Whether the body weight loss before the diagnosis of pancreatic cancer affects patients ' prognosis has been a controversial matter. The authors showed no association between body weight loss and prognosis in the patients with advanced pancreatic cancer who received either chemotherapy or best supportive care. I suspect that far advanced disease in almost entire patients in this series might have affected the results. I have several comments as below.

1. The authors collected the data regarding the body weight loss by the time of disease diagnosis. It should be described clearly how the authors collected the data of each patient's body weight before the referral. Were they dependent on the interview to each patient? In that case, how the data accuracy was ensured?
2. In addition, the international consensus defines cachexia as weight loss >5% over 6 months. Did the authors concerned the period that the weight loss occurred?
3. I wonder why the authors excluded the patients who received 119 patients with surgery. The influence of body weight loss in the patients with resectable pancreatic cancer is also a controversial issue, and presenting the data in the group of these patients would enhance the impact of this study.
4. I suspect cachexia would affect the tolerance of chemotherapy. Was there difference in the duration of chemotherapy or the incidence of abandoning treatment due to adverse events between the patients with and without body weight loss?
5. According to the presented survival curves, so many patients were censored within short period after the diagnosis. Were these patients lost to follow so early?

The authors adequately responded to reviewer's comments. I have no additional comments for furher revision.