TTF was defined as the time after the first dose of study therapy to discontinuation of study therapy due to disease progression, death by any cause, or the start of a new non-protocol-defined anticancer therapy/surgery. If a participant does not progress, die or start a new non-protocol-defined anticancer therapy, the participant was censored on the last tumor assessment date.

Tumor evaluations of CT or MRI scans were assessed by the investigative sites and response determined according to Response Evaluation Criteria in Solid Tumors (RECIST) guidelines, version 1.1.

The definition for progressive disease (PD) was >= 20% increase in the sum of diameters of target lesions from nadir, and the sum showed an absolute increase of >= 5 mm; the progression of a non-target lesion or the appearance of any new lesions is also considered progression.

Median and its 90% confidence interval (CI) of TTF were estimated using the method of Brookmeyer and Crowley.

Intent to treat population was defined as all participants enrolled into the study.

DCR was defined as the percentage of participants with a CR or PR or SD from of date of first treatment to 16 weeks. Tumor assessments after start of non-protocol-defined anticancer therapy were excluded.

RECIST 1.1 Definition:

• CR: disappearance of all target and non-target lesions; any pathological lymph nodes (target or non-target) must have reduction in short axis to < 10 mm and no new lesions diagnosed.
• PR: a >= 30% decrease in the sum of diameters of target lesions from baseline; no evidence of progression in any of the non-target lesions diagnosed at baseline; and no new lesions diagnosed.
• SD: neither sufficient shrinkage to qualify for PR nor sufficient increase of lesions to qualify for PD.

The two-sided 90% binomial confidence intervals (CIs) were estimated by Wilson score method.

Intent to treat population was defined as all participants enrolled into the study.

ORR was defined as the percentage of participants that achieved a combined incidence of complete (CR) and partial response (PR) using RECIST 1.1 guidelines as assessed by the investigator. Assessments after new non-protocol-defined anticancer therapy are excluded. For participants who had resectable surgery in Investigator Choice period, assessments after surgical intervention are excluded.

RECIST 1.1 Definition:

• CR: disappearance of all target and non-target lesions; any pathological lymph nodes (target or non-target) must have reduction in short axis to < 10 mm and no new lesions diagnosed.
• PR: a >= 30% decrease in the sum of diameters of target lesions from baseline; no evidence of progression in any of the non-target lesions diagnosed at baseline; and no new lesions diagnosed.

The two-sided 90% binomial confidence intervals (CIs) were estimated by Wilson score method

Intent to treat population was defined as all participants who were enrolled into the study.

Progression-free Survival (PFS) was defined as the time from the date of the first dose to the date of disease progression or death (by any cause), whichever is earlier. The analysis day was calculated from enrollment date for one participant who was not treated. Participants who have no disease progression or have not died were censored to last tumor assessment date with progression-free.

The definition for progressive disease (PD) was at least a 20% increase in the sum of diameters of target lesions from nadir; the sum must also demonstrate an absolute increase of >= 5 mm; the progression of a non-target lesion or the appearance of any new lesions is also considered progression.

Median and its 90% confidence interval of PFS were estimated using the method of Brookmeyer and Crowley.

Intent to treat population was defined as all participants who were enrolled into the study.

Overall survival was defined as the time from the date of first dose of study therapy to the date of death (by any cause). Participants who were alive at the end of study or clinical data cut were censored on the last known time that the participant was alive or the clinical cutoff date, whichever was earlier. Median and its 90% confidence interval of OS were estimated using the method of Brookmeyer and Crowley

Intent to treat population was defined as all participants who were enrolled into the study.

The European Organization for Research and Treatment of Cancer Quality-of-Life questionnaire (EORTC QLQ-C30) is a validated health-related quality of life (HRQoL) measure. The EORTC QLQ-C30 is composed of both multi-item scales and single-item measures, including 5 functional scales, 3 symptom scales, 6 single symptom items, and 1 global health status / quality of life scale. No item occurs in more than one scale. All reported measures are transformed to a 0 - 100 scale. In the Global Health Status and 5 functional scales, 0 = worst possible quality of life/health status and 100 = best possible quality of life/health status. The best score on treatment is the best score from all post-baseline visits and is compared to the baseline to get the following responder categories. Responder categories: - Improved: >=10 increase from baseline - Stable: neither increase nor decrease >10 - Worsened: >=10 decrease from baseline

Intent to treat population was defined as all participants enrolled into the study with both baseline and post baseline values.

The European Organization for Research and Treatment of Cancer Quality-of-Life questionnaire (EORTC QLQ-C30) is a validated health-related quality of life (HRQoL) measure. The EORTC QLQ-C30 is composed of both multi-item scales and single-item measures, including 5 functional scales, 3 symptom scales, 6 single symptom items, and 1 global health status / quality of life scale. No item occurs in more than one scale. All reported measures are transformed to a 0 to 100 scale. In the symptom scales and single symptom items, 0 = optimal health state and 100 = worst possible health state. The best score on treatment is the best score from all post-baseline visits and is compared to the baseline to get the following responder categories. Responder categories: - Improved: >=10 decrease from baseline - Stable: neither increase nor decrease >10 - Worsened: >=10 increase from baseline

Intent to treat population was defined as all participants enrolled into the study with both baseline and post baseline values.

The EORTC pancreatic cancer module is a validated tool intended for patients at all disease stages undergoing surgical resection, palliative surgical intervention, endoscopic palliation or palliative chemotherapy. The module includes 26 questions, organized into 7 scales and 10 individual item scores. All reported measures are transformed to a 0 to 100 scale. Six summary scales reported are: - Pancreatic Pain - Digestive Symptoms - Altered Bowel Habits - Hepatic Scale - Body Image - Sexuality Scores of 0 = optimal health state and 100 = worst possible health state. The best score on treatment is the best score from all post-baseline visits and is compared to the baseline. Responder categories: - Improved: >=MID decrease from baseline - Stable: no increase or decrease >MID - Worsened: >=MID increase from baseline MID = half the baseline standard deviation

Intent to treat population was defined as all participants enrolled into the study with both baseline and post baseline values.

The EORTC pancreatic cancer module is a validated tool intended for patients at all disease stages undergoing surgical resection, palliative surgical intervention, endoscopic palliation or palliative chemotherapy. The module includes 26 questions, organized into 7 scales and 10 individual item scores. The summary scale for Satisfaction with Health Care is reported. All reported measures are transformed to a 0 to 100 scale. Scores of 0 = not satisfied, worst possible health state and 100 = extremely satisfied, best possible health state. The best score on treatment is the best score from all post-baseline visits and is compared to the baseline to get the following responder categories. Responder categories: - Improved: >=MID increase from baseline - Stable: no increase or decrease >MID - Worsened: >=MID decrease from baseline MID = half the baseline standard deviation

Intent to treat population was defined as all participants enrolled into the study with both baseline and post baseline values.

The EORTC pancreatic cancer module is a validated tool intended for patients at all disease stages undergoing surgical resection, palliative surgical intervention, endoscopic palliation or palliative chemotherapy. The module includes 26 questions, organized into 7 scales and 10 individual item scores. The 10 individual item scores are reported. All reported measures are transformed to a 0 to 100 scale. Scores of 0 = best possible health state and 100 = worst possible health state. The best score on treatment is the best score from all post-baseline visits and is compared to the baseline to get the following responder categories. Responder categories: - Improved: >=MID decrease from baseline - Stable: no increase or decrease >MID - Worsened: >=MID increase from baseline MID = half the baseline standard deviation

Intent to treat population was defined as all participants enrolled into the study with both baseline and post baseline values.

TEAEs are defined as any adverse event (AE) that begin or worsen on or after the start of study drug or procedure of the study period through the maximum duration of the period plus 28 days. The severity of AEs was graded based on National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE), Version 4.0 and the scale: Grade 1 = Mild Grade 2 = Moderate Grade 3 = Severe Grade 4 = Life threatening Grade 5 = Death. Relation to study drug was determined by the investigator. A treatment-related TEAE is defined as TEAE which was considered to be related to one or both of the study drugs and reported as 'Suspected' on the case report form. AEs with a missing relationship were treated as 'treatment-related' in data summaries. IP (investigational product) refers to nab-Paclitaxel and/or Gemcitabine. "Related" TEAE refers to relation to study drug (IP).

The Treated population consists of all participants who received at least 1 dose of nab-paclitaxel or gemcitabine.