Advertisement

A Randomized Open-Label Phase III Trial Evaluating the Addition of Denosumab to Standard First-Line Treatment in Advanced NSCLC: The European Thoracic Oncology Platform (ETOP) and European Organisation for Research and Treatment of Cancer (EORTC) SPLENDOUR Trial

Open ArchivePublished:June 18, 2020DOI:https://doi.org/10.1016/j.jtho.2020.06.011

      Abstract

      Introduction

      Receptor activator of NF-kB ligand stimulates NF-kB–dependent cell signaling and acts as the primary signal for bone resorption. Retrospective analysis of a large trial comparing denosumab versus zoledronic acid in bone metastatic solid tumors suggested significant overall survival (OS) advantage for patients with lung cancer with denosumab (p = 0.01). The randomized open-label phase III SPLENDOUR trial was designed to evaluate whether the addition of denosumab to standard first-line platinum-based doublet chemotherapy improved OS in advanced NSCLC.

      Methods

      Patients with stage IV NSCLC were randomized in a 1:1 ratio to either chemotherapy with or without denosumab (120 mg every 3–4 wks), stratified by the presence of bone metastases (at diagnosis), Eastern Cooperative Oncology Group performance status, histology, and region. To detect an OS increase from 9 to 11.25 months (hazard ratio [HR] = 0.80), 847 OS events were required. The trial closed prematurely owing to decreasing accrual rate.

      Results

      A total of 514 patients were randomized, with 509 receiving one or more doses of the assigned treatment (chemotherapy: 252, chemotherapy-denosumab: 257). The median age was 66.1 years, 71% were men, and 59% were former smokers. Bone metastases were identified in 275 patients (53%). Median OS (95% confidence interval [CI]) was 8.7 (7.6–11.0) months in the control arm versus 8.2 (7.5–10.4) months in the chemotherapy-denosumab arm (HR = 0.96; 95% CI: 0.78–1.19; one-sided p = 0.36). For patients with bone metastasis, HR was 1.02 (95% CI: 0.77–1.35), whereas for those without, HR was 0.90 (95% CI: 0.66–1.23). Adverse events grade 3 or greater were observed in 40.9%, 5.2%, 8.7% versus 45.5%, 10.9%, 10.5% of patients. Conditional power for OS benefit was less than or equal to 10%.

      Conclusions

      Denosumab was well-tolerated without unexpected safety concerns. There was no OS improvement for denosumab when added to chemotherapy in the intention-to-treat population and the subgroups with and without bone metastases. Our data do not provide evidence of a clinical benefit for denosumab in patients with NSCLC without bone metastases.

      Keywords

      Introduction

      Lung cancer is the leading cause of cancer mortality, and histologically, NSCLC accounts for 85% of lung cancers. The life expectancy of individuals with lung cancer has increased with better treatments, including targeted therapies and immunotherapy.
      Bone metastases are a substantial cause of morbidity in advanced cancer and 30% to 45% of patients with advanced NSCLC will develop bone metastases during the course of their disease, with postmortem documentation in 36%. Retrospective data reveal that two-thirds of patients with bone metastatic disease already had bone metastases at the time of initial diagnosis.
      • Coleman R.E.
      Skeletal complications of malignancy.
      • Coleman R.E.
      Clinical features of metastatic bone disease and risk of skeletal morbidity.
      • Jaukovic L.
      • Ajdinovic B.
      • Jankovic Z.
      • Dugonjic S.
      Incidence and imaging characteristics of skeletal metastases detected by bone scintigraphy in lung cancer patients.
      • Tsuya A.
      • Kurata T.
      • Tamura K.
      • Fukuoka M.
      Skeletal metastases in non-small cell lung cancer: a retrospective study.
      Patients with metastatic bone disease may suffer from skeletal-related events (SREs), such as fractures, pain requiring radiation or bone operation, and spinal cord compression, severely affecting the quality of life. Patients with lung cancer are known to present with a high frequency of SREs
      • Rosen L.S.
      • Gordon D.
      • Tchekmedyian N.S.
      • et al.
      Long-term efficacy and safety of zoledronic acid in the treatment of skeletal metastases in patients with nonsmall cell lung carcinoma and other solid tumors: a randomized, phase III, double-blind, placebo-controlled trial.
      ,
      • Rosen L.S.
      • Gordon D.
      • Tchekmedyian S.
      • et al.
      Zoledronic acid versus placebo in the treatment of skeletal metastases in patients with lung cancer and other solid tumors: a phase III, double-blind, randomized trial-the zoledronic acid lung cancer and other solid tumors study group.
      and the occurrence of NSCLC-associated SREs has been reported to predict a very short life expectancy with virtually no long-term survivors.
      • Tsuya A.
      • Kurata T.
      • Tamura K.
      • Fukuoka M.
      Skeletal metastases in non-small cell lung cancer: a retrospective study.
      Signaling through binding of the receptor activator of NF-kB (RANK) to its ligand (RANKL) was first discovered as a means of communication between T-cells and dendritic cells. RANKL activates osteoclasts for bone resorption and enables mammary gland and secondary lymph node organogenesis.
      • Kong Y.Y.
      • Yoshida H.
      • Sarosi I.
      • et al.
      OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis.
      RANKL also has profound immune-modulating effects because binding to its receptor induces regulatory T-cells and chemoresistance through the activation of multiple signal transduction pathways.
      • Peters S.
      • Clézardin P.
      • Márquez-Rodas I.
      • Niepel D.
      • Gedye C.
      The RANK–RANKL axis: an opportunity for drug repurposing in cancer?.
      ,
      • van Dam P.A.
      • Verhoeven Y.
      • Trinh X.B.
      • et al.
      RANK/RANKL signaling inhibition may improve the effectiveness of checkpoint blockade in cancer treatment.
      As a consequence, RANKL inhibition enhances immune responses and holds promise as an immunotherapeutic agent to treat cancer. RANK and RANKL expressions have been observed in some tumor types with early clinical data, suggesting a potential antitumor effect of RANK pathway inhibitors.
      • Peters S.
      • Clézardin P.
      • Márquez-Rodas I.
      • Niepel D.
      • Gedye C.
      The RANK–RANKL axis: an opportunity for drug repurposing in cancer?.
      The RANKL inhibitor denosumab, a fully human monoclonal IgG2 antibody, is approved for the prevention of SREs in patients with advanced malignancies involving bone, including solid tumors and multiple myeloma.
      In a pivotal phase III trial of denosumab versus zoledronic acid for the treatment of bone metastases in advanced cancer, denosumab delayed first on-study SREs in NSCLC.
      • Henry D.H.
      • Costa L.
      • Goldwasser F.
      • et al.
      Randomized, double-blind study of denosumab versus zoledronic acid in the treatment of bone metastases in patients with advanced cancer (excluding breast and prostate cancer) or multiple myeloma.
      In a post hoc exploratory analysis of 811 patients with lung cancer, denosumab was associated with improved overall survival (OS) versus zoledronic acid. Specifically, in NSCLC, a hazard ratio (HR) of 0.79 (9.5 versus 8.1 mo, 95% confidence interval [CI]: 0.65–0.95, p = 0.01) was described.
      • Scagliotti G.V.
      • Hirsh V.
      • Siena S.
      • et al.
      Overall survival improvement in patients with lung cancer and bone metastases treated with denosumab versus zoledronic acid: subgroup analysis from a randomized phase 3 study.
      The SPLENDOUR trial was designed to address, in a randomized manner, whether the addition of denosumab to standard first-line platinum-based doublet chemotherapy improved OS in advanced NSCLC.

      Material and Methods

      Design

      SPLENDOUR (European Thoracic Oncology Platform [ETOP] 5-12 - European Organisation for Research and Treatment of Cancer [EORTC] 08111), an international, multicenter, randomized, open-label phase III trial, evaluated the addition of denosumab to standard first-line anticancer treatment in advanced NSCLC. Safety was monitored by the ETOP Independent Data Monitoring Committee. The trial is registered with ClinicalTrials.gov, number NCT02129699.

      Patients

      Patients were recruited from 55 centers in Austria, Belgium, France, Germany, Ireland, Italy, Slovenia, Spain, Switzerland, and the United Kingdom.
      Eligible patients were aged 18 years or older, had measurable or assessable stage IV NSCLC, (with or without bone metastases), Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0 to 2, available tumor tissue for translational research, life expectancy of at least 3 months, and adequate renal, hepatic, and hematologic functions. Patients with documented sensitizing EGFR-activating mutation or ALK rearrangements, symptomatic brain metastases, previous chemotherapy or targeted therapy for metastatic disease, or severe and uncorrected hypocalcemia or hypercalcemia were excluded. Screening for EGFR/ALK was optional (following local standards) but strongly encouraged in nonsquamous histology; computed tomography (CT) scan or magnetic resonance imaging of the brain was also not mandatory and only recommended in cases of clinically suspected brain metastasis. After a protocol amendment, which was enforced when most patients had already been randomized, one line of previous therapy with an immune checkpoint inhibitor was allowed.

      Study Treatment and Assessments

      Patients were randomized in a 1:1 ratio to receive either four to six cycles of platinum-based doublet chemotherapy (platinum compound plus gemcitabine or pemetrexed for nonsquamous cell histology) or platinum-based doublet chemotherapy plus denosumab at a dose of 120 mg subcutaneously every 3 to 4 weeks (chemotherapy-denosumab arm). Denosumab was continued beyond disease progression and given concomitantly with subsequent treatment lines for as long as it was tolerated. Zoledronic acid administration was only allowed in the chemotherapy arm when there were skeletal metastases at baseline, as this is a standard of care. Pemetrexed or erlotinib maintenance treatment after doublet chemotherapy was allowed in both arms.
      CT scans for tumor response assessments were performed at baseline, chemotherapy cycle 3, and then every 12 weeks until progression. All patients underwent a positron emission tomography CT or bone scan at baseline for specific bone metastasis status assignment. Bone imaging (including bone scan) during treatment and at progression were also performed if clinically indicated, that is, in the case of suspected bone metastasis. In the event of equivocal results, further confirmation using bone magnetic resonance imaging, CT, radiograph, or biopsy was recommended.
      Adverse events (AEs) were reported following the Common Terminology Criteria for Adverse Events version 4.0.

      Statistical Analysis

      SPLENDOUR used a centralized random assignment of patients using the block design technique, stratified by bone metastasis (presence versus absence), ECOG PS (0 or 1 [0/1] versus 2), histology (squamous versus other), and geographic region (Eastern versus Western versus Southern Europe) to balance pemetrexed availability, maintenance strategy, and standard of care.
      The primary endpoint was the OS. The secondary endpoints included progression-free survival (PFS), tumor response based on Response Evaluation Criteria in Solid Tumors v1.1, and safety. OS and PFS, estimated from the time of randomization, were conducted on the intention-to-treat (ITT) population and analyzed using Cox regression. Comparisons between treatment arms were made by test scores adjusted for stratification factors. Sensitivity analysis was performed on the basis of the log-rank test and Kaplan-Meier estimates, and plots were produced.
      For PFS analysis, an event was defined as disease progression or death, whichever occurred first. Patients without progression were censored at the date of the last follow-up.
      Using 90% power and one-sided type I error of 2.5%, demonstration of an increase in median OS to 11.25 months in the chemotherapy-denosumab arm relative to 9 months in the chemotherapy arm (equivalent to HR = 0.80) required observation of 847 deaths. Assuming an accrual rate of 15 patients per month in the first 6 months and 30 patients per month thereafter, 1000 patients were required to be recruited over a period of 37 months and followed up for an additional 14 months after the randomization of the last patient to reach the required number of events. The trial was designed with a futility interim analysis at 30% of the information time.
      Subgroup analyses for the treatment effect on OS and PFS by bone metastasis status at randomization, ECOG PS, and histology were predefined. The accrual in the bone metastases stratum was expected to be 30%. The same OS improvement was assumed for both bone metastases strata in the chemotherapy-denosumab arm.
      The planned interim analysis for futility was conducted when 274 events (32.4% of information) for the primary endpoint were available. Results were presented to the ETOP Independent Data Monitoring Committee in September 2017, which recommended the continuation of the study as per protocol. However, the steering committee closed the recruitment as of January 2018, considering that completion of accrual was not feasible in the context of recent immunotherapy advances and the multiplicity of immunotherapy-based ongoing clinical trials in this specific clinical scenario. In fact, subsequent survival improvement and regulatory approval of combined anti–programmed death-ligand 1 immunotherapy and chemotherapy would have ethically and practically prevented the continuation of this trial.

      Results

      Patients and Treatment

      Between December 11, 2014 and January 10, 2018, 514 patients were randomized from 55 institutions. The patient flow is summarized in the CONSORT diagram in Figure 1. The analysis includes data available as of January 25, 2018.
      Figure thumbnail gr1
      Figure 1Study design and CONSORT diagram. ITT, intention-to-treat; OS, overall survival; PFS, progression-free survival; PS, performance status; RECIST, Response Evaluation Criteria in Solid Tumors; TR, translational research.
      Of the 514 randomized patients, 509 started treatment (252 chemotherapy; 257 chemotherapy-denosumab). A total of 33 patients were retrospectively considered ineligible (17 chemotherapy, 16 chemotherapy-denosumab). The primary endpoint analysis was performed on the ITT population of all 514 randomized patients. Safety analysis was based on the 509 patients who started treatment. Of note, three patients randomized to chemotherapy actually received denosumab.
      Baseline characteristics were well-balanced between the two treatment arms (Table 1). Most patients had nonsquamous histology (72% overall; 73% chemotherapy; and 71% chemotherapy-denosumab), ECOG PS of 0/1 (89%; 90% chemotherapy, 89% chemotherapy-denosumab), were from Western Europe (61%; 60% chemotherapy, 63% chemotherapy-denosumab), and had a median age of 66.1 years (65.4 chemotherapy, 66.5 chemotherapy-denosumab). Bone metastases at baseline were observed in 54% of patients (54% chemotherapy, 53% chemotherapy-denosumab). Zoledronic acid was administered in 70 patients (27.5%) with chemotherapy.
      Table 1Stratification Factors and Baseline Characteristics of Patients (ITT Population)
      CharacteristicTreatment ArmAll Patients (N = 514)
      Chemotherapy (N = 255)Chemotherapy-Denosumab (N = 259)
      Histology, n (%)
       Squamous65 (25.5)70 (27.0)135 (26.3)
       Nonsquamous187 (73.3)183 (70.7)370 (72.0)
       Mixed3 (1.2)6 (2.3)9 (1.8)
      ECOG PS, n (%)
       0/1229 (89.8)230 (88.8)459 (89.3)
       226 (10.2)29 (11.2)55 (10.7)
      Region, n (%)
       Eastern Europe18 (7.1)10 (3.9)28 (5.4)
       Western Europe153 (60.0)162 (62.5)315 (61.3)
       Southern Europe84 (32.9)87 (33.6)171 (33.3)
      Sex, n (%)
       Male187 (73.3)179 (69.1)366 (71.2)
       Female68 (26.7)80 (30.9)148 (28.8)
      Age, y, n (%)
       Median65.466.566.1
       Range35.4–83.732.1–84.032.1–84.0
      Smoking status, n (%)
       Current smoker87 (34.1)87 (33.6)174 (33.9)
       Former smoker149 (58.4)156 (60.2)305 (59.3)
       Never smoker15 (5.9)15 (5.8)30 (5.8)
       Missing4 (1.6)1 (0.4)5 (1.0)
      Presence of documented sensitizing EGFR-activating mutation or ALK rearrangements,
      Screening following local standards was optional but strongly encouraged in nonsquamous histology.
      n (%)
       No147 (57.6)137 (52.9)284 (55.3)
       Yes2 (0.8)1 (0.4)3 (0.6)
       Not tested, missing106 (41.6)121 (46.7)227 (44.1)
      Bone metastasis, n (%)
       No118 (46.3)121 (46.7)239 (46.5)
       Yes137 (53.7)138 (53.3)275 (53.5)
      Documented brain metastasis, n (%)
       No254 (99.6)258 (99.6)512 (99.6)
       Yes1 (0.4)1 (0.4)2 (0.8)
      ECOG, Eastern Cooperative Oncology Group; ITT, intention-to-treat; PS, performance status.
      a Screening following local standards was optional but strongly encouraged in nonsquamous histology.
      The median duration (range) of the trial treatment was 12.4 weeks (3.0–24.9) in the chemotherapy arm and 21.4 weeks (3.0–148.6) in the chemotherapy-denosumab arm. The median duration of doublet chemotherapy was similar between the two arms (11.9 versus 12.0 weeks for cisplatin, 12.0 versus 12.0 weeks for carboplatin, 12.4 versus 12.0 weeks for gemcitabine, and 12.0 versus 12.0 weeks for pemetrexed), for the chemotherapy arm versus the chemotherapy-denosumab arm, respectively. The median duration for denosumab treatment was 21.4 weeks (range: 3.0–147.7), with a median of six cycles. At the time of the analysis, eight patients (3.2%) in the chemotherapy arm and 36 (14%) in the chemotherapy-denosumab arm were still on treatment (reasons for treatment discontinuation in Supplementary Table 1).

      Efficacy Analysis

      At the cutoff date for the final analysis (January 25, 2018), 159 patients (chemotherapy arm: 77; chemotherapy-denosumab arm: 82) were still on follow-up, with a median follow-up of 19.8 months (95% CI: 16.8–25.3) and 25.3 months (95% CI: 19.7–29.0) for the chemotherapy arm and chemotherapy-denosumab arm, respectively.
      In the ITT cohort of 514 randomized patients, 355 deaths were observed. Median OS was 8.7 months in the chemotherapy arm and 8.2 in the chemotherapy-denosumab arm, corresponding to an HR of 0.96 (95% CI: 0.78–1.19; stratified one-sided p = 0.36) (Fig. 2A), confirmed by sensitivity analysis (log-rank p = 0.34).
      Figure thumbnail gr2
      Figure 2(A) Overall survival by treatment arm in the intention-to-treat population. (B) Progression-free survival by treatment arm in the intention-to-treat population. CI, confidence interval; HR, hazard ratio; N, number of patients; O, observed events.
      On the basis of these results, the null hypothesis H0 cannot be rejected; thus, OS in the chemotherapy-denosumab arm was not found to be significantly superior to that in the chemotherapy arm. Conditional power calculations indicated that, even if recruitment had been completed, the power of detecting a significant OS benefit would be less than 10%.
      Similarly, PFS did not display a significant difference between the two treatment arms (Fig. 2B). The same number of 228 PFS events were observed in both arms (chemotherapy: 192 progressions and 36 deaths without progression; chemotherapy-denosumab: 186 and 42), with an almost identical median PFS (chemotherapy: 4.7 mo, 95% CI: 4.1–5.2; chemotherapy-denosumab: 4.7 mo, 95% CI: 4.2–5.3; stratified Cox one-sided p = 0.46).
      One complete response was observed in each treatment arm, whereas the objective response rate was 29.4% on chemotherapy arm and 30.5% on the chemotherapy-denosumab arm (Fisher’s exact p = 0.85). Stable disease occurred in 39.6% of the patients on chemotherapy and 33.6% on chemotherapy-denosumab, respectively (Table 2).
      Table 2Best Overall Response by Arm (ITT Population)
      Best Overall Response by ArmChemotherapy (N = 255)Chemotherapy-Denosumab (N = 259)Total (N = 514)
      N%N%N%
      Complete response10.410.420.4
      Partial response74297830.115229.6
      Stable disease10139.68733.618836.6
      Progressive disease5120702712123.5
      Not assessable2710.6238.9509.7
      Missing10.801.511.2
      ITT, intention-to-treat.
      Subgroup analyses on the primary endpoint OS are presented in Figure 3 (unadjusted Cox analysis). No significant treatment effect was found in any of the subgroups examined (all p > 0.05), including for the presence or absence of bone metastases at randomization (Supplementary Fig. 1). HRs for patients with and without bone metastases at randomization were 1.02 (95% CI: 0.77–1.35) and 0.90 (95% CI: 0.66–1.23), respectively (interaction p = 0.55). Of note, the interaction of treatment with ECOG PS was found significant (p = 0.027) (Fig. 3). Analogous stratified subgroup results for both OS and PFS are summarized in Supplementary Table 2.
      Figure thumbnail gr3
      Figure 3Overall Survival subgroup analysis (forest plot) of chemotherapy-denosumab and chemotherapy arms. OS HRs are based on unadjusted univariate Cox models. CI, confidence interval; df, degrees of freedom; ECOG PS, Eastern Cooperative Oncology Group Performance Status; HR, hazard ratio; OS, overall survival.
      Bone events were recorded in 48 patients (9.3%) overall, 7.7% and 11% on chemotherapy and chemotherapy-denosumab, respectively (Fisher exact p = 0.13, Supplementary Table 3).

      Safety Analysis

      The number of patients with grade 3, 4, 5 AEs were 103 (40.9%), 13 (5.2%), 22 (8.7%) on chemotherapy versus 117 (45.5%), 28 (10.9%), 27 (10.5%) on chemotherapy-denosumab, respectively. The most common AEs of grade 3 or greater on chemotherapy were hypertension in 20 of the patients (7.9%) (two grade 4), lung infection in 19 patients (7.5%) (three grade 5), dyspnea and fatigue, each in 13 patients (5.2%), and sepsis in 10 patients (4%) (seven grade 5; three grade 4). Similarly, on chemotherapy-denosumab, hypertension was the most frequent AE (40 patients; 15.6%; three grade 4). Lung infection was recorded for 28 patients (10.9%) (two grade 4; three grade 5). Other common AEs on chemotherapy-denosumab included fatigue in 31 (12.1%), dyspnea in 21 (8.2%; one grade 4), nausea in 17 (6.6%), and febrile neutropenia in 13 patients (5.1%, four grade 4; two grade 5). Sepsis was reported for 11 patients (4.3%) (seven grade 4, four grade 5). The full table of AEs of grade 3 or greater by treatment arm and laboratory toxicities are available in Supplementary Tables 4 and 5.
      A total of 22 patients (8.7%) on chemotherapy had serious AEs with a fatal outcome (nine of which were toxic deaths); whereas on chemotherapy-denosumab, 32 (12.4%) had a serious AE of fatal outcome, and nine toxic deaths were observed.

      Discussion

      Although a link between RANKL signaling and breast cancer bone and systemic progression has been established,
      • Gonzalez-Suarez E.
      • Jacob A.P.
      • Jones J.
      • et al.
      RANK ligand mediates progestin-induced mammary epithelial proliferation and carcinogenesis.
      • Schramek D.
      • Leibbrandt A.
      • Sigl V.
      • et al.
      Osteoclast differentiation factor RANKL controls development of progestin-driven mammary cancer.
      • Tan W.
      • Zhang W.
      • Strasner A.
      • et al.
      Tumour-infiltrating regulatory T cells stimulate mammary cancer metastasis through RANKL-RANK signalling.
      much less is known about RANKL signaling in primary tumors from other carcinomas such as lung cancer.
      NF-kB-signaling in tumor epithelial cells plays an important role in the development of lung tumors in NSCLC mouse models,
      • Bassères D.S.
      • Ebbs A.
      • Levantini E.
      • Baldwin A.S.
      Requirement of the NF-kappaB subunit p65/RelA for K-Ras-induced lung tumorigenesis.
      • Duran A.
      • Amanchy R.
      • Linares J.F.
      • et al.
      p62 is a key regulator of nutrient sensing in the mTORC1 pathway.
      • Meylan E.
      • Dooley A.L.
      • Feldser D.M.
      • et al.
      Requirement for NF-kappaB signalling in a mouse model of lung adenocarcinoma.
      • Stathopoulos G.T.
      • Sherrill T.P.
      • Cheng D.S.
      • et al.
      Epithelial NF-kappaB activation promotes urethane-induced lung carcinogenesis.
      • Xia Y.
      • Yeddula N.
      • Leblanc M.
      • et al.
      Reduced cell proliferation by IKK2 depletion in a mouse lung-cancer model.
      • Xue W.
      • Meylan E.
      • Oliver T.G.
      • et al.
      Response and resistance to NF-kB inhibitors in mouse models of lung adenocarcinoma.
      and RANKL could theoretically participate in the elaboration of an NF-kB response in lung tumor cells.
      • Peters S.
      • Meylan E.
      Targeting receptor activator of nuclear factor-kappa B as a new therapy for bone metastasis in non-small cell lung cancer.
      Of particular interest, and supporting the design of this trial, was the finding that RANKL-blocking agents could impair the growth of primary tumors in several mouse models of lung adenocarcinoma, with a predominant effect observed in the presence of KRAS mutation.
      • Faget J.
      • Contat C.
      • Zangger N.
      • Peters S.
      • Meylan E.
      RANKL signaling sustains primary tumor growth in genetically engineered mouse models of lung adenocarcinoma.
      SPLENDOUR was designed to evaluate whether denosumab, given in addition to standard first-line platinum-based doublet chemotherapy and continued across subsequent lines of treatments, improves OS in advanced NSCLC. Denosumab in addition to standard chemotherapy was well-tolerated without major safety concerns. However, the final analysis of SPLENDOUR did not reveal an improvement in OS for the addition of denosumab compared with chemotherapy alone. Subset analyses did not illustrate survival differences between patient cohorts with and without bone metastases and irrespective of histologic subtypes.
      Our results align with the data from a company-sponsored randomized phase II biomarker-driven trial of denosumab versus placebo in NSCLC (NCT01951586), which recruited a similar group of 226 patients in parallel with SPLENDOUR. The reported OS was better at 10.9 months for placebo and 10.7 months for denosumab, respectively (HR = 1.06, 95% CI: 0.75–1.59) compared with SPLENDOUR (median OS around 8 mo). There was no correlation between OS and RANK expression on the tissue (data obtained from ClinicalTrials.gov).
      The SPLENDOUR trial planned to randomize 1000 patients. However, the recruitment stopped prematurely in January 2018 after the randomization of 514 patients, owing to slow accrual. Recruitment, initially very fast, was negatively impacted by a rapidly changing treatment landscape, especially with the advent of frontline immunotherapy becoming available in competitive clinical trials and, subsequently, as the standard of care.
      Although chemotherapy, concomitantly or not with immunotherapy, will most probably remain a key component of lung cancer treatment, SPLENDOUR was unable to reveal any improvement of its activity by adding denosumab. An OS of less than 1 year is in keeping with the pretargeted historical results and preimmunotherapy, in a trial involving unselected patients with NSCLC, notably, including patients with PS 2.
      • Schiller J.H.
      • Harrington D.
      • Belani C.P.
      • et al.
      Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer.
      Serum and tumor samples were collected from patients and, translational analyses are ongoing, aimed at identifying a subset of patients who might benefit from the addition of denosumab.
      In conclusion, denosumab in combination with doublet chemotherapy in patients with stage IV NSCLC did not improve OS, objective response rate, or overall PFS in any subgroup analyzed. There were no new safety concerns.

      Acknowledgments

      The SPLENDOUR trial was sponsored by European Thoracic Oncology Platform (ETOP) and financed by a grant from Amgen. The trial was coordinated by ETOP and European Organisation for Research and Treatment of Cancer (EORTC) in collaboration with the Spanish Lung Cancer Group, Cancer Trials Ireland, the Central European Cooperative Oncology Group, and the Swiss Group for Clinical Cancer Research (SAKK). Funding was provided by Amgen Ltd ., United Kingdom, through a restricted research grant. The authors thank the 514 patients who participated in this trial and their families, the SPLENDOUR investigators at all clinical sites and their teams, the ETOP Independent Data Monitoring Committee, the people at the ETOP coordinating office, and EORTC headquarters and Amgen for supporting the trial.

      Supplementary Data

      References

        • Coleman R.E.
        Skeletal complications of malignancy.
        Cancer. 1997; 80: 1588-1594
        • Coleman R.E.
        Clinical features of metastatic bone disease and risk of skeletal morbidity.
        Clin Cancer Res. 2006; 12: 6243s-6249s
        • Jaukovic L.
        • Ajdinovic B.
        • Jankovic Z.
        • Dugonjic S.
        Incidence and imaging characteristics of skeletal metastases detected by bone scintigraphy in lung cancer patients.
        Vojnosanit Pregl. 2006; 63: 1001-1005
        • Tsuya A.
        • Kurata T.
        • Tamura K.
        • Fukuoka M.
        Skeletal metastases in non-small cell lung cancer: a retrospective study.
        Lung Cancer. 2007; 57: 229-232
        • Rosen L.S.
        • Gordon D.
        • Tchekmedyian N.S.
        • et al.
        Long-term efficacy and safety of zoledronic acid in the treatment of skeletal metastases in patients with nonsmall cell lung carcinoma and other solid tumors: a randomized, phase III, double-blind, placebo-controlled trial.
        Cancer. 2004; 100: 2613-2621
        • Rosen L.S.
        • Gordon D.
        • Tchekmedyian S.
        • et al.
        Zoledronic acid versus placebo in the treatment of skeletal metastases in patients with lung cancer and other solid tumors: a phase III, double-blind, randomized trial-the zoledronic acid lung cancer and other solid tumors study group.
        J Clin Oncol. 2003; 21: 3150-3157
        • Kong Y.Y.
        • Yoshida H.
        • Sarosi I.
        • et al.
        OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis.
        Nature. 1999; 397: 315-323
        • Peters S.
        • Clézardin P.
        • Márquez-Rodas I.
        • Niepel D.
        • Gedye C.
        The RANK–RANKL axis: an opportunity for drug repurposing in cancer?.
        Clin Transl Oncol. 2019; 21: 977-991
        • van Dam P.A.
        • Verhoeven Y.
        • Trinh X.B.
        • et al.
        RANK/RANKL signaling inhibition may improve the effectiveness of checkpoint blockade in cancer treatment.
        Crit Rev Oncol Hematol. 2019; 133: 85-91
        • Henry D.H.
        • Costa L.
        • Goldwasser F.
        • et al.
        Randomized, double-blind study of denosumab versus zoledronic acid in the treatment of bone metastases in patients with advanced cancer (excluding breast and prostate cancer) or multiple myeloma.
        J Clin Oncol. 2011; 29: 1125-1132
        • Scagliotti G.V.
        • Hirsh V.
        • Siena S.
        • et al.
        Overall survival improvement in patients with lung cancer and bone metastases treated with denosumab versus zoledronic acid: subgroup analysis from a randomized phase 3 study.
        J Thorac Oncol. 2012; 7: 1823-1829
        • Gonzalez-Suarez E.
        • Jacob A.P.
        • Jones J.
        • et al.
        RANK ligand mediates progestin-induced mammary epithelial proliferation and carcinogenesis.
        Nature. 2010; 468: 103-107
        • Schramek D.
        • Leibbrandt A.
        • Sigl V.
        • et al.
        Osteoclast differentiation factor RANKL controls development of progestin-driven mammary cancer.
        Nature. 2010; 468: 98-102
        • Tan W.
        • Zhang W.
        • Strasner A.
        • et al.
        Tumour-infiltrating regulatory T cells stimulate mammary cancer metastasis through RANKL-RANK signalling.
        Nature. 2011; 470: 548-553
        • Bassères D.S.
        • Ebbs A.
        • Levantini E.
        • Baldwin A.S.
        Requirement of the NF-kappaB subunit p65/RelA for K-Ras-induced lung tumorigenesis.
        Cancer Res. 2010; 70: 3537-3546
        • Duran A.
        • Amanchy R.
        • Linares J.F.
        • et al.
        p62 is a key regulator of nutrient sensing in the mTORC1 pathway.
        Mol Cell. 2011; 44: 134-146
        • Meylan E.
        • Dooley A.L.
        • Feldser D.M.
        • et al.
        Requirement for NF-kappaB signalling in a mouse model of lung adenocarcinoma.
        Nature. 2009; 462: 104-107
        • Stathopoulos G.T.
        • Sherrill T.P.
        • Cheng D.S.
        • et al.
        Epithelial NF-kappaB activation promotes urethane-induced lung carcinogenesis.
        Proc Natl Acad Sci U S A. 2007; 104: 18514-18519
        • Xia Y.
        • Yeddula N.
        • Leblanc M.
        • et al.
        Reduced cell proliferation by IKK2 depletion in a mouse lung-cancer model.
        Nat Cell Biol. 2012; 14: 257-265
        • Xue W.
        • Meylan E.
        • Oliver T.G.
        • et al.
        Response and resistance to NF-kB inhibitors in mouse models of lung adenocarcinoma.
        Cancer Discov. 2011; 1: 236-247
        • Peters S.
        • Meylan E.
        Targeting receptor activator of nuclear factor-kappa B as a new therapy for bone metastasis in non-small cell lung cancer.
        Curr Opin Oncol. 2013; 25: 137-144
        • Faget J.
        • Contat C.
        • Zangger N.
        • Peters S.
        • Meylan E.
        RANKL signaling sustains primary tumor growth in genetically engineered mouse models of lung adenocarcinoma.
        J Thorac Oncol. 2018; 13: 387-398
        • Schiller J.H.
        • Harrington D.
        • Belani C.P.
        • et al.
        Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer.
        N Engl J Med. 2002; 346: 92-98

      Linked Article