Randomized, phase 3 trial of inotuzumab ozogamicin plus rituximab <i>versus</i> chemotherapy plus rituximab for relapsed/refractory aggressive B‐cell non‐Hodgkin lymphoma

Inotuzumab ozogamicin (InO), a humanized anti-CD22 antibody–calicheamicin conjugate, demonstrated preliminary antitumour activity and manageable toxicity in phase 1/2 trials for the treatment of relapsed/refractory (R/R) B-cell non-Hodgkin lymphoma (B NHL), as a single- agent (Advani et al, 2010; Ogura et al, 2010) and in combination with rituximab (R-InO) (Advani et al, 2010; Ogura et al, 2010, 2012; Fayad et al, 2013). Given this preliminary evidence, a 2-arm, randomized, open-label, phase 3 study (NCT01232556) was conducted to compare the efficacy and safety of R-InO with investigator's choice (IC) of rituximab plus bendamustine (R-B) or rituximab plus gemcitabine (R-G), in adults with R/R CD20+/CD22+ aggressive B-NHL who were not candidates for high-dose chemotherapy, with or without transplant (see Supplemental Methods and Table SI for eligibility criteria, dose regimens and dose-delay/-reduction criteria). The primary endpoint was overall survival (OS); 2 interim analyses (IAs) were planned when 40% and 70% of OS events were reached (see Supplemental Methods for details of assessments and statistical methods). The trial was to be terminated for futility if P > 0·29 [hazard ratio (HR) > 0·9] or P > 0·10 (HR > 0·82) at the first or second IA, respectively, or if P < 0·0073 for efficacy at the second analysis. The planned IA based on ~40% of OS events (108 events) conducted in May 2013 yielded an estimated HR > 0·9 for OS in the R-InO versus IC arm; enrollment was thus stopped for futility. Reported here are the final data from this trial (locked on 24 July 2014) to inform future research and potential clinical studies in this difficult-to-treat patient population. Patient enrollment occurred between February 2011 and May 2013; 338 patients were randomized [R-InO, n = 166; IC, n = 172 (R-B, n = 137; R-G, n = 35); Fig S1]. Nearly all patients (91%) had diffuse large B-cell lymphoma (DLBCL) at baseline; 68% were aged ≥65 years (Table SII). Age was the primary reason why enrolled patients were not candidates for high-dose chemotherapy (R-InO, 77%; IC, 67%; Table SIII). Three-hundred and thirty-two patients received ≥1 dose of study drug [median (range) number of treatment cycles: 3·0 (1·0–6·0) for R-InO and R-G, 3·5 (1·0–6·0) for R-B; Table SIV]. Ninety-four patients completed treatment. Common reasons for discontinuing were progressive disease/relapse (R-InO, 50% vs. IC, 57%) and adverse events (AEs; R-InO, 32% vs. IC, 17%; Table SV). Median (range) duration of follow-up among surviving patients was 14·9 (0·4–32·8) months for R-InO and 15·9 (0·1–31·2) months for IC. Overall survival was not significantly different for R-InO versus IC (P = 0·708; HR [95% confidence interval (CI)] = 1·1 [0·8–1·4]; Fig S2); Kaplan–Meier estimated median (95% CI) OS was 9·5 (7·0–14·5) and 9·5 (7·7–14·1) months (estimated probabilities of OS [95% CI] at 18 months, 35% [27–43%] and 37% [29–45%]). Progression-free survival (PFS) was also not significantly different for R-InO versus IC (P = 0·27; HR [95% CI)] = 0·9 [0·7–1·2]; Fig S2). Median (95% CI) PFS with R-InO and IC were 3·7 (2·9–5·0) and 3·5 (2·8–4·9) months (estimated probabilities of PFS [95% CI] at 18 months, 19% [13–26%] and 17% [12–24%]). Notably, survival among patients receiving R-InO was prolonged for those with higher versus lower baseline CD22 expression levels (Fig S3). Among all randomized patients, the objective response rate (ORR; 95% CI) was 41% (33–49%) for R-InO and 44% (36–51%) for IC (arm difference, 3% [−8–13%]; P = 0·714; Table 1); Kaplan–Meier estimated median (95% CI) duration of response (DOR) for R-InO versus IC was 11·6 (7·8–not reached [NR]) versus 6·9 (5·5–10·8) months (HR = 0·76 [0·47–1·25]; P = 0·142). Median OS and PFS with R-InO were 9·5 [95% CI, 7·0–14·5] and 3·7 [2·9–5·0] months, respectively; ORR and DOR were 41% and 11·6 months. Although comparisons across studies require caution due to differences in design and patient characteristics, median OS and PFS with R-InO in the previous study with refractory aggressive B-NHL (n = 30) are shorter (OS, 8·8 [3·9–NR] months; PFS, 1·9 [1·0–4·8] months), the ORR is lower (20%), and the DOR is shorter (6·1 months) (Fayad et al, 2013). Conversely, the median OS and PFS with R-InO in the relapsed DLBCL cohort in the previous study (n = 47) are longer (OS, NR [34·7–NR] months; PFS, 17·1 [7·8–NR] months), the ORR is higher (74%), and the DOR is longer (17·7 months) (Fayad et al, 2013). The observed toxicity profile of R-InO is generally consistent with that reported previously for InO alone (Advani et al, 2010; Ogura et al, 2010) and for R-InO (Ogura et al, 2012; Fayad et al, 2013). Treatment-related grade ≥3 treatment-emergent AEs (TEAEs) differing by ≥10% of patients between treatment arms were hematologic (Table 2). All-cause TEAEs followed a similar pattern (Table SVI). Most common serious AEs (>5 patients in either arm) included febrile neutropenia (n = 5 vs. 7) and pneumonia (n = 8 vs. 1). Two treatment-related deaths occurred between treatment start and 56 days after last dose (R-InO, pneumonia [n = 1]; IC, fungal pneumonia, febrile neutropenia and septicaemia [n = 1]). Permanent discontinuations due to AEs were more frequent with R-InO versus IC (25% vs. 18%), most commonly due to thrombocytopenia in the R-InO arm (Table SVII). Treatment-related hepatic TEAEs were more frequent with R-InO versus IC, with hyperbilirubinaemia occurring in 8% vs. 2% of patients (Table 2). Two patients had grade 3 veno-occlusive disease/sinusoidal obstruction syndrome (VOD/SOS) during R-InO treatment (1 after the maximum 6 R-InO cycles; 1 after 3 cycles, resulting in treatment discontinuation). One additional patient developed VOD/SOS approximately 13 months after receiving a single R-InO dose and multiple subsequent therapies, including allogeneic stem cell transplantation after the single R-InO dose and before VOD/SOS onset. No VOD/SOS events occurred in the IC arm. Rituximab-InO treatment was associated with antitumour activity in patients with R/R aggressive B-NHL who were not candidates for high-dose chemotherapy, with or without transplant, for whom treatment options are limited. However, R-InO was not superior to IC with respect to OS; estimates of ORR and median PFS and OS were similar for the 2 treatments. Nevertheless, the efficacy observed here and in other studies (Fayad et al, 2013; Ogura et al, 2016) suggests an examination of InO-containing combination therapies may be appropriate in certain patient populations. A study of InO plus rituximab, cyclophosphamide, vincristine and prednisolone in chemotherapy-naïve patients with DLBCL who are not candidates for anthracycline-based treatment is currently recruiting (NCT01679119). NHD, MO, SC, LEF, MJ, JR, AP, IB, DAS, KT, RR, PAH, EG and KA participated in the collection, interpretation and analysis of data and the development of the manuscript draft. MS, SS and EV participated in study design, the interpretation and analysis of data and the development of the manuscript. NHD has received research funding from Pfizer, Eisai, Valor, Pharmacyclics, Seattle Genetics, and Oncomed. MO has served as a consultant/advisor for Mundipharma, MeijiSeika Pharma, and Celgene, and has received research funding from SymBio. SC has received honoraria and research funding and travel/accommodation expenses from, and served as a consultant/advisor for Pfizer. LEF, MJ, and IB have no relevant relationships to disclose. JR has served as a consultant/advisor for Takeda and received honoraria from Takeda and Seattle Genetics. AP has served as a consultant/advisor for Novartis, Celgene, Roche, Gilead, Janssen. DAS has served as a consultant/advisor for Lundbeck. MS, SS, and EV are employees of and own stock in Pfizer. KT has received research funding from Celgene. RR has received research funding from, and is a member of a speakers bureau for Seattle Genetics. PAH has served as a consultant/advisor for Genentech, Roche, Celgene, Portola and has received research funding from Spectrum, Portola, Molecular Templates, Novartis, GSK, and Seattle Genetics. EG has received research funding from Janssen. KA has received research funding from Kyowa Hakko Kirin. This study was sponsored by Pfizer Inc. Editorial support was provided by Simon J. Slater, PhD, of Complete Healthcare Communications, LLC, which was funded by Pfizer Inc. Data S1. Supplemental methods. Table SI. Redosing and dose reduction criteria. Table SII. Patient characteristics. Table SIII. Reasons that patients were not candidates for high-dose chemotherapy. Table SIV. Treatment summary. Table SV. Reasons for treatment discontinuation. Table SVI. Treatment-emergent adverse events occurring in >10% of patients in either R-InO or IC arm. Table SVII. Adverse events leading to permanent discontinuation or dose modification. Fig S1. Patient disposition. Fig S2. Kaplan–Meier estimate of (A) overall survival and (B) progression-free survival. Fig S3. Kaplan–Meier estimate of (A) overall survival and (B) progression-free survival among patients with low versus high CD22 expression.* Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.