synopsis rxponder
Transcription
synopsis rxponder
RxPONDER Etude de phase III randomisée comparant une hormonothérapie adjuvante standard +/- chimiothérapie chez des patientes atteintes de cancer du sein localisé avec 1-3 N+, RH+ et Her2- dont le score de rechute selon Oncotype DX™ est inférieur ou égal à 25 Sponsor Unicancer Breast group en France SWOG aux USA (et CALCG/NSABP) Coordonateur international : AM Gonzales-Angulo (MD Anderson Cancer Center, Houston) PI France Dr Suzette Delaloge, Institut Gustave Roussy, Villejuif Objectifs Objectif primaire: 1. Déterminer l’effet de la chimiothérapie adjuvante chez des patients atteintes de cancer du sein localisé avec 1 à 3 ganglions envahis, des récepteurs hormonaux positifs et Her2 négatif, qui ont un score de recurrence Oncotype DX® faible ou intermédiaire. Si le bénéfice dépend du score de récurrence (RS), l’étude déterminera le point à partir duquel une chimiothérapie adjuvante est recommandée. Objectifs secondaires: 2. Comparer la survie globale, la survie sans rechute métastatique et l’intervalle sans maladie locale selon l’administration ou non d’une chimiothérapie adjuvante et son interaction avec le RS. 3. Comparer la toxicité observée selon le bras de traitement 4. Développer d’autres tests biologiques pour mesurer le bénéfice potentiel de la chimiothérapie adjuvante et prédire la survie sans rechute, la survie sans rechute métastatique, l’intervalle sans maladie locale et comparer en fonction du score RS. 5. Evaluations prospectives de l’anxiété, la qualité de vie, la fatigue, la cognition, etc selon le bras de traitement. Etude Etude de phase III randomisée internationale ouverte Hypothèse clinique Le rapport bénéfice/risque n’est probablement pas en faveur de l’administration d’une chimiothérapie adjuvante chez les femmes ayant un cancer du sein localisé RH+ Her2- 1-3 N+ lorsque le RS est < 25. Background rationale and Indications of adjuvant chemotherapy after excision of localized breast cancer have been widely expanded. Indeed, prospective randomized trials indicate that patients with HR+ primary breast cancer benefit from the addition of chemotherapy to adjuvant endocrine treatment overall. However, 70-90% of patients who receive chemotherapy do not derive a direct individual benefit from it, since they would not have relapsed in the absence of this treatment, or they do relapse despite it. Adjuvant chemotherapy is also associated with major potential short-term, mid-term and eventually long-term harms. A considerable effort has therefore been made to identify those patients who really need chemotherapy and will really benefit form it. Several retrospective analyses of prospective clinical trials indicate that some patients may not benefit from chemotherapy. Specifically, patients with well-differentiated tumors, low grade, those with high expression of HR, or those with low or intermediate Recurrence Score (RS) as defined by the Oncotype DX® assay. Oncotype Dx® assay has reached level 2a of prognostic biomarkers in breast cancer and is currently used to decide for adjuvant chemotherapy among HR+ N- breast cancer patients. Beside prognosis, this score has been associated with a strong capability to predict for the expected benefit of chemotherapy (Albain et al 2010). The present trial aims at evaluating the safety of avoiding chemotherapy among patients with 1-3 positive nodes and HRpositive disease but an intermediate or low recurrence score (< 25). The trial is designed to determine whether there might be a benefit among sub-risk categories within that population. Background: Data from SWOG 8814 (Albain et al Lancet Oncol 2010) Références principales - 1 Paik S, Shak S, Tang G, et al. A multigene assay to predict recurrence of tamoxifen-treated, nodenegative breast cancer. N Engl J Med 2004 ;351:2817-26. PMID: 15591335. 2 Paik S, Tang G, Shak S, et al. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor–positive breast cancer. J Clin Oncol 2006 ; 24:3726-34. PMID: 16720680. 3 Albain KS, Barlow WE, Shak S, Hortobagyi GN, et al. Prognostic and predictive value of the 21gene recurrence score assay in postmenopausal women with node-positive, oestrogen-receptorpositive breast cancer on chemotherapy: a retrospective analysis of a randomized trial, Lancet Oncology 2010 ;11:55-65, PMID: 20005174. - 4 Berry DA, Cirrincione C, Henderson IC, et al. Estrogen-receptor status and outcomes of modern chemotherapy for patients with node-positive breast cancer. JAMA 2006;295:1658-67. PMID: 16609087. 5 Parker JS, Mullins M, Cheang MC, et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin Oncol 2009 ; 27 :1160-7. PMID : 1920420 Indication Cancer du sein localisé RH+ Her2- 1-3 N+ M0, décision de traitement adjuvant Critères de jugement Critère principal : survie sans rechute invasive Critères secondaires : - Survie globale Survie sans métastase à distance Intervalle libre sans rechute locale - Design de l’étude et plan de traitement Tolerance Qualité de vie Questionnaires spécifiques d’anxiété, fatigue, cognition, etc Toutes les femmes éligibles à l’étape 1 se voient proposer un test Oncotype DX centralisé. - Les femmes ayant un RS > 25 reçoivent une chimiothérapie - Les femmes ayant un RS < 25 se voient proposer une randomisation pour chimiothérapie versus pas de chimiothérapie. Divers schémas de chimiothérapie sont autorisés, de seconde ou troisième génération, dont - schéma type 3/4 FEC 100 suivis – 3/4 docetaxel/21 jours ou 12 paclitaxel hebdomadaire - 6 TAC - 4-6 TC Les patientes pourront être incluses dans d’autres essais cliniques ou études. Une radiothérapie loco-régionale adéquate selon les critères en vigueur doit être administrée. Toutes les patientes recevront une hormonothérapie pour au moins 5 ans adaptée aux standards en cours et à leur statut ménopausique. Le schéma global de l’étude est le suivant: • • • • N+ (1-3) RH-pos HER2-neg Pre et Post Menop. Test OncotypeDx SR > 25 SR ≤ 25 Chimio Discussion avec la patiente N=9400 (1200 en France dans 80 centres) Accepte (N=4000) Refus -> suivi Critères d’inclusion principaux CRITERES D INCLUSION PRINCIPAUX: ETAPE 1 POST OPERATOIRE - femme de 18 ans et plus - cancer du sein de type adénocarcinome infiltrant localise operable, T < T4D, 1-3 ganglions envahis (incluant pN1mi) - cancer du sein infiltrant ER+ - pas de surexpression ou amplification de Her2 - pas de métastase à distance au bilan - PS < 2, fonctions rénale, hématologique, cardiaque et hépatique adéquates - enregistrement dans les 56 jours de la chirurgie définitive - consentement éclairé signé - disponibilité d’un échantillon de tissu tumoral fixé en paraffine archive tissue pour la mesure centralisée de Oncotype DX® CRITERES D INCLUSION PRINCIPAUX: ETAPE 2 RANDOMISATION - RS < 25 Statistiques - Consentement éclairé signe - Randomisation à 84 jours maximum de la chirurgie - Les patientes incluses et randomisées dans le bras sans chimiothérapie peuvent être aussi incluses dans d’autres essais, incluant traitements locaux ou thérapies ciblées, mais à l’exclusion d’études de chimiothérapie. Les patientes incluses et randomisées dans le bras avec chimiothérapie peuvent aussi entrer dans des études de comparaison de chimiothérapies. 9400 femmes doivent être incluses, dont 5600 devraient avoir un RS < 25. 4000 d’entre elles devraient accepter la randomisation 1/1 soit 2000 recevant et 2000 ne recevant pas de chimiothérapie. L’étude est construite pour pouvoir démontrer un bénéfice éventuel de la chimiothérapie même dans certains sous-groupes de RS et pour évaluer le rapport bénéfice/risque de la chimiothérapie adjuvante selon le RS. Les principes généraux des hypothèses statistiques sont en annexe 1 en fin de synopsis Critères de stratification: Study duration - RS 0-13 (bas) vs 14-25 (intermédiaire) - Statut ménopausique (pré- versus post) - Type de geste axillaire : GS seul versus curage Durée des inclusions: 6 ans Durée de suivi: 15 ans Procédures Cft ci-dessous Annexe 1 : principes généraux du plan statistique The primary question is to test whether chemotherapy benefit (if it exists) depends on the Recurrence Score. This interaction is tested in a Cox regression model of DFS. If the interaction of chemotherapy and the linear RS term is statistically significant (two-sided α) and there is a point of equivalence between the two randomized treatments for some RS value in the range 025, then additional steps are undertaken. Based on simulation work, power to find a significant interaction with an equivalence point is 81%. Assuming there is a significant predictive effect of RS on chemotherapy benefit, then a clinical cutpoint for recommending chemotherapy will be estimated. This estimated cutpoint is the upper bound of the 95% confidence interval on the point of equivalence. Kaplan-Meier curves comparing randomized arms will be generated separately for RS values below and above this cutpoint and tested with stratified log-rank tests. Additionally, Kaplan-Meier curves will be generated for specific RS values (possibly grouped due to sparseness), and modeled 5-year and 10-year estimates by RS and treatment will be provided. If there is no statistical interaction between linear RS and chemotherapy, then chemotherapy will be tested in a Cox model adjusting for RS, but without an interaction term. This test will be conducted at a 1-sided α=0.025 since chemotherapy would be expected to improve outcomes. Compliance and dropout. Despite the required discussion between patient and physician prior to randomization, some patients will still be noncompliant after randomization. The estimated sample size includes a 5% crossover rate and assumes that noncompliance depends on RS. For patients randomized to chemotherapy, the assumption is that 5% do not receive chemotherapy and that a patient with RS 0-11 is twice as likely to refuse as one who has RS 12-25. For patients randomized to not receive chemotherapy, the assumption is 5% do receive chemotherapy and that a patient with RS 18-25 is twice as likely to receive chemotherapy as one who has RS 0-17. The assumption is that the noncompliant patients remain in the study and provide follow-up. Thus, in the intent to treat analysis 5% of patients in each treatment group have a treatment opposite to their randomized assignment. This crossover rate was incorporated into the simulations which used a sample size of 3,800. The 3,800 was increased to 4,000 to also accommodate dropout, ineligibility, or withdrawal of consent. Early termination for futility. Apart from statistical testing of outcomes, low accrual and crossovers post-randomization affect the viability of the trial. It is also possible that differential acceptance of randomization across the range of RS from 0 to 25 may affect the statistical power of this study. With respect to accrual the study will employ the usual CTEP guideline to judge whether accrual is within expectations and whether the protocol needs to be amended. After two years of accrual, a committee of five statisticians (one SWOG, two from other cooperative groups, and two from CTEP) will meet and discuss viability of the trial based on accrual, acceptance of randomization, and crossover rate. The result of this discussion will be presented to CTEP and the Data and Safety Monitoring Committee (DSMC) for action. With respect to crossover rate (randomized participants receiving the opposite treatment), the upper limit is set at 15% in the first year and 12% after 2 years. The trial will terminate if the crossover rate exceeds this unless NCI determines that the high crossover rate has been corrected and that the trial remains viable. Crossover rates between 5% and the upper limit can be addressed by longer follow-up without increasing accrual. The target crossover rate is less than 5% total. Interim analysis. Under the assumptions above, we would expect 706 events for the primary analysis of the interaction of RS and chemotherapy. The first interim analysis would be after 24% of the events have been observed or approximately 6 years after initiation of the study. This would correspond to the end of accrual if accrual is uniform and at the expected level. There would be subsequent annual interim analyses thereafter with 36%, 50%, 65%, and 82% in years 7-10 with the final analysis in year 11. The analyses will use the Lan-Demets spending function with a truncation bound. To achieve a cumulative 0.025 1-sided significance level, the interim test α‘s will be 0.0005, 0.0005, 0.00089, 0.00474, 0.01145, respectively, and the final α= 0.0206 so there is little loss of power due to the interim analyses. All of these analyses are expected to be after accrual has finished so a decision to publish early would be based on the interim analysis. We also want to monitor the upper RS group of 14-25 to avoid harming patients if there is early evidence of efficacy in this group. An analysis will be conducted at 4 years to evaluate efficacy of chemotherapy in patients with RS 14-25 to determine if there is a potential significant benefit of chemotherapy early in the trial. If this comparison is statistically significant at p=0.05 (2-sided) then further randomization in patients with RS 14-25 would be suspended. A similar comparison would then also be performed in the RS 0-13 group to determine if the trial should suspend accrual completely. Otherwise, all other analyses would occur after accrual is complete. A Data and Safety Monitoring Committee will oversee the conduct of the study. The Committee consists of four members from outside of the Southwest Oncology Group, 3 Southwest Oncology Group members, 3 non-voting representatives from the National Cancer Institute (NCI), and the Group Statistician (non-voting). The members of this Committee will receive confidential reports every 6 months from the Southwest Oncology Group Statistical Center, and will meet at the Group's bi-annual meetings as necessary. The Committee will be responsible for decisions regarding possible termination and/or early reporting of the study.