If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author. Address for correspondence: Natasha B. Leighl, MD, MMSc, FRCPC, FASCO, Division of Medical Oncology, Princess Margaret Cancer Centre, University of Toronto, 7-913 700 University Avenue, Toronto, Ontario, Canada.
Canadian Cancer Trials Group, Kingston, Ontario, CanadaThe Ottawa Hospital Cancer Centre and Ottawa Hospital Research Institute, Ottawa, Ontario, Canada
Canadian Cancer Trials Group, Kingston, Ontario, CanadaThe Ottawa Hospital Cancer Centre and Ottawa Hospital Research Institute, Ottawa, Ontario, Canada
First-line therapy for patients with metastatic NSCLC includes checkpoint inhibitor monotherapy, dual checkpoint inhibition, or combination with chemotherapy. We compared outcomes with combination chemoimmunotherapy versus dual checkpoint inhibition as first-line treatment for patients with metastatic NSCLC.
Methods
This open-label, randomized clinical trial was conducted at 44 sites in Canada and Australia. Patients with treatment-naive, metastatic NSCLC without sensitizing EGFR or ALK alterations were randomized (1:1) to receive treatment with durvalumab plus tremelimumab with or without platinum-doublet chemotherapy. The primary end point was overall survival (OS). Secondary end points were progression-free survival, overall response rate, and safety.
Results
A total of 301 patients were randomized. Median OS was 16.6 months (95% confidence interval [CI]: 12.6–19.1) with chemotherapy plus immunotherapy and 14.1 months (95% CI: 10.6–18.3) with immunotherapy (hazard ratio = 0.88, 90% CI: 0.67–1.16, p = 0.46). Median progression-free survival with chemotherapy plus immunotherapy was 7.7 months (95% CI: 5.5–8.5) and 3.2 months (95% CI: 2.7–5.1) with immunotherapy (hazard ratio = 0.67, 95% CI: 0.52–0.88). The overall response rate with chemoimmunotherapy was 42.4% and 29.3% with immunotherapy (adjusted OR = 1.69, 95% CI: 1.04–2.76). The percentage of patients with grade 3 or higher adverse events was 82% in the chemotherapy plus immunotherapy group and 70% in the immunotherapy group. Exploratory analyses of programmed death-ligand 1 expression and blood-based tumor mutation burden revealed no differential treatment effect on OS.
Conclusions
The addition of chemotherapy to durvalumab plus tremelimumab in the first-line treatment of stage IV NSCLC did not improve survival compared with durvalumab plus tremelimumab alone. Further study is warranted to identify patients that benefit from initial immunotherapy alone versus combination chemotherapy plus immunotherapy as first-line treatment.
Programmed cell death protein-1 (PD-1) axis inhibitors are now standard in the first-line treatment of patients with metastatic NSCLC without oncogenic driver alterations. Current treatments include PD-1 axis inhibitor monotherapy in patients with high programmed death-ligand 1 (PD-L1) tumor expression
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomized, open-label, controlled, phase 3 trial.
Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower 130): a multicentre, randomised, open-label, phase 3 trial.
Nivolumab, a PD-1 inhibitor, plus ipilimumab, (CTLA-4 inhibitor), is also approved for use in patients with PD-L1 tumor expression greater than or equal to 1%
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.
A recent study comparing pembrolizumab plus ipilimumab versus pembrolizumab alone in patients with PD-L1 tumor expression greater than or equal to 50% revealed no improvement in overall and progression-free survival (PFS), with higher toxicity in the dual immunotherapy arm.
Pembrolizumab plus ipilimumab or placebo for metastatic non-small-cell lung cancer with PD-L1 tumor proportion score ≥ 50%: randomized, double-blind phase III KEYNOTE-598 study.
Nevertheless, understanding how to select the optimal regimen for each patient, maximizing benefit while minimizing toxicity, remains a clinical challenge.
Durvalumab, a human IgG1κ anti–PD-L1 monoclonal antibody, and tremelimumab, a fully humanized IgG2 anti–CTLA-4 monoclonal antibody, activate and enhance T cell function by means of distinct and complementary mechanisms.
Although the combination did not improve survival compared with chemotherapy in patients with advanced NSCLC in the MYSTIC trial, outcomes were similar.
Durvalumab with or without tremelimumab vs standard chemotherapy in first-line treatment of metastatic non-small cell lung cancer: the MYSTIC phase 3 randomized clinical trial.
High blood-based tumor mutation burden (bTMB), greater than or equal to 20 mutations per megabase, was identified as a potential predictor of benefit from combination anti–PD-L1 and anti–CTLA-4 therapy.
Durvalumab with or without tremelimumab vs standard chemotherapy in first-line treatment of metastatic non-small cell lung cancer: the MYSTIC phase 3 randomized clinical trial.
Durvalumab plus tremelimumab has been safely combined with platinum chemotherapy with an objective response rate of 51% in patients with advanced NSCLC.
A phase IB study of durvalumab with or without tremelimumab and platinum-doublet chemotherapy in advanced solid tumors: Canadian Cancer Trials Group Study IND226.
Data from the POSEIDON trial reveal that durvalumab, tremelimumab plus platinum chemotherapy yield superior survival, PFS, and response rate compared with chemotherapy alone.
Johnson ML, Cho BC, Luft A, et al. Durvalumab +/− tremelimumab + chemotherapy as first-line treatment for mNSCLC: results from the phase 3 POSEIDON study. Presented at the World Conference on Lung Cancer 2021. September 9, 2021. Virtual. Presidential Symposium PL02.01.
Nevertheless, the relative contribution of chemotherapy added to first-line checkpoint inhibition remains unclear in patients who might benefit from checkpoint inhibitor therapy alone.
We evaluated the efficacy and safety of adding platinum-doublet chemotherapy to first-line treatment with durvalumab and tremelimumab in patients with metastatic NSCLC irrespective of tumor PD-L1 expression. Herein, we report the final analysis of overall survival (OS) in this phase 2 randomized trial.
Materials and Methods
Patients
This phase 2 study was conducted at 44 centers in Canada and Australia. Eligible patients were adults with stage IV (American Joint Commission on Cancer, eighth edition), pathologically confirmed squamous or nonsquamous NSCLC, an Eastern Cooperative Oncology Group performance status score of 0 or 1 (using a 5-point scale with higher scores indicating greater disability),
Patients had not received previous systemic anticancer therapy for advanced or metastatic disease. Key exclusion criteria were the presence of sensitizing EGFR mutations or ALK translocations and a history of active autoimmune disease. Further eligibility criteria are detailed in the Methods section of the Supplementary Material.
The Canadian Cancer Trials Group (CCTG) sponsored this study, maintained the trial database, and collaborated with trial committee members and the Australasian Lung Cancer Trials Group on the study design, data collection, analysis, and interpretation. The CCTG received trial funding from the Canadian Cancer Society Research Institute and AstraZeneca, but these had no role in the study design nor analysis. The trial was conducted in accordance with the International Conference on Harmonization Good Clinical Practice guidelines and the Declaration of Helsinki. All trial participants provided written informed consent. The institutional review board or independent ethics committee at each site approved the conduct of the study. The independent CCTG Data Safety and Monitoring Committee reviewed safety data and planned interim analysis results.
Study Design and Treatment
Patients were randomly assigned in a 1:1 ratio (open label) to receive four cycles of initial therapy with durvalumab (1500 mg every 3 wk) plus tremelimumab (75 mg every 3 wk) plus platinum-doublet chemotherapy (every 3 wk), followed by maintenance therapy with durvalumab (1500 mg every 4 wk starting week 13), plus, in those with nonsquamous carcinoma, pemetrexed (500 mg/m2 every 4 wk) until disease progression or four cycles of durvalumab (1500 mg every 4 wk) plus tremelimumab (75 mg every 4 wk) followed by maintenance therapy with durvalumab alone until disease progression, unacceptable toxicity, or patient withdrawal (1500 mg every 4 wk starting week 17) (Supplementary Fig. 1). Details regarding the treatment regimens are included in the Methods section of the Supplementary Material.
Randomization was performed centrally using Mango software (Montreal, Quebec, Canada) and was stratified according to patient pathologic subtype (squamous or nonsquamous), disease stage (stage IVA or IVB), and smoking status (never versus previous versus current). Treatment continued until confirmed disease progression by immune-related RECIST (iRECIST),
unacceptable toxicity, or patient withdrawal. Selected patients were permitted to continue treatment beyond disease progression if clinical benefit was confirmed. The protocol recommended that patients allocated to the durvalumab plus tremelimumab arm receive platinum-doublet chemotherapy as the next line of treatment.
End Points
The primary end point of this phase 2 study was OS (the time from randomization to the date of death from any cause) in the intent-to-treat population. Key secondary end points included investigator-assessed PFS (the time from randomization to date of the first documented disease progression according to RECIST version 1.1 or death, whichever came first), investigator-assessed overall response rate (ORR) using RECIST 1.1 and iRECIST, and adverse events assessed by the investigator and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0.
Tumor assessments were performed at baseline, week 6, week 12, and then every 12 weeks thereafter. Confirmation of disease progression according to iRECIST occurred at least 4 weeks after disease progression (immune unconfirmed disease progression) was identified.
Prespecified, exploratory analyses included the assessment of efficacy according to tumor PD-L1 tumor cell (TC) expression, using the Ventana PD-L1 SP263 assay (Ventana Medical Systems, Tucson, AZ) and bTMB in baseline plasma samples using the Guardant OMNI next-generation sequencing assay (Guardant Health, Redwood City, CA), as described previously and in the Methods section of the Supplementary Material.
Durvalumab with or without tremelimumab vs standard chemotherapy in first-line treatment of metastatic non-small cell lung cancer: the MYSTIC phase 3 randomized clinical trial.
The primary end point was assessed in the intent-to-treat population by assigned treatment, irrespective of the actual treatment received. Data were censored for patients who were alive at the time of last contact on or before the data cutoff date of December 1, 2019. Assuming a 2-year OS of 40% in the control group, a sample size of 300 was required with 155 survival events to provide a power of 80% to detect a hazard ratio (HR) of 0.67 at a one-sided significance level of 5%, using a log-rank test stratified by stratification factors at randomization. One planned interim analysis of PFS was performed when 154 progression events or deaths had occurred (data cutoff February 28, 2019). The guideline for study termination for futility was based on HR for progression or death of 0.86 or greater, which did not affect the type 1 error of the primary analysis.
The Kaplan-Meier method was used to estimate the probability of OS, PFS, and duration of response, and the Brookmeyer and Crowley method was used to construct the 90% confidence interval (CI) for the median OS values and 95% CI for other median values using a two-sided significance level of 5% for PFS. The Cochran-Mantel-Haenszel test, stratified for stratification factors at randomization, was used to compare objective response rates. Differential treatment effects between subgroups were assessed using the Cox proportional hazards model with interaction action terms included.
Safety and drug exposure analyses were performed in the as-treated population that received at least one dose of protocol therapy. The final analysis was conducted in May 2020.
Results
Patients and Treatment
From February 15, 2017, to November 7, 2018, a total of 301 patients were enrolled at 44 sites in Canada and Australia and were randomly allocated to receive durvalumab, tremelimumab plus chemotherapy (151 patients) or durvalumab plus tremelimumab (150 patients) (Fig. 1). Baseline characteristics were well balanced between the groups, with the exception of a greater proportion of patients over the age of 65 years in the group allocated to receive chemotherapy plus immunotherapy (Table 1). The median duration of treatment in the group that received chemotherapy plus immunotherapy was 7.4 months (range: 0.03–25) and 3.3 months (range: 0.03–28) in the group that received immunotherapy (Supplementary Table 1).
In the intent-to-treat population, 50 patients in the chemotherapy plus immunotherapy group and 77 in the immunotherapy group received at least one subsequent treatment after discontinuing protocol therapy. A total of 54 patients in the immunotherapy group received subsequent systemic therapy; 49 received platinum-doublet chemotherapy (Supplementary Table 2). At the time of data cutoff (December 1, 2019), 37 patients in the chemotherapy plus immunotherapy group and 27 patients in the immunotherapy group were still receiving protocol therapy.
Efficacy
Overall Survival
As of December 1, 2019, the median follow-up time was 16.6 months. A total of 77 patients (51%) in the chemotherapy plus immunotherapy group and 80 (53%) in the immunotherapy group had died. OS was not significantly improved with the addition of chemotherapy (median = 16.6 mo, 95% CI: 12.6–19.1) compared with immunotherapy alone (median = 14.1 mo, 95% CI: 10.6–18.3). The stratified HR for death was 0.88 (90% CI: 0.67–1.16, p = 0.46) (Fig. 2A). The 1-year OS rate was 59% (95% CI: 51%–63%) in the group that received chemotherapy plus immunotherapy and 55% (95% CI: 46%–63%) in the group that received immunotherapy. Survival outcomes were similar across key subgroups (Fig. 2C).
Figure 2OS and PFS in study participants (intention-to-treat analysis). (A) K-M survival by treatment arm. (B) K-M PFS by treatment arm. (C) Risk of death in prespecified subgroups. #, number; CI, confidence interval; CNS, central nervous system; DT, durvalumab + tremelimumab; DT + Chemo, durvalumab + tremelimumab + chemotherapy; HR, hazard ratio; K-M, Kaplan-Meier; NR, not reached; OS, overall survival; PFS, progression-free survival.
As of December 1, 2019, a total of 235 patients had developed disease progression or died, 108 (72%) in the chemotherapy plus immunotherapy arm and 127 (85%) in the immunotherapy arm. PFS was significantly longer with the addition of chemotherapy (median = 7.7 mo, 95% CI: 5.5–8.5) than with immunotherapy alone (median = 3.2 mo, 95% CI: 2.7–5.1). The stratified HR for disease progression or death was 0.67 (95% CI: 0.52–0.88) (Fig. 2B). PFS outcomes were similar across key subgroups (Supplementary Fig. 2).
Tumor Response
As of December 1, 2019, investigator-assessed response rates were higher with the addition of chemotherapy, 42.4% in the group receiving chemotherapy plus immunotherapy and 29.3% in the group receiving immunotherapy (adjusted OR = 1.69, 95% CI: 1.04–2.76, Supplementary Table 3). No significant differences in response rates were found across key subgroups, including by PD-L1 TC expression and bTMB (Supplementary Table 4). Duration of response was similar in both treatment groups with a median duration of partial response of 8.2 months with chemotherapy plus immunotherapy (95% CI: 6.6–10.0) and 7.7 months with immunotherapy (95% CI: 4.1–12.3). Response assessment using iRECIST revealed similar findings (Supplementary Table 3).
Exploratory Analyses
Pretreatment tumor tissue samples were available for 261 patients for central PD-L1 TC testing (87% of those randomized) and bTMB data for 236 patients (78.4% of those randomized). The remaining patients had insufficient tissue samples for submission or PD-L1 scoring or insufficient plasma or cell-free DNA input for bTMB analysis or quality control failures.
Subgroup analysis in prespecified PD-L1 subgroups with TC expression levels of 1%, 25%, and 50% revealed no differential impact of the addition of chemotherapy on OS (Fig. 3A). A consistent benefit in PFS was found (Supplementary Fig. 3), with no significant differential effect across PD-L1 subgroups (Fig. 3B). Subgroup analysis by bTMB using a prespecified threshold of 20 mutations per megabase revealed no evidence of chemotherapy effect on survival outcome in either those with lower or higher bTMB (Fig. 3A). Longer OS was observed in patients with bTMB greater than or equal to 20 mutations per megabase compared with those with lower bTMB in both treatment groups (HR = 1.99, 95% CI: 1.3–3.1). Longer PFS was found with the addition of chemotherapy in those with bTMB less than 20 mutations per megabase (stratified HR = 0.59, 95% CI: 0.41–0.84; Supplementary Fig. 4). Patients with bTMB greater than or equal to 20 mutations per megabase did not derive significant benefit from additional chemotherapy; however, the test for interaction of bTMB with survival and treatment effect was not significant (p = 0.20; Supplementary Fig. 4). Exploratory analysis of combining subgroups with bTMB less than 20 mutations per megabase or higher and PD-L1 TC expression greater than or equal to 25% did not identify a population that derived survival benefit with the addition of chemotherapy (Supplementary Fig. 5).
Figure 3Risk of death and progression according to tumor PD-L1 expression and TMB. (A) Survival in patients with tumor PD-L1 greater than or equal to 25%, TMB less than 20 or greater than or equal to 20 mutations/Mb. (B) PFS in patients with tumor PD-L1 greater than or equal to 25%, TMB less than 20 or greater than or equal to 20 mutations/Mb. CI, confidence interval; DT + Chemo, durvalumab + tremelimumab + chemotherapy; DT, durvalumab + tremelimumab; mutations/Mb, mutations per megabase; NR, not reached; OS, overall survival; PD-L1, programmed death-ligand 1; PFS, progression-free survival; TMB, tumor mutation burden.
Of the 301 patients enrolled in the study, safety was evaluated in the 148 patients who received at least one dose of chemotherapy, durvalumab plus tremelimumab and 149 patients who received at least one dose of durvalumab plus tremelimumab.
All patients who received treatment experienced at least one adverse event. More patients in the chemotherapy plus immunotherapy group experienced grade 3 or greater treatment-related side effects including myelosuppression, 121 (82.0%) compared with 105 (70.5%) (Table 2). The most common adverse events with a potential immunologic cause were rash, endocrine disorders, and diarrhea, with similar frequency in both treatment groups (Supplementary Table 5). Renal injury was more often found in patients receiving chemotherapy plus immunotherapy. More patients discontinued treatment for adverse events attributed to therapy in the chemotherapy plus immunotherapy group, 34 patients (23%) and 21 patients (14%) in those receiving immunotherapy. Treatment-related deaths occurred in five patients who received chemotherapy plus immunotherapy and three patients who received durvalumab plus tremelimumab.
Table 2Adverse Events (Worst Grade Ever on Study) at Least Possibly, Probably, or Definitely Attributed to Protocol Therapy in Greater Than or Equal to 15% of Study Participants Who Received at Least One or More Doses of Durvalumab Plus Tremelimumab With or Without Chemotherapy
Deaths attributed to treatment in the immunotherapy plus chemotherapy arm were from pneumonitis (1), functional decline (1), gastric perforation (1), acute kidney injury plus pneumonia (1), and not otherwise specified (1). Deaths attributed to treatment in the immunotherapy arm were from pneumonitis (1), pneumonitis plus sepsis (1), and not otherwise specified (1).
5 (3)
3 (2)
a Adverse events that required high-dose steroid initiation were deemed serious; adverse events related to cancer progression were excluded.
b Deaths attributed to treatment in the immunotherapy plus chemotherapy arm were from pneumonitis (1), functional decline (1), gastric perforation (1), acute kidney injury plus pneumonia (1), and not otherwise specified (1). Deaths attributed to treatment in the immunotherapy arm were from pneumonitis (1), pneumonitis plus sepsis (1), and not otherwise specified (1).
To our knowledge, this is one of the first randomized studies in patients with metastatic NSCLC to ensure that all patients receive checkpoint inhibitors as part of first-line therapy. The addition of chemotherapy to durvalumab plus tremelimumab as first-line therapy did not improve survival compared with combination immunotherapy alone. This effect was consistent across patient subgroups and in prespecified exploratory analyses using PD-L1 TC expression and bTMB using a cutoff of 20 mutations per megabase. The addition of chemotherapy to immunotherapy did result in significantly better PFS and ORR, although the median duration of response was similar in the two groups. Adverse events were more frequent in patients who received chemotherapy plus immunotherapy, with a higher proportion of patients discontinuing treatment for adverse events in the chemotherapy combination arm and more dose modifications. We confirmed a prognostic impact of bTMB using a prespecified cutoff of 20 mutations per megabase, with significantly longer median survival in those with high bTMB compared with those with low bTMB. In exploratory analysis, PFS was prolonged with the addition of chemotherapy in patients with bTMB less than 20 mutations per megabase, whereas this benefit was not found in those with bTMB greater than or equal to 20 mutations per megabase. Nevertheless, the test for interaction was not significant. This finding warrants further confirmation in independent randomized cohorts.
Whether all patients with advanced NSCLC require initial chemotherapy plus checkpoint inhibition remains an unanswered question. Many patients in our study successfully received a sequential approach, with initial immunotherapy followed by chemotherapy on progression, with the potential for fewer adverse events during first-line therapy. A recent network meta-analysis has suggested that combination chemoimmunotherapy does not confer a survival benefit over single or combination checkpoint inhibitor therapy.
This is being tested in the Eastern Cooperative Oncology Group 5163 trial, comparing first-line immunotherapy followed by sequential (or additional) chemotherapy upon disease progression with first-line combination chemotherapy plus immunotherapy in patients with advanced, PD-L1–expressing, nonsquamous NSCLC (NCT03793179).
PD-L1 tumor expression has been associated with benefit from anti–PD-1 monotherapy and anti–PD-1 plus chemotherapy combinations in patients with advanced NSCLC.
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomized, open-label, controlled, phase 3 trial.
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.
Durvalumab with or without tremelimumab vs standard chemotherapy in first-line treatment of metastatic non-small cell lung cancer: the MYSTIC phase 3 randomized clinical trial.
The recent KEYNOTE-598 study revealed no improvement in OS, PFS, or ORR and higher toxicity when ipilimumab was added to pembrolizumab compared with pembrolizumab alone in patients with high (≥50%) tumor PD-L1 expression.
Pembrolizumab plus ipilimumab or placebo for metastatic non-small-cell lung cancer with PD-L1 tumor proportion score ≥ 50%: randomized, double-blind phase III KEYNOTE-598 study.
The association between TMB in tissue or blood with benefit from combination CTLA-4 plus PD-1 inhibitors remains controversial, although there is growing evidence that high TMB may be associated with greater benefit from checkpoint inhibitors compared with chemotherapy.
The prognostic impact of tumor mutational burden (TMB) in the first-line management of advanced non-oncogene addicted non-small cell lung cancer (NSCLC): a systematic review and meta-analysis of randomized controlled trials.
Patients with higher tissue TMB (cutoff 10 mutations per megabase) had longer survival compared with those with lower TMB (median survival = 23.0 mo and 16.2 mo with nivolumab plus ipilimumab; 16.4 mo and 12.6 mo with chemotherapy, respectively).
In CheckMate 9LA, an exploratory analysis of tissue and bTMB revealed that patients with higher tissue and bTMB experienced greater ORR and PFS with nivolumab, ipilimumab, and short-course chemotherapy than chemotherapy alone, although OS differences were not affected (tissue TMB 10 mutations per megabase, bTMB 16, 20 mutations per megabase).
First-line nivolumab plus ipilimumab plus two cycles chemotherapy versus four cycles chemotherapy in advanced non-small-cell lung cancer: association of blood and tissue mutational burden with efficacy in CheckMate 9LA.
In MYSTIC, patients with higher bTMB (cutoff 20 mutations per megabase) who received durvalumab plus tremelimumab experienced longer survival, (median OS = 21.9 mo) compared with patients who received chemotherapy (median OS = 8.5 mo), suggesting a predictive effect. In a preliminary report from the NEPTUNE study, AstraZeneca reported that in contrast to findings in the MYSTIC trial, treatment with first-line durvalumab plus tremelimumab did not improve OS in patients with bTMB greater than or equal to 20 mutations per megabase versus chemotherapy although study data have not yet been presented.
In our phase 2 randomized study, higher bTMB was prognostic, associated with longer survival duration in all patients receiving durvalumab plus tremelimumab irrespective of the addition of chemotherapy (median OS = 18.7 and 22.0 mo with chemotherapy plus immunotherapy and immunotherapy, respectively). Nevertheless, we are unable to conclude that bTMB greater than or equal to 20 mutations per megabase is a predictive biomarker for durvalumab plus tremelimumab in our study, as we did not include a group who received chemotherapy alone.
Our study has several limitations. Neither treatment arm is currently considered standard of care, although the POSEIDON trial recently revealed improved OS, PFS, and response with durvalumab, tremelimumab plus chemotherapy versus chemotherapy alone (HR OS = 0.77, 95% CI: 0.65–0.92, p = 0.003; HR PFS = 0.72, 95% CI: 0.60–0.86, p = 0.0003, OR response 2.0, 95% CI: 1.4–2.8).
Johnson ML, Cho BC, Luft A, et al. Durvalumab +/− tremelimumab + chemotherapy as first-line treatment for mNSCLC: results from the phase 3 POSEIDON study. Presented at the World Conference on Lung Cancer 2021. September 9, 2021. Virtual. Presidential Symposium PL02.01.
The low rate of subsequent chemotherapy in our study, 16% in patients allocated to chemoimmunotherapy and 44% of those receiving dual immunotherapy, may have also contributed to the negative survival result. Shorter survival times compared with other studies such as CheckMate 227
with dual immunotherapy may be related to the heavy burden of disease in our study population, with a median number of five metastatic sites and 70% with stage IVB disease and lack of diversity. Another limitation was the toxicity of the four-drug regimen which involved four cycles of platinum-based therapy and ongoing pemetrexed maintenance in patients with nonsquamous NSCLC. The rate of grade 3 or higher adverse events in our study was high with chemoimmunotherapy at 82%. Whether a less toxic four-drug regimen could have led to better outcomes is unknown but may be relevant. For example, in the CheckMate 9LA trial,
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.
patients received only short-course chemotherapy plus combination nivolumab and ipilimumab, which was better tolerated than ongoing chemotherapy until disease progression. Although OS was the primary end point of our trial, PFS was a secondary end point assessed every 6 weeks for the first 12 weeks and every 12 weeks thereafter, to account for the varying schedules of treatment administration and patient visits in each arm. Although this could lead to an overestimation of PFS, the median PFS in the dual immunotherapy arm was 3.2 months, similar to what was reported in the durvalumab plus tremelimumab arm of the MYSTIC trial at 3.9 months.
Durvalumab with or without tremelimumab vs standard chemotherapy in first-line treatment of metastatic non-small cell lung cancer: the MYSTIC phase 3 randomized clinical trial.
Our study was powered to detect a HR for survival of 0.67 using a one-sided significance test, on the basis of 2-year survival projections. We believe that OS remains the most clinically relevant end point when assessing the impact of combination chemoimmunotherapy versus immunotherapy alone as initial therapy with sequential chemotherapy poststudy. We were unable to report 2-year survival rates given the very small number of patients alive at this time point. A larger sample size may have had the power to detect smaller differences, although these differences would be modest and of potentially limited clinical relevance. This further underscores the need for validation of biological selection, such as validating TMB and other markers in independent datasets. Finally, the bTMB analyses in our study were exploratory and informed by the literature and technology available at the time of study conduct. Preanalytical and analytical considerations may further affect bTMB results across trials, such as bioinformatic correction for leukocyte variants. Similarly, several factors may lead to variability in the measurement of tissue TMB and established cutoffs, complicating the conclusions that may be drawn from multiple studies.
The promises and challenges of tumor mutation burden as an immunotherapy biomarker: a perspective from the International Association for the Study of Lung Cancer Pathology Committee.
In conclusion, we believe that our results provide additional insight into the relative contribution of chemotherapy to first-line immunotherapy combinations. In the final analysis from this study, median duration of OS with chemotherapy plus durvalumab and tremelimumab was not longer than with durvalumab plus tremelimumab alone. PFS was significantly longer with the addition of chemotherapy to immunotherapy. Although exploratory biomarker analyses in our phase 2 study suggested that patients with bTMB less than 20 mutations per megabase may benefit preferentially from the addition of chemotherapy to combination immunotherapy as initial treatment, this hypothesis-generating finding warrants validation in prospective randomized cohorts.
CRediT Authorship Contribution Statement
Natasha B. Leighl: Conceptualization, Methodology, Validation, Investigation, Resources, Data curation, Writing—original draft, Writing—review and editing, Visualization, Supervision.
Scott Laurie, Brett Hughes, Christopher Lee: Conceptualization, Methodology, Investigation, Resources, Writing—review and editing.
Martin Stockler, Ming Tsao: Conceptualization, Methodology, Validation, Investigation, Resources, Data curation, Writing—review and editing, Visualization, Supervision.
David Hwang, Phillipe Joubert: Methodology, Validation, Investigation, Resources, Data curation, Writing—review and editing.
Keyue Ding: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing—review and editing, Visualization, Supervision.
Francisco Vera-Badillo: Conceptualization, Methodology, Validation, Investigation, Data curation, Writing—review and editing, Visualization, Supervision, Project administration, Funding acquisition.
Acknowledgments
This study (NCT03057106) was supported by grants to the Canadian Cancer Trials Group from the Canadian Cancer Society Research Institute and AstraZeneca. We gratefully acknowledge the participating patients and their families, Central Offices of Canadian Cancer Trials Group, National Health and Medical Research Council, Clinical Trial Centre, and Australasian Lung Cancer Trials Group, all participating investigators and site staff, and Ms. Janice Li, University Health Network, for her assistance in editing figures, tables, and references.
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomized, open-label, controlled, phase 3 trial.
Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower 130): a multicentre, randomised, open-label, phase 3 trial.
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.
Pembrolizumab plus ipilimumab or placebo for metastatic non-small-cell lung cancer with PD-L1 tumor proportion score ≥ 50%: randomized, double-blind phase III KEYNOTE-598 study.
Durvalumab with or without tremelimumab vs standard chemotherapy in first-line treatment of metastatic non-small cell lung cancer: the MYSTIC phase 3 randomized clinical trial.
A phase IB study of durvalumab with or without tremelimumab and platinum-doublet chemotherapy in advanced solid tumors: Canadian Cancer Trials Group Study IND226.
Johnson ML, Cho BC, Luft A, et al. Durvalumab +/− tremelimumab + chemotherapy as first-line treatment for mNSCLC: results from the phase 3 POSEIDON study. Presented at the World Conference on Lung Cancer 2021. September 9, 2021. Virtual. Presidential Symposium PL02.01.
The prognostic impact of tumor mutational burden (TMB) in the first-line management of advanced non-oncogene addicted non-small cell lung cancer (NSCLC): a systematic review and meta-analysis of randomized controlled trials.
First-line nivolumab plus ipilimumab plus two cycles chemotherapy versus four cycles chemotherapy in advanced non-small-cell lung cancer: association of blood and tissue mutational burden with efficacy in CheckMate 9LA.
The promises and challenges of tumor mutation burden as an immunotherapy biomarker: a perspective from the International Association for the Study of Lung Cancer Pathology Committee.
Disclosure: Dr. Leighl reports receiving travel support (CME lectures) from AstraZeneca, Teva Oncotest, and Roche and personal fees (CME lectures, advisory) from Merck Sharp & Dohme, Bristol-Myers Squibb, EMD Serono, Takeda, Novartis, and Puma Biotechnology. Dr. Laurie reports receiving personal fees from Bristol-Myers Squibb. Dr. Goss reports receiving travel support from AstraZeneca and Boehringer Ingelheim; honoraria from Bristol-Myers Squibb, Celgene, and Pfizer; and financial support from AstraZeneca unrelated to the study. Dr. Hughes reports participating in the advisory boards for AstraZeneca, Roche, Merck Sharp & Dohme, Bristol-Myers Squibb, Pfizer, Elsai, and Takeda and receiving institutional research grants from Amgen. Dr. Stockler reports receiving grants from the Canadian Cancer Trials Group, Cancer Australia, and Australian National Health and Medical Research Council, during the conduct of the study, and grants from Astellas, Amgen, AstraZeneca, Bayer, Bionomics, Bristol-Myers Squibb, Celgene, Medivation, Merck Sharp & Dohme, Pfizer, Roche, Sanofi, and Tilray. Dr. Tsao reports receiving personal fees (honoraria) from Merck, Bristol-Myers Squibb, and AstraZeneca. Dr. Hwang reports receiving grants from Merck, AstraZeneca, Novartis, Takeda, and Boehringer Ingelheim and personal fees from Pfizer, Merck, Novartis, Takeda, Roche, and Bayer. Dr. Kulkarni reports receiving personal fees (honoraria) from AstraZeneca, Roche, and Merck. Dr. Blais reports receiving consultancy honoraria from Bristol-Myers Squibb, Merck, AstraZeneca, Roche, and Pfizer. Dr. Joy reports receiving consulting fees from AstraZeneca. Dr. Yadav reports receiving personal fees (honoraria) from AstraZeneca, Merck, Pfizer, Bristol-Myers Squibb, Roche, Takeda, and Sanofi. Dr. Underhill reports receiving grants from AstraZeneca. Dr. Lee reports receiving personal fees (advisory) from AstraZeneca. Dr. Bradbury reports receiving personal fees (advisory, honoraria) from Merck, AbbVie, Eli Lilly, and Boehringer Ingelheim. Dr. Hiltz reports receiving grants from the Canadian Cancer Society Research Institute and AstraZeneca and nonfinancial support from AstraZeneca. Dr. Dancey reports receiving research grants from AstraZeneca. Dr. Vera-Badillo reports receiving research grants from AstraZeneca. The remaining authors declare no conflict of interest.
These data have been presented in part at the American Society of Clinical Oncology 2020 Annual Meeting (Virtual, May 2020, Chicago, Illinois).
Programmed death 1 (PD-1)/programmed death-ligand 1 (PD-L1) inhibitors in combination with platinum-based chemotherapies are currently established as standard care of first-line treatment in NSCLC regardless of PD-L1 expression.1–4 However, the development and the successful outcomes of cytotoxic T-lymphocyte–associated antigen 4 (CTLA-4) inhibition combined with PD-1 or PD-L1 inhibitors in lung cancer urged clinical researchers to further test these agents with chemotherapy (chemo). In this issue of the journal, Leighl et al.
03322-0&id=gr1.jpg){kind=link}
03322-0&id=gr2.jpg){kind=link}
03322-0&id=gr3.jpg){kind=link}