying this article, visit the online version of the Journal of Thoracic Oncology at www.jto.org and at https://doi.org/10.1016/j.jtho.2023.01.091.

CRediT Authorship Contribution Statement

Joao V. Alessi: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Visualization; Roles/Writing—original draft; Writing—review and editing.

Arielle Elkrief: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Visualization; Roles/Writing—original draft; Writing—review and editing.

Biagio Ricciuti: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Visualization; Roles/Writing—original draft; Writing—review and editing.

Xinan Wang: Methodology; Formal analysis; Validation; Writing—review and editing.

Alessio Cortellini: Data curation; Methodology; Validation; Writing—review and editing.

Victor R. Vaz: Data curation; Methodology; Writing—review and editing.

Giuseppe Lamberti: Data curation; Methodology; Validation; Writing—review and editing.

Rosa L. Frias: Data curation; Methodology; Validation; Writing—review and editing.

Deepti Venkatraman: Data curation; Methodology; Validation; Writing—review and editing.

Claudia A. M. Fulgenzi: Data curation; Methodology; Validation; Writing—review and editing.

Federica Pecci: Data curation; Methodology; Validation; Writing—review and editing.

Alessandro Di Federico: Data curation; Methodology; Writing—review and editing.

Gonzalo Recondo: Data curation; Methodology; Validation; Writing—review and editing.

Adriana Barrichello: Data curation; Methodology; Validation; Writing—review and editing.

Hyesun Park: Data curation; Methodology; Validation; Writing—review and editing.

Mizuki Nishino: Data curation; Methodology; Validation; Writing—review and editing.

Grace M. Hambelton: Data curation; Methodology; Validation; Writing—review and editing.

Jacklynn V. Egger: Methodology; Validation; Writing—review and editing.

Marc Ladanyi: Methodology; Validation; Writing—review and editing.

Subba Digumarthy: Data curation; Methodology; Validation; Writing—review and editing.

Bruce E. Johnson: Writing—review and editing.

David C. Christiani: Formal analysis; Validation; Writing—review and editing.

Xihong Lin: Methodology; Formal analysis; Validation; Writing—review and editing.

Justin F. Gainor: Writing—review and editing.

Jessica J. Lin: Data curation; Writing—review and editing.

David J. Pinato: Data curation; Methodology; Validation; Writing—review and editing.

Adam J. Schoenfeld: Conceptualization; Methodology; Project administration; Resources; Supervision; Validation; Writing—review and editing.

Mark M. Awad: Conceptualization; Methodology; Project administration; Resources; Supervision; Validation; Writing—review and editing.

Data sharing

Data are available on reasonable request. The data that support the finding of our study are available on request from the corresponding author.

References

1. Reck M, Rodríguez-Abreu D, Robinson AG, et al. Updated analysis of KEYNOTE-024:pembrolizumab versus platinum-based chemotherapy for advanced non-small-cell lung cancer with PD-L1 tumor proportion score of 50% or greater. J Clin Oncol. 2019;37:537–546. [PubMed] [Google Scholar]

2. Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial. Lancet. 2019;393:1819–1830. [PubMed] [Google Scholar]

3. Sezer A, Kilickap S, Gümüş M, et al. Cemiplimab monotherapy for first-line treatment of advanced non-small-cell lung cancer with PD-L1 of at least 50%: a multicentre, open-label, global, phase 3, randomised, controlled trial. Lancet. 2021;397:592–604. [PubMed] [Google Scholar]

4. Herbst RS, Giaccone G, de Marinis F, et al. Atezolizumab for first-line treatment of PD-L1-selected patients with NSCLC. N Engl J Med. 2020;383:1328–1339. [PubMed] [Google Scholar]

5. Hellmann MD, Paz-Ares L, Bernabe Caro R, et al. Nivolumab plus ipilimumab in advanced non-small-cell lung cancer. N Engl J Med. 2019;381:2020–2031. [PubMed] [Google Scholar]

6. Gandhi L, Rodríguez-Abreu D, Gadgeel S, et al. Pembrolizumab plus chemotherapy in metastatic non-small-cell lung cancer. N Engl J Med. 2018;378:2078–2092. [PubMed] [Google Scholar]

7. Paz-Ares L, Luft A, Vicente D, et al. Pembrolizumab plus chemotherapy for squamous non-small-cell lung cancer. N Engl J Med. 2018;379:2040–2051. [PubMed] [Google Scholar]

8. Socinski MA, Jotte RM, Cappuzzo F, et al. Atezolizumab for first-line treatment of metastatic nonsquamous NSCLC. N Engl J Med. 2018;378:2288–2301. [PubMed] [Google Scholar]

9. Paz-Ares L, Ciuleanu TE, Cobo M, et al. First-line nivolumab plus ipilimumab combined with two cycles of chemotherapy in patients with non-small-cell lung cancer (CheckMate 9LA): an international, randomised, open-label, phase 3 trial. Lancet Oncol. 2021;22:198–211. [PubMed] [Google Scholar]

10. Rizvi H, Sanchez-Vega F, La K, et al. Molecular determinants of response to anti-programmed cell death (PD)-1 and anti-programmed death-ligand 1 (PD-L1) blockade in patients with non-small-cell lung cancer profiled with targeted next-generation sequencing. J Clin Oncol. 2018;36:633–641. [PMC free article] [PubMed] [Google Scholar]

11. Mazieres J, Drilon A, Lusque A, et al. Immune checkpoint inhibitors for patients with advanced lung cancer and oncogenic driver alterations: results from the immuno-target registry. Ann Oncol Off J Eur Soc Oncol. 2019;30:1321–1328. [PMC free article] [PubMed] [Google Scholar]

12. Skoulidis F, Goldberg ME, Greenawalt DM, et al. STK11/LKB1 mutations and PD-1 inhibitor resistance in KRAS-mutant lung adenocarcinoma. Cancer Discov. 2018;8:822–835. [PMC free article] [PubMed] [Google Scholar]

13. Aguilar EJ, Ricciuti B, Gainor JF, et al. Outcomes to first-line pembrolizumab in patients with non-small-cell lung cancer and very high PD-L1 expression. Ann Oncol. 2019;30:1653–1659. [PubMed] [Google Scholar]

14. Ricciuti B, Arbour KC, Lin JJ, et al. Diminished efficacy of programmed death-(ligand)1 inhibition in STK11- and KEAP1-mutant lung adenocarcinoma is affected by KRAS mutation status. J Thorac Oncol. 2022;17:399–410. [PMC free article] [PubMed] [Google Scholar]

15. Alessi JV, Ricciuti B, Jiménez-Aguilar E, et al. Outcomes to first-line pembrolizumab in patients with PD-L1-high (≥50%) non-small cell lung cancer and a poor performance status. J Immunother Cancer. 2020;8:e001007. [PMC free article] [PubMed] [Google Scholar]

16. Gainor JF, Rizvi H, Jimenez Aguilar E, et al. Clinical activity of programmed cell death 1 (PD-1) blockade in never, light, and heavy smokers with non-small-cell lung cancer and PD-L1 expression ≥50%. Ann Oncol. 2020;31:404–411. [PMC free article] [PubMed] [Google Scholar]

17. Alessi JV, Ricciuti B, Alden SL, et al. Low peripheral blood derived neutrophil-to-lymphocyte ratio (dNLR) is associated with increased tumor T-cell infiltration and favorable outcomes to first-line pembrolizumab in non-small cell lung cancer. J Immunother Cancer. 2021;9:e003536. [PMC free article] [PubMed] [Google Scholar]

18. Herbst RS, Baas P, Kim DW, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial. Lancet. 2016;387:1540–1550. [PubMed] [Google Scholar]

19. Marabelle A, Fakih M, Lopez J, et al. Association of tumour mutational burden with outcomes in patients with advanced solid tumours treated with pembrolizumab: prospective biomarker analysis of the multicohort, open-label, phase 2 KEYNOTE-158 study. Lancet. Oncol2020;21:1353–1365. [PubMed] [Google Scholar]

20. Yarchoan M, Hopkins A, Jaffee EM. Tumor mutational burden and response rate to PD-1 inhibition. N Engl J Med. 2017;377:2500–2501. [PMC free article] [PubMed] [Google Scholar]

21. Ricciuti B, Wang X, Alessi JV, et al. 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. JAMA Oncol. 2022;8:1160–1168. [PMC free article] [PubMed] [Google Scholar]

22. Skoulidis F, Heymach JV. Co-occurring genomic alterations in non-small cell lung cancerbiology and therapy. Nat Rev Cancer. 2019;19:495–509. [PMC free article] [PubMed] [Google Scholar]

23. Gadgeel S, Rodríguez-Abreu D, Speranza G, et al. Updated analysis from KEYNOTE-189: pembrolizumab or placebo plus pemetrexed and platinum for previously untreated metastatic nonsquamous non-small-cell lung cancer. J Clin Oncol. 2020;38:1505–1517. [PubMed] [Google Scholar]

24. Garassino M, Rodriguez-Abreu D, Gadgeel S, et al. OA04.06 evaluation of TMB in KEYNOTE-189: pembrolizumab plus chemotherapy vs placebo plus chemotherapy for nonsquamous NSCLC. J Thorac Oncol. 2019;14:S216–S217. [Google Scholar]

25. Alessi JV, Ricciuti B, Spurr LF, et al. SMARCA4 and other SWItch/Sucrose nonfermentable family genomic alterations in NSCLC: clinicopathologic characteristics and outcomes to immune checkpoint inhibition. J Thorac Oncol. 2021;16:1176–1187. [PubMed] [Google Scholar]

26. Schoenfeld AJ, Bandlamudi C, Lavery JA, et al. The genomic landscape of SMARCA4 alterations and associations with outcomes in patients with lung cancer. Clin Cancer Res. 2020;26:5701–5708. [PMC free article] [PubMed] [Google Scholar]

27. Vokes NI, Liu D, Ricciuti B, et al. Harmonization of tumor mutational burden quantification and association with response to immune checkpoint blockade in non-small-cell lung cancer. JCO Precis Oncol. 2019;3:PO.19.00171. [PMC free article] [PubMed] [Google Scholar]

28. Garassino MC, Gadgeel S, Novello S, et al. Associations of tissue tumor mutational burden and mutational status with clinical outcomes with pembrolizumab plus chemotherapy versus chemotherapy for metastatic NSCLC. JTO Clin Res Rep. 2022;4:100431. [PMC free article] [PubMed] [Google Scholar]

29. West HJ, McCleland M, Cappuzzo F, et al. Original research: clinical efficacy of atezolizumab plus bevacizumab and chemotherapy in KRAS-mutated non-small cell lung cancer with STK11, KEAP1, or TP53 commutations: subgroup results from the phase III IMpower150 trial. J Immunother Cancer. 2022;10:e003027. [PMC free article] [PubMed] [Google Scholar]

30. Song S, Nguyen V, Schrank T, et al. Loss of SWI/SNF chromatin remodeling alters NRF2 signaling in non-small cell lung carcinoma. Mol Cancer Res. 2020;18:1777–1788. [PMC free article] [PubMed] [Google Scholar]

31. Best SA, Sutherland KD. “Keaping” a lid on lung cancer: the Keap1-Nrf2 pathway. Cell Cycle. 2018;17:1696–1707. [PMC free article] [PubMed] [Google Scholar]

32. Tian Y, Wu K, Liu Q, et al. Modification of platinum sensitivity by KEAP1/NRF2 signals in non-small cell lung cancer. J Hematol Oncol. 2016;9:1–14. [PMC free article] [PubMed] [Google Scholar]

33. Samstein RM, Lee CH, Shoushtari AN, et al. Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat Genet. 2019;51:202–206. [PMC free article] [PubMed] [Google Scholar]

34. Valero C, Lee M, Hoen D, et al. The association between tumor mutational burden and prognosis is dependent on treatment context. Nat Genet. 2021;53:11–15. [PMC free article] [PubMed] [Google Scholar]

35. Paz-Ares L, Langer CJ, Novello S, et al. Pembrolizumab (pembro) plus platinum-based chemotherapy (chemo) for metastatic NSCLC: Tissue TMB (tTMB) and outcomes in KEYNOTE-021. Ann Oncol. 2019;30:v917–v918. [Google Scholar]

36. Cristescu R, Aurora-Garg D, Albright A, et al. Tumor mutational burden predicts the efficacy of pembrolizumab monotherapy: a pan-tumor retrospective analysis of participants with advanced solid tumors. J Immunother Cancer. 2022;10:e003091. [PMC free article] [PubMed] [Google Scholar]

37. Merino DM, McShane LM, Fabrizio D, et al. Establishing guidelines to harmonize tumor mutational burden (TMB): in silico assessment of variation in TMB quantification across diagnostic platforms: phase I of the Friends of Cancer Research TMB Harmonization Project. J Immunother Cancer. 2020;8:e000147. [PMC free article] [PubMed] [Google Scholar]

38. Ayers M, Lunceford J, Nebozhyn M, et al. IFN-γ-related mRNA profile predicts clinical response to PD-1 blockade. J Clin Invest. 2017;127:2930–2940. [PMC free article] [PubMed] [Google Scholar]

39. Stenzinger A, Allen JD, Maas J, et al. Tumor mutational burden standardization initiatives: recommendations for consistent tumor mutational burden assessment in clinical samples to guide immunotherapy treatment decisions. Genes. Chromosomes Cancer. 2019;58:578–588. [PMC free article] [PubMed] [Google Scholar]