The primary variable in this study was ORR, defined as the proportion of patients with a BOR of CR or PR (by Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST v1.1)). A BOR was defined as the best response (in the order of CR, PR, stable disease (SD), no evidence of disease (NED), progressive disease (PD), and not evaluable (NE)) among all post-baseline disease assessments that occurred until progression, or last evaluable assessment in the absence of progression prior to the initiation of subsequent anti-cancer therapy, irrespective of whether or not patients discontinued the study treatment. The 95% Pearson-Clopper confidence interval (CI) of ORR for each treatment arm was provided. Per RECIST v1.1 for target lesions and assessed by computed tomography (CT) or, if contraindicated, magnetic resonance imaging (MRI): CR=disappearance of all target lesions; PR=at least 30% decrease from baseline in the sum of diameters of target lesions. Overall Response=CR+PR.

In order to meet the Food and Drug Administration (FDA) requirement, the primary efficacy analysis of ORR was performed using the Intent-to-Treat (ITT) population (patients included in each treatment arm as randomized), and the blinded independent central review (BICR) radiological assessments.

ORR (by RECIST v1.1) at Week 19 was defined as the proportion of patients with a BOR of CR or PR assessed at Week 19. Only tumor assessments performed up until 19 weeks (i.e. Week 18 assessment + 7 day assessment window) from randomization were considered in this analysis. Per RECIST v1.1 for target lesions and assessed by CT or, if contraindicated, MRI: CR=disappearance of all target lesions; PR=at least 30% decrease from baseline in the sum of diameters of target lesions. Overall Response=CR+PR.

In order to meet the FDA requirement, the secondary efficacy analysis of ORR was performed using the ITT population (patients included in each treatment arm as randomized), and the BICR radiological assessments.

The event of interest for PFS was defined as the interval from the date of randomization until first documented disease progression or death from any cause, whichever occurs first. Disease progression was based on tumor assessments according to RECIST v1.1 criteria. The items of the overall response CR, PR, SD and NED were taken as progression-free whereas PD denoted disease progression. PFS was summarized using Kaplan-Meier estimates of the quartiles for each treatment arm, and 95% CIs for the medians were calculated. Per RECIST v1.1 for target lesions and assessed by CT or, if contraindicated, MRI: PD=at least a 20% increase in the sum of the longest diameter of target lesions, unequivocal progression of existing non-target lesions, or the appearance of one or more new lesions.

This assessment was performed using the ITT population (patients included in each treatment arm as randomized) and the BICR radiological assessments.

The event of interest was defined as death from any cause. OS was defined as the interval from date of randomization until the date of death due to any cause. OS was summarized using Kaplan-Meier estimates of the quartiles for each treatment arm, and 95% CIs for the medians were calculated.

This assessment was performed using the ITT population (patients included in each treatment arm as randomized).

DOR was evaluated in this study as a secondary efficacy endpoint. Only the patients defined as responders in the primary analysis of ORR were taken into account for the analysis of DOR. The event of interest was defined as first documented disease progression or death due to any reason, whichever occurred first. DOR was defined as the interval from the first documented response (as defined per RECIST v1.1) until the earlier date of the first documented disease progression or death due to any reason. The date of first documented response was taken as the date of the first tumor assessment with an overall visit response of CR or PR. Per RECIST v1.1 for target lesions and assessed by CT or, if contraindicated, MRI: CR=disappearance of all target lesions; PR=at least 30% decrease from baseline in the sum of diameters of target lesions. DOR was calculated in units of months.

This assessment of DOR was based on patients with events in the ITT population (patients included in each treatment arm as randomized), and the BICR radiological assessments.

The DCR was defined as the proportion of patients defined as responders. The number and percentage of responders and non-responders and the 95% Pearson-Clopper CI of DCR for each treatment arm was provided. The odds ratio for treatment (FKB238 arm versus Avastin arm) and the corresponding 95% Wald CI were produced based on a logistic regression analysis of DCR. Per RECIST v1.1 for target lesions and assessed by CT or, if contraindicated, MRI: CR=disappearance of all target lesions; PR=at least 30% decrease from baseline in the sum of diameters of target lesions; SD=neither sufficient shrinkage to qualify for PR nor sufficient increase to qualify for PD, taking as reference the smallest sum of diameters while on study. DCR=CR+PR+SD (≥ 6 weeks).

This assessment of DCR was based on patients in the ITT population (patients included in each treatment arm as randomized), and the BICR radiological assessments.

Ctrough (pre-infusion) and serum maximum concentration (Cmax; at completion of infusion) were compared between treatment arms and time points, and descriptive statistics provided. Ctrough and Cmax concentrations were summarized using the pharmacokinetics (PK) population for each visit at which samples were taken. The pre-dose serum concentrations at Cycles 2, 4, and 6 were considered as Ctrough values and the post-dose serum concentrations at Cycles 1 and 4 were considered as Cmax values. PK data at Cycle 1 Day 1 pre-infusion were not calculable and are therefore not presented in the outcome measure data table. Data are only provided for the time points at which the serum trough concentration was measured.

All patients randomized to treatment who received at least 1 dose of investigational product with no important protocol deviations, and at least 1 serum drug concentration data after investigational product administration (PK population).

The ADA levels were summarized at baseline and post-baseline time points using descriptive statistics.

Includes all patients randomized to treatment who received at least 1 dose of investigational product with no important protocol deviations, and who had non-missing baseline ADA and at least 1 non-missing post-baseline ADA result (ADA evaluable population).