European Journal of Cancer (2015) 51, 45– 54

Available at www.sciencedirect.com

ScienceDirect journal homepage: www.ejcancer.com

Multicentre randomised phase II trial of gemcitabine + platinum, with or without trastuzumab, in advanced or metastatic urothelial carcinoma overexpressing Her2 q Ste´phane Oudard a,⇑, Ste´phane Culine b, Yann Vano a, Franc¸ois Goldwasser c, Christine The´odore d, Thierry Nguyen e, Eric Voog f, Eugeniu Banu a, Annick Vieillefond c, Franck Priou g, Gae¨l Deplanque h, Gwenae¨lle Gravis i, Alain Ravaud j, Jean Michel Vannetzel k, Jean-Pascal Machiels l, Xavier Muracciole m, Marie-France Pichon n, Jacques-Olivier Bay o, Reza Elaidi p, Corine Teghom p, Franc¸ois Radvanyi q, Philippe Beuzeboc r a

Department of Medical Oncology, Georges Pompidou Hospital, University Paris Descartes, Paris, France Department of Medical Oncology, Saint Louis Hospital, Paris, France c Department of Medical Oncology, Cochin Hospital, Paris, France d Department of Medical Oncology, Foch Hospital, Suresnes, France e Department of Medical Oncology, Besanc¸on University Hospital, Besanc¸on, France f Department of Medical Oncology, Victor Hugo Clinic, Le Mans, France g Department of Oncology-Hematology, La Roche-Sur-Yon Hospital, La Roche-Sur-Yon, France h Department of Medical Oncology, Saint Joseph Foundation, Paris, France i Department of Medical Oncology, Paoli Calmettes Institute, Marseille, France j Department of Medical Oncology, Bordeaux University Hospital, Bordeaux, France k Department of Oncology, Clinique Hartman, Neuilly-sur-Seine, France l Department of Medical Oncology, Saint Luc University Clinics, Catholic University of Leuven (IREC/MIRO, pole ONCO), Brussels, Belgium m Department of Medical Oncology, La Timone Hospital, Marseille, France n Oncobiology Laboratory, Rene´ Huguenin Center, St Cloud, France o Department of Oncology, Saint Perrin Hospital, Clermont Ferrand, France p ARTIC Group (Association de Recherche sur les The´rapeutiques innovantes en Cance´rologie), France q Molecular Oncology, Curie Institute, UMR 144, CNRS, Paris, France r Department of Medical Oncology, Curie Institute, Paris, France b

Registry number: clinicaltrials.gov/ct/show/NCT01828736. ⇑ Corresponding author at: Department of Medical Oncology, Ho ˆ pital Europe´en Georges Pompidou, Rene´ Descartes University, 20 rue Leblanc, 75908 Paris Cedex 15, France. Tel.: +33 1 56 09 34 47; fax: +33 1 56 09 24 15. E-mail address: [email protected] (S. Oudard). http://dx.doi.org/10.1016/j.ejca.2014.10.009 0959-8049/Crown Copyright Ó 2014 Published by Elsevier Ltd. All rights reserved.

46

S. Oudard et al. / European Journal of Cancer 51 (2015) 45–54

Received 4 July 2014; received in revised form 10 September 2014; accepted 6 October 2014 Available online 15 November 2014

KEYWORDS Advanced urothelial carcinoma Gemcitabine Platinum salts Trastuzumab Her2 overexpression

Abstract Aim: To investigate the efficacy and safety of gemcitabine and platinum salt, with or without trastuzumab, in patients with locally advanced or metastatic urothelial carcinoma overexpressing Her2. Methods: The main eligibility criterion was Her2 overexpression on immunohistochemistry (IHC 2+ or 3+) of primary tumour tissue confirmed by fluorescence in situ hybridisation (FISH). Patients were randomised to Arm A: gemcitabine 1000 mg/m2 (days 1 and 8) plus either cisplatin (70 mg/m2) or carboplatin (AUC = 5) (day 1 every 3 weeks) or Arm B: added trastuzumab (8 mg/kg loading dose, then 6 mg/kg every 21 days until progression). The primary end-point was progression-free survival (PFS). Results: Among 563 screened patients, 75 (13.3%) were Her2 positive (IHC 2+/3+ and FISH+) and 61 met all eligibility criteria (median age, 64 years; 54/61 males; 50/61 baseline ECOG-PS 0-1; 11 locally advanced and 50 metastatic). There was no significant difference between Arms A and B in median PFS (10.2 versus 8.2 months, respectively, p = 0.689), objective response rate (65.5% versus 53.2%, p = 0.39), and median overall survival (15.7 versus 14.1 months, respectively, p = 0.684). In an exploratory analysis, trastuzumab-treated patients receiving cisplatin rather than carboplatin-based chemotherapy fared better (PFS: 10.6 versus 8.0; OS: 33.1 versus 9.5 months). Myelosuppression was the main grade 3/4 toxicity. A case of grade 3 cardiotoxicity and one death from febrile neutropenia occurred in arm B. Conclusion: The unexpectedly low incidence of Her2 overexpression precluded the detection of a significant difference in efficacy on addition of trastuzumab to platinum-based chemotherapy with gemcitabine. However, the satisfactory tolerance of the combination warrants further studies, especially of the cisplatin-based combination, in well-defined patient subsets. Crown Copyright Ó 2014 Published by Elsevier Ltd. All rights reserved.

1. Introduction Urothelial carcinoma of the bladder is estimated to be the 5th most prevalent cancer in Europe, with a 3- to 4-fold higher likelihood of occurrence in men than in women. The estimated age-standardised rate per 100,000 was 14.4 for incidence and 4.5 for mortality in 2012 (data for both sexes in Europe) [1]. The prognosis of patients with advanced or metastatic urothelial carcinoma (UC) is poor. The standard first-line regimen for advanced UC is cisplatin-based combination therapy, initially MVAC (methotrexate, vinblastine, doxorubicin and cisplatin) and, more recently, the combination gemcitabine + cisplatin (GC). MVAC and GC provide similar median overall survival (OS) (14 and 15.2 months, respectively) but GC has a more favourable toxicity profile [2]. A large proportion (30–50%) of patients, however, is deemed ineligible for cisplatin-based chemotherapy [3]. The alternative gemcitabine + carboplatin combination has given a 30–40% overall response rate (ORR) and a median 4- to 5-month progression-free survival (PFS) [4–6]. The only randomised phase III trial comparing two carboplatin-based chemotherapy regimens is EORTC study 30986 (gemcitabine + carboplatin versus methotrexate/carboplatin/vinblastine) in patients unfit for cisplatin chemotherapy. In the gemcitabine-carbo-

platin arm, median PFS and OS were 5.8 and 9.3 months, respectively [7]. It has been suggested that a way forward in the treatment of UC patients with advanced or metastatic disease may be on the lines of the advances made in the targeted treatment of metastatic breast and gastric carcinoma, where trastuzumab (HerceptinÒ) based therapy has shown substantial benefit in patients presenting tumours with overexpression and/or amplification of the ERBB2 gene, aka Her2 (from human epidermal growth factor receptor 2). Trastuzumab is a recombinant humanised anti-Her2 monoclonal antibody. Unfortunately, the reported incidence of Her2 protein overexpression and gene amplification in UC varies extremely widely (9–81% for overexpression; 0–32% for amplification) and its prognostic value has not been definitively established (Supplementary Table S1) [8,9]). Early case reports reported encouraging results in metastatic (m) UC patients with trastuzumab in combination with gemcitabine/carboplatin or cisplatin [10,11]. These results were, however, not fully borne out in later open studies with either trastuzumab or the dual epidermal growth factor/Her2 tyrosine kinase inhibitor, lapatinib [12,13]. No comparative randomised studies have been reported. Back in 2004, we decided to conduct a randomised phase II study comparing gemcitabine and

S. Oudard et al. / European Journal of Cancer 51 (2015) 45–54

platinum salt, with or without trastuzumab, in patients with advanced or metastatic UC with Her2 overexpression. The primary end-point was PFS. Secondary endpoints were ORR, OS, toxicity, and quality of life. 2. Patients and methods 2.1. Study design This was an open-label comparative, randomised, multicentre phase II trial performed in 17 centres (16 in France and 1 in Belgium) (Registry number: NCT01828736). It was sponsored by ARTIC (Association pour la Recherche de The´rapeutiques Innovantes en Cance´rologie) and approved by the French national ethics committee (Comite´ Consultatif de Protection des Personnes dans la Recherche Biome´dicale) of Ile-deFrance X and the Ethics Committee of the Catholic University of Leuven in Belgium. A signed written informed consent form was obtained from all patients before starting therapy. The study was conducted in accordance with good clinical practice and the ethics principles of the Declaration of Helsinki. 2.2. Eligibility To be eligible, patients had to have stage IV American Joint Committee on Cancer (AJCC) UC (M1 or locally advanced T4b, unresectable), histologically or cytologically documented disease, measurable disease, and to have undergone no prior chemotherapy for mUC. Prior neoadjuvant or adjuvant chemotherapy at least 6 months before enrolment was allowed. Patients who had undergone radiation therapy had to have completed treatment by at least 4 weeks before enrolment. Other eligibility criteria included Eastern Cooperative Oncology Group Performance Status (ECOG-PS) 62, life expectancy P12 weeks, adequate haematological, renal and hepatic function, adequate cardiac function (left ventricular ejection fraction (LVEF) of at least 50% on Multiple Gated Acquisition (MUGA) scan or 2D-echocardiogram) within 4 weeks of randomisation and no evidence of acute ischaemic changes on baseline electrocardiogram. Patients who had received prior trastuzumab therapy or presenting brain metastases or severe lung, liver or renal disease were ineligible. Prestudy laboratory assessments were to be completed within 15 days of randomisation and tumour measurements (computed tomography (CT) scans) within 4 weeks of randomisation. The key eligibility criterion was Her2 positivity as given by Her2 overexpression on immunohistochemistry (IHC) of primary tumour tissue (intensity of staining 2+ or 3+) confirmed by gene amplification testing on the same tissue specimen by fluorescence in situ hybridisation (FISH) [14,15]. The FDA-approved companion

47

diagnostic tests Dako’s HercepTeste (A0485 polyclonal antibody, 1/1500) and HER2 FISH pharmDxe Kit (Dako, Glostrup, Denmark) for the identification of patients with HER2-positive breast cancer were used. Serum was collected for Her2 extracellular domain testing using a quantitative enzyme-linked immunosorbent assay (ELISA) [15]. All Her2 tumour tissue assays were performed by the Pathology Unit of Cochin Hospital (Paris) and all Her2 serum assays by the OncoBiology Laboratory of the Rene´ Huguenin Centre (Saint Cloud). 2.3. Treatment Patients were randomly assigned to receive either chemotherapy without (arm A) or with trastuzumab (arm B) by central randomisation using Pocock and Simon’s minimisation method [16]. Arm A patients were administered gemcitabine (1000 mg/m2, intravenous (IV), 30 min) on days 1 and 8, every 21 days. Administration on day 8 took place only if absolute granulocyte count was P1500/mm3 and platelet count >100,000/mm3. Following gemcitabine administration, patients with a creatinine clearance P60 ml/min/m2 received cisplatin every 21 days (70 mg/m2, 60 min) whereas those with a creatinine clearance