original articles

Annals of Oncology Annals of Oncology 25: 2156–2162, 2014 doi:10.1093/annonc/mdu384 Published online 19 August 2014

A phase II randomized study evaluating the addition of iniparib to gemcitabine plus cisplatin as first-line therapy for metastatic non-small-cell lung cancer S. Novello1,†, B. Besse2,†*, E. Felip3, F. Barlesi4, J. Mazieres5, G. Zalcman6, J. von Pawel7, M. Reck8, F. Cappuzzo9, D. Ferry10,‡, E. Carcereny11, A. Santoro12, I. Garcia-Ribas13, G. Scagliotti1 & J.-C. Soria2 1

Department of Oncology, University of Turin, AOU San Luigi, Orbassano, Italy; 2Thoracic Cancer Unit, Department of Medicine, Gustave-Roussy, Villejuif, France; Department of Medical Oncology, Vall d’Hebron Institute of Oncology, Barcelona, Spain; 4Multidisciplinary Oncology and Therapeutic Innovations Department, Aix Marseille University, Assistance Publique Hôpitaux de Marseille, Marseille; 5Department of Pneumology and Allergies, CHU Toulouse Hôpital Larrey, Toulouse; 6 Department of Pneumology, CHU Côte de Nacre, Caen, France; 7Department of Oncology, Asklepios Fachkliniken München-Gauting, Gauting; 8Thoracic Oncology, LungenClinic Grosshansdorf, Member of the German Center for Lung Research (DZL), Grosshansdorf, Germany; 9Department of Medical Oncology, Istituto Toscano Tumori-Ospedale Civile, Livorno, Italy; 10Department of Oncology, New Cross Hospital, Wolverhampton, UK; 11Department of Medical Oncology, Catalan Institute of Oncology, Hospital Germans Trias i Pujol, Badalona, Spain; 12Department of Medical Oncology, Humanitas Cancer Center IRCCS, Milan, Italy; 13 Oncology Division, Sanofi Aventis, Barcelona, Spain 3

Received 22 May 2014; revised 29 July 2014; accepted 5 August 2014

Background: Iniparib is a novel anticancer agent initially considered a poly (ADP-ribose) polymerase (PARP) inhibitor, but subsequently shown to act via non-selective protein modification through cysteine adducts. This randomized phase II study investigated the addition of iniparib to gemcitabine–cisplatin in metastatic non-small-cell lung cancer (NSCLC) patients. Patients and methods: Patients with histologically confirmed stage IV NSCLC were randomized 2 : 1 to receive gemcitabine (1250 mg/m2, days 1/8) and cisplatin (75 mg/m2, day 1) with [gemcitabine/cisplatin/iniparib (GCI)] or without [gemcitabine/cisplatin (GC)] iniparib (5.6 mg/kg, days 1/4/8/11) every 3 weeks for six cycles. The primary end point was the overall response rate (ORR). Secondary objectives included progression-free survival (PFS), overall survival (OS), and safety. The study was not designed for formal efficacy comparison, the control arm being to benchmark results against the literature. Results: One hundred and nineteen patients were randomized (39 GC and 80 GCI). More GCI patients were male (80% GCI and 67% GC) and had PS 0 (61% GCI and 49% GC). The ORR was 25.6% [95% confidence interval (CI) 13.0%– 42.1%] with GC versus 20.0% (95% CI 11.9%–30.4%) with GCI, which did not allow rejection of the null hypothesis (ORR with GCI ≤20%; P = 0.545). Median PFS was 4.3 (95% CI 2.8–5.6) months with GC and 5.7 (95% CI 4.6–6.6) months with GCI (hazard ratio 0.89, 95% CI 0.56–1.40). Median OS was 8.5 (95% CI 5.5 to not reached) months with GC, and 12.0 (95% CI 8.9–17.1) months with GCI (hazard ratio 0.78, 95% CI 0.48–1.27). More GCI patients received second-line treatment (51% GC and 68% GCI). Toxicity was similar in the two arms. Grade 3–4 toxicities included asthenia (28% GC and 8% GCI), nausea (3% GC and 14% GCI), and decreased appetite (10% in each). Conclusions: Addition of iniparib to GC did not improve ORR over GC alone. The GCI safety profile was comparable to GC alone. Imbalances in PS and gender distribution may have impacted study results regarding PFS and OS. Trial Registration: ClinicalTrial.gov Identifier NCT01086254. Key words: iniparib, non-small-cell lung cancer, advanced disease, first-line therapy

introduction Iniparib (4-iodo-3-nitrobenzamide, BSI-201) was originally investigated as a poly (ADP-ribose) polymerase (PARP) *Correspondence to: Dr Benjamin Besse, Thoracic Cancer Unit, Department of Medicine, Gustave Roussy, 114 rue Edouard Vaillant, 94805 Villejuif, France. Tel: +33-1-42-11-43-22; E-mail: [email protected] †

These authors contributed equally to this study. Present address: Lilly Research Centre, Windlesham, UK.

‡

inhibitor. However, later preclinical studies showed that it does not possess typical characteristics of PARP inhibitors, instead inducing cell response by non-selective modification of numerous proteins via cysteine adducts [1]. It is believed that iniparib acts as a pro-drug whose nitro-group is converted into either a nitroso-group or a nitrosyl radical by two alternative reduction processes. Proteomic and transcriptional profiling experiments and short-hairpin RNA synergy screens are consistent with a mechanism, in which the Nrf2-mediated antioxidant response and/or the mitochondrial electron transport chain converts

© The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].

original articles

Annals of Oncology

iniparib into its putative active metabolite. This metabolite was observed to uncouple electron transport from oxidative phosphorylation, leading to the production of reactive oxygen species at cytotoxic levels in an in vitro breast cancer model [2]. Despite promising data from an earlier phase II study, the addition of iniparib to a gemcitabine–carboplatin doublet failed to show a significant benefit in terms of the clinical benefit rate, overall response rate (ORR), progression-free survival (PFS), or overall survival (OS) in a phase III randomized study in metastatic triple-negative breast cancer patients [3, 4]. Iniparib administered either as a single agent or in combination with chemotherapy has shown a predictable and manageable safety profile at the proposed dose and schedule. Standard of care systemic therapy for inoperable non-smallcell lung cancer (NSCLC) is platinum-based doublet chemotherapy. A survival benefit was reported in a meta-analysis comparing platinum agents combined with gemcitabine with other platinum doublets [5]. With a median OS of ∼10 months in this population, additional therapeutic approaches are keenly awaited. The phase II study presented here investigated the potential benefit and safety of adding iniparib to the standard cisplatin (Bristol-Myers Squibb)–gemcitabine (Eli Lilly and Co) doublet for the treatment of metastatic NSCLC.

patients and methods study design This phase II, randomized, open-label, non-comparative study was carried out in five European countries between May 2010 and December 2011. Patients were randomized to gemcitabine 1250 mg/m2 (days 1 and 8) plus cisplatin 75 mg/m2 (day 1), with [gemcitabine/cisplatin/iniparib (GCI)] or without [gemcitabine/cisplatin (GC)] iniparib 5.6 mg/kg (1-h intravenous infusion, days 1, 4, 8, and 11) every 3 weeks for six cycles. A 2 : 1 randomization ratio in favor of GCI was used and randomization was stratified for histological type (squamous versus non-squamous) and smoking status (smoker versus never-smoker). Two dose reductions for gemcitabine or cisplatin were permitted for toxicity and a maximum 2-week treatment delay. Hematology, biochemistry, and vital signs were evaluated on days 1 and 8 of every cycle. Adverse events were graded according to NCI-CTCAE, v4.0. Tumor evaluation was carried out at screening and then every 6 weeks until progression, death, other anticancer treatment, or study cutoff date. After treatment, patients were followed up for survival every 3 months until cutoff. Response was evaluated by investigators according to Response Evaluation Criteria in Solid Tumors (RECIST, v1.1). A confirmatory scan was carried out in patients with complete response (CR) or partial response (PR) at least 4 weeks after the initial documentation of response. The study was approved by the local and national ethics committee and conducted in accordance with the Declaration of Helsinki.

patient eligibility Patients had to be at least 18 years old and have histologically confirmed, squamous or non-squamous stage IV (UICC TNM 7th edition) NSCLC with no prior systemic therapy, measurable disease according to RECIST v1.1, an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0 or 1, and adequate hematologic, renal, and hepatic function. Patients with prior definitive radiotherapy for locally advanced NSCLC were eligible. Patients with a history of cardiac disease or active brain metastases were not eligible. All patients gave written informed consent before enrollment.

Volume 25 | No. 11 | November 2014

statistical analyses The study was not designed for formal comparisons of efficacy end points between arms, the control arm being used to benchmark results against available data for combined cisplatin–gemcitabine. For the GCI arm, 70 patients were needed to provide ∼90% power to reject the null hypothesis that the true ORR with GCI was ≤20%, assuming the true response rate was 36%, using a one-sided exact binomial test at a significance level of 0.05. Thirty-five patients (50% of the GCI arm sample size) were planned in the control arm. The ORR with the 95% confidence interval (CI) was calculated in the intent to treat (ITT) population (i.e. all randomized patients). PFS was defined as the time from randomization to progressive disease (RECIST) or death, whichever was earlier. In the absence of progression or death, patients

Table 1. Patient and disease characteristics at study entry, ITT population GC N patients randomized 39 Males, n (%) 26 (66.7) Age in years, median (range) 58 (29–73) Caucasian, n (%) 38 (97.4) Black, n (%) 1 (2.6) ECOG performance status, n (%) 0 19 (48.7) 1 20 (51.3) Smoker (current or former), n (%) 35 (89.7) Histology, n (%) Adenocarcinoma 28 (71.8) Large cell carcinoma 5 (12.8) Poorly differentiated non-squamous 1 (2.6) Squamous cell carcinoma 5 (12.8) Stage at diagnosis, n (%) I 1 (2.6) III 3 (7.7) IV 35 (89.7) Mean time in months since diagnosis 6.9 (15.6) (SD) N metastatic sites, n (%) 1 4 (10.3) 2–3 23 (59.0) >3 12 (30.8) Location metastases, n (%) Lungs 38 (97.4) Lymph nodes 25 (64.1) Bone 13 (33.3) Pleura 10 (25.6) Adrenal 9 (23.1) Liver 9 (23.1) Brain 1 (2.6) Other 5 (12.8) Prior therapy, n (%) Radiotherapy 10 (25.6) Surgery 7 (17.9)

GCI 80 64 (80.0) 59 (37–73) 77 (96.3) 3 (3.8) 49 (61.3) 31 (38.8) 71 (88.8) 57 (71.3) 6 (7.5) 7 (8.8) 10 (12.5) 1 (1.3) 2 (2.5) 77 (96.3) 2.2 (8.6)

3 (3.8) 58 (72.5) 19 (23.8) 75 (93.8) 60 (75.0) 20 (25.0) 15 (18.8) 25 (31.3) 17 (21.3) 6 (7.5) 14 (17.5) 8 (10.0) 3 (3.8)

ECOG, Eastern Cooperative Oncology Group; ITT, intent to treat; SD, standard deviation.

doi:10.1093/annonc/mdu384 | 

original articles

Annals of Oncology

were censored at the last tumor assessment before cutoff or starting other anticancer therapy. OS was defined as the time from randomization to death. In the absence of death, patients were censored at the cutoff date or the last date they were alive, whichever was earlier. PFS and OS were analyzed with the Kaplan–Meier method. The cutoff date was 1 year after the first dose of the last treated patient. Safety data were summarized with descriptive statistics in patients who received at least one dose of study treatment.

antitumor activity The ORR in the ITT population was lower in the GCI arm than in the GC arm: 20.0% (95% CI 11.9%– 30.4%) versus 25.6% (95% CI 13.0%– 42.1%), and did not allow rejection of the null hypothesis (ORR with GCI ≤20%; P = 0.545; Table 2). Best overall responses included 1 CR and 15 PRs in the 80 GCI patients and 10 PRs in the 39 GC patients. Stable disease was more frequent in the GCI arm than in the GC arm (55% versus 44%, respectively).

results A total of 119 patients were randomized, 39 to the GC arm, and 80 to the GCI arm. Two patients randomized to the GCI arm were not treated due to protocol deviations. Patient and disease baseline characteristics are summarized in Table 1. Characteristics were mostly well balanced between the two treatment arms, although more GCI patients were male (80% GCI and 67% GC) and had PS 0 (61% GCI and 49% GC), while fewer GCI patients had an initial diagnosis of stage I or III (4% GCI and 10% GC). Median treatment duration was 13.9 weeks (range, 3–23 weeks) for the GC arm and 15.0 weeks (range, 3–23 weeks) with GCI. A median number of four cycles was administered in both arms, and a similar proportion of patients in each arm completed the six planned treatment cycles (41% GC and 45% GCI). Median relative dose intensity was 88% and 85% for gemcitabine in the GC and GCI arms, respectively, 85% for cisplatin in both arms, and 91% for iniparib.

survival At the cutoff date, progression or death was reported in 32 GC (82.1%) and 52 GCI patients (65.0%). Median PFS was 4.3 (95% CI 2.8–5.6) months in the GC arm and 5.7 (95% CI 4.6–6.6) months in the GCI arm, with a hazard ratio (HR) of 0.89 (95% CI 0.56–1.40) favoring the GCI arm (Figure 1A). Fourteen GC (35.9%) and 33 GCI patients (41.3%) were alive at last contact. Median OS was 12.0 (95% CI 8.9–17.1) months in the GCI arm and 8.5 (95% CI 5.5 to not reached) months in the GC arm, with an HR of 0.78 (95% CI 0.48–1.27; Figure 1B).

safety Safety was analyzed in the 117 treated patients. Toxicity profiles were similar for the two arms (Table 3). Grade 1–2 hematologic toxicity was widespread, and approximately 50% of the patients

Table 2. Summary of efficacy parameters, ITT population

Best overall response, n (%)a Complete response Partial response Stable disease Progressive disease Not evaluableb Overall response rate (95% CI) ORR by subgroup PS PS 0 (n = 19/n = 49) PS 1 (n = 20/n = 31) Gender Male (n = 26/n = 64) Female (n = 13/n = 16) Progression-free survival Median in months (95% CI) Hazard ratio (95% CI) Overall survival Median in months (95% CI) Hazard ratio (95% CI)

GC (N = 39)

GCI (N = 80)

0 10 (25.6) 17 (43.6) 8 (20.5) 4 (10.3) 25.6% (13.0%–42.1%)

1 (1.3) 15 (18.8) 44 (55.0) 11 (13.8) 9 (11.3) 20.0% (11.9%–30.4%)c

36.8% (16.3%–61.6%) 15.0% (3.2%–37.9%)

22.4% (11.8%–36.6%) 16.1% (5.5%–33.7%)

19.2% (6.6%–39.4%) 38.5% (13.9%–68.4%)

15.6% (7.8%–26.9%) 37.5% (15.2%–64.6%)

4.3 (2.8–5.6) 0.89 (0.56–1.40)

5.7 (4.6–6.6)

8.5 (5.5–NE) 0.78 (0.48–1.27)

12.0 (8.9–17.1)

a

CR and PR confirmed 4 weeks after initial observation. Two patients not treated, two patients withdrew due to AE with tumor assessment 20% of patients and grade 3–4) and hematologic laboratory findings, safety population

Hematologic, n (%) Anemia Neutropenia Thrombocytopenia Leukopenia Febrile neutropenia Non-hematologic, n (%) Asthenia Fatigue Nausea Decreased appetite Vomiting Constipation Pyrexia Diarrhea Weight decreased Tinnitus/hypoacusis Stomatitis Abdominal pain upper Hypertension Dyspnea Pulmonary embolism Deep vein thrombosis Pleural effusion Hyponatremia Hyperkalemia Hypokalemia

GC (N = 39) All grades

Grade 3–4

GCI (N = 78) All grades

Grade 3–4

38 (97.4) 22 (56.4) 26 (66.7) 27 (69.2) 3 (7.7)

4 (10.3) 17 (43.6) 8 (20.5) 10 (25.6) 3 (7.7)

72 (92.3) 49 (62.8) 47 (60.3) 59 (75.6) 0

12 (15.4) 29 (37.2) 18 (23.1) 13 (16.7) 0

31 (79.5) 4 (10.3) 34 (87.2) 16 (41.0) 16 (41.0) 17 (43.6) 5 (12.8) 7 (17.9) 7 (17.9) 7 (17.9) 8 (20.5) 3 (7.7) 9 (23.1) 8 (20.5) 3 (7.7) 2 (5.1) 3 (7.7) 25 (64.1) 13 (33.3) 5 (12.8)

11 (28.2) 2 (5.1) 1 (2.6) 4 (10.3) 3 (7.7) 1 (2.6) 0 0 0 1 (2.6) 0 0 3 (7.7) 4 (10.3) 3 (7.7) 1 (2.6) 0 9 (23.1) 1 (2.6) 2 (5.1)

52 (66.7) 11 (14.1) 49 (62.8) 33 (42.3) 29 (37.2) 28 (35.9) 20 (25.6) 19 (24.4) 16 (20.5) 17 (21.8) 14 (17.9) 13 (16.7) 12 (15.4) 9 (11.5) 5 (6.4) 3 (3.8) 2 (2.6) 42 (53.8) 33 (42.3) 14 (17.9)

6 (7.7) 4 (5.1) 11 (14.1) 8 (10.3) 6 (7.7) 3 (3.8) 0 1 (1.3 1 (1.3) 0 0 1 (1.3) 6 (7.7) 3 (3.8) 5 (6.4) 1 (1.3) 1 (1.3) 9 (11.5) 2 (2.6) 5 (6.4)

(10% GC and 4% GCI), pulmonary embolism (8% GC and 6% GCI), and fatigue (5% in both arms). Differences between arms (including grade 3–4) were apparent for asthenia, nausea, dyspnea, hyponatremia, and febrile neutropenia, which were more frequent in GC patients, while pyrexia and abdominal pain were more common with GCI. Five patients died due to adverse events, three of which were treatment-related (two GC patients, one with septic shock and the other with mental status changes, and cardiac failure in a GCI patient). Seven GC (18%) and 16 GCI (21%) patients discontinued the study due to a wide range of AEs (mostly grade 3–4). AEs led to dose reductions in 56% of the patients and delay in approximately one-third, generally due to hematologic toxicity or asthenia (Table 4). Approximately half of the patients had at least one gemcitabine or iniparib dose omission.

therapy after GC or GCI treatment Second-line therapy was given at the investigator’s discretion and could be started before documented progression. More than half of the patients received further systemic therapy on-study (51% GC and 68% GCI; Table 5), notably pemetrexed (GC 18% and GCI 20%) and docetaxel (GC 18% and GCI 16%). In PFS analyses, more GCI patients were censored for initiating new

 | Novello et al.

Table 4. Treatment modifications

n patients with dose reductions (%) ≥1 gemcitabine dose reduction ≥1 cisplatin dose reduction ≥1 iniparib dose reduction ≥1 AE leading to dose reduction ≥1 G3–4 AE leading to dose reduction n patients with delays (%) ≥1 cisplatin cycle delay ≥1 AE leading to dose delay ≥1 G3–4 AE leading to dose delay n patients with omitted doses (%)a ≥1gemcitabine dose missed ≥1 cisplatin dose missed ≥1 iniparib dose missed

GC (N = 39)

GCI (N = 78)

11 (28.2) 9 (23.1) – 22 (56.4) 15 (38.5)

16 (20.5) 13 (16.7) 1 (1.3) 44 (56.4) 26 (33.3)

19 (48.7) 14 (35.9) 9 (23.1)

36 (46.2) 26 (33.3) 18 (23.1)

18 (46.2) 0 –

38 (48.7) 0 35 (44.9)

a

In a given cycle, if one drug was omitted but another drug in the combination was administered, then the missing drug was considered as omitted for that cycle. AE, adverse event.

Volume 25 | No. 11 | November 2014

original articles

Annals of Oncology

Table 5. Therapy after GC or GCI treatment

Systemic antitumor therapy, n (%) Antineoplastic/immunomodulating agents Investigational agent Zoledronic acid Radiotherapy Surgery

GC (N = 39)

GCI (N = 80)

20 (51.3) 20 (51.3)

54 (67.5) 53 (66.3)

2 (5.1) 0 8 (20.5) 0

7 (8.8) 2 (2.5) 24 (30.0) 3 (3.8)

antitumor therapy (22 patients, 27.5%) compared with GC (4 patients, 10.3%).

discussion Addition of iniparib to standard gemcitabine–cisplatin therapy did not improve the activity of this chemotherapy doublet in metastatic NSCLC patients in terms of the ORR (25.6%, 95% CI 13.0%–42.1% with GC; 20.0%, 95% CI 11.9%–30.4% with GCI). The 25.6% ORR in the GC arm is close to the expected 28% rate [6], showing that patients enrolled in this study were representative of the targeted NSCLC population. The failure of iniparib to add clinical benefit in this context may be influenced by its mechanism of action which is different from that of PARP inhibitors [1]. Selection of PFS rather than ORR may have been a more appropriate end point for measuring benefit, the value of the latter having been questioned, notably in studies with targeted agents [7]. In addition, interpretation of the primary efficacy results may have been influenced by imbalances in various key baseline characteristics, such as PS, gender, and stage. For the secondary efficacy parameters, median PFS was longer in the GCI than in the GC arm (5.7 versus 4.3 months, HR 0.89). However, median PFS in GC-treated patients was shorter than reported in phase III studies conducted in the same population (5.1–6.1 months) [6, 8–11]. Similarly, a marginal trend towards a survival benefit was seen with GCI (median OS of 12.0 versus 8.5 months with GC alone, HR 0.78). Though here again median OS in GC-treated patients was shorter than published reports (9.6–12.5 months). Median OS in this small study may have been confounded by other factors such as subsequent therapy or tumor heterogeneity. To date, no pharmacodynamic markers predicting efficacy have been identified. The addition of iniparib did not significantly alter the GC safety profile, notably with respect to withdrawals due to toxicity, severe events, and reductions or delays for toxicity. Furthermore, most reported differences favored the GCI arm. Addition of iniparib did not impact the incidence of neutropenia or serious systemic infections, with no cases of febrile neutropenia in the GCI arm. The excess of severe asthenia/fatigue reported with GC may be due to the higher proportion of patients with PS 1 at baseline. This proof-of-concept study was not designed for formal comparison of efficacy end points between the test and the

Volume 25 | No. 11 | November 2014

control arms. The purpose of the control arm was to benchmark results against historical data. However, due to the lack of efficacy of the test arm and since the patient population was representative of the studied disease, a single-arm trial design would have led to a similar conclusion. This trial enrolled patients approximately in parallel with a phase III study of gemcitabine and carboplatin with or without iniparib in patients with metastatic squamous NSCLC, the results of which were recently communicated as negative [12]. In conclusion, treatment of metastatic NSCLC patients with 5.6 mg/kg iniparib added to gemcitabine–cisplatin did not give benefit over the chemotherapy doublet alone, and no further clinical development of iniparib in this indication is planned.

acknowledgements We thank Sebastien Lavialle (Sanofi, France) for study management support, Qihua Feng and Eric Charpentier (Sanofi, USA) for statistical programming support, and Sarah MacKenzie (Medi.Axe, France, funded by Sanofi, France) for medical writing.

funding This work was supported by Sanofi.

disclosure IGR is an employee of Sanofi. JCS and BB received institutional financial research support for this study from Sanofi. DF received travel monies and acted as an external consultant to Sanofi. All other authors have no conflicts of interest.

references 1. Liu X, Shi Y, Maag DX et al. Iniparib nonselectively modifies cysteine-containing proteins in tumor cells and is not a bona fide PARP inhibitor. Clin Cancer Res 2012; 18(2): 510–523. 2. Licht S, Cao H, Li Z et al. Mechanism of action of iniparib: stimulation of reactive oxygen species (ROS) production in an iniparib-sensitive breast cancer cell line. Mol Cancer Ther 2011; 10(Suppl 1): Abstract 226. 3. O’Shaughnessy J, Osborne C, Pippen JE et al. Iniparib plus chemotherapy in metastatic triple-negative breast cancer. N Engl J Med 2011; 364(3): 205–214. 4. O’Shaughnessy J, Schwartzberg L, Danso M et al. A randomized phase III study of iniparib (BSI-201) in combination with gemcitabine/carboplatin (G/C) in metastatic triple-negative breast cancer (TNBC). J Clin Oncol 2011; 29(Suppl): Abstract 1007. 5. Le Chevalier T, Scagliotti G, Natale R et al. Efficacy of gemcitabine plus platinum chemotherapy compared with other platinum containing regimens in advanced non-small-cell lung cancer: a meta-analysis of survival outcomes. Lung Cancer 2005; 47(1): 69–80. 6. Scagliotti GV, Parikh P, von Pawel J et al. Phase III study comparing cisplatin plus gemcitabine with cisplatin plus pemetrexed in chemotherapy-naive patients with advanced-stage non-small-cell lung cancer. J Clin Oncol 2008; 26(21): 3543–3551. 7. Di Maio M, Gallo C, De Maio E et al. Methodological aspects of lung cancer clinical trials in the era of targeted agents. Lung Cancer 2010; 67(2): 127–135. 8. Paz-Ares L, Douillard J-Y, Koralewski P et al. Phase III study of gemcitabine and cisplatin with or without aprinocarsen, a protein kinase C-alpha antisense oligonucleotide, in patients with advanced-stage non-small-cell lung cancer. J Clin Oncol 2006; 24(9): 1428–1434.

doi:10.1093/annonc/mdu384 | 

original articles 9. Paz-Ares LG, Biesma B, Heigener D et al. Phase III, randomized, double-blind, placebo-controlled trial of gemcitabine/cisplatin alone or with sorafenib for the first-line treatment of advanced, nonsquamous non-small-cell lung cancer. J Clin Oncol 2012; 30(25): 3084–3092. 10. Reck M, von Pawel J, Zatloukal P et al. Phase III trial of cisplatin plus gemcitabine with either placebo or bevacizumab as first-line therapy for nonsquamous nonsmall-cell lung cancer: AVAiL. J Clin Oncol 2009; 27(8): 1227–1234.

Annals of Oncology 11. Manegold C, van Zandwijk N, Szczesna A et al. A phase III randomized study of gemcitabine and cisplatin with or without PF-3512676 (TLR9 agonist) as first-line treatment of advanced non-small-cell lung cancer. Ann Oncol 2012; 23(1): 72–77. 12. Spigel DR, Kim ES, Lynch TS et al. Randomized phase III trial of gemcitabine (G)/ carboplatin (C) with or without iniparib (I) in patients ( pts) with previously untreated stage IV squamous lung cancer. J Thor Oncol 2013; 8(Suppl 2): Abstract #O15.06.

Annals of Oncology 25: 2162–2166, 2014 doi:10.1093/annonc/mdu442 Published online 5 September 2014

Adjuvant cisplatin-based chemotherapy in nonsmall-cell lung cancer: new insights into the effect on failure type via a multistate approach F. Rotolo1*, A. Dunant1, T. Le Chevalier2, J. -P. Pignon1 & R. Arriagada3 on behalf of the IALT Collaborative Group Departments of 1Biostatistics and Epidemiology; 2Medical Oncology; 3Radiation Oncology, Gustave Roussy, Villejuif, France

Received 17 June 2014; revised 13 August 2014; accepted 20 August 2014

Background: Adjuvant cisplatin-based chemotherapy has become the standard therapy against resected nonsmall-cell lung cancer (NSCLC). Because of variable results on its late effect, we reanalyze the long-term data of the International Adjuvant Lung Cancer Trial (IALT) to describe in details the role of adjuvant chemotherapy. Patients and methods: In the IALT, 1867 patients were randomized between adjuvant cisplatin-based chemotherapy and control, who were followed up for a median of 7.5 years. Of these, 1687 patients were enrolled from 132 centers accepting to report the times to cancer events. We used event history methodology to estimate the effects of adjuvant chemotherapy on the risks of local relapse, distant metastasis, and death. Results: Adjuvant chemotherapy was highly effective against local relapses [HR = 0.73; 95% confidence interval (CI) 0.60– 0.90; P = 0.003] and nonbrain metastases (HR = 0.79; 95% CI 0.66–0.94; P = 0.008) but not against brain metastases (HR = 1.1; 95% CI 0.82–1.4; P = 0.61). The effect on noncancer mortality was nonsignificant during the first 5 years (HR = 1.1; 95% CI 0.81–1.5; P = 0.29), whereas the risk of noncancer mortality was subsequently higher with treatment (HR = 3.6; 95% CI 2.2–5.9; P < 0.001). This harmful effect, however, potentially concerned only about 2% of the patients at 8 years. Conclusion: Adjuvant cisplatin-based chemotherapy reduced the risk of local relapse and of nonbrain metastasis, thereby improving survival. This treatment exerted no residual effect on mortality during the first 5 years, but a higher risk of noncancer mortality was found thereafter. Detailed long-term follow-up is strongly recommended for all patients in randomized trials evaluating adjuvant treatments in NSCLC. Key words: cisplatin, nonsmall-cell lung cancer, adjuvant chemotherapy, randomized trial, multistate model, failure pattern

introduction During the last decade, adjuvant cisplatin-based chemotherapy has been established as an effective treatment in resected nonsmall-cell lung cancer (NSCLC) and it has become the standard therapy in this patient population [1, 2]. The International Adjuvant Lung Cancer Trial (IALT) [3], the first and largest *Correspondence to: Dr Federico Rotolo, Department of Biostatistics and Epidemiology, Gustave Roussy, 114 Rue Edouard-Vaillant – 94805 Villejuif Cedex, France. Tel: +33-142-11-61-28; E-mail: [email protected]

randomized study on this issue, showed a significant overall survival advantage of 4.1% at 5 years in patients receiving adjuvant chemotherapy compared with surgery alone. The joint (Lung Adjuvant Cisplatin Evaluation [4]) LACE analysis included the IALT and the four other largest randomized trials [5–8]. LACE and, more recently, a worldwide individual patient-based metaanalysis [9] confirmed a survival benefit of about 5% at 5 years obtained with adjuvant chemotherapy. Even though this improvement was observed at the classic milestone of 5 years, long-term results should also be explored

© The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].