Real-World Treatment Patterns and Clinical Outcomes After Platinum-Doublet Chemotherapy and Immunotherapy in Metastatic Non–Small Cell Lung Cancer: A Multiregional Chart Review in the United States, Europe, and Japan

PURPOSE To characterize treatment patterns and real-world clinical outcomes of patients with metastatic non–small cell lung cancer (mNSCLC) who developed progression on an anti–PD-1/anti–PD-L1, herein referred to as anti–PD-(L)1, and platinum-doublet chemotherapy. METHODS Eligible oncologists/pulmonologists in the United States, Europe (France, Germany, and United Kingdom), and Japan completed electronic case report forms for patients with mNSCLC (no evidence of EGFR/ALK/ROS1 alterations). Eligible patients had disease progression on/after an anti–PD-(L)1 and platinum-doublet chemotherapy (received concurrently or sequentially), initiated a subsequent line of therapy (LOT) between 2017 and 2021, and had an Eastern Cooperative Oncology Group (ECOG) performance status 0-2 at this subsequent LOT initiation (index date). Overall survival (OS), time to treatment discontinuation (TTD), and real-world progression-free survival (rwPFS) after index were assessed using Kaplan-Meier analysis. RESULTS Overall, 160 physicians (academic, 54.4%; community, 45.6%) provided deidentified data from 487 patient charts (United States, 141; Europe, 218; Japan, 128; at mNSCLC diagnosis: median age 66 years, 64.7% male, 81.3% nonsquamous, 86.2% de novo mNSCLC; at line of interest initiation: 86.0% ECOG 0-1, 39.6% liver metastases, 18.9% brain metastases, 79.1% smoking history). The most common treatment regimens upon progression after anti–PD-(L)1/platinum-doublet chemotherapy were nonplatinum chemotherapy (50.5%), nonplatinum chemotherapy plus vascular endothelial growth factor receptor inhibitor (12.9%), and platinum-doublet chemotherapy (6.6%). Median OS was 8.8 months (squamous, 7.8 months; nonsquamous, 9.5 months). Median TTD was 4.3 months (squamous, 4.1 months; nonsquamous, 4.3 months). Median rwPFS was 5.1 months (squamous, 4.6 months; nonsquamous, 5.4 months). CONCLUSION In this multiregional, real-world analysis of pooled patient chart data, patients with mNSCLC who had disease progression after anti–PD-(L)1/platinum-doublet chemotherapy had poor clinical outcomes with various treatment regimens, demonstrating an unmet clinical need for effective options after failure on anti–PD-(L)1 and platinum-doublet chemotherapy treatments.

34%-38% of patients discontinued treatment because of disease progression and 32%-39% required a subsequent line of therapy (LOT) over a median follow-up time of 22-31 months. 8,9ere are limited effective treatment options and recommendations for later lines of therapy after progression on anti-PD-(L)1 and platinum-doublet chemotherapy. 1,2,10[16][17] Studies have evaluated treatment patterns and real-world outcomes following progression after anti-PD-(L)1 or platinum-doublet chemotherapy, [18][19][20][21][22][23] but there are limited analyses of patients who have progression after both.A few previously published studies reported median overall survival (OS) of 8.1-11.6 months among patients with advanced NSCLC who developed progression on 1L anti-PD-(L)1 plus platinum-doublet chemotherapy in clinical practice, but these studies included small sample sizes or focused on second-line (2L) chemotherapy regimens only. 24,25Preliminary reports of a number of US-based studies have described real-world treatment patterns and median OS (6.2-14.37][28] There remains a need for additional analyses of treatment outcomes associated with various regimens among a global patient population.This retrospective, physician panel-based medical chart review study characterized treatment patterns and real-world outcomes of patients with mNSCLC who developed progression after an anti-PD-(L)1 and platinum-doublet chemotherapy in the United States, Europe, and Japan.

Data Source
A noninterventional, physician panel-based medical chart review was undertaken in the United States, Europe (ie, Germany, France, and United Kingdom), and Japan.Eligible medical oncologists and/or pulmonologists were recruited from an existing physician panel in each country to complete a survey and collect data from eligible patient charts via an online case report form (CRF). Data were collected from June 06, 2022, to April 28, 2023, and were anonymized; no personal identifiable information were accessible to the study sponsor or researchers at any time.Therefore, the study received an exemption from the Pearl Independent Institutional Review Board.

Inclusion Criteria
Physicians were eligible if they had (1) a medical oncology or pulmonology subspecialty, (2) experience treating adult patients with mNSCLC for ≥5 years, (3) treated ≥5 patients with squamous/nonsquamous mNSCLC in the past 5 years,

CONTEXT Key Objective
What are the real-world outcomes and treatment patterns of patients with metastatic non-small cell lung cancer (mNSCLC) who develop progression after treatment with both anti-PD-(L)1 and platinum-doublet chemotherapy?

Knowledge Generated
In one of the first analyses of clinical outcomes after previous exposure to both anti-PD-(L)1 and platinum-doublet chemotherapy, patients with mNSCLC in the United States, Europe, and Japan in this study largely received nonplatinum chemotherapy after progression.Real-world clinical outcomes were suboptimal, with a median overall survival of 8.8 months, time to treatment discontinuation of 4.3 months, and real-world progression-free survival of 5.1 months overall.

Relevance
These findings highlight the large unmet need for additional treatment options that may improve survival in later lines of therapy for patients with mNSCLC who have experienced progression on both anti-PD-(L)1 and platinum-doublet chemotherapy.Clinical trials of novel agents that may offer benefit in this patient population are needed.
including ≥2 patients in the past year, (4) experience using immunotherapy-based regimens for mNSCLC, and (5) direct access to charts of patients with mNSCLC at the time of CRF completion, including those of deceased patients or patients no longer under care.
Patient charts from eligible physicians were included if they (1) were from adults with a histologically or cytologically confirmed NSCLC diagnosis, (2) had mNSCLC with no history of other primary malignancies, (3) had not participated in a therapeutic clinical trial for mNSCLC, (4) had documentation of squamous (had no known EGFR-, ALK-, and ROS1-positive test results and no use of an EGFR-, ALK-, or ROS1-directed targeted therapy for mNSCLC) or nonsquamous (had documentation of EGFRand ALK-negative test results and no known ROS1-positive test result) histology, ( 5) had progression after one anti-PD-(L)1 and platinum-doublet chemotherapy regimen (in combination or sequence for metastatic disease), (6) were initiated on a subsequent LOT (ie, 2L1 therapy; line of interest) between June 01, 2017 and September 30, 2021 (United States), or November 30, 2021 (Europe and Japan; given later availability of anti-PD-(L)1 in these regions), and ( 7) had an Eastern Cooperative Oncology Group (ECOG) performance status of 0-2 at this subsequent LOT initiation (ie, index date).Given patients with squamous histology are not commonly tested for ALK and EGFR mutations, negative test results were not required for eligibility.When physicians had multiple eligible patient charts, they were instructed in the CRF to select patients with last names beginning with programmed randomized letters.The study design scheme is presented in Appendix Figure A1.

Measures and Study Outcomes
Physician characteristics, including biomarker testing patterns and prescription practices for mNSCLC (at the drug level), and patient demographic/clinical characteristics were assessed.The most common treatment regimens at the class level and treatment sequences for mNSCLC were reported by LOT, which included all treatments received for initial systemic and maintenance therapy.Temporary interruption or drop in treatment(s) of a regimen was not considered as a change in LOT.No minimum number of cycles was required to define a LOT.The identification of LOTs was based on physician assessment using these instructions included in the CRF.
OS was defined as the number of days from index date until death (any cause).Patients without death were censored at the end of the follow-up period, defined as the earliest of loss to follow-up or data collection time.Time to treatment discontinuation (TTD) was defined as the number of days from index date until treatment discontinuation (including death).Patients without discontinuation were censored at the end of follow-up.Real-world progression-free survival (rwPFS) was defined as the number of days from the index date until real-world progression or death (any cause), whichever occurred first.Patients who did not experience real-world progression were censored at the earliest of the initiation of a next LOT or end of follow-up.Real-world progression was defined using a clinician-anchored approach and was based on radiographic/imaging and/or physical examination/clinical evidence, as documented in the medical chart.

Statistical Analyses
All analyses were descriptive, with no hypothesis testing performed.OS, TTD, and rwPFS were assessed using Kaplan-Meier analyses, with the median time to event along with 95% CIs reported if achieved.Subgroup analysis of end points was also conducted by histology type (ie, squamous and nonsquamous) and region (ie, United States, Europe, and Japan).

Physician Characteristics
Overall, 160 physicians (United States, 42; Europe, 71; Japan, 47) provided 487 patient charts (United States, 141; Europe, 218; Japan, 128).Most physicians were from academic-based practices (54.4%), treated >80 patients in the last year (43.8%), and had >10 years of experience treating patients with mNSCLC (82.5%;Appendix Table A1).Almost all physicians (99.5%) reported biomarker testing (next-generation sequencing [NGS], hotspot, or single gene) at initial mNSCLC diagnosis; physicians performed NGS testing for a mean of 85.2% of their patients.Physicians not testing patients at initial diagnosis reported histology (tissue/cell tumor type) among the reasons not to test at diagnosis (17.1%; higher in Europe [25.8%]than the United States [8.6%] or Japan [7.7%]), with fewer patients with squamous histology tested (mean of 48.2% of their patients with squamous histology).Approximately half (52.3%) of the physicians reported biomarker testing when tumor response was suboptimal/disease progression; these physicians performed NGS testing for a mean of 32.2% of their patients (18.9% in Japan).

DISCUSSION
This contemporary study of patients with mNSCLC represents one of the first analyses of real-world clinical outcomes after previous exposure to both anti-PD-(L)1 and platinum-doublet chemotherapy, in combination or sequentially, among a large, pooled patient sample from the United States, Europe, and Japan.2][3] Following progression after anti-PD-(L)1 and platinum-doublet chemotherapy, over half of the included patients received nonplatinum chemotherapy, with or without a VEGF(R) inhibitor, which was relatively consistent across histology and region.Real-world clinical outcomes were suboptimal, with a median OS of 8.8 months, TTD of 4.3 months, and rwPFS of 5.1 months overall, signaling an unmet need for more efficacious treatment options.
In this study, 50.5% of the patients received nonplatinum chemotherapy and 12.9% received nonplatinum chemotherapy plus VEGF(R) inhibitors after progression with an anti-PD-(L)1 and platinum-doublet chemotherapy.This observation is consistent with the multicenter study conducted by Auclin et al, 24 where 46.0% of patients received taxane monotherapy and 20.1% received taxane plus antiangiogenic therapy in the 2L after progression on 1L immune checkpoint inhibitors plus platinum-doublet chemotherapy.Aside from the proportion of patients using nonplatinum chemotherapy in the current study, the remaining observed treatment patterns were quite heterogeneous, including retreatment with anti-PD-(L)1 agents and/or platinum-doublet chemotherapy.Few patients were observed to receive targeted treatment, because of the large intake of frontline genotyping in these centers (99.5% of physicians reported biomarker testing) and the study design excluding patients with common actionable mutations.Although reasons for treatment selection were not available in the data, it is possible that physicians tried regimens with no proven efficacy or guideline recommendation in the 2L1 setting because of the limited treatment options available.Indeed, Kundu et al 26 observed similarly heterogeneous treatment patterns among patients with mNSCLC, with less than half receiving approved treatment options after progression on anti-PD-(L)1 agents and platinumdoublet chemotherapy.
The limited therapeutic options and use of nonrecommended regimens in 2L1 may have contributed to the poor clinical outcomes observed in this study.The overall median OS (8.8 months) was lower than what has been observed in the standard-of-care arms of recent clinical trials evaluating NSCLC after progression on anti-PD-(L)1 and platinum-doublet chemotherapy. 14,29,30For instance, median OS of the control arm was 11.6 months in Lung-MAP S1800A (72% received docetaxel with or without ramucirumab), 14 11.3 months in CANOPY-2 (docetaxel plus placebo arm), 30 and 10.5 months in CONTACT-01 (docetaxel plus placebo arm). 29This difference may be due to the inclusion of patients with ECOG performance status of two (14.0%), and proportion of patients with brain (18.9%) or liver (39.6%) metastases in this study.Moreover, the definition of anti-PD-(L)1 resistance and extent of its use in early-stage disease may have varied across studies.
5][26][27][28] However, variability in rwPFS can be affected by different disease monitoring practices (eg, frequency of follow-ups) and definitions of progression used in clinical practice, which may limit comparability between real-world studies. 31Patients in the Japanese REACTIVE study had the longest median OS (11.6 months), which was longer than that of the Japan subgroup in the current study (9.6 months), potentially because REACTIVE focused on the approved and guideline-recommended regimen of 2L docetaxel plus ramucirumab (only 12.9% of patients in the current study received nonplatinum chemotherapy plus VEGF[R] inhibitor). 2,25,32][28] Although Auclin et al 24 evaluated patients from several countries across North America and Europe, they did not report findings stratified by region.However, the authors did conduct a subgroup analysis by histology, with patients with adenocarcinoma having longer median OS and rwPFS than patients with squamous cell carcinoma.These results are consistent with the current subgroup analysis showing a trend of better clinical outcomes for patients with nonsquamous relative to squamous histology.
Although TTD is typically associated with PFS in clinical trials of mNSCLC, 33 median TTD in the current study (4.3 months overall) was slightly shorter than rwPFS (5.1 months overall).This discrepancy may be explained by the aforementioned variability in disease progression monitoring in clinical practice 31 or the fact that TTD was estimated on the basis of the last administration date of treatment, where progression would preclude a subsequent dose (and thus discontinuation would appear to occur earlier than disease progression).Additionally, since most patients initiated their line of interest in 2020-2021, the COVID-19 pandemic may have caused delays in patient follow-up to identify progression. 34ven the heterogeneous monitoring and treatment approaches in real-world clinical practice, these study findings may not be directly comparable with clinical trial results.
There were limited patients observed in 3L treatment or later after progression on anti-PD-(L)1 and platinum-doublet chemotherapy received sequentially from 1L-3L since most patients received 1L anti-PD-(L)1 plus platinumdoublet chemotherapy in combination.2][3] Capture of sequential treatment with anti-PD-(L)1 and platinum-doublet chemotherapy may have been limited by early deaths or the available follow-up time, warranting further research in this 3L1 setting.
As with all chart reviews, the findings may have been influenced by selection bias, recall bias (eg, physicians may not have specifically remembered the reasons for including specific notes), and incomplete data on the basis of the omission of a particular answer option.
In conclusion, in this large, real-world analysis of pooled patient chart data from the United States, Europe, and Japan, patients with mNSCLC previously treated with anti-PD-(L)1 and platinum-doublet chemotherapy had heterogeneous treatment patterns after progression.This study contributes to the growing literature describing the poor clinical outcomes among patients with NSCLC who experience progression on anti-PD-(L)1 and platinum-doublet chemotherapy, highlighting the large unmet need for additional treatment options that may improve survival in later lines of therapy.Therefore, clinical trials of novel agents that may offer benefit in this patient population are needed.

TABLE 1 .
Characteristics of Patients With mNSCLC Who Initiated a Subsequent Treatment After Disease Progression on Anti-PD-(L)1 and Platinum-Doublet Chemotherapy Received Concurrently or Sequentially for Metastatic Disease JCO Global Oncology ascopubs.org/journal/go|3Real-World Outcomes After Progression With PDC and IO in mNSCLC a Race/ethnicity was only captured for patients in the United States.b Theoretical follow-up was defined as the period of time from the index date (ie, line of interest initiation) to the date the data were collected.c Actual follow-up was defined as the period of time from the index date (ie, line of interest initiation) to death or the earliest between the date the data were collected or the date of the last clinical visit, if patient remained alive.d Location of cancer metastases categories was not mutually exclusive.

TABLE 2 .
Treatment Regimens for the Subsequent Line of Therapy After Disease Progression on Anti-PD-(L)1 and Platinum-Doublet Chemotherapy Received Concurrently or Sequentially for Metastatic Disease Real-World Outcomes After Progression With PDC and IO in mNSCLC Abbreviations: mNSCLC: metastatic non-small cell lung cancer; PDC, platinum-doublet chemotherapy; VEGF(R), vascular endothelial growth factor receptor.JCO Global Oncologyascopubs.org/journal/go| 5

TABLE A1 .
Physician Characteristics Abbreviation: mNSCLC: metastatic non-small cell lung cancer.JCO Global Oncology ascopubs.org/journal/goReal-World Outcomes After Progression With PDC and IO in mNSCLC