Atezolizumab with or without chemotherapy in metastatic
urothelial cancer (IMvigor130): a multicentre, randomised,
placebo-controlled phase 3 trial
Matthew D Galsky, José Ángel Arranz Arija, Aristotelis Bamias, Ian D Davis, Maria De Santis, Eiji Kikuchi*, Xavier Garcia-del-Muro, Ugo De Giorgi, Marina Mencinger, Kouji Izumi, Stefano Panni, Mahmut Gumus, Mustafa Özgüroğlu, Arash Rezazadeh Kalebasty, Se Hoon Park, Boris Alekseev, Fabio A Schutz, Jian-Ri Li, Dingwei Ye, Nicholas J Vogelzang, Sandrine Bernhard, Darren Tayama, Sanjeev Mariathasan, Almut Mecke, AnnChristine Thåström, Enrique Grande, for the IMvigor130 Study Group†
Summary
Background Atezolizumab can induce sustained responses in metastatic urothelial carcinoma. We report the results of IMvigor130, a phase 3 trial that compared atezolizumab with or without platinum-based chemotherapy versus placebo plus platinum-based chemotherapy in first-line metastatic urothelial carcinoma.
Methods In this multicentre, phase 3, randomised trial, untreated patients aged 18 years or older with locally advanced or metastatic urothelial carcinoma, from 221 sites in 35 countries, were randomly assigned to receive atezolizumab plus platinum-based chemotherapy (group A), atezolizumab monotherapy (group B), or placebo plus platinum-based chemotherapy (group C). Patients received 21-day cycles of gemcitabine (1000 mg/m² body surface area, administered intravenously on days 1 and 8 of each cycle), plus either carboplatin (area under the curve of 4·5 mg/mL per min administered intravenously) or cisplatin (70 mg/m² body surface area administered intravenously) on day 1 of each cycle with either atezolizumab (1200 mg administered intravenously on day 1 of each cycle) or placebo. Group B patients received 1200 mg atezolizumab, administered intravenously on day 1 of each 21-day cycle. The co-primary efficacy endpoints for the intention-to-treat population were investigator-assessed Response Evaluation Criteria in Solid Tumours 1.1 progression-free survival and overall survival (group A vs group C) and overall survival (group B vs group C), which was to be formally tested only if overall survival was positive for group A versus group C. The trial is registered with ClinicalTrials.gov, NCT02807636.
Findings Between July 15, 2016, and July 20, 2018, we enrolled 1213 patients. 451 (37%) were randomly assigned to group A, 362 (30%) to group B, and 400 (33%) to group C. Median follow-up for survival was 11·8 months (IQR 6·1–17·2) for all patients. At the time of final progression-free survival analysis and interim overall survival analysis (May 31, 2019), median progression-free survival in the intention-to-treat population was 8·2 months (95% CI 6·5–8·3) in group A and 6·3 months (6·2–7·0) in group C (stratified hazard ratio [HR] 0·82, 95% CI 0·70–0·96; one-sided p=0·007). Median overall survival was 16·0 months (13·9–18·9) in group A and 13·4 months (12·0–15·2) in group C (0·83, 0·69–1·00; one-sided p=0·027). Median overall survival was 15·7 months (13·1–17·8) for group B and 13·1 months (11·7–15·1) for group C (1·02, 0·83–1·24). Adverse events that led to withdrawal of any agent occurred in 156 (34%) patients in group A, 22 (6%) patients in group B, and 132 (34%) patients in group C. 50 (11%) patients in group A, 21 (6%) patients in group B, and 27 (7%) patients in group C had adverse events that led to discontinuation of atezolizumab or placebo.
Interpretation Addition of atezolizumab to platinum-based chemotherapy as first-line treatment prolonged progression-free survival in patients with metastatic urothelial carcinoma. The safety profile of the combination was consistent with that observed with the individual agents. These results support the use of atezolizumab plus platinum-based chemotherapy as a potential first-line treatment option for metastatic urothelial carcinoma.
Funding F Hoffmann-La Roche and Genentech. Copyright © 2020 Elsevier Ltd. All rights reserved.
Introduction
Cisplatin-based chemotherapy became the standard first-line treatment for metastatic urothelial carcinoma in the 1980s,1 with long-term remission possible in around
10% of patients. Although tolerability has improved,2 no
further substantial improvements in efficacy have been reported.1,2 Furthermore, around 50% of patients with
metastatic urothelial carcinoma are ineligible for cisplatin treatment because of poor performance status, comor-bidities, or impaired renal function.3,4 These patients
generally receive inferior carboplatin-based regimens.4
Programmed death-ligand 1 (PD-L1) and programmed death 1 (PD-1) inhibitors are the first new systemic therapies for metastatic urothelial carcinoma, both for
Lancet 2020; 395: 1547–57 See Comment page 1522 *Current affiliation: Department of Urology, St Marianna University School of Medicine, Kawasaki, Japan
†Members are listed in the appendix
Icahn School of Medicine at Mount Sinai, Tisch Cancer Institute, New York, NY, USA
(Prof M D Galsky MD); Hospital
General Universitario Gregorio Marañón, Madrid, Spain
(J A A Arija MD); National and
Kapodistrian University of Athens, Athens, Greece
(Prof A Bamias MD); Eastern
Health, Monash University, Melbourne, VIC, Australia
(Prof I D Davis PhD); Charité
University Hospital, Berlin, Germany (M De Santis MD); Department of Urology, Medical University, Vienna, Austria (M De Santis); Keio University School of Medicine, Tokyo, Japan (E Kikuchi MD); Catalan Institute of Oncology, IDIBELL, University of Barcelona, Barcelona, Spain
(X Garcia-del-Muro MD); Istituto
Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST), IRCCS, Meldola, Italy
(U De Giorgi MD); Institute of
Oncology Ljubljana, Ljubljana, Slovenia (M Mencinger MD); Kanazawa University Hospital, Kanazawa, Japan (K Izumi MD); Istituti Ospitalieri Cremona, Cremona, Italy (S Panni MD); Istanbul Medeniyet University, Goztepe Research Hospital, Istanbul, Turkey
(Prof M Gumus MD); Istanbul
University-Cerrahpaşa, Cerrahpaşa School of Medicine, Istanbul, Turkey
(Prof M Özgüroğlu MD); Norton
Cancer Institute, Louisville, KY, USA (A R Kalebasty MD); Sungkyunkwan University Samsung Medical Center, Seoul, Korea (S H Park MD);
P Herzen Oncology Research Institute, Moscow, Russia
(Prof B Alekseev MD);
Beneficencia Portuguesa de São Paulo, São Paulo, Brazil
(F A Schutz MD); Taichung
Veterans General Hospital, HungKuang University, Taichung, Taiwan (J-R Li MD); Fudan University Shanghai Cancer Center, Shanghai,China
(Prof D Ye MD); Comprehensive
Cancer Centers of Nevada, Las Vegas, NV, USA
(Prof N J Vogelzang MD); Roche,
Welwyn Garden City, UK
(S Bernhard PharmD);
Genentech, San Francisco, CA, USA (D Tayama MD,
S Mariathasan PhD, A Thåström PhD);
F Hoffmann-La Roche, Basel, Switzerland (A Mecke PhD); and MD Anderson Cancer Center Madrid, Madrid, Spain
(E Grande MD) Correspondence to: Dr Matthew D Galsky, Icahn School of Medicine at Mount Sinai, Tisch Cancer Institute, New York, NY 10029, USA
matthew.galsky@mssm.edu
See Online for appendix
first-line treatment of cisplatin-ineligible patients and for patients with disease progression despite platinum-based chemotherapy.5–12
Regimens that combine platinum-based chemother-apy and PD-L1 and PD-1 inhibitors are appealing for several reasons. Platinum-based chemotherapy can induce immunomodulatory effects, thereby enhancing concomitant PD-L1 and PD-1 blockade.13,14 This
combi-nation might also be beneficial because of the absence of clinical cross-resistance between these therapeutic classes, because a minority of patients receive treatment beyond first-line therapy.15,16
IMvigor130, a phase 3, global, multicentre, randomised, partly-blinded trial evaluated the efficacy of atezolizumab alone or combined with platinum-based chemotherapy versus placebo plus platinum-based chemotherapy in patients with untreated metastatic urothelial carcinoma. Preliminary results have already affected clinical practice. As of July, 2018, based on unplanned assessments by the independent data monitoring committees (IDMC) of both IMvigor130 and KEYNOTE-361, the US Food and Drug Administration (FDA) and European Medicines Agency (EMA) revised the labels for atezolizumab and pembrolizumab monotherapy in the first-line setting
to limit treatment to PD-L1 biomarker-expressing tumours.17–19 Here, we report the primary progression-free
survival and interim overall survival results from IMvigor130.
Methods
Study design and participants
We did a multicentre, phase 3, randomised trial at 221 sites in 35 countries (Australia, Belgium, Bosnia and Herzegovina, Brazil, Canada, Chile, China, Czech Republic, Estonia, Finland, Georgia, Greece, Hong Kong, Israel, Italy, Japan, South Korea, Malaysia, Mexico, Netherlands, Poland, Portugal, Romania, Russia, Serbia, Singapore, Slovenia, South Africa, Spain, Taiwan, Thailand, Turkey, Ukraine, the UK, and the USA). Eligible patients were aged 18 years or older with locally advanced or metastatic urothelial carcinoma and had not received previous systemic therapy in the metastatic setting (adjuvant or neoadjuvant therapy was allowed if >12 months before), were eligible for platinum-based chemotherapy (cisplatin or carboplatin with gemcitabine) based on the investigator’s judgment, had measurable disease according to Response Evaluation Criteria in Solid Tumours (RECIST) version 1.1, and had Eastern
Research in context
Evidence before this study
We searched PubMed and major international oncology and
urology congressesfor articles published in English pertaining
to metastatic urothelial carcinoma between June 1, 2011, and June 1, 2016, with MeSH search terms “metastatic” AND “bladder cancer”, “urothelial carcinoma”, “urothelial cancer”, “programmed cell death 1”, “PD-1”, “programmed cell death ligand 1”, and “PD-L1”. This search identified an unmet clinical need for effective and tolerable therapies for metastatic urothelial carcinoma based on the paucity of positive data from randomised phase 3 studies. Platinum-based combination chemotherapy has been the preferred regimen in the first-line setting and single-agent chemotherapy is typically reserved for the second-line setting. However, 40–50% of patients with metastatic urothelial carcinoma are ineligible for cisplatin because of poor performance status, comorbidities, or impaired renal function, and these patients generally receive carboplatin-based regimens, an inferior option. Although tolerability has improved with new treatments over the past 30 years, efficacy has not. Checkpoint inhibitors that target programmed death-ligand 1 (PD-L1) and programmed death 1 (PD-1) are the first new systemic therapies for metastatic urothelial carcinoma, both for first-line treatment of cisplatin-ineligible patients and for patients with disease progression despite platinum-based chemotherapy.
Added value of this study
To our knowledge, IMvigor130 is the first phase 3, randomised trial to report results for platinum-based chemotherapy plus
PD-L1 or PD-1 blockade, or PD-L1 blockade alone or PD-1 blockade alone, versus standard platinum-based chemotherapy plus placebo as a first-line treatment for metastatic urothelial carcinoma. IMvigor130 met its co-primary progression-free survival endpoint, showing a significant improvement when atezolizumab was added to platinum-based chemotherapy versus platinum-based chemotherapy alone. At the interim analysis, the result for the co-primary endpoint of overall survival did not cross the interim efficacy boundary. The combination of atezolizumab plus platinum-based chemotherapy had a safety profile consistent with previous studies and with the safety profiles of each individual agent. Implications of all the available evidence
The results of IMvigor130, to our knowledge the largest phase 3 trial reporting results from first-line treatment for metastatic urothelial cancer, included both cisplatin-eligible and cisplatin-ineligible patients. Preliminary results have already impacted clinical practice; recommendations from the US Food and Drug Administration and European Medicines Agency based on an unplanned early analysis by independent data monitoring committees of IMvigor130 led to revision of the prescribing label for atezolizumab for metastatic urothelial cancer in the first-line setting. Results from IMvigor130 support clinical activity of the atezolizumab plus chemotherapy combination as a potential first-line treatment option for metastatic urothelial carcinoma.
Cooperative Oncology Group (ECOG) performance status of 2 or less.
Adequate haematological and end-organ function were required. Patients were screened and enrolled at major academic medical centres and community oncology practices. Per protocol, cisplatin ineligibility was guided by the Galsky criteria,20 although the final decision to use
cisplatin or carboplatin was left at the investigator’s discretion. Patients needed a representative tumour specimen that could be evaluated for prospective central testing of PD-L1 expression on tumour-infiltrating immune cells as a percentage of tumour area (SP142 PD-L1 immunohistochemical assay; Ventana Medical Systems; Tucson, AZ, USA). Additional eligibility criteria are presented in the appendix (p 11).
The trial was conducted according to Good Clinical Practice and the Declaration of Helsinki. All patients provided written informed consent. Protocol approval was obtained from independent review boards or ethics committees at each site. For the study protocol, see the appendix (p 27).
Randomisation and masking
We randomly assigned patients (1:1:1) with a stratified permuted block method (fixed block size) and an interactive voice–web response system to receive blinded atezolizumab plus open-label platinum-based chemo-therapy (group A), open-label atezolizumab monochemo-therapy (group B), or masked placebo plus open-label platinum-based chemotherapy (group C). Before randomisation, investigators specified which chemo therapy (cisplatin or carboplatin) patients would receive if randomly assigned to group A or group C. Randomisation was stratified by PD-L1 immune cell expression status (IC0 [<1%] vs IC1 [≥1% and <5%] vs IC2/3 [≥5%]), Bajorin risk factor score, comprising Karnofsky performance status less than 80% versus 80% or more and presence of visceral metastases (none vs one vs two or patients with liver metastases), and investigator choice of platinum-based chemotherapy (cisplatin plus gemcitabine or carboplatin plus gemcitabine).
Procedures
Patients received 21-day cycles of gemcitabine (1000 mg/m² body surface area, administered intra-venously on days 1 and 8 of each cycle), plus either carboplatin (area under the curve of 4·5 mg/mL per min administered intravenously) or cisplatin (70 mg/m² body surface area administered intravenously) on day 1 of each cycle with either atezolizumab (1200 mg administered intravenously on day 1 of each cycle) or placebo. Group B patients received 1200 mg atezolizumab, administered intra venously on day 1 of each 21-day cycle. Patients were allowed to continue atezolizumab as a single agent after completing chemotherapy according to institutional guidelines (appendix p 11). Treatment was admin istered until investigator-assessed disease progression per
RECIST 1.1 or unacceptable toxicity. Single-patient unblinding was permissible for adverse event manage-ment or treatmanage-ment discontinuation after RECIST 1.1 progression, provided the treatment plan for the next
Group A (n=451) Group B (n=362) Group C (n=400)
Age, years 69 (62–75) 67 (62–74) 67 (61–73)
Age group, years
<65 153 (34%) 142 (39%) 153 (38%) ≥65 298 (66%) 220 (61%) 247 (62%) Sex Female 113 (25%) 82 (23%) 102 (26%) Male 338 (75%) 280 (77%) 298 (75%) Race White 346 (77%) 260 (72%) 304 (76%)
Black or African American 6 (1%) 1 (<1%) 1 (<1%)
Asian 90 (20%) 94 (26%) 85 (21%)
American Indian or Alaska Native 4 (1%) 2 (1%) 3 (1%)
Native Hawaiian or other Pacific Islander 0 1 (<1%) 0
Multiple races 0 1 (<1%) 0
Unknown 5 (1%) 3 (1%) 7 (2%)
Tobacco use history
Never 150 (33%) 134 (37%) 157 (39%)
Current 76 (17%) 51 (14%) 69 (17%)
Former 225 (50%) 177 (49%) 174 (44%)
Primary tumour site
Bladder 312 (69%) 265 (73%) 293 (73%) Urethra 10 (2%) 6 (2%) 5 (1%) Renal pelvis 64 (14%) 53 (15%) 58 (15%) Ureter 59 (13%) 36 (10%) 42 (11%) Other 2 (<1%) 0 1 (<1%) Missing 4 (1%) 2 (1%) 1 (<1%) Disease status Locally advanced 51 (11%)* 43 (12%) 34 (9%) Metastatic 401 (89%)* 319 (88%) 366 (92%)
Site of metastatic disease
Lymph node only 80 (18%) 70 (19%) 67 (17%)
Visceral† 259 (57%) 201 (56%) 239 (60%)
Liver 95 (21%) 85 (23%) 91 (23%)
ECOG performance status
0 182 (40%) 157 (43%) 173 (43%)
1 209 (46%) 174 (48%) 187 (47%)
2 60 (13%) 31 (9%) 40 (10%)
Haemoglobin concentration <100 g/L‡ 53 (12%) 38 (11%) 53 (13%) Bajorin risk factor score§
0 176 (39%) 151 (42%) 162 (41%)
1 169 (37%) 134 (37%) 149 (37%)
2 106 (24%) 77 (21%) 89 (22%)
Previous adjuvant or neoadjuvant regimen 54 (12%) 43 (12%) 64 (16%) Programmed death-ligand 1 status on immune cells
IC2/3 108 (24%) 88 (24%) 91 (23%)
IC1 195 (43%) 160 (44%) 179 (45%)
IC0 148 (33%) 114 (31%) 130 (33%)
line of therapy included a checkpoint inhibitor or plan to enrol the patient in a subsequent clinical trial requiring knowledge of current study treatment assignment to confirm eligibility. Unblinding did not result in study withdrawal and patients were still followed up for survival.
Tumour assessments were done at baseline and every 9 weeks thereafter (every 12 weeks after 54 weeks of treatment) until disease progression per RECIST 1.1 as determined by the investigator, unacceptable toxicity, death, study termination, or consent withdrawal, which-ever occurred first.
Additional trial details, including key protocol amend-ments, are available in the appendix (pp 11–15).
Outcomes
The co-primary efficacy endpoints for the intention-to-treat population were investigator-assessed RECIST 1.1 progression-free survival and overall survival (group A vs group C) and overall survival (group B vs group C), which was formally tested only if overall survival was positive for group A versus group C.
An additional primary outcome was evaluation of safety, comprisingthe incidence, nature, and severity of adverse events or adverse events of special interest suggesting an autoimmune cause, assessed with National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0.
Secondary endpoints were investigator-assessed con-firmed response rate (according to RECIST 1.1); duration of response; progression-free survival (according to RECIST 1.1) determined by blinded independent central review; investigator-assessed progression-free survival (according to RECIST 1.1) specific to Group B (atezolizumab monotherapy) compared with Group C (patients treated with placebo plus platinum-based chemotherapy); overall survival at 1 year; progression-free
survival at 1 year; time to deterioration in global health status as measured by the European Organisation for Research and Treatment of Cancer (EORTC) Quality-of-Life Questionnaire Core 30 (QLQ-C30); and time to deterioration in physical function as measured by the EORTC QLQ-C30. Secondary endpoints reported here are investigator-assessed confirmed objective response rate (according to RECIST 1.1) and duration of response. The following will be reported elsewhere or as applicable with additional study follow-up: progression-free survival (according to RECIST 1.1) determined by blinded inde-pendent central review, investigator-assessed progres sion-free survival (according to RECIST 1.1) specific to Group B (atezolizumab monotherapy) compared with Group C (patients treated with placebo plus platinum-based chemotherapy), overall survival at 1 year, progression-free survival at 1 year, time to deterioration in global health status as measured by the EORTC QLQ-C30, and time to deterioration in physical function as measured by the EORTC QLQ-C30.
Statistical analysis
This study was initially implemented with random assignment to two treatment groups (carboplatin plus gemcitabine with or without atezolizumab; group A and group C; 2:1) in patients who were ineligible for cisplatin-based chemotherapy. A protocol amendment added an open-label atezolizumab monotherapy group (group B; changed randomisation to 1:1:1; appendix p 15) and expanded the eligibility criteria to also allow enrolment of patients who were eligible for cisplatin-based chemo-therapy. This protocol amendment was implemented after Sept 21, 2016,while recruitment was ongoing and, as a result of this design change, the study consists of two stages—patients recruited into the study before this amendment (stage 1) and patients recruited after this amendment (stage 2). Patients from both stages were included in the final analysis. For comparison of efficacy in group A versus group C, the intention-to-treat popu-lation was defined as all patients randomly assigned to group A or group C in stages 1 and 2, whether or not the assigned study treatment was received. For comparison of efficacy in group B versus group C, the intention-to-treat population included only patients concurrently enrolled in stage 2 before implementation of a protocol amendment that limited group B enrolment to patients with PD-L1 IC2/3 tumours. The type I error (0·025; one-sided) was split between co-primary endpoints progression-free survival and overall survival comparing group A versus group C (appendix p 15). Formal treat-ment comparisons of overall survival in group B versus group C (intention-to-treat population and PD-L1 IC2/3 subgroup) were done in a hierarchal fashion (appendix p 15) if overall survival comparing group A versus group C was statistically significant.
This study was designed to enrol around 1200 patients (appendix p 12). Sample size and power calculations are
Group A (n=451) Group B (n=362) Group C (n=400)
(Continued from previous page)
Cisplatin ineligibility per Galsky criteria¶ 263 (58%) 192 (53%) 222 (56%)
Renal impairment|| 222 (49%) 172 (48%) 202 (51%)
Hearing loss of ≥25 dB 16 (4%) 11 (3%) 8 (2%)
Peripheral neuropathy grade ≥2 6 (1%) 2 (1%) 3 (1%)
ECOG performance status 2 60 (13%) 31 (9%) 40 (10%)
Investigator choice of chemotherapy**
Carboplatin 314 (70%) 227 (63%) 264 (66%)
Cisplatin 137 (30%) 135 (37%) 136 (34%)
Data are median (IQR) or n (%). ECOG=Eastern Cooperative Oncology Group. *One patient in group A reported both locally advanced and metastatic disease. †Defined as lung, liver, and bone metastases. ‡n=444 for group A; n=359 for group B; n=394 for group C. §Bajorin risk factor 0=score 0 and no liver metastases, 1=score 1 and no liver metastases, 2=score 2 or liver metastases. ¶Per protocol, excluding New York Heart Association functional classification. ||Creatinine clearance could not be evaluated for one patient without a bodyweight measurement in group B. **Investigator’s choice of platinum therapy refers to the choice before randomisation; patients randomly assigned to group B did not receive platinum as part of the initial study design.
explained in the appendix (pp 12–13). The timing of this final progression-free survival analysis in the intention-to-treat population was event-driven, and a first interim overall survival analysis in the intention-to-treat population was planned at the same time. Progression-free survival and overall survival were compared between trial groups using the stratified log-rank test. For comparison of group A with group C, we used the inverse normal combination approach to account for a possible effect of study stage (appendix p 14).21 We used Kaplan-Meier methodology to
estimate progression-free survival and overall survival and the median time for each group from randomisation to disease progression or death (for progression-free survival) and median time from randomisation to death (for overall survival). We estimated the hazard ratios (HRs) and 95% CIs for progression-free survival and overall survival by stratified Cox regression, with the same stratification factors used in the stratified log-rank test plus enrolment stage. The safety population was defined as patients who received any amount of any component of study treatment. For safety evaluation, patients were analysed according to the treatment received. An IDMC reviewed safety and trial conduct data every 6 months.
Sponsor-derived cisplatin ineligibility status according to Galsky criteria used the following data entered into the database by each site: renal impairment (determined from calculated baseline creatinine clearance [Cockcroft-Gault Formula <60 mL/min] at patient screening or day 1, cycle 1), hearing loss and peripheral neuropathy (collected from the medical history form), and ECOG performance status 2 data (obtained from a separate ECOG performance status form; table 1).
Statistical analyses were done using SAS (version 9.4 TS1M3). The trial is registered with ClinicalTrials.gov, NCT02807636.
Role of the funding source
F Hoffmann-La Roche and Genentech sponsored the study and collaborated with academic authors regarding study design, data collection, data analysis, data inter-pretation, and writing of the report. All authors verify that this study was done according to the protocol and attest to data accuracy and completeness. All authors had full access to all the data in the study, contri-buted to drafts of the manuscript, and gave final approval for publication. The corresponding author had
451 assigned to receive atezolizumab plus platinum-based chemotherapy 444 received treatment
184 treatment ongoing 105 in survival follow-up
79 continuing to receive intervention
451 included in intention-to-treat analysis
453 included in safety analysis†
267 discontinued treatment 231 died
24 withdrew 11 lost to follow-up
1 symptomatic deterioration
360* assigned to receive atezolizumab 359 received treatment
150 treatment ongoing 84 in survival follow-up
66 continuing to receive intervention
360 included in intention-to-treat analysis
354 included in safety analysis† 1213 randomly assigned 1898 patients assessed for eligibility
212 discontinued treatment 189 died
19 withdrew 4 lost to follow-up 685 excluded
513 did not meet eligibility criteria 63 withdrew
18 withdrawn by physician 91 other
400 assigned to receive placebo plus platinum-based chemotherapy 394 received treatment
139 treatment ongoing 88 in survival follow-up 51 continuing to receive intervention
400 included in intention-to-treat analysis
390 included in safety analysis†
261 discontinued treatment 218 died
34 withdrew 7 lost to follow-up 2 deviated from protocol
Figure 1: Trial profile
*Two additional patients were enrolled under protocol version 6 and not included in the intention-to-treat population. †Some patients switched treatment groups for safety analyses because they received a different treatment than they were initially randomly assigned, either by mistake or because of protocol version 6.
final responsibility for the decision to submit for publication.
Results
Between July 15, 2016, and July 20, 2018, we enrolled 1213 patients (intention-to-treat population). 451 (37%) were randomly assigned to group A, 362 (30%) to group B, and 400 (33%) to group C (figure 1). Median follow-up for survival was 11·8 months (IQR 6·1–17·2) for all patients. Most baseline characteristics were similar
between treatment groups (table 1). The proportions of cisplatin-ineligible patients in group A, group B, and group C were 58%, 53%, and 56%, respectively. The differences in these proportions could be due to the initial two-group design of IMvigor130, which targeted only cisplatin-ineligible patients, wherein a total of 129 patients were randomly assigned in a ratio of 2:1 to group A versus group C. The trial protocol was later amended to enrol all platinum-eligible patients at a ratio of 1:1:1. Baseline characteristics for the subpopulations of
Number at risk (number censored) Atezolizumab + platinum-based chemotherapy (group A) Placebo + platinum-based chemotherapy (group C) 0 451 (0) 400 (0) 3 345 (22) 317 (13) 6 282 (27) 246 (14) 9 160 (34) 116 (19) 12 111 (49) 73 (31) 15 74 (62) 40 (45) 18 42 (81) 18 (58) 21 22 (98) 11 (63) 24 10 (108) 4 (70) 30 2 (115) NE 27 4 (114) NE 33 NE NE Time (months) Atezolizumab + platinum-based chemotherapy (group A) Placebo + platinum-based chemotherapy (group C) 0 10 20 30 40 50 60 70 80 90 100 Progression-free survival (%)
Patients Median progression-free survival
months HR for progression or death (95% CI)
Group A (n=451) Group C (n=400) All patients 851 8·2 6·3 Age, years <65 306 8·3 7·0 ≥65 545 7·5 6·3 Sex Female 215 7·4 6·2 Male 636 8·2 6·7 Race Asian 175 8·3 8·2 White 650 7·4 6·3
ECOG performance status score
0 355 9·4 6·7 1 396 6·8 6·3 2 100 4·8 6·1 PD-L1 status 0 278 6·5 6·2 1 374 8·1 6·7 2 or 3 199 8·6 6·3
Bajorin risk factor score
0 338 9·8 8·3
1 318 8·2 6·2
2 195 5·0 6·1
Investigator choice of chemotherapy
Cisplatin 273 8·8 6·4 Carboplatin 578 7·1 6·3
Previous adjuvant or neoadjuvant regimen
Yes 118 8·3 6·6 No 733 8·2 6·3 0·81 (0·69–0·94) 0·82 (0·63–1·06) 0·80 (0·66–0·97) 0·75 (0·55–1·02) 0·83 (0·69–0·99) 0·86 (0·61–1·23) 0·78 (0·65–0·93) 0·69 (0·54–0·89) 0·86 (0·69–1·07) 0·94 (0·61–1·44) 0·79 (0·61–1·03) 0·89 (0·70–1·13) 0·68 (0·49–0·95) 0·79 (0·61–1·03) 0·74 (0·58–0·95) 0·94 (0·69–1·29) 0·73 (0·55–0·97) 0·84 (0·70–1·02) 0·73 (0·48–1·12) 0·82 (0·69–0·97) 0·3 3·0 Favours group C Favours group A 1·0
Median progression-free survival (95% CI), months Number of events/ number of patients A B 334/451 326/400 8·2 (6·5–8·3) 6·3 (6·2–7·0) Number at risk (number censored) Atezolizumab + platinum-based chemotherapy (group A) Placebo + platinum-based chemotherapy (group C) 0 451 (0) 400 (0) 3 408 (11) 359 (13) 6 360 (17) 308 (19) 9 301 (23) 255 (24) 12 229 (51) 182 (48) 15 163 (88) 123 (79) 18 117 (118) 79 (108) 21 72 (154) 49 (131) 24 36 (184) 25 (151) 30 3 (213) NE 27 16 (200) 8 (164) 33 NE NE Time (months) Atezolizumab + platinum-based chemotherapy (group A) Placebo + platinum-based chemotherapy (group C) 0 10 20 30 40 50 60 70 80 90 100 Overall survival (%)
Median overall survival (95% CI), months Number of events/ number of patients C 235/451 228/400 16·0 (13·9–18·9) 13·4 (12·0–15·2) Number at risk (number censored) Atezolizumab monotherapy (group B) Placebo + platinum-based chemotherapy (group C) 0 360 (0) 359 (0) 3 285 (10) 322 (13) 6 245 (14) 274 (19) 9 216 (16) 224 (24) 12 173 (41) 158 (48) 15 120 (75) 103 (79) 18 72 (107) 62 (107) 21 42 (128) 35 (128) 24 16 (153) 15 (147) 30 NE NE 27 NE 3 (158) 33 NE NE Time (months) Atezolizumab monotherapy (group B) Placebo + platinum-based chemotherapy (group C) 0 10 20 30 40 50 60 70 80 90 100 Overall survival (%)
Median overall survival (95% CI), months Number of events/ number of patients D 191/360 198/359 15·7 (13·1–17·8) 13·1 (11·7–15·1)
Stratified HR for death, 1·02 (95% CI 0·83–1·24) Stratified HR for death,
0·83 (95% CI 0·69–1·00) p=0·027
Stratified HR for prgression or death, 0·82 (95% CI 0·70–0·96) p=0·007
groups B and C with PD-L1 IC2/3 tumours are listed in the appendix (p 18). 314 (70%) patients in group A and 264(66%) patients in group C received carboplatin. 75 (40%) cisplatin-eligible patients in group A and 70(39%) in group C received carboplatin per investigator choice. For all 535 patients across groups who were considered cisplatin eligible at study entry according to Galsky criteria, the investigator choice was carboplatin in 213 (40%) patients and was cisplatin for 86(13%) of 677 patients who were cisplatin ineligible.
The numbers of patients still receiving the trial intervention at data cutoff (May 31, 2019) are shown in figure 1. Subsequent anticancer therapy was administered to 118 (26%) patients in group A, 144 (40%) patients in group B, and 164 (41%) patients in group C (appendix pp 19–20).
334 (74%) patients in group A and 326 (82%) patients in group C had disease progression or had died at the time of the data cutoff. Median duration of progression-free survival in the intention-to-treat population was 8·2 months (95% CI 6·5–8·3) in group A and 6·3 months (6·2–7·0) in group C (stratified HR 0·82, 95% CI 0·70–0·96; one-sided p=0·007; figure 2). Over-all, the observed treatment effect on progression-free survival was consistent across subgroups (figure 2).
At the data cutoff and first interim analysis of overall survival in the intention-to-treat population, 235 (52%) patients in group A and 228 (57%) patients in group C had died. Median overall survival was
16·0 months (95% CI 13·9–18·9) in group A and 13·4 months (12·0–15·2) in group C (stratified HR 0·83, 95% CI 0·69–1·00; one-sided p=0·027; figure 2). Because this result did not cross the prespecified interim efficacy boundary for statistical significance (p=0·007), no further statistical hypothesis testing was done for group B versus group C (appendix p 15). Overall survival results across key subgroups were consistent with the treatment effect observed in the intention-to-treat population (appendix pp 16–17).
191 (53%) patients in group B and 198 (55%) patients in group C (specific to the predefined analysis popula-tion for the group B comparison) died. The median overall survival was 15·7 months (95% CI 13·1–17·8) for group B and 13·1 months (11·7–15·1) for group C (stratified HR 1·02, 95% CI 0·83–1·24; figure 2). Median overall survival was not estimable (95% CI 17·7–not estimable) for the PD-L1 IC2/3 subgroup in group B versus 17·8 months (10·0–not estimable) in group C (stratified HR 0·68, 95% CI 0·43–1·08; figure 2). Median overall survival for the PD-L1 IC0/1 subgroup was 13·5 months (95% CI 11·1–16·4) in group B versus 12·9 months (11·3–15·0) in group C (unstratified HR 1·07, 95% CI 0·86–1·33; figure 2).
In the intention-to-treat population, 212 (47%) of 447 patients (95% CI 43–52) in group A, 82 (23%) of 359 patients (19–28) in group B, and 174 (44%) of 397 patients (39–49) in group C had an investigator-assessed objective response (table 2). A complete
Figure 2: Investigator-assessed progression-free survival and overall survival
(A) Kaplan-Meier estimates of progression-free survival for groups A and C. (B) Subgroup analysis of progression-free survival according to baseline characteristics. Bajorin risk factor score 0=score 0 and no liver metastases, 1=score 1 and no liver metastases, and 2=score 2 or liver metastases. ECOG performance status ranges from 0 to 5, with higher scores reflecting greater disability.
(C) Kaplan-Meier estimates of overall survival for groups A and C. (D) Kaplan-Meier estimates of overall survival for groups B and C. (E) Kaplan-Meier estimates of overall survival for groups B and C for patients with PD-L1 IC2/3 tumours. (F) Kaplan-Meier estimates of overall survival for groups B and C for patients with PD-L1 IC0/1 tumours. Tick marks on Kaplan-Meier plots indicate censored data. NE=not estimable. HR=hazard ratio. ECOG=Eastern Cooperative Oncology Group. PD-L1=programmed death-ligand 1.
Number at risk (number censored) Atezolizumab monotherapy (group B) Placebo + platinum-based chemotherapy (group C) 0 88 (0) 85 (0) 3 75 (3) 76 (3) 6 70 (4) 62 (3) 9 64 (4) 51 (4) 12 49 (14) 42 (8) 15 35 (26) 30 (17) 18 24 (33) 21 (24) 21 14 (41) 14 (30) 24 5 (50) 5 (38) 30 NE NE 27 NE 1 (42) 33 NE NE Time (months) Atezolizumab monotherapy (group B) Placebo + platinum-based chemotherapy (group C) 0 10 20 30 40 50 60 70 80 90 100 Overall survivl (%)
Median overall survival (95% CI), months Number of events/ number of patients E 33/88 42/85 NE (17·7–NE) 17·8 (10·0–NE) Number at risk (number censored) Atezolizumab monotherapy (group B) Placebo + platinum-based chemotherapy (group C) 0 272 (0) 274 (0) 3 210 (7) 246 (10) 6 175 (10) 212 (16) 9 152 (12) 173 (20) 12 124 (27) 116 (40) 15 85 (49) 73 (62) 48 (74) 41 (83) 18 28 (87) 21 (98) 21 11 (103) 10 (109) 24 30 NE NE 27 NE 2 (116) 33 NE NE Time (months) Atezolizumab monotherapy (group B) Placebo + platinum-based chemotherapy (group C) 0 10 20 30 40 50 60 70 80 90 100 Overall survival (%)
Median overall survival (95% CI), months Number of events/ number of patients F 158/272 156/274 13·5 (11·1–16·4) 12·9 (11·3–15·0)
Stratified HR for death,
response was seen in 56 (13%) patients in group A, 22 (6%) patients in group B, and 27 (7%) patients in group C (table 2). In patients who had a reponse, the median duration of response was 8·5 months (95% CI 7·2–10·4) in group A, not estimable in group B (15·9–not estimable), and 7·6 months (6·3–8·5) in group C (table 2). Objective responses for the PD-L1 IC2/3 subpopulations of group B and group C are shown in the appendix (p 21).
79 (20%) patients in group C received non-protocol immunotherapy after discontinuing study treatment compared with 21(5%) patients in group A. Non-protocol immunotherapy use was lower in group B, in which only nine(2%) patients received non-protocol immunotherapy after discontinuing study treatment.
The safety population included 453 patients in group A, 354 patients in group B, and 390 patients in group C (table 3). Median durations of treatment are presented in the appendix (p 22). The median number of doses of platinum-based chemotherapy were similar in group A and group C (appendix p 22).
Adverse events related to any component of the trial regimen occurred in 434 (96%) patients in group A, 211 (60%) patients in group B, and 373 (96%) patients in group C (table 3). Grade 3 or 4 adverse events related to any component of the trial regimen occurred in 367 (81%) patients in group A, 54 (15%) patients in group B, and 315 (81%) patients in group C (table 3; appendix p 23). For patients receiving chemotherapy, the most common grade 3 or 4 adverse events related to atezolizumab or placebo were neutropenia, anaemia, decreased platelets, decreased neutrophils, and thrombocytopenia, regardless of the addition of atezolizumab (appendix p 25). Serious adverse events are reported in table 3. 227 (50%) patients in group A, 132 (37%) in group B, and 135 (35%) in group C had an adverse event of special interest for atezolizumab, suggesting an immune-related cause (table 3; appendix p 24). Adverse events of special interest requiring the use of systemic corticosteroids occurred in 55 (12%) patients in group A, 29 (8%) patients in group B, and 22 (6%) patients in group C. Grade 3 or 4 adverse events of special interest occurred in 34 (8%) patients in group A, 29 (8%) in group B, and 17 (4%) in group C (table 3; appendix p 24). All-cause fatal adverse events occurred in 29 (6%) patients in group A, 28 (8%) in group B, and 20 (5%) in group C (table 3). Adverse events attributed to the trial regimen by the investigators were reported (appendix p 25). Deaths related to the trial regimen according to the investigators occurred in nine (2%) patients in group A, three (1%) in group B, and four (1%) in group C (table 3). Six fatal adverse events of special interest occurred (one case each of hepatic failure in group A and group B, one case of interstitial lung disease in group B, one case each of hepatitis in group A and group C, and one case of pneumonitis in group A; appendix p 24). Adverse events that led to withdrawal of any agent occurred in 156 (34%) patients in group A, 22 (6%) patients in group B, and 132 (34%) patients in group C (table 3). 50 (11%) patients in group A, 21 (6%) patients in group B, and 27 (7%) patients in group C had adverse events that led to discontinuation of atezolizumab or placebo (table 3).
Discussion
In this study, we report the final analysis of progression-free survival and the interim analysis of overall survival
Group A (n=447) Group B (n=359) Group C (n=397) Response
Objective confirmed response 212 (47%) [43–52] 82 (23%) [19–28] 174 (44%) [39–49] Complete response 56 (13%) [10–16] 22 (6%) [4–9] 27 (7%) [5–10] Partial response 156 (35%) [30–40] 60 (17%) [13–21] 147 (37%) [32–42] Stable disease 132 (30%) [25–34] 91 (25%) [21–30] 139 (35%) [30–40] Progressive disease 54 (12%) [9–15] 134 (37%) [32–43] 53 (13%) [10–17]
Duration of response
Patients with event 132/212 (62%) 27/82 (33%) 122/174 (70%)
Duration of response, months 8·5 (7·2–10·4] NE (15·9–NE) 7·6 (6·3–8·5)
Data are n (%) [95% CI], n/N (%), or median (95% CI). NE=not estimable.
Table 2: Response rate, duration of response, and disease progression
Group A (n=453) Group B (n=354) Group C (n=390)
Total deaths 236 (52%) 190 (54%) 223 (57%)
Adverse events regardless of attribution
Any grade adverse events 451 (>99%) 329 (93%) 386 (99%)
Grade 3 or 4 adverse events 383 (85%) 148 (42%) 334 (86%)
Grade 5 adverse events 29 (6%) 28 (8%) 20 (5%)
Treatment-related adverse events 434 (96%) 211 (60%) 373 (96%) Treatment-related grade 3 or 4 adverse events 367 (81%) 54 (15%) 315 (81%) Treatment-related grade 5 adverse events 9 (2%) 3 (1%) 4 (1%) Serious adverse events
Regardless of attribution 234 (52%) 152 (43%) 191 (49%)
Treatment-related serious adverse events 144 (32%) 44 (12%) 101 (26%) Adverse events leading to any treatment
discontinuation 156 (34%) 22 (6%) 132 (34%)
Adverse events leading to discontinuation of
atezolizumab or placebo 50 (11%) 21 (6%) 27 (7%)
Adverse events leading to discontinuation of
cisplatin 53 (12%) 0 52 (13%)
Adverse events leading to discontinuation of
carboplatin 90 (20%) 1 (<1%)* 79 (20%)
Adverse events leading to discontinuation of
gemcitabine 117 (26%) 1 (<1%)* 100 (26%)
Adverse events leading to any dose reduction or
interruption 363 (80%) 112 (32%) 304 (78%)
Any grade adverse events of special interest 227 (50%) 132 (37%) 135 (35%) Grade 3 or 4 adverse events of special interest 34 (8%) 29 (8%) 17 (4%) Grade 5 adverse events of special interest 3 (1%) 2 (1%) 1 (<1%)
Data are n (%). *This patient was randomly assigned to group A and received atezolizumab; they had an adverse event of pyrexia that day, and gemcitabine and carboplatin were marked as drug withdrawn. Since no chemotherapy was given, this patient was included in group B for safety analysis.
for IMvigor130, the first phase 3 trial to assess a PD-L1 and PD-1 inhibitor alone or in combination with platinum-based chemotherapy versus platinum-based chemotherapy alone in untreated metastatic urothelial carcinoma, and the first trial to include both cisplatin-ineligible and cisplatin-eligible patients. Addition of atezolizumab to platinum-based chemotherapy resulted in a significant prolongation of progression-free survival. A near-doubling of complete responses was also seen, and although the prespecified boundary for declaring a statistically significant overall survival advantage was not crossed in this interim analysis, numerically longer median overall survival was observed; these data are immature and survival follow-up will continue.As with previous atezolizumab and chemotherapy combination trials,22–24 no new adverse event signals were observed, the
combination safety profile was consistent with those of the individual therapeutic agents, and the safety profile was similar to platinum-based chemo therapy alone. These data establish the benefit of adding atezolizumab to standard platinum-based chemotherapy for first-line treatment of metastatic urothelial carcinoma.
For several decades, the gold-standard treatment for metastatic urothelial carcinoma has been cisplatin-based chemotherapy. However, many patients are ineligible for cisplatin treatment and instead receive inferior carboplatin-based regimens. IMvigor130 was initially designed to solely test addition of atezolizumab to carboplatin-based chemotherapy in cisplatin-ineligible patients. Based on updated efficacy data for first-line atezolizumab in metastatic urothelial carcinoma, the study design was modified to include an atezolizumab monotherapy group and broaden the patient population to include cisplatin-eligible patients.25,26 The improvement
in progression-free survival and encouraging interim overall survival data with addition of atezolizumab to platinum-based chemotherapy were achieved across key patient subgroups, although a larger effect size was observed in the subset of patients who received cisplatin compared with those who received carboplatin. However, the 95% CI overlapped the overall effect size for all subgroups analysed and the possible imbalance in baseline prognostic factors complicates dissection of the effect of cisplatin-based versus carboplatin-based chemo-therapy. Longer survival follow-up is needed to better understand the potential interaction between platinum-based regimens and atezolizumab and the effect on outcomes.
There are potential limitations to the study. In the treatment groups including platinum-based chemo-therapy, 145 (40%) of 365 patients deemed cisplatin eligible received carboplatin-based chemotherapy, and 53 (11%) of 485 patients deemed cisplatin ineligible received cisplatin. Nuances to real-world cisplatin use probably extend beyond established cisplatin ineligibility criteria. Furthermore, cisplatin eligibility is dynamic and patients with metastatic urothelial carcinoma deemed
eligible for cisplatin-based chemotherapy on initial evaluation might not be suitable candidates for cisplatin by the time of treatment initiation. As this trial was among the largest to explore first-line treatment of metastatic urothelial cancer, a pragmatic approach was taken to permit investigator discretion regarding use of cisplatin versus carboplatin to facilitate accrual and improve generalisability. In the time since this study was designed, the nuances of cisplatin use versus carboplatin use have been further highlighted in a retrospective analysis of patients treated in major academic centres, showing that around 26% of patients considered cisplatin eligible were treated with carboplatin-based chemo-therapy.3 At the time of the analysis, 79(20%) patients in
the placebo group had received subsequent immune checkpoint blockade, although this figure will almost certainly change with longer follow-up. However, previous studies show that only a subset of patients with metastatic urothelial carcinoma are able to receive subsequent therapy which might, in part, underlie the potential benefit of first-line combination therapy.27
Atezolizumab monotherapy was granted accelerated approval by the FDA in 2017 for first-line treatment of cisplatin-ineligible patients with metastatic urothelial carcinoma based on a single-group phase 2 trial. An unplanned IDMC review of the early IMvigor130 survival data led to recommendations to cease further accrual of patients with low tumour PD-L1 expression (IC0/1) to the atezolizumab monotherapy group.28
Due to these events, and because of analogous findings with pembrolizumab in the KEYNOTE-361 trial, in June, 2018, the FDA and EMA revised both labels to limit PD-L1 and PD-1 blockade therapies in the first-line metastatic urothelial carcinoma setting to patients who are cisplatin ineligible with increased tumour PD-L1 expression or ineligible for any platinum-based chemo-therapy (USA only).
In the planned interim analysis, the overall survival of patients randomly assigned to atezolizumab monotherapy versus platinum-based chemotherapy was not formally tested given the hierarchical testing procedure. Survival curves in the intention-to-treat population favoured platinum-based chemotherapy earlier but atezolizumab later, and ultimately also appeared to cross in the PD-L1 IC0/1 subgroup. This phenomenon was not apparent in the PD-L1 IC2/3 subgroup, with the survival curves favouring atezolizumab, similar to real-world results for atezolizumab.29 Other clinical and biological parameters
beyond PD-L1 expression could affect benefit from immune checkpoint blockade; this observation is reinforced by the crossing of the survival curves, even in the IC0/1 subgroup. The median duration of response in the intention-to-treat monotherapy group is not yet estimable, with the lower limit of the 95% CI at 15·9 months. Although the median response duration in the combination group was shorter compared with the median response duration in the atezolizumab alone
group, this is probably because the median in the combination group was predominantly influenced by patients responding to chemotherapy, whereas all the responses in the atezolizumab group were secondary to immunotherapy, which is generally associated with more durable responses. Additionally, patients treated with atezolizumab monotherapy had substantially fewer grade 3 or 4 adverse events versus those treated with platinum-based chemotherapy. These findings underscore the complex response kinetics of immune checkpoint blockade relative to platinum-based chemotherapy, high-light the limitations of focusing solely on early periods of survival follow-up, and reinforce the role for first-line atezolizumab monotherapy in select patients with metastatic urothelial carcinoma, potentially guided by platinum-based chemotherapy eligibility, PD-L1 expres-sion, and patient preference.
In conclusion, this phase 3 trial of first-line metastatic urothelial carcinoma treatment showed that addition of atezolizumab to platinum-based chemotherapy was asso-ciated with a significant prolongation of progression-free survival, encouraging interim overall survival data, an increased complete response rate, and a safety profile consistent with those seen with the individual agents. The efficacy and safety of first-line atezolizumab monotherapy in metastatic urothelial carcinoma observed in a previous phase 2 study limited to cisplatin-ineligible patients was recapitulated in this large, randomised study. IMvigor130 has already affected clinical practice, providing context for an evolving landscape of treatment approaches for first-line metastatic urothelial carcinoma, fortifying a potential role for monotherapy in select patients and, based on the totality of data, supporting atezolizumab plus platinum-based chemotherapy as a potential treatment option.
Contributors
AB, SB, MDS, MDG, EG, AT, AM, and NJV conceived and designed the study. IDD, MDS, MDG, and AM developed the methodology. ARK did the literature search. BA, AB, SB, IDD, MDG, XG-d-M, UDG, KI, J-RL, SM, MM, MO, SP, SHP, ARK, FAS, AT, NJV, and DY collected the data. BA, JAAA, SB, IDD, MDS, MDG, XG-d-M, UDG, EG, EK, J-RL, SM, AM, MM, MO, SP, SHP, ARK, FAS, AT, DY, and DT analysed and interpreted the data. AM did the statistical analysis. JAAA, IDD, MDS, MDG, UDG, EG, MG, J-RL, AM, MM, MO, ARK, FAS, AT, and NJV drafted and revised the manuscript. BA, JAAA, AB, SB, IDD, MDS, MDG, XG-d-M, UDG, EG, MG, KI, EK, SM, AM, MM, MO, SHP, FAS, AT, NJV, DY, and DT critically revised the manuscript for intellectual content. IDD, MG, and AT oversaw author review of the report. JAAA and MG provided administrative, technical, or material support. JAAA selected candidates and recruited and treated patients. JAAA, AB, and IDD were part of the steering committee. All authors approved the final version of the submitted report and agree to be accountable for all aspects. All authors verify that this study was done per protocol and vouch for data accuracy and completeness.
Declaration of interests
All authors report editorial support from F Hoffmann-La Roche. MDG has received personal fees from BioMotiv, Janssen, Merck, GlaxoSmithKline, Eli Lilly, Astellas, Genentech, Bristol-Myers Squibb, Pfizer, EMD Serono, AstraZeneca, Seattle Genetics, Aileron Therapeutics, Dracen, Inovio Pharmaceuticals, and Dragonfly Therapeutics outside the submitted work; grants from Merck, Genentech, Bristol-Myers Squibb, AstraZeneca, and Dendreon outside the submitted work; and is a co-founder of Rappta Therapeutics outside the submitted work. JAAA has received third-party
service as collaboration in the selection and management of administrative requirements for the participation of SOGUG centres in the study from SOGUG during the conduct of the study; speaking or consulting fees from Bristol-Myers Squibb, MSD, Roche, Astellas, Janssen Cilag, Pfizer, Novartis, and Bayer outside the submitted work; travel support from Bristol-Myers Squibb, MSD, Roche, and Janssen Cilag outside the submitted work; research funding (from SOGUG) outside the submitted work from Bristol-Myers Squibb, Novartis, and Pierre Fabre; and has participated in industry-sponsored clinical trials for Bristol-Myers Squibb, MSD, Roche, Astellas, Janssen Cilag, Pfizer, and Novartis outside the submitted work. AB has received personal fees from Roche, Bristol-Myers Squibb, MSD, AstraZeneca, and Pfizer outside the submitted work; grants from Bristol-Myers Squibb, AstraZeneca, and Pfizer outside the submitted work; and non-financial support from Pfizer outside the submitted work. IDD has received institutional payments for conduct of the IMvigor130 trial and other clinical trials from Roche during the conduct of the study; has been a member of an international steering committee
(unremunerated); and has been a member of a local Roche genitourinary cancers advisory board in 2016 (unremunerated—all honoraria paid to Australian and New Zealand Urogenital and Prostate Cancer Trials Group). UDG has received personal fees from Pfizer, Janssen, Astellas, Sanofi, Bristol-Myers Squibb, Bayer, Ipsen, and Merck; and travel support from Bristol-Myers Squibb, Ipsen, Janssen, Pfizer, and Roche. MDS has received honoraria for advisory functions and lecture fees from MSD, Roche, Merck, Pfizer, and AstraZeneca during the conduct of the study; study fees from MSD and Roche during the conduct of the study; honoraria for advisory functions from Janssen, Astellas, Sanofi, Sandoz, Ipsen, Clovis, Bayer, and Bristol-Myers Squibb outside the submitted work; lecture fees from Janssen, Astellas, and Sanofi outside the submitted work; and study fees from Janssen, Astellas, Clovis, and Bayer outside the submitted work. EK has received honoraria from Chugai during the conduct of the study; grants from MSD, Nippon Kayaku, Astellas, Ono, Takeda, Taiho, and AstraZeneca outside the submitted work; honoraria from Kissei, Nippon Shinyaku, and Pfizer outside the submitted work; and has a financial relationship with ASKA outside the submitted work. XG-d-M has received consulting fees from Pfizer, Ipsen, Bristol-Myers Squibb, Roche, Lilly, Pharmamar, EUSA, and Eisai outside the submitted work; speaker fees from Pfizer, Ipsen, Bristol-Myers Squibb, Pharmamar, and Eisai outside the submitted work; and a research grant from AstraZeneca. MM has received speaker fees for national meetings from Roche outside the submitted work. KI has received grants from Chugai during the conduct of the study; personal fees for lectures from Chugai, Novartis, Bristol-Myers Squibb, Astellas, Ono, Yakult, Pfizer, Taiho, MSD, and AstraZeneca outside the submitted work; personal fees for expert testimony from Bayer outside the submitted work; and grants from MSD and AstraZeneca outside the submitted work. MG has received institutional and personal investigator fees from Roche during the conduct of the study; institutional and personal advisory fees from Roche, MSD, and Lilly outside the submitted work; and investigator fees from MSD and Lilly outside the submitted work. MO has received honoraria for themselves and their institution as a speaker or moderator in scientific meetings from Novartis, Roche, Janssen, Sanofi, and Astellas; travel, accommodation, or expenses support from Bristol-Myers Squibb and Janssen; and has been an advisory board member for Janssen, Sanofi, and Astellas. ARK has received personal fees for being on advisory boards and speakers bureaus from Exelixis, AstraZeneca, Bayer, Pfizer, Novartis, Genentech, Roche, Bristol-Myers Squibb, EMD Sorono, Janssen, Astellas Medivation, Sanofi, Eisai, Amgen, and Merck outside the submitted work. BA has received grants from P Herzen Oncology Research Institute during the conduct of the study; grants from AstraZeneca, Bayer, Bristol-Myers Squibb, Janssen, Astellas, MSD, Eisai, and Roche outside the submitted work; personal fees from AstraZeneca, Bayer, Bristol-Myers Squibb, Janssen, Astellas, MSD, Sanofi, Ferring, Ipsen, Eisai, and Roche outside the submitted work; and non-financial support from AstraZeneca, Bayer, Bristol-Myers Squibb, Janssen, Astellas, MSD, Sanofi, Ferring, and Roche outside the submitted work. FAS has received personal fees from Roche, MSD, Bristol-Myers Squibb, and Janssen during the conduct of the study; and personal fees from Janssen, Bayer, Astellas, Ipsen, and Pfizer outside the submitted work. J-RL has received invited speaker or moderator honoraria from Roche, MSD, AstraZeneca, Astella, Pfizer, Ipsen, Johnson & Johnson, Novartis, Bristol-Myers Squibb, Ono, Amgen, and Bayer
outside the submitted work. NJV has received grants for clinical trial support and advisory board fees from Genentech, Roche, Pfizer, Merck, Eisai, and Exelixis during the conduct of the study; personal fees for legal consultation and advisory board fees from Novartis during the conduct of the study; grants for clinical trial support from Calithera during the conduct of the study; advisory board fees from Caris during the conduct of the study; and fees for being a part of the editorial board from UpToDate during the conduct of the study. SB is an employee and has stock ownership of F Hoffmann-La Roche. DT is an employee of Roche, Genentech and has stock ownership of F Hoffmann-La Roche. SM has a patent (US-8217149B2) issued describing atezolizumab invention (anti-PD-L1 antibodies, compositions, and articles of manufacture). AM is an employee and has non-voting stock ownership of F Hoffmann-La Roche. AT is an employee and has stock ownership of Genentech. EG has received advisory fees from Roche, Pfizer, Bristol-Myers Squibb, Ipsen, EUSA, MSD, Sanofi, Janssen, Astellas, and Adacap during the conduct of the study; research grants from Roche, Pfizer, Bristol-Myers Squibb, Ipsen, EUSA, Lexicon, MTEM Threshold, and AstraZeneca during the conduct of the study; and lecture fees from Pierre Fabre during the conduct of the study. SP, SHP, and DY declare no other competing interests.
Data sharing
Qualified researchers can request access to individual patient-level data through the clinical study data request platform. Further details on Roche’s criteria for eligible studies are available online. Further details on Roche’s Global Policy on the Sharing of Clinical Information and how to request access to related clinical study documents are available online.
Acknowledgments
This study was sponsored by F Hoffmann-La Roche and Genentech, a member of the Roche Group. We thank the patients who participated in the trial and the clinical site investigators (appendix pp 2–10). We also thank Himika Patel (Genentech) for her contributions to the study. IDD is supported by an National Health and Medical Research Council Practitioner Fellowship (APP1102604). Support for third-party writing assistance for this manuscript, given by Paige S Davies(Health Interactions), was provided by F Hoffmann-La Roche.
References
1 Loehrer PJ Sr, Einhorn LH, Elson PJ, et al. A randomized comparison of cisplatin alone or in combination with methotrexate, vinblastine, and doxorubicin in patients with metastatic urothelial carcinoma: a cooperative group study. J Clin Oncol 1992; 10: 1066–73. 2 von der Maase H, Sengelov L, Roberts JT, et al. Long-term survival
results of a randomized trial comparing gemcitabine plus cisplatin, with methotrexate, vinblastine, doxorubicin, plus cisplatin in patients with bladder cancer. J Clin Oncol 2005; 23: 4602–08. 3 Bamias A, Tzannis K, Harshman LC, et al. Impact of contemporary
patterns of chemotherapy utilization on survival in patients with advanced cancer of the urinary tract: a retrospective international study of invasive/advanced cancer of the urothelium (RISC). Ann Oncol 2018; 29: 361–69.
4 Galsky MD, Chen GJ, Oh WK, et al. Comparative effectiveness of cisplatin-based and carboplatin-based chemotherapy for treatment of advanced urothelial carcinoma. Ann Oncol 2012; 23: 406–10. 5 Gartrell BA, He T, Sharma J, Sonpavde G. Update of systemic immunotherapy for advanced urothelial carcinoma. Urol Oncol 2017; 35: 678–86.
6 Balar AV, Galsky MD, Rosenberg JE, et al. Atezolizumab as first-line treatment in cisplatin-ineligible patients with locally advanced and metastatic urothelial carcinoma: a single-arm, multicentre, phase 2 trial. Lancet 2017; 389: 67–76.
7 Balar AV, Castellano D, O’Donnell PH, et al. First-line
pembrolizumab in cisplatin-ineligible patients with locally advanced and unresectable or metastatic urothelial cancer (KEYNOTE-052): a multicentre, single-arm, phase 2 study. Lancet Oncol 2017;
18: 1483–92.
8 Powles T, Durán I, van der Heijden MS, et al. Atezolizumab versus chemotherapy in patients with platinum-treated locally advanced or metastatic urothelial carcinoma (IMvigor211): a multicentre, open-label, phase 3 randomised controlled trial. Lancet 2018;
391: 748–57.
9 Rosenberg JE, Hoffman-Censits J, Powles T, et al. Atezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed following treatment with platinum-based chemotherapy: a single-arm, multicentre, phase 2 trial. Lancet 2016;
387: 1909–20.
10 Massard C, Gordon MS, Sharma S, et al. Safety and efficacy of durvalumab (MEDI4736), an anti-programmed cell death ligand-1 immune checkpoint inhibitor, in patients with advanced urothelial bladder cancer. J Clin Oncol 2016; 34: 3119–25.
11 Sharma P, Retz M, Siefker-Radtke A, et al. Nivolumab in metastatic urothelial carcinoma after platinum therapy (CheckMate 275): a multicentre, single-arm, phase 2 trial. Lancet Oncol 2017; 18: 312–22. 12 Apolo AB, Infante JR, Balmanoukian A, et al. Avelumab,
an anti-programmed death-ligand 1 antibody, in patients with refractory metastatic urothelial carcinoma: results from a multicenter, phase Ib study. J Clin Oncol 2017; 35: 2117–24. 13 Chen DS, Mellman I. Oncology meets immunology: the
cancer-immunity cycle. Immunity 2013; 39: 1–10.
14 Hato SV, Khong A, de Vries IJ, Lesterhuis WJ. Molecular pathways: the immunogenic effects of platinum-based chemotherapeutics. Clin Cancer Res 2014; 20: 2831–37.
15 Palmer AC, Sorger PK. Combination cancer therapy can confer benefit via patient-to-patient variability without drug additivity or synergy. Cell 2017; 171: 1678–91.
16 Flannery K, Boyd M, Black-Shinn J, Robert N, Kamat AM. Outcomes in patients with metastatic bladder cancer in the USA: a retrospective electronic medical record study. Future Oncol 2019; 15: 1323–34. 17 Genentech. Tecentriq (atezolizumab). Highlights of prescribing
information. https://www.gene.com/download/pdf/tecentriq_ prescribing.pdf (accessed Jan 27, 2020).
18 TECENTRIQ (atezolizumab) [summary of product characteristics]. Welwyn Garden City, UK: Roche Registration Limited; 2019. https://www.ema.europa.eu/en/documents/product-information/ tecentriq-epar-product-information_en.pdf (accessed March 10, 2020). 19 Merck. Keytruda (pembrolizumab). Highlights of prescribing
information. https://www.merck.com/product/usa/pi_circulars/k/ keytruda/keytruda_pi.pdf (accessed Jan 27, 2020).
20 Galsky MD, Hahn NM, Rosenberg J, et al. A consensus definition of patients with metastatic urothelial carcinoma who are unfit for cisplatin-based chemotherapy. Lancet Oncol 2011; 12: 211–14. 21 Wassmer G. Planning and analyzing adaptive group sequential
survival trials. Biom J 2006; 48: 714–29.
22 Socinski MA, Jotte RM, Cappuzzo F, et al. Atezolizumab for first-line treatment of metastatic nonsquamous NSCLC. N Engl J Med 2018; 378: 2288–301.
23 Schmid P, Adams S, Rugo HS, et al. Atezolizumab and nab-paclitaxel in advanced triple-negative breast cancer. N Engl J Med 2018; 379: 2108–21.
24 Horn L, Mansfield AS, Szczęsna A, et al. First-line atezolizumab plus chemotherapy in extensive-stage small-cell lung cancer. N Engl J Med 2018; 379: 2220–29.
25 Sonpavde G, Galsky MD, Vogelzang NJ. First-line systemic therapy trials for advanced transitional-cell carcinoma of the urothelium: should we stop separating cisplatin-eligible and -ineligible patients? J Clin Oncol 2010; 28: e441–42.
26 Balar AV, Dreicer R, Loriot Y, et al. Atezolizumab (atezo) in first-line cisplatin-ineligible or platinum-treated locally advanced or metastatic urothelial cancer (mUC): long-term efficacy from phase 2 study IMvigor210. J Clin Oncol 2018; 36: 4523.
27 Galsky MD, Pal SK, Mortazavi A, et al. Randomized double-blind phase 2 study of maintenance pembrolizumab versus placebo after first-line chemotherapy in patients (pts) with metastatic urothelial cancer (mUC): HCRN GU14-182. J Clin Oncol 2019; 37: 4504. 28 Suzman DL, Agrawal S, Ning YM, et al. FDA approval summary:
atezolizumab or pembrolizumab for the treatment of patients with advanced urothelial carcinoma ineligible for cisplatin-containing chemotherapy. Oncologist 2019; 24: 563–69.
29 Vander Velde N, Guerin A, Ionescu-Ittu R, et al. Comparative effectiveness of non-cisplatin first-line therapies for metastatic urothelial carcinoma: phase 2 IMvigor210 study versus US patients treated in the Veterans Health Administration. Eur Urol Oncol 2019;
2: 12–20.
For the clinical study data
request platform see www.
clinicalstudydatarequest.com For more on Roche’s criteria for
eligible studies see
https://clinicalstudydatarequest. com/Study-Sponsors/Study-Sponsors-Roche.aspx For more on Roche’s Global
Policy on the Sharing of Clinical information and how to request access to related clinical study documents see
http://www.roche.com/research_ and_development/who_we_are_ how_we_work/clinical_trials/ our_commitment_to_data_ sharing.htm
