Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control I ³

Follow-up

Outcome measures and effect size ⁴

Comments

Antonia 2017 (PACIFIC study)

Type of study: RCT, phase 3

Setting and country: Australia, Belgium, Canada, Chile, France, Germany, Greece, Hungary, Italy, Israel, Japan, Korea, Mexico, The Netherlands, Peru, Poland, Singapore, Slovakia, South Africa, Spain, Taiwan, Thailand, Turkey, UK, USA, Vietnam

Funding and conflicts of interest:

Funded by AstraZeneca.

The study was designed by representatives of the sponsor (AstraZeneca) and academic advisors. Multiple authors received grants and fees from pharmaceutical companies. Some authors are employed by pharmaceutical companies / the funder.

Inclusion criteria:

Eligible patients had histologically or cytologically

documented stage III, locally advanced, unresectable

NSCLC according to the Staging Manual

in Thoracic Oncology, version 7, of the International

Association for the Study of Lung Cancer. These

patients had received two or more cycles (defined

according to local practice) of platinum-based

chemotherapy (containing etoposide, vinblastine,

vinorelbine, a taxane (paclitaxel or docetaxel), or

pemetrexed) concurrently with definitive radiation therapy (54 to 66 Gy), in which the mean dose to the lung was less than 20 Gy, the V20 (the

volume of lung parenchyma that received 20 Gy

or more) was less than 35%, or both.

Additional

inclusion criteria were no disease progression

after this treatment, an age of 18 years or older, a World Health Organization performance status

of 0 or 1 (on a 5-point scale in which higher numbers indicate greater disability), an estimated

life expectancy of 12 weeks or longer, and completion of the last radiation dose within 1 to

14 days before randomization

Exclusion criteria:

Key exclusion criteria were previous exposure

to anti–PD-1 or PD-L1 antibodies; receipt of immunotherapy

or an investigational drug within 4 weeks before the first dose (6 weeks for monoclonal

antibodies); active or previous autoimmune

disease (within the past 2 years) or a history of primary immunodeficiency; evidence of uncontrolled,

concurrent illness or ongoing or active

infections; unresolved toxic effects of grade 2 or

higher (according to the Common Terminology

Criteria for Adverse Events (CTCAE)); and grade 2 or higher pneumonitis from previous chemoradiotherapy.

N total at baseline:

Intervention: 476

Control: 237

Important prognostic factors²:

age median (range):

I: 64 (31-84)

C: 64 (23-90)

Sex:

I: 70.2% M

C: 70.0% M

WHO performance-status score 0/ 1

I: 49.2%/ 50.4%

C: 48.1%/ 51.5%

Previous radiotherapy:

I:

<54 Gy: 0.6%

≥54 to ≤66 Gy: 92.9%

>66 to ≤74 Gy :6.3%

C:

<54 Gy: 0%

≥54 to ≤66 Gy: 91.6%

>66 to ≤74 Gy :8.0%

Previous chemotherapy

I:

Induction: 25.8%

Concurrent with radiation therapy: 99.8%

C:

Induction: 28.7%

Concurrent with radiation therapy: 99.6%

Groups comparable at baseline?

Yes

Describe intervention (treatment/procedure/test):

Patients received at least 1 dose of durvalumab (at a dose of 10 mg per kilogram of body weight intravenously)

Describe control (treatment/procedure/test):

Patients received at least 1 dose of placebo.

Length of follow-up:

The overall median

follow-up was 14.5 months (range, 0.2 to 29.9) at the data cut-off point for the interim analysis.

Loss-to-follow-up:

None

Incomplete outcome data:

None

Four patients (3 in the durvalumab group and 1 in the placebo group) were randomized but did not

receive treatment because of patient decision (n=2), neutropenia (n=1), and worsening chronic

obstructive pulmonary disease (n=1). 475 patients in the durvalumab and 234 patients in the placebo group were included in the safety analysis.

Outcome measures and effect size (include 95%CI and p-value if available):

Outcome measure 1 – Overall survival

Overall survival was unplanned for the interim analysis.

Outcome measure 2 – Progression free survival

Stratified hazard ratio for disease progression or death:

0.52 (95% CI, 0.42 to 0.65)

Median PFS (months)

I: 16.8 (13.0-18.1)

C: 5.6 (4.6-7.8)

12 months PFS (%)

I: 55.9 (51.0-60.4)

C: 35.3 (29.0-41.7)

18 months PFS (%)

I: 44.2 (37.7-50.5)

C: 27.0 (19.9-34.5)

The median time to death or distant metastasis (months)

I: 23.2 months (95% CI, 23.2 to not reached)

C: 14.6 months (95% CI,

10.6 to 18.6)

HR: 0.52 (95% CI, 0.39 to 0.69)

Outcome measure 3 – Response rate

Objective response (%):

I: 28.4 (95% CI 24.3–32.9) (n=126)

C: 16.0 (95% CI 11.3–21.6) (n=34)

Duration of response, median months

I: Not reached

C: 13.8 (95% 6.0-not reached)

Outcome measure 4 – Quality of life

NR

Outcome measure 5 - Safety (adverse events and toxicity)

Any cause and grade

I: 460 (96.8%) of the 475

C: 222 (94.9%) of the 234

Grade 3 or 4

I: 142 (29.9%)

C: 61 (26.1%)

Treatment related adverse events

Any grade

I: 322 (67.8%)

C: 125 (53.4%)

Grade 3 or 4

I: 56 (11.8%)

C: 10 (4.4%)

Author’s conclusion

Progression-free survival was significantly longer with durvalumab than with placebo.

The secondary end points also andomiz durvalumab, and safety was similar between

the groups.

Subgroups

The progression-free survival

benefit with durvalumab was observed irrespective

of PD-L1 expression before chemoradiotherapy

(hazard ratio, 0.59 (95% CI, 0.43 to 0.82) for a PD-L1 expression level of <25% and 0.41 (95% CI, 0.26 to 0.65) for a PD-L1 expression level of

≥25%).

Remarks

These are results from an interim analysis of

the randomized, double-blind, international, phase

3 PACIFIC study.

In an ongoing analysis, an external independent

data and safety monitoring committee is assessing

safety. This committee assessed the interim efficacy

analyses.

Antonia 2018 (PACIFIC study, updated results)

See Antonia 2017

Length of follow-up:

The overall median

follow-up was 25.2 months

Loss-to-follow-up:

None

Incomplete outcome data:

None

Outcome measures and effect size (include 95%CI and p-value if available):

Outcome measure 1 – Overall survival

At 12 months

I: 83.1% (95% CI 79.4-86.2)

C: 75.3% (95% CI 69.2-80.4)

At 24 months

I: 66.3% (95% CI 61.7-70.4)

C: 55.6% (95% CI 48.9-61.8)

HR for death 0.68 (99.73% CI 0.47–0.997)

P=0.0025

Outcome measure 2 – Progression free survival

Hazard ratio for disease progression or death:

0.51 (95% CI, 0.41 to 0.63)

Median PFS (months)

I: 17.2 (13.1-23.9)

C: 5.6 (4.6-7.7)

The median time to death or distant metastasis (months)

I: 28.3 months (95% CI, 24.0 to 34.9)

C: 16.2 months (95% CI,

12.5 to 21.1)

HR: 0.53 (95% CI, 0.41 to 0.68)

Outcome measure 3 – Response rate

Objective response (%):

I: 30.0% (95% CI 25.8–34.5) (n=133)

C: 17.8% (95% CI 13.0–34.5) (n=38)

Duration of response, median months

I: Not reached (27.4-not reached)

C: 18.4 (95% 6.7-24.5)

Outcome measure 4 – Quality of life

NR

Outcome measure 5 - Safety (adverse events and toxicity)

Any cause and grade

I: 460 (96.8%) of the 475

C: 222 (94.9%) of the 234

Grade 3 or 4

I: 145 (30.5%)

C: 61 (26.1%)

Grade 5 adverse events of any cause

I: 21 patients (4.4%)

C: 14 patients (6.0%)

Author’s conclusion

Durvalumab therapy resulted in significantly longer overall survival than placebo.

No new safety signals were identified.

Butts 2014 (START study)

Type of study: RCT, phase 3

Setting and country: patients from 33 countries

Worldwide were recruited.

Funding and conflicts of interest: Merck KgaA (Darmstadt, Germany).

Multiple authors received grants and fees from pharmaceutical companies. Some authors are employed by pharmaceutical companies / the funder.

Merck KgaA, the study sponsor, designed the trial in

collaboration with the investigators. The sponsor developed

the protocol and statistical analysis plan, provided the

study drug, coordinated the management of study sites

and the clinical data management, did statistical analyses, and participated in the interpretation of data.

Inclusion criteria:

Eligible patients were those aged 18 years or

older with histologically or cytologically unresectable

stage III non-small-cell lung cancer and an Eastern

Cooperative Oncology Group (ECOG) per formance status

of 0 or 1. Stage was confirmed and documented by CT,

MRI, or PET. We did not require pathological confirmation

of mediastinal nodal involvement and we included all

histological subtypes of non-small-cell lung cancer. Between 4 and 12 weeks before randomisation, patients

had to have completed at least two cycles of platinum based

chemotherapy (given sequentially or concurrently)

with a minimum of 50 Gy of radiation, and have received

confirmation of stable disease or an objective response

after chemoradiotherapy. All patients underwent brain

imaging during screening to exclude brain metastases.

Exclusion criteria:

having undergone any

therapy for lung cancer (other than primary

chemoradiotherapy), including surgery; receipt of any

immunotherapy 28 days before randomisation; and

having metastatic disease or any autoimmune disease.

Further details of eligibility and exclusion criteria are

listed in the appendix.

N total at baseline:

Intervention: 1006

Control: 507

Important prognostic factors²:

Sex:

I: 68% M

C: 680% M

ECOG performance-score 0/ 1

I: 48%/ 52%

C: 41%/ 58%

Previous chemoradiotherapy

I:

Concurrent: 65%

Sequential:35

C:

Concurrent: 65%

Sequential:35

Groups comparable at baseline? Yes

Describe intervention (treatment/procedure/test):

Tecemotide

After randomisation, and 3 days before administration of study drug, one dose of intravenous cyclophosphamide

(300 mg/m², maximum dose 600 mg) was administered.

Patients then received tecemotide (contract manufacturer:

Baxter Pharmaceutical Solutions LLC, Bloomington, IN, USA) or placebo (appendix). Tecemotide consists of the MUC1-derived 25-aminoacid BLP25 lipopeptide, the

immunoadjuvant mono phosphoryl lipid A, and three liposome-forming lipids (cholesterol, dimyristoyl phosphatidyl

glycerol, and dipalmitoyl phosphatidyl choline).

Initial therapy consisted of eight consecutive weekly

sub cutaneous injections of tecemotide (806 μg lipopeptide)

Describe control (treatment/procedure/test):

Corresponding placebo

Length of follow-up:

Median follow-up was 39.9 months (IQR 21.2–48.7) in the tecemotide group and 37.7 months (19.9–49.7) in the placebo group.

Loss-to-follow-up:

Intervention:

5 (0.6%)

Control:

3 (0.7%)

Incomplete outcome data:

Intervention:177 (17.6%)

Control: 97 (19.1%)

Excluded from primary

efficacy analysis population

due to clinical hold (see remarks)

I: 829 assessed in the efficacy analysis (mITT primary analysis population) 1024 assessed in the safety analysis

C: 410 assessed in the efficacy analysis (mITT primary analysis population) 477 assessed in the safety analysis

Outcome measures and effect size (include 95%CI and p-value if available):

Outcome measure 1 – Overall survival

I: 25.6 months (95% CI 22.5-29.2)

C: 22.3 months (95% CI 19.6-25.5)

HR 0.88 95% CI 0.75–1.03)

Subgroup analysis

Concurrent: HR 0.78 (95% CI 0.64-0.96)

Sequential: HR 1.11 (95% CI 0.86-1.46)

Outcome measure 2 – Progression free survival

NR

Outcome measure 3 – Response rate

NR

Outcome measure 4 – Quality of life

NR

Outcome measure 5 - Safety (adverse events and toxicity)

All adverse events

I: 938 (91.6%) of the 1024

C: 432 (90.6%) of the 477

Grade 3 or 4

I: 342 (33.4%)

C: 171 (35.8%)

Treatment related adverse events

Any grade

I: 353 (34%)

C: 129 (27%)

Grade 3 or 4

I: 15 (1%)

C: 5 (1%)

Adverse event leading to death

I: 46 (4%)

C: 35 (7%)

Author’s conclusion

We found no significant difference in overall survival with the administration of tecemotide after

chemoradiotherapy compared with placebo for all patients with unresectable stage III non-small-cell lung cancer.

However, tecemotide might have a role for patients who initially receive concurrent chemoradiotherapy, and further study in this population is warranted.

Remarks

In March, 2010, clinical trials of tecemotide, including

the START trial, were put on hold for enrolment and

treatment after a case of encephalitis occurred in a

phase 2 trial of tecemotide for multiple myeloma.

Subsequent investigations of this patient, an overall

safety analysis of the use of tecemotide in non-small-cell

lung cancer, and introduction of safety measures by

protocol amendment led to the clinical hold being lifted

in June, 2010. At the time of the clinical hold, we had

randomly assigned 1182 of the planned 1322 patients. We

designed a modified intention-to-treat population for the primary analysis by prospectively excluding patients randomly assigned within the 6 months preceding the clinical hold. This approach was based on the assumption

that a minimum of eight weekly doses and two 6-weekly doses (corresponding to about 6 months of treatment) were needed for tecemotide to induce an immunotherapeutic effect on survival. As a result, the sample size was adjusted and 274 excluded patients were replaced.

The sensitivity analysis to assess the potential

effect of the clinical hold suggested that the hold had a

negative impact on the treatment effect of tecemotide in terms of survival that extended beyond the 6 months selected for in the primary analysis population.

Mitchell 2015 (START study, updated results)

See Butts 2014

Length of follow-up: 20 months’ additional

observation beyond the primary analysis, increasing the median follow-up time to 58.7 months (tecemotide) and 57.3 months (placebo).

Outcome measures and effect size (include 95%CI and p-value if available):

Outcome measure 1 – Overall survival

I: 25.8 months (95% CI 23.1-29.4)

C: 22.4 months (95% CI 19.6-25.4)

(HR 0.89 95% CI 0.77–1.03)

Subgroup analysis

Concurrent: HR 0.81 (95% CI 0.68-0.98)

Sequential: HR 1.04 (95% CI 0.82-1.31)

Author’s conclusion

Updated OS data support potential treatment benefit with tecemotide in patients treated with concurrent chemoradiotherapy.

Exploratory biomarker analyses suggest that elevated sMUC1 or ANA levels correlate with tecemotide

benefit.

Subgroup analyses

High sMUC1 and ANA correlated with a possible

survival benefit with tecemotide (interaction P = 0.0085 and 0.0022) and might have future value as biomarkers. Interactions between lymphocyte count, NLR, or prespecified HLA alleles and treatment effect were not observed.

Study reference

(first author, publication year)

Describe method of randomisation¹

Bias due to inadequate concealment of allocation?²

(unlikely/likely/andomi)

Bias due to inadequate blinding of participants to treatment allocation?³

(unlikely/likely/andomi)

Bias due to inadequate blinding of care providers to treatment allocation?³

(unlikely/likely/andomi)

Bias due to inadequate blinding of outcome assessors to treatment allocation?³

(unlikely/likely/andomi)

Bias due to selective outcome reporting on basis of the results?⁴

(unlikely/likely/andomi)

Bias due to loss to follow-up?⁵

(unlikely/likely/andomi)

Bias due to violation of

intention to treat analysis?⁶

(unlikely/likely/andomi)

Antonia 2017/ Antonia 2018 (PACIFIC study)

Patients were randomly assigned in a 2:1 ratio according to the randomisation scheme produced by a computer software program generated by the Biostatistics Group, AstraZeneca, or delegate called Grand (AstraZeneca Global Randomisation system) that incorporates a standard procedure for generating randomisation numbers. One randomisation list will be produced for each of the randomisation strata. A blocked randomisation will be generated and all centres will use the same list in order to minimise any imbalance in the number of patients assigned to each treatment arm. (Information extracted from study protocol)

Unlikely

Unlikely

Unlikely

Unlikely

Unlikely

Unlikely

Unlikely

Butts 2014/ Mitchell 2015 (START study)

Patients were randomly assigned on a double-blind basis

in a 2:1 ratio to receive tecemotide or placebo using a

central interactive voice randomisation system (Almac,

Craigavon, Northern Ireland, UK); the interactive voice

randomisation system staff assigned patients and were

not involved in the rest of the trial. Block randomisation

was used to ensure balanced populations in 24 prespecifi ed

strata consisting of disease stage (IIIA versus IIIB), response to primary chemoradiotherapy (stable disease versus objective response), type of primary chemo radiotherapy (concurrent versus sequential), and region (North America

(Canada, USA) and Australia versus western Europe versus rest of

world (Mexico, Central and South America, eastern Europe, and Asia)).

Unlikely

Unlikely

Unlikely

Unlikely

Unlikely

Unlikely

Likely