Biomarker analysis patient cohort characteristics and ITT patient characteristics. Clinical and biomarker characteristics are similar between treatment arms of the biomarker analysis patient cohort evaluated in this study. Clinical and biomarker characteristics are similar between the ITT population in IMpower131 and the biomarker analysis patient cohort evaluated in this study.

Supplementary Figure S2: ctDNA decrease ≥ 50% or ≥ 90% at C4D1 on chemoIO identifies patients with poorer outcomes on IO maintenance

Supplementary Figure S5: ctDNA monitoring on chemo requires greater levels of ctDNA clearance to effectively predict patient outcomes.

ctDNA decrease from any subsequent cycle was associated with greater objective response in patients treated with chemoIO. Number of patients per best objective response with associated increase or decreases in ctDNA levels between subsequent cycles.

Supplementary Figure S3: Any ctDNA decrease or full clearance at C4D1 on chemoIO identifies patients with poorer outcomes on IO maintenance

Supplementary Figure S8: Number of variants tracked per patient did not impact prognostic outcomes of ctDNA monitoring

Supplementary Figure S4: Full clearance of ctDNA at C4D1 on chemoIO identifies patients with poorer outcomes on IO maintenance, independent of PD-L1 status

ctDNA decrease from C1 to C4 was associated with greater objective response in patients treated with chemoIO at all cutoffs. Number of patients per best objective response with associated increase or decreases in ctDNA between C1 and C4. *Progression-free rate was calculated from the start of therapy rather than the start of C4D1.

Supplementary Figure S7: ctDNA detection was associated with worse outcomes across all monitoring timepoints

After the global approval of atezolizumab plus bevacizumab and chemotherapy as first-line metastatic nonsquamous non-small-cell lung cancer (nsqNSCLC) treatment, the IMpower151 ( NCT04194203 ) trial was conducted in China to address regional differences. Chemotherapy-naive patients with metastatic nsqNSCLC (N = 305) were randomized 1:1 to receive either atezolizumab, bevacizumab, carboplatin and paclitaxel or pemetrexed (ABCPem/Pac; n = 152) or placebo plus bevacizumab, carboplatin and pemetrexed or paclitaxel (BCPem/Pac; n = 153). The primary endpoint was investigator-assessed progression-free survival (INV-PFS); secondary endpoints included subgroup analyses of INV-PFS, independent review facility-assessed PFS, overall survival, and investigator-assessed objective response rate and duration of response per RECIST v.1.1. Most patients (97%) received pemetrexed, and 53% had EGFR+ tumors. Median INV-PFS for ABCPem/Pac versus BCPem/Pac was 9.5 versus 7.1 months (stratified hazard ratio: 0.84; 95% confidence interval: 0.65, 1.09; P = 0.184). INV-PFS across subgroups and independent review facility-assessed PFS were consistent with INV-PFS in the intention-to-treat population. Median overall survival was 20.7 versus 18.7 months in the ABCPem/Pac versus BCPem/Pac arms, respectively (stratified hazard ratio: 0.93; 95% confidence interval: 0.67, 1.28). Confirmed objective response rate with ABCPem/Pac versus BCPem/Pac was 48% versus 50%, respectively; median duration of response was 11.3 versus 8.3 months. Adverse events of special interest for atezolizumab were observed in 68% (grades 3 and 4: 11%) and 71% (grades 3 and 4: 7%) of patients receiving ABCPem/Pac and BCPem/Pac, respectively. The most common adverse events of special interest for atezolizumab in the ABCPem/Pac and BCPem/Pac arms were hepatitis (driven by laboratory abnormalities; mostly low grade), hypothyroidism and rash. Overall, IMpower151 did not meet its primary endpoint (INV-PFS) in metastatic nsqNSCLC. ABCPem/Pac was generally well tolerated, with no new safety signals. Trial registration number: ClinicalTrials.gov, NCT02366143.

INTRODUCTION:Treatment selection in patients with advanced NSCLC is based on programmed death-ligand 1 (PD-L1) expression, which is usually scored manually and is subject to intra- and inter-pathologist variability. A PD-L1 clone-agnostic artificial intelligence (AI) model for AI-based measurement of PD-L1 (AIM-PD-L1) was developed and assessed in advanced NSCLC using clinical samples from two phase 3 trials. METHODS:IMpower110 evaluated atezolizumab versus chemotherapy in PD-L1-positive metastatic, stage IV, squamous or nonsquamous NSCLC. IMpower150 evaluated atezolizumab, carboplatin, and paclitaxel, with or without bevacizumab, versus carboplatin, paclitaxel, and bevacizumab in patients with metastatic nonsquamous NSCLC. AIM-PD-L1 was developed and deployed on SP263-stained whole slide images (IMpower110, n = 509; IMpower150, n = 766) for digital scoring of tumor cell (TC) PD-L1 expression and identification of human-interpretable features (HIFs) associated with survival outcomes. RESULTS:Overall percentage agreements between scoring methods for TC more than or equal to 50% and more than or equal to 1% cutoffs were high. Survival analyses were similar for PD-L1 subgroups between scoring methods at both TC cutoffs. A nonsignificant improvement in survival outcomes was observed in patients treated with atezolizumab-containing regimens and classified as positive by digital scoring but missed by manual scoring. Two HIFs in the cancer epithelium-density of all PD-L1-positive TC and immune cells-were nominally associated with overall survival. Many HIFs were identified to be predictive of significantly improved progression-free survival with atezolizumab-containing regimens versus control. CONCLUSIONS:AIM-PD-L1 digital SP263 PD-L1 scoring is concordant with manual scoring, revealing similar predictivity for benefit, and could potentially be used as a predictive marker for patient stratification and selection for anti-PD-(L)1 therapy.

Importance:Treating locally advanced squamous cell carcinoma of the head and neck (LA SCCHN) involves any combination of surgery, radiation, and chemotherapy, followed by routine monitoring for local recurrence or distant metastases. Given the poor patient outcomes, a significant unmet clinical need for improved treatment options remains. Objective:To evaluate efficacy and safety of maintenance atezolizumab in patients with LA SCCHN at high risk of disease progression after multimodal definitive treatment. Design, Setting, and Participants:IMvoke010 was a phase 3, global, double-blind, randomized clinical trial. Patients were recruited at 128 sites in 23 countries between April 3, 2018, and February 14, 2020 (clinical cutoff date: September 27, 2023). Eligible patients had LA SCCHN (stage IVa/IVb involving the oral cavity, larynx, hypopharynx, or human papillomavirus-negative oropharynx, or stage III human papillomavirus-positive oropharynx [AJCC Cancer Staging Manual, eighth edition]) without disease progression after multimodal definitive treatment. Intervention:Patients were randomized (1:1) to receive atezolizumab 1200 mg or placebo every 3 weeks for 1 year or until disease recurrence, disease progression, unacceptable toxicity, or consent withdrawal. Main Outcomes and Measures:The primary end point was investigator-assessed event-free survival. Other end points included overall survival and safety. Results:Overall, 406 patients were randomized to receive atezolizumab (n = 203) or placebo (n = 203); baseline demographics were balanced between both treatment groups (<65 years, 142 [70.0%] vs 155 [76.4%]; male, 168 [82.8%] vs 174 [85.7%]; Asian, 68 [35.6%] vs 61 [31.0%]; Black, 1 [0.5%] vs 1 [0.5%]; and White, 121 [63.4%] vs 135 [68.5%], respectively). At clinical cutoff (median follow-up, 46.5 months), median investigator-assessed event-free survival was 59.5 months (95% CI, 46.8 to not estimable) with atezolizumab vs 52.7 months (95% CI, 41.4 to not estimable) with placebo (hazard ratio, 0.94; 95% CI, 0.70-1.26; P = .68). There was no difference in overall survival between atezolizumab and placebo (24-month overall survival, 82.0% vs 79.2%, respectively). No new or unexpected safety signals were identified. Conclusions and Relevance:In this study, atezolizumab did not improve clinical outcomes in patients with LA SCCHN at high risk of disease progression after multimodal definitive treatment. These data contribute to evidence on the limited activity of checkpoint inhibitors in the global population of this disease setting. Overall, the role of immunotherapy for patients with LA SCCHN remains to be determined. Trial Registration:ClinicalTrials.gov Identifier: NCT03452137.

The phase 3 IMpower150 trial in treatment-naïve patients with metastatic non-small-cell lung cancer (NSCLC) demonstrates significantly longer progression-free (PFS) and overall survival (OS) with first-line atezolizumab (anti-PD-L1)-bevacizumab (anti-VEGF)-carboplatin-paclitaxel (ABCP) than with bevacizumab-carboplatin-paclitaxel (BCP). We characterise four molecular NSCLC subtypes identified by unsupervised clustering of transcriptomes of 564 pre-treatment primary tumour samples from IMpower150 using non-negative matrix factorization (NMF1-4). Each subtype has distinct tumour PD-L1 expression levels, epithelial characteristics, immune composition, and treatment outcomes. Both NMF2 (enriched in tumour proliferation signal, macrophages, and monocytes) and NMF4 (enriched in B cells and T cells) have elevated tumour PD-L1 expression. Of these two, only NMF4 demonstrates PFS and OS benefits with ABCP versus either BCP or atezolizumab-carboplatin-paclitaxel (ACP). Patients with NMF1 (enriched in basal and squamous-like cells) have improved outcomes on ABCP compared with ACP or BCP; those with NMF3 (enriched in adenocarcinoma signatures) show similar outcomes among treatments. These insights could help inform individualised first-line treatment for metastatic NSCLC.

BACKGROUND:Despite improved efficacy with first-line immune checkpoint inhibitors plus platinum-based chemotherapy for extensive-stage small-cell lung cancer (ES-SCLC), survival remains poor. In this study, we aimed to compare lurbinectedin plus atezolizumab and atezolizumab alone as maintenance therapies in patients with ES-SCLC without progression after induction therapy with atezolizumab, carboplatin, and etoposide. METHODS:IMforte was a randomised, open-label, phase 3 trial done at 96 hospitals and medical centres in 13 countries (Belgium, Germany, Greece, Hungary, Italy, Mexico, Poland, South Korea, Spain, Taiwan, Türkiye, the UK, and the USA). Eligible patients were aged 18 years or older with treatment-naive ES-SCLC. Patients received four 21-day cycles of induction treatment (atezolizumab, carboplatin, and etoposide). After completing induction treatment, eligible patients without disease progression were randomly assigned (1:1) using permuted blocks (Interactive Voice/Web Response System) to receive maintenance treatment intravenously every 3 weeks with lurbinectedin (3·2 mg/m2; with granulocyte colony-stimulating factor prophylaxis) plus atezolizumab (1200 mg) or atezolizumab (1200 mg). The two primary endpoints were independent review facility-assessed (IRF) progression-free survival and overall survival, measured from randomisation into the maintenance phase. Efficacy endpoints were assessed in the full analysis set, which included all patients who were randomly assigned to maintenance phase treatment, regardless of whether they received their assigned study treatment. Safety was assessed in all patients who received at least one dose of lurbinectedin or atezolizumab, and was analysed according to the treatment received. This study is registered with ClinicalTrials.gov, NCT05091567, and is closed for recruitment. FINDINGS:Between Nov 17, 2021, and Jan 11, 2024, 895 patients were screened for enrolment, of whom 660 (74%) were enrolled into the induction phase. Between May 24, 2022, and April 30, 2024, 483 (73%) of 660 patients entered the maintenance phase and were randomly assigned to lurbinectedin plus atezolizumab (n=242) or atezolizumab (n=241). At the data cutoff (July 29, 2024), IRF progression-free survival was longer in the lurbinectedin plus atezolizumab group than the atezolizumab group (stratified hazard ratio [HR] 0·54 [95% CI 0·43-0·67]; p<0·0001), as was overall survival (stratified HR 0·73 [0·57-0·95]; p=0·017). 92 (38%) of 242 patients in the lurbinectedin plus atezolizumab group and 53 (22%) of 240 patients in the atezolizumab group had grade 3-4 adverse events. The most common grade 3-4 events in the lurbinectedin plus atezolizumab group were anaemia (20 [8%] of 242 patients), decreased neutrophil count (18 [7%] patients), and decreased platelet count (18 [7%] patients) and the most common events in the atezolizumab group were hyponatremia (five [2%] of 240 patients), dyspnoea (four [2%] patients), and pneumonia (four [2%] patients). Grade 5 adverse events occurred in 12 (5%) of 242 patients in the lurbinectedin plus atezolizumab group and six (3%) of 240 patients in the atezolizumab group. The incidence of myelosuppressive toxicities (eg, neutropenia and leukopenia) was higher in the lurbinectedin plus atezolizumab group than the atezolizumab group. INTERPRETATION:IRF progression-free survival and overall survival were longer in the lurbinectedin plus atezolizumab group than the atezolizumab group for patients with ES-SCLC, albeit with a higher incidence of adverse events. Lurbinectedin plus atezolizumab represents a novel therapeutic option for first-line maintenance treatment in this setting. FUNDING:F Hoffmann-La Roche and Jazz Pharmaceuticals.

IMpower010 (ClinicalTrials.gov identifier: NCT02486718 ) previously showed that atezolizumab improved disease-free survival (DFS) versus best supportive care (BSC) after adjuvant chemotherapy in patients with resected non–small cell lung cancer (NSCLC). We report DFS final analysis, second overall survival (OS) interim analysis, and safety with a ≥5-year follow-up. Patients with completely resected stage IB-IIIA NSCLC were randomly assigned to atezolizumab (1,200 mg once every 3 weeks, 16 cycles) or BSC after platinum-based chemotherapy. At clinical cutoff (January 26, 2024), stratified hazard ratios (HRs; 95% CI) for DFS were 0.85 (95% CI, 0.71 to 1.01; P = .07) in the intention-to-treat (n = 1,005), 0.83 (95% CI, 0.69 to 1.00) in the all-randomized stage II-IIIA (n = 882), and 0.70 (95% CI, 0.55 to 0.91) in stage II-IIIA PD-L1 tumor cell (TC) ≥1% (n = 476) populations. Stratified HRs (95% CI) for OS were 0.97 (95% CI, 0.78 to 1.22), 0.94 (95% CI, 0.75 to 1.19), and 0.77 (95% CI, 0.56 to 1.06), respectively. The unstratified HRs (95% CI) in the stage II-IIIA PD-L1 TC ≥50% population (n = 229) were 0.48 (95% CI, 0.32 to 0.72) for DFS and 0.47 (95% CI, 0.28 to 0.77) for OS, and the unstratified HRs in the stage II-IIIA PD-L1 TC ≥50% without EGFR / ALK alterations (n = 209) population were 0.49 (95% CI, 0.32 to 0.75) and 0.44 (95% CI, 0.26 to 0.74). No new safety signals were reported. IMpower010 is the first study to report survival outcomes with a ≥5-year follow-up and continued to show benefit with atezolizumab versus BSC after adjuvant chemotherapy in patients with resected stage II-IIIA PD-L1–selected NSCLC.

8022 Background: In metastatic NSCLC (mNSCLC), genomic alterations (MUT) are well studied and have led to the development of efficacious new drugs. However, the association of MUT in the adjuvant setting in eNSCLC is not as well understood. Here, we describe an exploratory, retrospective analysis of genomic profiling by whole exome sequencing and clinical association in IMpower010. Methods: Whole exome sequencing was done on baseline tumor samples and germline DNA from whole blood from the biomarker-evaluable population (n=623). Multiple MUT were identified in the full population, including by histology. Disease-free survival (DFS) and overall survival (OS) were assessed in the most common gene subgroups. Results: The distribution of MUT and co-occurrence patterns were similar to expected non-squamous and squamous patterns in mNSCLC, except for lower prevalence of STK11 (17%) and KEAP1 (12%) MUT. In non-squamous disease, STK11 , EGFR , and KEAP1 MUT were associated with increased prevalence in PD-L1–negative tumors, and TP53 MUT with PD-L1–positive tumors (adjusted P <0.1). KRAS MUT were associated with Stage II; STK11 MUT were associated with Stage I more than with Stages II and III (adjusted P <0.1). Increased enrichment of KRAS , STK11 , KEAP1 , and TP53 MUT was seen in those with previous or current smoking status, and EGFR MUT were enriched in those who never smoked (adjusted P <0.1). In non-squamous disease, STK11 MUT were a poor prognostic for OS but not DFS, whereas KEAP1 MUT were not significantly associated with poor prognosis for DFS or OS (Table). Neither STK11 or KEAP1 MUT were significantly associated with differential atezolizumab vs best supportive care DFS or OS benefit (interaction P >0.05). Conclusions: This analysis represents the largest dataset evaluating the genomic profile of patients with eNSCLC who were treated with cancer immunotherapy. The prevalences of STK11 and KEAP1 MUT were lower than in mNSCLC and were enriched for PD-L1–negative in non-squamous NSCLC. Unlike in mNSCLC, patients with tumors that harbored KEAP1 MUT did not have poor prognosis in IMpower010. Data are hypothesis generating and require validation in independent eNSCLC datasets with larger numbers. Clinical trial information: NCT02486718 . STK11 and KEAP1 associations with DFS and OS in combined arms (non-squamous). STK11 MUT STK11 WT KEAP1 MUT KEAP1 WT n 71 342 49 364 DFS Median, mo 43.1 41.8 45.3 41.8 HR (95% CI) 1.03 (0.73, 1.46) 0.91 (0.60, 1.39) OS Median, mo NR NR NR NR HR (95% CI) 1.66 (1.10, 2.52) 1.08 (0.63, 1.85) CI, confidence interval; mo, month; NR, not reached; WT, wild type.

LBA8035 Background: IMpower010 (NCT02486718) met its primary endpoint of significant DFS improvement with atezo vs BSC after adj chemotherapy in resected NSCLC in the PD-L1 TC ≥1% and all-randomized stage II-IIIA populations, leading to worldwide approval of adj atezo for PD-L1 TC ≥1% or PD-L1 TC ≥50% stage II-IIIA NSCLC. At OS IA1, a trend favoring atezo was seen in the PD-L1 TC ≥1% stage II-IIIA population. Here we report findings from the DFS FA and OS IA2. Methods: The IMpower010 study design has been previously described (Felip et al, Lancet 2021). The primary DFS and secondary OS endpoints were tested hierarchically: DFS in the PD-L1 TC ≥1% (SP263) stage II-IIIA, then in the all-randomized stage II-IIIA, and then in the intent-to-treat (ITT; stage IB-IIIA) populations, followed by OS in the ITT population. Secondary endpoints included 3- and 5-y DFS and DFS in the PD-L1 TC ≥50% (SP263) stage II-IIIA population. OS in the ITT population could only be formally tested if the significance boundary for DFS in that population was crossed. Results: At the DFS FA and OS IA2 (clinical cutoff date: Jan 26, 2024), with a minimum follow-up of 60 mo, DFS and OS for the PD-L1 TC ≥1% and TC ≥50% stage II-IIIA populations were consistent with previously observed benefit; the difference in median (m) DFS between arms in the PD-L1 TC ≥1% population was 31.2 mo (Table). In the ITT population, the significance boundary for DFS was not crossed and OS was similar between arms, although data were immature. The safety profile of atezo was consistent with prior analyses. Conclusions: These results provide the first cancer immunotherapy data with ≥5 y of follow-up from a Phase III study in resectable NSCLC. Although the statistical boundary for the ITT population was not crossed, DFS benefit with adj atezo continues to translate into a positive OS trend vs BSC in the PD-L1 TC ≥1% and TC ≥50% stage II-IIIA populations. These results further support the use of adj atezo in PD-L1–selected populations. Clinical trial information: NCT02486718 . [Table: see text]