A Study of Atezolizumab in Locally Advanced or Metastatic Urothelial or Non-Urothelial Carcinoma of the Urinary Tract

AEs were defined as any untoward medical occurrence in a subject administered a pharmaceutical product, regardless of causal attribution. An AE can be any unfavorable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product, whether or not considered related to the medicinal product. New disease, exacerbation of existing disease, recurrence of an intermittent medical condition not present at baseline, any deterioration in a laboratory value or other clinical test associated with symptoms or leading to a change in study/concomitant treatment or discontinuation from study drug as well as events related to protocol-mandated interventions are considered AEs.

Time frame: Baseline up to end of study (up to approximately 6 years)

Population: The safety population included all enrolled participants who received at least one dose of study medication.

The EuroQol 5-Dimension Questionnaire (EQ-5D-5L) is a self-report health status questionnaire that consists of 6 questions used to calculate a health utility score for use in health economic analysis. There are two components to the EuroQol EQ-5D: 1) five health dimensions that assess mobility, self-care, usual activities, pain/discomfort, and anxiety/depression; 2) a visual analogue scale (VAS) that measures health state. There are 5 response levels for each dimension (1=no problems, 2=slight problems, 3=moderate problems, 4=severe problems, and 5=extreme problems) with the highest level representing the worst outcome. The VAS is scored on a scale from 0 to 100, with 0 representing the worst imaginable health and 100 representing the best imaginable health.

Time frame: Baseline, Day 1 of Cycles 1, 2, 3 and thereafter every 9 weeks for 54 weeks from study treatment start; and then every 12 weeks until progression/study discontinuation (up to approximately 6 years) (Cycle length = 21 days)

Population: The ITT population included all enrolled participants who completed the questionnaire at baseline and had 1 post-baseline assessment.

The EORTC QLQ-C30 included global health status, functional scales (physical, role, emotional, cognitive, and social), symptom scales (fatigue, nausea/vomiting, and pain) and single items (dyspnoea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties). Most questions used a 4-point scale (1 'Not at all' to 4 'Very much'; 2 questions used 7-point scale \[1 'very poor' to 7 'Excellent'\]). Scores were averaged and transformed to 0 - 100 scale. Higher scores on the global health status and functional scales indicated better health status/function. Higher scores on the symptoms scales and symptom items indicated greater symptom burden.

Time frame: Baseline, Day 1 of Cycles 1, 2, 3 and thereafter every 9 weeks for 54 weeks from study treatment start; and then every 12 weeks until progression/study discontinuation (up to approximately 6 years) (Cycle length = 21 days)

Population: The ITT population included all enrolled participants who completed the questionnaire at baseline and had 1 post-baseline assessment.

Duration of response was determined separately on disease status using modified RECIST by the investigator. For overall responders, DoR was defined as the time from the date of first occurrence of a confirmed response (complete response or partial response) to date of tumor progression following that confirmed response or death from any cause, or to censoring date: 1) end of response was the date of tumor progression after that confirmed response or death (in the absence of tumor progression), 2) for a participant without disease progression or death following a response, the censored end of response was the latest RECIST assessment or start of subsequent cancer therapy, whichever occurred first.

Time frame: Time from first occurrence of a documented response to disease progression or death from any cause, whichever occurred first (up to approximately 6 years)

Population: The ITT population included all enrolled participants. Only participants with a response were analyzed for this outcome measure.

Duration of response was determined separately on disease status using RECIST v1.1 by the investigator. For overall responders, DoR was defined as the time from the date of first occurrence of a confirmed response (complete response or partial response) to date of tumor progression or death from any cause, or to censoring date: 1) end of response coincided with the date of tumor progression or death (in the absence of tumor progression) used for the PFS endpoint, 2) for a participant without disease progression or death following a response, the censored end of response coincided with the PFS censoring date (that was latest RECIST assessment or start of subsequent cancer therapy, whichever occurred first).

Time frame: Time from first occurrence of a documented response to disease progression or death from any cause, whichever occurred first (up to approximately 6 years)

Population: The ITT population included all enrolled participants. Only participants with a response were analyzed for this outcome measure.

OS was defined as date of death (due to any cause) or censoring minus date of start of study treatment plus 1.

Time frame: Randomization until death from any cause (up to approximately 6 years)

Population: The intent-to-treat (ITT) population included all enrolled participants.

BOR was assessed by the investigators according to the RECIST v1.1. BOR was defined as a complete response (CR) or partial response (PR) determined on two consecutive investigator assessments \>= 4 weeks apart in participants with measurable disease at baseline. CR = Disappearance of all target lesions. Any pathological lymph nodes (whether target or nontarget) must have reduction in short axis to \<10 millimeters (mm); PR = At least a 30% decrease in the sum of diameters of target lesions, taking as reference the baseline sum of diameters. Progressive Disease (PD) = At least a 20% increase in the sum of diameters of target lesions, taking as reference the smallest sum on study (nadir), including baseline. In addition to the relative increase of 20%, the sum must have demonstrated an absolute increase of at least 5 mm. Stable Disease (SD) = Neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD, taking as reference the smallest sum on study.

Time frame: Randomization up to disease progression or death from any cause, whichever occurred first (up to approximately 6 years)

Population: The ITT population included all enrolled participants.

BOR was assessed by the investigators according to the modified RECIST. BOR was defined as complete response (CR) or partial response (PR). CR includes complete disappearance of all tumor lesions and no new measurable or unmeasurable lesions confirmed by a consecutive assessment \>=4 weeks from the first documented date. PR is a decrease in the sum of the diameters of all target and all new measurable lesions \>=30%, relative to baseline, in the absence of CR confirmed by a consecutive assessment \>=4 weeks from the first documented date. The assessment of BOR included post-screening RECIST assessments obtained up to: 1) death from any cause, 2) last evaluable RECIST assessment in the absence of death, 3) start of a subsequent anti-cancer therapy, whichever occurred first.

Time frame: Randomization up to disease progression or death from any cause, whichever occurred first (up to approximately 6 years)

Population: The ITT population included all enrolled participants.

Disease control was determined separately on disease status using modified RECIST by the investigator. Disease control rate was defined as the sum of the complete response, partial response, and stable disease rates.

Time frame: Randomization up to disease progression or death from any cause, whichever occurred first (up to approximately 6 years)

Population: The ITT population included all enrolled participants.

Disease control was determined separately on disease status using RECIST v1.1 by the investigator. Disease control rate was defined as the sum of the complete response, partial response, and stable disease rates.

Time frame: Randomization up to disease progression or death from any cause, whichever occurred first (up to approximately 6 years)

Population: The ITT population included all enrolled participants.

PFS as per Modified RECIST was defined as: * date of first occurrence of tumor progression after a modified confirmed response if the participant was a responder according to modified RECIST or * date of first occurrence of tumor progression in case the participant was not a responder according to modified RECIST or * date of death (in the absence of tumor progression) by any cause, or * date of censoring whichever occurred first, minus date of start of study treatment plus 1

Time frame: Randomization up to disease progression or death from any cause, whichever occurred first (up to approximately 6 years)

Population: The ITT population included all enrolled participants.

PFS was defined as the date of first occurrence of tumor progression (earliest of the dates of the RECIST component indicating tumor progression) or date of death (in the absence of tumor progression) by any cause, whichever occurred first, or date of censoring minus date of start of study treatment plus 1.

Time frame: Randomization up to disease progression or death from any cause, whichever occurred first (up to approximately 6 years)

Population: The ITT population included all enrolled participants.