Study design
The DURATION study is an open label, treatment stratified, and randomized phase II study (Figure 1) enrolling patients with histologically confirmed NSCLC (adenocarcinoma and squamous) stage four (metastatic) prior to any systemic treatment. All procedures and time frames displaced in figure 1 and shedule of assessment (Table 3) are developed according to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT). Additional file 1 contains the complete SPIRIT checklist.
Study setting
The DURATION trial is a multicenter trial recruiting patients from approximately 30 sites across Germany. A full list of sites can be obtained at clinicaltrials.gov (NCT03345810).
Study objectives
Primary objective
The primary objective is to investigate the safety and tolerability of sequential therapy consisting of standard of care single agent or doublet chemotherapy followed by durvalumab in comparison to standard of care single-agent or doublet chemotherapy in frail and/or elderly patients.
Secondary objectives
Secondary objectives are to collect information on efficacy, safety and quality of life parameters and to investigate the utility of geriatric assessments for treatment guidance.
Exploratory objectives
Exploratory objectives are to identify potential predictive biomarkers for efficacy variables. To this end, tissue collection and blood sampling will be performed before and during course of disease/treatment. The blood and tissue samples will be subjected to molecular analyses to search for markers of immune response in this population.
Analysis of biomarker data will include correlations with clinical phenotype and tumor PD-L1 expression.
Characteristics of patients
200 frail and/or elderly patients with metastatic non-small cell lung cancer with no targetable molecular alterations (EGFRwt, EML4ALKtransl-) before 1st-line treatment will be included.
Key inclusion criteria are age ≥70 years and/or CCI >1 and/or ECOG >1, previously untreated NSCLC with no targetable molecular alterations (EGFRwt, EML4ALKtransl-) and the availability of a formalin-fixed, paraffin-embedded (FFPE) tumor tissue block (fresh or archival less than three years old or recent) or a minimum of ten unstained slides of tumor sample for biomarker (PD-L1) evaluation. Key exclusion criteria include mixed small cell lung cancer and NSCLC histology, and history of another primary active malignancy or active autoimmune disease.
For a full list of inclusion and exclusion criteria see Table 1.
Procedures for stratification
Patients are stratified according to modified Cancer and Age Research Group (CARG) with the following cut-offs (9,21):
- Total risk score ≤ 3 → doublet chemotherapy
- Total risk score > 3 → single-agent chemotherapy
The aim is to prevent >50% of standard chemotherapy toxicities (CTCAE grade III/IV). The risk score will be determined according to Table 2.
Study procedures
According to a chemotherapy toxicity tool (modified CARG score, Table 2), patients are classified as suitable for doublet chemotherapy (arm A and B) or suitable for a single-agent chemotherapy (arm C and D). Patients are then randomized to receive either four cycles of single agent or doublet chemotherapy or two cycles of single-agent or doublet chemotherapy followed by two cycles of immunotherapy. After four cycles of standard chemotherapy patients receive either follow-up care (arm A and D) or a maintenance therapy with durvalumab for a maximum of two years in the experimental arms B and C. Dose modification and Toxicity management is detailed described in the chapter “Treatment plan” of the protocol. Furthermore, detailed information about permitted or prohibited concomitant treatment is obtained in the protocol. A data safety monitoring board (DSMB) is installed to control, modify or terminate the study. Details are provided in the protocol referring to the DSMB Charta. An overview of all study procedures is presented in table 3.
Treatment arms A and D: standard of care single agent or doublet chemotherapy
Arm A: nab-paclitaxel 100mg/m² on days d1 and d8 and carboplatin area under the curve (AUC) 5 on day 1, every 3 weeks up to four cycles
Arm D: Gemcitabine 1000mg/m² on days d1 and d8, given every 3 weeks, or vinorelbine 30mg/m² on days d1 and d8 every 3 weeks up to four cycles
Treatment arms B and C: 2 cycles of single agent or doublet chemotherapy followed by durvalumab
Arm B: Two cycles of nab-paclitaxel 100mg/m² on days d1 and d8 and carboplatin area under the curve (AUC) 5 on day 1, every 3 weeks followed by durvalumab 1125 mg every 3 weeks for two cycles followed by maintenance with durvalumab 1500 mg every 4 weeks
, Arm C: Two cycles of gemcitabine 1000mg/m² on days d1 and d8, given every 3 weeks or vinorelbine 30mg/m² on days d1 and d8 every 3 weeks up to four cycles followed by durvalumab 1125 mg every 3 weeks for two cycles followed by maintenance with durvalumab 1500 mg every 4 weeks
Tissue and blood collection for exploratory endpoints
Tissue collection
For each patient a FFPE tumor tissue block (archival or recent) or a minimum of ten unstained slides of tumor sample (2-3 µm sections; slices must be recent and collected on slides provided by the sponsor) must be available for biomarker (PD-L1) evaluation as stated in the inclusion criteria. Biopsy should be excisional, incisional or core-needle. Fine-needle aspiration is insufficient. Tumor PD-L1 expression is measured by an immunohistochemistry assay using SP263 antibody. If a re-biopsy upon tumor progression under study treatment is performed, submission of this tumor material is highly valued.
Blood collection
Participation of patients in the biomarker program is voluntary and must be documented in the informed consent form. The time points for blood sampling are before start of any treatment at baseline and after two cycles of chemotherapy, as well as after 20 weeks of study participation for all patients with stable disease or tumor regression and at the time point of detection of tumor growth in patients with disease progression. In arms B and C, blood is additionally collected after the first cycle of durvalumab.
Study endpoints
Primary endpoint
The primary endpoint will be the rate of treatment related grade III/IV adverse events (CTCAE V4.03).
Secondary endpoints
Secondary endpoints will be:
- ORR according to RECIST 1.1 criteria
- Progression-free survival (PFS)
- Overall survival (OS)
- AEs/SAEs according to CTCAE 4.03
- Health-related quality of life (HR-QoL)
Exploratory endpoints
Exploratory analysis on tissue samples
Tumor PD-L1 assessment using SP263 antibody will be performed as part of the clinical study. The results will be used to correlate PD-L1 staining intensity (proportion of positive tumor and immune cells) with durvalumab efficacy.
Exploratory analysis on blood samples
Blood samples that are collected at different time points will be used to characterize the immune response and investigate biological processes before, during and after the administration of the treatment. Phenotypic fluorescence activated cell sorter (FACS) analysis will be used to analyse whole blood samples with respect to the changes in the T-cell composition. Abundances of immunostimulatory cytokines will be quantified by measuring serum pro- and anti-inflammatory cytokines. Analysis of mutational load on cfDNA will be performed.
Statistical analysis
The primary safety endpoint for the study is the occurrence of CTC grade III/IV toxicities assessed from the first dose to 90 days after the last dose of durvalumab. This is also the primary study endpoint on which the sample size calculation is based. According to the results presented at ASCO 2015 by Rizvi it is assumed that the probability for a CTC grade III/IV toxicity for patients from the pooled experimental arms B+C receiving durvalumab amounts to PB+C=0.18 (22). Based on reported data of selected treatment related adverse events (combination chemotherapy nab-paclitaxel/carboplatin (23), mono-chemotherapy gemcitabine/vinorelbine: (6) it is furthermore assumed that the rate of patients with a CTC grade III/IV toxicity in the pooled control arms A+D receiving chemotherapy only amounts to PA+D=0.35. With the planned number of patients of N=200, the assumed difference between these two groups can be detected using a Chi-square test at a two-sided significance level of α=10% with a probability of 1-β=0.80, also taking into account a dropout rate of 15%. Sample size calculation was performed using ADDPLAN v6.1.
It should be noted that the study is not powered to detect significant differences with regard to the efficacy endpoints, since its primary aim is to assess safety and tolerability. Hence, no confirmatory evidence can be drawn from the efficacy evaluation. Accordingly, all p-values for efficacy outcomes are only to be interpreted descriptively and no adjustment for multiple testing will be done.
The null hypothesis for the primary (safety) endpoint of the trial is defined as H0: PB+C = PA+D (i.e., the rate of patients with a CTC grade III/IV toxicity is equal in the pooled experimental arms B+C and the pooled control arms A+D), which is tested against its alternative H1: PB+C ≠ PA+D (i.e., there is a difference between the pooled experimental arms B+C and the pooled control arms A+D with regard to the rate of patients with a CTC grade III/IV toxicity). These hypotheses will be assessed at a two-sided significance level of α=0.1 using a Mantel-Haenszel Chi-square test adjusting for the stratum “adopted combination/not prone to combination”. Missing data for the primary outcome variable will be replaced by using multiple imputation (24). The analysis of the primary endpoint will be based on the Safety Population comprising all patients enrolled who received at least one dose of study medication. Secondary endpoints will be analyzed descriptively. Subgroup analyses according to PD-L1 expression will be performed. A detailed methodology for the statistical analysis will be described in the statistical analysis plan (SAP), which will be finalized before data base lock. Statistical analysis will be done using SAS v9.4 or higher (SAS Institute, Cary, NC).